Sustained T cell Rap1 signaling is protective in the collagen-induced arthritis model of rheumatoid...

Sm

all GTPases: em

erging targets in rheumatoid arthritis

Joana Abreu 2009

Uitnodiging

voor het bijwonen van de openbare verdediging van het proefschrift van

Joana Abreu

op

vrijdag 11 december 2009 om 12.00 uur

in

de AgnietenkapelOudezijds Voorburgwal

231Amsterdam

Receptie

ter plaatse na afloop van de promotie

Paranymfen

Tineke Cantaert([email protected])

Gabriela Salinas([email protected])

Parkeergarages: Muziektheater, Bijenkorf

Openbaar vervoer:Tram 4, 9, 16, 24, 25

Small GTPases:

emerging targets in rheumatoid arthritis

Joana Abreu

Small GTPases:emerging targets

in rheumatoid arthritis

Joana Abreu

Small GTPases: emerging targets in rheumatoid arthritisThesis, University of Amsterdam

© 2009 Joana Abreu, Amsterdam, The NetherlandsPrinted by Wöhrmann Print Service, Zutphen

The studies described in this thesis were performed at the Department of Clinical Immunology and Rheumatology and the Department of Experimen-tal Immunology, Academic Medical Center, University of Amsterdam, Am-sterdam, The Netherlands

The printing of this thesis was financially supported by: MRC-Holland, Abbott B.V., Roche Nederland B.V., Amgen B.V., UCB Pharma B.V., Wyeth Pharmaceuticals B.V., BD Biosciences, Clean Air Techniek B.V., PerkinElmer, Beckman Coulter, Greiner Bio-One, University of Amsterdam

Small GTPases:emerging targets

in rheumatoid arthritis

ACADEMISCH PROEFSCHRIFT

Ter verkrijging van de graad van doctoraan de Universiteit van Amsterdamop gezag van de Rector Magnificus

prof. dr. D.C. van den Boomten overstaan van een door het college voor promoties ingestelde

commissie, in het openbaar te verdedigen in de Agnietenkapel

op

vrijdag 11 december 2009, te 12.00 uur

door

Joana Raquel Ferreira de Abreu

geboren te Braga, Portugal

Promotiecommissie

Promotor: prof. dr. P.P. Tak

Co-promotor: dr. K.A. Reedquist

Overige leden: prof. dr. R.A.W. van Lier prof. dr. R.J.M. ten Berge prof. dr. S. Florquin prof. dr. B.J. Prakken dr. P.L. Hordijk dr. M.M. Maurice

Faculteit der Geneeskunde, Universiteit van Amsterdam

para os meus paispara a Lilavoor Rogier

Contents

Chapter 1 Introduction 9 Chapter 2 The presumed hyporesponsive behaviour of rheumatoid 35 arthritis T lymphocytes can be attributed to spontaneous ex vivo apoptosis rather than defects in T cell receptor signaling

Chapter 3 Sustained T cell Rap1 signaling is protective in the collagen 63 induced arthritis model of rheumatoid arthritis

Chapter 4 Antigen receptor and co-stimulatory signals differentially 87 regulate RapGAP family protein expression in human T lymphocytes

Chapter 5 The Ras guanine nucleotide exchange factor RasGRF1 109 promotes MMP-3 production in rheumatoid arthritis synovial tissue

Chapter 6 A Rac1 inhibitory peptide suppresses antibody production 137 and paw swelling in the murine collagen-induced arthritis model of rheumatoid arthritis

Chapter 7 General discussion and summary 157

Chapter 8 English summary 173

Chapter 9 Nederlandse samenvatting 179

Acknowledgements 187

Curriculum vitae 193

List of publications 197

Introduction

1

11

Introduction

Rheumatoid arthritis

Rheumatoid arthritis (RA) is a complex, chronic, autoimmune disease that affects approximately 1% of the population and has a higher incidence in women1. The pri-mary manifestations are joint pain, stiffness and swelling, and when it is not adequa-tely treated the synovial inflammation leads to erosion and destruction of cartilage, bone and periarticular structures. The lining of the synovium, which is adjacent to the joint space and is composed of differentiated macrophages and fibroblast-like synoviocytes (FLS), attaches to the joint at the bone–cartilage junction, and forms a destructive pannus. The synovium in patients with RA is characterized by cellu-lar hyperplasia, prominent angiogenesis and an influx of inflammatory leukocytes, mainly consisting of T and B lymphocytes, macrophages and plasma cells2. These cells contribute to the perpetuation of inflammation by producing pro-inflammatory cytokines, such as interleukin (IL)-1β and tumor necrosis factor (TNF)-α, that to-gether with secreted chemokines, promote the infiltration and activation of more leukocytes into the joints. Locally expressed degradative enzymes, including matrix metalloproteinases (MMPs), digest the extracellular matrix and destroy the articu-lar structures. Although the mechanisms involved in the initiation of RA are still unknown, increased knowledge on the pathogenesis of this disease has implicated several cell types as key players in autoimmune inflammation.

T lymphocytes

One of the strongest pieces of evidence for the involvement of T cells in RA is the in-creased frequency of expression of specific major histocompatibility complex (MHC) alleles in patients with RA. Shared epitopes present on HLA-DR1 and HLA-DR4 alleles are thought to present arthritogenic peptides to lymphocytes, which might initiate RA3. The presence of the shared epitope might also influence the severity of disease, as the risk of extra-articular and erosive disease is greater in patients that have the genes, and is further increased by homozygosity4. A number of possi-ble auto-antigens have been identified including citrullinated proteins, heavy chain binding proteins, human cartilage glycoprotein 39 (GP39), heat shock proteins and type II collagen5. In experimental arthritis models, T cell-dependence has also been demonstrated. Animal models of arthritis, such as collagen-induced arthritis (CIA) or adjuvant arthritis, are clearly T cell dependent and transfer of CD4+ T cells from sick animals into healthy recipients is enough to trigger the development of disease,

12

while CD4 depletion is able to diminish inflammation6;7. T cells are also important for the production of auto-antibodies, as they provide help for B cell activation. Li-gation of CD154 (CD40L) on activated T cells with CD40 on B cells stimulates the last to proliferate, produce antibodies and switch isotype8.Recently, Th17 cells characterized by the production of the highly inflammatory IL-17 cytokine have been implicated in RA. Several studies have described the presence of high levels of IL-17 in RA synovial fluid (SF)9;10 and the spontaneous production of IL-17 by RA synovial tissue (ST) T cells11;12. Also, increased numbers of Th17 cells in RA SF and ST have been found when compared to healthy donor and disease con-trols11;12. However, different studies have shown the predominate presence of Th1, rather than Th17, cells in inflamed joints, raising the possibility that other cells in the synovium are responsible for local IL-17 production13;14.The hypothesis that RA synovial T cells have defective TCR signaling cascades is still controversial. Synovial T cells are highly differentiated CD45RO+ T cells, and pre-sent markers of recent activation such as CD69 and HLA-DR15;16. However, chronic exposure to cytokines and/or oxidative stress has been proposed to downregulate TCRzeta expression17-19 and displace linker for activation of T cells (LAT) from the plasma membrane20. In this case, T cells would no longer be able to respond to TCR ligation and would contribute to inflammation by TCR-independent mechanisms.In normal immune responses, regulatory T cells (Tregs) are important for the suppres-sion of abnormal and excessive inflammation. However in RA, studies have showed that defective Treg function and/or resistance of effector RA T cells to Treg-mediated suppression may contribute to the persistence of chronic inflammation21;22.Besides specific antigen-driven activation, T cells may also contribute to synovial in-flammation via cell-contact interactions with neighboring macrophages, FLS and B cells, thereby promoting their activation. Activated T cells are able to induce macro-phage TNF-α, IL-1β and MMP production via cell-cell contacts23-26 and interaction with FLS induces IL-6, IL-8 and MMP-1 production27;28.

B lymphocytes

The first clue that B cells may play an important role in autoimmunity came from the identification of antibodies specific for IgG in the blood of RA patients, known as rheumatoid factors (RF). Up to 75% of RA patients are seropositive for RF, and its presence predicts a more aggressive and destructive course of disease29. Anti-citru-linated protein antibodies (ACPA) have also been identified, and these can be detec-

Chapter 1

13

ted years before clinically onset of inflammatory arthritis. ACPA are directed against epitopes in which the L-arginine amino acid has been postranslationally modified to L-citruline30. While RF can also be detected in other autoimmune diseases, systemic infections and even 5% of healthy individuals, the presence of ACPA is very specific to RA, so its detection is an important diagnostic tool in RA29.In the synovium, lymphoid neogenesis can be found in one third of RA patients31. Up to 20% of these patients display lymphoid aggregates with germinal centers (GC)32. Although the presence of synovial lymphoid neogenesis is associated with the degree of synovial inflammation, it does not support antigen-driven clonal B cell expansion or antibody affinity maturation33;34. Additionally, no relation is found between synovial lymphoid neogenesis and the levels of RF or ACPA, so the rele-vance of these organized structures in the synovium to the severity of clinical signs and symptoms is uncertain. Besides auto-antibody producers, it is thought that B cells in the joints play an im-portant role in antigen presentation, as B cell depletion prevents the formation of ectopic germinal centers and impairs T cell activation35. It is therefore not surprising that B cell directed therapies, such as the anti-CD20 monoclonal antibody Rituxi-mab, has shown promising results. This antibody depletes all B cell subsets, except plasma cells, which lack CD20 expression. Besides a possible role in diminishing antigen presentation, Rituximab may exert its effects by an indirect effect on anti-body producing plasma cells as well36. Since B cells are precursors of plasma cells, B cell depletion may indirectly result in a decrease in numbers of the short-lived plasma cells. In agreement, treatment with anti-CD20 monoclonal antibody reduces the levels of serum RF and ACPA, inducing significant clinical benefit in patients with RA36;37.

Macrophages

Macrophages are important players in joint inflammation. They are present in high numbers in the inflamed synovium and display signs of cellular activation, such as expression of MHC class II molecules. They are also important sources of pro-inflammatory cytokines, including TNF-α and IL-1β, as well as chemokines and MMPs38. Moreover, a positive correlation has been found between the degree of synovial macrophage infiltration, TNF-α expression and clinical features, such as radiological progression of joint destruction39;40.In the synovium, macrophages are important components in the recruitment and

Introduction

14

activation of inflammatory cells. They produce chemokines that will attract other cells to the joint such as T and B cells, neutrophils and other macrophages, and se-crete cytokines that will activate these newly arrived cells41. Additionally, cellular contacts between macrophages and T cells, FLS, and endothelial cells, constitute an important component of macrophage effector responses, even in the absence of anti-gen. Cell-cell contacts between macrophages and FLS elicits the production of IL-6, granulocyte macrophage colony-stimulating factor (GM-CSF) and IL-842, while in-teraction with activated T cells can induce TNF-α, IL-1β and MMP production23-26.Importantly, ST macrophages are used as biomarkers of clinical response in RA cli-nical trials, as their numbers decrease in response to successful anti-rheumatic tre-atment43;44.

Fibroblast-like Synoviocytes

In normal synovium, fibroblast-like synoviocytes (FLS) are mesenchymal cells that produce extracellular matrix and secrete hyaluronan and lubricin, key components of synovial fluid and important for joint lubrication45. However in RA, FLS in the synovial lining layer display numerous features of cellular activation that ultima-tely contribute to their aggressive and invasive behavior. In the lining, layer FLS increase in number, thickening the synovial lining layer into a hyperplastic tissue. In vitro, they grow in an anchorage-independent manner and lack contact inhibition46. Activated FLS contribute to the degradation of extracellular matrix. They attach to cartilage and release matrix degrading enzymes, particularly MMPs, allowing them to deeply invade the extracellular matrix45. Both MMP-1 and MMP-3 are elevated in the synovial fluid and serum of RA patients, and their production has been mainly attributed to FLS47;48. In vitro studies have showed the capacity of RA FLS to divide more rapidly than cells from normal or osteoarthritic joints49. Different mechanisms have been pro-posed to explain the hyperplastic growth of FLS in RA synovium. Exposure to cy-tokines and growth factors in the synovium are thought to play a crucial role in this process. Furthermore, FLS have a high expression of transcription factors and molecules that regulate cell cycle, which might be involved in their increased proli-feration. Proto-oncogene products of ras, myc, and others can be found abundantly in FLS, especially at sites of invasion into cartilage and bone50. Another mechanism that may underlie synovial hyperplasia is a decline of cells undergoing apoptosis. In this context, high levels of phosphorylated protein kinase B (PKB, also known

Chapter 1

15

as Akt), are found in RA FLS51. PKB is an important protein in the phosphatidy-linositol 3-kinase (PI3K) signaling cascade, which in turn regulates processes such as cell growth, differentiation, survival and proliferation52. The fact that PKB levels can be further increased by TNF-α stimulation, together with the findings that RA synovium lacks expression of the tumor suppressor PTEN (PI3K inhibitor) at sites of invasive FLS growth53, might explain in part the impaired apoptosis associated with the proliferating synovium in RA.

Clearly, there are many pieces of evidence in RA pointing to a de-regulation of in-tracellular signaling pathways involved in inflammation, cell proliferation and sur-vival. Targeting these pathways and restoring normal cellular behavior seems there-fore an elegant and promising manner of reducing cellular inflammation and joint destruction. To accomplish this, a detailed examination of how signaling pathways might be de-regulated in RA is necessary.

Small GTPases: Key regulators of cellular functions

The Ras superfamily of small GTPases constitutes a large group of structurally and functionally related proteins. They are important components of signal transduction pathways used by antigen receptors, costimulatory, cytokine and chemokine recep-tors to regulate the immune response. Small GTPases control fundamental biologi-cal processes including cell division, differentiation, shape changes, and survival. The Ras superfamily of proteins is divided into five major subfamilies on the basis of their sequence and functional similarities: Ras, Rho, Rab, Ran and Arf54.Despite a high level of homology, members of the Ras GTPase superfamily display major differences in their signaling specificity. These functional differences can be explained in part by their different cellular localization, which is mainly determined by the hypervariable domain at their C-termini and specific effector domains, which bind to and regulate downstream signaling proteins55;56.

Regulation of GTPase activation

Small GTPases are highly conserved throughout all eukaryotes, and their activity is regulated by common biochemical mechanisms. All Ras-related proteins cycle between an inactive guanosine diphosphate (GDP) and an active guanosine trip-

Introduction

16

Chapter 1

hosphate (GTP) bound form, operating as binary switches that control cell activation in response to environmental cues57. GTPases adopt different conformations when binding GTP vs. GDP. The active GTP-bound conformation allows GTPases to inter-act with downstream effectors and thereby initiate downstream signaling pathways, which regulate many important biological processes. Two main classes of regula-tory proteins control this cycle: guanine nucleotide exchange factors (GEFs) promote the exchange of GDP for GTP, and GTPase-activating proteins (GAPs) stimulate the otherwise slow intrinsic GTPase activity, promoting the formation of the inactive GDP-bound configuration. Canonical mutations that affect the GTPase cycle lead to constitutively- active or dominant-negative molecules58. Mutations that abolish GT-Pase activity (e.g., glycine 12 to valine; G12V), such as RasV12 or RapV12, result in constitutive activation of the small GTPase, and mutations that affect interaction of the GTPase with its GEFs and effectors (e.g., threonine 17 to asparagine; T17N) result in dominant-negative molecules. Both activated and dominant- negative GTPases can dominantly perturb cellular processes in which the GTPases are involved.

The Ras subfamily of small GTPases

Members of the subfamily of Ras GTPases include the Ras proteins (H-, K-, and N-Ras), R-Ras, M-Ras, Rap, Ral and Rheb proteins. These have been recognized for their involvement in signal transduction cascades that regulate cell growth, prolife-ration, differentiation and survival, primarily through the modulation of gene ex-pression59.

Ras GTPases

The Ras proteins are ubiquitously expressed 21-kDa proteins. The three homolo-gues of Ras, H-Ras, K-Ras, and N-Ras, share a high degree of sequence homology (>85%), especially in the effector domain that directly couples them to downstream signaling proteins60. As all members of the Ras GTPase family, these proteins are regulated by GEFs and GAPs. The selectivity of GEFs in activating distinct Ras ho-mologues, as well as the differential coupling of activating GEFs, such as Son-of-sevenless (Sos) and Ras guanine nucleotide-releasing factor (RasGRF) 1 to tyrosine kinase-dependent and G protein-coupled receptors, respectively, contributes to the signaling specificity of each Ras homologue. All three homologues are important

17

Introduction

in activating intracellular downstream pathways such as mitogen-activated protein kinase (MAPK) cascades, PI3K and Ral family GTPases. However, genetic and de-tailed cell biology studies were able to demonstrate that differential subcellular lo-calization of each Ras homologue confers to them distinct signaling properties61;62. Biochemical studies have demonstrated the different specificity of the three Ras ho-mologues in activating downstream signaling pathways. While K-Ras was shown to be a more potent activator of Raf-1 and the downstream MAPK cascade than H-Ras, H-Ras seemed to be more efficient in activating PI3K63. Furthermore, targeting expe-riments have showed that whereas h-ras and n-ras single or double knock-out mice are completely viable, targeted disruption of the k-ras gene leads to lethality64-66. In addition, different ras genes are found mutated in different types of tumors.

Ras GTPase effectors and signaling

In the active GTP-bound conformation, Ras GTPases can activate a large panel of downstream effectors in response to diverse extracellular stimuli. Members of the Raf family, the PI3K and members of a family of exchange factors for the small GT-Pase Ral, e.g., RalGDS, have been established as Ras effectors. One of the best studied effectors of Ras is the serine-threonine kinase Raf. Raf is involved in a signaling pathway where activation of the MAPK cascade culminates in the regulation of cell proliferation. In this signaling pathway, active Ras binds to and promotes the translocation of Raf to the plasma membrane, where additi-onal phosphorylation events promote full Raf kinase activation67. Once active, Raf phosphorylates and activates MEK (MAPK/Erk kinase), a dual specificity tyrosine-threonine kinase, which in turn phosphorylates and activates the Erk1/2 MAPK. Activated ERK translocates to the nucleus, where it phosphorylates transcription factors involved in cell proliferation and differentiation59. In this way, signals arising from an extracellular growth factor or cytokine are transmitted from the cell surface to the nucleus, ultimately changing the activity of nuclear transcription factors.

Ras GTPases and tumors

Ras proteins have been the subject of intense research, partly because of their critical roles in human oncogenesis. A high frequency of ras mutations has been detected in a variety of tumors. The commonly occurring mutations (at codons 12, 13 and 61)

18

make the GTPase insensitive to the action of GAPs and thereby lock it in the GTP-bound, active state68. The highest incidence of ras mutations is found in pancreatic adenocarcinomas where almost 90% of the tumors are associated with a mutation in kras. Also, in 50% of colon, lung and thyroid tumors, mutations in ras genes have been observed68.

Ras GTPases and rheumatoid arthritis

In RA synovial tissue, abundant expression of Ras proteins can be found, predo-minantly in synovial lining cells attached to cartilage and bone at the site of joint destruction50. In systemic lupus erythematosus (SLE) T cells, de-regulated Ras ex-pression has also been observed. In a subset of patients, Ras expression and function is reduced, while mice defficient for the Ras GEF Ras guanine nucleotide-releasing protein 1 (Ras GRP1) develop a spontaneous SLE-like disease69-71. In arthritic synovium, point mutations in h-ras were initially described72. However, when assessing higher number of samples in later studies, no activating mutations could be found73. However, the relevance of Ras signaling pathways in synovial in-flammation has been underscored by several studies. Activation of Ras effector pa-thways, including MAPK, PI3K, and nuclear factor-kappa B (NF-kB), is enhanced in RA compared to disease controls51;74;75. In vitro, over-expression of dominant-nega-tive Ras has been shown to suppress RA FLS proliferation, IL-6 production and IL-1-induced ERK activation76. Moreover, RA FLS stably expressing a dominant negative version of c-Raf, which can bind to and interfere with signaling of Ras GTPases, have reduced MMP-1 and MMP-3 production77. In these cells c-Raf inhibition also decreases ERK and JNK phosphorylation as well as FLS invasiveness. Importantly, inhibition of Ras family function in vivo has been shown to be protective in experi-mental arthritis models76-78.

Rap1 GTPase

In mammals there are two isoforms of Rap1, Rap1a and Rap1b, which are encoded by distinct genes but share 95% amino acid identity79. In T cells, Rap1 is transiently activated upon TCR ligation80, and like other GTP-binding proteins its activation is dependent on the action of GEFs. C3G, PDZ-GEF, exchange protein directly activa-ted by cyclic AMP (EPAC) and calcium diacylglycerol regulated guanine nucleotide

Chapter 1

19

exchange factor (CalDAG-GEF, RasGRP) are some of the GEFs identified for Rap179. Two groups of GAPs regulate inactivation of Rap1, the RapGAP and the Spa1 fa-milies81. Members of the RapGAP family include RapGAP1A, RapGAP1B and Rap-GAP2. The Spa1 family of GAPs, consists of Spa1 and E6TP1.

Rap1 and the regulation of cell adhesion

One of the best characterized functions of Rap1 is the regulation of cell adhesion. Integrin activation82, cadherin-mediated adhesion83 and cell-cell junction formation84

are adhesion processes regulated by Rap1. Several studies have investigated the mechanisms by which Rap1 controls integrin activation. In its active state, bound to GTP, Rap1 can associate with RAPL (regulator of adhesion and cell polarization enriched in lymphoid tissues)85, which in turn, associates with and activates the se-rine-threonine kinase Mst1 (mammalian sterile twenty-like-1)86. This allows the spa-tial distribution of LFA1 to the leading edge, or to the immunological synapse, and integrin clustering85. Other Rap1 effectors include PKD1 (protein kinase D1)87 and RIAM (Rap1-interacting adaptor molecule)88. Association of active Rap1 with these effectors has also been demonstrated to induce cellular adhesion. In vitro studies have demonstrated that afadin (AF-6) can also act downstream of Rap1 activation, inhibiting endocytosis of E-cadherin, and allowing the maintenance of cellular junc-tions89. Many studies have tried to unravel the mechanism of regulation of integrins by Rap1, but it is not clear yet whether RAPL, PDK and RIAM control separate pathways that are required for Rap1-induced integrin activation, or whether they participate in the same pathway that leads to Mst1 function, and ultimately to inte-grin clustering.Rap1 regulation of integrin activation has important consequences in the immune system, as effective T cell responses are critically dependent on appropriate T cell-antigen presenting cell (APC) interactions. The formation of a stable immunological synapse requires proper integrin activation. Upon TCR triggering Rap1 becomes rapidly activated and induces conformational changes in integrin structure, which increases their avidity (clustering) as well as ligand affinity, thereby potentiating T cell-APC interactions90;91. Ligation of chemokine receptors is also able to activate Rap1, which in turn enhances integrin function. This allows T cell trafficking and migration to lymphoid organs and sites of inflammation91;92.

Introduction

20

Rap1 and the regulation of ROS production

In T lymphocytes, Rap1 activation is able to suppress Ras-dependent reactive oxy-gen species (ROS) production. ROS are proposed to act as important second mes-sengers in T cell activation93. TCR triggering results in transient ROS production, and scavenging intracellular ROS with antioxidants has been shown to suppress TCR-induced NF-kB, AP-1 and IL-2 promoter transcription94. On the other hand, chronic oxidative stress can lead to constitutive activation of NF-kB-dependent in-flammatory gene products95. The intracellular production of ROS involves the activation of the small GTPase Ras and its downstream target Ral96. Simultaneous activation of Rap1 is able to attenuate ROS production, distally from Ral, and in a PI3K-dependent manner. When both Ras and Rap1 are transiently activated, limited ROS production is used as a second messenger, optimizing Ras-dependent activation of ERKs and transcription factors.

Rap1 and rheumatoid arthritis

In a number of human autoimmune diseases, including RA97, multiple sclerosis (MS)98 and SLE99, chronic oxidative stress triggered by ROS is thought to underlie the pathogenic T cell behavior. In RA, destructive proliferative synovitis has been related to oxidative stress100;101. In SF T lymphocytes, chronic oxidative stress derived from increased intracellular ROS production may induce constitutive activation of NF-kB-dependent gene transcription. This results in the upregulation of pro-inflam-matory cytokines, which will contribute to the perpetuation of synovial joint inflam-mation102.In T cells from the SF of RA patients, a high rate of endogenous ROS production cor-relates with a constitutive activation of Ras and an inhibition of Rap1 activation. In vitro experiments have showed that a restored redox balance could be achieved in RA SF T cells by introduction of a dominant-negative form of Ras96, indicating that deregulated Ras and Rap1 signaling underlies the chronic oxidative stress observed in RA SF T cells.

Function of Rap1 in vivo

Genetic manipulation of Rap1 signaling has been of great importance to under-

Chapter 1

21

stand the role of Rap1 in the regulation of the immune system. As mentioned be-fore, Rap1 is an important mediator of integrin activation, and this is confirmed in Rap1 knockout mice, where T cell polarization and integrin-dependent adhesion is impaired103;104. Previous in vitro studies have also suggested that Rap1 regulates po-sitive and negative thymocyte selection105 and in agreement, transgenic expression of Spa1 renders animals with a defect in α/β thymocyte development at the double negative stage106.An accumulating amount of evidence suggests Rap1 as a critical mediator of T cell responses. RapGAP1 transgenic mice display an age-dependent accumulation of activated T cells107. On the contrary, mice lacking Spa1 expression exhibit age-de-pendent defects in T cell responses108. B cell responses are also diminished in these animals due to reduced T helper cell function. Similarly, transgenic mice expressing the active Rap1E63 mutant show decreases in T helper and effector cell functions, although LFA-mediated adhesion is increased109. In these mice the defective respon-ses correlate with increased numbers and function of CD4+CD103+ regulatory T cells. Finally, RapV12 transgenic mice, expressing constitutive active Rap1 in the T cell compartment, show that increased Rap1 -dependent adhesion can enhance T cell function in conditions where TCR-MHC interactions are of low affinity110.

The Rho subfamily of small GTPases

Like Ras, Ras homologous (Rho) proteins also serve as key regulators of extracellu-lar-stimulus-mediated signaling networks that regulate actin organization, cell cycle progression and gene expression111. Most studies on this family have been focused on three of the 22 mammalian Rho GTPases: Rac1, RhoA and Cdc42. Rho GTPases have been shown to control cellular motility and polarity in migrating cells by regulating actin and myosin organization. In this context, RhoA was shown to promote actin stress fiber formation and focal adhesion assembly, Cdc42 actin mi-crospikes and filopodium formation and Rac1 lamellipodium formation and mem-brane ruffling111. In this thesis we will focus on one of the Rho GTPases: Rac1.

Rac1 GTPase

Rac1 is a key protein in the regulation of cell migration, as well as in the adhesion of cells to the underlying protein matrix, or to other cells. It is ubiquitously expressed

Introduction

22

and found activated at the leading edge of migrating cells112. Rac1 is activated by the action of distinct GEFs including Vav-1, Tiam1 and β-Pix113-115. Once bound to GTP, Rac1 can interact with target effector proteins, such as Pak and PI3K, and regulate several signaling pathways including JNK, p38, NF-κB and PKB, in different cell types116-120. Genetic studies in mice have showed that rac1 deletion provokes embryonic letha-lity121, so conditional gene disruption has been used to unravel the role of Rac1 in dif-ferent cell types. Rac1 is important for the optimal reconstitution of the hematopoie-tic system, having roles both in the engraftment and retention of hematopoietic stem cells (HSCs) in the bone marrow. This GTPase is essential for the entry of HSCs into cell cycle upon extracellular stimulation, as well as for their progression through S and G2/M phases122. In neutrophils, deficiency for Rac1 makes them defective in in-flammatory recruitment in vivo, migration to chemotactic stimuli, and chemoattrac-tant-mediated actin assembly123. Upon monocyte activation, Rac1 is used to assemble the activated NADPH oxidase complex124, and its deficiency in macrophages leads to defects in cell spreading and membrane ruffling125. Rac1 activation is also of great importance in the formation of immunological synapses. Dendritic cells lacking ex-pression of both Rac1 and Rac2 show defective cytoskeletal re-arrangements, migra-tion and antigen presentation that, as a result, prevent adequate T cell priming126. Finally in B cells, Rac1, together with Rac2, plays an important role in BCR-induced activation, transducing BCR signals that control survival and cell cycle entry127. Many of the pathways regulated by Rac1 seem to be involved in the inflammatory process that occurs in RA. In agreement, in vitro inhibition of Rac1 signaling is able to reduce FLS proliferation, invasiveness and JNK activation. In the same way, Rac1 has been found to regulate both in vitro and in vivo osteoclastogenesis and bone resorption128;129. Outline of this thesis

In this thesis we analyse the involvement of distinct small GTPases in RA, and in-vestigate the consequences of their activation or inhibition in vitro and in vivo, in a murine model for arthritis.T cells from the synovial fluid (SF) of RA patients are believed to behave in a hypo-responsive manner and signaling abnormalities that impair T cell activation have been extensively described. In chapter 2 we perform single-cell analysis of these T cells to evaluate where de-regulated small GTPase function might impair TCR-dependent responses.

Chapter 1

23

Previous studies in RA SF T cells have demonstrated a constitutively block in Rap1 activation. In chapter 3 we explore the consequences of Rap1 activation in vivo in a collagen-induced arthritis model. We induce arthritis in transgenic mice expressing a constitutively active form of Rap1 within the T cell lineage, and assess susceptibi-lity and severity of arthritis.Chapter 4 describes how the expression of inactivators of Rap1, the RapGAPs, is regulated upon T cell activation.In chapter 5 we examine the expression of an important Ras GEF, RasGRF1, in RA synovial tissue, and evaluate the influence of RasGRF1 expression and function on RA FLS and synovial tissue MMP and pro-inflammatory cytokine production.In chapter 6 we investigate the in vivo effects of blocking Rac1 signaling in arthritis.

Introduction

24

Chapter 1

References

Sangha O. Epidemiology of rheumatic diseases. Rheumatology.(Oxford) 2000;39 1. Suppl 2:3-12.Tak PP, Bresnihan B. The pathogenesis and prevention of joint damage in rheu-2. matoid arthritis: advances from synovial biopsy and tissue analysis. Arthritis Rheum. 2000;43:2619-2633.Gregersen PK, Silver J, Winchester RJ. The shared epitope hypothesis. An ap-3. proach to understanding the molecular genetics of susceptibility to rheumatoid arthritis. Arthritis Rheum. 1987;30:1205-1213.Weyand CM, Hicok KC, Conn DL, Goronzy JJ. The influence of HLA-DRB1 ge-4. nes on disease severity in rheumatoid arthritis. Ann.Intern.Med. 1992;117:801-806.Blass S, Engel JM, Burmester GR. The immunologic homunculus in rheumatoid 5. arthritis. Arthritis Rheum. 1999;42:2499-2506.Banerjee S, Webber C, Poole AR. The induction of arthritis in mice by the car-6. tilage proteoglycan aggrecan: roles of CD4+ and CD8+ T cells. Cell Immunol 1992;144:347-357.Klimiuk PA, Yang H, Goronzy JJ, Weyand CM. Production of cytokines and 7. metalloproteinases in rheumatoid synovitis is T cell dependent. Clin.Immunol. 1999;90:65-78.Lee BO, Moyron-Quiroz J, Rangel-Moreno J et al. CD40, but not CD154, expres-8. sion on B cells is necessary for optimal primary B cell responses. J.Immunol. 2003;171:5707-5717.Ziolkowska M, Koc A, Luszczykiewicz G et al. High levels of IL-17 in rheuma-9. toid arthritis patients: IL-15 triggers in vitro IL-17 production via cyclosporin A-sensitive mechanism. J.Immunol. 2000;164:2832-2838.Kotake S, Udagawa N, Takahashi N et al. IL-17 in synovial fluids from patients 10. with rheumatoid arthritis is a potent stimulator of osteoclastogenesis. J.Clin.In-vest 1999;103:1345-1352.Shahrara S, Huang Q, Mandelin AM, Pope RM. TH-17 cells in rheumatoid 11. arthritis. Arthritis Res.Ther. 2008;10:R93.Chabaud M, Durand JM, Buchs N et al. Human interleukin-17: A T cell-derived 12. proinflammatory cytokine produced by the rheumatoid synovium. Arthritis Rheum. 1999;42:963-970.Raza K, Falciani F, Curnow SJ et al. Early rheumatoid arthritis is characterized 13. by a distinct and transient synovial fluid cytokine profile of T cell and stromal

25

Introduction

cell origin. Arthritis Res.Ther. 2005;7:R784-R795.Yamada H, Nakashima Y, Okazaki K et al. Th1 but not Th17 cells predominate in 14. the joints of patients with rheumatoid arthritis. Ann.Rheum.Dis. 2008;67:1299-1304.Afeltra A, Galeazzi M, Ferri GM et al. Expression of CD69 antigen on synovial 15. fluid T cells in patients with rheumatoid arthritis and other chronic synovitis. Ann.Rheum.Dis. 1993;52:457-460.Thomas R, McIlraith M, Davis LS, Lipsky PE. Rheumatoid synovium is enriched 16. in CD45RBdim mature memory T cells that are potent helpers for B cell differen-tiation. Arthritis Rheum. 1992;35:1455-1465.Cope AP, Londei M, Chu NR et al. Chronic exposure to tumor necrosis factor 17. (TNF) in vitro impairs the activation of T cells through the T cell receptor/CD3 complex; reversal in vivo by anti-TNF antibodies in patients with rheumatoid arthritis. J.Clin.Invest 1994;94:749-760.Isomaki P, Panesar M, Annenkov A et al. Prolonged exposure of T cells to TNF 18. down-regulates TCR zeta and expression of the TCR/CD3 complex at the cell surface. J.Immunol. 2001;166:5495-5507.Maurice MM, Lankester AC, Bezemer AC et al. Defective TCR-mediated signa-19. ling in synovial T cells in rheumatoid arthritis. J.Immunol. 1997;159:2973-2978.Gringhuis SI, Leow A, Papendrecht-van der Voort EA et al. Displacement of lin-20. ker for activation of T cells from the plasma membrane due to redox balance alterations results in hyporesponsiveness of synovial fluid T lymphocytes in rheumatoid arthritis. J.Immunol. 2000;164:2170-2179.Ehrenstein MR, Evans JG, Singh A et al. Compromised function of regulatory T 21. cells in rheumatoid arthritis and reversal by anti-TNFalpha therapy. J.Exp.Med. 2004;200:277-285.van Amelsfort JM, Jacobs KM, Bijlsma JW, Lafeber FP, Taams LS. CD4(+)CD25(+) 22. regulatory T cells in rheumatoid arthritis: differences in the presence, pheno-type, and function between peripheral blood and synovial fluid. Arthritis Rhe-um. 2004;50:2775-2785.Isler P, Vey E, Zhang JH, Dayer JM. Cell surface glycoproteins expressed on acti-23. vated human T cells induce production of interleukin-1 beta by monocytic cells: a possible role of CD69. Eur.Cytokine Netw. 1993;4:15-23.Lacraz S, Isler P, Vey E, Welgus HG, Dayer JM. Direct contact between T lymp-24. hocytes and monocytes is a major pathway for induction of metalloproteinase expression. J.Biol.Chem. 1994;269:22027-22033.McInnes IB, Leung BP, Sturrock RD, Field M, Liew FY. Interleukin-15 mediates 25.

26

Chapter 1

T cell-dependent regulation of tumor necrosis factor-alpha production in rheu-matoid arthritis. Nat.Med. 1997;3:189-195.Sebbag M, Parry SL, Brennan FM, Feldmann M. Cytokine stimulation of T lymp-26. hocytes regulates their capacity to induce monocyte production of tumor necro-sis factor-alpha, but not interleukin-10: possible relevance to pathophysiology of rheumatoid arthritis. Eur.J.Immunol. 1997;27:624-632.Burger D, Rezzonico R, Li JM et al. Imbalance between interstitial collagenase 27. and tissue inhibitor of metalloproteinases 1 in synoviocytes and fibroblasts upon direct contact with stimulated T lymphocytes: involvement of membrane-associated cytokines. Arthritis Rheum. 1998;41:1748-1759.Yamamura Y, Gupta R, Morita Y et al. Effector function of resting T cells: activa-28. tion of synovial fibroblasts. J.Immunol. 2001;166:2270-2275.De RL, Peene I, Hoffman IE et al. Rheumatoid factor and anticitrullinated pro-29. tein antibodies in rheumatoid arthritis: diagnostic value, associations with ra-diological progression rate, and extra-articular manifestations. Ann.Rheum.Dis. 2004;63:1587-1593.Schellekens GA, de Jong BA, van den Hoogen FH, van de Putte LB, van Venrooij 30. WJ. Citrulline is an essential constituent of antigenic determinants recognized by rheumatoid arthritis-specific autoantibodies. J.Clin.Invest 1998;101:273-281.Klimiuk PA, Goronzy JJ, Bjor nJ, Beckenbaugh RD, Weyand CM. Tissue cy-31. tokine patterns distinguish variants of rheumatoid synovitis. Am.J.Pathol. 1997;151:1311-1319.Wagner UG, Kurtin PJ, Wahner A et al. The role of CD8+ CD40L+ T cells in the for-32. mation of germinal centers in rheumatoid synovitis. J.Immunol. 1998;161:6390-6397.Cantaert T, Kolln J, Timmer T et al. B lymphocyte autoimmunity in rheuma-33. toid synovitis is independent of ectopic lymphoid neogenesis. J.Immunol. 2008;181:785-794.Thurlings RM, Wijbrandts CA, Mebius RE et al. Synovial lymphoid neogene-34. sis does not define a specific clinical rheumatoid arthritis phenotype. Arthritis Rheum. 2008;58:1582-1589.Takemura S, Klimiuk PA, Braun A, Goronzy JJ, Weyand CM. T cell activation in 35. rheumatoid synovium is B cell dependent. J.Immunol. 2001;167:4710-4718.Thurlings RM, Vos K, Wijbrandts CA et al. Synovial tissue response to rituxi-36. mab: mechanism of action and identification of biomarkers of response. Ann.Rheum.Dis. 2008;67:917-925.Edwards JC, Cambridge G. B-cell targeting in rheumatoid arthritis and other 37.

27

Introduction

autoimmune diseases. Nat.Rev.Immunol. 2006;6:394-403.Hamilton JA, Tak PP. The dynamics of macrophage lineage populations in in-38. flammatory and autoimmune diseases. Arthritis Rheum. 2009;60:1210-1221.Mulherin D, Fitzgerald O, Bresnihan B. Synovial tissue macrophage populati-39. ons and articular damage in rheumatoid arthritis. Arthritis Rheum. 1996;39:115-124.Tak PP, Smeets TJ, Daha MR et al. Analysis of the synovial cell infiltrate in early 40. rheumatoid synovial tissue in relation to local disease activity. Arthritis Rheum. 1997;40:217-225.Kinne RW, Stuhlmuller B, Burmester GR. Cells of the synovium in rheumatoid 41. arthritis. Macrophages. Arthritis Res.Ther. 2007;9:224.Chomarat P, Rissoan MC, Pin JJ, Banchereau J, Miossec P. Contribution of IL-1, 42. CD14, and CD13 in the increased IL-6 production induced by in vitro monocyte-synoviocyte interactions. J.Immunol. 1995;155:3645-3652.Haringman JJ, Gerlag DM, Zwinderman AH et al. Synovial tissue macropha-43. ges: a sensitive biomarker for response to treatment in patients with rheumatoid arthritis. Ann.Rheum.Dis. 2005;64:834-838.Bresnihan B, Gerlag DM, Rooney T et al. Synovial macrophages as a biomarker 44. of response to therapeutic intervention in rheumatoid arthritis: standardization and consistency across centers. J.Rheumatol. 2007;34:620-622.Muller-Ladner U, Ospelt C, Gay S, Distler O, Pap T. Cells of the synovium in 45. rheumatoid arthritis. Synovial fibroblasts. Arthritis Res.Ther. 2007;9:223.Lafyatis R, Remmers EF, Roberts AB et al. Anchorage-independent growth of sy-46. noviocytes from arthritic and normal joints. Stimulation by exogenous platelet-derived growth factor and inhibition by transforming growth factor-beta and retinoids. J.Clin.Invest 1989;83:1267-1276.Klimiuk PA, Sierakowski S, Latosiewicz R et al. Serum matrix metalloproteina-47. ses and tissue inhibitors of metalloproteinases in different histological variants of rheumatoid synovitis. Rheumatology.(Oxford) 2002;41:78-87.Walakovits LA, Moore VL, Bhardwaj N, Gallick GS, Lark MW. Detection of 48. stromelysin and collagenase in synovial fluid from patients with rheumatoid arthritis and posttraumatic knee injury. Arthritis Rheum. 1992;35:35-42.Mohr W, Beneke G, Mohing W. Proliferation of synovial lining cells and fibro-49. blasts. Ann.Rheum.Dis. 1975;34:219-224.Trabandt A, Aicher WK, Gay RE et al. Expression of the collagenolytic and Ras-50. induced cysteine proteinase cathepsin L and proliferation-associated oncogenes in synovial cells of MRL/I mice and patients with rheumatoid arthritis. Matrix

28

Chapter 1

1990;10:349-361.Zhang HG, Wang Y, Xie JF et al. Regulation of tumor necrosis factor alpha-medi-51. ated apoptosis of rheumatoid arthritis synovial fibroblasts by the protein kinase Akt. Arthritis Rheum. 2001;44:1555-1567.Okkenhaug K, Vanhaesebroeck B. PI3K in lymphocyte development, differenti-52. ation and activation. Nat.Rev.Immunol. 2003;3:317-330.Pap T, Franz JK, Hummel KM et al. Activation of synovial fibroblasts in rheu-53. matoid arthritis: lack of Expression of the tumour suppressor PTEN at sites of invasive growth and destruction. Arthritis Res. 2000;2:59-64.Colicelli J. Human RAS superfamily proteins and related GTPases. Sci.STKE. 54. 2004;2004:RE13.Hancock JF, Paterson H, Marshall CJ. A polybasic domain or palmitoylation is 55. required in addition to the CAAX motif to localize p21ras to the plasma mem-brane. Cell 1990;63:133-139.Omerovic J, Prior IA. Compartmentalized signalling: Ras proteins and signal-56. ling nanoclusters. FEBS J. 2009;276:1817-1825.Bos JL, Rehmann H, Wittinghofer A. GEFs and GAPs: critical elements in the 57. control of small G proteins. Cell 2007;129:865-877.Bourne HR, Sanders DA, McCormick F. The GTPase superfamily: conserved 58. structure and molecular mechanism. Nature 1991;349:117-127.Mor A, Philips MR. Compartmentalized Ras/MAPK signaling. Annu.Rev.Im-59. munol. 2006;24:771-800.Omerovic J, Prior IA. Compartmentalized signalling: Ras proteins and signal-60. ling nanoclusters. FEBS J. 2009;276:1817-1825.Castellano E, De Las RJ, Guerrero C, Santos E. Transcriptional networks of 61. knockout cell lines identify functional specificities of H-Ras and N-Ras: sig-nificant involvement of N-Ras in biotic and defense responses. Oncogene 2007;26:917-933.Hancock JF. Ras proteins: different signals from different locations. Nat.Rev.62. Mol.Cell Biol 2003;4:373-384.Yan J, Roy S, Apolloni A, Lane A, Hancock JF. Ras isoforms vary in their ability 63. to activate Raf-1 and phosphoinositide 3-kinase. J.Biol Chem. 1998;273:24052-24056.Esteban LM, Vicario-Abejon C, Fernandez-Salguero P et al. Targeted geno-64. mic disruption of H-ras and N-ras, individually or in combination, reveals the dispensability of both loci for mouse growth and development. Mol.Cell Biol 2001;21:1444-1452.

29

Introduction

Johnson L, Greenbaum D, Cichowski K et al. K-ras is an essential gene in the 65. mouse with partial functional overlap with N-ras. Genes Dev. 1997;11:2468-2481.Koera K, Nakamura K, Nakao K et al. K-ras is essential for the development of 66. the mouse embryo. Oncogene 1997;15:1151-1159.Morrison DK, Cutler RE. The complexity of Raf-1 regulation. Curr.Opin.Cell 67. Biol. 1997;9:174-179.Minamoto T, Mai M, Ronai Z. K-ras mutation: early detection in molecular di-68. agnosis and risk assessment of colorectal, pancreas, and lung cancers--a review. Cancer Detect.Prev. 2000;24:1-12.Layer K, Lin G, Nencioni A et al. Autoimmunity as the consequence of a sponta-69. neous mutation in Rasgrp1. Immunity. 2003;19:243-255.Cedeno S, Cifarelli DF, Blasini AM et al. Defective activity of ERK-1 and ERK-2 70. mitogen-activated protein kinases in peripheral blood T lymphocytes from pa-tients with systemic lupus erythematosus: potential role of altered coupling of Ras guanine nucleotide exchange factor hSos to adapter protein Grb2 in lupus T cells. Clin.Immunol. 2003;106:41-49.Yasuda S, Stevens RL, Terada T et al. Defective expression of Ras guanyl nucleo-71. tide-releasing protein 1 in a subset of patients with systemic lupus erythemato-sus. J.Immunol. 2007;179:4890-4900.Roivainen A, Jalava J, Pirila L et al. H-ras oncogene point mutations in arthritic 72. synovium. Arthritis Rheum. 1997;40:1636-1643.Roivainen A, Zhu F, Sipola E, Yli-Jama T, Toivanen P. Failure to verify H-ras 73. mutations in arthritic synovium: comment on the article by Roivainen et al. Arthritis Rheum. 2001;44:2705.Schett G, Tohidast-Akrad M, Smolen JS et al. Activation, differential localiza-74. tion, and regulation of the stress-activated protein kinases, extracellular signal-regulated kinase, c-JUN N-terminal kinase, and p38 mitogen-activated protein kinase, in synovial tissue and cells in rheumatoid arthritis. Arthritis Rheum. 2000;43:2501-2512.Tak PP, Firestein GS. NF-kappaB: a key role in inflammatory diseases. J.Clin.75. Invest 2001;107:7-11.Yamamoto A, Fukuda A, Seto H et al. Suppression of arthritic bone destruction 76. by adenovirus-mediated dominant-negative Ras gene transfer to synoviocytes and osteoclasts. Arthritis Rheum. 2003;48:2682-2692.Pap T, Nawrath M, Heinrich J et al. Cooperation of Ras- and c-Myc-dependent 77. pathways in regulating the growth and invasiveness of synovial fibroblasts in

30

Chapter 1

rheumatoid arthritis. Arthritis Rheum. 2004;50:2794-2802.Na HJ, Lee SJ, Kang YC et al. Inhibition of farnesyltransferase prevents colla-78. gen-induced arthritis by down-regulation of inflammatory gene expression through suppression of p21(ras)-dependent NF-kappaB activation. J.Immunol 2004;173:1276-1283.Bos JL, de RJ, Reedquist KA. Rap1 signalling: adhering to new models. Nat.Rev.79. Mol.Cell Biol. 2001;2:369-377.Amsen D, Kruisbeek A, Bos JL, Reedquist K. Activation of the Ras-related GT-80. Pase Rap1 by thymocyte TCR engagement and during selection. Eur.J.Immunol. 2000;30:2832-2841.Minato N, Kometani K, Hattori M. Regulation of immune responses and hema-81. topoiesis by the Rap1 signal. Adv.Immunol. 2007;93:229-264.Bos JL. Linking Rap to cell adhesion. Curr.Opin.Cell Biol. 2005;17:123-128.82. Fukuhara S, Sakurai A, Sano H et al. Cyclic AMP potentiates vascular endothe-83. lial cadherin-mediated cell-cell contact to enhance endothelial barrier function through an Epac-Rap1 signaling pathway. Mol.Cell Biol. 2005;25:136-146.Knox AL, Brown NH. Rap1 GTPase regulation of adherens junction positioning 84. and cell adhesion. Science 2002;295:1285-1288.Katagiri K, Maeda A, Shimonaka M, Kinashi T. RAPL, a Rap1-binding molecule 85. that mediates Rap1-induced adhesion through spatial regulation of LFA-1. Nat.Immunol 2003;4:741-748.Katagiri K, Imamura M, Kinashi T. Spatiotemporal regulation of the kinase Mst1 86. by binding protein RAPL is critical for lymphocyte polarity and adhesion. Nat.Immunol 2006;7:919-928.Medeiros RB, Dickey DM, Chung H et al. Protein kinase D1 and the beta 1 inte-87. grin cytoplasmic domain control beta 1 integrin function via regulation of Rap1 activation. Immunity. 2005;23:213-226.Lafuente EM, van Puijenbroek AA, Krause M et al. RIAM, an Ena/VASP and 88. Profilin ligand, interacts with Rap1-GTP and mediates Rap1-induced adhesion. Dev.Cell 2004;7:585-595.Kooistra MR, Dube N, Bos JL. Rap1: a key regulator in cell-cell junction forma-89. tion. J.Cell Sci. 2007;120:17-22.Mueller KL, Daniels MA, Felthauser A et al. Cutting edge: LFA-1 integrin-90. dependent T cell adhesion is regulated by both ag specificity and sensitivity. J.Immunol 2004;173:2222-2226.Kinashi T. Intracellular signalling controlling integrin activation in lymphocy-91. tes. Nat.Rev.Immunol 2005;5:546-559.

31

Introduction

Laudanna C, Kim JY, Constantin G, Butcher E. Rapid leukocyte integrin activa-92. tion by chemokines. Immunol Rev. 2002;186:37-46.Tatla S, Woodhead V, Foreman JC, Chain BM. The role of reactive oxygen spe-93. cies in triggering proliferation and IL-2 secretion in T cells. Free Radic.Biol Med. 1999;26:14-24.Goldstone SD, Fragonas JC, Jeitner TM, Hunt NH. Transcription factors as tar-94. gets for oxidative signalling during lymphocyte activation. Biochim.Biophys.Acta 1995;1263:114-122.Hehner SP, Breitkreutz R, Shubinsky G et al. Enhancement of T cell receptor 95. signaling by a mild oxidative shift in the intracellular thiol pool. J.Immunol 2000;165:4319-4328.Remans PH, Gringhuis SI, van Laar JM et al. Rap1 signaling is required for sup-96. pression of Ras-generated reactive oxygen species and protection against oxida-tive stress in T lymphocytes. J.Immunol 2004;173:920-931.Maurice MM, Nakamura H, van d, V et al. Evidence for the role of an altered 97. redox state in hyporesponsiveness of synovial T cells in rheumatoid arthritis. J.Immunol 1997;158:1458-1465.Gilgun-Sherki Y, Melamed E, Offen D. The role of oxidative stress in the patho-98. genesis of multiple sclerosis: the need for effective antioxidant therapy. J.Neurol. 2004;251:261-268.Fernandez D, Bonilla E, Phillips P, Perl A. Signaling abnormalities in systemic 99. lupus erythematosus as potential drug targets. Endocr.Metab Immune.Disord.Drug Targets. 2006;6:305-311. Tak PP, Zvaifler NJ, Green DR, Firestein GS. Rheumatoid arthritis and p53: how 100. oxidative stress might alter the course of inflammatory diseases. Immunol To-day 2000;21:78-82. Fairburn K, Stevens CR, Winyard PG et al. Oxidative stress and its control: a pa-101. thogenetic role in inflammatory joint disease. Biochem.Soc.Trans. 1993;21:371-375. Bowie A, O'Neill LA. Oxidative stress and nuclear factor-kappaB activation: a 102. reassessment of the evidence in the light of recent discoveries. Biochem.Pharma-col. 2000;59:13-23. Li Y, Yan J, De P et al. Rap1a null mice have altered myeloid cell functions sug-103. gesting distinct roles for the closely related Rap1a and 1b proteins. J.Immunol. 2007;179:8322-8331. Duchniewicz M, Zemojtel T, Kolanczyk M et al. Rap1A-deficient T and B cells 104. show impaired integrin-mediated cell adhesion. Mol.Cell Biol. 2006;26:643-653.

32

Chapter 1

Amsen D, Kruisbeek A, Bos JL, Reedquist K. Activation of the Ras-related GT-105. Pase Rap1 by thymocyte TCR engagement and during selection. Eur.J.Immunol. 2000;30:2832-2841. Kometani K, Moriyama M, Nakashima Y et al. Essential role of Rap signal in 106. pre-TCR-mediated beta-selection checkpoint in alphabeta T-cell development. Blood 2008;112:4565-4573. Dillon TJ, Carey KD, Wetzel SA, Parker DC, Stork PJ. Regulation of the small 107. GTPase Rap1 and extracellular signal-regulated kinases by the costimulatory molecule CTLA-4. Mol.Cell Biol. 2005;25:4117-4128. Ishida D, Yang H, Masuda K et al. Antigen-driven T cell anergy and defec-108. tive memory T cell response via deregulated Rap1 activation in SPA-1-deficient mice. Proc.Natl.Acad.Sci.U.S.A 2003;100:10919-10924. Li L, Greenwald RJ, Lafuente EM et al. Rap1-GTP is a negative regulator of Th 109. cell function and promotes the generation of CD4+CD103+ regulatory T cells in vivo. J.Immunol. 2005;175:3133-3139. Sebzda E, Bracke M, Tugal T, Hogg N, Cantrell DA. Rap1A positively regulates 110. T cells via integrin activation rather than inhibiting lymphocyte signaling. Nat.Immunol. 2002;3:251-258. Jaffe AB, Hall A. Rho GTPases: biochemistry and biology. Annu.Rev.Cell Dev.111. Biol. 2005;21:247-269. Kraynov VS, Chamberlain C, Bokoch GM et al. Localized Rac activation dyna-112. mics visualized in living cells. Science 2000;290:333-337. Michiels F, Habets GG, Stam JC, van der Kammen RA, Collard JG. A role for Rac 113. in Tiam1-induced membrane ruffling and invasion. Nature 1995;375:338-340. Crespo P, Schuebel KE, Ostrom AA, Gutkind JS, Bustelo XR. Phosphotyrosine-114. dependent activation of Rac-1 GDP/GTP exchange by the vav proto-oncogene product. Nature 1997;385:169-172. ten Klooster JP, Jaffer ZM, Chernoff J, Hordijk PL. Targeting and activation of 115. Rac1 are mediated by the exchange factor beta-Pix. J.Cell Biol. 2006;172:759-769. Genot EM, Arrieumerlou C, Ku G et al. The T-cell receptor regulates Akt (pro-116. tein kinase B) via a pathway involving Rac1 and phosphatidylinositide 3-kinase. Mol.Cell Biol. 2000;20:5469-5478. Hashimoto A, Okada H, Jiang A et al. Involvement of guanosine triphosphata-117. ses and phospholipase C-gamma2 in extracellular signal-regulated kinase, c-Jun NH2-terminal kinase, and p38 mitogen-activated protein kinase activation by the B cell antigen receptor. J.Exp.Med. 1998;188:1287-1295.

33

Introduction

Jacinto E, Werlen G, Karin M. Cooperation between Syk and Rac1 leads to sy-118. nergistic JNK activation in T lymphocytes. Immunity. 1998;8:31-41. Marinari B, Costanzo A, Viola A et al. Vav cooperates with CD28 to induce NF-119. kappaB activation via a pathway involving Rac-1 and mitogen-activated kinase kinase 1. Eur.J.Immunol. 2002;32:447-456. Montaner S, Perona R, Saniger L, Lacal JC. Multiple signalling pathways lead to 120. the activation of the nuclear factor kappaB by the Rho family of GTPases. J.Biol.Chem. 1998;273:12779-12785. Sugihara K, Nakatsuji N, Nakamura K et al. Rac1 is required for the formation 121. of three germ layers during gastrulation. Oncogene 1998;17:3427-3433. Gu Y, Filippi MD, Cancelas JA et al. Hematopoietic cell regulation by Rac1 and 122. Rac2 guanosine triphosphatases. Science 2003;302:445-449. Glogauer M, Marchal CC, Zhu F et al. Rac1 deletion in mouse neutrophils has 123. selective effects on neutrophil functions. J.Immunol. 2003;170:5652-5657. Zhao X, Carnevale KA, Cathcart MK. Human monocytes use Rac1, not Rac2, in 124. the NADPH oxidase complex. J.Biol.Chem. 2003;278:40788-40792. Wells CM, Walmsley M, Ooi S, Tybulewicz V, Ridley AJ. Rac1-deficient macro-125. phages exhibit defects in cell spreading and membrane ruffling but not migra-tion. J.Cell Sci. 2004;117:1259-1268. Benvenuti F, Hugues S, Walmsley M et al. Requirement of Rac1 and Rac2 expres-126. sion by mature dendritic cells for T cell priming. Science 2004;305:1150-1153. Walmsley MJ, Ooi SK, Reynolds LF et al. Critical roles for Rac1 and Rac2 GTPa-127. ses in B cell development and signaling. Science 2003;302:459-462. Fukuda A, Hikita A, Wakeyama H et al. Regulation of osteoclast apoptosis and 128. motility by small GTPase binding protein Rac1. J.Bone Miner.Res. 2005;20:2245-2253. Wang Y, Lebowitz D, Sun C et al. Identifying the relative contributions of Rac1 129. and Rac2 to osteoclastogenesis. J.Bone Miner.Res. 2008;23:260-270.

The presumed hyporespon-sive behaviour of rheuma-toid arthritis T lympho-

cytes can be attributed to spontaneous ex vivo apop-tosis rather than defects

in T cell receptor signaling

Joana RF Abreu1, Aleksander M Grabiec1, Sarah Krausz1, René Spijker2, Tomasz Burakowski3, Wlodzi-

mierz Maslinski3, Eric Eldering2, Paul P Tak1, and Kris A Reedquist1

1Division of Clinical Immunology and Rheumatology and 2De-partment of Experimental Immunology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands. 3Depart-ment of Pathophysiology and Immunology, Institute of Rheuma-

tology, Warsaw, Poland.

The Journal of Immunology, 2009, 183(1):621-630

2

37

Competent TCR signaling is maintained in RA SF T cells

Abstract

Genetic associations and the clinical success of compounds targeting TCR costimu-latory proteins suggest an active role for TCR signaling in the initiation and perpetu-ation of rheumatoid arthritis (RA). Paradoxically, T cells isolated from affected joints in RA show impaired proliferative and cytokine responses following stimulation with mitogens and recall antigens, attributed in part to chronic T cell exposure to oxidative stress and inflammatory cytokines. Therefore, it is uncertain how local autoreactive TCR signaling contributes to pathology in established RA. Using single cell analysis, we show that in contrast to results obtained in bulk culture assays, T cells from the synovial fluid (SF) of RA patients proliferate and produce cytokines (IL-2, TNF-α and IFN-g) as efficiently, if not more so, than T cells isolated from heal-thy donors and RA patient peripheral blood following TCR/CD28 stimulation. RA SF T cell hyporesponsiveness observed in bulk cultures can be attributed to spon-taneous apoptosis ex vivo, which is associated with altered ratios of pro-apoptotic Noxa and anti-apoptotic Mcl-1 expression. The absence of RA synovial T cell pro-liferation and cytokine production in situ, despite the capacity of these cells to sup-port productive TCR signaling, suggests that T cells contribute to local pathology in established RA by TCR-independent mechanisms.

38

Chapter 2

Introduction

T lymphocytes are thought to contribute to synovitis and joint destruction in rheu-matoid arthritis (RA) through multiple mechanisms. These include the pleiotropic activation of other synovial cells, such as macrophages and stromal fibroblast-like synoviocytes (FLS) via cell-cell contacts and interleukin IL-17 production, stimula-tion of B lymphocytes producing autoimmune Abs, and promotion of osteoclast dif-ferentiation1. Human genetic studies, experimental animal models of arthritis, and recent clinical experience have provided evidence consistent with a role for impro-per engagement of auto-reactive TCRs in the initiation and progression of disease in RA1. However, our inability to detect significant T cell proliferation or cytokine production at sites of inflammation in RA, combined with observed defects in TCR-proximal signaling and TCR-dependent functional responses of RA synovial T cells, has raised questions regarding the role of TCR signaling in established RA, and the mechanism of action of therapies targeting TCR costimulatory pathways2-4.The strongest evidence supporting an active role for TCR engagement in the initiati-on, if not the perpetuation, of disease in RA is data underlying the “shared epitope” hypothesis. Expression of specific HLA- DR1 and DR4 shared epitope alleles enhan-ces the risk of the development of RA and contributes to disease severity5-8. Additi-onally, polymorphisms in T cell gene products which influence the quality of TCR responses, such as PTPN22, PD-1, CTLA-4, TRAF1-C5 and CD40, have also been identified as candidate susceptibility genes in RA1;9-12. Finally, evidence consistent with an active role for TCR engagement in established arthritis is observed in clinical trials using soluble CTLA-4Ig fusion protein (abatacept) to disrupt interaction of the TCR costimulatory protein CD28 with CD80/CD86 ligands expressed on synovial APCs and FLS. Initial clinical trials using abatacept to treat RA have demonstrated clear clinical benefits, even in patients refractory to therapy with biologicals that block TNF-α signaling13-15.T cells derived from both RA synovial tissue and synovial fluid (SF) display simi-lar phenotypic and functional abnormalities. These T cells express surface markers characteristic of recent TCR stimulation, including CD44, CD45RO, CD69, HLA-DR, and VLA-43;16-20. These cells are primarily of a Th1 phenotype, and resistant to Th2 polarization ex vivo21-24. Despite this, little direct evidence is available demonstra-ting that the TCR is functionally engaged in RA. Proliferation of synovial tissue T cells is not observed in situ, and the relatively low levels of IL-2 and IFN-g which can be detected in RA synovial T cells are inconsistent with a contributory pathological role for TCR signaling in established RA25-29. Ex vivo, RA synovial tissue and SF

39


T cells are hyporesponsive to stimulation by pharmacological mitogens and recall antigens, both in terms of proliferation and cytokine secretion2;3;25;30-32. This may be a consequence of chronic T cell exposure to inflammatory cytokines and/or oxidative stress, which can lead to altered expression or mis-folding of critical TCR signaling proteins, such as the CD3z chain30;33 and LAT31;32;34.The inability to provide evidence of direct TCR engagement in RA synovium, in combination with identified TCR-proximal signaling defects in RA synovial T lymp-hocytes in vitro, has led to the suggestion that these cells contribute to pathology through TCR-independent mechanisms2;3. Many properties of RA T cells, including surface expression of activation markers and cell-cell contact–dependent activation of macrophages and FLS, can be recapitulated by peripheral blood (PB) T cell expo-sure to IL-15 or a combination of inflammatory cytokines19;35;36. Similar effects are observed in PB T cells chronically exposed to TNF-α20;30. Additionally, stimulation of PB T cells with a number of inflammatory cytokines, in combination with CD28 ligation, can reproduce oxidative stress observed in synovial T cells37.Here, we report the unexpected finding that TCR signaling is functionally intact in freshly isolated RA SF T lymphocytes, and fully capable of initiating cytokine production and T cell proliferation. Previously observed RA SF T cell hyporespon-siveness in bulk culture assays is due to spontaneous apoptosis of these cells ex vivo, associated with changes in the relative expression of the pro-apoptotic protein Noxa and anti-apoptotic Mcl-1. Our results suggest that the inability to detect evidence of TCR engagement in RA synovial tissue is unlikely a result of defects in TCR signa-ling, but rather, lack of TCR engagement.

Materials and Methods

Patients

Paired PB and SF samples were obtained from patients attending our out-patient clinics, with clinically active RA fulfilling the American College of Rheumatology (ACR) revised criteria for RA38. Patient characteristics are presented in Table I. At the time of sample collection eleven patients were receiving methotrexate (2.5-25 mg/week), four prednisolone (2.5-10 mg/day), one leflunomide (20 mg/day), five TNF-a antagonist therapy (adalimumab, 40 mg/2 weeks; etanercept, 25 mg twice a week or 50 mg/week; infliximab, 3 mg/kg intravenously [iv] every 8 weeks), one had received rituximab treatment, and one patient was not receiving any medication at

40

Chapter 2

the time of arthrocentesis. All patients provided informed written consent, and the study was approved by the Medical Ethics Committees of the Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands and the Institute of Rheumatology, Warsaw, Poland.

Cell isolation and culture

PB mononuclear cells (PBMCs) from healthy volunteers and PBMC and SF mononu-clear cells (SFMCs) from RA patients were isolated by Ficoll-Isoplaque density gra-dient centrifugation (Nycomed, Pharma, Oslo, Norway). PB and SF T cells were pu-rified from PBMC and SFMC using a negative isolation procedure (T Cell Negative Isolation Kit, Dynal Biotech, Oslo, Norway) in accordance with the manufacturer’s instructions. Purified T cells were >95% CD3+ as assessed by FACS analysis (see be-low). T cells were cultured at 1x106/ml in IMDM medium supplemented with 10% FCS, L-glutamine, 25 mM HEPES, streptomycin (100 ng/ml) and penicillin (10U/ml) (all from Gibco, Invitrogen, Carlsbad, CA). T cells were stimulated with either 1 µg/ml anti-CD3 (clone 1XE) and 1 µg/ml anti-CD28 (clone 15E8) mAbs (both from Sanquin, Amsterdam, The Netherlands) or 1 ng/ml PMA and 1 µg/ml ionomycin (I) (both from Sigma-Aldrich, St. Louis, MO).

Measurement of T cell proliferation

T cell proliferation in bulk cultures was assessed by culturing T cells for 72 hours in a 96-well plate in the absence or presence of activating Abs. During the last 20 hours of culture, cells were pulsed with 1 mCi [3H] thymidine (Amersham Biosciences, Pis-cataway, NJ). Cells were harvested and incorporated radioactivity measured using a 1450 Microbeta Plus Liquid Scintillation counter (Perkin Elmer, Waltham, MA). For single-cell analysis of T cell proliferation, PBTC and SFTC were resuspended at 5-10x106 cells/ml in PBS and labeled with 2.5 mM CFSE (Molecular Probes Eu-rope BV, Leiden, The Netherlands) for 10 minutes at 37°C. Cells were washed and subsequently resuspended in complete culture medium. T cells (1x106/ml) were left unstimulated or stimulated for 72 hours at 37°C with anti-CD3 and anti-CD28 Abs. Proliferation was detected using a FACSCalibur flow cytometer (BD Biosci-ences, San Jose, CA) and CellQuest Pro software (BD Biosciences). The precursor frequency (percentage of cells in the initial population that underwent one or more

41


divisions), and the mean number of divisions per proliferating cells were calculated as previously described39.

Detection of cytokine production

T cell culture supernatants were collected for cytokine analysis 24h (for IL-2) or 72h (TNF-α, IFN-g) post-stimulation, and cytokine concentrations measured using a Bio-Plex Human 27-plex panel (Bio-Rad Laboratories, Hercules, CA) according to the manufacturer’s instructions. For single-cell analysis of T cell cytokine production, T cells were stimulated for six hours with anti-CD3/CD28 Abs or PMA/I, with 10 µg/ml Brefeldin A (Sigma-Aldrich) included for the last four hours of stimulation. Cells were fixed with 4% (w/v) paraformaldehyde/PBS and permeabilized with 0.5% (w/v) BSA/PBS containing 0.1% (w/v) saponin (Sigma-Aldrich). Cells were then in-cubated with conjugated anti- IL-2-APC, IFN-g-PE, or TNF-α-APC Abs (all from BD Biosciences). The percentage of positively stained cells and the mean fluorescent intensity of staining were measured by flow cytometry as above.

Apoptosis detection

Cells were washed in ice-cold HEPES buffer and incubated with APC- or FITC- la-beled Annexin-V (IQ Products, Groningen, The Netherlands) for 30 minutes. Pro-pidium iodide (PI; 5 mg/ml, Sigma-Aldrich) was added prior to analysis, and the percentage of viable cells quantified by flow cytometry.

RT-MLPA Procedure and Analysis

Total T cell mRNA was isolated using a GenElute RNA isolation kit (Sigma-Aldrich). Reverse transcription multiplex ligation-dependent probe amplification (RT-MLPA) of pro- and anti- apoptotic genes was performed as previously described40. Briefly, RNA was reverse-transcribed using a gene-specific probe mix (MRC Holland, Am-sterdam, The Netherlands). The obtained cDNA was annealed to MLPA probes and covalently linked with Ligase-65 (MRC Holland). Ligation products were amplified and fluorescently labeled by PCR using one unlabeled and one 6-carboxy-fluores-cein-labeld primer. PCR products were applied to an ABI 3100 capillary sequen-

42

Chapter 2

cer (Applied Biosystems, Warrington, United Kingdom), and data processed with Genescan and Genotyper software (both from Applied Biosystems). Final analyses were conducted with Microsoft Excel spreadsheet software (Microsoft, Redman, WA). The sum of all peak data was set at 100% to normalize for fluctuations in total signal between samples. Individual peaks for each gene product were then calcula-ted relative to the total value.

SDS-Page and Western Blotting

T cells were washed with ice-cold PBS and lysed in buffer containing 1% Chaps, 20 mM Tris-HCl (pH 7.4), 135 mM NaCl, 1.5 mM MgCl2, 10% glycerol, 1 mM EGTA, 2 mM Na3VO4, 10 mM NaF, 2 mg/ml leupeptin, 1 mM PMSF, 0.1 mM TLCK, and 2 mg/ml trypsin inhibitor. Lysates were cleared by centrifugation at 13k rpm for 15 minutes, and protein expression analyzed by standard western blotting procedures as previously reported in detail37. Proteins were resolved by SDS-PAGE, transfer-red to PVDF membrane (Bio-Rad Laboratories), and blots probed with antibodies against Noxa (Imgenex, San Diego, CA), Bim (Chemicon, Temecula, CA), Bcl-2 (Alexis, San Diego, CA), Bcl-XL (BD Transduction Laboratories, Lexington, KY), Mcl-1 (BD Biosciences Pharmingen, San Diego, CA), b-actin and ERK 1/2 (Santa Cruz Biotechnology, Santa Cruz, CA).

Statistical Analysis

Values between groups were compared using one-way ANOVA followed by Dun-nett’s post hoc test using RA SF T lymphocytes as the reference group. Variables that were not normally distributed were rank transformed prior to the analyses. Comparisons within groups were done using a t-test or Mann-Whitney test where appropriate. To correct for multiple testing, overall p-values were corrected using the False Discovery rate (FDR) and p-values < 0.05 were considered statistically sig-nificant.

43


Results

RA SF TCR is competent to support T cell proliferative responses

In initial analyses of RA SF T cell responses to TCR ligation, we first attempted to confirm previously published reports that RA SF T lymphocytes were refractory to TCR/CD28-induced proliferation. Freshly isolated HD PB, and paired RA PB and SF T cells were stimulated for 72 hours in the presence of anti-TCR/CD28 Abs, and proliferation measured by incorporation of [3H]-thymidine during the last 20 hours of culture (Figure 1). In the absence of TCR/CD28 stimulation, [3H]-thymidine incor-poration was barely detectable in all samples, and did not differ statistically between HD PB, RA PB, and RA SF T cell populations. In accordance with previous inde-

Age (years)Male:femaleDisease duration (years)Erythrocyte sedimentation rate (mm/hour)C-reactive protein (mg/L)Rheumatoid factor (kU/L)Anticitrullinated peptide Abs (kAU/L)

52 (25-85)3:10

9 (1-17)57 (23-120)64 (2-117)

102 (<1-438)2438 (2-15951)

Characteristic Median (range)

Table I. Clinical features of RA patients (n=13) included in the study.

0300006000090000

120000150000180000

medium TCR/CD28

3H-th

ymid

ine

inc

(cpm

) HDRA PBRA SF

**

Figure 1. RA SF T lymphocyte proliferative responses are depressed in bulk cultures. HD PB, RA PB, and RA SF T lymphocytes were cultured in the presence of anti-TCR/CD28 an-tibodies for 72 hours, pulsed for an additional 20 hours with [3H]-thymidine, and [3H]-thymidine incorporation quantified (n=3). Mean values of counts per minute (cpm) +/- SEM are indicated. * p < 0.05.

44

Chapter 2

pendent observations, RA SF T lymphocyte proliferation was significantly reduced following TCR/CD28 stimulation, compared to HD (82% reduction, p < 0.05) or RA PB T cells (79% reduction, p < 0.05).

1 10 100 1000 100000

500

1000

1500

1 10 100 1000 100000

500

1000

1500

1 10 100 1000 100000

300

600

900

1200

1 10 100 1000 100000

200

400

600

800

1 10 100 1000 100000

200

400

600

1 10 100 1000 100000

100

200

300

400

500

TCR/CD28Unstimulated

HD

RA

PB

RA

SF

CD4 CD8

CFSE

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

0

20

40

60

80

100

100 101 102 103 104 0

1

2

3

CD3 CD4 CD8Mea

n nu

mbe

rof d

ivis

ions

0

20

40

60

80

CD3 CD4 CD8Prec

urso

r Fre

quen

cy(%

)

HDRA PBRA SF

D

CBA

Figure 2. Single cell analysis fails to reveal defects in RA SF T lymphocyte proliferative responses. T cell proliferative responses were assessed by FACS analysis of T cell CFSE dilution fol-lowing stimulation with anti-TCR/CD28 antibodies. A, Representative histograms of CFSE labeled T cells cultured in the absence (left panels) or presence (right panels) of activating anti-TCR/CD28 for 72 hours (n=5). B, Representative histograms of CFSE dilution in CD4+ (n=2) and CD8+ (n=2) T lympho-cytes in the absence (broken line) and presence (solid line) of anti-TCR/CD28 stimulation. C, Precur-sor frequency and D, mean number of HD PB (white bars), RA PB (gray bars) and RA SF (black bars) T cell divisions calculated after anti-TCR/CD28 stimulation. Mean values +/-SEM are indicated.

We next attempted to confirm RA SF T cell proliferative hyporesponsiveness at the single-cell level, and determine whether residual proliferative responses might be limited to specific T cell subpopulations. To accomplish this, HD PB, RA PB, and RA SF T cells were labeled with CFSE, and cultured in the absence or presence of anti-TCR/CD28 Abs for three days prior to analysis by flow cytometry. Surprisingly, in contrast to results from bulk T cell analysis, CFSE dilution in total RA SF CD3+

T lymphocytes was not reduced compared to HD PB T lymphocytes (Figure 2A). Indeed, proliferation was most robustly observed in RA SF T lymphocytes, followed by RA PB T lymphocytes. This relative enhanced TCR/CD28-induced proliferation of RA SF T cells was observed in both CD4+ and CD8+ T cell subsets (Figure 2B). After

45


calculating the percentage of T cells in the initial populations that underwent one or more cell divisions, we found that the precursor frequency of RA SF T cells was not depressed compared to HD or RA PB T cells (Figure 2C). Rather, the precursor fre-quency of RA SF T cells (45%) was increased compared to HD T cells (27%), although this difference did not reach statistical significance (p = 0.152). Again, this trend was observed in both CD4+ and CD8+ T cell subsets (Figure 2C). No differences were observed between HD PB, RA PB and RA SF T cells in terms of the mean number of cell divisions achieved following TCR/CD28 stimulation, either in total CD3+ T lymphocytes or in CD4+ and CD8+ T cell subsets (Figure 2D). Differences in RA SF T cell proliferation observed between bulk cultures and single cell analyses were unlikely due to patient heterogeneity or drug treatment, as RA SF T cells from two of the three patients studied in [3H]-thymidine incorporation experiments (Figure 1) were assessed in parallel by CFSE dilution. Thus, in contrast to observations made in analyses of bulk T cell populations, single cell analysis of RA SF T cell responses to TCR/CD28 stimulation reveals that these cells can proliferate as well, if not better, than their HD PB or RA PB counterparts. This may be due to an increased frequency of precursors competent to initiate proliferation.

RA SF T cells display increased cytokine production following TCR stimulation

Given the discordance of our results with previous observations regarding RA SF T cell proliferative responses to TCR/CD28 stimulation, we next assessed SF T cell

IFN-γ

0

35000

70000

105000

140000

medium TCR/CD28

pg/m

l

TNF-α

0

2000

4000

6000

8000

10000

medium TCR/CD28

pg/m

l

*

IL-2

0

20000

40000

60000

80000

100000

medium TCR/CD28

pg/m

l

*HDRA PBRA SF

*

Figure 3. IL-2 and TNF-α production is decreased in cell culture supernatants of TCR/CD28-stimulated RA SF T lymphocytes. HD PB (white bars), RA PB (gray bars) and RA SF T lymphocytes (black bars) were left unstimulated or stimulated with anti-TCR/CD28 antibodies, and medium collected for cytokine analysis by ELISA. Cells were stimulated for 24 hours to detect IL-2 pro-duction (left panel, n=3) and for 72 hours to detect TNF-α (middle panel, n=3) and IFN-g (right panel, n=3) production. Values represent the mean and SEM of cytokine concentrations (pg/ml). * p < 0.05.

46

Chapter 2

cytokine responses. In agreement with previous reports, following TCR/CD28 sti-mulation, IL-2 production in supernatants of bulk RA SF T cell cultures was severely impaired compared to autologous RA PB T cells (p< 0.001) (Figure 3). Additionally, TNF-α production in RA SF T cells was decreased compared to RA PB T cells (p < 0.005). A similar trend was observed when comparing RA SF T cells with HD PB T cells. In contrast, HD PB, RA PB, and RA SF T cells all produced similar levels of IFN-g. We next analyzed TCR/CD28-induced cytokine production by intracellular staining and flow cytometry (Figures 4 and 5). In line with previous reports22;24, pharma-cological stimulation of HD PB, RA PB, and RA SF T cells with PMA/I resulted in IL-2 production in a similar frequency of cells, while the frequency of TNF-α- and IFN-g- producing T lymphocytes was elevated in RA SF (Figure 5A). Unexpectedly, following TCR/CD28 stimulation, RA SF T cell IL-2 production was significantly in-creased when compared to TCR/CD28-stimulated RA PB (p < 0.005) or HD T cells (p

Figure 4. Single cell analysis reveals robust TCR-dependent cytokine production by RA SF T lymphocytes. Representative histograms of IL-2 (top row), TNF-α (middle row) and IFN-g production (bottom row) in HD, RA PB and RA SF T lymphocytes, following T cell stimulation in the absence (dashed black lines) or presence (red solid lines) of PMA/I or anti-TCR/CD28 antibodies. Cells were stimulated for 6 hours and Brefeldin A included for the last 4 hours of culture. Cells were fixed, permeabilized, stained with anti -IL-2, -TNF-α , and -IFN-g antibodies, and cytokine-producing cells detected by FACS analysis.

IL-2

TNF-α

IFN

-γ

PMA-I TCR-CD28

HD

PMA-I TCR-CD28

RA PB

PMA-I TCR-CD28

RA SF

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

IFN-γ

TNF-α

IL-2

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

100 101 102 103 1040

20

40

60

80

100

47


< 0.005) (Figure 5B). Also TNF-α and especially IFN-g production was significantly increased in SF T cells compared to HD or RA PB T cells. The frequency of IFN-g producing T cells after TCR/CD28 stimulation was as high as 13 times that of HD T cells (p < 0.005). The same pattern of cytokine production observed in the CD3+ T cell population could be observed in both CD4+ (Figure 5C) and CD8+ cells (Figure 5D). The mean fluorescent intensity of cytokine staining in RA SF T cells was similar to or higher than that observed in HD PB and RA PB T cells (data not shown), indicating that decreased cytokine production observed in RA SF T cell bulk cultures was not due to inefficient cytokine production by responding cells. Additionally, differences

IL-2

TNF-α

IFN

-γ

010203040506070

medium PMA/I

% P

ositi

ve C

ells

05

101520

TCR/CD28

% P

ositi

ve C

ells

05

10152025

% P

ositi

ve C

ells

0153045607590

medium PMA/I

% P

ositi

ve C

ells

010203040

TCR/CD28

% P

ositi

ve C

ells

010203040506070

medium PMA/I

% P

ositi

ve C

ells

02468

10

TCR/CD28

% P

ositi

ve C

ells

01020304050

% P

ositi

ve C

ells

02468

10

% P

ositi

ve C

ells

0

10

20

30

40

% P

ositi

ve C

ells

0

5

10

15

% P

ositi

ve C

ells

05

10152025

% P

ositi

ve C

ells

medium TCR/CD28 medium TCR/CD28

TCR/CD28 medium TCR/CD28

medium TCR/CD28 medium TCR/CD28

**

**

**

**

**

**

**

**

**

**

**

CD4 CD8CD3 CD3

medium

HDRA PBRA SF

Figure 5. TCR-dependent cytokine production is enhanced in RA SF T lymphocytes. Cyto-kine production in HD PB (white bars), RA PB (gray bars) and RA SF T lymphocytes (black bars) in the absence or presence of PMA/I or TCR/CD28 stimulation was assessed by intracellular staining for IL-2 (upper panels), TNF-α (middle panels), and IFN-g (lower panels) and FACS analysis. A, T cells were left unstimulated or stimulated with PMA/I for 6 hours, and Brefeldin A included for the last 4 hours of culture. Cells were fixed, permeabilized, stained with anti– IL-2, -TNF-α, and -IFN-g antibodies, and cy-tokine-producing cells detected by FACS analysis. B. Cytokine production in T lymphocytes stimulated with anti-TCR/CD28. Data was obtained in the same experiments presented in A and presented in sepa-rate graphs to facilitate visualization of differences in T cell cytokine responses. C, Cytokine production in CD4+ and D, CD8+ T cell subsets. Values represent the mean and SEM of the percentage of positive cells in CD3+ (n=6), CD4+ (n=4), and CD8+ (n=3) populations from independent experiments. * p < 0.05.

DCBA

48

Chapter 2

in cytokine production observed between ELISA analysis of bulk cultures (Figure 3) and single cell analysis were not attributable to patient heterogeneity or drug treat-ment, as all three of the patients assessed by ELISA were studied in parallel single cell analyses. Together these results demonstrated that at the single cell level, the frequency of TCR-responsive lymphocytes is elevated in RA SF.

Figure 6. RA SF T lymphocytes undergo spontaneous apoptosis ex vivo. A, Representative forward scatter (FSC)/side scatter (SSC) dot plots of HD PB (upper panels), RA PB (middle panels), and RA SF T lymphocytes (lower panels) after 72 hours in culture in the absence (unstimulated) or presence of activating anti-TCR/CD28 antibodies. Numbers indicate the percentage of viable gated cells. B, Representative FSC/SSC dot plots and stainings of HD PB (upper panels), RA PB (middle panels), and RA SF T lymphocytes (lower panels) with annexin V and propidium iodide (PI) immedi-ately after isolation (T=0) and after 24 hours (T=24h) in culture. C and D, The percentage of apoptotic HD PB (open diamonds), RA PB (gray squares), and RA SF T lymphocytes (black circles) as detected by annexin V/PI staining and FACS analysis after 0, 24, 48, and 72 hours in culture in the absence (C) or presence (D) of activating anti-TCR/CD28 antibodies. Values represent the mean percentage of apoptotic cells and SEM of 9 independent experiments. *p < 0.05, **p < 0.001.

DC

BA

0

15

3045

60

75

90

0 h 24 h 48 h 72 h

Dea

d C

ells

(%)

HDRA PBRA SF

** * *

0

15

30

45

60

75

90

0 h 24 h 48 h 72 h

Dea

d C

ells

(%) **

* *

FSC-Height Annexin

TCR/CD28Unstimulate d

0 50 100 150 200 2500

50

100

150

200

250

71.6

0 50 100 150 200 2500

50

100

150

200

250

75.8

0 50 100 150 200 2500

50

100

150

200

250

35.5

0 50 100 150 200 2500

50

100

150

200

250

78.7

0 50 100 150 200 2500

50

100

150

200

250

83.8

0 50 100 150 200 2500

50

100

150

200

250

41.1

SS

C-H

eigh

t

T=0h

0 50 100 150 200 2500

50

100

150

200

250

80.6

0 50 100 150 200 2500

50

100

150

200

250

97.2

0 50 100 150 200 2500

50

100

150

200

250

96.6

PI

100 101 102 103 104100

101

102

103

104 0.76 0.59

0.74

100 101 102 103 104100

101

102

103

104 0.53 0.33

0.46

100 101 102 103 104100

101

102

103

104 4.59 12

4.32

FSC-Height Annexin

T=24h

0 50 100 150 200 2500

50

100

150

200

250

74.5

0 50 100 150 200 2500

50

100

150

200

250

94.6

0 50 100 150 200 2500

50

100

150

200

250

97.3

PI

100 101 102 103 104100

101

102

103

104 2.36 3.68

3.58

100 101 102 103 104100

101

102

103

104 1.12 0.66

0.93

100 101 102 103 104100

101

102

103

104 0.53 27.6

24.5

HD

RA SF

RA PB

SS

C-H

eigh

t

49


RA SF T cell hyporesponsiveness in bulk cultures is secondary to spontaneous ex-vivo apoptosis

One possible explanation for the observed discrepancy between RA SF T cell pro-liferative and cytokine responses in bulk cultures and single cell analyses could be changes in RA SF T cell viability, as these cells have been reported to quickly un-dergo apoptosis ex vivo41;42. Assays using [3H]-thymidine incorporation and tissue culture supernatant ELISA analyses can not accurately account for apoptosis which may occur during extended cell culture. Initial examination of live cell gating of HD PB, RA PB and RA SF T lymphocytes 72 hours after isolation and CFSE labeling sug-gested a significant loss of viability of RA SF T cells under these culture conditions (Figure 6A). We confirmed this by performing Annexin-V/PI stainings on HD PB, RA PB, and RA SF T cells. Apoptosis was measured immediately after T cell isola-tion and after 24 hours (Figure 6B). Immediately following T cell isolation, similar percentages of apoptotic cells were observed in all samples. However, after 24 hours in culture, RA SF T cells displayed almost four times higher levels of apoptosis as compared to HD T cells (p < 0.001) and to RA PB T cells (p < 0.001). RA SF T cells continued to undergo apoptosis at a higher rate than the other T cell populations after 48 and 72 hours in culture (Figure 6C), and remained elevated compared to HD (p < 0.001) and RA PB (p< 0.001) controls in the presence of TCR/CD28 stimulation (Figure 6D).

RA SF T cells have altered expression levels of Noxa, Bcl-2, and Bcl-XL

To investigate in more detail the mechanisms which might be responsible for the increased susceptibility of RA SF T cells to apoptosis ex vivo, we quantified the re-lative expression of gene products known to be direct regulators of apoptosis. To accomplish this, total mRNA from freshly isolated HD PB, RA PB, and RA SF T lymphocytes was subjected to a RT-MLPA assay, allowing simultaneous quantifi-cation of expression of 34 important regulators of apoptosis. Expression profiles of HD PB, RA PB, and RA SF T cells were remarkably similar (Figure 7A). However, mRNA expression of the pro-apoptotic BH3-only family member Noxa was incre-ased approximately two-fold in RA SF T lymphocytes compared to HD (p < 0.05) and RA PB (p < 0.05) T cell populations (Figure 7B). The expression of Bim, NIP3 and Puma, three other pro-apoptotic Bcl-2 family members known to regulate T cell apoptosis was equivalent between RA SF and other T cells (Figure 7A). Among

50

Chapter 2

anti-apoptotic gene products associated with T cell survival, we saw no differences in the expression levels of Bcl-XL or the Noxa-binding partner Mcl-1 (Figure 7B). An approximately 50% reduction in Bcl-2 expression was observed, but did not reach statistical significance.As many Bcl-2 family members are also subjected to post-translational modifications affecting protein stability, we examined protein expression of pro- and anti- apop-totic proteins in whole cell lysates (Figures 7C and 7D). Consistent with mRNA data, Noxa protein expression was elevated in RA SF T cells compared to HD PB T cells. Surprisingly, protein expression of Noxa was even higher in RA PB T cells.

RA

PB

Mcl-1

Noxa

Bcl-2

Bcl-XL

actinH

DR

A SF

RA

PB

RA

SF

Bcl-2

Bim

Erk

0

4

8

12

16

20Bc

l-W Bcl2

Bcl-X

LA1

MC

L1-L

Pum

aN

oxa

Bim Bid

HR

KBA

DBM

FBI

KN

IXN

IP3

MAP

1Ba

kBa

x1Ba

x2Bc

l-Rm

bBc

l-GM

CL1

-SN

IAP

IAP1

IAP2

XIAP

SUR

VIVI

NAP

OLL

OLI

VIN

APAF

APAF

-XL

PI-9

Flip

AIF

DIA

BLO

PAR

NB2

MFL

TG

USR

elat

ive

Expr

essi

on (%

) HDRA PBRA SF

**

Bcl-2-like BH3-only Bax-like IAP Miscellaneous HKG

0.0

0.5

1.0

1.5

2.0

2.5

Noxa Bcl-2 Bcl-XL Mcl-1 Bim

Fold

Exp

ress

ion

toH

D

** HD

RA PBRA SF

Figure 7. Expression of pro- and anti-apoptotic proteins is altered in RA SF T lymphocytes. A, MLPA analysis of gene expression in mRNA from HD PB (n=4), RA PB (n=3) and SF T lymphocytes (n=3). The results were calculated and expressed as percentage of the total signal (relative expression) of all genes examined. Mean values and SEM are indicated. * p < 0.05. B, Expression of selected genes pre-sented as the mean and SEM relative to HD PB T lymphocytes. Relative expression of genes in HD PB T cells has been normalized to a value of 1. C, Western blotting of whole cell lysates of freshly isolated HD PB, RA SF and RA PB T lymphocytes with antibodies against Noxa, Bcl-2, Bcl-XL, Mcl-1 and actin. The arrow indicates Mcl-1 expression. One of three representative independent experiments is shown. D, Western blotting of RA PB and RA SF T cell lysates as in C, using antibodies against Bcl-2, Bim and ERK.

DCB

A

51


Mcl-1, which antagonizes Noxa-induced apoptosis, was only detectable in RA PB T cells, but not HD PB or RA SF T cells. Although previous analysis of RA SF T cells by intracellular FACS staining identified elevated Bcl-XL expression as a proposed compensatory pro-survival mechanism in RA SF T cells41, we found that Bcl-XL was hardly detectable by western blotting (Figure 7C). Anti-apoptotic Bcl-2 protein ex-pression in RA SF T cells was severely depressed (Figures 7C and 7D), but this occur-red in parallel with decreased expression of the pro-apoptotic Bcl-2 binding partner Bim (Figure 7D).

The ratio of Noxa and Mcl-1 expression in RA SF T cells ex vivo favors apoptosis

Our collective data suggested a link between the susceptibility of RA SF T cells to apoptosis, and the relative expression levels of Noxa versus Mcl-1. However, while the relative balance of expression of these proteins has been previously shown to re-gulate T cell apoptosis under various conditions, including environmental stress, cy-tokine withdrawal, and antigen stimulation43;44, we noted no differences in apoptotic rates of HD PB, RA PB and RA SF T cells immediately post-isolation. Therefore, we performed a comparative analysis of HD PB, RA PB, and RA SF T cell gene expres-sion immediately following isolation, and after 24 hours in culture. Ex vivo culture of HD PB T cells led to a significant down-regulation of Noxa mRNA expression as compared to RA SF T cells (p < 0.05) (Figure 8A). A trend towards down-regulation of Mcl-1 expression was observed in RA SF T cells, while Bcl-XL expression remained comparable in each T cell population after 24 hours (Figure 8A). Paired analysis of the ratio of Noxa expression relative to Mcl-1 in each sample immediately after iso-lation and after 24 hours culture demonstrated that apoptosis in RA SF T cells was associated with an inability of RA SF T cells to down-regulate the Noxa/Mcl-1 ratio (Figure 8B, left panel). Relative expression levels of Bim to Bcl-2, especially in the context of Fas signaling, have recently been demonstrated to regulate T cell survival and prevent autoimmunity in mice45-48, so we therefore assessed potential changes in HD PB, RA PB and RA SF T cell expression of Bim and Bcl-2 over time (Figure 8A). After 24 hours in culture, little differences were observed in the relative expression of Bim. However, significant down-regulation of Bcl-2 expression was observed in HD PB (p < 0.05) and RA PB T cells (p < 0.05), and a similar trend observed in RA SF T cells. Ratios of Bim mRNA expression relative to Bcl-2 increased by a similar degree in HD PB, RA PB and RA SF T cells (Figure 8B, right panel). Thus, within the context of general increases in Bim/Bcl-2 ratios in T cells during culture, the failure of RA SF

52

Chapter 2

T cells to down-regulate Noxa expression relative to Mcl-1 may promote selective apoptosis of this T cell population ex vivo.

Discussion

Here, we demonstrate that at the single cell level, RA SF T lymphocytes produce IL-2 and proliferate as well as, if not more so, than HD and RA PB T cells. Our stu-dies provide evidence that the intrinsic capacity of T lymphocytes to undergo TCR-dependent activation and proliferation remains intact in the SF of RA patients. This finding stands in sharp contrast to currently held notions regarding the cellular and

0.00.20.40.60.81.01.21.4

HD RA PB RA SFFold

exp

ress

ion

to0h

Bcl-XLNoxa

0.00.20.40.60.81.01.2

HD RA PB RA SFFold

exp

ress

ion

to0h *

*Mcl-1

0.00.20.40.60.81.01.2

HD RA PB RA SFFold

exp

ress

ion

to0h *

Bim

0.0

0.5

1.0

1.5

HD RA PB RA SFFold

exp

ress

ion

to0h *

Bcl-20h24h

0.00.20.40.60.81.01.21.4

HD RA PB RA SFFold

exp

ress

ion

to0h * *

0.0

0.2

0.4

0.6

0.8

1.0

HD RA PB RA SF

Rat

io N

oxa/

Mcl

1

* *

010

2030

4050

60

HD RA PB RA SF

Rat

io B

im/B

cl2

0h24h

Figure 8. RA SF T lymphocytes fail to downregulate expression of Noxa relative to Mcl-1 fol-lowing isolation. A, Expression of selected genes from HD PB, RA PB, and RA SF T lymphocytes imme-diately after isolation (0h, white bars) and after 24 hours (24h, black bars) in culture. Values are normalized to 1 for each T cell source at 0h. B, The ratio of relative expression of Noxa/Mcl-1 (left panel) and Bim/Bcl-2 (right panel) in HD PB, RA PB and RA SF T lymphocytes immediately after isolation (0h, white bars) or after 24 hours in culture (24h, black bars). Shown are the means and SEM of 3 independent experiments.

B

A

53


molecular mechanisms underlying our inability to detect T cell cytokine production and proliferation in RA SF and synovial tissue. Initial studies characterizing the res-ponsiveness of RA SF and synovial tissue T cells to TCR triggering, recall antigens, and pharmacological mitogens presented clear evidence of depressed IL-2 produc-tion and proliferation in these lymphocytes (reviewed in2;3). However, the accuracy and reliability of thymidine incorporation measurements, as well as cytokine deter-mination in supernatants, are critically dependent upon the comparison of equiva-lent cell numbers. Our data suggest that previous reports of RA SF TCR signaling defects, at least in terms of proliferation and cytokine production, reflect limitations of the experimental systems used in these studies in accounting for apoptosis, and not the functional capacity of RA SF T cells. Although previous studies have demon-strated that T cells derived from RA synovial tissue, like those derived from SF, are hyporesponsive to TCR and mitogenic triggering, further studies will be needed to reassess the functional capacity of RA synovial TCR signaling in situ.Productive TCR engagement of autoreactive T lymphocytes should readily lead to detectable local T cell cytokine generation and proliferation in RA synovial tissue. However, in established RA, no significant synovial tissue T cell cytokine produc-tion or proliferation has been observed in situ27;28;49, and no significant spontaneous production of IL-2, IFN-g or TNF-α has been observed in freshly isolated RA SF T cells (this manuscript and22-24;49). Spontaneous IL-17 production has been observed in RA synovial tissue T cells50;51, although the presence of spontaneous IL-17 produc-tion in RA SF T cells is controversial49;52. Similarly contradictory reports exist regar-ding whether IL-17 levels are elevated in RA SF compared to disease controls29;36;51;52. The lack of obvious T cell cytokine production and proliferation in established RA has led to the idea that TCR signaling is repressed in the presence of chronic inflam-matory mediators, and that at this stage of disease, synovial T cells contribute to pathology by TCR-independent mechanisms2;3. A number of TCR-independent in vitro model systems have been developed, each of which recapitulates at least some phenotypic characteristics of RA synovial T cells. In both human PB T cells and in murine T cell hybridomas, chronic TNFa or cytokine cocktail exposure results in de-creased CD3z protein expression and concomitant diminished proliferative respon-ses to CD3/CD28 stimulation, yet promotes T cell activation of monocytes by cell-cell contact19;30;33. Additionally, coincubation of RA SF T cells with IL-15 can induce IL-17 production36, while a subset of RA synovial tissue and SF T cells respond to cytokine cocktails by producing IFN-g53. Studies in model systems may thus benefit most from functional readouts distinct from TCR-dependent cytokine production and proliferative responses.

54

The balance between survival and apoptosis in T lymphocytes, under homeostatic or inflammatory conditions, is tightly regulated by expression and post-translational modification of Bcl-2 family proteins. Bcl-2 family proteins known to regulate T cell survival include pro-apoptotic Bax, Bim, NIP3, Noxa, and Puma, and anti-apoptotic Bcl-2, Bcl-XL, and Mcl-154. Specific association of Bim to its anti-apoptotic binding partners, Bcl-2 and Mcl-1, and Noxa binding to Mcl-1, has been demonstrated55. Therefore, shifts in expression levels of these interacting Bcl-2 members are expected to affect the threshold for T cell apoptosis. Although no significant apoptosis of RA synovial T lymphocytes is observed in situ or immediately following isolation from tissue or SF (data presented here and41;42), previous studies indicated that RA syno-vial T cells display a phenotype favoring rapid apoptosis. Using intracellular FACS staining, it was observed that synovial T cells expressed low levels of anti-apoptotic Bcl-241. Under these circumstances, it was proposed that synovial T cell apoptosis in situ was actively suppressed by cell-cell contacts, signaling of IL-2 and IL-15 via the IL-2 receptor common g chain, and/or CD28 costimulation, each of which could enhance Bcl-XL protein expression41;42;56. In our analysis of mRNA expression in RA SF T cells, we did observe a modest decrease in Bcl-2 expression, but no differences in Bcl-XL expression, compared to HD and RA PB T lymphocytes. Consistent with mRNA data, Bcl-2 protein expression was selectively down-regulated in RA SF T cells. Surprisingly, Bcl-XL protein expression in RA SF T cells was hardly detectable, in contrast with previous reports. Although the reason for this discrepancy is un-known, Bcl-XL expression in RA SF T cells was previously assessed by intracellular FACS staining without independent verification by western blotting. mRNA expres-sion of the pro-apoptotic Bcl-2 binding partner Bim was similar in freshly isolated HD PB, RA PB, and RA SF T cells, but at the protein level was depressed in RA SF T cells.During T cell culture ex vivo, we observed a dramatic increase in the ratio of Bim mRNA to that of Bcl-2, although this occurred to a similar degree in HD PB, RA PB, and RA SF T cells and could not be clearly linked with the selective induction of apoptosis in RA SF T cells. However, we did note a significant increase in expression of pro-apoptotic Noxa, which is up-regulated following TCR triggering or IL-7/IL-15 stimulation and determines the apoptosis susceptibility of T cells exposed to envi-ronmental stress43. Also in line with mRNA data, we readily detected elevated Noxa protein expression in RA SF T cells as compared to HD PB T cells. However, Noxa protein was even more elevated in RA PB T lymphocytes, although apoptosis of these cells ex vivo was no greater than observed in HD PB T lymphocytes. The rea-

Chapter 2

55

son for this inconsistency may lie in the observation that RA PB T cells, unlike their SF counterparts, also express elevated levels of Mcl-1, protecting them from Noxa-mediated apoptosis. In murine T cells, Mcl-1 plays a general role in protecting T cells against apoptosis during development, activation, and differentiation44. While HD and RA PB T cells quickly down-regulated transcription of Noxa ex vivo, this process was delayed in RA SF T cells, leading to a persistently increased Noxa/Mcl-1 ratio. This specific failure to decrease Noxa/Mcl-1 ratios in RA SF T cells, in com-bination with general increases of Bim/Bcl-2 ratios in T cells during culture, might push RA SF T cells into apoptosis. In CLL, increases in the ratio of Noxa/Mcl-1 when Bcl-2 expression is limited also drive cellular apoptosis57.Although numerous studies noting the clinical efficacy of abatacept in RA are now available, there are no reports describing the direct effects of this compound on T cell function in RA4. At least implicitly, clinical benefits are usually interpreted in terms of the ability of abatacept to block requisite costimulatory signaling of CD28 during the TCR-dependent activation of autoreactive T cells. Given that TCR sig-naling is intact in SF T cells, it will be of interest to determine if abatacept exerts its effects through the inhibition of rare TCR-dependent activation events, currently below our detection threshold, or suppresses inflammation in RA by alternative me-chanisms, such as reverse signaling to CD80/86-expressing antigen presenting cells and FLS58, direct targeting of RA synovial and SF T cells, which abundantly express CD80/8659;60, or effects on peripheral mononuclear cell populations. Continued ef-forts to understand the molecular mechanisms by which abatacept achieves clinical efficacy in RA may identify additional immune-mediated inflammatory diseases to which this compound might be applied therapeutically.

Acknowledgments

We would like to thank M. E. Sanders for providing technical assistance, Dr. M. Tanck for providing expertise in statistical analyses, and Dr. R. van Lier for helpful discussions and critical reading of the manuscript.


56

Chapter 2

References

Lundy SK, Sarkar S, Tesmer LA, Fox DA. Cells of the synovium in rheumatoid 1. arthritis - T lymphocytes. Arthritis Research & Therapy 2007;9:Firestein GS, Zvaifler NJ. How important are T cells in chronic rheumatoid sy-2. novitis?: II. T cell-independent mechanisms from beginning to end. Arthritis Rheum. 2002;46:298-308.Clark J, Vagenas P, Panesar M, Cope AP. What does tumour necrosis factor ex-3. cess do to the immune system long term? Ann.Rheum.Dis. 2005;64 Suppl 4:iv70-iv76.Malmstrom V, Trollmo C, Klareskog L. Modulating co-stimulation: a rational 4. strategy in the treatment of rheumatoid arthritis? Arthritis Research & Therapy 2005;7:S15-S20.Winchester R, Dwyer E, Rose S. The Genetic-Basis of Rheumatoid-Arthritis - 5. the Shared Epitope Hypothesis. Rheumatic Disease Clinics of North America 1992;18:761-783.de Vries N, Tijssen H, van Riel PLCM, van de Putte LBA. Reshaping the shared 6. epitope hypothesis - HLA-associated risk for rheumatoid arthritis is encoded by amino acid substitutions at positions 67-74 of the HLA-DRB1 molecule. Arthritis and Rheumatism 2002;46:921-928.Padyukov L, Silva C, Stolt P, Alfredsson L, Klareskog L. A gene-environment 7. interaction between smoking and shared epitope genes in HLA-DR provides a high risk of seropositive rheumatoid arthritis. Arthritis and Rheumatism 2004;50:3085-3092.Gourraud PA, Boyer JF, Barnetche T et al. New classification of HLA-DRB1 al-8. leles differentiates predisposing and protective alleles for rheumatoid arthritis structural severity. Arthritis and Rheumatism 2006;54:593-599.van der Helm-van Mil, Wesoly JZ, Huizinga TWJ. Understanding the gene-9. tic contribution to rheumatoid arthritis. Current Opinion in Rheumatology 2005;17:299-304.Plenge RM, Cotsapas C, Davies L et al. Two independent alleles at 6q23 associa-10. ted with risk of rheumatoid arthritis. Nature Genetics 2007;39:1477-1482.Plenge RM, Seielstad M, Padyukov L et al. TRAF1-C5 as a risk locus for rheu-11. matoid arthritis - A genomewide study. New England Journal of Medicine 2007;357:1199-1209.Raychaudhuri S, Remmers EF, Lee AT et al. Common variants at CD40 and other 12. loci confer risk of rheumatoid arthritis. Nat.Genet. 2008

57


Moreland LW, Alten R, Van den BF et al. Costimulatory blockade in patients 13. with rheumatoid arthritis: a pilot, dose-finding, double-blind, placebo-control-led clinical trial evaluating CTLA-4Ig and LEA29Y eighty-five days after the first infusion. Arthritis Rheum. 2002;46:1470-1479.Kremer JM, Westhovens R, Leon M et al. Treatment of rheumatoid arthritis by se-14. lective inhibition of T-cell activation with fusion protein CTLA4Ig. N.Engl.J.Med. 2003;349:1907-1915.Genovese MC, Schiff M, Luggen M et al. Efficacy and safety of the selective co-15. stimulation modulator abatacept following 2 years of treatment in patients with rheumatoid arthritis and an inadequate response to anti-tumour necrosis factor therapy. Annals of the Rheumatic Diseases 2008;67:547-554.Morimoto C, Romain PL, Fox DA et al. Abnormalities in CD4+ T-lymphocyte 16. subsets in inflammatory rheumatic diseases. Am.J.Med. 1988;84:817-825.Tak PP, Hintzen RQ, Teunissen JJ et al. Expression of the activation antigen CD27 17. in rheumatoid arthritis. Clin.Immunol.Immunopathol. 1996;80:129-138.Cush JJ, Lipsky PE. Phenotypic analysis of synovial tissue and peripheral blood 18. lymphocytes isolated from patients with rheumatoid arthritis. Arthritis Rheum. 1988;31:1230-1238.Brennan FM, Hayes AL, Ciesielski CJ et al. Evidence that rheumatoid arthritis 19. synovial T cells are similar to cytokine-activated T cells: involvement of phosp-hatidylinositol 3-kinase and nuclear factor kappaB pathways in tumor necrosis factor alpha production in rheumatoid arthritis. Arthritis Rheum. 2002;46:31-41.Zhang ZL, Gorman CL, Vermi AC et al. TCR zeta(dim) lymphocytes define 20. populations of circulating effector cells that migrate to inflamed tissues. Blood 2007;109:4328-4335.Smeets TJ, Dolhain RJ, Breedveld FC, Tak PP. Analysis of the cellular infiltrates 21. and expression of cytokines in synovial tissue from patients with rheumatoid arthritis and reactive arthritis. J.Pathol. 1998;186:75-81.Van der Graaff WL, Prins AP, Niers TM, Dijkmans BA, van Lier RA. Quantitation 22. of interferon gamma- and interleukin-4-producing T cells in synovial fluid and peripheral blood of arthritis patients. Rheumatology.(Oxford) 1999;38:214-220.Isomaki P, Luukkainen R, Lassila O, Toivanen P, Punnonen J. Synovial fluid T 23. cells from patients with rheumatoid arthritis are refractory to the T helper type 2 differentiation-inducing effects of interleukin-4. Immunology 1999;96:358-364.Berner B, Akca D, Jung T, Muller FA, Reuss-Borst MA. Analysis of Th1 and Th2 24. cytokines expressing CD4+and CD8+T cells in rheumatoid arthritis by flow cy-

58

Chapter 2

tometry. Journal of Rheumatology 2000;27:1128-1135.Combe B, Pope RM, Fischbach M et al. Interleukin-2 in rheumatoid arthritis: 25. production of and response to interleukin-2 in rheumatoid synovial fluid, syno-vial tissue and peripheral blood. Clin.Exp.Immunol. 1985;59:520-528.Firestein GS, Zvaifler NJ. Peripheral blood and synovial fluid monocyte acti-26. vation in inflammatory arthritis. II. Low levels of synovial fluid and synovial tissue interferon suggest that gamma-interferon is not the primary macrophage activating factor. Arthritis Rheum. 1987;30:864-871.Firestein GS, Xu WD, Townsend K et al. Cytokines in chronic inflammatory 27. arthritis. I. Failure to detect T cell lymphokines (interleukin 2 and interleukin 3) and presence of macrophage colony-stimulating factor (CSF-1) and a novel mast cell growth factor in rheumatoid synovitis. J.Exp.Med. 1988;168:1573-1586.Smeets TJ, Dolhain RJEM, Miltenburg AM et al. Poor expression of T cell-derived 28. cytokines and activation and proliferation markers in early rheumatoid synovial tissue. Clin.Immunol.Immunopathol. 1998;88:84-90.Raza K, Falciani F, Curnow SJ et al. Early rheumatoid arthritis is characterized 29. by a distinct and transient synovial fluid cytokine profile of T cell and stromal cell origin. Arthritis Res.Ther. 2005;7:R784-R795.Cope AP, Londei M, Chu NR et al. Chronic exposure to tumor necrosis factor 30. (TNF) in vitro impairs the activation of T cells through the T cell receptor/CD3 complex; reversal in vivo by anti-TNF antibodies in patients with rheumatoid arthritis. J.Clin.Invest 1994;94:749-760.Maurice MM, Lankester AC, Bezemer AC et al. Defective TCR-mediated signa-31. ling in synovial T cells in rheumatoid arthritis. J.Immunol. 1997;159:2973-2978.Maurice MM, Nakamura H, van der Voort EA et al. Evidence for the role of 32. an altered redox state in hyporesponsiveness of synovial T cells in rheumatoid arthritis. J.Immunol. 1997;158:1458-1465.Isomaki P, Panesar M, Annenkov A et al. Prolonged exposure of T cells to TNF 33. down-regulates TCR zeta and expression of the TCR/CD3 complex at the cell surface. J.Immunol. 2001;166:5495-5507.Gringhuis SI, Leow A, Papendrecht-van der Voort EA et al. Displacement of lin-34. ker for activation of T cells from the plasma membrane due to redox balance alterations results in hyporesponsiveness of synovial fluid T lymphocytes in rheumatoid arthritis. J.Immunol. 2000;164:2170-2179.Foey A, Green P, Foxwell B, Feldmann M, Brennan F. Cytokine-stimulated T 35. cells induce macrophage IL-10 production dependent on phosphatidylinosi-tol 3-kinase and p70S6K: implications for rheumatoid arthritis. Arthritis Res.

59


2002;4:64-70.Ziolkowska M, Koc A, Luszczykiewicz G et al. High levels of IL-17 in rheuma-36. toid arthritis patients: IL-15 triggers in vitro IL-17 production via cyclosporin A-sensitive mechanism. J.Immunol. 2000;164:2832-2838.Remans PH, Wijbrandts CA, Sanders ME et al. CTLA-4IG suppresses reactive 37. oxygen species by preventing synovial adherent cell-induced inactivation of Rap1, a Ras family GTPASE mediator of oxidative stress in rheumatoid arthritis T cells. Arthritis Rheum. 2006;54:3135-3143.Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Associa-38. tion 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988;31:315-324.He X, Janeway CA, Levine M et al. Dual receptor T cells extend the immune 39. repertoire for foreign antigens. Nature Immunology 2002;3:127-134.Eldering E, Spek CA, Aberson HL et al. Expression profiling via novel multi-40. plex assay allows rapid assessment of gene regulation in defined signalling pa-thways. Nucleic Acids Res. 2003;31:e153.Salmon M, Scheel-Toellner D, Huissoon AP et al. Inhibition of T cell apoptosis in 41. the rheumatoid synovium. J.Clin.Invest 1997;99:439-446.Maurice MM, van der Voort EA, Leow A et al. CD28 co-stimulation is intact and 42. contributes to prolonged ex vivo survival of hyporesponsive synovial fluid T cells in rheumatoid arthritis. Eur.J.Immunol. 1998;28:1554-1562.Alves NL, Derks IA, Berk E et al. The Noxa/Mcl-1 axis regulates susceptibility 43. to apoptosis under glucose limitation in dividing T cells. Immunity. 2006;24:703-716.Dzhagalov I, Dunkle A, He YW. The anti-apoptotic bcl-2 family member mcl-1 44. promotes T lymphocyte survival at multiple stages. J.Immunol. 2008;181:521-528.Wojciechowski S, Tripathi P, Bourdeau T et al. Bim/Bcl-2 balance is critical for 45. maintaining naive and memory T cell homeostasis. J.Exp.Med. 2007;204:1665-1675.Hughes PD, Belz GT, Fortner KA et al. Apoptosis regulators Fas and Bim coo-46. perate in shutdown of chronic immune responses and prevention of autoimmu-nity. Immunity. 2008;28:197-205.Hutcheson J, Scatizzi JC, Siddiqui AM et al. Combined deficiency of proapopto-47. tic regulators Bim and Fas results in the early onset of systemic autoimmunity. Immunity. 2008;28:206-217.Weant AE, Michalek RD, Khan IU et al. Apoptosis regulators Bim and Fas func-48.

60

Chapter 2

tion concurrently to control autoimmunity and CD8+ T cell contraction. Immu-nity. 2008;28:218-230.Yamada H, Nakashima Y, Okazaki K et al. Th1 but not Th17 cells predominate in 49. the joints of patients with rheumatoid arthritis. Ann.Rheum.Dis. 2008;67:1299-1304.Chabaud M, Durand JM, Buchs N et al. Human interleukin-17: A T cell-derived 50. proinflammatory cytokine produced by the rheumatoid synovium. Arthritis Rheum. 1999;42:963-970.Kotake S, Udagawa N, Takahashi N et al. IL-17 in synovial fluids from patients 51. with rheumatoid arthritis is a potent stimulator of osteoclastogenesis. J.Clin.In-vest 1999;103:1345-1352.Shahrara S, Huang Q, Mandelin AM, Pope RM. TH-17 cells in rheumatoid 52. arthritis. Arthritis Res.Ther. 2008;10:R93.Sattler A, Wagner U, Rossol M et al. Cytokine-induced human IFN-gamma-se-53. creting effector-memory Th cells in chronic autoimmune inflammation. Blood 2009;113:1948-1956.Alves NL, van Lier RAW, Eldering E. Withdrawal symptoms on display: Bcl-2 54. members under investigation. Trends in Immunology 2007;28:26-32.Chen L, Willis SN, Wei A et al. Differential targeting of prosurvival Bcl-2 pro-55. teins by their BH3-only ligands allows complementary apoptotic function. Mol.Cell 2005;17:393-403.Wu LX, La Rose J, Chen L et al. CD28 regulates the translation of Bcl-xL via 56. the phosphatidylinositol 3-kinase/mammalian target of rapamycin pathway. J.Immunol. 2005;174:180-194.Hallaert DYH, Spijker R, Jak M et al. Crosstalk among Bcl-2 family members in 57. B-CLL: seliciclib acts via the Mcl-1/Noxa axis and gradual exhaustion of Bcl-2 protection. Cell Death and Differentiation 2007;14:1958-1967.Boasso A, Herbeuval JP, Hardy AW, Winkler C, Shearer GM. Regulation of indo-58. leamine 2,3-dioxygenase and tryptophanyl-tRNA-synthetase by CTLA-4-Fc in human CD4+ T cells. Blood 2005;105:1574-1581.Azuma M, Yssel H, Phillips JH, Spits H, Lanier LL. Functional expression of B7/59. BB1 on activated T lymphocytes. J.Exp.Med. 1993;177:845-850.Verwilghen J, Lovis R, De Boer M et al. Expression of functional B7 and CTLA4 60. on rheumatoid synovial T cells. J.Immunol. 1994;153:1378-1385.

3Sustained T cell Rap1 sig-

naling is protective in the collagen-induced arthritis

model of rheumatoid arthritis

Joana RF Abreu1, Sarah Krausz1, Wendy Dontje1, Aleksander M Grabiec1, Daphne deLaunay1, Martijn A

Nolte2, Paul P Tak1 and Kris A Reedquist1

1Division of Clinical Immunology and Rheumatology and 2De-partment of Experimental Immunology, Academic Medical Center,

University of Amsterdam, Amsterdam, The Netherlands

Altered version submitted for publication

65

T cell Rap1 activation prevents collagen-induced arthritis

Abstract

The small GTPase Rap1 plays a critical role in T lymphocyte trafficking and integrin-dependent adhesion to antigen presenting cells. Genetic inactivation of Rap1 in mu-rine T cells leads to the accumulation of potentially auto-reactive T cells in lymphoid compartments, and Rap1 signaling is blocked in synovial T cells from rheumatoid arthritis (RA) patients. As these data imply that Rap1 inactivation might contribute to T cell-mediated autoimmunity, we examined the consequences of maintaining T cell Rap1 activation in an experimental murine model of rheumatoid arthritis. We find that disease incidence and severity of collagen-induced arthritis (CIA) is drama-tically reduced in mice expressing active RapV12, an active mutant of Rap1, in T cells. Protection against pathology in CIA is accompanied by defective TNF-α production in CD8+ T cells, reminiscent of clonal T cell exhaustion observed during chronic viral infection. Additionally, B cell immunoglobulin class switching of auto-antibodies is diminished in RapV12 mice. Defects in qualitative T cell immune responses during CIA are paralleled by inadequate up-regulation of the T cell co-stimulatory molecu-les ICOS and CD40L. Our results suggest that modulation of T cell Rap1 signaling may be beneficial in attenuating pathologic contributions of T cells to RA and other immune-mediated inflammatory diseases.

66

Chapter 3

Introduction

T cells contribute to synovitis and joint destruction in rheumatoid arthritis (RA), a prototype immune-mediated inflammatory disease, through the pleiotropic acti-vation of macrophages and synovial stromal cells via cell-cell contacts and inter-leukin-17 production, the stimulation of B cells producing autoimmune antibodies, and the promotion of osteoclast differentiation1. An active role for the T cell antigen receptor (TCR) in the initiation and perpetuation of disease in RA is suggested by associations between expression of specific MHC HLA-DR1 and DR4 epitope alleles with enhanced disease risk and disease severity in RA patients2. T cells derived from RA synovial tissue or synovial fluid (SF) display characteristics suggestive of recent TCR stimulation, including surface expression of CD45RO, CD69, CD154, HLA-DR, ICOS and VLA-4 proteins1;3. These T cells are primarily pro-inflammatory Th1 and Th17 cells, and display hyperresponsive cytokine responses to TCR/CD28 stimula-tion. In established RA, inflammatory cytokines present in the synovial tissue, such as IL-6, IL-12, IL-8 and TNF-α, rather than antigen stimulation, may drive T cell contributions to the perpetuation of inflammation4;5. Although the molecular mechanisms underlying altered T cell function in RA are unknown, recent studies have indicated that inactivation of the small GTPase Rap1 may contribute to the pathogenic behavior of T cells in the synovium6-8. TCR stimu-lation results in the activation of guanine nucleotide exchange factors, such as C3G and CalDAG-GEFs which promote accumulation of Rap1 in an active GTP-bound form9;10. TCR-dependent Rap1 activation is exquisitely sensitive to costimulatory sig-nals provided by antigen-presenting cells (APCs) such as CD28, which acts through the Rap1 GTPase activating protein RapGAP1 and suppresses Rap1 activation10;11. Conversely, CTLA-4 ligation, which opposes CD28 signaling, promotes accumulati-on of GTP-bound Rap112;13. Once activated, Rap1 regulates several distinct signaling pathways predicted to contribute to the quality of T cell immune responses in vivo. Activation of Rap1 by TCR ligation, chemokines and adhesion molecules promotes remodeling of the cytoskeleton and integrin activation, needed for T cell trafficking and adhesion to APCs14. Additionally, under certain experimental conditions, Rap1 suppresses TCR-dependent ERK activation and IL-2 production, either directly, through blocking Ras-dependent Raf kinase activation9;11, or indirectly, through di-minishing TCR-dependent reactive oxygen species production6. Genetic manipulation of Rap1 signaling pathways in vivo has provided further evi-dence that the activation status of Rap1 in T cells can have qualitative effects on immune responses. Mice deficient for Spa-1, a RapGAP expressed in T cells, demon-

67

strate age-dependent defects in T cell proliferative responses to stimulation by the TCR, mitogens, and recall antigen15. Transgenic mice expressing an active Rap1E63 mutant in T cells exhibit defects in both primary and secondary T cell proliferative responses, as well as defects in B cell immunoglobulin (Ig) class switching16. Defec-tive T cell responses in these mice is attributed to both suppression of ERK activa-tion in effector cells, as well as increases in the frequency and functional capacity of CD103-expressing regulatory T cells (Tregs). Conversely, transgenic expression of RapGAP1 in T cells, suppressing Rap1 function, leads to an accumulation of T cells in lymph nodes17. These T cells express high levels of CD69 under homeostatic con-ditions, possibly reflecting an autoreactive phenotype. Alternatively, accumulation of activated T cells in lymph nodes of RapGAP1 transgenic mice may indicate de-fects in T cell trafficking, a phenotype displayed in Rap1A knockout mice18;19. Despite these studies, it is still unclear if Rap1 T cell functions are relevant to immune responses in human disease. We have previously demonstrated that Rap1 activation is suppressed in T cells obtained from the synovium of RA patients, likely a result of CD28-dependent interactions with synovial APCs6;7. Moreover, suppressed Rap1 function in RA synovial T cells was associated with enhanced T cell ROS produc-tion and TCR-dependent cytokine responsiveness, indicating that defects in Rap1 signaling may contribute to T cell-dependent pathology in RA7;8. To determine if maintenance of T cell Rap1 signaling might protect against autoimmunity, we here examined the effects of transgenic expression of active Rap1 on pathogenesis in the murine collagen-induced model of RA (CIA). We find that transgenic expression of an active mutant of Rap1, RapV12, in murine T cells potently suppresses disease incidence and severity. Protection against disease was not associated with general defects in lymphocyte trafficking or immune responsiveness, but rather specific de-fects in TNF-α production by CD8+ T cells and ineffective upregulation of the costi-mulatory proteins ICOS and CD154 on T helper cells, needed for Ig class-switching by autoimmune B cells. Indeed, production of autoantibodies was also reduced in RapV12 transgenic mice. These results suggest that strategies aimed at enhancing T cell Rap1 function may be beneficial in the treatment of human immune-mediated inflammatory diseases.


68

Chapter 3

Materials and methods

Animals

RapV12 transgenic C57BL/6 mice were kindly provided by Dr. D. Cantrell (Universi-ty of Dundee, Dundee, U.K.)20. RapV12 and WT littermate control mice were housed under conventional conditions at the animal facility of the Academic Medical Center (Amsterdam, The Netherlands). Feeding was ad libitum. The animal ethical commit-tee of the Academic Medical Center approved all experiments.

Cell staining and flow cytometry

Single cell suspensions were obtained from spleen and axial lymph nodes (LN) by grinding tissue over 40 µm cell strainers (BD Biosciences). Erythrocytes were remo-ved using erylysis buffer (155 mM NH4Cl, 10 mM KHCO3, and 1 mM EDTA, pH 7.4). Cells were surface stained with the indicated fluorochrome-conjugated antibo-dies for 30 minutes at 4°C in PBS containing 0.5% BSA. For Foxp3 intracellular stai-nings cells were stained according to manufacturer’s instructions. For assessment of T cell cytokine expression, splenocytes and LN cells were stimulated for 1 hour with PMA (10 ng/ml, Sigma-Aldrich) and ionomycin (1 µM, Sigma-Aldrich) or anti-CD3 (10 µg/ml) and anti-CD28 (4 µg/ml) antibodies (kindly provided by Dr. L. Boon, Bioceros BV, Utrecht). Brefeldin A (10 µg/ml, Sigma- Aldrich) was added for the fi-nal 4 hours of stimulation, and cells were harvested and stained with CD4 and CD8 antibodies. Cells were then fixed and permeabilized using Cytofix/Cytoperm (BD Biosciences) and labeled for intracellular cytokines using specific antibodies listed in Supplemental Materials and Methods. For examination of T cell surface molecules following in vitro activation, splenocytes and LN cells were stimulated for 24 hours in the presence of 10 µg/ml anti-CD3 and 4 µg/ml anti-CD28 antibodies. Surface marker and cytokine expression were monitored using FACSCalibur or Canto flow cytometers (BD Biosciences).

In vitro T cell differentiation

For T cell isolation, splenocytes were incubated 30 min at 4°C with a saturating mix-

69

ture of hybridoma culture supernatants of the following rat anti-mouse antibodies: anti-CD11b (clone M1/70), anti-Ly6G (clone RB6-8C5), anti-CD45R (B220; clone RA3-6B2), anti−MHC class II (clone M5-114) and anti-TER119. After washing, cells were incubated with goat-anti-rat Ig microbeads (Miltenyi) and isolated by MACS. Puri-fied T cells were stimulated with plate-bound anti-CD3 and anti-CD28 antibodies as above for 3 days in 96 well round-bottom plates (Greiner Bio-One, Frickenhausen, Germany) in the presence of 50 U/ml recombinant murine IL-2 (Invitrogen) sup-plemented with 3 ng/ml human TGF-β and 20 ng/ml murine IL-6 (both from R&D Systems) when indicated.

Induction and assessment of CIA

Chicken collagen type II (cCII) (Sigma-Aldrich, St. Louis, MO) was dissolved over-night at 4oC with 0.1 M acetic acid at a final concentration of 2 mg/ml and then mixed with 4mg/ml complete Freund’s adjuvant (CFA) (Chondrex, Inc., Redmond, WA). 10- to 12- week old mice were injected intradermally on day 0 at the base of the tail with 100 µl of the cCII-CFA emulsion (133 µg cCII and 133 ug CFA in a total volume of 100 µl emulsion). The immunization was repeated on day 21, and animals monitored 3 times weekly through sacrifice at day 60. Methodologies for the scoring of arthritis severity, paw swelling, synovial infiltration, cartilage erosion and radiological damage, as well as determination of serum anti-cCII antibodies, are described in detail in Supplemental Materials and Methods.

Statistical Analysis

Statistical significance was determined using a two tailed Student’s t-test. P values < 0.05 were considered statistically significant.

Supplemental Materials and Methods

Assessment of CIA disease scores and paw swelling

Arthritis severity was assessed in a blinded manner, using a semi-quantitative sco-


70

Chapter 3

ring system (0 to 4): 0, normal; 1, redness and/or swelling in one joint; 2, redness and/or swelling in more than one joint; 3, redness and/or swelling in the entire paw; and 4, deformity and/or ankylosis. Hind paw ankle joint swelling was measured using a dial caliper (POCO 2T 0- to 10-mm test gauge; Kroeplin Längenmesstechnik, Schlüchtern, Germany).

Histological analysis

Animals were sacrificed and hind paws fixed in 10% buffered formalin for 48 hours, decalcified in 15% ethylenediaminetetraacetic acid, and embedded in paraffin. Tis-sue sections (5 µm sagittal serial sections) were stained with haematoxylin and eo-sin. Inflammation was graded on a scale from 0 (no inflammation) to 3 (severely in-flamed joint) based on infiltration by inflammatory cells in the synovium. Cartilage erosions were scored using a semi-quantitative scoring system from 0 (no erosions) to 3 (extended erosions and destruction of bone).

Radiological analysis

Joint destruction of hind paws was analyzed by radiological assessment of x-rays by two observeres blinded to mouse genotype, using a semi-quantitative score (0-4): 0, no damage; 1, minor bone destruction observed in one enlightened spot; 2, moderate changes, two to four spots in one area; 3, severe erosions afflicting the joint; and 4, complete destruction of the joints.

Antibodies for FACS analysis

The following antibodies from eBiosciences were used: anti-CD3 -FITC, -APC, -Alexa 700; anti-CD8 -PE, -FITC, -Alexa 750; anti-CD69-PE; anti-CD44 -FITC; anti-CD62L -APC; anti-CD103 -PE; anti-Foxp3 -PE, -APC; IL-17 –Alexa 488; anti-TNF-α -APC and anti-IFN-γ - APC, -PerCP Cy5.5. Anti-CD4 –FITC, -PE, -PECy7; anti-CD25 –FITC; anti-TNF-α –PE; anti-IL-4 –PE and anti-IL-10 –PE were purchased from BD Biosciences.

71

Determination of anti-collagen antibodies by ELISA

Maxisorb 96-well plates (Nunc, Roskilde, Denmark) were coated with 5 µg/ml of chicken collagen type II (cCII) (Sigma-Aldrich, St. Louis, MO) in 0.1 M sodium carbonate buffer (pH 9.7) overnight at 4°C. After blocking for 1 h with 2% milk in phosphate-buffered saline (PBS) at room temperature (RT), sera were added at an initial dilution of 1/100 in 2% milk in PBS and 1/3 serial dilutions, and incubated overnight at 4°C. Plates were subsequently washed and incubated with 1 µg/ml biotinylated rat anti-mouse Ig (Southern Biotechnology Associates, Birmingham, Alabama) of the indicated isotype in 2% milk in PBS for 1 hour at RT. After wa-shing, plates were incubated with streptavidin-conjugated alkaline phosphatase (AP) (Jackson ImmunoResearch, Newmarket, Suffolk, UK) for 1 hour at RT, washed, and developed with p-Nitrophenyl Phosphate (pNPP) substrate (Sigma-Aldrich, St Louis, MO). The reaction was stopped with 2M H2SO4, and the optical density (OD) was measured at 415 nm.

Results and discussion

In RA, a block in synovial T cell Rap1 activation is associated with their pathogenic behavior6-8. To determine if maintenance of T cell Rap1 signaling might limit in-flammation and joint destruction in an experimental model of RA, we examined the influence of T cell-specific expression of active RapV12, driven by the human CD2 promoter, in murine CIA20. We chose to examine RapV12 transgenic mice for several specific reasons. RapV12 is expressed in T cells of these mice at levels equivalent to endogenous Rap1A, and Rap1 activity contributed by the transgene product is similar in magnitude to that obtained by PMA/I-stimulated endogenous Rap1A. Although RapV12 T cells have enhanced integrin function, no defects are observed in TCR-dependent ERK activation or proliferative responses, and unlike Rap1E63 and RapGAP1 transgenic mice, RapV12 mice have no obvious alterations in T cell homeostasis12;16;20. Finally, unlike the active Rap1E63 mutant, RapV12 can still cycle between active and inactive states, albeit at a highly reduced rate21. We therefore reasoned that responses of RapV12 mice in CIA might most closely mimic therapeu-tic interventions aimed at restoring the function of endogenous T cell Rap1.As phenotypes of genetically modified mice can be strongly influenced by differen-ces in housing conditions, we examined thymocyte development and T cell matu-ration in RapV12 mice housed in our facilities. Consistent with previous studies20,


72

Chapter 3

we observed no differences between RapV12 mice and wild-type (WT) littermates in terms of percentages of double negative (DN), double positive (DP), or CD4 and CD8 single positive thymocytes (Supplementary Figure 1A). Additionally, within the DN thymocyte compartment, RapV12 failed to influence CD44 and CD25 expression (Supplementary Figure 1B), as previously described20. As thymocyte development and T cell maturation appeared normal in RapV12 mice, we extended our analyses to peripheral T cell compartments. CD4+ and CD8+ T lymphocytes were represented at normal percentages in both spleens and lymph nodes (LN) of RapV12 mice (Fi-gure 1A). RapV12 failed to influence the activation status of splenic and LN T cells under homeostatic conditions, as assessed by CD25 and CD69 staining (Figure 1B). Normal proportions of naive, effector/memory (EM), and central memory (CM) T cells were also observed (Figure 1C). Moreover, we detected no influence of RapV12 on the frequency of CD4+FoxP3+ Tregs or activated CD4+ CD103+ T cells (Figure 1D). Additionally, RapV12 T cells were capable of producing similar amounts of IL-2, TNF-α, IFN-γ and IL-17 following TCR/CD28 in vitro stimulation (Figure 2A). To rule out the possibility that RapV12 may regulate the induction of peripheral Tregs and/or Th17 cells, we isolated T cells from the spleens of healthy WT and RapV12

Spleen LN Spleen LN

CD4 CD80

5

10

15

20

% P

ositi

ve c

ells

CD4 CD80

10

20

30

40

% P

ositi

ve c

ells

CD4 CD4 CD80

4

8

12

16

% P

ositi

ve c

ells

CD25 CD69 CD69CD4 CD4 CD8

0

4

8

12

16

% P

ositi

ve c

ells

CD25 CD69 CD69

WTRapV12

WTRapV12

Spleen LN

01020304050607080

% P

ositi

ve c

ells

N EM CM0

10

20

30

40

50

60

% P

ositi

ve c

ells

N EM CM Foxp3 CD1030

5

10

15

20

25

30

% P

ositi

ve c

ells

WTRapV12

Spleen

Foxp3 CD10305

1015202530354045

% P

ositi

ve c

ells

LNWTRapV12

DC

BA

Figure 1. Peripheral T cell homeostasis is normal in RapV12 Tg mice. (A) Expression levels of CD4+ and CD8+ T cells, (B) CD25 and CD69 cell surface makers on CD4+ and on CD4+ and CD8+ T cells, respectively, and (C) expression levels of naive CD44-CD62L+ (N), effector memory CD44+CD62L- (EM), and central memory CD44+CD62L+ (CM) CD4+ T cells, present in the spleen and LN of WT or RapV12 Tg mice (n=3). (D) Foxp3 and CD103 expression in CD4+ T cells derived from spleens (Foxp3 n=13; CD103 n=10) and LN (Foxp3 n=10; CD103 n=8) of 8-10 weeks old animals. Values are expressed as mean ± SEM.

73

mice. In vitro stimulation of these cells with TGF-β induced a robust increase in FoxP3+ T cell numbers, equivalent to that observed in WT T cells (Figure 2B). RapV12 Th17 cells could also be induced in vitro at frequencies equivalent to those observed in WT mice (Figure 2C). Thus, unlike RapGAP1 transgenic, Rap1E63 transgenic and Spa-1 knockout mice, RapV12 mice display no apparent alterations in the T cell com-partment. We concluded from these results that potential differences in responses of RapV12 mice in CIA would not be secondary to altered T cell homeostasis, but rather, specific influences of active Rap1 in this disease model.Following induction of CIA, all WT mice developed clinical signs of disease wit-hin two weeks (Figure 3A). Remarkably, only 20% of the RapV12 Tg mice develo-ped arthritis, while the rest remained disease free until the end of the experiment. Throughout the experiment, clinical arthritis scores of RapV12 mice were drama-tically lower than those of WT mice (P < 0.0005) (Figure 3B), and paw swelling in RapV12 mice was almost completely suppressed compared to WT mice (P < 0.005) (Figure 3C). Histologic analysis of murine hind paws by hematoxylin staining re-

Foxp3

medium TGF-β TGF-β+IL60

10

20

30

40

% P

ositi

ve c

ells

WTRapV12

IL-17

medium TGF-β TGF-β+IL60

1

2

3

4

% P

ositi

ve c

ells

WTRapV12

0

5

10

15

20

25

30

35

% P

ositi

ve c

ells

medium CD3/CD28CD4 CD8 CD4 CD8

0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

% P

ositi

ve c

ells


0

4

8

12

16

% P

ositi

ve c

ells


IL-2 TNF-α IFN-γ

Figure 2. RapV12 Tg mice have normal cytokine production and induction of regulatory and Th17 T cells in vitro. (A) Splenocytes from WT and RapV12 mice were stimulated for 5h with anti-CD3/CD28 and analyzed for the intracellular expression of IL-2, TNF-α and IFN-γ (n=3). (B,C) T cells were isolated from spleens and stimulated with anti-CD3 and anti-CD28 antibodies for 3 days in the presence of the indicated cytokines. (B) Percentage of Foxp3+ expressing and (C) IL-17 producing T cells induced in the different conditions. Values are expressed as mean ± SEM (n=3).

CB

A


74

Chapter 3

vealed that unlike WT mice, RapV12 mice displayed an almost complete absence of joint infiltration by white blood cells (Figure 3D). This was confirmed by semi-quan-titative analysis of synovial cell infiltration (Figure 3E) (P < 0.001). RapV12 mice also escaped destruction of cartilage (Figures 3E) (P < 0.001) and erosive bone damage, as determined by radiology, which was reduced by approximately 75% compared to WT mice (Figure 3F) (P < 0.001). To better characterize how sustained Rap1 activation in T cells protected against disease induction, we sacrificed mice 42 days after primary immunization, prior to the peak of clinical arthritis in WT mice, and examined splenic and LN T cells. RapV12 mice displayed normal numbers of CD3+CD4+ and CD3+CD8+ T cells (Fi-gure 4A) as well as naive (N) (CD44-CD62L+), effector memory (EM) (CD44+CD62L-) and central memory (CM) (CD44+CD62L+) T cells (Figure 4B). We also examined

RapV12WT

0

2

4

6

8

10

19 24 29 34 39 44 49 54 59 64

Days after immunization

Art

hriti

sSc

ore

* * **

* * * * * * * * * * **

0

20

40

60

80

100

19 24 29 34 39 44 49 54 59 64


Inci

denc

e(%

)

*

0.0

0.5

1.0

1.5

2.0

2.5

WT RapV12

Rad

iolo

gica

l sc

ore

WTRapV12

WTRapV12


*

* * * * * * * * * ** *

*

0.0

0.2

0.4

0.6

0.8

1.0

19 24 29 34 39 44 49 54 59 64Del

ta P

aw S

wel

ling

(mm

)

WTRapV12

0.0

0.51.0

1.52.0

2.53.0

3.5

Cellular Infiltrate

Cartilage Erosion

n.d. n.d.

Sem

i-qua

ntita

tive

scor

e

* *WTRapV12

E

CBA

D F

Figure 3. RapV12 Tg mice are protected from arthritis. WT (open squares) and RapV12 mice (black squares) were immunized with cCII in CFA and boosted on day 21. (A) Hind paw swelling and inflam-mation of the four limbs were monitored for each mouse during the course of disease and the percentage of affected animals (incidence) as well as their (B) disease scores were plotted at the indicated times. (C) Paw swelling was measured during the course of disease using calipers. Delta hind paw swelling was calculated by subtracting the paw diameter before onset of disease from the measured diameter. (D) Representative pictures of paraffin-embedded sections of hind paws stained with haematoxylin and eosin analyzed for (E) cellular infiltration and cartilage erosion. (F) X-rays from hind paws scored for bone damage. Values are depicted as mean ± SEM (n=10) and are representative of two independent experiments. * P < 0.05.

75

potential influences of RapV12 on the Treg compartment. However, no significant differences in the numbers of FoxP3+ T cells or CD4+CD103-CD62L+ naive T regs and CD4+CD103+CD62L- activated/EM Treg subsets were observed (Figure 4C).We next examined the quality of T cell responses in WT and RapV12 mice during disease onset. Splenocytes and LN cells were harvested from mice at day 42 and re-stimulated in vitro with PMA/I. Production of Th1 and Th17 cytokines involved in the pathology of CIA was measured by intracellular staining and FACS analysis. No differences in the number of CD4+ and CD8+ T cells producing IFN-γ were observed between RapV12 and WT mice, either in the spleen or LN (Figure 5A). Remarkably, hardly any CD8+ T cells producing TNF-α were detected in RapV12 LN compared to WT mice (P < 0.05) (Figure 5B). A similar trend toward reduced TNF-α production was also observed in RapV12 LN CD4+ T cells, but did not reach statistical signifi-cance. In contrast, similar numbers of TNF-α-producing T cells were detected in WT and RapV12 splenocytes. Depressed TNF-α production by LN CD8+ T cells from arthritic RapV12 mice appeared to be dependent on in vivo inflammatory con-

0

10

20

30

40

0

2

4

6

8

10

0

1

2

3

4

5

0

2

4

6

8

10

CD4+ CD8+

Cel

l num

bers

(106

)C

ell n

umbe

rs (1

06)

Spleen

Foxp3

Spleen LN

CD4+ CD8+

Cel

l num

bers

(106

)

LN

Foxp3 - Spleen

CD103-

CD62L+CD103+

CD62L-

Cel

l num

bers

(106

)

N EM CM

LN

N EM CM

Spleen

Foxp3 - LN

CD103-

CD62L+CD103+

CD62L-

0

5

10

15

Cel

l num

bers

(106

)

0

2

4

6

8

10

0

1

2

3

4

Cel

l num

bers

(106

)

Cel

l num

bers

(106

)

WTRapV12

WTRapV12

WTRapV12

Figure 4. RapV12 Tg mice have normal numbers of effector and regulatory T cells. Spleens and LN were collected from mice 42 days after the induction of arthritis. (A) Absolute numbers of CD3+CD4+ and CD3+CD8+ T cell subsets and (B) absolute numbers of naive CD44-CD62L+ (N), ef-fector memory CD44+CD62L- (EM) and central memory CD44+CD62L+ (CM) CD4+ T cells, present in the spleen and LN of WT or RapV12 Tg mice. (C) Absolute numbers of regulatory Foxp3+CD4+ T cells and of Foxp3+CD103-CD62L+ and Foxp3+CD103+CD62L- T cells. Values are depicted as mean ± SEM (WT n=5, RapV12 Tg n=4).

C

BA


76

Chapter 3

ditions, as no such defect was observed in T cells isolated from the LN of healthy mice (data not shown). Defects in RapV12 LN T cell TNF-α production were highly selective, as in addition to IFN-γ, no differences were observed between WT and RapV12 mice in regards to IL-17 (Figure 5C) or IL-10 production (Figure 5D). Thus, qualitative analysis of arthritic WT and RapV12 mice T cell responses indicates that the protective effects of RapV12 are not a result of generalized T cell anergy or un-responsiveness, but rather a selective defect in T cell TNF-α production.As previous studies have demonstrated that chronic T cell Rap1 activation can sup-press B cell antibody production and Ig class switching15;16, we examined anti-chic-ken collagen type II (c-CII) production in arthritic WT and RapV12 mice. We found a significant reduction in the serum levels of anti-cCII IgG2a and IgG2b in RapV12 mice relative to WT mice (relative IgG2a: WT 100% ± 6.4, RapV12 60.5% ± 8.5; P < 0.005; relative IgG2b: WT 100% ± 6.7, RapV12 65.8% ± 9.7; P < 0.05) (Figure 6A). In Rap1E63 transgenic mice and Spa-1 knockout mice, defects in Ig class switching are associated with increased Treg function and accumulation of anergic CD44high CD4+

T cells, respectively15;16. However, we did not observe these altered T cell phenoty-pes in arthritic RapV12 mice (Figures 4B and 4C).

0

1

2

3

0.0

0.1

0.2

0.3

0.4

0.5

0.0

0.5

1.0

1.5

0

1

2

3

4

CD4+ CD8+

Cel

l num

bers

(106

)C

ell n

umbe

rs (1

06)

CD4+ CD8+

Cel

l num

bers

(106

)

Cel

l num

bers

(106

)C

ell n

umbe

rs (1

06)

Cel

l num

bers

(106

)

CD4+ CD8+ CD4+ CD8+

0

2

4

6

8

10

CD4+ CD8+ CD4+ CD8+

*WTRapV12

WTRapV12

IFN-γ - Spleen IFN-γ - LN TNF-α - Spleen TNF-α - LN

IL-17 - Spleen IL10 - Spleen

Cel

l num

bers

(106

)

0.0

0.1

0.2

CD4+ CD8+

WTRapV12

IL10 - LN

Cel

l num

bers

(106

)

CD4+ CD8+

WTRapV12

IL-17 - LN

0.0

0.1

0.2

0.3

0

2

4

6

8

10

DC

BA

Figure 5. RapV12 Tg mice have defective TNF-α production. (A) Intracellular expression of IFN-γ, (B) TNF-α, (C) IL-17 and (D) IL-10 cytokines in CD3+CD4+ and CD3+CD8+ T cells upon 5h PMA/I stimulation (WT n=5, RapV12 Tg n=4). Values are depicted as mean ± SEM. * P < 0.002.

77

Recent studies have demonstrated essential roles for the T cell costimulatory proteins ICOS and CD40L expressed by follicular Th cells in effective B cell activation and Ig class switching22. ICOS, a CD28-related protein which is upregulated on activated T cells and provides costimulatory signals distinct from CD28, is required for effective T cell-dependent immune responses23. ICOS may contribute to Ig class switching directly, or indirectly by promoting CD40L expression on T cells24. CD40L binding to B cell CD40 in turn induces proliferation, Ig production, isotype switching and upre-gulation of B cell costimulatory molecules, important for T cell activation. Blockade of ICOS signaling in murine CIA or knock down of ICOS renders mice resistant to disease, with decreased anti-CII antibodies and decreased T cell cytokine produc-tion25;26. Notably, disruption of ICOS signaling in CIA results in defective follicular

ICO

SC

D40

L

WT RapV12

CD69

CD4+ CD8+CD4+ CD8+

CD4 CD80

20

40

60

80

100

Rel

ativ

eC

D69

exp

ress

ion

(%)

CD4 CD80

20

40

60

80

100

Rel

ativ

eC

D40

L ex

pres

sion

(%)

* *

CD4 CD80

20

40

60

80

100

Rel

ativ

eIC

OS

expr

essi

on(%

)

* * WTRapV12

IgG1 IgG2a IgG2b IgM0.0

0.5

1.0

1.5

2.0

2.5

Ave

rage

OD

* *

WTRapV12

Figure 6. RapV12 Tg mice have defective anti-collagen antibody production. (A) Serum from WT (white bars) or RapV12 Tg mice (black bars) was collected at day 60 and the levels of specific anti-collagen IgG detected (n=10). Represen-ted values (mean ± SEM) were obtained within linear regions of the serum dilution curve. *P < 0.05. (B, C) T cells were isola-ted from spleens and stimulated with anti-CD3 and anti-CD28 antibodies for 24 hours, followed by analysis of the expression le-vels of CD69, ICOS and CD40L surface molecules Values are ex-pressed as mean ± SEM (n=5).C

B

A


78

Chapter 3

Th function27. Interference with CD40L signaling in CIA also blocks development of disease and production of anti-CII antibodies28. To examine if RapV12 expression may influence activation-dependent expression of proteins required for functional interactions with B cells, we isolated splenocytes and LN cells from non-arthritic WT and RapV12 mice. Cells were stimulated for 24 hours in the presence of activa-ting TCR/CD28 antibodies, and T cells analyzed for expression of CD69, ICOS, and CD40L. In line with previous studies, RapV12 had no influence on Ras-dependent expression of CD69 (Figure 6B and 6C)20. In contrast, TCR/CD28-dependent up-regulation of both ICOS and CD40L (Figure 6B and 6C) was specifically suppressed in both CD4+ and CD8+ RapV12 T cells.ICOS and CD40L signaling are required for induction of pathology in murine mo-dels of arthritis, and are also thought to contribute to T cell pathogenic behavior in RA. RA synovial T cells have increased ICOS expression compared to disease controls29. Several studies have shown the ability of ICOS signaling to enhance CD8 effector T cell responses. Importantly, in vivo studies have shown that ICOS deli-vers important signals for TNF-α production30. In this context, decreased TNF-α production in RapV12 Tg CD8+ T cells, and to a lesser extent in CD4+ T cells, could be a consequence of decreased ICOS expression. Similarly, CD40L is also frequently expressed on RA synovial T cells31. Although CD40L expression on RA synovial T cells has historically been interpreted as an indication of recent antigen-dependent stimulation, it is now recognized that inflammatory cytokines present in RA syno-vial fluid and the synovium are sufficient to induce T cell CD40L expression3;4. T cell CD40L expression might not only promote auto-antibody production in RA, but also promote antigen-independent, cell contact-dependent activation of macropha-ges and stromal cells32-34.Decreased activation-dependent expression of ICOS and CD40L on RapV12 T cells offers a possible explanation for the protective effects of T cell Rap1 activation in CIA. However, the magnitude of disease resistance in RapV12 mice suggests that other mechanisms may be at play. Previous studies have predicted that Rap-dependent effects on T cell trafficking and anergy induction might influence T cell-dependent responses in vivo. However, although Rap1 regulates T cell trafficking in vivo19, we observed no overt changes in effector T cell or Treg numbers in the spleen or LN of RapV12 mice. Our data are also in discordance with other models suggesting that Rap1 promotes T cell anergy. Elevated Rap1 activity has been observed in anergized murine and human T cells9;35. Additionally, Spa-1 knockout mice display an age-de-pendent accumulation of CD44high T cells which are unresponsive to antigen and mi-togen stimulation15. However, in RapV12 T cells, where Rap1 activity approximates

79

that of pharmacological activation of endogenous Rap1, no defects are observed in antigen-dependent IL-2 production or T cell proliferation20. In vitro, we found that RapV12 and WT T cells also produce similar amounts of IL-2, TNF-α, IFN-γ, and IL-17 following TCR/CD28 stimulation. In vivo, RapV12 T cell defects are limited to TNF-α production. Together, these data argue that T cell anergy is not the underly-ing mechanism in regard to protection against CIA. A more likely possibility might be that we are detecting the deletion or exhaustion of specific auto-antigen –specific T cell clonal populations. Elegant studies in vitro have previously illustrated that sustained Rap1 activation can promote Fas-dependent apoptosis of antigen-specific T cells36. Interestingly, the specific loss of TNF-α production by CD8+ cells is also observed during chronic viral infection. LCMV infection of mice results in the clonal exhaustion of antigen-specific CD8+ T cell clones. Here, this process is accompanied by a hierarchal loss of cytokine responses, in which TNF-α defects precede loss of IFN-γ production37;38. A similar phenomenon has been observed in human CD8+ T cells following HIV infection39.Here we show that maintenance of T cell Rap1 activation decreases arthritis inci-dence and severity in the CIA model, and provide the first evidence that T cell Rap1 function is important in the regulation of inflammatory disease. Strategies disrup-ting CD28 co-stimulation, such as CTLA4-Ig therapy, have proven clinical efficacy in the treatment of RA40. The pathogenic behavior of synovial T cells is associated with a block in Rap1 activation, which is dependent upon synovial cell stimulation of CD28 and can be disrupted by CTLA4-Ig6;7. We propose that the development of specific therapies that prevent Rap1 inactivation in RA T cells might be of clinical benefit for RA and other auto-immune diseases driven by improper activation of T cells.

Acknowledgments

We would like to thank Dr. D. Cantrell (University of Dundee, UK) for providing the RapV12 Tg mice, and Dr. R. van Lier and Dr. H. Schuitemaker (our institute) for critical reading of the manuscript. This work was supported in part by a grant from the Dutch Arthritis Association to KAR.


80

Chapter 3

CD4-CD8- CD4+CD8+

% P

ositi

ve c

ells

0

20

40

60

80

CD4+ CD8+ CD44-CD25-CD44+CD25+

% P

ositi

ve c

ells

0

10

20

30

40

50

CD44+ CD25+

WTRapV12

WTRapV12

BA

Supplementary figure 1. Thymocyte development is normal in RapV12 Tg mice. Thymus from WT (white bars) or RapV12 Tg mice (black bars) were collected and analyzed for cell surface expression of (A) CD4 and CD8 and (B) CD25 and CD44 cell surface markers. Values are expressed as mean ± SEM (n=3).

Supplementary figure

81

References

Lundy SK, Sarkar S, Tesmer LA, Fox DA. Cells of the synovium in rheumatoid 1. arthritis - T lymphocytes. Arthritis Research & Therapy 2007;9:Gourraud PA, Boyer JF, Barnetche T et al. New classification of HLA-DRB1 al-2. leles differentiates predisposing and protective alleles for rheumatoid arthritis structural severity. Arthritis and Rheumatism 2006;54:593-599.Clark J, Vagenas P, Panesar M, Cope AP. What does tumour necrosis factor ex-3. cess do to the immune system long term? Ann.Rheum.Dis. 2005;64 Suppl 4:iv70-iv76.Zhang ZL, Gorman CL, Vermi AC et al. TCR zeta(dim) lymphocytes define 4. populations of circulating effector cells that migrate to inflamed tissues. Blood 2007;109:4328-4335.Sattler A, Wagner U, Rossol M et al. Cytokine-induced human IFN-gamma-se-5. creting effector-memory Th cells in chronic autoimmune inflammation. Blood 2009;113:1948-1956.Remans PH, Gringhuis SI, van Laar JM et al. Rap1 signaling is required for sup-6. pression of Ras-generated reactive oxygen species and protection against oxida-tive stress in T lymphocytes. J.Immunol. 2004;173:920-931.Remans PH, Wijbrandts CA, Sanders ME et al. CTLA-4IG suppresses reactive 7. oxygen species by preventing synovial adherent cell-induced inactivation of Rap1, a Ras family GTPASE mediator of oxidative stress in rheumatoid arthritis T cells. Arthritis Rheum. 2006;54:3135-3143.Abreu JR, Grabiec AM, Krausz S et al. The presumed hyporesponsive beha-8. vior of rheumatoid arthritis T lymphocytes can be attributed to spontaneous ex vivo apoptosis rather than defects in T cell receptor signaling. J.Immunol. 2009;183:621-630.Boussiotis VA, Freeman GJ, Berezovskaya A, Barber DL, Nadler LM. Mainte-9. nance of human T cell anergy: blocking of IL-2 gene transcription by activated Rap1. Science 1997;278:124-128.Reedquist KA, Bos JL. Costimulation through CD28 suppresses T cell receptor-10. dependent activation of the Ras-like small GTPase Rap1 in human T lymphocy-tes. J.Biol.Chem. 1998;273:4944-4949.Carey KD, Dillon TJ, Schmitt JM et al. CD28 and the tyrosine kinase lck stimulate 11. mitogen-activated protein kinase activity in T cells via inhibition of the small G protein Rap1. Mol.Cell Biol. 2000;20:8409-8419.Dillon TJ, Carey KD, Wetzel SA, Parker DC, Stork PJ. Regulation of the small 12.


82

Chapter 3

GTPase Rap1 and extracellular signal-regulated kinases by the costimulatory molecule CTLA-4. Mol.Cell Biol. 2005;25:4117-4128.Schneider H, Valk E, da Rocha DS, Wei B, Rudd CE. CTLA-4 up-regulation of 13. lymphocyte function-associated antigen 1 adhesion and clustering as an alterna-te basis for coreceptor function. Proc.Natl.Acad.Sci.U.S.A 2005;102:12861-12866.Menasche G, Kliche S, Bezman N, Schraven B. Regulation of T-cell antigen re-14. ceptor-mediated inside-out signaling by cytosolic adapter proteins and Rap1 ef-fector molecules. Immunol.Rev. 2007;218:82-91.Ishida D, Yang H, Masuda K et al. Antigen-driven T cell anergy and defective 15. memory T cell response via deregulated Rap1 activation in SPA-1-deficient mice. Proc.Natl.Acad.Sci.U.S.A 2003;100:10919-10924.Li L, Greenwald RJ, Lafuente EM et al. Rap1-GTP is a negative regulator of Th 16. cell function and promotes the generation of CD4+CD103+ regulatory T cells in vivo. J.Immunol. 2005;175:3133-3139.Dillon TJ, Carey KD, Wetzel SA, Parker DC, Stork PJ. Regulation of the small 17. GTPase Rap1 and extracellular signal-regulated kinases by the costimulatory molecule CTLA-4. Mol.Cell Biol. 2005;25:4117-4128.Duchniewicz M, Zemojtel T, Kolanczyk M et al. Rap1A-deficient T and B cells 18. show impaired integrin-mediated cell adhesion. Mol.Cell Biol. 2006;26:643-653.Li Y, Yan J, De P et al. Rap1a null mice have altered myeloid cell functions sug-19. gesting distinct roles for the closely related Rap1a and 1b proteins. J.Immunol. 2007;179:8322-8331.Sebzda E, Bracke M, Tugal T, Hogg N, Cantrell DA. Rap1A positively regulates 20. T cells via integrin activation rather than inhibiting lymphocyte signaling. Nat.Immunol. 2002;3:251-258.Brinkmann T, Daumke O, Herbrand U et al. Rap-specific GTPase activating pro-21. tein follows an alternative mechanism. J.Biol.Chem. 2002;277:12525-12531.Reinhardt RL, Liang HE, Locksley RM. Cytokine-secreting follicular T cells sha-22. pe the antibody repertoire. Nat.Immunol. 2009;10:385-393.Yoshinaga SK, Whoriskey JS, Khare SD et al. T-cell co-stimulation through B7RP-23. 1 and ICOS. Nature 1999;402:827-832.McAdam AJ, Greenwald RJ, Levin MA et al. ICOS is critical for CD40-mediated 24. antibody class switching. Nature 2001;409:102-105.Iwai H, Kozono Y, Hirose S et al. Amelioration of collagen-induced arthritis 25. by blockade of inducible costimulator-B7 homologous protein costimulation. J.Immunol. 2002;169:4332-4339.Nurieva RI, Treuting P, Duong J, Flavell RA, Dong C. Inducible costimulator is 26.

83

essential for collagen-induced arthritis. J.Clin.Invest 2003;111:701-706.Hu YL, Metz DP, Chung J, Siu G, Zhang M. B7RP-1 blockade ameliorates autoim-27. munity through regulation of follicular helper T cells. J.Immunol. 2009;182:1421-1428.Durie FH, Fava RA, Foy TM et al. Prevention of collagen-induced arthritis with 28. an antibody to gp39, the ligand for CD40. Science 1993;261:1328-1330.Ruth JH, Rottman JB, Kingsbury GA et al. ICOS and B7 costimulatory molecule 29. expression identifies activated cellular subsets in rheumatoid arthritis. Cytome-try A 2007;71:317-326.Gonzalo JA, Delaney T, Corcoran J et al. Cutting edge: the related molecules 30. CD28 and inducible costimulator deliver both unique and complementary sig-nals required for optimal T cell activation. J.Immunol. 2001;166:1-5.MacDonald KP, Nishioka Y, Lipsky PE, Thomas R. Functional CD40 ligand is 31. expressed by T cells in rheumatoid arthritis. J.Clin.Invest 1997;100:2404-2414.Cho CS, Cho ML, Min SY et al. CD40 engagement on synovial fibroblast 32. up-regulates production of vascular endothelial growth factor. J.Immunol. 2000;164:5055-5061.Foey AD, Feldmann M, Brennan FM. CD40 ligation induces macrophage IL-10 33. and TNF-alpha production: differential use of the PI3K and p42/44 MAPK-pa-thways. Cytokine 2001;16:131-142.Brennan FM, Hayes AL, Ciesielski CJ et al. Evidence that rheumatoid arthritis 34. synovial T cells are similar to cytokine-activated T cells: involvement of phosp-hatidylinositol 3-kinase and nuclear factor kappaB pathways in tumor necrosis factor alpha production in rheumatoid arthritis. Arthritis Rheum. 2002;46:31-41.Morton AM, McManus B, Garside P, Mowat AM, Harnett MM. Inverse Rap1 35. and phospho-ERK expression discriminate the maintenance phase of tolerance and priming of antigen-specific CD4+ T cells in vitro and in vivo. J.Immunol. 2007;179:8026-8034.Katagiri K, Hattori M, Minato N, Kinashi T. Rap1 functions as a key regulator of 36. T-cell and antigen-presenting cell interactions and modulates T-cell responses. Mol.Cell Biol. 2002;22:1001-1015.Wherry EJ, Blattman JN, Murali-Krishna K, van der MR, Ahmed R. Viral persis-37. tence alters CD8 T-cell immunodominance and tissue distribution and results in distinct stages of functional impairment. J.Virol. 2003;77:4911-4927.Fuller MJ, Khanolkar A, Tebo AE, Zajac AJ. Maintenance, loss, and resurgen-38. ce of T cell responses during acute, protracted, and chronic viral infections.


84

Chapter 3

J.Immunol. 2004;172:4204-4214.Appay V, Nixon DF, Donahoe SM et al. HIV-specific CD8(+) T cells produce an-39. tiviral cytokines but are impaired in cytolytic function. J.Exp.Med. 2000;192:63-75.Genovese MC, Schiff M, Luggen M et al. Efficacy and safety of the selective co-40. stimulation modulator abatacept following 2 years of treatment in patients with rheumatoid arthritis and an inadequate response to anti-tumour necrosis factor therapy. Annals of the Rheumatic Diseases 2008;67:547-554.

Antigen receptor and co-stimulatory signals diffe-rentially regulate RapGAP family protein expression in human T lymphocytes

Joana RF Abreu1, Marjolein E Sanders1, Silvia Ariotti1, Tania C Cruz1, Albert P Smolenski2, Paul P Tak1, and Kris

A Reedquist1

1Division of Clinical Immunology and Rheumatology, Academic Medical Center, University of Amsterdam, Amsterdam, The Ne-therlands. 2UCD Conway Institute, University College Dublin,

Dublin, Ireland

Manuscript in preparation

4

89

RapGAP family proteins are differentially regulated during T cell activation

Abstract

The quality of T cell immune responses is exquisitely regulated by the coordinated triggering of the T cell antigen receptor (TCR) and costimulatory proteins, such as CD28-like proteins and TNF family ligands. Costimulatory proteins are differenti-ally expressed depending on the activation and differentiation status of the T cell. Recent independent observations have suggested that T cell costimulatory proteins may exert their effects via Rap1GAP-dependent regulation of Rap1 activation. Ho-wever, little is known regarding the regulation of distinct Rap1GAPs in human T cells during T cell activation and differentiation. Here, we provide evidence that each of the five Rap1GAPs is expressed in human T lymphocytes in both lineage- and activation- dependent manners, regulated at both transcriptional and post-translati-onal levels. Our results indicate that control of Rap1 function is tightly orchestrated during both T cell differentiation and activation, and each of the Rap1GAPs plays a role in this process.

90

Chapter 4

Introduction

Antigen-specific T cells are activated following T cell receptor (TCR) engagement by antigenic peptide complexed to major histocompatibility (MHC) proteins on antigen-presenting cells (APCs). Resultant effector T cell cytokine secretion, proliferation, differentiation into effector and memory T cells, and contraction of the T cell pool by apoptosis are tightly regulated processes1;2. Signaling by co-stimulatory cues, especially ligands of the CD28-like and tumor necrosis factor receptor family mem-bers, dictate the difference between antigenic unresponsiveness, successful immune responses, and chronic inflammatory responses. Specific T cell functional responses are coupled to these co-stimulatory cues via intracellular signaling pathways.The intracellular signaling protein Rap1, a member of the Ras superfamily of GT-Pases, has emerged as playing a central role in interpreting TCR and co-stimulatory signals to coordinate T cell immune responses in vitro and in vivo. Like other small GTPases, Rap1 is activated by guanine nucleotide exchange factors (GEFs) which catalyze dissociation of the GTPase from GDP3. Subsequent binding of Rap1 to cyto-solic GTP activates Rap1, allowing it to interact with downstream effector proteins4. A large number of potential Rap1 effectors have been identified, including adap-tor proteins involved in integrin function, exchange factors for Rho family GEFs which mediate cytoskeletal rearrangements, and components of Ras signaling pa-thways such as Ras GEFs and Raf kinases4. Rap1 signaling is ultimately terminated by GTPase-activating proteins (GAPs) which enhance the intrinsic GTPase activity of Rap1, converting GTP to GDP and returning Rap1 to its inactive state3. Known GAPs for Rap1 include Rap1GAP1A, Rap1GAP1B, Rap1GAP2, Spa-1, and E6TP1 (Spa-L1/SPAR)3;5;6. Activation of Rap1 by chemokines drives polarization of T cells and activates in-tegrins needed for chemotaxis and initial attachment to APCs7. TCR-dependent activation of Rap1 also drives cytoskeletal rearrangements and integrin activation required for establishment of a productive immunological synapse with the APC8;9. T cell co-stimulatory receptors, such as CD28 and CTLA-4, coordinate with the TCR to modulate Rap1 activity, which in turn influences TCR-dependent ERK activation 10-12 and reactive oxygen species production13;14. These latter functions of Rap1 are needed for optimal T cell cytokine transcription and proliferative responses, and two transcription factors sensitive to Rap1 signaling in T cells are NFAT and Elk10;12. CD28 and CTLA4 may modulate Rap1 activation by, respectively, stimulating and inactivating Rap1GAP1A15-17.Influences of Rap1 on T cell function in vitro have suggested that Rap1 might regu-

91


late the quality of T cell responses in vivo, and this has largely been confirmed in stu-dies utilizing genetic manipulation of Rap1 signaling in mice5. Atlhough Rap1 parti-cipates in positive and negative selection of thymocytes, expression of constitutively active Rap1 in the T cell compartment has little overt effect on thymocyte develop-ment, but can enhance positive selection for low affinity antigens11;18;19. Transgenic expression of the active Rap1 mutant Rap1E63, but not the weaker RapV12 active mutant, decreases T helper cell function and increases the frequency and function of CD103-expressing regulatory T cells11;19. A similar phenotype is observed in mice lacking Spa-1, where thymocyte development appears normal20. However, if en-dogenous Rap1 is further activated by exogenous expression of a Rap1 GEF, Spa-1 –deficient mice display expanded DP thymocyte populations and develop T-cell leukemia21. Spa-1 –deficiency also results in an age-dependent acquisition of T cell hypo-responsiveness to mitogens and antigen recall challenges20, and in human T cells, elevated Rap1 activation has been linked to T cell anergy10;22. Reciprocally, genetic ablation of Rap1A has no observable effect on T cell development, but does impair T cell polarization and integrin-dependent adhesion23;24. More severe pheno-types are observed in mice transgenically over-expressing Rap1GAPs in the T cell compartment. Spa-1 transgenic mice display a block in α/β thymocyte development at the DN stage25. Rap1GAP1A transgenic mice display normal thymocyte develop-ment and peripheral T cell populations, but accumulate activated, hyper-responsive lymph node T lymphocytes during aging17. A similar effect of T cell Rap1 inactiva-tion with pathological consequences may occur in humans. In patients with rheu-matoid arthritis, a block in Rap1 activation in observed in synovial fluid T cells13;14, associated with enhanced TCR-dependent cytokine and proliferative responses26.Coupling of extracellular stimuli to Rap1 activation in T cells is mediated primarily by two Rap1 GEFs, C3G and CalDAG-GEFI. In general, GEF activity is modulated by inducible conformational changes in GEFs3. TCR stimulation recruits Crk-bound C3G to the TCR27, which in turn stimulates GEF activity of C3G28;29. The conformati-on and activity of CalDAG-GEF I is regulated by the soluble second messengers cal-cium and diacyl glycerol (DAG), both of which are generated following phospholi-pase C γ1 activation30. CalDAG-GEF is required for TCR and chemokine -dependent Rap1 activation in human T cells30;31. In contrast to GEFs, Rap1GAPs are thought to be constitutively active3, and their ability to inactivate Rap1 is instead regulated by changes in expression levels and recruitment to the site of activated Rap132. Three Rap1GAPs have been reported in T lymphocytes, RapGAP1A, RapGAP1B, and Spa-133;34. TCR-dependent Rap1 activation is inhibited by co-ligation of CD2815, which is RapGAP1A-dependent16;17. RapGAP1B can bind to G-coupled receptors34. Spa-1 is

92

Chapter 4

expressed in proliferating lymphoid cells33;35 and its participation in TCR signaling is suggested by its recruitment to the immunological synapse during antigen-specific stimulation of T cells36. Variation in phenotypes of mice transgenically expressing different Rap1GAPs raises the possibility that each Rap1GAP family member may make distinct contributions to T cell activation. Here, we find that Rap1GAPs are differentially expressed in resting and TCR/CD28-stimulated human T lympho-cytes, and that expression of Rap1GAPs is regulated by both transcriptional and post-translational mechanisms. Distinct Rap1GAPs may thus differentially couple external stimuli to Rap1 regulation, or inactivate Rap1 in distinct cellular subcom-partments, dependent upon the activation status of the T cell.

Results

Differential expression of Rap1GAP family member mRNA in resting and activa-ted human T lymphocytes

We initiated our studies by investigating mRNA expression of each of the Rap1GAP family members in freshly isolated resting and CD3/CD28-stimulated human pe-ripheral blood T lymphocytes. In unstimulated T cells, Spa-1 mRNA was expressed at low to undectable levels, but increased following CD3/CD28 stimulation (Fig. 1). Rap1GAP1A mRNA was readily detected in resting T cells, and increased further following activation. A similar pattern of expression was observed for Rap1GAP1B. In contrast, Rap1GAP2 and E6TP1 mRNA, both expressed in unstimulated T cells,

Spa-1

E6TP1

RapGAP1A

RapGAP1B

RapGAP2

CD3/CD28C- 24 72 W

Figure 1. Rap1GAP family member mRNA ex-pression is differentially regulated by CD3/CD28 stimulation. Qualitative PCR analysis of Spa-1, Rap-1GAP1A, Rap1GAP1B, Rap1GAP2 and E6TP1 mRNA expression in freshly isolated unstimulated (-) human T lymphocytes, and lymphocytes stimulated for 24 or 72 hours with anti-CD3/CD28 antibodies. Positive controls (C) consisted of mRNA from COS-7 cells transfected with 10 ng of expression plasmid for the appropriate Rap1GAP, while water (W) served as a negative control. Results are representative of three independent experiments.

93


were down-regulated within 24 hours of CD3/CD28 ligation. Together, these re-sults suggested that each of the Rap1GAP family members could be expressed in T lymphocytes, and that regulation of specific Rap1GAPs was a dynamic process dependent upon T cell activation status.

Spa-1 protein expression is specifically up-regulated in activated CD4+ T lymp-hocytes

We next examined if changes in Rap1GAP family member mRNA expression fol-lowing T cell activation resulted in changes in protein expression. Purified CD3+

T lymphocytes were cultured in medium alone or stimulated with activating anti-CD3/CD28 antibodies for up to 72 hours, and cellular lysates examined for Spa-1 protein expression by immunoblotting (Fig. 2). Consistent with mRNA expression, Spa-1 protein was also detected at low levels in freshly isolated T cells (Fig. 2A). Following CD3/CD28 stimulation, Spa-1 protein expression increased in a time-dependent manner throughout the time course of activation (Fig. 2A). We further investigated if Spa-1 expression was similarly regulated in CD4+ T helper cells and CD8+ cytotoxic T cells. Purified T cell subsets were isolated by negative selection,

Total CD4 Total CD4

medium CD3/CD28

CD8 CD8

+ +-

Total

CD3/CD28

+ + +

24h 48h 72h

CD3/CD28:

Spa-1

actin

- - - -

Figure 2. Spa-1 protein protein is selectively up-regulated in CD4+ T lymphocytes following CD3/CD28 stimulation. (A) Total T lymphocytes were left unstimulated (-) or stimulated (+) for the indicated number of hours with anti-CD3/CD28 antibodies and Spa-1 and actin expres-sion determined by immunoblotting of cell lysates. (B) Immunoblotting of total and purified CD4+ T lymphocyte lysates with Spa-1 antibodies after 72 hour incubation in medium alone or in the presence of anti-CD3/CD28 an-tibodies. (C) Immunoblotting of total

and purified CD8+ T lymphocyte lysates with Spa-1 antibodies after 72 hour incubation in the absence (-) or presence (+) of anti-CD3/CD28 antibodies. Results shown are representative of 3-6 independent experiments. Experiments shown are representative of three to five independent experiments.

B C

A

94

Chapter 4

and cultured in the absence or presence of anti-CD3/CD28 antibodies for 72 hours. Remarkably, Spa-1 protein expression was up-regulated only in stimulated CD4+ T lymphocytes (Fig. 2B), but not CD8+ cytotoxic T cells (Fig. 2C). Spa-1 thus repre-sents a lineage and activation status –specific regulator of Rap1 function in human T lymphocytes.

Figure 3. Spa-1 up-regulation requires CD28 costimulation and involvement of PI3-kinase and NF-kB signaling pathways. (A) Immunblotting of lysates from total T lymphocytes with anti-Spa-1 and anti-tubulin antibodies following 72 hours culture in medium alone (-), or medium con-taining anti-CD3 and anti-CD28 antibodies, alone or in combination. (B) Immunoblotting of lysates from total T lymphocytes with anti-Spa-1 and anti-actin antibodies following 72 hours culture in the presence of increasing concentrations of anti-CD3 and anti-CD28 antibodies. (C) Immunoblotting of lysates from total T lymphocytes with anti-Spa-1 and anti-CrkL antibodies following incubation for the indicated number of hours in the absence (-) or presence (+) of anti-CD3/CD28 antibodies and vehicle control (med), LY294002 (LY, 20 μM), PD98059 (PD, 50 μM), or CAPE (40 μM). Experiments shown are representative of three to five independent experiments.

B

C

A

CD3 CD28 CD3/CD28

Spa-1

tubulin

- anti-CD3:(μg/ml)

0

anti-CD28 (mg/ml)

0.1

0.5

1

0 1.25 2.5 5

anti-CD28 (mg/ml)

0 1.25 2.5 5

Blot AB: Spa-1 actin

+ + +CD3/CD28:

24h 48h 72h

med

LY

PD

CAPE

+ + +

24h 48h 72h

Blot AB: Spa-1 CrkL

--- - - -

95


T lymphocyte Spa-1 expression requires CD28 costimulatory activation of PI3-kinase and NF-κB signaling pathways

To further dissect the signaling requirements needed for induction of Spa-1 expres-sion in T cells, we examined the relative contributions of CD3 and CD28 stimula-tion. Purified T cells were cultured for 72 hours in medium alone, or anti-CD3 and anti-CD28 antibodies, alone or in combination (Fig. 3A). Stimulation via CD3, or CD28 ligation in the absence of CD3 stimulation, failed to induce Spa-1 expression. Dose-dependency experiments, titrating in increasing amounts of anti-CD3 and an-ti-CD28 antibodies, demonstrated that induction of Spa-1 expression was sensitive to the strengthes of both CD3 and CD28 signals (Fig. 3B). To gain insight into which CD3 and CD28-dependent intracellular signaling events might contribute to Spa-1 induction, isolated T lymphocytes were pre-incubated with pharmacological inhibi-tors of PI3-kinase catalytic subunits (LY294002), the MEK/ERK pathway (PD98059), and NF-κB activation (CAPE) (Fig. 3C). Inhibition of either PI3-kinase or NF-κB sig-naling pathways almost completely abolished Spa-1 induction, while suppression of MEK/ERK signaling components had no effect. As both PI3-kinase and NF-κB signaling pathways are known important downstream mediators of CD28 signaling, this may explain the requirement for CD28 costimulation in the induction of Spa-1.

Mitogenic stimuli and homeostatic cytokines fail to induce Spa-1 expression

Initial descriptions of Spa-1 characterized it as a protein expressed specifically in proliferating lymphoid cells33;35. To examine if there was a strict relationship between T cell proliferation and Spa-1 expression, we compared Spa-1 protein expression in

Total Total CD4 Total CD4

CD3/CD28 IL-15 PHA + IL-2

Total CD4

IL-7

Spa-1

ERK

Figure 4. Homeostatic cytokines and mitogenic stimuli fail to induce Spa-1 expression. Total and CD4+ T lymp-hocytes were stimulated for 72 hours with anti-CD3/CD28 antibodies, or 7 days with IL-15 (10 ng/ml), IL-7 (10 ng/ml), or PHA (1 ng/ml) + IL-2 (50 U/ml). Cel-lular lysates were prepared and Spa-1 ex-

pression detected by immunoblotting. Experiments shown are representative of three independent ex-periments.

96

Chapter 4

CD3/CD28-stimulated T cells with T cells exposed to homeostatic cytokines and mi-togens. Total purified T cells and CD4+ T cells were treated for 72 hours with anti-CD3/CD28 antibodies, or for seven days with IL-7, IL-15, or PHA+IL-2 (Figure 4). Of these stimuli, only CD3/CD28 antibodies induced Spa-1 expression. IL-7, IL-15, and PHA+IL-2 also failed to upregulate Spa-1 expression in CD8+ T cells, and Spa-1 induction was not evident at shorter time points tested, including 48 and 96 hours (data not shown). Thus, in freshly isolated T lymphocytes, Spa-1 induction appears to be exquisitely dependent upon CD28 costimulatory signals, rather than associa-ted with T cell proliferation.

CD3/CD28 stimulation promotes degradation of Rap1GAP1 and Rap1GAP2 pro-teins

Finally, we turned our attention to potential protein expression of other Rap1GAP family members in T cells. Total purified T cells were left unstimulated or stimulated for 24 and 72 hours in the presence of anti-CD3/CD28 antibodies, and cellular lysates prepared for immunoblotting. Although E6TP1 mRNA expression was detected in resting T cells (Figure 1), and the E6TP1 antibodies used for immunoblotting could detect exogenous E6TP1 in transfected COS cells, we could not detect any endoge-nous E6TP1 protein expression in T lymphocytes (data not shown). mRNA analysis indicated that both Rap1GAP1A and Rap1GAP1B expression was induced follo-wing CD3/CD28 stimulation (Figure 1). Surprisingly, using an antibody recogni-zing both Rap1GAP1A and Rap1GAP1B, we observed that expression of Rap1GAP1 proteins was suppressed in a time-dependent manner following CD3/CD28 stimula-tion (Figure 5A). This was accompanied by the appearance of apparent degradation products of Rap1GAP1 proteins recognized by the antibody. Similar results were also observed in purified CD4+ and CD8+ T cell subsets following stimulation. Rap-1GAP2 protein expression, like Rap1GAP2 mRNA, also decreased following CD3/CD28 stimulation (Figure 5B). At 24 hours post-stimulation, decreases in full-length Rap1GAP2 expression were already observed, accompanied by an upward shift in gel motility of the remaining proteins. This may indicate activation-dependent phosphorylation and/or ubiquitination of Rap1GAP2. Additionally, disappearance of full-length Rap1GAP2, like Rap1GAP1s, was also associated with the appearance of degradation products. Activation-dependent mobility shifts and degradation of Rap1GAP2 occurred in both CD4+ and CD8+ T lymphocytes. Together, our results indicate that post-translational modifications play an important role in the suppres-

97


sion of Rap1GAP1 and Rap1GAP2 expression in activated T lymphocytes.

Discussion

Our results demonstrate that expression of Rap1GAP family members is dynami-cally regulated by both transcriptional and post-translational mechanisms in human T lymphocytes. Previous studies have independently but collectively provided evi-dence for the expression or potential expression of Spa-1, Rap1GAP1A, Rap1GAP1B, Rap1GAP2 and E6TP1 in murine or human T lymphocytes6;20;33;34;37. However, com-parative analysis of Rap1GAP expression within the T cell lineage has been confined to analysis of Rap1GAP1A and Spa-1 expression in murine thymocytes and splenic T cells20. Here, both GAPs were expressed in thymocytes, while Spa-1 was relatively enriched in peripheral T lymphocytes. In human T lymphocytes, we find that ex-pression of each Rap1GAP is dependent upon T cell lineage and activation status.

total CD4 CD8

RapGAP2

ERK

CD3/CD28: 24 720 24 720 24 720

RapGAP1

ERK

total CD4 CD8

CD3/CD28: 24 720 24 720 24 720

Figure 5. CD3/CD28 stimulation indu-ces degradation of Rap1GAP1 and Rap-1GAP2 proteins in T lymphocytes. Total, CD4+ and CD8+ T lymp-hocytes were immedia-tely lysed or stimulated for 24 or 72 hours with anti-CD3/CD28 anti-bodies. Lysates were prepared and assessed by immunoblotting with anti-Rap1GAP1, anti-Rap1GAP2, and control anti-ERK antibodies. (A) Rap1GAP1 expression. (B) Rap1GAP2 expres-sion. Full-length Rap-1GAP expression is noted

B

A

by solid arrows to the right of immunoblots. Altered protein mobility and/or degradation products are noted by dotted arrows. Experiments shown are representative of three independent experiments.

98

Chapter 4

Spa-1 was originally identified in proliferating murine lymphoid cells and T cell hybridomas33;35. Spa-1 expression in lymphoid cells was suppressed following IL-2 withdrawal, associated with cell cycle arrest. In T cell hybridomas, stimulation with anti-CD3 antibodies also suppressed Spa-1 expression, again accompanied by cell cycle arrest. Conversely, mitogenic stimulation of splenic cells with concanavalin A stimulated Spa-1 expression35. However, it is unclear as to how far these observa-tions can be extended to normal murine and human T lymphocytes, as subsequent studies from the same group found that Spa-1 could be readily detected in freshly isolated peripheral murine T lymphocytes, only a small percentage of which would be expected to be actively cycling20. Here, we find that in human T lymphocytes, Spa-1 is expressed at low to undetectable levels in resting cells, and upregulated only in CD4+ T lymphocytes following CD3/CD28 stimulation. Spa-1 expression requires specific input from both CD3 and CD28, signals which can not be recapitulated by homeostatic cytokines and mitogens used in our studies. Two signaling pathways which are cooperatively activated by CD3 and CD28, PI3-kinase and NF-κB, appear to be important in the induction of Spa-1 expression. The requirement for CD28 co-stimulation in the induction of Spa-1 expression may be relevant for reports of a role for elevated Rap1 activity in anergic T cells. One common model for inducing anergy or tolerance in mature T cells is to stimulate naive cells with antigen in the absence of appropriate co-stimulatory input, such as that provided by CD28. Induc-tion of anergy in human T cells in vitro is associated with constitutive activation of Rap110. Similarly, CD4+ T lymphocytes isolated from mice following in vivo antigen tolerization also display elevated levels of Rap122. As T cells from mice lacking Spa-1 develop age-dependent hyporesponsiveness to mitogens and recall antigen20, it may be the case that up-regulation of Spa-1 expression, and subsequent dampening of Rap1 activity, may be a necessary element of appropriate T cell costimulation.Although we could detect increased mRNA expression of Rap1GAP1A and Rap-1GAP1B in T lymphocytes following CD3/CD28 stimulation, this did not correlate with changes in protein expression. Instead, Rap1GAP1 protein expression was suppressed in activated T cells, accompanied by protein degradation. Post-trans-lational regulation of protein stability has appeared as an emerging theme in Rap-1GAP expression. Thyroid-stimulating hormone (TSH) stabilizes Rap1GAP1A pro-tein in thyroid cells by inactivating glycogen synthase kinase 3β, which otherwise phosphorylates Rap1GAP1A and promotes its proteasomal degradation38. Gαo and Gαi –coupled receptors can also promote Rap1GAP1B proteasomal degradation39. Finally, E6TP1 is targeted by the papilloma virus E6 oncogene for degradation37;40. Here, we extend these studies to show that both Rap1GAP1 and Rap1GAP2 proteins

99


are degraded following CD3/CD28 stimulation in T cells. The mechanism(s) leading to degradation of these Rap1GAPs in T cells remains to be determined. In platelets, cAMP and cGMP can promote phosphorylation of Rap1GAP26;41. However, these events do not appear to affect Rap1GAP2 stability, but rather inhibits suppression of Rap1GAP2 by 14-3-3 proteins42. The c-Cbl and Cbl-b proto-oncogene products have previously been indentified as E3 ubiquitin ligases regulating Rap1 function in T cells. However, these proteins suppress Rap1 activation by promoting degradation of the Crk-bound Rap1 GEF C3G43;44.It has been generally thought that diversity in Rap1GAP expression represented tissue-specific enrichment of enzymes performing redundant functions in the in-activation of Rap1. Indeed, in over-expression systems Rap1GAP1A, Rap1GAP1B, Spa-145 and E6TP1 (our unpublished observation) can each effectively suppress integrin-dependent adhesion in T cells. However, mice transgenically expressing RapGAP1A and Spa-1 have distinct phenotypes17;25. Inappropriate expression of Rap1GAP1A, driven by the β-actin promoter, has no obvious effect on thymocyte development, but leads to accumulation of T cells with an activated phenotype in peripheral lymph nodes46. In contrast, transgenic expression of Spa-1 under control of the lck promoter, expressed early in thymocyte development, blocks thymocyte maturation at an early CD4-CD8- stage. Delaying Spa-1 over-expression until la-ter in development, using a CD4 promoter, failed to modulate further thymocyte maturation25. These observations, along with our findings, suggest that each Rap-1GAP may make distinct contributions during T cell development and activation. One possibility is that each Rap1GAP may couple to distinct cell surface signaling proteins. In vitro studies have indicated that Rap1GAPs inactivate Rap1 primarily at the cell membrane32. This may be due to specific recruitment of GAPs to cell-surface proteins. Initial evidence supporting this model has already been provided. For example, Rap1GAP1A is regulated directly or indirectly by CD28 and CTLA-4 ligation in T cells16;17. Rap1GAP1B, which compared to Rap1GAP1A, contains an N-terminal amino acid extension encoding a GoLoCo motif, can bind to Gi and Ga subunits of G-coupled receptors34. Indirect evidence that Spa-1 interacts with cell-surface proteins in T cells is seen in observations that Spa-1 is recruited to the T cell immunological synapse following antigen stimulation36. A second possibility is that in addition to inactivating Rap1, each Rap1GAP may also serve independent functi-ons. In the case of Spa-1, protein-protein interactions with Brd447 and aquaporin-248

have been observed, although the relevance of these interactions to T cell function remains to be investigated. The wealth and variation of co-stimulatory proteins and other signaling receptors expressed on distinct T cell subsets during activation

100

Chapter 4

and differentiation, in conjunction with the differential expression of Rap1GAPs we observe in T cells, raises the possibility that Rap1 function can be therapeutically modulated in distinct T cell populations to clinically enhance or suppress T cell-dependent immune responses.


Reagents

Recombinant human cytokines used in these studies included IL-2 (Dreiech, Ger-many), IL-7 (Strathmann Biotech GMBH, Germany), and IL-15 (R&D, Abingdon, UK). PHA was purchased from Sigma, LY294002 (used at 20 µM), and PD 98059 (50 µM) from Calbiochem, and CAPE (40 µM) from Biomol.

T cell isolation and culture

Peripheral blood mononuclear cells (PBMC) were isolated from healthy volunteers by Ficoll-Isopaque density gradient centrifugation (Nycomed, Pharma, Oslo, Nor-way). CD3+ T lymphocytes were purified from PBMC using a negative isolation procedure (T Cell Negative Isolation Kit, Dynal Biotech, Oslo, Norway) according to the manufacturer’s instructions. In some experiments, purified CD4+ and CD8+

T lymphocytes were also obtained using negative isolation procedures (MACSiso-lation kits, Miltenyi). T cells and T cell subset purity was greater than 95% as as-sessed by FACS analysis (described below). T cells were cultured at 37°C under 5% CO2 at a density of 1x106/ml in IMDM medium supplemented with 10% FCS, 200 µM L-glutamine, 25 mM HEPES, streptomycin (100 ng/ml) and penicillin (10 U/ml) (all from Invitrogen, Carlsbad, CA). T cells were cultured in medium alone or stimulated for up to 72 hours with activating anti-CD3 antibodies 1XE (Sanquin, Amsterdam, The Netherlands) or 16E9 (provided by Dr. R. van Lier, our institute), and/or anti-CD28 antibody (clone 15E8, Sanquin) antibodies. Alternatively, T cells were stimulated for up to 7 days with IL-7 (10 ng/ml), IL-15 (10 ng/ml), or PHA (1 ng/ml) + IL-2 (50 U/ml).

101


Flow Cytometric Analysis

Purity of T cell populations was assessed by staining with anti-CD3 antibody. Ef-fects of antibodies and cytokines on T cell proliferation was assessed by CFSE dilu-tion analysis. T cells were resuspended at 5-10x106 cells in PBS and labeled with 2.5 µM CFSE (Molecular Probes Europe BV, Leiden, The Netherlands) for 10 minutes at 37°C. Cells were then washed, resuspended at 1x106 cells/ml in complete culture medium, and left untreated, or stimulated with anti-CD3 and anti-CD28 antibodies, alone or in combination, for up to 72 hours, or PHA+IL-2, IL-7 or IL-15 for 7 days. Proliferation was measured using a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA) and CellQuest Pro software (BD Biosciences).

RNA extraction and PCR analysis

T lymphocytes were left untreated or stimulated for 24 or 72 hours with anti-CD3/CD28 antibodies. T lymphocytes were then harvested, washed with PBS and total RNA isolated using a GenElute RNA isolation kit (Sigma-Aldrich). RNA was reverse-transcribed using SuperScript™ II Reverse Transcriptase (Invitrogen) and cDNA amplified by polymerase chain reaction (PCR) using primers specific for GADPH (Eurogentec, Philadelphia, PA), Rap1GAP1a, Rap1GAP1b, Rap1GAP2, Spa-1 and E6TP1α (all synthesized by Invitrogen). RapGAP primer sequences and PCR condi-tions were as previously described6. Positive controls used were total RNA isolated from COS-7 cells transfected with 10 ng pCMV-myc-E6TP1 (kindly provided by Dr. V. Band, Northwestern University, Chicago, IL), pMT2HA-Rap1GAP1A, pMT2HA-Rap1GAP1B, pSR-His-Spa-1, and pCDNA3-Rap1GAP2 expression vectors37;45 using Lipofectamine 2000 transfection reagent (Invitrogen). PCR products were separated by electrophoresis and visualized using a Gene Flash imaging system (Westburg, Leusden, The Netherlands).

Western blot analysis

Cells were counted, equivalent numbers of T lymphocytes were lysed in 1x Laem-mli’s buffer, and clarified protein lysates were resolved by electrophoresis on 3-8% gradient Bis-Tris SDS NuPAGE® gels (Invitrogen). Proteins were then transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad Laboratories, Hercules, CA)

102

Chapter 4

using a semi-dry transfer apparatus (Invitrogen). Membranes were washed in Tris-buffered saline (pH 8.0) containing 0.05% Tween-20 (Bio-Rad) (TBS/T), blocked in 2% milk (Bio-Rad)/TBS/T, and incubated overnight at 4°C in primary antibody dilu-ted in TBS/T. Primary antibodies used included antibodies specific for Rap1GAP1a, Rap1GAP1b, E6TP1, actin, CrkL (all from Santa Cruz Biotechnology, Santa Cruz, CA), Spa-1, ERK 1/2 (Cell Signaling, Beverly, MA), and tubulin (Sigma-Aldrich). Rap1GAP2 rabbit anti-serum and purified rabbit anti-Rap1GAP2 antibodies have been previously described6;42. Following incubation with primary antibodies, blots were washed in TBS/T and then incubated in TBS/T containing IRDye 680 or IRDye 800 -conjugated anti-rabbit or anti-mouse immunoglobulin antibodies (LI-COR, Bad Homburg, Germany) and staining detected using an Odyssey Imager (LI-COR) and Odyssey 3.0 software.

103


References

Marrack P, Kappler J. Control of T cell viability. Annu.Rev.Immunol. 2004;22:765-1. 787.van Lier RA, ten Berge IJ, Gamadia LE. Human CD8(+) T-cell differentiation in 2. response to viruses. Nat.Rev.Immunol. 2003;3:931-939.Bos JL, Rehmann H, Wittinghofer A. GEFs and GAPs: critical elements in the 3. control of small G proteins. Cell 2007;129:865-877.Raaijmakers JH, Bos JL. Specificity in Ras and Rap signaling. J.Biol.Chem. 20084. Minato N, Kometani K, Hattori M. Regulation of immune responses and hema-5. topoiesis by the Rap1 signal. Adv.Immunol. 2007;93:229-264.Schultess J, Danielewski O, Smolenski AP. Rap1GAP2 is a new GTPase-activa-6. ting protein of Rap1 expressed in human platelets. Blood 2005;105:3185-3192.Shimonaka M, Katagiri K, Nakayama T et al. Rap1 translates chemokine signals 7. to integrin activation, cell polarization, and motility across vascular endothe-lium under flow. J.Cell Biol. 2003;161:417-427.Katagiri K, Hattori M, Minato N, Kinashi T. Rap1 functions as a key regulator of 8. T-cell and antigen-presenting cell interactions and modulates T-cell responses. Mol.Cell Biol. 2002;22:1001-1015.Bivona TG, Wiener HH, Ahearn IM et al. Rap1 up-regulation and activation on 9. plasma membrane regulates T cell adhesion. J.Cell Biol. 2004;164:461-470.Boussiotis VA, Freeman GJ, Berezovskaya A, Barber DL, Nadler LM. Mainte-10. nance of human T cell anergy: blocking of IL-2 gene transcription by activated Rap1. Science 1997;278:124-128.Li L, Greenwald RJ, Lafuente EM et al. Rap1-GTP is a negative regulator of Th 11. cell function and promotes the generation of CD4+CD103+ regulatory T cells in vivo. J.Immunol. 2005;175:3133-3139.Czyzyk J, Leitenberg D, Taylor T, Bottomly K. Combinatorial effect of T-cell re-12. ceptor ligation and CD45 isoform expression on the signaling contribution of the small GTPases Ras and Rap1. Mol.Cell Biol. 2000;20:8740-8747.Remans PH, Gringhuis SI, van Laar JM et al. Rap1 signaling is required for sup-13. pression of Ras-generated reactive oxygen species and protection against oxida-tive stress in T lymphocytes. J.Immunol. 2004;173:920-931.Remans PH, Wijbrandts CA, Sanders ME et al. CTLA-4IG suppresses reactive 14. oxygen species by preventing synovial adherent cell-induced inactivation of Rap1, a Ras family GTPASE mediator of oxidative stress in rheumatoid arthritis T cells. Arthritis Rheum. 2006;54:3135-3143.

104

Chapter 4

Reedquist KA, Bos JL. Costimulation through CD28 suppresses T cell receptor-15. dependent activation of the Ras-like small GTPase Rap1 in human T lymphocy-tes. J.Biol.Chem. 1998;273:4944-4949.Carey KD, Dillon TJ, Schmitt JM et al. CD28 and the tyrosine kinase lck stimulate 16. mitogen-activated protein kinase activity in T cells via inhibition of the small G protein Rap1. Mol.Cell Biol. 2000;20:8409-8419.Dillon TJ, Carey KD, Wetzel SA, Parker DC, Stork PJ. Regulation of the small 17. GTPase Rap1 and extracellular signal-regulated kinases by the costimulatory molecule CTLA-4. Mol.Cell Biol. 2005;25:4117-4128.Amsen D, Kruisbeek A, Bos JL, Reedquist K. Activation of the Ras-related GT-18. Pase Rap1 by thymocyte TCR engagement and during selection. Eur.J.Immunol. 2000;30:2832-2841.Sebzda E, Bracke M, Tugal T, Hogg N, Cantrell DA. Rap1A positively regulates 19. T cells via integrin activation rather than inhibiting lymphocyte signaling. Nat.Immunol. 2002;3:251-258.Ishida D, Yang H, Masuda K et al. Antigen-driven T cell anergy and defective 20. memory T cell response via deregulated Rap1 activation in SPA-1-deficient mice. Proc.Natl.Acad.Sci.U.S.A 2003;100:10919-10924.Wang SF, Aoki M, Nakashima Y et al. Development of Notch-dependent T-cell 21. leukemia by deregulated Rap1 signaling. Blood 2008;111:2878-2886.Morton AM, McManus B, Garside P, Mowat AM, Harnett MM. Inverse Rap1 22. and phospho-ERK expression discriminate the maintenance phase of tolerance and priming of antigen-specific CD4+ T cells in vitro and in vivo. J.Immunol. 2007;179:8026-8034.Duchniewicz M, Zemojtel T, Kolanczyk M et al. Rap1A-deficient T and B cells 23. show impaired integrin-mediated cell adhesion. Mol.Cell Biol. 2006;26:643-653.Li Y, Yan J, De P et al. Rap1a null mice have altered myeloid cell functions sug-24. gesting distinct roles for the closely related Rap1a and 1b proteins. J.Immunol. 2007;179:8322-8331.Kometani K, Moriyama M, Nakashima Y et al. Essential role of Rap signal in 25. pre-TCR-mediated beta-selection checkpoint in alphabeta T-cell development. Blood 2008;112:4565-4573.Abreu JR, Grabiec AM, Krausz S et al. The presumed hyporesponsive beha-26. vior of rheumatoid arthritis T lymphocytes can be attributed to spontaneous ex vivo apoptosis rather than defects in T cell receptor signaling. J.Immunol. 2009;183:621-630.Reedquist KA, Fukazawa T, Panchamoorthy G et al. Stimulation through the T 27.

105


cell receptor induces Cbl association with Crk proteins and the guanine nucleo-tide exchange protein C3G. J.Biol.Chem. 1996;271:8435-8442.Ichiba T, Kuraishi Y, Sakai O et al. Enhancement of guanine-nucleotide exchange 28. activity of C3G for Rap1 by the expression of Crk, CrkL, and Grb2. J.Biol.Chem. 1997;272:22215-22220.Ichiba T, Hashimoto Y, Nakaya M et al. Activation of C3G guanine nucleoti-29. de exchange factor for Rap1 by phosphorylation of tyrosine 504. J.Biol.Chem. 1999;274:14376-14381.Katagiri K, Shimonaka M, Kinashi T. Rap1-mediated lymphocyte function-asso-30. ciated antigen-1 activation by the T cell antigen receptor is dependent on phosp-holipase C-gamma1. J.Biol.Chem. 2004;279:11875-11881.Ghandour H, Cullere X, Alvarez A, Luscinskas FW, Mayadas TN. Essential role 31. for Rap1 GTPase and its guanine exchange factor CalDAG-GEFI in LFA-1 but not VLA-4 integrin mediated human T-cell adhesion. Blood 2007;110:3682-3690.Ohba Y, Kurokawa K, Matsuda M. Mechanism of the spatio-temporal regulation 32. of Ras and Rap1. EMBO J. 2003;22:859-869.Kurachi H, Wada Y, Tsukamoto N et al. Human SPA-1 gene product selectively 33. expressed in lymphoid tissues is a specific GTPase-activating protein for Rap1 and Rap2. Segregate expression profiles from a rap1GAP gene product. J.Biol.Chem. 1997;272:28081-28088.Mochizuki N, Ohba Y, Kiyokawa E et al. Activation of the ERK/MAPK pathway 34. by an isoform of rap1GAP associated with G alpha(i). Nature 1999;400:891-894.Hattori M, Tsukamoto N, Nur-e-Kamal MS et al. Molecular cloning of a novel 35. mitogen-inducible nuclear protein with a Ran GTPase-activating domain that affects cell cycle progression. Mol.Cell Biol. 1995;15:552-560.Harazaki M, Kawai Y, Su L et al. Specific recruitment of SPA-1 to the immuno-36. logical synapse: involvement of actin-bundling protein actinin. Immunol.Lett. 2004;92:221-226.Gao Q, Srinivasan S, Boyer SN, Wazer DE, Band V. The E6 oncoproteins of high-37. risk papillomaviruses bind to a novel putative GAP protein, E6TP1, and target it for degradation. Mol.Cell Biol. 1999;19:733-744.Tsygankova OM, Feshchenko E, Klein PS, Meinkoth JL. Thyroid-stimulating 38. hormone/cAMP and glycogen synthase kinase 3beta elicit opposing effects on Rap1GAP stability. J.Biol.Chem. 2004;279:5501-5507.Jordan JD, He JC, Eungdamrong NJ et al. Cannabinoid receptor-induced neurite 39. outgrowth is mediated by Rap1 activation through G(alpha)o/i-triggered prote-asomal degradation of Rap1GAPII. J.Biol.Chem. 2005;280:11413-11421.

106

Chapter 4

Gao Q, Singh L, Kumar A et al. Human papillomavirus type 16 E6-induced de-40. gradation of E6TP1 correlates with its ability to immortalize human mammary epithelial cells. J.Virol. 2001;75:4459-4466.Danielewski O, Schultess J, Smolenski A. The NO/cGMP pathway inhibits Rap 41. 1 activation in human platelets via cGMP-dependent protein kinase I. Thromb.Haemost. 2005;93:319-325.Hoffmeister M, Riha P, Neumuller O et al. Cyclic nucleotide-dependent protein 42. kinases inhibit binding of 14-3-3 to the GTPase-activating protein Rap1GAP2 in platelets. J.Biol.Chem. 2008;283:2297-2306.Shao Y, Elly C, Liu YC. Negative regulation of Rap1 activation by the Cbl E3 43. ubiquitin ligase. EMBO Rep. 2003;4:425-431.Zhang W, Shao Y, Fang D et al. Negative regulation of T cell antigen recep-44. tor-mediated Crk-L-C3G signaling and cell adhesion by Cbl-b. J.Biol.Chem. 2003;278:23978-23983.de Bruyn KM, Rangarajan S, Reedquist KA, Figdor CG, Bos JL. The small GT-45. Pase Rap1 is required for Mn(2+)- and antibody-induced LFA-1- and VLA-4-mediated cell adhesion. J.Biol.Chem. 2002;277:29468-29476.Ishida D, Kometani K, Yang H et al. Myeloproliferative stem cell disorders by 46. deregulated Rap1 activation in SPA-1-deficient mice. Cancer Cell 2003;4:55-65.Farina A, Hattori M, Qin J et al. Bromodomain protein Brd4 binds to GTPase-47. activating SPA-1, modulating its activity and subcellular localization. Mol.Cell Biol. 2004;24:9059-9069.Noda Y, Horikawa S, Furukawa T et al. Aquaporin-2 trafficking is regulated by 48. PDZ-domain containing protein SPA-1. FEBS Lett. 2004;568:139-145.

The Ras guanine nucleo-tide exchange factor Ras-

GRF1 promotes MMP-3 pro-duction in rheumatoid

arthritis synovial tissue

Joana RF Abreu*, Daphne de Launay*, Marjolein E San-ders, Aleksander M Grabiec, Marleen van de Sande,

Paul P Tak and Kris A Reedquist

Division of Clinical Immunology and Rheumatology, Academic Medical Center, University of Amsterdam, Amsterdam, The Ne-

therlands

Arthritis Research and Therapy, 2009, 11(4):R121

5

*Contributed equally

111

RasGRF1 promotes MMP-3 production in rheumatoid arthritis synovial tissue

Abstract

Introduction

Fibroblast-like synoviocytes (FLS) from rheumatoid arthritis (RA) patients share many similarities with transformed cancer cells, including spontaneous production of matrix metalloproteinases (MMPs). Altered or chronic activation of proto-onco-genic Ras family GTPases is thought to contribute to inflammation and joint destruc-tion in RA, and abrogation of Ras family signaling is therapeutic in animal models of RA. Recently, expression and post-translational modification of Ras guanine nu-cleotide releasing factor 1 (RasGRF1) was found to contribute to spontaneous MMP production in melanoma cancer cells. Here, we examined the potential relationship between RasGRF1 expression and MMP production in RA, inflammatory osteo-arthritis (OA), and reactive arthritis (ReA) synovial tissue and FLS.

Methods

Expression of RasGRF1, MMP-1, MMP-3, and interleukin (IL)-6 was detected in synovial tissue by immunohistochemistry and stained sections were evaluated by digital image analysis. Expression of RasGRF1 in FLS and synovial tissue was also assessed by immunoblotting. Double staining was performed to detect proteins in specific cell populations, and cells producing MMP-1 and MMP-3. RasGRF1 ex-pression was manipulated in RA FLS by cDNA transfection and gene silencing, and effects on MMP-1, TIMP-1, MMP-3, IL-6, and IL-8 production measured by enzyme-linked immunosorbent assay (ELISA).

Results

Expression of RasGRF1 was significantly enhanced in RA synovial tissue, and de-tected in FLS and synovial macrophages in situ. In cultured FLS and synovial bi-opsies, RasGRF1 was detected by immunoblotting as a truncated fragment lacking its negative regulatory domain. Production of MMP-1 and -3 in RA but not non-RA synovial tissue positively correlated with expression of RasGRF1 and colocalized in cells expressing RasGRF1. RasGRF1 over-expression in FLS induced production of

112

Chapter 5

MMP-3, and RasGRF1 silencing inhibited spontaneous MMP-3 production.

Conclusions

Enhanced expression and post-translational modification of RasGRF1 contributes to MMP-3 production in RA synovial tissue and the semi-transformed phenotype of RA FLS.

113


Introduction

Inflammation of affected joints in rheumatoid arthritis (RA) is characterized by infil-tration of the synovial sublining by macrophages, lymphocytes, and other immune cells, and intimal lining layer hyperplasia due to increased numbers of intimal ma-crophages and fibroblast-like synoviocytes (FLS)1 . Initial in situ and in vitro studies of invasive RA FLS revealed striking similarities with transformed cells expressing mutated proto-oncogene and tumor suppressor gene products2. Hyperplastic FLS invading the joints of RA patients resemble proliferating tumor cells and in vitro, RA FLS proliferate more rapidly than FLS from inflammatory non-RA patients or healthy individuals3. Characteristic of transformed cells, they spontaneously secrete autocrines and matrix metalloproteinases (MMPs), display anchorage-independent growth, and are resistant to contact inhibition of proliferation4;5. While transforming mutations in gene products involved in cellular transformation, such as Ras and PTEN, have not been detected in RA FLS6;7, it is appreciated that signaling pathways regulated by proto-oncogene and tumor suppressor gene products are constituti-vely activated due to stimulation by inflammatory cytokines, chemokines, growth factors, and oxidative stress in RA synovial tissue8.Ras superfamily small GTPases are expressed throughout mammalian tissue, and play essential roles in coupling extracellular stimuli to multiple downstream signa-ling pathways9. Cellular stimulation results in the activation of guanine nucleotide exchange factors (GEFs), which catalyze the exchange of GDP on inactive GTPase for GTP. The binding of GTP to Ras superfamily GTPases leads to a conformatio-nal change in the GTPase, allowing signaling to downstream effector proteins10. Of these small GTPases, Ras family homologues (H-, K-, and N-Ras) are important in coupling extracellular stimuli to activation of a shared set of signaling pathways re-gulating cell proliferation and survival, including mitogen-activated protein (MAP) kinase cascades, phosphoinositide 3-kinase (PI3K) and Ral GTPases9;11. The related but distinct family of Rho GTPases (including Rac, Cdc42 and Rho proteins) regulate cellular polarization and chemotactic responses, MAP kinase cascades, and oxida-tive burst machinery12;13. Specificity in GEF activation of GTPase families, and even GEF selectivity in activating different Ras homologs, as well as differential coupling of GEFs to specific types of cellular receptors, such as Son-of-sevenless to tyrosine kinase-dependent receptors, and Ras guanine nucleotide-releasing factor (RasGRF) 1 to G protein-coupled receptors, achieves specificity in Ras superfamily GTPase signaling.Previous studies have demonstrated that Ras family homologs are present in RA

114

Chapter 5

synovial tissue, and preferentially expressed in the intimal lining layer14;15. Acti-vation of Ras effector pathways, including MAP kinases, PI3K, and nuclear factor (NF)-kB, is enhanced in RA compared to disease controls16-18. In RA synovial fluid T cells constitutive activation of Ras, in conjunction with inactivation of the related GTPase Rap1, contributes to persistent reactive oxygen species production by these cells19;20. In RA FLS, ectopic expression of dominant-negative (DN) H-Ras suppres-ses interleukin (IL)-1-induced extra-cellular signal-regulated kinase (ERK) activa-tion and IL-6 production21. DN Raf kinase, which broadly binds to and inhibits Ras family members and related GTPases, suppresses epidermal growth factor-induced ERK and c-jun N-terminal kinase (JNK) activation in RA FLS, and reduces constitu-tive expression of MMPs22. Additionally, strategies which broadly inhibit Ras family function in vivo are protective in animal models of arthritis21-23. Evidence is now emerging that altered expression of Ras GEFs may contribute to autoimmune diseases. Mice lacking expression of the Ras GEF Ras guanine nucleo-tide-releasing protein 1 develop a spontaneous systemic lupus erythematosis (SLE)-like disease, and similar defects are observed in a subset of SLE patients24-26. Recent evidence has shown that expression levels of the GEF RasGRF1 regulate constitutive MMP-9 production in human melanoma cells27. RasGRF1 displays in vitro and in vivo exchange activity against H-Ras28, as well as the Rho family GTPase Rac29;30. RasGRF1 activity can also be regulated by protease-dependent post-translational modification, as calpain-dependent cleavage of RasGRF1 enhances its Ras-activating capacity in vitro and in vivo31. Given similarities between FLS and transformed cancer cells, we examined the expression of RasGRF1 in RA and non-RA synovial tissue and FLS, providing evidence that elevated RasGRF1 expression and post-translational modification of this protein in RA synovial tissue may contribute to joint destruction by stimulating MMP-3 production.


Patients and synovial tissue samples

Synovial biopsy samples were obtained by arthroscopy as previously described32

from an actively inflamed knee or ankle joint, defined by both pain and swelling, of patients with RA (n=10)33, inflammatory osteoarthritis (OA) (n=4)34, or reactive arthritis (ReA) (n=7)35. Patient characteristics are detailed in Table 1. All patients

115


provided written informed consent prior to the start of this study, which was appro-ved by the Medical Ethics Committee of the Academic Medical Center, University of Amsterdam, The Netherlands.

Immunohistochemical analysis

Serial sections from six different biopsy samples per patient were cut with a cryostat (5 mm), fixed with acetone, and endogenous peroxidase activity blocked with 0.3% hydrogen peroxide in 0.1% sodium azide/phosphate-buffered saline. Sections were stained overnight at 4oC with monoclonal antibodies against MMP-1 (MAB 1346) and MMP-3 (MAB 1339) (both from Chemicon International, Temicula, CA) and rab-bit polyclonal antibodies recognizing RasGRF1 (SC-863) (Santa Cruz Biotechnology,

Age (years)Male:femaleDisease duration (months)Erythrocyte sedimentation rate (mm/hour)Rheumatoid factor (kU/L)

55 (30-68)6:484 (2-360)64 (2-107)21 (0-138)

Characteristic Median (range)

Table I. Clinical features of RA, ReA and OA patients included in the study.

Diagnosis

Rheumatoid arthritis

Reactive arthritis

Osteoarthritis


33 (22-39)4:32.5 (1-14)5 (0-14)0 (0-1)


72.5 (54-83)2:266 (6-180)9.5 (5-43)0 (0-1)

116

Chapter 5

Santa Cruz, CA), and anti-IL-6 (Department of Nephrology, Leiden University Medi-cal Center, Leiden, The Netherlands). For control sections, primary antibodies were omitted or irrelevant immunoglobulins were applied. Sections were then washed and incubated with goat anti-mouse horseradish peroxidase (HRP)-conjugated- or swine anti-rabbit-HRP-conjugated antibodies (from Dako, Glostrup, Denmark), fol-lowed by incubation with biotinylated tyramide and streptavidin-HRP, and deve-lopment with amino-ethylcarbazole (AEC, Vector Laboratories, Buringame, CA)36. Sections were then counterstained with Mayer’s hematoxylin (Perkin Elmer Life Sciences, Boston, MA) and mounted in Kaiser’s glycerol gelatin (Merck, Darmstadt, Germany).

Digital image analysis

For quantitative analysis of protein expression, stained slides were randomly co-ded by an independent observer, blinded to antibodies used and clinical diagnosis. Stained sections were analyzed by computer-assisted image analysis using the Qwin analysis system (Leica, Cambridge, UK) as previously described in detail37. Values of integrated optical densities (IOD)/mm2 and number of positive cells/mm2 were obtained for both the intimal lining layer and the synovial sublining, and corrected for total number of nucleated cells/mm2.

Immunohistochemical double staining

To detect potential cell-specific expression of RasGRF1 in synovial tissue, tissue sections were incubated with anti-RasGRF1 antibodies overnight at 4oC, followed by serial incubation with swine anti-rabbit-HRP antibodies, biotinylated tyramine, and streptavidin-HRP. Sections were then labeled for one hour at room tempera-ture with FITC-conjugated antibodies to detect T lymphocytes (anti-CD3, clone SK7, Becton Dickinson, San Jose, CA), FLS (anti-CD55, mAB67, Serotec, Oxford, UK), and macrophages (anti-CD68, clone DK25, Dako), followed by incubation with alkaline phosphatase (AP)-conjugated goat anti-mouse antibody (Dako). HRP staining was developed as above, and AP staining was developed using an AP Substrate III kit (SK-5300, Vector Laboratories) according to the manufacturer’s instructions.

117


FLS culture and transfection with cDNA and locked nucleic acids (LNA)

RA FLS and OA FLS were cultured as previously described38. FLS were used between passages 4 and 9 and cultured in medium containing 10% fetal calf serum (FCS). To examine the influence of RasGRF1 overexpression on FLS MMP production, 2 x 105

RA FLS were plated overnight in 6-well plates and then transfected with 7.5 mg con-trol pCDNA3 or pCDNA3 encoding full-length human RasGRF1 (provided by Dr. R. Zippel, University of Milan, Milan, Italy) using Lipofectamine 2000 transfection reagent (Invitrogen, Verviers, Belgium) as per the manufacturer’s instructions. Cul-ture medium was replaced with medium containing 1.0% FCS after 24 hours, and cells harvested 48 hours post-transfection.

RasGRF1 expression in FLS was silenced using RasGRF1-specific and control LNA designed with on-line software (https://rnaidesigner.invitrogen.com/rnaiexpress/design.do) (synthesized by Exiqon A/S, Vedbaek, Denmark). LNA oligonucleotides used were RasGRF1 (TTGcgttaccttTGCt – LNA nucleotides in capital letters, DNA nucleotides in lower case letters), and as a negative control, a scrambled RasGRF1 sequence (GTAcagcaagatTGGg). LNA transductions were performed with Lipofec-tamine 2000 transfection reagent and 50 nM LNA. Culture medium was replaced with starvation medium (1% FCS in DMEM) after 24 hours and cells harvested after an additional 24 hours.

Protein preparation and immunoblotting

FLS were lysed in Laemli’s buffer. Frozen synovial biopsies were homogenized and proteins solubilized using a ReadyPrep™ Sequential Extraction Kit (BioRad, Her-cules, CA) and protein content quantified using a BCA Protein Assay Kit (Pierce, Rockford, IL). Equivalent amounts of protein were resolved by electrophoresis on NuPage 4-12% Bis-Tris gradient gels (Invitrogen) and transferred to polyvinylidene difluoride membrane (BioRad). Proteins were detected by immunoblotting with anti- RasGRF1 (SC-863 and SC-224, Santa Cruz), actin (Santa Cruz) or tubulin (Sig-ma Aldrich, St. Louis, MO) antibodies, followed by extensive washing, incubation with HRP-conjugated anti-rabbit or anti-mouse immunoglobulin antibodies (Bio-Rad) and enhanced chemiluminescence detection (Pierce). For quantitative analysis of RasGRF1 expression, staining was detected using IRDye 680 or 800 –labelled an-tibodies and an Odyssey Imager (LI-COR, Bad Homburg, Germany), and quantified

118

Chapter 5

using Odyssey 3.0 software.

Measurement of MMP-1, MMP-3, TIMP-1, IL-6 and IL-8 production by FLS

Medium was removed from FLS 24 hours after introduction of cDNA or LNA, and replaced with starvation medium. After 24 hours, cell-free tissue culture superna-tants were harvested and analyzed using ELISA kits for MMP-1, MMP-3, TIMP-1 (all from R&D Systems Europe Ltd., Abingdon, UK), IL-6 and IL-8 (both from Sanquin Reagents, Amsterdam, The Netherlands), according to the manufacturer’s instruc-tions.

Immunofluorescence staining

Synovial tissue sections were incubated with primary anti-RasGRF1 antibodies over-night at 4oC, followed by incubation for 30 minutes with Alexa-594-conjugated goat anti-rabbit antibodies (Molecular Probes Europe, Leiden, the Netherlands). Sections were then incubated with mouse monoclonal antibodies against MMP-1, MMP-3, or IL-6, followed by incubation with Alexa-488-conjugated goat anti-mouse antibody (Molecular Probes Europe), mounting in Vectashield (Vector Laboratories) and ana-lysis using a fluorescence microscope (Leica DMRA) coupled to a CCD camera and Image-Pro Plus software (Media Cybernetics, Dutch Vision Components, Breda, the Netherlands).

Statistical analysis

Wilcoxon’s nonparametric signed ranks test was used to compare protein expres-sion between intimal lining layer and the synovial sublining layer within diagnos-tic groups. As no trend towards a difference in RasGRF1 expression was found between inflammatory OA and ReA synovial tissues, these two non-erosive groups were combined as non-RA for further analyses. The Mann-Whitney U test was used for the comparison of RasGRF1 expression between diagnostic groups. Correlations between RasGRF1 expression and MMP-1, MMP-3 and IL-6 expression in synovial tissue were assessed by Spearman’s rank correlation coefficient. ELISA results were examined using Student’s t-test. P values less than 0.05 were considered statistically

119


significant. There was no correction for multiple comparisons due to the explora-tory nature of the study.

Results

Expression of RasGRF1 in RA and non-RA synovial tissue

To gain insight into potential involvement of RasGRF1 in RA, immunohistochemical staining was performed on RA synovial tissue using RasGRF-1 specific antibodies. While no specific staining was observed with irrelevant control rabbit antibodies, ro-bust staining was observed in RA synovial tissue with anti-RasGRF1 antibodies (Fi-gure 1A). RasGRF1 staining was most apparent throughout the intimal lining layer,

RA

OA

control rabbit

RasGRF1

Lin Sub Lin Sub0

100200300400500600700800

IOD

/ mm

2

***

***

RA non-RA

B

C

A Figure 1. Detection of RasGRF1 protein ex-pression in RA and non-RA synovial tissue. (a) Representative stainings of RA synovial tissue with control and anti-RasGRF1 antibodies. (b) Representative stainings of RA and OA synovial tissue with anti-RasGRF1 antibodies. Stainings were developed with AEC (red), and counterstained with Mayer’s hematoxyline. Magnification x 100. (c) Quantitative analysis of Ras signaling protein expression in RA and non-RA (OA and ReA) sy-novial tissue. Integrated optical densities (IOD)/mm2, corrected for nucleated cells, for staining of synovial sublining (sub) and intimal lining (lin) layer of 10 RA and 11-non-RA (4 inflammatory OA, 7 ReA) patients with anti-RasGRF1 antibo-dies. IOD were calculated by computer-assisted image analysis. Box plots represent the 25th to 75th percentiles, the lines within each box the median, and lines outside the boxes designate the 10th and 90th percentiles. Bars indicate statisti-cally significant differences in protein expression between sublining and intimal lining layer tissues within diagnostic groups and between diagnostic groups. * P < 0.05, ** P < 0.01, *** P < 0.005.

120

Chapter 5

but was also observed in infiltrating mononuclear cells found in the synovial subli-ning. Initial qualitative analysis of RasGRF1 expression in RA and inflammatory OA synovial tissue suggested that RasGRF1 expression was elevated in RA synovial tissue (Figure 1B). We therefore compared RasGRF1 expression in RA and non-RA (inflammatory OA and ReA) synovial tissue quantitatively, using digital image ana-lysis (Figure 1C). Preliminary analyses indicated no differences in RasGRF1 expres-sion between inflammatory OA and ReA synovial tissue, either in the intimal lining layer (mean integrated optical density [IOD]/mm2 ± standard error of the mean OA, 259.0 ± 131.6; ReA, 263.4 ± 77.0) or the synovial sublining layer (OA, 113.3 ± 55.7; ReA, 135.6 ± 51.9) (data not shown). Therefore, these two non-erosive groups were combined as non-RA for further analyses. Comparing RA with non-RA synovial tissue, RasGRF1 expression was elevated in RA (P < 0.05) and non-RA (P < 0.01) in-timal lining layer as compared to the synovial sublining. RasGRF1 expression was enhanced in the synovial sublining of RA tissue as compared to non-RA synovial tissue (P < 0.01), and a trend towards enhanced RasGRF1 expression was observed in the RA intimal lining layer. Correction of RasGRF1 expression for the number of RasGRF1-positive cells, confirmed that RasGRF1 expression was enhanced in both the synovial sublining (P < 0.005) and intimal lining layer (P < 0.05) of RA patients compared to non-RA patients (data not shown). Qualitative double labeling of RA synovial tissue with antibodies recognizing RasGRF1 and markers for T lymphocy-tes (CD3), FLS (CD55), and macrophages (CD68) revealed that RasGRF1 expression was restricted to FLS and macrophages (Figure 2).

Figure 2. Representative double stainings of RA synovial tissue with antibodies against Ras-GRF1 and cell-specific markers. Synovial tissue sections were stained overnight with antibodies against RasGRF1, followed by antibodies against CD3, CD55, and CD68. After biotin tyramide enhancement, stai-ning was developed with AEC (red, RasGRF1) and Fast blue (blue, cell-specific markers). Magnification x 100.

CD3/RasGRF1 CD55/RasGRF1 CD68/RasGRF1

121


RasGRF1 expression in RA and non-RA FLS

To independently confirm RasGRF1 expression in synovial tissue and FLS detected by immunohistochemistry, we performed immunoblotting experiments on lysates derived from intact RA and OA synovial biopsies, and RA and OA FLS. In protein lysates derived from intact RA and OA synovial biopsies (Figure 3), we were unable to detect full-length 140 kDa RasGRF1. However, we did observe prominent ex-pression of a 98 kDa truncation product, and lower and variable levels of 75 and 54 kDa truncation products. These C-terminal fragments are thought to be generated by calpain-dependent cleavage, resulting in constitutive activation of RasGRF127;31. In analyses of FLS lysates, full-length 140 kDa RasGRF1 was detected by immuno-blotting in only one of six RA FLS lines (RA FLS5), and neither of two OA FLS lines tested (Figure 4A). In contrast, a 54 kDa RasGRF1 C-terminal fragment was detected in all RA and OA FLS lines, a 75 kDa fragment in three of five RA FLS and both OA FLS lines, and a 98 kDa C-terminal fragments in four of six RA and both OA lines. Quantitative analysis of RasGRF1 protein expression in 5 RA and 5 OA FLS lines revealed no significant difference in total RasGRF1 expression (Figure 4B). With the exception of the 74 kDa RasGRF1 fragment, which was detected at lower levels in RA FLS (P < 0.05), other RasGRF1 truncation fragments, as well as full-length Ras-GRF1, were expressed at similar levels in RA and OA FLS.To verify that the observed truncation products were derived from RasGRF1, rather than non-specific interactions with the antibodies, we performed additional expe-riments. First, RA FLS were transfected with cDNA encoding full-length RasGRF1 (Figure 4C and Figure 4D). Quantitative analysis of proteins detected by immuno-blotting demonstrated that transfection of RA FLS with RasGRF1 cDNA encoding

1 2 3 4 1 2RA OA

18898

6249

Mw (kDa):Biopsies

actin

Figure 3. RasGRF1 is expressed as a truncated protein in synovial tissue. Immunoblot analysis of Ras-GRF1 and actin in RA and OA syno-vial biopsy lysates. 98, 75 and 54 kDa proteins reacting with RasGRF1 anti-bodies, and expected position of full-length 140 kDa RasGRF1, are indica-ted on the left by arrowheads. Relative mobility of molecular weight standards (kDa) are indicated to the right.

122

Chapter 5

B

C

A

D

E F

Figure 4. RasGRF1 is expressed as a truncated protein in FLS. (a) Immunoblot analysis of RasGRF1 in RA and OA FLS. 140, 98, 75 and 54 kDa proteins reacting with RasGRF1 antibodies are indicated on the left by arrowheads. Relative mobility of molecular weight standards (kDa) are indicated to the right. (b) Expression of 140, 98, 75, and 54 kDa RasGRF1 polypeptides, as well as total RasGRF1 signal, normalized to tubulin expression was quantified in RA (n = 5) and OA (n = 5) FLS lines, and expressed as mean optical density ± SEM. (c) Overexpression of RasGRF1 in RA FLS. RA FLS were treated with transfection reagent alone (mock) or transfected with empty (control) vector or vector encoding RasGRF1, and cell lysates immunoblotted with antibodies against RasGRF1 (up-per panel) and tubulin (lower panel). Expression of full-length and truncated RasGRF1 polypeptides is indicated with arrows, and a 60 kDa polypeptide with an asterisk. (d) Expression of 140, 98, 75, and 54 kDa RasGRF1 polypeptides following transfection of RA FLS with empty vector or RasGRF1, normalized to tubulin expression, was quantified and expressed as mean optical density ± SEM (middle panel) (n=4). (e) Silencing of RasGRF1 expression with LNA. RA FLS were treated with transfection reagent alone (mock) or transduced with control or RasGRF1 LNA and lysates assessed for expression of RasGRF1 (upper panel) and tubulin (lower panel) by immunoblotting and (f) quantitative analysis as in (d). * P < 0.05, ** P< 0.01 compared to controls.

98

62

49

Mw (kDa):

tubulin

moc

k

empt

y

Ras

GR

F1

98

62

49

Mw (kDa):

tubulin

moc

k

cont

rol L

NA

Ras

GR

F1 L

NA

Mw (kDa):RA OA FLS

1 2 3 4 5 6 1 2188

986249

* *

****

00.20.40.60.81.01.21.4

140 98 75 54RasGRF1 product (kDa)

Mea

nop

tical

dens

ity

(arb

itrar

yun

its)

mockcontrol LNARasGRF1 LNA

** *

*

0

0.5

1.0

1.5

2.0

140 98 75 54RasGRF1 product (kDa)

Mea

nop

tical

dens

ity

(arb

itrar

yun

its)

mockcontrol vectorRasGRF1

0

1

2

3

4

140 98 74 54 total

OA FLS

RasGRF1 product (kDa)

RA FLS

Mea

nop

tical

dens

ity

(arb

itrar

yun

its)

*

*

*

123


full-length RasGRF1 resulted in the enhanced expression of the 140 (P < 0.01), 98, 75 (P < 0.05), and 54 kDa (P < 0.05) forms of RasGRF1. Second, we silenced RasGRF1 expression by transduction of RA FLS with RasGRF1-specific LNA. LNA are anti-sense nucleotide analogs containing methylene bridges which mimic RNA monomer structure, disrupt gene expression by promoting mRNA degradation and/or preven-ting gene product translation39. RasGRF1-specific LNA decreased RasGRF1 expres-sion in RA FLS compared to control scrambled LNA (Figure 4E), while leaving tubu-lin expression unaffected. Significant decreases in the expression of full-length 140 kDa RasGRF1 (P < 0.05), and 98 (P < 0.01), 75 (P < 0.05) and 54 kDa (P < 0.01) forms were achieved (Figure 4F). Exposure of FLS to transfection reagent alone resulted in the generation of an additional 60 kDa polypeptide (mock-treated FLS in Figures 4C and 4E, noted by an asterisk next to figures) not observed in synovial biopsies or untreated FLS, possibly due to activation of an unidentified cellular protease.

Figure 5. Effect of RasGRF1 over-expression on RA FLS MMP and cytokine production. Tissue culture supernatants from RA FLS transfected with empty vector or RasGRF1 were harvested and assessed for production of (a) MMP-1, (b) TIMP-1, (c) the ratio of TIMP-1 to MMP-1, (d) MMP-3, (e) IL-6 (n = 4 each) and (f) IL-8 (n = 3) by ELISA. * P < 0.05 compared to controls.

B CA

D E F

TIMP-1

0

0.5

1.0

1.5

RasGRF1empty

IL-6

0

20

40

60

80

empty RasGRF1

IL-6

(pg/

ml)

TIMP-1:MMP-1

0

0.2

0.4

0.6

0.8

empty RasGRF1

ratio

TIM

P-1:

MM

P-1

(arb

itrar

yun

its)

MMP-1

0

2

4

6

8

empty RasGRF1

MM

P-1

(μg/

ml)

TIM

P-1

(μg/

ml)

0

10

20

30

40

MM

P3 (n

g/m

l)

empty RasGRF1

MMP-3

*

010203040506070

empty RasGRF1

IL-8

(ng/

ml)

IL-8 *

124

Chapter 5

Effects of changes in RasGRF1 expression on RA FLS MMP-3 production in vitro

As RasGRF1 expression levels regulate MMP production in cancer cell lines27, we examined if modulation of RasGRF1 expression in RA FLS might also regulate con-stitutive MMP and cytokine production. Quantitative analysis of FLS tissue cul-ture supernatants demonstrated that RasGRF1 over-expression had no effect on FLS production of MMP-1 (Figure 5A) or TIMP-1 (Figure 5B). Additionally, the ratio of TIMP-1 expression relative to MMP-1 was unaffected (Figure 5C). However, forced expression of RasGRF1 induced an approximately 150% increase in MMP-3 pro-duction (27.99 ng/ml ± 5.62) (mean ± SEM) compared to FLS transfected with empty control vector alone (11.47 ng/ml ± 2.02) (P < 0.05) (Figure 5D). Enhancing RasGRF1 expression had no effect on spontaneous IL-6 production by RA FLS (Figure 5E),

Figure 6. Effect of RasGRF1 gene silencing on RA FLS MMP and cytokine production. Tis-sue culture supernatants from RA FLS treated with transfection reagent alone (mock) or transfected with control or RasGRF1 LNA were harvested and assessed for production of MMP-1 (a), TIMP-1 (b), the ratio of TIMP-1 to MMP-1 (c), MMP-3 (d), IL-6 (e) (n = 4 each) and (f) IL-8 (n = 3) by ELISA. * P < 0.05 compared to controls.

0

1

2

3

4

MM

P-1

(μg/

ml)

MMP-1

mock RasGRF1 control

LNA

TIMP-1

0

0.2

0.4

0.6

0.8

1.0


LNA

TIM

P-1

(μg/

ml)

ratio

TIM

P-1:

MM

P-1

(arb

itrar

yun

its)

0

0.2

0.4

0.6

0.8


LNA

TIMP-1:MMP-1

0

5

10

15

20

MM

P-3

(ng/

ml)


LNA

MMP-3

0

10

20

30

IL-6

(pg/

ml)


LNA

IL-6

**

0

5

10

15

20

25

30


IL-8

(ng/

ml)

LNA

IL-8

B CA

D E F

125


but did increase spontaneous IL-8 secretion by approximately 2-fold (P < 0.05) (Fi-gure 5F). To determine if RasGRF1 was required for spontaneous MMP or cytokine production, we silenced RasGRF1 gene expression using LNA. Again, modulati-on of RasGRF1 expression failed to influence MMP-1 and TIMP-1 production, or the ratio of TIMP-1 relative to MMP-1 (Figure 6A-C). A significant suppression of spontaneous MMP-3 production was observed in tissue culture supernatants of FLS transduced with RasGRF1-specific LNA (Figure 6D) (P < 0.05), as compared to FLS treated with transfection reagent alone or in combination with control scrambled LNA. Although over-expression of RasGRF1 in RA FLS failed to enhance basal IL-6

MMP-1

MMP-3

IL-6

control mouse

Ras

GR

F1

Lini

ng

MMP-1 MMP-3 IL-6R = 0.70

P = 0.043

0 100 200 300 400 500 600 700 8000

100

200

300

400

IOD/ mm2

IOD

/ mm

2

0 100 200 300 400 500 600 700 8000

1000

2000

3000

4000

5000

6000

IOD/ mm20 100 200 300 400 500 600 700 800

0

50

100

150

200

IOD/ mm 2

Subl

inin

g

R = 0.81

P = 0.022

R = 0.67

P = 0.059

0 100 200 300 400 5000

100

200

300

IOD/ mm2

IOD

/ mm

2

0 100 200 300 400 5000

1000200030004000500060007000

IOD/ mm20 100 200 300 400 500

0

5000

10000

15000

IOD/ mm2

B

A

Figure 7. Association of RasGRF1 expression with MMP production in RA synovial tissue. (a) Representative stainings of RA synovial tissue with control and anti- MMP-1, MMP-3, and IL-6 antibodies (magnification x 100). (b) Correlation of Ras signaling protein expression with MMP-1 and MMP-3 production in RA synovial tissue. Pearson R values (R) and P values are indicated.

126

Chapter 5

production (Figure 5E), IL-6 levels were significantly decreased following silencing of RasGRF1 expression (Figure 6E) (P < 0.05). An apparent 67% reduction in sponta-neous IL-8 production was also noted, but this did not reach statistical significance (P = 0.069) (Figure 6F).

Relationship between RasGRF1 expression and MMP production in RA synovial tissue

Our in vitro data indicated an important role for RasGRF1 in regulating MMP-3 expression in RA FLS. We therefore examined if expression of RasGRF1 was associ-ated with MMP-3 production in RA synovial tissue. Immunohistochemical analysis demonstrated that MMP-1, MMP-3, and IL-6 were readily detected in RA synovial tissue (Figure 7A). RasGRF1 expression demonstrated a strong positive correlation (R= 0.81, P = 0.022) with MMP-1 in the RA synovial sublining, but not in the intimal lining layer (Figure 7B). Instead, a positive correlation between RasGRF1 and MMP-

RasGRF1 MMP-1 Merge

RasGRF1 MMP-3 Merge

Figure 8. Double immunofluorescence labeling of RasGRF1, MMP-1 and MMP-3 in RA synovial tissue. RA synovial tissue was stained with combinations of anti-RasGRF1 and either anti-MMP-1 (upper panels) or anti-MMP-3 (lower panels). Sections were then stained with fluorochrome-conjugated anti-rabbit Ig (red) and anti-mouse IgG (green) antibodies to visualize RasGRF1 and MMP expression, respectively. Colocalization of RasGRF1 with MMPs is visualized by yellow staining in merged images (right panels).

127


3 expression was observed in the intimal lining layer (R = 0.70, P = 0.043). In non-RA patients, no association between RasGRF1 and MMP-1 (synovial sublining: R = 0.17, P = 0.703; intimal lining layer: R = -0.89, P = 0.083) or MMP-3 (synovial sublining: R = 0.83, P = 0.058; intimal lining layer: R = -0.20, P= 0.917) expression was observed (data not shown). No correlation was observed between RasGRF1 expression and IL-6 expression in either RA or non-RA patient cohorts (Figure 7B, and data not shown). Double immunofluorescent staining revealed colocalization of RasGRF1 with MMP-1 and MMP-3 in RA synovial tissue (Figure 8). Colocalization of Ras-GRF1 with MMP-1 was observed in the synovial sublining (Figure 8, upper panels), while RasGRF1 colocalization with MMP-3 was restricted to the intimal lining layer (Figure 8, lower panels). Together, these data indicate that RasGRF1 may contribute to RA FLS MMP-3 production in vivo.

Discussion

Our results demonstrate that RasGRF1 regulates spontaneous MMP-3 production in RA FLS, and suggest that over-expression of RasGRF1 sensitizes signaling pathways promoting MMP-3 production and joint destruction in RA. RasGRF1 specifically ac-tivates H-Ras, but not other Ras homologues in vivo28, and RasGRF1 activation of H-Ras induces constitutive MMP-9 production in human melanoma cells27. RasGRF1 can also activate the Rho family GTPase Rac129;30, and a role for Rac1, potentially via activation of JNK, has been recently shown in the regulation of RA FLS proliferation and invasiveness40. Data has been reported indicating that RasGRF1 can also stimu-late GTP exchange on R-Ras in vitro, although this GEF activity has yet to be verified in vivo41;42.

Our data raise the possibility that changes in expression of GEFs, such as RasGRF1, or negatively regulatory GAPs, may be more relevant to the pathology of RA than GTPase expression levels. We observe a strong positive correlation between Ras-GRF1 expression in RA synovial tissue on one hand, and production of MMP-1 and MMP-3 on the other. Such an association is not clearly observed in non-RA syno-vial tissue. Consistent with the notion that RasGRF1 is involved in the regulation of MMPs, we find that RasGRF1 expression colocalizes to synovial cells producing MMP-1 and MMP-3 in situ, and modulation of RasGRF1 in RA FLS in vitro regulates spontaneous MMP-3 production by these cells. The inability of RasGRF1 modu-lation to regulate MMP-1 production in RA FLS, despite the positive association

128

Chapter 5

of expression of these proteins in the synovial sublining in vivo, may indicate that other RasGRF1-expressing cells, namely macrophages, are a more important source of MMP-1 in vivo. Consistent with this, we observe a relationship between RasGRF1 and MMP-1 in the synovial sublining, rather than the intimal lining layer where FLS predominate. Additionally, co-localization of cells expressing RasGRF1 and MMP-1 is most apparent in the synovial sublining layer. Further direct studies will be nee-ded to examine if RasGRF1 regulates MMP-1 production in synovial macrophages. Alternatively, RasGRF1-dependent secretion of IL-8 or other as yet unidentified in-flammatory cytokines may indirectly promote MMP-1 production in vivo through the recruitment and/or activation of leukocytes.

We provide additional in vitro evidence that although many FLS stimuli regulate both MMP-1 and MMP-3 expression, regulation of these two proteases is not requi-sitely coupled. For instance, adhesion of RA FLS to laminin-111 in the presence of tumor growth factor-b induces expression of MMP-3 but not MMP-143. Inhibition of JNK can partially block TNF-α -induced MMP-1 production by RA FLS, MMP-3 production is independent of JNK44. Reciprocally, mitogen-activated protein kinase-activated protein kinase 2 (MK2) regulates MMP-3 secretion, but not MMP-1, in OA chondrocytes45. That regulation of MMP-1 is uncoupled from that of MMP-3 likely reflects differential utilization of NF-kB, AP-1, Ets, and hypoxia-inducible factor-1α transcription factors by the promoters of the MMP-1 and MMP-3 genes43;46;47. Simi-larly, we find that RasGRF1 is necessary for spontaneous IL-6 production by RA FLS, but over-expression of RasGRF1 is not sufficient to augment IL-6 secretion. This may reflect a necessary coordination of RasGRF1 signaling with other signaling pa-thways, such as previously reported cooperative effects between Ras GTPase and c-myc pathways in the regulation of RA FLS activation22. Further definition of pa-thways by which RasGRF1 modulates MMP and cytokine production will require identification of the immediate downstream target(s) of this GEF in FLS.

While RasGRF1 expression is sufficient and required for spontaneous MMP-3 pro-duction in RA FLS, similar effects of RasGRF1 on MMP-1, TIMP-1 and IL-6 are not observed. Stimuli which activate RasGRF1 include ligands for both tyrosine kinase receptors and G-protein-coupled receptor48. Examples of receptors known to regu-late RasGRF1 and expressed in RA synovial tissue include those for lysophosphati-dic acid and muscarinic acid, NMDA, and nerve growth factor49-52. In preliminary studies, we have found that silencing of RasGRF1 in RA FLS has no effect on TNF-α or IL-1b -induced MMP-3 production (data not shown). RasGRF1 activity can also

129


be regulated by post-translational modification, as calpain-dependent cleavage of RasGRF1 enhances Ras-activating capacity in vitro and in vivo27;31. Enhanced ex-pression of RasGRF1 in RA compared to non-RA may sensitize RA FLS to produce MMPs in response to extracellular stimuli. This would result from disease-specific extracellular stimuli activating full-length RasGRF1, as well as constitutive signaling from post-translationally modified RasGRF1, such as the predominantly expressed 96 kDa carboxy-terminal fragment we observe in synovial tissue. Identification of the protease(s) responsible for RasGRF1 cleavage in vivo may lead to new therapeu-tic strategies in the treatment of arthritis.

Conclusions

RasGRF1 expression and post-translational modifications regulate spontaneous MMP-3 production in RA FLS, and is associated with MMP-3 production in RA sy-novial tissue. Contributions of RasGRF1 to MMP-3 production in RA and other forms of arthritis will likely depend upon 1) RasGRF1 expression levels, 2) the extent of activating post-translational modifications of RasGRF1, and 3) the strength of ex-tracellular stimuli leading to activation of residual full-length RasGRF1. Our data suggest a molecular mechanism by which Ras signaling pathways might contribute to the semi-transformed and invasive phenotype of RA FLS in the absence of onco-genic mutations in Ras superfamily GTPases.

Acknowledgments

We would like to thank Dr. TJM Smeets and Ms. M. Vinkenoog for assistance with digital image analysis experiments, and D. Groot for technical assistance. This re-search was supported by a Dutch Arthritis Association project grant (NR 04-1-301) to KAR.

130

Chapter 5

References

Tak PP, Smeets TJ, Daha MR et al. Analysis of the synovial cell infiltrate in early 1. rheumatoid synovial tissue in relation to local disease activity. Arthritis Rheum. 1997;40:217-225.Tak PP, Zvaifler NJ, Green DR, Firestein GS. Rheumatoid arthritis and p53: how 2. oxidative stress might alter the course of inflammatory diseases. Immunol.To-day 2000;21:78-82.Fassbender HG. Histomorphological basis of articular cartilage destruction in 3. rheumatoid arthritis. Coll.Relat Res. 1983;3:141-155.Lafyatis R, Remmers EF, Roberts AB et al. Anchorage-independent growth of sy-4. noviocytes from arthritic and normal joints. Stimulation by exogenous platelet-derived growth factor and inhibition by transforming growth factor-beta and retinoids. J.Clin.Invest 1989;83:1267-1276.Huber LC, Distler O, Tarner I et al. Synovial fibroblasts: key players in rheuma-5. toid arthritis. Rheumatology.(Oxford) 2006;45:669-675.Roivainen A, Zhu F, Sipola E, Yli-Jama T, Toivanen P. Failure to verify H-ras 6. mutations in arthritic synovium: comment on the article by Roivainen et al. Arthritis Rheum. 2001;44:2705.Pap T, Franz JK, Hummel KM et al. Activation of synovial fibroblasts in rheu-7. matoid arthritis: lack of Expression of the tumour suppressor PTEN at sites of invasive growth and destruction. Arthritis Res. 2000;2:59-64.Sweeney SE, Firestein GS. Signal transduction in rheumatoid arthritis. Curr.8. Opin.Rheumatol. 2004;16:231-237.Colicelli J. Human RAS superfamily proteins and related GTPases. Sci.STKE. 9. 2004;2004:RE13.Bos JL, Rehmann H, Wittinghofer A. GEFs and GAPs: critical elements in the 10. control of small G proteins. Cell 2007;129:865-877.Raaijmakers JH, Bos JL. Specificity in Ras and Rap signaling. J.Biol.Chem. 200811. Jaffe AB, Hall A. Rho GTPases: Biochemistry and biology. Annual Review of Cell 12. and Developmental Biology 2005;21:247-269.Iden S, Collard JG. Crosstalk between small GTPases and polarity proteins in 13. cell polarization. Nat.Rev.Mol.Cell Biol. 2008;9:846-859.Trabandt A, Aicher WK, Gay RE et al. Expression of the collagenolytic and Ras-14. induced cysteine proteinase cathepsin L and proliferation-associated oncogenes in synovial cells of MRL/I mice and patients with rheumatoid arthritis. Matrix 1990;10:349-361.

131


Roivainen A, Soderstrom KO, Pirila L et al. Oncoprotein expression in human 15. synovial tissue: an immunohistochemical study of different types of arthritis. Br.J.Rheumatol. 1996;35:933-942.Tak PP, Firestein GS. NF-kappaB: a key role in inflammatory diseases. J.Clin.16. Invest 2001;107:7-11.Schett G, Tohidast-Akrad M, Smolen JS et al. Activation, differential localiza-17. tion, and regulation of the stress-activated protein kinases, extracellular signal-regulated kinase, c-JUN N-terminal kinase, and p38 mitogen-activated protein kinase, in synovial tissue and cells in rheumatoid arthritis. Arthritis Rheum. 2000;43:2501-2512.Zhang HG, Wang Y, Xie JF et al. Regulation of tumor necrosis factor alpha-medi-18. ated apoptosis of rheumatoid arthritis synovial fibroblasts by the protein kinase Akt. Arthritis Rheum. 2001;44:1555-1567.Remans PH, Gringhuis SI, van Laar JM et al. Rap1 signaling is required for sup-19. pression of Ras-generated reactive oxygen species and protection against oxida-tive stress in T lymphocytes. J.Immunol. 2004;173:920-931.Remans PH, Wijbrandts CA, Sanders ME et al. CTLA-4IG suppresses reactive 20. oxygen species by preventing synovial adherent cell-induced inactivation of Rap1, a Ras family GTPASE mediator of oxidative stress in rheumatoid arthritis T cells. Arthritis Rheum. 2006;54:3135-3143.Yamamoto A, Fukuda A, Seto H et al. Suppression of arthritic bone destruction 21. by adenovirus-mediated dominant-negative Ras gene transfer to synoviocytes and osteoclasts. Arthritis Rheum. 2003;48:2682-2692.Pap T, Nawrath M, Heinrich J et al. Cooperation of Ras- and c-Myc-dependent 22. pathways in regulating the growth and invasiveness of synovial fibroblasts in rheumatoid arthritis. Arthritis Rheum. 2004;50:2794-2802.Na HJ, Lee SJ, Kang YC et al. Inhibition of farnesyltransferase prevents colla-23. gen-induced arthritis by down-regulation of inflammatory gene expression through suppression of p21(ras)-dependent NF-kappaB activation. J.Immunol. 2004;173:1276-1283.Layer K, Lin G, Nencioni A et al. Autoimmunity as the consequence of a sponta-24. neous mutation in Rasgrp1. Immunity. 2003;19:243-255.Cedeno S, Cifarelli DF, Blasini AM et al. Defective activity of ERK-1 and ERK-2 25. mitogen-activated protein kinases in peripheral blood T lymphocytes from pa-tients with systemic lupus erythematosus: potential role of altered coupling of Ras guanine nucleotide exchange factor hSos to adapter protein Grb2 in lupus T cells. Clin.Immunol. 2003;106:41-49.

132

Chapter 5

Yasuda S, Stevens RL, Terada T et al. Defective expression of Ras guanyl nucleo-26. tide-releasing protein 1 in a subset of patients with systemic lupus erythemato-sus. J.Immunol. 2007;179:4890-4900.Zhu TN, He HJ, Kole S et al. Filamin A-mediated Down-regulation of the Ex-27. change Factor Ras-GRF1 Correlates with Decreased Matrix Metalloproteinase-9 Expression in Human Melanoma Cells. J.Biol.Chem. 2007;282:14816-14826.Jones MK, Jackson JH. Ras-GRF activates Ha-Ras, but not N-Ras or K-Ras 4B, 28. protein in vivo. J.Biol.Chem. 1998;273:1782-1787.Kiyono M, Satoh T, Kaziro Y. G protein beta gamma subunit-dependent Rac-gu-29. anine nucleotide exchange activity of Ras-GRF1/CDC25(Mm). Proc.Natl.Acad.Sci.U.S.A 1999;96:4826-4831.Innocenti M, Zippel R, Brambilla R, Sturani E. CDC25(Mm)/Ras-GRF1 regulates 30. both Ras and Rac signaling pathways. FEBS Lett. 1999;460:357-362.Baouz S, Jacquet E, Bernardi A, Parmeggiani A. The N-terminal moiety of 31. CDC25(Mm), a GDP/GTP exchange factor of Ras proteins, controls the activity of the catalytic domain. Modulation by calmodulin and calpain. J.Biol.Chem. 1997;272:6671-6676.Kraan MC, Reece RJ, Smeets TJ et al. Comparison of synovial tissues from the 32. knee joints and the small joints of rheumatoid arthritis patients: Implications for pathogenesis and evaluation of treatment. Arthritis Rheum. 2002;46:2034-2038.Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Associa-33. tion 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988;31:315-324.Altman R, Asch E, Bloch D et al. Development of criteria for the classification 34. and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Di-agnostic and Therapeutic Criteria Committee of the American Rheumatism As-sociation. Arthritis Rheum. 1986;29:1039-1049.Inman RD. Classification criteria for reactive arthritis. J.Rheumatol. 1999;26:1219-35. 1221.Tak PP, van der Lubbe PA, Cauli A et al. Reduction of synovial inflammation 36. after anti-CD4 monoclonal antibody treatment in early rheumatoid arthritis. Arthritis Rheum. 1995;38:1457-1465.Haringman JJ, Vinkenoog M, Gerlag DM et al. Reliability of computerized image 37. analysis for the evaluation of serial synovial biopsies in randomized controlled trials in rheumatoid arthritis. Arthritis Res.Ther. 2005;7:R862-R867.Kasperkovitz PV, Timmer TC, Smeets TJ et al. Fibroblast-like synoviocytes de-38. rived from patients with rheumatoid arthritis show the imprint of synovial tis-

133


sue heterogeneity: evidence of a link between an increased myofibroblast-like phenotype and high-inflammation synovitis. Arthritis Rheum. 2005;52:430-441.Mook OR, Baas F, De Wissel MB, Fluiter K. Evaluation of locked nucleic acid-mo-39. dified small interfering RNA in vitro and in vivo. Mol.Cancer Ther. 2007;6:833-843.Chan A, Akhtar M, Brenner M et al. The GTPase Rac regulates the proliferation 40. and invasion of fibroblast-like synoviocytes from rheumatoid arthritis patients. Molecular Medicine 2007;13:297-304.Ohba Y, Mochizuki N, Yamashita S et al. Regulatory proteins of R-Ras, TC21/R-41. Ras2, and M-Ras/R-Ras3. J.Biol.Chem. 2000;275:20020-20026.Overbeck AF, Brtva TR, Cox AD et al. Guanine nucleotide exchange factors: acti-42. vators of Ras superfamily proteins. Mol.Reprod.Dev. 1995;42:468-476.Warstat K, Pap T, Klein G, Gay S, Aicher WK. Co-activation of synovial fibro-43. blasts by laminin-111 and transforming growth factor-beta induces expression of matrix metalloproteinases 3 and 10 independently of nuclear factor-kappaB. Ann.Rheum.Dis. 2008;67:559-562.Kunisch E, Gandesiri M, Fuhrmann R et al. Predominant activation of MAP ki-44. nases and pro- destructive/pro-inflammatory features by TNF-alpha in early-passage, rheumatoid arthritis and osteoarthritis synovial fibroblasts via tumor necrosis factor receptor- 1: Failure of p38 inhibition to suppress matrix metal-loproteinase-1 in rheumatoid arthritis. Ann.Rheum.Dis. 2007Jones SW, Brockbank SM, Clements KM et al. Mitogen-activated protein kinase-45. activated protein kinase 2 (MK2) modulates key biological pathways associated with OA disease pathology. Osteoarthritis.Cartilage. 2009;17:124-131.Ahn JK, Koh EM, Cha HS et al. Role of hypoxia-inducible factor-1alpha in hypo-46. xia-induced expressions of IL-8, MMP-1 and MMP-3 in rheumatoid fibroblast-like synoviocytes. Rheumatology.(Oxford) 2008;47:834-839.Buttice G, Duterque-Coquillaud M, Basuyaux JP et al. Erg, an Ets-family mem-47. ber, differentially regulates human collagenase1 (MMP1) and stromelysin1 (MMP3) gene expression by physically interacting with the Fos/Jun complex. Oncogene 1996;13:2297-2306.Cullen PJ, Lockyer PJ. Integration of calcium and Ras signalling. Nat.Rev.Mol.48. Cell Biol. 2002;3:339-348.Pozza M, Guerra M, Manzini E, Calza L. A histochemical study of the rheuma-49. toid synovium: focus on nitric oxide, nerve growth factor high affinity receptor, and innervation. J.Rheumatol. 2000;27:1121-1127.Zhao C, Fernandes MJ, Prestwich GD et al. Regulation of lysophosphatidic acid 50.

134

Chapter 5

receptor expression and function in human synoviocytes: implications for rheu-matoid arthritis? Mol.Pharmacol. 2008;73:587-600.Nochi H, Tomura H, Tobo M et al. Stimulatory role of lysophosphatidic acid 51. in cyclooxygenase-2 induction by synovial fluid of patients with rheumatoid arthritis in fibroblast-like synovial cells. J.Immunol. 2008;181:5111-5119.Flood S, Parri R, Williams A, Duance V, Mason D. Modulation of interleukin-6 52. and matrix metalloproteinase 2 expression in human fibroblast-like synoviocy-tes by functional ionotropic glutamate receptors. Arthritis Rheum. 2007;56:2523-2534.

A Rac1 inhibitory peptide suppresses antibody pro-

duction and paw swelling in the murine collagen-in-duced arthritis model of

rheumatoid arthritis

JRF Abreu1, D deLaunay1, PB van Hennik2, AM van Stal-borgh2, JP ten Klooster2, ME Sanders1, KA Reedquist1,

MJ Vervoordeldonk1,3, PL Hordijk2, and PP Tak1

1Division of Clinical Immunology and Rheumatology and 2De-partment of Molecular Cell Biology, Sanquin Research and

Landsteiner Laboratory, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands. 3Arthrogen BV, Amster-

dam, The Netherlands

submitted for publication

6

139

Rac1 inhibition suppresses antibody production and paw swelling in CIA

Abstract

Introduction

The Rho family GTPase Rac1 regulates cytoskeletal rearrangements crucial for the re-cruitment, extravasation and activation of leukocytes at sites of inflammation. Rac1 signaling also promotes the activation and survival of lymphocytes and osteoclasts. Therefore, we assessed the ability of a cell-permeable Rac1 carboxy-terminal inhibi-tory peptide to modulate disease in mice with collagen-induced arthritis (CIA).

Methods

CIA was induced in DBA/1 mice, and either in early or chronic disease, mice were treated three times per week by intraperitoneal injection with control peptide or Rac1 inhibitory peptide. Effects on disease progression were assessed by measure-ment of paw swelling. Inflammation and joint destruction were examined by histo-logy and radiology. Serum levels of anti-collagen type II antibodies were measured by enzyme-linked immunosorben assay (ELISA). Results were analyzed using un-paired Student’s t-tests.

Results

Treatment of mice with Rac1 inhibitory peptide resulted in a decrease in paw swel-ling in early disease and to a lesser extent in more chronic arthritis. Of interest, while joint destruction was unaffected by Rac1 inhibitory peptide, anti-collagen type II antibody production was significantly diminished in treated mice, both in early and chronic arthritis.

Conclusions

The data suggest that targeting of Rac1 with the Rac1 carboxy-terminal inhibitory peptide may suppress autoantibody production in autoimmune disease. Whether this could translate into clinically meaningful improvement remains to be shown.

140

Chapter 6

Introduction

Rheumatoid arthritis (RA) is marked by de-regulated recruitment, activation, and retention of inflammatory white blood cells in affected joints1. Subsequent auto-antibody production, release of cytokines and cell-cell contacts may perpetuate in-flammation and lead to joint destruction through activation of stromal fibroblast-like synoviocytes (FLS) and osteoclasts2. Many of the cellular processes required for perpetuation of inflammation and joint destruction in RA are regulated by Rac GTPases, members of the Rho-like family of small GTPase signaling proteins3.Rac1 is ubiquitously expressed in mammalian tissues, while expression of Rac2 is limited to cells of hematopoietic lineage4;5. Rac GTPases are activated by a broad array of extracellular stimuli relevant to RA, including chemokines, lymphocyte an-tigen receptor ligation, inflammatory cytokines, and cell-cell adhesion6-11. Following activation, Rac proteins initiate multiple signaling pathways that regulate cytoske-letal rearrangements, kinase cascades needed for gene transcription, and assembly of the NADPH oxidase6-12. Transfection of active and dominant-negative mutants of Rac1, as well as genetic studies, have demonstrated that lymphocytes and neutrop-hils require Rac1 signaling for efficient polarized chemotactic responses and traffic-king in vivo13-19. Although macrophages do not require Rac1 and Rac2 function for chemotactic responses, macrophage invasion of tissue is dependent upon Rac1 and Rac220. Rac signaling is also important for productive interactions between lymp-hocytes and antigen presenting cells (APCs). After antigen recognition by T-cells, ezrin-radixin-moesin (ERM) proteins are dephosphorylated through a Rac1-depen-dent pathway, favoring relaxation of the cytoskeleton and subsequently promoting T-cell −APC conjugate formation21. Reciprocally, Rac activity in dendritic cells is re-quired for effective antigen presentation to T-cells and subsequent T-cell priming22. Antigen receptor-dependent activation of Rac signaling also stimulates activation of mitogen-activated protein kinase, phosphatidylinositol 3-kinase, and NF-κB signa-ling pathways important for lymphocyte activation, proliferation, and survival8-10. Many of these downstream signaling pathways are now being explored as potential therapeutic targets in RA23. Rac proteins also serve additional important functions in cells of myeloid lineage which contribute to inflammation and joint destruction in RA. Oxidative bursts of macrophages and neutrophils rely upon Rac1-depen-dent assembly of the NADPH oxidase machinery12. Additionally, in vitro studies of osteoclasts transfected with plasmid encoding dominant-negative Rac, and in vivo studies in Rac-deficient mice have identified essential but redundant roles for Rac1 and Rac2 proteins in osteoclastogenesis, osteoclast motility, and bone resorption24;25.

141


Together, these studies indicate that therapeutic strategies targeting Rac1 function may be of clinical benefit in RA. However, pre-clinical assessment of Rac1 inhibi-tion has been hampered by a lack of compounds specifically targeting Rac1, and limited analyses of Rac1 in animal models of arthritis, a consequence of early fin-dings demonstrating that genetic deletion of Rac1 in mice results in early embryonic lethality26. NSC23766, a pharmacological compound which inhibits all Rac GTPa-ses, suppresses RA FLS proliferation and invasiveness in vitro, effects mimicked by siRNA silencing of Rac1 expression in these cells27. This may indicate that specific inhibition of Rac1 may be therapeutically beneficial in RA. However, mice in which Rac1 has been conditionally deleted in mature neutrophils and macrophages, on a Rac2-deficient background, show a complex phenotype in a Chlamydia-induced infection model of arthritis28. In these animals, Rac1 has a bimodal effect on disease progression. In the acute phase, Rac1-deficiency delays recruitment and activation of inflammatory neutrophils in the joint, while in the chronic phase, disease is exa-cerbated due to an inability of neutrophils to clear the pathogen. In this study, we targeted Rac1 in mice with CIA, using a Rac1-specific cell-permeable carboxy-termi-nal inhibitory peptide which we have previously shown to block Rac1 function in human lymphocytes, endothelial cells and epithelial cells11;29;30.

Materials and methods

Animals

Male DBA/1 mice were purchased from Harlan (Horst, The Netherlands), housed under conventional conditions at the animal facility of the Academic Medical Center (Amsterdam, The Netherlands), and fed ad libitum. The animal ethical committee of the Academic Medical Center approved all experiments.

Peptide synthesis

For this study, peptides encoding a protein transduction domain31 alone (indicated as control, Ctrl throughout the manuscript) or fused to the carboxy-terminal domain of Rac1, excluding the Rac1 CAAX box (indicated as Rac1 throughout the manus-cript), were synthesized using N-(9-fluorenyl)methoxycarbonyl (fMoc) solid phase

142

Chapter 6

chemistry30. Peptide synthesis was performed using a Syro II (MultiSynTec GmbH, Bochum, Germany).

T-cell isolationMurine spleens were crushed through a 40-µm cell strainer (BD Pharmingen, Fran-klin Lakes, NJ) to obtain single cell suspensions. Erythrocytes were lysed with ice-cold isotonic NH4Cl solution (155 mM NH4Cl, 10 mM KHCO3, and 100 mM EDTA, pH 7.4). To purify T-cells, splenic cell suspensions were incubated with anti-muri-ne CD4 and CD8 antibody-coated magnetic microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany) for 20 min at 4°C and positively selected by magnetic separa-tion with MACS. Purified T-cells were >90% CD3+, as analyzed by flow cytometry.

Actin polymerization assays

Actin polymerization assays were performed as previously described32. Briefly, pu-rified T-cells were pre-incubated for 15 minutes in medium containing 200 µg/ml Ctrl or Rac1 peptide. T-cells were then exposed to 100 ng/ml SDF-1, and at the indi-cated time points 100 µl aliquots of cell suspensions were transferred to an equiva-lent volume of fixation solution (Intraprep Fixation Reagent, Coulter Immunotech, Marseille, France). After 15 minutes, cells were washed in 0.5% bovine serum albu-min/phosphate-buffered saline (PBS) and resuspended in 100 µl of permeabiliza-tion reagent (Coulter Immunotech) for 5 minutes. Cells were stained for 20 minutes with 1 unit/ml Alexa 488 Phalloidin (Invitrogen, Molecular Probes, Eugene, OR) to visualize F-actin. The mean fluorescence intensity (MFI) of polymerized actin was measured by FACS (BD Biosciences), and the fold increase in actin polymerization was calculated by dividing the MFI generated at a particular time point by the MFI at t = 0 of that particular condition.

Induction and assessment of CIA

Bovine collagen type II (bCII) (2 mg/ml in 0.05 M acetic acid, Chondrex, Inc., Red-mond, WA) was mixed with complete Freund's adjuvant (CFA) (2 mg/ml of Myco-bacterium tuberculosis, Chondrex, Inc.) and injected intradermally on day 0 at the base of the tail with 100 µl of emulsion into 8- to 11-week old mice. On day 21, mice received an intraperitoneal (ip) booster injection with 100 µg of bCII in PBS. To in-

143


vestigate the treatment efficacy of Rac1 peptide at disease onset, mice were treated at day 20 with either 2mg, 1mg or 0.5mg of Ctrl or Rac1 peptide, three times weekly until sacrifice. Alternatively, to explore the effect of Rac1 peptide treatment in chro-nic disease, the animals were randomly assigned at day 29 to one of two groups, and treated ip with 4 mg Ctrl, or Rac1 peptide. Treatments were continued three times weekly until sacrifice at day 39. The severity of arthritis was assessed in a blinded manner, using a semi-quantitative scoring system (0 to 4): 0, normal; 1, redness and/or swelling in one joint; 2, redness and/or swelling in more than one joint; 3, redness and/or swelling in the entire paw; and 4, deformity and/or ankylosis. Hind paw ankle joint thickness was measured using a dial caliper (POCO 2T 0- to 10-mm test gauge; Kroeplin Längenmesstechnik, Schlüchtern, Germany). Experiments were performed using 8–16 mice per group.

Histologic analysis

Hind paws were fixed in 10% buffered formalin for 48 hours and decalcified in 15% ethylenediaminetetraacetic acid (EDTA). The paws were then embedded in paraffin, and 5-µm saggital serial sections of whole hind paws were cut. Tissue sections were stained with hematoxylin and eosin. Inflammation was graded on a scale from 0 (no inflammation) to 3 (severely inflamed joint) based on infiltration of the synovium by inflammatory cells. Cartilage erosion was scored using a semi-quantitative scoring system from 0 (no erosions) to 3 (extended erosions). The tissue was examined by microscopic evaluation in a blinded manner by two independent observers (JA and MV).

Radiological analysis

Hind paws were used for radiographic evaluation. Two observers without know-ledge of the treatment groups scored the X-rays. Joint destruction was scored on a scale from 0 to 4: 0, no damage; 1, minor bone destruction observed in one enligh-tened spot; 2, moderate changes, two to four spots in one area; 3, severe erosions afflicting the joint; and 4, complete destruction of the joints.

144

Chapter 6

Determination of anti-collagen antibodies by ELISA

Maxisorb 96-well plates (Nunc, Roskilde, Denmark) were coated with 5 µg/ml of bCII in 0.1 M sodium carbonate buffer (pH 9.7) overnight at 4°C. After blocking for 1 h with 2% milk in PBS at room temperature, sera were added in serial dilu-tions in 2% milk/PBS, and incubated overnight at 4°C. Plates were subsequently washed and incubated with 1 µg/ml biotinylated rat anti-mouse immunoglobulin (Ig) (Southern Biotechnology Associates, Birmingham, AL) of the indicated isotype in 2% milk/PBS for 1 h at room temperature. After washing, plates were incubated with streptavidin-conjugated alkaline phosphatase (Jackson ImmunoResearch, Ne-wmarket, Suffolk, UK) for 1h at room temperature, washed, and developed with p-nitrophenyl phosphate substrate (Sigma-Aldrich, St Louis, MO). The reaction was stopped with 2M H2SO4, and optical density (OD) at 415 nm was measured.

F-actin (MFI)F-actin (MFI)

Rac1Ctrl

Cou

nts

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

0 15 30 60 120 300

Time after SDF-1 (sec)

F-ac

tin(fo

ldin

crea

se) Ctrl

Rac1

**

**

Figure 1. The Rac1 carboxy-terminal peptide efficiently blocks actin polymerization in mu-rine cells. Representative histograms of F-actin staining in T-cells isolated from mice spleens that were exposed for 15 min to 200 ug/ml of Ctrl or Rac1 peptide followed by SDF-1 stimulation (a). Fold increase in F-actin (b). Data are depicted as mean fold increase of F-actin ± standard error of the mean (SEM) (n=3); *P < 0.05.

B

A

145


Statistical analysis

To evaluate the effects of different treatments, we determined the change in paw swelling scores (delta) of each mouse from the start of treatment until the end of the experiment. Areas under the curve (AUC) were calculated for the delta paw swel-ling. The significance of the differences between the means of delta paw swelling, radiological, and histologic scores between groups was determined by using the unpaired Student’s t-test. P values ≤ 0.05 were considered statistically significant.

Results

The Rac1 inhibitory peptide blocks murine T-cell actin polymerization

The Rac1 peptide is able to block endogenous Rac1 signaling within minutes by

-0.2

0.0

0.2

0.4

0.6

0.8

20 25 30 35 40 45


Del

ta P

awSw

ellin

g

CtrlRac1

0.0

0.2

0.4

0.6

0.8

1.0

20 25 30 35 40 45


Del

ta P

awSw

ellin

g

CtrlRac1

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

20 25 30 35 40 45


Del

ta P

awSw

ellin

g

CtrlRac1

2 mg 1 mg 0.5 mg

2 mg 1 mg 0.5 mg

*

0

5

10

15

Ctrl Rac1

AU

C

0

5

10

15

Ctrl Rac1

AU

C

0

5

10

15

Ctrl Rac1

AU

C

B

A

Figure 2. Reduced paw swelling after treatment with 2mg of Rac1 carboxy-terminal peptide. Mice were treated with 2 mg, 1 mg or 0.5 mg Ctrl (□) or Rac1 (■) peptide at the indicated time points. Paw swelling and inflammation of the four limbs were determined for each mouse. Delta hind paw ankle joint swelling was calculated by subtracting the paw diameter on the day of initiation of treatment from the measured diameter (a). Values are depicted as mean ± SEM (n=8). Area under the curve (AUC) was cal-culated for the delta paw swelling in each mouse (b). Values are depicted as mean AUC ± SEM. * P < 0.01.

146

Chapter 6

competing with Rac1 effector proteins, and in vitro has been demonstrated to pos-sess a potent capacity to block actin polymerization and migration of human cells stimulated with SDF-111;29;30. The biological activity of the peptide batches used for in vivo experiments, and their ability to influence murine cellular responses, was first examined in an actin polymerization assay using murine splenic T-cells. Pre-treat-ment of murine T-cells with Rac1 peptide, but not Ctrl peptide, completely blocked actin polymerization following SDF-1 stimulation, as measured by increases in T cell F-actin content (Figures 1A and 1B). This indicated that the Rac1 peptide was effec-tive in blocking Rac1 signaling not only in human cells, but murine cells as well.

Rac1 inhibitory peptide treatment reduces paw swelling and anti-bCII antibody production in early arthritis

After confirming the in vitro efficiency of the Rac1 peptide in inhibiting murine Rac1

0

1

2

3

Ctrl Rac1

Infil

trat

esc

ore

0

1

2

3

Ctrl Rac1

Car

tilag

eer

osio

nsc

ore

0

1

2

3

Ctrl Rac1

Rad

iolo

gica

lsco

re

* *

0

20

40

60

80

100

120

140

160

IgG1 IgG2a IgG2b IgG3

Rel

ativ

eIg

Gpr

oduc

tion

(%) Ctrl

Rac1

Figure 3. Treat-ment with the Rac1 carboxy-terminal peptide at onset of disease reduces anti-bovine collagen IgG production but does not protect against joint destruction. Mice were treated with Ctrl (white bars) or Rac1 (black bars) pep-tide. Sections from mice paws were stained with haematoxylin and eosin (n=8). X-rays of hind paws were analyzed for

for bone damage (n=8). Cellular infiltration scores (a). Cartilage erosion scores (b). X-ray scores sho-wing no differences in bone damage between groups (c). Values are depicted as mean ± SEM. Serum from mice that started treatment at day 20 (n=8) with Ctrl (white bars) or Rac1 (black bars) peptide were collected and the levels of specific anti-collagen IgG detected (d). IgG levels in the serum of Ctrl-treated mice were set to 100% and the levels obtained in the serum of Rac1 -treated mice were then calculated relative to Ctrl, and expressed as mean ± SEM. Represented values were calculated within linear regions of the serum dilution curve. *P ≤ 0.05.

BA

DC

147


signaling, we examined the in vivo potential of this peptide when mice are treated at the onset of disease. One day before the booster, at day 20, we started treatment of the animals with 2 mg, 1 mg or 0.5 mg Ctrl or Rac peptide, three times weekly, until sacrificing. Treatment of animals with Rac1 peptide at all doses failed to influ-ence disease incidence, clinical scores of disease severity or animal weight (data not shown) but animals treated with 2 mg of Rac1 peptide showed a highly significant decrease in paw swelling when compared to treatment with Ctrl peptide (61% re-duction, P = 0.009) (Figures 2A and 2B). The effect was dose-dependent; treatment with lower doses of Rac1 peptide showed a trend towards improvement, which did not reach statistical significance (Figures 2A and 2B).We next examined the influence of Rac1 peptide on synovial inflammation, cartilage degradation, and bone destruction. Quantification of synovial inflammation revea-led a minor decrease in cellularity, which did not reach statistical significance, pos-sibly due to the relatively small number of mice (Figure 3A). Treatment with Rac1 peptide did not protect against joint destruction (Figures 3B and 3C).Finally, we examined the influence of Rac1 peptide on anti-bCII antibody produc-tion. We collected sera from mice at the time of sacrifice and measured specific anti-bCII antibody levels by ELISA. Mice treated with Rac1 peptide showed a significant

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

29 31 33 35 37 39


Del

ta p

awsw

ellin

g(m

m)

Ctrl

Rac1

Ctrl Rac10

1

2

3

AU

C

Figure 4. Effect of Rac1 carboxy-terminal peptide treatment of mice with chronic arthritis on paw swelling. Hind paw swelling was monitored during the course of disease using calipers. Del-ta hind paw ankle joint swelling of mice treated at day 29, three times weekly, with 4 mg Ctrl or Rac1 peptide (a). Values are depicted as mean ± SEM and are representative of two independent experiments (n= 16 per group). AUC calculated as in Figure 2b (b).

BA

148

Chapter 6

reduction in the serum levels of anti-bCII IgG1 (Ctrl 100% ± 12.9; Rac1 62.1% ± 11.8; P < 0.05) and IgG2a (Ctrl 100% ± 2.7; Rac1 83.3% ± 6.8; P = 0.05) antibodies compared to mice treated with Ctrl peptide (Figure 3D).

Treatment of chronic CIA with Rac1 peptide reduces anti-bCII antibody produc-tion

We next investigated the effect of Rac1 peptide treatment of mice with chronic

0.0

0.4

0.8

1.2

1.6

2.0

2.4

Ctrl Rac1

Infil

trat

esc

ore

0.0

0.4

0.8

1.2

1.6

2.0

2.4

Ctrl Rac1Car

tilag

eer

osio

nsc

ore

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Ctrl Rac1

Rad

iolo

gica

lsco

re

B

A

D

C

Ctrl Rac1

Figure 5. Rac1 carboxy-terminal peptide treatment of mice with chronic arthritis does not protect against joint destruction. Mice were treated with Ctrl (white bars) or Rac1 peptide (black bars). Paraffin-embedded sections of the hind paws were stained with haematoxylin and eosin and ana-lyzed for synovial inflammation and cartilage destruction. X-rays were analyzed for bone damage (n = 16 per group). Representative pictures of joints showing extensive cellular infiltration and cartilage erosions (a). Cellular infiltration scores (b). Cartilage erosion scores (c). X-rays from mice paws were taken and scored for bone damage (d). Values are depicted as mean ± SEM.

149


arthritis. For this we performed an independent CIA experiment wherein 29 days after the initial immunization, mice having clinical signs of arthritis were randomly assigned to one of two groups. Mice within groups were treated with 4 mg of Ctrl or 4 mg of Rac1 peptide, three times weekly, until sacrificing. Administration of Rac1 peptide had no influence on clinical disease severity (data not shown). However, we observed a clear trend towards reduced paw swelling in mice treated with Rac1 peptide (Figure 4A and 4B) in two independent experiments, although the diffe-rences did not reach statistical significance (experiment 1: 49% reduction, P = 0.528; experiment 2: 22% reduction, P = 0.193) (Figure 4B).We also analyzed the effects of the Rac1 peptide on synovial inflammation, cartilage degradation, and bone destruction. Consistent with the trend towards reduced paw swelling, there was a trend towards decreased histologic signs of inflammation and cartilage destruction, which did not reach statistical significance (Figure 5A, 5B and 5C). Analysis of X-rays taken from the mice paws revealed that Rac1 peptide treat-ment did not protect against erosive disease (Figure 5D). However, Rac1 peptide treatment of mice with chronic arthritis resulted in a significant reduction in the se-rum levels of anti-bCII IgG2a (Ctrl 100% ± 5.6; Rac1 81.6% ± 6.5; P = 0.05) and IgG2b (Ctrl 100% ± 10.9; Rac1 56.0% ± 6.7; P < 0.005), whereas no differences were observed for IgG1 or IgG3 (Figure 6).

* *

0

20

40

60

80

100

120

140

IgG1 IgG2a IgG2b IgG3

Rel

ativ

eIg

Gpr

oduc

tion

(%)

CtrlRac1

Figure 6. Rac1 carboxy-terminal peptide treatment of mice with chronic arthritis reduces anti-collagen IgG production. Serum from mice that started treatment at day 29 (n=7) with Ctrl (white bars) or Rac1 peptide (black bars) were collected and the levels of specific anti-collagen IgG detected. IgG levels in the serum of Ctrl-treated mice were set to 100% and the levels obtained in the serum of Rac1 pep-tide-treated mice were then calcula-ted relative to Ctrl, and expressed

as mean ± SEM. Represented values were calculated within linear regions of the serum dilution curve. *P ≤ 0.05.

150

Chapter 6

Discussion

In this report we provide evidence that the inhibitory Rac1 carboxy-terminal pep-tide suppresses anti-collagen antibody production in mice with CIA, associated with reduced paw swelling. We have found that the administration of Rac1 peptide in vivo, either as an early treatment or as treatment of mice with more chronic arthritis, results in a significant reduction of circulating levels of anti-bCII IgG1 and IgG2a or IgG2a and IgG2b antibodies, respectively, dependent on the stage of the disease. In both early and chronic arthritis, there was an effect on IgG2a levels. In many murine in vivo model systems, IgG2a and IgG2b antibodies display greater pro-inflamma-tory properties than IgG1 and IgG333. Genetic studies indicate that the suppressive effect of Rac1 peptide on anti-collagen antibody production observed in our studies is unlikely to be due to direct effects on B-cell trafficking or activation. Rac1, toge-ther with Rac2, is critical in transducing B-cell receptor signals required for both survival and efficient cell cycle entry9;34. Rac2 deletion in mice results in decreased B-cell maturation and T-cell -independent antigen responses13;14. In contrast, condi-tional deletion of Rac1 in the B-cell compartment has no observable effect on B-cell maturation or function unless Rac2 is simultaneously deleted34. Similarly redundant but critical roles for Rac1 and Rac2 are observed in T-cells17. In addition, antibody production is critically dependent on efficient T-cell priming. Deletion of Rac1, but not Rac2, inhibits migration of CD8α+ dendritic cells (DCs) to secondary lymphoid organs22. Moreover, Rac-deficient DCs fail to establish stable contacts with naïve T-cells, leading to suboptimal T-cell priming22. DCs from mindin-/- mice, which have reduced expression of Rac1 and Rac2, also have impaired priming capacity due to inefficient engagement with T-cells, in turn leading to defective humoral responses to T-cell -dependent antigens in these mice35. This may suggest that Rac1 peptide suppresses anti-collagen antibody production in CIA via inhibition of lymphocyte interactions with antigen presenting cells. Surprisingly, we observed little if any effect of Rac1 peptide on cartilage and joint destruction in murine arthritis, although a significant decrease in paw swelling was observed in mice when treated at the onset of disease, and a reproducible trend to-wards reduced paw swelling was noted in mice treated with Rac1 peptide at a more chronic phase of disease. There was only a limited effect of Rac1 peptide treatment on synovial cellularity, which might be in part due to the often redundant roles of Rac1 and Rac2 in lymphocytes and myeloid lineage cells, and the high specificity of the Rac1 peptide for interfering with signaling from Rac1 but not other Rho family GTPases30. Murine neutrophils express both Rac1 and Rac2, and genetic deletion of

151


each GTPase has revealed their important but redundant contributions to neutrophil chemotactic responses in vitro and in vivo16;19. Rac1-deficient murine neutrophils retain chemokinetic responses, but are unable to orient and migrate towards che-mokine gradients. In contrast, Rac2 is required for efficient neutrophil migration19. Although similar direct comparative analyses have not been performed on B and T -cells in these mice, initial studies indicate that lymphocyte trafficking is regulated primarily by Rac213. Lack of effect of Rac1 peptide on joint destruction in CIA may reflect a redundant role for Rac2 in supporting osteoclastogenesis24;25. Decreases in paw swelling after treatment with Rac1 peptide which we observed might be due to effects on the formation of edema. Experiments conducted in vitro have demonstrated that exposure of human endothelial cells to reactive oxygen spe-cies, or engagement of the integrin ligand VCAM, leads to Rac1 dependent loss of cadherin-mediated endothelial cell-cell adhesion. In the presence of the Rac1 pepti-de, endothelial cell-cell adhesion is maintained (data not shown)11;36. A role for Rac1 in maintaining vascular endothelial integrity in vivo is also indirectly suggested in studies of c-Jun knockout mice, where inhibition of c-Jun, a downstream target of Rac signaling, suppresses edema, paw swelling and inflammation in an experimen-tal model of arthritis37. Our results suggest that future studies should also consider the potential contributions of Rac2, independently or in conjunction with Rac1, to pathology in CIA. Structure-based studies have recently led to the development of small molecular weight compounds which can specifically prevent interaction of Rac1 and Rac2 with activating guanine nucleotide exchange factors27;38. These compounds can block RA FLS growth and matrix invasion in vitro, although their efficacy in the treatment of arthritis in vivo remains to be established.

Conclusions

We demonstrate that a cell-permeable inhibitory Rac1 carboxy-terminal peptide can reduce paw swelling and antibody production during murine experimental arthritis. Conceivably, Rac1 peptide treatment could augment the pharmacologic activity to-ward B-lineage cells of other immunosuppressive therapies, like rituximab or ataci-cept, which may theoretically increase therapeutic activity. An alternative approach that might perhaps result in a beneficial effect both on clinical signs and symptoms, as well as joint destruction could be the suppression of Rac signaling in RA by com-pounds targeting both Rac1 and Rac2 signaling. The present study supports the rationale for future studies exploring these approaches.

152

Chapter 6

Acknowledgments

Grant support: This research was supported by Dutch Arthritis Association project grants (NR 04-1-301) to KAR and (NR 05-2-102) to PLH.

153


References

Vergunst CE, van de Sande MG, Lebre MC, Tak PP. The role of chemokines in 1. rheumatoid arthritis and osteoarthritis. Scand.J.Rheumatol. 2005;34:415-425.McInnes IB, Schett G. Cytokines in the pathogenesis of rheumatoid arthritis. 2. Nat.Rev.Immunol. 2007;7:429-442.Jaffe AB, Hall A. Rho GTPases: biochemistry and biology. Annu.Rev.Cell Dev.3. Biol. 2005;21:247-269.Moll J, Sansig G, Fattori E, van der PH. The murine rac1 gene: cDNA cloning, 4. tissue distribution and regulated expression of rac1 mRNA by disassembly of actin microfilaments. Oncogene 1991;6:863-866.Shirsat NV, Pignolo RJ, Kreider BL, Rovera G. A member of the ras gene super-5. family is expressed specifically in T, B and myeloid hemopoietic cells. Oncogene 1990;5:769-772.Bokoch GM. Regulation of innate immunity by Rho GTPases. Trends Cell Biol. 6. 2005;15:163-171.Coso OA, Chiariello M, Yu JC et al. The small GTP-binding proteins Rac1 7. and Cdc42 regulate the activity of the JNK/SAPK signaling pathway. Cell 1995;81:1137-1146.Genot EM, Arrieumerlou C, Ku G et al. The T-cell receptor regulates Akt (pro-8. tein kinase B) via a pathway involving Rac1 and phosphatidylinositide 3-kinase. Mol.Cell Biol. 2000;20:5469-5478.Hashimoto A, Okada H, Jiang A et al. Involvement of guanosine triphosphata-9. ses and phospholipase C-gamma2 in extracellular signal-regulated kinase, c-Jun NH2-terminal kinase, and p38 mitogen-activated protein kinase activation by the B cell antigen receptor. J.Exp.Med. 1998;188:1287-1295.Marinari B, Costanzo A, Viola A et al. Vav cooperates with CD28 to induce NF-10. kappaB activation via a pathway involving Rac-1 and mitogen-activated kinase kinase 1. Eur.J.Immunol. 2002;32:447-456.van Wetering S, van den BN, van Buul JD et al. VCAM-1-mediated Rac signaling 11. controls endothelial cell-cell contacts and leukocyte transmigration. Am.J.Physiol Cell Physiol 2003;285:C343-C352.Hordijk PL. Regulation of NADPH oxidases: the role of Rac proteins. Circ.Res. 12. 2006;98:453-462.Croker BA, Tarlinton DM, Cluse LA et al. The Rac2 guanosine triphosphatase 13. regulates B lymphocyte antigen receptor responses and chemotaxis and is re-

154

Chapter 6

quired for establishment of B-1a and marginal zone B lymphocytes. J.Immunol 2002;168:3376-3386.Croker BA, Handman E, Hayball JD et al. Rac2-deficient mice display perturbed 14. T-cell distribution and chemotaxis, but only minor abnormalities in T(H)1 re-sponses. Immunol.Cell Biol. 2002;80:231-240.del Pozo MA, Vicente-Manzanares M, Tejedor R, Serrador JM, Sanchez-Madrid 15. F. Rho GTPases control migration and polarization of adhesion molecules and cytoskeletal ERM components in T lymphocytes. Eur.J.Immunol. 1999;29:3609-3620.Glogauer M, Marchal CC, Zhu F et al. Rac1 deletion in mouse neutrophils has 16. selective effects on neutrophil functions. J.Immunol. 2003;170:5652-5657.Guo F, Cancelas JA, Hildeman D, Williams DA, Zheng Y. Rac GTPase isoforms 17. Rac1 and Rac2 play a redundant and crucial role in T-cell development. Blood 2008;112:1767-1775.Nijhara R, Van Hennik PB, Gignac ML et al. Rac1 mediates collapse of microvilli 18. on chemokine-activated T lymphocytes. J.Immunol. 2004;173:4985-4993.Sun CX, Downey GP, Zhu F et al. Rac1 is the small GTPase responsible for regu-19. lating the neutrophil chemotaxis compass. Blood 2004;104:3758-3765.Wheeler AP, Wells CM, Smith SD et al. Rac1 and Rac2 regulate macrophage mor-20. phology but are not essential for migration. J.Cell Sci. 2006;119:2749-2757.Faure S, Salazar-Fontana LI, Semichon M et al. ERM proteins regulate cytoske-21. leton relaxation promoting T cell-APC conjugation. Nat.Immunol. 2004;5:272-279.Benvenuti F, Hugues S, Walmsley M et al. Requirement of Rac1 and Rac2 expres-22. sion by mature dendritic cells for T cell priming. Science 2004;305:1150-1153.Tas SW, Remans PH, Reedquist KA, Tak PP. Signal transduction pathways and 23. transcription factors as therapeutic targets in inflammatory disease: towards in-novative antirheumatic therapy. Curr.Pharm.Des 2005;11:581-611.Fukuda A, Hikita A, Wakeyama H et al. Regulation of osteoclast apoptosis and 24. motility by small GTPase binding protein Rac1. J.Bone Miner.Res. 2005;20:2245-2253.Wang Y, Lebowitz D, Sun C et al. Identifying the relative contributions of Rac1 25. and Rac2 to osteoclastogenesis. J.Bone Miner.Res. 2008;23:260-270.Sugihara K, Nakatsuji N, Nakamura K et al. Rac1 is required for the formation of 26. three germ layers during gastrulation. Oncogene 1998;17:3427-3433.Chan A, Akhtar M, Brenner M et al. The GTPase Rac regulates the proliferation 27. and invasion of fibroblast-like synoviocytes from rheumatoid arthritis patients.

155


Mol.Med. 2007;13:297-304.Zhang X, Glogauer M, Zhu F et al. Innate immunity and arthritis: neutrophil 28. Rac and toll-like receptor 4 expression define outcomes in infection-triggered arthritis. Arthritis Rheum. 2005;52:1297-1304.Lorenowicz MJ, Fernandez-Borja M, van Stalborch AM et al. Microtubule dyna-29. mics and Rac-1 signaling independently regulate barrier function in lung epi-thelial cells. Am.J.Physiol Lung Cell Mol.Physiol 2007;293:L1321-L1331.Van Hennik PB, ten Klooster JP, Halstead JR et al. The C-terminal domain of 30. Rac1 contains two motifs that control targeting and signaling specificity. J.Biol.Chem. 2003;278:39166-39175.Ho A, Schwarze SR, Mermelstein SJ, Waksman G, Dowdy SF. Synthetic protein 31. transduction domains: enhanced transduction potential in vitro and in vivo. Cancer Res. 2001;61:474-477.Voermans C, Anthony EC, Mul E, van der SE, Hordijk P. SDF-1-induced actin 32. polymerization and migration in human hematopoietic progenitor cells. Exp.Hematol. 2001;29:1456-1464.Nimmerjahn F, Ravetch JV. Fcgamma receptors as regulators of immune respon-33. ses. Nat.Rev.Immunol. 2008;8:34-47.Walmsley MJ, Ooi SK, Reynolds LF et al. Critical roles for Rac1 and Rac2 GTPa-34. ses in B cell development and signaling. Science 2003;302:459-462.Li H, Oliver T, Jia W, He YW. Efficient dendritic cell priming of T lymphocy-35. tes depends on the extracellular matrix protein mindin. EMBO J. 2006;25:4097-4107.van Wetering S, van Buul JD, Quik S et al. Reactive oxygen species mediate Rac-36. induced loss of cell-cell adhesion in primary human endothelial cells. J.Cell Sci. 2002;115:1837-1846.Fahmy RG, Waldman A, Zhang G et al. Suppression of vascular permeability 37. and inflammation by targeting of the transcription factor c-Jun. Nat.Biotechnol. 2006;24:856-863.Gao Y, Dickerson JB, Guo F, Zheng J, Zheng Y. Rational design and characteriza-38. tion of a Rac GTPase-specific small molecule inhibitor. Proc.Natl.Acad.Sci.U.S.A 2004;101:7618-7623.

General discussion and summary

7

159


Rheumatoid arthritis (RA) is an auto-immune disease for which no cure is yet avai-lable. Recent insights into the pro-inflammatory cytokines and cells present in the inflamed joint, and their contributions to pathology in RA, have led to the develop-ment of biological therapies aimed at their neutralization. Anti-TNF-α blockade for example, is one of the most widely used and successful biological therapies1. There is a large group of patients however, that don’t respond sufficiently to treatment. Also, therapies with biologicals can have serious side effects as the immune sys-tem of treated patients become less efficient in clearing pathogens and opportunistic infections. Therefore, there is still the need for the development of new and more specific therapies. Cellular activation, proliferation, survival and migration are tightly regulated at the intracellular level by Ras-superfamily small GTPases. Small GTPases are important modulators of cellular responses and crucial mediators of effective immune respon-ses, and initial evidence has suggested these enzymes may contribute to pathology in RA. Ras family GTPase effector pathways such as MAP kinases, PI3K and NF-kB are highly activated in RA synovial tissue (ST)2-4. Moreover, strategies that block Ras function are able to suppress fibroblast-like synoviocyte (FLS) proliferation, invasi-veness and pro-inflammatory cytokine production in vitro, and exert protective ef-fects in animal models of arthritis in vivo5-7. Similarly, blockage of Rho family small GTPases reduces FLS proliferation, invasiveness, JNK activation and IL-6 producti-on in vitro8;9. Finally, inactivation of the Ras-related small GTPase Rap1 in synovial T cells is associated with increased intracellular oxidative stress10. Modulation of small GTPase signaling may therefore be an attractive tool to inhibit the inflammatory pro-cess in RA. In this thesis we analyzed the in vitro and in vivo effects of manipulating the activity of representative small GTPases in respect to inflammatory activation of cells relevant to RA, and animal models of this disease.

Much evidence indirectly suggests an active involvement of T cells in RA. Most im-portantly, the expression of the shared epitope (SE), thought to present self antigen(s) to autoreactive T cells, is one of the strongest indications of TCR involvement in RA11. Synovial T cells have a highly differentiated CD45RO+ phenotype, promote the activation of other synovial cells through cell-cell contacts, and in many animal models T cell dependency has been shown12;13. The clinical success of CTLA-4Ig the-rapy has suggested that TCR engagement may take place in RA, as disruption of the T cell costimulatory CD28 protein with its ligands CD80/86 is of clinical benefit14. However, despite this evidence, whether there is TCR triggering in RA is still a mat-

160

Chapter 7

ter of debate. Several studies have described impaired T cell responses in RA due to deficiencies downstream of the T cell receptor (TCR). Increases in intracellular oxidative stress due to Rap1 inactivation10, impaired phosphorylation of the adaptor protein linker for activation of T cells (LAT)15 and chronic downregulation of TCR-zeta expression16 are some of the defects described. Based on the inability to detect significant T cell cytokine production and proliferation in the synovium, together with the reported defects in TCR-proximal signaling responses, several studies have proposed that TCR signaling in RA is impaired17. In chapter 2 we analyze synovial T cell responses at a single cell level, and demonstrate that TCR signaling is fully functional and completely capable of initiating T cell proliferation and cytokine pro-duction. We provide evidence that the reported hyporesponsiveness in RA synovial T cell bulk cultures is secondary to ex vivo spontaneous apoptosis. Additionally, we find that the increase in susceptibility to apoptosis ex vivo is associated with altered ratios of pro-apoptotic Noxa and anti-apoptotic Mcl-1 expression. These data indi-cate that although there is no evidence of T cell apoptosis in vivo, once isolated from the synovium T cells are extremely susceptible to apoptosis. It will be interesting to address which soluble factors or specific cellular contacts are responsible for the in vivo inhibition of synovial apoptosis, as abrogation of these signals might exert a therapeutic benefit by selectively promoting T cell apoptosis at the site of inflam-mation.The data in chapter 2 clearly shows that synovial T cells are capable of supporting productive TCR signaling. Our results suggest that the inability to detect T cell cy-tokines and proliferation in the synovium of established RA might be due to the absence of TCR engagement, rather than a consequence of TCR signaling defects. In early stages of disease, where T cell cytokines are observed, TCR engagement may play a prominent role, as autoreactive T cells could initiate inflammation. Ho-wever in the effector phase of disease, T cells may contribute to inflammation by TCR-independent mechanisms. In this context it will be interesting to investigate the specific mechanism of action by which CTLA-4Ig therapy exerts its effects. In light of our results we propose that alternative mechanisms, independent of effects on classical TCR costimulation, might be responsible for the clinical improvements observed in treated patients. One of these mechanisms could be upregulation of im-munomodulatory IDO on CD80/86 expressing cells. Besides antigen presenting cells (APC), T cells in the synovium have also been shown to express functional CD80/86 costimulatory molecules18;19. In vitro studies have shown that triggering of CD80/86 on T cells by CTLA-4 ligation is capable of upregulating IDO expression, reducing their capacity to respond to antigen triggering20. Whether T cells in the synovium,

161


APCs, or both are targeted in this manner by CTLA-4 Ig therapy remains to be de-termined.

The small GTPase Rap1 is transiently activated upon TCR triggering and is an im-portant regulator of immune responses. In RA synovial fluid (SF) T cells a block in Rap1 activation is believed to underlie the altered T cell behaviour in this disease10. While RA synovial T cells were first described as hyporesponsive, we demonstrated in chapter 2 that T cells from the SF of RA patients are hyper-responsive to TCR trig-gering. In chapter 3 we investigate whether inactivation of T cell Rap1 might play an inflammatory role in RA. To accomplish this we used transgenic (Tg) mice that express constitutively active Rap1a, RapV12, within the T cell lineage. We perfor-med a collagen-induced arthritis (CIA) experiment and evaluated the effect of Rap1 activation during experimental arthritis. We observed that arthritis incidence was decreased more than 70% in RapV12 mice and that both paw swelling as well clini-cal signs of disease were significantly decreased. Histological evaluation of the paws revealed that cellular infiltration and cartilage destruction were significantly redu-ced in Tg mice, as was joint damage. Different animal models in which the activity of Rap1 has been genetically modulated have also found defects in T cell responses. In mice expressing the active mutant Rap1E63 defects in T cell responses have been attributed to increased numbers of regulatory T cells (Tregs)21. In RapV12 mice we found no differences in the numbers of Tregs either under homeostatic conditions or after CIA induction, and Il-10 production was equivalent in RapV12 and WT mice. We found however, that TNF-α production was severely impaired in RapV12 mice, particularly in cytotoxic CD8+ T cells. Defective T cell responses have also been noted in mice lacking Spa1 expression (Spa1-/-)22. In the absence of this GAP, Rap1 activity is elevated in T cells of these mice. These mice show an age-dependent accumulation of unresponsive CD44high T cells that drive defective T helper cell responses. Young Spa1-/- mice with normal primary T cell responses display impaired T cell responses to recall antigens, with reduced antibody production. We investigated whether this was also the case for RapV12 mice. CD44 expression levels in RapV12 mice were comparable to WT mice and no differences were observed between naive, effector memory or central memory populations. We found however, that the circulating levels of specific anti-collagen antibody production were significantly decreased in RapV12 mice. We further investigated the mechanism underlying the defective an-tibody production and found that RapV12 T cells were impaired in their capacity to upregulate CD40L and ICOS co-stimulatory molecules, which could explain why

162

Chapter 7

RapV12 T cells are less effective in providing help for B cell activation and antibody production. Analysis of B cell responses in experimental models where T cells are dispensable for B cell activation (TNP Ficoll) and in T cell-dependent assays (TNP KLH) may be useful to confirm the defective T helper phenotype of RapV12 mice. Additionally, as CIA is a model of a chronic immune stimulation, it will be of interest to determine if the defective T cell responses in RapV12 Tg mice (lower TNF-α and antibody production) occur only in a chronic setting, or also in primary and secon-dary acute immune responses. Finally, it is as yet uncertain if the ability of RapV12 to protect against autoimmune disease is specific for CIA, or will be a general effect in other immune-mediated inflammatory models such as EAE.

TCR engagement in the presence of CD28 costimulation is a critical step in the initi-ation of effector T cell responses, while CTLA-4 ligation is essential at the end of an immune response to dampen cellular activation. Costimulation is therefore a cru-cial step in determining the fate of T cell responses, however the specific proces-ses downstream of costimulatory molecule ligation are not yet clear. Accumulating studies have shown the involvement of the small GTPase Rap1 in this process. Rap1 is transiently activated upon TCR triggering and further enhanced by CTLA-4 en-gagement, whereas CD28 ligation blocks Rap1 activation23-25. These data suggests Rap1 as a central regulator of T cell responses to co-stimulatory cues. Rap1 is acti-vated by guanine nucleotide exchange factors (GEFs) and inactivated by GTPase-activating proteins (GAPs)26. CD28-dependent inactivation of Rap1 is believed to be the result of Rap1GAP activation, while CTLA-4-dependent Rap1 activation the result of Rap1GAP inactivation23-25. In vivo studies have confirmed the idea of Rap1 as a critical mediator of T cell responses27. Different phenotypes of transgenic (Tg) Rap1 GAP animals suggest that each Rap1 GAP may contribute differentially to T cell activation24;28. In chapter 4 we show that the five mammalian Rap1 GAPs are differentially expressed in resting and TCR/CD28 activated T cells. We found Spa-1 to be upregulated both at the mRNA and protein levels after TCR/CD28 stimulation in CD4 but not CD8 T cells. We found this upregulation to be PI3-kinase and NF-κB dependent, as inhibitors for each signaling pathway were able to block Spa-1 upre-gulation. We show that rather than dependent on proliferation, Spa1 upregulation is dependent on TCR/CD28 specific signals, as stimulation with IL-15, IL-17 or IL-2/PHA were able to induce proliferation but not Spa1 expression. Additionally, we found Rap1GAP1A and Rap1GAP1B mRNA to be equally upregulated upon TCR/CD28. Surprisingly we found the opposite at the protein level, suppression of Rap-

163


1GAP1 expression accompanied by protein degradation. Rap1GAP2 was detected at the mRNA level in rested cells, and downregulated upon TCR/CD28 stimulation. At the protein level Rap1GAP2 expression followed the same pattern as Rap1GAP1, ex-pression in rested cells followed by suppression and protein degradation upon TCR/CD28 ligation. Distinct studies have described that post-translational regulation of Rap1GAP protein stability is necessary to avoid proteasomal degradation29-31. The mechanism through which Rap1GAP proteins are degraded during T cell stimula-tion, and the consequences of inhibiting this process, remains to be determined.It will be interesting to address which Rap1 GAP regulates Rap1 inactivation in CD8 T cells upon TCR/CD28, as Spa-1 upregulation occurs in CD4 T cells only. RAP-1GAP1 would be a good candidate, as it becomes upregulated at the mRNA level. Although we observed RAP1GAP1 protein degradation after 24h of TCR/CD28 sti-mulation, shorter time-course stimulations may elucidate on the dynamics of RAP-1GAP1 activation and degradation. The findings presented in this chapter suggest that each Rap1 GAP may differen-tially contribute to T cell activation. Each Rap1 GAP may couple distinct extracel-lular stimuli to Rap1 regulation, associate to different costimulatory molecules and/or inactivate Rap1 in distinct cellular subcompartments. In this chapter we study how TCR/CD28 stimulation modulates Rap1 GAP expression. It will be interesting to evaluate how other co-stimulatory molecules, either positive (ICOS) or negative (CTLA-4 and PD-1) regulators of T cell activation influence Rap1 activation. Inves-tigating the activation patterns of each Rap1GAP may allow to the development of compounds that will be able to specifically regulate Rap1 activation in distinct T cell compartments, with potential application to diseases like RA where a constitutive block in Rap1 expression is observed.

Fibroblast-like synoviocytes (FLS) from RA patients display proliferative and inva-sive properties reminiscent of malignant tumor cells32. Ras small GTPases play im-portant roles in tumor cell proliferation and invasion. Although no transforming mutations have been detected in RA FLS33;34, pathways mediated by Ras GTPases are known to be de-regulated7;35;36. Recent evidence suggests that changes in GEF expression or activity might be relevant to disease. This is the case in human me-lanoma cells, where expression levels of Ras guanine nucleotide-releasing factor (RasGRF) 1 regulates constitutive MMP-9 production37. In chapter 5 we analyze the expression patterns of RasGRF1, a specific H-Ras activator38, in RA and non-RA ST and FLS. RasGRF1 expression was significantly increased in RA ST and found to

164

Chapter 7

positively correlate and co-localize with MMP-1 and MMP-3 producing cells. Ad-ditionally, we found that in vitro modulation of RasGRF1 expression in RA FLS was able to regulate spontaneous MMP-3 production. We could not find an association between RasGRF1 expression and MMP-1 production in RA FLS, which could imply that maybe other cell types, such as macrophages might be more important sources of MMP-1 production. It will therefore be interesting to analyze the effects of knock down and overexpression of RasGRF1 in macrophages in terms of MMP-1 produc-tion. In lysates from FLS and ST we observed the expression of post-translationally modified RasGRF1. These truncated fragments are known to be derived from pro-tease-dependent cleavage and these fragments have been shown to have enhanced Ras-activating capacity39. The results presented in this chapter suggest that enhan-ced RasGRF1 expression and activating post-translational modifications in RA FLS sensitize signaling pathways that promote MMP-3 production and consecutively joint damage. It is not yet known which stimuli present in the joint promotes the post-translational modification of RasGRF1, but approaches targeting these acti-vating stimuli may prove of clinical interest in reducing RasGRF1 activation and MMP-3 production.

Cellular migration, proliferation, activation and immunological synapse formation are processes tightly regulated by Rho family Rac small GTPases40. In RA these cel-lular processes are thought to be de-regulated as high numbers of activated im-mune cells are observed in the synovium, reflecting increased migration into the joints, local proliferation or both41. In chapter 6 we investigated whether treatment of arthritic mice with inhibitory Rac1 C-terminal peptide, which specifically blocks Rac1 signaling, would ameliorate disease severity. Different studies have reported the involvement of members of the Rho GTPase family in the pathogenesis of RA. Inhibition of Rac1 signaling in vitro is able to reduce FLS proliferation and inva-siveness8 while modulation of Rho activation regulates FLS proliferation and IL-6 secretion9. In vivo, in a model of reactive arthritis, Rac signaling in neutrophils has been shown to regulate joint inflammation42. In this chapter we treated arthritic mice during early or chronic disease, with Rac1 C-terminal peptide and found that inhi-bition of Rac1 signaling in vivo significantly reduces anti-collagen antibody produc-tion. The reduction in levels of immunoglobulin (Ig) G1 and IgG2a (early treatment) or IgG2a and IgG2b (treatment during chronic arthritis) correlated with an amelio-ration of paw swelling in treated animals. Decreased antibody production could be the result of defective T cell priming, as B cell responses are critically dependent on

165


proper T cell activation. In order to provide help for B cell activation, T cells need to have a productive engagement with APC. Studies in vivo have shown that Rac deficient dendritic cells (DC) have defective migration to secondary lymphoid or-gans and fail to establish stable contacts with naive T cells43. In this case, suboptimal T cell priming leads to defective humoral responses to T cell-dependent antigens44. Rac1 peptide may similarly inhibit anti-collagen antibody production by disturbing T cell-APC interactions. Further studies addressing antibody production in T cell -dependent and -independent assays may elucidate whether treatment with Rac1 C-terminal peptide targets T cell priming or the antibody producing B cells directly. In the first case, it will be interesting to address whether also T cell effector responses, besides T helper, are impaired. The observed effect on the reduction of paw swelling could be due to effects in edema formation, as inhibition of Rac1-mediated signaling leads to a decrease in leukocyte extravasation45. However, although in vitro Rac1 C-terminal peptide was able to inhibit actin polymerization, we found no evidence for a decrease in cellular infiltration or joint destruction in vivo. The limited clinical effect of Rac1 peptide treatment may reflect the redundant role of Rac2 in promoting osteoclastogenene-sis46;47 and cellular migration48, as the inhibitory peptide we used is highly specific for Rac1 and does not interfere with Rac2 signaling49. Although the data presented in this chapter shows that treatment of arthritic mice with Rac1 peptide can safely be used to reduce paw swelling and the circulating levels of anti-collagen antibodies, future studies will need to directly evaluate the relative contributions of both Rac1 and Rac2 in CIA.

In conclusion, the work presented in this thesis shows that modulation of small GTPase signaling might have good therapeutic potential in the treatment of RA. Strategies aimed at the inhibition or maintenance of small GTPase signaling could be applied at different levels of intracellular signaling cascades and tested for their efficacy in the amelioration of arthritis. Several compounds targeting effectors of small GTPases are already available, and some have been tested in clinical trials with successful results. In cancer clinical tri-als, blockage of the Raf/MEK/ERK pathway with compounds targeting Raf (BAY 43-9006, ISIS 5132) and MEK (PD184352, PD0325901 and ARRY-142886) have showed promising results50. In RA clinical trials, the use of VX-702 and RO-440-2257 as p38 inhibitors have showed only limited success51;52, and further analysis of the distinct contributions of Ras effector pathways to RA is urgently needed.

166

Chapter 7

Direct targeting of small GTPases has also been investigated in different disease set-tings. We have shown in chapter 6 the possibility of using C-terminal peptides to interfere with specific small GTPase signaling in vivo for the treatment of CIA. In chapter 5 we used locked nucleic acid (LNA) as a tool to efficiently inhibit RasGRF1 expression in vitro. Compounds that interfere with small GTPase post-translational modification, such as farnesyl transferase inhibitors (FTIs), have received special at-tention in cancer settings. The FTIs Tipifarnib and lonafarnib are currently under clinical investigation with encouraging results in the treatment of hematological malignancies and breast cancer53. Although not yet tested in clinical trials, bacte-rial toxins have also been shown to target individual small GTPases. Clostridium difficile toxin A and B specifically glucosylate Rho proteins, while bacterial C3-like ADP-ribosyltransferase induces ADP-ribosylation of RhoA, both inhibiting in a spe-cific manner small GTPase downstream signaling54;55. Maintenance or activation of small GTPases may also be interesting to modulate inflammation in RA. We have shown in chapter 3 that maintenance of Rap1 signaling ameliorates disease in CIA. 8-(4-chloro-phenylthio)-2'-O-methyladenosine-3',5'-cyclic monophosphate (8CPT-2Me-cAMP) is a cAMP analog that can selectively activate Epac, a Rap1 GEF56. Alt-hough not yet tested in a disease setting, it would be interesting to analyze the effect of this analogue in mice with CIA.Finally, targeting cell surface receptors upstream of the small GTPases may allow specific modulation of GTPase activity, through the regulation of GEFs and GAPs. One such candidate may already be in the clinic. CTLA-4Ig (abatacept), which dis-rupts the interaction between the co-stimulatory CD28 molecule and CD80/86 on APCs, has shown clinical benefit in RA patients. In vitro, this compound has been shown to prevent T cell Rap1 inactivation by synovial fluid adherent cells57. Future work will need to determine how restored T cell Rap1 function may contribute to the clinical benefit of CTLA-4Ig therapy. In sum, the data presented in this thesis provides strong pre-clinical evidence that targeting Ras superfamily GTPases may be beneficial in the treatment of RA, a potential awaiting the development of com-pounds specifically targeting these enzymes in the clinic.

167


References

Olsen NJ, Stein CM. New drugs for rheumatoid arthritis. N.Engl.J.Med. 1. 2004;350:2167-2179.Schett G, Tohidast-Akrad M, Smolen JS et al. Activation, differential localiza-2. tion, and regulation of the stress-activated protein kinases, extracellular signal-regulated kinase, c-JUN N-terminal kinase, and p38 mitogen-activated protein kinase, in synovial tissue and cells in rheumatoid arthritis. Arthritis Rheum. 2000;43:2501-2512.Tak PP, Firestein GS. NF-kappaB: a key role in inflammatory diseases. J.Clin.3. Invest 2001;107:7-11.Zhang HG, Wang Y, Xie JF et al. Regulation of tumor necrosis factor alpha-medi-4. ated apoptosis of rheumatoid arthritis synovial fibroblasts by the protein kinase Akt. Arthritis Rheum. 2001;44:1555-1567.Na HJ, Lee SJ, Kang YC et al. Inhibition of farnesyltransferase prevents colla-5. gen-induced arthritis by down-regulation of inflammatory gene expression through suppression of p21(ras)-dependent NF-kappaB activation. J.Immunol 2004;173:1276-1283.Pap T, Nawrath M, Heinrich J et al. Cooperation of Ras- and c-Myc-dependent 6. pathways in regulating the growth and invasiveness of synovial fibroblasts in rheumatoid arthritis. Arthritis Rheum. 2004;50:2794-2802.Yamamoto A, Fukuda A, Seto H et al. Suppression of arthritic bone destruction 7. by adenovirus-mediated dominant-negative Ras gene transfer to synoviocytes and osteoclasts. Arthritis Rheum. 2003;48:2682-2692.Chan A, Akhtar M, Brenner M et al. The GTPase Rac regulates the proliferation 8. and invasion of fibroblast-like synoviocytes from rheumatoid arthritis patients. Mol.Med. 2007;13:297-304.Nakayamada S, Kurose H, Saito K, Mogami A, Tanaka Y. Small GTP-binding 9. protein Rho-mediated signaling promotes proliferation of rheumatoid synovial fibroblasts. Arthritis Res.Ther. 2005;7:R476-R484.Remans PH, Gringhuis SI, van Laar JM et al. Rap1 signaling is required for sup-10. pression of Ras-generated reactive oxygen species and protection against oxida-tive stress in T lymphocytes. J.Immunol. 2004;173:920-931.Winchester R, Dwyer E, Rose S. The genetic basis of rheumatoid arthritis. The 11. shared epitope hypothesis. Rheum.Dis.Clin.North Am. 1992;18:761-783.Lundy SK, Sarkar S, Tesmer LA, Fox DA. Cells of the synovium in rheumatoid 12.

168

Chapter 7

arthritis. T lymphocytes. Arthritis Res.Ther. 2007;9:202.Thomas R, McIlraith M, Davis LS, Lipsky PE. Rheumatoid synovium is enriched 13. in CD45RBdim mature memory T cells that are potent helpers for B cell differen-tiation. Arthritis Rheum. 1992;35:1455-1465.Genovese MC, Schiff M, Luggen M et al. Efficacy and safety of the selective co-14. stimulation modulator abatacept following 2 years of treatment in patients with rheumatoid arthritis and an inadequate response to anti-tumour necrosis factor therapy. Ann.Rheum.Dis. 2008;67:547-554.Gringhuis SI, Leow A, Papendrecht-van der Voort EA et al. Displacement of lin-15. ker for activation of T cells from the plasma membrane due to redox balance alterations results in hyporesponsiveness of synovial fluid T lymphocytes in rheumatoid arthritis. J.Immunol. 2000;164:2170-2179.Zhang Z, Gorman CL, Vermi AC et al. TCRzetadim lymphocytes define po-16. pulations of circulating effector cells that migrate to inflamed tissues. Blood 2007;109:4328-4335.Firestein GS, Zvaifler NJ. How important are T cells in chronic rheumatoid sy-17. novitis?: II. T cell-independent mechanisms from beginning to end. Arthritis Rheum. 2002;46:298-308.Azuma M, Yssel H, Phillips JH, Spits H, Lanier LL. Functional expression of B7/18. BB1 on activated T lymphocytes. J.Exp.Med. 1993;177:845-850.Verwilghen J, Lovis R, De BM et al. Expression of functional B7 and CTLA4 on 19. rheumatoid synovial T cells. J.Immunol. 1994;153:1378-1385.Boasso A, Herbeuval JP, Hardy AW, Winkler C, Shearer GM. Regulation of indo-20. leamine 2,3-dioxygenase and tryptophanyl-tRNA-synthetase by CTLA-4-Fc in human CD4+ T cells. Blood 2005;105:1574-1581.Li L, Greenwald RJ, Lafuente EM et al. Rap1-GTP is a negative regulator of Th 21. cell function and promotes the generation of CD4+CD103+ regulatory T cells in vivo. J.Immunol. 2005;175:3133-3139.Ishida D, Yang H, Masuda K et al. Antigen-driven T cell anergy and defective 22. memory T cell response via deregulated Rap1 activation in SPA-1-deficient mice. Proc.Natl.Acad.Sci.U.S.A 2003;100:10919-10924.Carey KD, Dillon TJ, Schmitt JM et al. CD28 and the tyrosine kinase lck stimulate 23. mitogen-activated protein kinase activity in T cells via inhibition of the small G protein Rap1. Mol.Cell Biol. 2000;20:8409-8419.Dillon TJ, Carey KD, Wetzel SA, Parker DC, Stork PJ. Regulation of the small 24. GTPase Rap1 and extracellular signal-regulated kinases by the costimulatory molecule CTLA-4. Mol.Cell Biol. 2005;25:4117-4128.

169


Reedquist KA, Bos JL. Costimulation through CD28 suppresses T cell receptor-25. dependent activation of the Ras-like small GTPase Rap1 in human T lymphocy-tes. J.Biol Chem. 1998;273:4944-4949.Bos JL, Rehmann H, Wittinghofer A. GEFs and GAPs: critical elements in the 26. control of small G proteins. Cell 2007;129:865-877.Minato N, Kometani K, Hattori M. Regulation of immune responses and hema-27. topoiesis by the Rap1 signal. Adv.Immunol. 2007;93:229-264.Kometani K, Moriyama M, Nakashima Y et al. Essential role of Rap signal in 28. pre-TCR-mediated beta-selection checkpoint in alphabeta T-cell development. Blood 2008;112:4565-4573.Gao Q, Srinivasan S, Boyer SN, Wazer DE, Band V. The E6 oncoproteins of high-29. risk papillomaviruses bind to a novel putative GAP protein, E6TP1, and target it for degradation. Mol.Cell Biol. 1999;19:733-744.Jordan JD, He JC, Eungdamrong NJ et al. Cannabinoid receptor-induced neurite 30. outgrowth is mediated by Rap1 activation through G(alpha)o/i-triggered prote-asomal degradation of Rap1GAPII. J.Biol.Chem. 2005;280:11413-11421.Tsygankova OM, Feshchenko E, Klein PS, Meinkoth JL. Thyroid-stimulating 31. hormone/cAMP and glycogen synthase kinase 3beta elicit opposing effects on Rap1GAP stability. J.Biol.Chem. 2004;279:5501-5507.Tak PP, Zvaifler NJ, Green DR, Firestein GS. Rheumatoid arthritis and p53: how 32. oxidative stress might alter the course of inflammatory diseases. Immunol To-day 2000;21:78-82.Pap T, Franz JK, Hummel KM et al. Activation of synovial fibroblasts in rheu-33. matoid arthritis: lack of Expression of the tumour suppressor PTEN at sites of invasive growth and destruction. Arthritis Res. 2000;2:59-64.Roivainen A, Zhu F, Sipola E, Yli-Jama T, Toivanen P. Failure to verify H-ras 34. mutations in arthritic synovium: comment on the article by Roivainen et al. Arthritis Rheum. 2001;44:2705.Na HJ, Lee SJ, Kang YC et al. Inhibition of farnesyltransferase prevents colla-35. gen-induced arthritis by down-regulation of inflammatory gene expression through suppression of p21(ras)-dependent NF-kappaB activation. J.Immunol. 2004;173:1276-1283.Pap T, Nawrath M, Heinrich J et al. Cooperation of Ras- and c-Myc-dependent 36. pathways in regulating the growth and invasiveness of synovial fibroblasts in rheumatoid arthritis. Arthritis Rheum. 2004;50:2794-2802.Zhu TN, He HJ, Kole S et al. Filamin A-mediated down-regulation of the ex-37. change factor Ras-GRF1 correlates with decreased matrix metalloproteinase-9

170

Chapter 7

expression in human melanoma cells. J.Biol.Chem. 2007;282:14816-14826.Jones MK, Jackson JH. Ras-GRF activates Ha-Ras, but not N-Ras or K-Ras 4B, 38. protein in vivo. J.Biol.Chem. 1998;273:1782-1787.Baouz S, Jacquet E, Bernardi A, Parmeggiani A. The N-terminal moiety of 39. CDC25(Mm), a GDP/GTP exchange factor of Ras proteins, controls the activity of the catalytic domain. Modulation by calmodulin and calpain. J.Biol.Chem. 1997;272:6671-6676.Jaffe AB, Hall A. Rho GTPases: biochemistry and biology. Annu.Rev.Cell Dev.40. Biol. 2005;21:247-269.Buckley CD. Michael Mason prize essay 2003. Why do leucocytes accumulate 41. within chronically inflamed joints? Rheumatology.(Oxford) 2003;42:1433-1444.Zhang X, Glogauer M, Zhu F et al. Innate immunity and arthritis: neutrophil 42. Rac and toll-like receptor 4 expression define outcomes in infection-triggered arthritis. Arthritis Rheum. 2005;52:1297-1304.Benvenuti F, Hugues S, Walmsley M et al. Requirement of Rac1 and Rac2 expres-43. sion by mature dendritic cells for T cell priming. Science 2004;305:1150-1153.Li H, Oliver T, Jia W, He YW. Efficient dendritic cell priming of T lymphocy-44. tes depends on the extracellular matrix protein mindin. EMBO J. 2006;25:4097-4107.van Wetering S, van den BN, van Buul JD et al. VCAM-1-mediated Rac signaling 45. controls endothelial cell-cell contacts and leukocyte transmigration. Am.J.Physiol Cell Physiol 2003;285:C343-C352.Fukuda A, Hikita A, Wakeyama H et al. Regulation of osteoclast apoptosis and 46. motility by small GTPase binding protein Rac1. J.Bone Miner.Res. 2005;20:2245-2253.Wang Y, Lebowitz D, Sun C et al. Identifying the relative contributions of Rac1 47. and Rac2 to osteoclastogenesis. J.Bone Miner.Res. 2008;23:260-270.Sun CX, Downey GP, Zhu F et al. Rac1 is the small GTPase responsible for regu-48. lating the neutrophil chemotaxis compass. Blood 2004;104:3758-3765.Van Hennik PB, ten Klooster JP, Halstead JR et al. The C-terminal domain of 49. Rac1 contains two motifs that control targeting and signaling specificity. J.Biol.Chem. 2003;278:39166-39175.Kohno M, Pouyssegur J. Targeting the ERK signaling pathway in cancer therapy. 50. Ann.Med. 2006;38:200-211.Damjanov N, Kauffman RS, Spencer-Green GT. Efficacy, pharmacodynamics, 51. and safety of VX-702, a novel p38 MAPK inhibitor, in rheumatoid arthritis: results of two randomized, double-blind, placebo-controlled clinical studies.

171


Arthritis Rheum. 2009;60:1232-1241.Cohen SB, Cheng TT, Chindalore V et al. Evaluation of the efficacy and safety 52. of pamapimod, a p38 MAP kinase inhibitor, in a double-blind, methotrexate-controlled study of patients with active rheumatoid arthritis. Arthritis Rheum. 2009;60:335-344.Karp JE, Lancet JE. Development of farnesyltransferase inhibitors for clinical 53. cancer therapy: focus on hematologic malignancies. Cancer Invest 2007;25:484-494.Just I, Selzer J, Wilm M et al. Glucosylation of Rho proteins by Clostridium dif-54. ficile toxin B. Nature 1995;375:500-503.Fritz G, Kaina B. Rho GTPases: promising cellular targets for novel anticancer 55. drugs. Curr.Cancer Drug Targets. 2006;6:1-14.Enserink JM, Christensen AE, de RJ et al. A novel Epac-specific cAMP analo-56. gue demonstrates independent regulation of Rap1 and ERK. Nat.Cell Biol. 2002;4:901-906.Remans PH, Wijbrandts CA, Sanders ME et al. CTLA-4IG suppresses reactive 57. oxygen species by preventing synovial adherent cell-induced inactivation of Rap1, a Ras family GTPASE mediator of oxidative stress in rheumatoid arthritis T cells. Arthritis Rheum. 2006;54:3135-3143.

english summary

8

175

Summary

Summary

Rheumatoid arthritis (RA) is a chronic autoimmune disease for which no cure is yet available. At the moment therapies achieving clinical success are not effective in all patients, and occurrence of undesirable side effects may lead to the discontinuation of treatment. Therefore, there is still a need to develop novel therapies which may result in clinical benefit in patients who do not respond well to current therapies. In this context, investigating cellular mechanisms that are altered in RA may contri-bute to the development of these new therapeutic strategies.

Small GTPases are important mediators of the immune system. When a cell receives an external trigger, small GTPases are transiently activated and transmit signals that result in a cellular response. Abnormal GTPase function may thus alter proper cel-lular function. In this thesis we analyzed the contributions of representative small GTPases to inflammation and joint destruction in cells from RA patients and animal models of arthritis.

Previous studies in RA synovial T cells have shown several defects downstream of the T cell receptor (TCR) such as constitutive inactivation of the small GTPase Rap1, chronic downregulation of TCRzeta expression and impaired phosphorylation of TCR downstream proteins. While many of those reports described T cell responses in RA to be defective, we demonstrate in chapter 2 that T cells isolated from RA sy-novial fluid are fully functional and capable of initiating proliferation and cytokine production upon TCR triggering. We show that previous studies concluding that there were defects in RA synovial T cells were compromised by increased sponta-neous apoptosis of these cells upon removal from the joint. We show that there are no defects in RA synovial T cell TCR signaling. Furthermore, we found that altered ratios of pro-apoptotic Noxa and anti-apoptotic Mcl-1 expression were associated with the increased susceptibility of these cells to apoptosis. These findings sugge-sted two important conclusions. First, the previously noted defects in TCR pathway components, such as Rap1, might contribute to disease in RA. Second, since only low levels of T cell cytokines are found in the joints of established RA, our findings suggest that in the effector phase of disease, T cells might contribute to inflammation by TCR-independent mechanisms.

T cells derived from the synovial fluid of RA patients have a block in Rap1 activa-

176

Chapter 8

tion. In chapter 3 we evaluated whether constitutive T cell Rap1 activation would be able to suppress disease in an animal model of RA. Using transgenic (Tg) mice ex-pressing constitutively active Rap1 we performed collagen-induced arthritis (CIA) experiments and observed that Tg mice were significantly protected from arthritis development and severity. We found that Tg T cells with constitutive active Rap1 had defective production of the pro-inflammatory cytokine TNF-α. Additionally, Tg T cells were not capable of providing proper help for B cell activation and produc-tion of pathogenic anti-collagen antibodies. This was associated with diminished up-regulation of ICOS and CD40L T cell co-stimulatory molecules. The results from this chapter show that enhancement of T cell Rap1 activation might be an attractive strategy for therapy in RA.

The small GTPase Rap1 is inactivated by GTPase-activating proteins (GAPs). In chapter 4 we show that the five Rap1GAPs are differentially expressed in resting and TCR/CD28 -activated T cells. We demonstrate that Spa-1 upregulation occurs after TCR/CD28 specific signals and is PI3-kinase and NF-κB dependent. The data in this chapter suggests that each Rap1GAP protein may differentially contribute to T cell activation. We propose that modulation of distinct Rap1GAPs may allow the specific regulation of Rap1 activation in diseases such as RA, where a block in Rap1 is observed.

Small GTPases are activated by guanine nucleotide exchange factors (GEFs). Chan-ges in the expression patterns of these activators might therefore result in altered cellular responses. In chapter 5 we explored the expression levels of Ras guanine nucleotide-releasing factor (RasGRF) 1, a specific activator of the GTPase H-Ras, in RA synovial tissue (ST) and fibroblast-like synoviocytes (FLS). We found RasGRF1 expression to be significantly increased in RA ST and to co-localize with matrix me-talloproteinase (MMP)-1 and MMP-3 expression. In vitro modulation of RasGRF1 expression in RA FLS was able to regulate MMP-3 production. These findings indi-cate that modulation of GEF activation or expression levels might be of interest in protecting against joint destruction in RA.

The small GTPase Rac1 regulates many of the cellular processes required for the per-petuation of inflammation and joint destruction in RA. In chapter 6 we treated ani-mals with Rac1 inhibitory peptide and analyzed effects of this peptide on arthritis development. Treated animals displayed less swelling of the paws and reduced cir-culating levels of anti-collagen antibodies. We found, however, no amelioration in

177

Summary

bone damage. These results suggest that while Rac1 inhibitory peptide may be used to reduce auto-antibody production, additional studies targeting Rac2 as well, for instance, might achieve greater therapeutic efficacy.

In conclusion, the studies provided in this thesis show that modulation of small GT-Pase function regulates cellular activation and inflammation in RA and animal mo-dels of the disease. Small GTPases are therefore emerging targets in RA and further studies should explore the development of novel small GTPase modulators and eva-luate their potential in the treatment of patients with RA.

Nederlandse samenvatting

9

181

Samenvatting

Samenvatting

Reumatoïde artritis (RA) is een chronische auto-immuunziekte; dit is een ziekte waarbij het afweersysteem zich tegen het eigen lichaam keert. Hierbij worden door chronische ontsteking de gewrichten van RA patiënten zo aangedaan dat er, zonder medisch ingrijpen, onherstelbare schade aan de gewrichten ontstaat. Dit gaat ge-paard met hevige pijnen in de aangedane gewrichten. Ook kunnen de gewrichten door het chronische ontstekingsproces vervormd raken.Momenteel zijn er goede medicijnen beschikbaar voor patiënten met RA, maar helaas hebben ze niet bij alle patiënten een voldoende effect. Bovendien krijgen patiënten soms last van ernstige bijwerkingen waardoor de behandeling moet worden beëin-digd. Er is dus nog steeds behoefte aan nieuwe medicijnen waarmee ook patiënten die niet reageren op de huidige therapie behandeld kunnen worden. Momenteel vloeien veel nieuwe medicijnen voort uit onderzoek naar processen die zich afspe-len in de cellen. Ook in cellen van RA patiënten blijken er allerlei processen anders te verlopen dan in cellen van gezonde personen. Een bepaalde groep moleculen, de zogenaamde kleine GTPases, lijkt een belangrijke rol in deze veranderde processen te spelen. Dit zal nader worden toegelicht in de volgende alinea’s.

Kleine GTPases zijn belangrijke signaalmoleculen binnen het afweersysteem. Zij kunnen zich in twee toestanden bevinden: geactiveerd of inactief. Hun werkings-mechanisme is als volgt: wanneer een cel een signaal van buitenaf ontvangt, leidt dit tot voorbijgaande activatie van kleine GTPases; na deze activatie worden zij snel weer geïnactiveerd. Hierdoor wordt een signaal van buitenaf vertaald in een reac-tie binnen in de cel. Het is dan ook goed voor te stellen dat processen binnenin de cel anders verlopen wanneer kleine GTPases abnormaal functioneren. In dit proef-schrift wordt de rol van specifieke kleine GTPases in het proces van ontsteking en gewrichtsbeschadiging geanalyseerd. Hiervoor werden zowel patiëntmateriaal als proefdiermodellen gebruikt.

Bevindingen van andere onderzoeksgroepen suggereren dat verschillende signa-leringsprocessen in T cellen (een bepaald soort witte bloedcellen) die zich in de gewrichtsvloeistof van RA patiënten bevinden, verstoord zijn. Hierdoor zouden T cellen van RA patiënten minder goed in staat zijn te reageren wanneer ze worden geactiveerd via een belangrijk molecuul op hun oppervlak, de T cel receptor (TCR). Deze hypothese wordt ondersteund door het feit dat moleculen die het signaal van de TCR moeten doorgeven binnenin de cel (waaronder kleine GTPase Rap1) min-

182

Chapter 9

der activiteit vertonen in T cellen uit gewrichtsvloeistof van RA patiënten. Wij laten echter in hoofdstuk 2 zien dat uit gewrichtsvloeistof van RA patiënten geïsoleerde T cellen volledig normaal functioneren. Dit blijkt uit het feit dat ze gaan delen en verschillende soorten signaalstoffen (zogenaamde cytokines) gaan produceren wan-neer ze worden geactiveerd via de TCR. Deze bevindingen lijken dus volledig in tegenspraak met de bevindingen van andere onderzoeksgroepen. De verklaring hiervoor ligt in het feit dat dit op een geheel nieuwe manier hebben onderzocht ver-geleken met eerder onderzoek. In ons onderzoek toonden wij aan dat T cellen van RA patiënten meteen nadat zij uit het gewricht zijn afgenomen spontaan sterven (in apoptose gaan). De overlevende cellen produceren samen dus minder cytokines dan verwacht, maar omdat wij de opbrengst per cel konden meten lieten wij zien dat ze per cel een normale hoeveelheid cytokines produceren na activatie via de TCR. De sterke neiging tot apoptose blijkt samen te hangen met een verlaagde hoeveelheid Mcl-1 (een molecuul dat celdood remt) en een verhoogde hoeveelheid Noxa (een molecuul dat celdood stimuleert) in T cellen uit het gewricht van RA patiënten.

Zoals hierboven al kort genoemd werd, is Rap1 activatie in T cellen uit de gewrichts-vloeistof van RA patiënten verminderd of zelfs geblokkeerd. In hoofdstuk 3 hebben we daarom getest of een gemuteerde vorm van Rap1 die continu actief is artritis kan voorkómen. Hiervoor gebruikten wij een proefdiermodel met genetisch gemanipu-leerde muizen die een continu actieve vorm van Rap1 in hun cellen hebben. In deze muizen probeerden wij artritis te veroorzaken door ze in te spuiten met collageen (dit proefdiermodel wordt vaak gebruikt om RA na te bootsen). Het bleek inderdaad dat deze muizen beschermd waren tegen het krijgen van artritis en dat wanneer zij de ziekte wel kregen deze veel minder ernstig verliep. Dit ging gepaard met ver-minderde productie van een ontstekings-bevorderend cytokine (TNF-α). Bovendien bleken T cellen van deze muizen minder goed in staat hulp te bieden aan B cellen, waardoor deze minder artritis-bevorderende antilichamen produceerden.Hiermee laten wij zien dat het beloop van RA mogelijk kan worden beïnvloed door medicijnen te ontwikkelen die Rap1 activatie in T cellen van RA patiënten bewerk-stelligen.

Het lijkt dus alsof inactivatie van Rap1 cruciaal is voor het ontstaan van artritis. Inactivatie van Rap1 gebeurt door GTPase-inactiverende eiwitten, in het Engels ook wel GAPs genoemd. Voor ieder klein GTPase zijn er specifieke GAPs; voor Rap1 zijn dit er vijf. Wij keken daarom in hoofdstuk 4 wat er gebeurt met deze GAPs na stimulatie van T cellen via de TCR. Allereerst blijkt dat de vijf Rap1-GAPs verschil-

183

Samenvatting

lend reageren op TCR stimulatie: van sommige (zoals Spa-1) neemt de concentratie in de cel toe, terwijl deze van andere juist afneemt. Hoewel nog niet geheel duidelijk is wat de exacte rol van ieder van de vijf Rap1-GAPs in de inactivatie van Rap1 is, is het wel voor te stellen dat het manipuleren van ieder van deze GAPs uiteenlopende gevolgen kan hebben. Door in de toekomst nader te bestuderen wat het exacte effect van ieder van deze GAPs op Rap1 activatie is, zou bekeken kunnen worden welke van deze GAPs een mogelijke rol speelt in RA of andere ziekten waarin Rap1 geïn-activeerd is. Hopelijk wordt het met die verworven kennis in de toekomst mogelijk specifieke medicijnen te ontwikkelen die aangrijpen op een van de Rap1-GAPs.

De activatie van kleine GTPases wordt verricht door zogenaamde guanine nucle-otide uitwisselingsfactoren, in het Engels afgekort tot GEFs. Naast GAPs zouden veranderingen in activiteit van GEFs dus ook een rol kunnen spelen in het ziektepro-ces van RA. In hoofdstuk 5 onderzochten wij een van de GEFs die betrokken is bij activatie van klein GTPase H-Ras, RasGRF1. Wij keken hierbij naar de expressie van RasGRF1 in weefsel afkomstig uit gewrichten van RA patiënten en in specifieke cel-len uit het gewricht, de zogenaamde fibroblast-achtige synoviocyten (FLS). RasGRF1 bleek in duidelijk verhoogde concentratie aanwezig te zijn in het gewrichtsweefsel van RA patiënten. Bovendien bevindt het zich op dezelfde plek in het gewricht als twee gewrichtsbeschadigende enzymen, matrix metalloproteinase-1 (MMP-1) en MMP-3. Vervolgens bevestigden wij in celkweken van FLS dat via RasGRF1 de pro-ductie van MMP-3 is te reguleren. Hiermee toonden wij aan dat, naast manipulatie van GAPs, ook manipulatie van GEFs een interessante optie is voor het ontwikkelen van nieuwe behandelingen voor RA.

Tot slot werd een ander belangrijk klein GTPase in RA, Rac1 onderzocht in hoofd-stuk 6. Het is bekend dat Rac1 veel van de cellulaire processen reguleert die essenti-eel zijn voor het in stand houden van de ontsteking en de uiteindelijke beschadiging van het bot in gewrichten van RA patiënten. Daarom onderzochten wij wat het effect zou zijn van het toedienen van een klein eiwit (peptide) dat Rac1 remt op de ontwik-keling van artritis bij proefdieren. Tijdens deze experimenten bleek dat proefdieren die met het Rac1 remmende peptide behandeld waren minder gezwollen gewrich-ten hadden en minder artritisbevorderende antilichamen in het bloed hadden dan proefdieren die niet behandeld waren met het peptide. Er was echter geen effect op beschadiging van het bot in de gewrichten. Deze resultaten suggereren dat er een gunstig effect op artritisbevorderende antilichamen bereikt zou kunnen worden door patiënten met RA te behandelen met een Rac1 remmend peptide. Daarnaast

184

Chapter 9

zou kunnen worden onderzocht of simultane remming van een ander GTPase, Rac2, een sterker anti-artritis effect heeft dan alleen remming van Rac1.

De in dit proefschrift beschreven experimenten tonen aan dat kleine GTPases de ac-tivering van cellen en de ontsteking in gewrichten van RA patiënten en in proefdier-modellen voor artritis reguleren. Daarmee zijn kleine GTPases interessante doelwit-ten voor de ontwikkeling van nieuwe behandelingen voor RA. Toekomstige studies moeten uitwijzen of er inderdaad een rol is weggelegd voor stoffen als het Rac1 remmende peptide in de behandeling van RA.

Acknowledgments

189

Acknowledgments

During this PhD there were many people who helped me along the way, so now comes the opportunity to thank everyone who was directly or indirectly involved in the completion of this thesis.

First of all my promotor and co-promotor who guided me and gave me the opportu-nity to produce this thesis. Paul-Peter, thank you for your guidance and for always reminding me to think in terms of chapters/manuscripts. Thanks for the friendly words during our work discussions, especially during the down moments of my PhD. I really admire that although always so busy and traveling around the world you would still make time to correct my manuscripts on short notice. Kris, thanks for your guidance and supervision. Although our personalities did not always match, you helped me develop my scientific spirit and shared your knowledge with me. Thanks for always making time to answer my questions and of course, for the help in ‘pimping’ my manuscripts!

Then, my team without whom this thesis would not have been possible. My special thanks to Marjolein Sanders who through all these years has become the indispen-sable piece in our team. Thanks for all the help whenever an extra pair of hands was necessary, I enjoyed working with you! Daphne, Sarah, Alek and Wendy, thanks for all the extra hours spent in the ARIA with the mouse experiments! I really appreciate your help during this thesis, I could not have made it without you! Thanks also to the ‘ex-team’ Dion and Tânia. It was a pleasure working with you all!

My two good friends and paranimfen, Gabriela and Tineke, thanks for accepting this job! I enjoyed the moments spent together at work, but especially the nice mo-ments outside work! Thanks for the nice dinners and ‘uitjes’ - we should enjoy them more often!

Then all the roomies I had during these years: Klaas (thanks for all the advices with the mouse experiments!), Alex (thanks for cheering up the office with your singing :) ), Amber, Marianne, Bernard, Lena (спасибо за Вашу дружбу!), Mourad, Sander, Pol and Natasha. Thanks for making the work environment fun and ‘gezellig’!

The Portuguese community: Cristina, Nuno e Inês. Cristina, obrigada por seres a minha conselheira durante o doutoramento. Obrigada pelas dicas, conselhos e pela amizade! Nuno, obrigada pelos bons momentos e pela amizade. Boa sorte em Por-tugal!

190

Acknowledgments

Felix and Paul K, I really enjoyed our dinners and going out, hope we’ll have many more to come! Ronald, working with you made the weekend and 23.00h mouse ex-periments a bit more fun (although it may seem hard to believe!). Good luck with your promotion!

Martijn, your passion and enthusiasm for science are contagious. Thanks for all your nice suggestions and input in this thesis.

René van Lier and Eric, thanks for the critical reading of my manuscripts and for all the helpful suggestions during our work discussions. René, thanks for accepting to take part in the committee for my thesis defense.

Tom Smeets and Margriet, thanks for all the help with analyzing and interpreting my infinite immunohistochemistry slides.

I could not forget to thank the whole KIR team, including all doctors (who were al-ways on alert for the precious synovial fluid samples), all the technicians and PhDs, and everyone at EXIM (including G1). Thanks for all the tips, help and specially for the nice environment to work in!

My thanks to our collaborators from Sanquin (Peter, Paula, Jean-Paul and Anne) for the nice discussions and help in making chapter 6 possible. My special thanks to Peter Hordijk for accepting to take part in the committee for my thesis defense.João Eurico, muito obrigada pela hospitalidade em Lisboa!

I had the privilege to join a very nice team at the LUMC. Bart, Gaby, Sandra, Arno, Fleur, Menno, Tanja, Yvonne, Kay, Martijn, Arnaud, Stefan and Bobby, thanks for making me feel at home. Bart, your enthusiasm for science is really motivating! Thanks for the opportunity you’re giving me to develop as a scientist.

How boring would life be without friends?… During my stay in The Netherlands I was fortunate to become friends with a special group of people, Fra, Edo, Dani, Ali, Agnieszka, Bert, Massi, Bas, Femke, Michiel, Volker, Imke, Vivianda, Ron, Jantienne, Clemens, thanks for all the great dinners, party’s and for your friendship! Santa, Tai, Cristiana, Hugo e Hugo André, quem diria que depois do estágio ficariamos por cá para o doutoramento? Obrigada pelo vosso apoio, amizade e claro, pelos jantares portugueses!

191

Acknowledgments

Os meus amigos em Portugal… que saudades dos meus amigos portugueses… Ca-tarina, Ana Catarina, Luísa, Tiago, Lu, Ana e Póvoas, vocês sabem que são os meus amigos de sempre, muito especiais! Apesar de passar tanto tempo sem nos vermos, quando nos re-encontramos a intimidade está sempre presente e parece que foi on-tem, a última vez que estivemos juntos. Catarina e Luísa, espero que me perdoem pelas festas perdidas!

Obrigada a toda a minha família (tios, tias, primos e primas) e quase família (Zé e Alice), pelo carinho e amizade. António e Nuno, obrigada pela vossa amizade e por fazerem a minha mãe e irmã felizes!

Ik wil mijn nederlandse familie, Hans, Marianne, Marieke en Joost, bedanken omdat jullie mij het gevoel geven dat ik deel van jullie familie uitmaak. Naast een gewel-dige schoonfamilie zijn jullie ook goede vrienden!

Oma Mous, sinds de eerste keer dat ik jou ontmoet heb, voelt het alsof ik jouw klein-kind ben. Dank je wel voor jouw liefde. Ik wil ook de hele familie Mous & Distel-brink bedanken voor alle warmte die jullie mij geven.

E agora as pessoas mais especiais, que mais me apoiaram durante este projecto…

Mãe, obrigada pelo teu carinho, apoio e compreensão. Obrigada por, apesar de ser difícil ver-me partir para longe, sempre me teres apoiado e dado força durante este projecto. Sabes que apesar de estar longe, tu estás sempre no meu coração! Lila, min-ha manita, sabes que sem o teu incentivo talvez nem teria começado este projecto. Obrigada por sempre me fazeres acreditar em mim, especialmente nos momentos mais difíceis. Vocês sabem que eu tenho muito orgulho de vos ter como mãe e irmã e que vos adoro! Muito obrigada por tudo!

Finally I would like to thank this PhD for giving me the opportunity to meet Ro-gier… Lieve Rogier, bedankt voor jouw liefde en vriendschap, dat je altijd in mij bent blijven geloven en mij gesteund hebt tijdens de moeilijke momenten. Ik ben heel gelukkig om iemand zo speciaal te hebben naast mij. Eu amo-te muito muito!

Joana

Curriculum Vitae

195

Curriculum vitae

Curriculum vitae

Joana Abreu was born in Braga, Portugal, on the 18th May 1980. After graduating from highschool in 1998 at the Escola Secundária de Carlos Amarante in Braga she started her university studies. In 2002 she completed the Applied Biology degree at the School of Sciences from Minho University in Braga. During the last year of her studies she performed a traineeship at the University of Amsterdam where she stu-died the role of the PHB complex of proteins under oxidative stress conditions in the development of C. elegans, under supervision of Dr. H. van der Spek. She initiated her PhD project in 2002 at the department of Clinical Immunology and Rheumato-logy, at the Academic Medical Center, University of Amsterdam, under supervision of Dr. K.A. Reedquist and Prof. Dr. P.P. Tak. Since 2008 she has joined the group of Prof. Dr. Bart Roep at the department of Immunohematology and Blood Transfu-sion, Division of Diabetes, at the Leiden University Medical Center as a postdoctoral fellow to study T cell autoreactivity in type 1 diabetes.

197

List of publications


Abreu JRF, Grabiec AM, Krausz S, Spijker R, Burakowski T, Maslinski W, Eldering E, Tak PP and Reedquist KA. The presumed hyporesponsive behaviour of rheu-matoid arthritis T lymphocytes can be attributed to spontaneous ex vivo apoptosis rather than defects in T cell receptor signaling. The Journal of Immunology, July 2009, 183(1):621-630

Abreu JRF, deLaunay D, Sanders ME, Grabiec AM, Sande M, Tak PP and Reedquist KA. The Ras guanine nucleotide exchange factor RasGRF1 promotes MMP-3 pro-duction in rheumatoid arthritis synovial tissue. Arthritis Research and Therapy, August 2009, 11(4):R121

Abreu JRF, Krausz S, Dontje W, Grabiec AM, deLaunay D, Nolte MA, Tak PP and Reedquist KA. Sustained T cell Rap1 signaling protects mice against collagen-indu-ced arthritis. Submitted for publication

Abreu JRF, deLaunay D, van Hennik PB, van Stalborgh AM, ten Klooster JP, Sanders ME, Reedquist KA, Vervoordeldonk MJ, Hordijk PL and Tak PP. A Rac1 inhibitory peptide suppresses antibody production and paw swelling in the murine collagen-induced arthritis model of rheumatoid arthritis. Submitted for publication

deLaunay D, Vreijling J, Abreu JRF, van Maanen MA, Sanders ME, Oerum H, Ver-voordeldonk MJ, Fluiter K, Tak PP and Reedquist KA. Silencing expression of Ras family GTPase homologues decreases inflammation and joint destruction in experi-mental arthritis. Submitted for publication

Unger WW, Velthuis JH, Abreu JRF, Laban S, Quinten E, Kester M, Hadrup SR, Bak-ker AH, Duinkerken G, Eijsink C, Mulder A, Franken K, Hilbrands R, Keymeulen B, Pipeleers DG, Peakman M, Ossendorp F, Drijfhout JW, Schumacher TN and Roep BO. Discovery of low-affinity preproinsulin epitopes and detection of autoreactive CD8+ T cells using combinatorial MHC multimers. Submitted for publication

Velthuis JH, Unger WW, Abreu JRF, Duinkerken G, Franken K, Peakman M, Bakker AH, Reker-Hadrup S, Keymeulen B, Pipeleers D, Drijfhout JW, Schumacher TN and Roep BO. Simultaneous detection of circulating CD8+ T cells specific for HLA-A2

198

restricted islet cell-associated epitopes using multidimensional encoded MHC mul-timers. Submitted for publication


Date post:	07-Nov-2023
Category:	Documents
Upload:	academicmedicalcentreuniversiteitvanamsterdam
View:	2 times
Download:	0 times

Sustained T cell Rap1 signaling is protective in the collagen-induced arthritis model of rheumatoid...

Documents