Polyamines and somatic embryogenesis in two Vitis vinifera cultivars

Daniela Bertoldia,b, Annalisa Tassoni

a, Lucia Martinelli

band Nello Bagni

a,*

aDepartment of Biology e.s. and Interdepartmental Centre for Biotechnology, University of Bologna, Via Irnerio 42, 40126, Bologna, ItalybLaboratorio di Biotecnologie, Istituto Agrario, San Michele all’Adige, 38010, Trento, Italy*Corresponding author, e-mail: bagninel@kaiser.alma.unibo.it

Received 26 May 2003; revised 3 September 2003

Polyamine content and activities of enzymes of polyaminebiosynthesis were assayed during somatic embryogenesis in

Vitis vinifera callus cultures of Chardonnay and Brachetto ‘a

grappolo lungo’ (Brachetto g.l.) cultivars. The analyses were

carried out on embryogenic callus samples, embryos at differ-ent stages and developing plants. Polyamine content, both in

the free and PCA-soluble conjugated form, was higher in

Brachetto g.l. than in Chardonnay, and putrescine was presentat higher concentrations than the other polyamines. In all

samples of both cultivars, ornithine decarboxylase activity

(ODC, EC 4.1.1.17) was higher than arginine decarboxylase

(ADC, EC 4.1.1.19), with a maximum in developing plantroots. S-Adenosylmethionine decarboxylase (SAMDC, EC

4.1.1.50) activity displayed a similar trend. The activities of

all three enzymes were detected both in the supernatant and

pellet fractions, indicating for the first time the presence ofSAMDC activity in the particulate fraction. Particularly in

the Chardonnay cultivar, an increase in the mRNAs expres-

sion patterns of ODC and SAMDC during morphoge-nesis from small embryos to plantlets was detected by

northern blot, suggesting a direct correlation with enzymatic

activities.

Introduction

The complex developmental events leading to somaticembryogenesis in higher plants require an orderedexpression of morphogenetic stimuli arising from endo-genous genotype-specific factors, physiological status,culture medium and the physical environment. The hor-monal balance plays a decisive role in directing somaticembryo development, as the composition of growth regu-lators is one of the most critical aspects of the formula-tion of media (Choi and Sung 1989).

Modulation of polyamine metabolism has been stud-ied in different systems in relation to somatic embryo-genesis, demonstrating that polyamines are crucialendogenous factors during in vitro embryo formation(Feirer et al. 1984, Mengoli et al. 1989, Faure et al.1991, Feirer 1995, Minocha et al. 1999). Aliphatic poly-amines are ubiquitous compounds in both prokaryoticand eukaryotic organisms and are generally consideredto be necessary in cell division and to be involved in

developmental processes in plants such as growth, mor-phogenesis and differentiation (Bagni 1989).

In vitro embryogenesis ofDaucus carota has been shownto be associated with a high polyamine titre (Feireret al. 1984), probably related to a general increase inarginine decarboxylase (ADC) and S-adenosylmethioninedecarboxylase (SAMDC) activities (Fienberg et al. 1984).In addition, studies on Phaseolus embryogenesis (Nagl1990) indicate that embryo development in vivo requirespolyamines, which may be one of the limiting factorsfor in vitro somatic embryogenesis in Hevea brasiliensis(El Hadrami et al. 1989). Moreover, inhibitors of poly-amine biosynthesis, such as a-difluoromethylarginine,block somatic embryogenesis in Daucus carota andeggplant (Fobert and Webb 1988, Mengoli et al.1989, Robie and Minocha 1989), whereas in Cichoriumthis effect was reversed by putrescine (Helleboid et al.1995).

PHYSIOLOGIA PLANTARUM 120: 657–666. 2004 Copyright# Physiologia Plantarum 2004

Printed in Denmark – all rights reserved

Abbreviations – ADC, arginine decarboxylase; Brachetto g.l., Brachetto ‘a grappolo lungo’; dap, 1,3-diaminopropane; NN medium, Nitschand Nitsch medium; ODC, ornithine decarboxylase; PCA, perchloric acid; put, putrescine; SAMDC, S-adenosylmethionine decarboxylase;spd, spermidine; spm, spermine.

Physiol. Plant. 120, 2004 657

Polyamines are key factors during morphogenesis alsoin grapevine, and the putrescine/spermidine ratioprovides a good marker of the subsequent steps of somaticand zygotic embryo development (Faure et al. 1991). In thecultivar Grenache noir, polyamines accumulated duringthe late torpedo stage in which enhanced ADC andornithine decarboxylase (ODC) activities were measured.In this cultivar the low rate of conversion of embryos intoplantlets (,5%) seemed to be related to the highpolyamine content (about 3mM) (Faure et al. 1991).Somatic embryogenesis in grapevine is hindered by lowefficiency and a strong genotype effect. Only a few molecu-lar studies of the developmental steps of morphogenesishave been conducted in this plant (Gianazza et al. 1992,Martinelli et al. 1993). A better understanding of thefactors involved in somatic embryogenesis would berequired for determining central aspects of grapevinebiology and for improving this plant using a biotech-nological approach (Martinelli and Gribaudo 2001).

In the present study of somatic embryogenesis ingrapevine, we have measured the polyamine content, aswell as activities and transcript levels of the mainenzymes involved in polyamine synthesis during well-defined steps, from embryogenic callus to rooted plant-lets, in two grapevine cultivars of economic interest(Chardonnay and Brachetto ‘a grappolo lungo’) that aregenetically quite different. In addition, Chardonnay iscultivated all over the world whereas Brachetto g.l. isonly cultivated in the north of Italy.

Materials and methods

Plant material, induction of somatic embryogenesis and

germination of plants

Embryogenic callus obtained from immature anthers(Chardonnay) and ovaries (Brachetto ‘a grappololungo’) according to Martinelli et al. (2001) was kindlyprovided by Dr I. Gribaudo (Istituto di Virologia Vege-tale, CNR, Grugliasco, Italy). The embryogenic calluswas propagated on a medium containing Nitsch andNitsch (1969) (NN) macroelements and iron, Murashigeand Skoog (1962) microelements, Gamborg et al. (1968)vitamins, 6% (w/v) sucrose, 4.5 mM 2,4-dichlorophenoxy-acetic acid (2,4-D), 8.9 mM benzyladenine (BA) and0.25% (w/v) phytagel, pH 5.8 (medium A). Every 2months, this medium was alternated with an embryodifferentiation medium (medium B) with the same basalcomposition but supplemented with 10 mM naphthoxy-acetic acid (NOA), 1mM BA, 20 mM indoleacetic acid(IAA), 0.25% (w/v) phytagel and 0.25% (w/v) activatedcharcoal, pH 6.2. Media were placed in 90-mm Petridishes and the cultures were incubated at 26�C in thedark. For maturation and multiplication, 0.1 g FW ofembryos differentiated on medium B were placed in125-ml flasks containing 50ml of a NN-based mediumwith 3% (w/v) sucrose and 0.5 mM indol-3-butyric acid(IBA), pH5.75 (medium C). These cultures were incu-bated at 25�C with a 16-h photoperiod (light intensity

70 mmolm�2 s�1, 35W Philips lamps, Philips, Monza,Italy) and shaken continuously at 90 r.p.m. The mediawere replaced every week. For germination, maturesomatic embryos were isolated from the liquid cultureand planted on a NN-based medium (medium D) con-taining 1.5% (w/v) sucrose and 0.9% (w/v) bacto-agarsupplemented with 4.4 mM BA and 0.5 mM IBA, pH5.75.About 11 somatic embryos were placed into Magentaboxes and incubated at 25�C with a 16-h photoperiod(light intensity 70mmol m�2 s�1).

During sample harvesting, embryogenic callus at dif-ferent stages, globular, torpedo, small (hypocotyl length0.1–0.5 cm) and large embryos (hypocotyl length0.5–1 cm) and developing plantlets (separated into leavesplus stems and roots), were collected (Fig. 1) and manu-ally counted (Table 1). Developing plantlets of Brachettog.l. were, at the same collection time, larger than thosefrom Chardonnay, probably due to a higher watercontent. In particular, Brachetto g.l. leaves plus stems(S1L) and divided roots (R) were, respectively, two-and five-fold higher in gFW than the same samples ofChardonnay. All samples were frozen in liquid nitrogenand stored at �80�C until use.

Polyamine analysis by HPLC

Polyamine analyses were performed according to Tassoniet al. (2000). Grapevine samples (about 0.2 gFW) werehomogenized in 10 volumes of 4% (v/v) cold perchloricacid (PCA) and centrifuged at 20 000 g for 30min at 4�C.The pellet was washed three times and re-suspended inthe original volume of PCA. Triplicates of this suspen-sion and of the supernatant were hydrolysed with 6MHCl in flame-sealed vials at 110�C for 20 h in order torelease polyamines from their conjugates. Aliquots(0.2ml) of supernatant, hydrolysed supernatant andhydrolysed pellet were derivatized with dansyl-chloride(3mgml�1 of acetone), extracted with toluene and ana-lysed by HPLC (Jasco, Großumstad, Germany) with areverse phase C18 column (Spherisorb ODS2, 5mM par-ticle diameter, 4.6� 250mm) as described by Tassoniet al. (2000). Standard polyamines were subjected to thesame procedure.

ADC, ODC and SAMDC enzyme activities

Arginine decarboxylase (ADC; EC 4.1.1.19), ornithinedecarboxylase (ODC; EC 4.1.1.17) and S-adenosyl-methionine decarboxylase (SAMDC; EC 4.1.1.50) activ-ities were determined by a radiochemical method asdescribed by Tassoni et al. (2000). Grapevine tissues(0.3 g FW) were homogenized in an ice-cold mortarwith five volumes of the assay buffer (100mM Tris-HCl(pH8.5), 50 mM pyridoxal phosphate) and centrifuged at20 000 g for 30min at 4�C. Aliquots of 0.2ml of bothsupernatant and re-suspended pellet (containing cellwall, nuclei, plastids and mitochondria) were used todetermine ADC and ODC activities (0.5ml final assayvolume). The arginine decarboxylase assay was performed

658 Physiol. Plant. 120, 2004

by measuring the 14CO2 evolution from 7.4 kBq (about1.3 mM) of L-[U-14C]arginine (specific activity11.7GBqmmol�1, Amersham Biosciences, Milan, Italy)whereas the ODC assay was performed by measuring the14CO2 evolution from 7.4 kBq (about 7.3 mM) of D,L-[1–14C]ornithine (specific activity 2.11GBqmmol�1;Amersham Biosciences). To determine SAMDC activity,grapevine samples (0.2 g FW) were homogenized in fivevolumes of 100mM Tris-HCl (pH7.6) plus 50 mMEDTA and centrifuged at 20 000 g for 30min at 4�C.The supernatant and there-suspended pellet (0.2ml

aliquots) were incubated separately with 3.7 kBq (about3.6 mM) [1–14C]S-adenosylmethionine (specific activity2.04GBqmmol�1 Amersham Biosciences) and 14CO2

evolution was meas-ured for 2 h (Altamura et al. 1993).Protein content of supernatant and pellet fractions was

determined according to Lowry et al. (1951).

RNA extraction and northern blotting

Total RNA was extracted from about 100mg of tissue asdescribed by Chang et al. (1993). Northern blot analysis

Fig. 1. Induction of somaticembryogenesis and plant germinationof Vitis vinifera. (A), undifferentiatedcallus; (B), intermediate callus withembryogenic nuclei; (C), callus withembryos; (D), embryo at globularstage; (E), embryo at torpedo stagewith developing cotyledons; (F),embryo of the ‘small’ category; (G),embryo of the ‘large’ category; (H),developing plant with primary leaves.

Table 1. Chardonnay and Brachetto g.l. samples reported as gFW and unit numbers. One unit corresponds to one embryo or one plantletwithout root or one root apparatus. C, undifferentiated callus; IC, intermediate callus; C1E, callus with embryos; G, globular embryos;T, torpedo embryos; SE, small embryos; LE, large embryos; S1L, plant stems plus leaves; R, plant roots

Chardonnay Brachetto g.l.

Sample gFW Unit number gFWunit�1 g FW Unit number gFWunit�1

C 0.25 – – 0.24 – –IC 0.24 – – 0.28 – –C1E 0.27 – – 0.28 – –G 0.16 2200 7.3� 10�5 0.21 3000 7.0� 10�5

T 0.22 900 2.4� 10�4 0.22 1100 1.9� 10�4

SE 0.43 60 0.007 0.41 57 0.007LE 0.55 20 0.028 0.45 32 0.014S1L 0.45 3 0.15 0.61 2 0.31R 0.22 3 0.07 0.71 2 0.36

Physiol. Plant. 120, 2004 659

was performed by standard procedures (Sambrook et al.1989) using Hybond1-N nylon membranes (AmershamBiosciences, Milan, Italy). The loading amount waschecked both by ethidium bromide staining and byhybridization with an 18S rRNA probe (kindly providedby Dr Marianne van Buuren, Agricultural Faculty,University of Bologna, Italy).

Arginine decarboxylase, ODC and SAMDC hybrid-ization probes were amplified by PCR using degenerateprimers (Table 2) designed on the conserved regions ofaligned database sequences (data not shown). Totalextracted RNA (0.5mg) was used for reverse transcrip-tion polymerase chain reaction (RT-PCR; Ready-To-GoRT-PCR Beads kit, Amersham Biosciences, Milan, Italy)according to the manufacturer’s instructions. Amplifi-cation was carried out in a PTC-150 MiniCycler (MJResearch, Waltham, MA, USA) with the followingtemperature parameters: 30min at 42�C; 5min at 95�C;followed by several cycles of 1min at 95�C, 2min at theannealing temperature (Table 2), 2min at 72�C; finalextension 10min at 72�C. Aliquots (0.5–1ml) of thesereaction mixes were used as template for nested PCRamplification (Ready-To-Go PCR Beads kit, AmershamBiosciences) using different degenerate primers with thefollowing temperature parameters: 5min at 95�C, fol-lowed by several cycles (Table 2) of 1min at 95�C,2min at the annealing temperature (Table 2), 2min at72�C; final extension 10min at 72�C.

The amplified ADC (300 bp) and ODC (450 bp) PCRproducts were extracted from the gel using the QIAquickgel extraction kit (QIAGEN, Hilden, Germany), directlysequenced and the fragment identities were confirmedby database searching. In particular, ADC cDNA frag-ment showed about 98% identity with the known databasesequence (from 758 to 1038bp, Accession N� X96791,Primikirios and Roubelakis-Angelakis 1999). The SAMDCcDNA fragment (900bp) was extracted from the gel, clonedwith the PCR-Script1 Amp Cloning Kit (Stratagene,La Jolla, CA, USA) and sequenced. The identities of the

fragments were confirmed by database searching andSAMDC cDNA sequence was reported in the EMBLNucleotide Sequence database, Accession N� AJ427634.

The purified cDNA fragments were successivelylabelled with a[32P]dCTP (Ready-To-Go DNA LabellingBeads kit, Amersham Biosciences) and used as probesfor Northern blot hybridization. Pre-hybridization(2 h) and hybridization (16 h) were carried out at 42�Cfollowing Sambrook et al. (1989). Membranes werewashed at high stringency for 2 h at 65�C and exposedto Kodak Biomax films at �80�C for an appropriatetime. Densitometry of exposed films was performed byimage analysis software (Phoretix International Ltd,Newcastle-upon-Tyne, UK).

Results

Polyamine content

Changes in free and PCA-soluble conjugated polyaminecontent were analysed in different samples of embryo-genic callus, somatic embryos at later developmentalstages and in developing plants (about 100 days frombeginning of culture) of both Chardonnay and Brachettog.l. cultivars and expressed both as nmol g FW�1

(Table 3) and as nmol unit�1 (Figs 2 and 3).Putrescine (put), spermidine (spd) and spermine (spm)

were found in the free and in the PCA-solubleconjugated form in most of the samples. Put, both inthe free and PCA-soluble conjugated form, was the mostabundant polyamine in all the samples when expressedas either nmol gFW�1 (Table 3) or nmol unit�1 (Figs 2and 3). When expressing data as nmol gFW�1, in bothcultivars, a progressive decrease of the free put contentwas measured going from undifferentiated callus (C) tocallus with differentiated embryos (C1E) (Table 3).When testing for the free spd, an increase was detectedin the C1E sample of Chardonnay, whereas a reductionwas observed in Brachetto g.l. (Table 3). In the subsequent

Table 2. Degenerate primers for the amplification of the hybridization probes. The annealing temperature (Ta) and number of PCR cycles areindicated. The primers were designed on conserved regions of aligned database sequences. The nucleotide mixtures are indicated by thefollowing one-letter code: R: a, g; Y: c, t; M: a, c; K: g, t; S: g, c; W: a, t; H: a, t, c; B: g, t, c; V: g, a, c; D: g, a, t; N: a, g, c, t. aSpecific primerdesigned from base 1055 to base 1038 of ADC Vitis vinifera database sequence (Accession N� X96791, Primikirios and Roubelakis-Angelakis1999).

Probes RT-PCR primers Ta; cycles PCR primers Ta; cycles

50 pmol each 25 pmol eachADC (300 bp) ADC-sense 1 52.5�C ADC-sense 1 52�C

50-GGDGGKGGKYTBGGDAT-30 40 cycles 50-GGDGGKGGKYTBGGDAT-30 35 cyclesADC-antisense 1 ADC-antisense 2 a

50-CATYRCYCKRAGVACATC-30 50-TGGTAATCAACACGGGC-30

50 pmol each 50 pmol eachODC (450 bp) ODC-sense 1 47�C ODC-sense 2 48�C

50-TCCVTTYTAYGCHGTHAARTG-30 45 cycles 50-ATYRTYTWYGCHMABCCNT-30 40 cyclesODC-antisense 1 ODC-antisense 250-ADNSCRTCRCANGTDGG-30 50-NGGYTCDGCDATDATDKBHA-30

25 pmol each 25 pmol eachSAMDC (900 bp) SAMDC-sense 1 52.5�C SAMDC-sense 2 50�C

50-CTYGGHTACAGGATTGARGAC-30 40 cycles 50-TCMRRTCWKCTGYHTACTCCAAC-30 35 cyclesSAMDC-antisense 1 SAMDC-antisense 250-TCATDGARTARCCRCAGGG-30 50-GYTCRAAVTCAAARTCRCA-30

660 Physiol. Plant. 120, 2004

samples, the three polyamines generally appeared todecrease during the transition from globular embryos(G) to developing plantlets (Table 3). However, thesedata could be explained by the higher water content ofthe tissues that increases the fresh weight, especially ofthe large embryos (LE) and plantlets. This could beovercome by expressing data, when possible, asnmol unit�1, where a unit is one embryo, one plantletwithout root or one root apparatus (Figs 2 and 3). Dataexpressed as nmol unit�1 showed that the free polyaminecontent increased in both cultivars during the later stagesof embryo development, with maximum accumulation inlarge embryos (LE) (Fig. 2). In comparison with LE,developing plants had lower free polyamine content,however, the spm level was almost 50% higher in theplant stem plus leaves sample (S1L) than in LE (Fig. 2).Free 1,3-diaminopropane (dap), already present in callussamples, progressively increased in concentration fromglobular embryos to developing plants (Fig. 3) althoughits PCA-soluble form was present only in trace amounts(data not shown).

In embryogenic callus samples, the PCA-solubleconjugated put, spd and spm were detected only in theintermediate callus (IC) and callus with embryos (C1E)of Brachetto g.l. (Table 3). In general, during embryodevelopment and plant germination, the accumulationtrend of conjugated polyamines was similar to thatobserved for the free polyamines (Table 3 and Fig. 3).

PCA-insoluble conjugated polyamines were detectedin trace amounts only in the developing plants (datanot shown).

Enzyme activities of ADC, ODC and SAMDC

Figure 4 shows the ADC and ODC total (supernatantplus pellet) activities measured in different samples of

embryogenic callus, isolated somatic embryos and devel-oping plants of both Chardonnay and Brachetto g.l.cultivars. ADC activity was higher in the pellet than inthe supernatant fraction, whereas ODC activity wasalmost equal in both fractions (data not shown). ODCtotal activity was generally higher compared with ADC.In particular, in developing plants ADC and ODC levelsgreatly increased, reaching a maximum in the rootswhere ODC was about four times higher than ADC(Fig. 4). The SAMDC activity displayed a pattern similarto that of ADC and ODC in both cultivars. For the firsttime in plants (cf. Torrigiani et al. 1986), SAMDC activ-ity was found to be two- to three-fold higher in the pellet(Fig. 5) than in the supernatant.

mRNA expression of ADC, ODC and SAMDC

The mRNAs expression levels of ADC, ODC andSAMDC were detected in different stages of somaticembryogenesis by northern blot. After having normalizedthe band relative intensity to the rRNA 18S control probe,the mRNA expression pattern of the three enzymes wassimilar in both cultivars (Fig. 6). Hybridization with theADC probe revealed the presence of a 3.4kb mRNA,quite similar to the 3.1kb transcript detected by Primikiriosand Roubelakis-Angelakis (1999). Densitometric analysisof the mRNA expression level of ADC showed, in parti-cular in Chardonnay, a decreasing pattern ingoing fromsmall embryos (SE) to plantlet roots (R). In contrast, anincrease of ODC and SAMDC expression levels was foundin the same samples. In developing plant samples (S1Land R), the expression level for ODC was higher than forthe other enzymes (Fig. 6). The mRNA levels of theenzymes in Brachetto g.l. S1L and R samples were notclearly detectable due to the difficulty of extracting a suffi-cient amount of total RNA for loading in the northern gel.

Table 3. Polyamine content (nmol gFW�1) in different samples of Chardonnay and Brachetto g.l. cultivars. C, undifferentiated callus; IC,intermediate callus; C1E, callus with embryos; G, globular embryos; T, torpedo embryos; SE, small embryos; LE, large embryos; S1L, plantstems plus leaves; R, plant roots; nd, not detected. Values represent the mean� SE of two experiments with three determinations each.

Free polyamines PCA-soluble conjugated polyamines

Sample put spd spm dap put spd spm

ChardonnayC 5200� 169 699� 63 126� 7 29.0� 0.3 nd nd ndIC 4658� 479 764� 146 55� 9 18.1� 0.2 nd nd ndC1E 1706� 213 1317� 119 114� 7 51.0� 0.5 nd nd ndG 4401� 451 894� 5 159� 5 30.7� 1.0 nd nd ndT 5260� 262 570� 9 80� 9 29.4� 0.1 nd nd ndSE 544� 67 237� 27 7.5� 0.9 7.8� 0.5 504� 25 237� 7 13.3� 1.6LE 1853� 136 257� 25 7.5� 1.2 27.3� 1.1 377� 11 155� 6 5.6� 0.4S1L 8.6� 0.3 11.2� 0.9 3.3� 0.1 0.9� 0.1 49.9� 2.5 43.1� 0.1 14.1� 0.8R 26.2� 0.4 12.4� 0.2 1.7� 0.1 5.0� 0.6 49.7� 7.5 25.4� 1.0 4.0� 0.3

Brachetto g.l.C 5786� 459 419� 25 417� 35 27.2� 0.3 nd nd ndIC 4614� 67 167� 265 32� 5 20.9� 0.3 223� 7 132� 11 14.1� 1.3C1E 3104� 32 273� 56 40� 9 16.9� 0.4 682� 171 236� 31 1.6� 0.4G 6671� 942 521� 99 110� 41 nd nd nd ndT 5423� 260 373� 33 87� 4 nd 4121� 124 519� 31 81.6� 4.9SE 2692� 373 601� 56 20� 1 17.35� 0.2 1145� 158 807� 105 31.3� 5.6LE 3157� 286 368� 65 46� 3 10.8� 0.5 2933� 235 1662� 116 47.8� 3.8S1L 25.5� 1.5 40.1� 3.8 3.7� 0.6 0.70� 0.05 38.3� 0.7 7.2� 0.6 0.6� 0.01R 30.1� 5.6 37.2� 3.7 2.1� 0.1 33.3� 0.8 50.6� 0.1 23.7� 0.2 nd

Physiol. Plant. 120, 2004 661

Discussion

In this work, several aspects of polyamine metabolismhave been studied in two important Italian Vitis viniferacultivars, Chardonnay and Brachetto ‘a grappolo lungo’.

Total polyamine content was higher in Brachetto g.l.than in Chardonnay. In all samples of both cultivars, put(in the free and PCA-soluble conjugated form) was themore abundant polyamine, as already observed for thecultivar Grenache noir (Faure et al. 1991). The poly-amine content was expressed both as nmol g FW�1

(Table 3) and nmol unit�1 (Figs 2 and 3). For embryo-genic callus, putrescine reached its maximum concentra-

tion, over 5mM in both cultivars, in undifferentiatedcallus (C), whereas the spd differed in the Chardonnayand Brachetto cultivars (1.5mM in callus with embryosof Chardonnay) (Table 3). When data were expressed asnmol unit�1 to overcome the dilution problem due to theincreasing amount of water in large embryos (LE) andplantlets, put generally increased during embryo devel-opment, reaching a maximum concentration in largeembryos (Fig. 2). These results agree with the polyaminelevels measured during somatic embryogenesis of Daucuscarota (Mengoli et al. 1989) and Papaver somniferum(Nabha et al. 1999). In contrast, during somatic embryodevelopment of Pinus radiata (Minocha et al. 1999),Picea abies (Santanen and Simola 1992) and Quercuspetrea (Cvikrova et al. 1998), spd was the most abundantpolyamine. A relationship between data given in

Fig. 2. Free polyamine content (nmol unit�1) in Chardonnay ( )and Brachetto g.l. ( ) cultivars. G, globular embryos; T, torpedoembryos; SE, small embryos; LE, large embryos; S1L, plant stemsplus leaves; R, plant roots. Values represent two experiments withthree determinations each.

Fig. 3. PCA-soluble conjugated polyamine content (nmol unit�1) inChardonnay ( ) and Brachetto g.l. (&) cultivars. T, torpedoembryos; SE, small embryos; LE, large embryos; S1L, plant stemsplus leaves; R, plant roots. PCA-soluble conjugated polyamineswere not detectable in globular embryos. Values represent twoexperiments with three determinations each.

662 Physiol. Plant. 120, 2004

nmol gFW�1 and in nmol unit�1 could be made only forLE and plantlets, not for callus samples. The dataexpressed in nmol unit�1 probably reflect the realincrease in polyamine accumulation during embryodevelopment. These data were in fact not influenced bythe increase in water and cell wall components, but onlyby the cell number. In contrast, the data expressed asnmol gFW�1 allowed us to compare embryogenic calluswith embryos and developing plants. The results showeda large accumulation of polyamines, especially in the freeform, suggesting a high polyamine synthesis in cell cul-tures as stated by several authors (Bagni 1989, Mengoliet al. 1989).

The free put/(spd1 spm) ratio indicates, in general, ahigh putrescine accumulation and a low utilization ofthis diamine for spermidine synthesis (Table 4). Previouspapers (Faure et al. 1991, Sass-Kiss et al. 2001) havereported the presence of only free polyamines in grape-vine and in wine. The present ratios between conjugatedput/(spd1 spm) indicate a detectable accumulation ofthe conjugated form of put (Table 4). In all samples,this value was lower than the free put/(spd1 spm)ratio as a consequence of the higher amount of conju-gated spd than of free spd. Polyamines conjugated withhydroxycinnamic acids, even if not directly involved inplant growth and development, are in fact importantregulators of the level of free polyamines that are con-sidered to be the only active form (Bagni et al. 1994). Thelow ratio (1 to 4) between free and conjugated poly-

amines during embryo development (data not shown),confirmed the presence of a high level of conjugatedpolyamines.

Our results suggest a significant role of polyamines inembryo development of the Vitis vinifera cvs. Chardonnayand Brachetto g.l. The consistent decrease of the freepolyamine amount as well as the large conjugationprocess could be related to the high efficiencies(30–40%) of embryo conversion into plantlets (Martinelliet al. 2001). Accordingly, a much lower conversion effi-ciency (,5%) was obtained in the cv. Grenache noirin which a high free polyamine content and/or aninadequate put/spd ratio was measured in non-germinatingteratological embryos with abnormal proliferation(Faure et al. 1991).

The activities of the polyamine biosynthetic enzymesADC, ODC and SAMDC increased during the progres-sion from small embryos to developing plants, with amaximum in the root samples (Figs 4 and 5). The contentof polyamines showed the same trend during embryodevelopment. In developing plants, in contrast, theincrease in enzyme activities was not associated with anenhancement of polyamines. This finding could beexplained by the dilution of proteins occurring in devel-oping plants and might suggest the degradation of poly-amines due to polyamine oxidases (PAO; EC 1.4.3.3) asdescribed by Federico and Angelini (1991). Accordingly,the levels of free 1,3-diaminopropane (Fig. 2) may

Fig. 4. ADC ( ) and ODC ( ) total (supernatant plus pellet)activities (pmol 14CO2 2h

�1mgprot�1) in Chardonnay and Brachettog.l. cultivars. C, undifferentiated callus; IC, intermediate callus; C1E,callus with embryos; G, globular embryos; T, torpedo embryos; SE,small embryos; LE, large embryos; S1L, plant stems plus leaves; R,plant roots. Values represent the mean� SE of two experiments withthree determinations each.

Fig. 5. SAMDC activity (pmol 14CO2 2 h�1 mg prot�1) in the

supernatant (&) and pellet ( ) fractions of Chardonnay andBrachetto g.l. cultivars. C, undifferentiated callus; IC, intermediatecallus; C1E, callus with embryos; G, globular embryos; T, torpedoembryos; SE, small embryos; LE, large embryos; S1L, plant stemsplus leaves; R, plant roots. Values represent the mean� SE of twoexperiments with three determinations each.

Physiol. Plant. 120, 2004 663

indicate the oxidation of spd, the concentration of whichdecreases in stem plus leaves and roots samples (S1L andR) of developing plants with respect to large embryos.

Arginine decarboxylase and ODC activities weredetected both in the soluble and in the particulate frac-tions (data not shown), thus supporting the hypothesis ofa different localization of ADC and ODC in plant cells

(Panagiotidis et al. 1982, Torrigiani et al. 1986, Borrelet al. 1995).

In contrast to what has been reported in other plantsystems (Torrigiani et al. 1986) in which SAMDC activ-ity was only detected in the soluble fraction, in grapevinecells this activity was predominant in the pellet fraction,indicating a different localization of this enzyme.

0

1

2

C IC C+E SE LE S+L R C IC C+E SE LE S+L R

Chardonnay Brachetto g.l.

Chardonnay Brachetto g.l.

C IC C+E SE LE S +L R C IC C+E SE LE S +L R

SAMDC (2.1 kb)

rRNA 18S (2 kb)

ODC (1.8 kb)

ADC (3.4 kb)

Fig. 6. Northern blot analysis ofADC, ODC and SAMDC mRNAexpression levels in differentsamples of Chardonnay andBrachetto g.l. cultivars. Theamount of hybridizingradioactivity was normalized tothe 18S rRNA signal andexpressed as relative signalintensity. ( ) ADC (&) ODC (&)SAMDC. C, undifferentiatedcallus; IC, intermediate callus;C1E, callus with embryos; SE,small embryos; LE, largeembryos; S1L plant stems plusleaves; R, plant roots. The mRNAlevels of Brachetto S1L and Rsamples were not clearlydetectable. The hybridizationswere repeated twice.

Table 4. Free and PCA-soluble conjugated polyamine ratios in Chardonnay and Brachetto g.l. cultivars. C, undifferentiated callus; IC,intermediate callus; C1E, callus with embryos; G, globular embryos; T, torpedo embryos; SE, small embryos; LE, large embryos; S1L, plantstems plus leaves; R, plant roots.

Chardonnay Brachetto g.l.

Sampleput/(spd1 spm)free

put/(spd1 spm)conjugated

put/(spd1 spm)free

put/(spd1 spm)conjugated

C 6.30 – 6.92 –IC 5.69 – 23.21 1.52C1E 1.19 – 9.92 2.88G 4.18 – 10.58 –T 8.10 – 11.79 6.87SE 2.22 2.01 4.34 1.37LE 7.02 2.31 7.63 1.72S1L 0.59 0.73 2.00 4.89R 1.86 1.69 0.90 2.14

664 Physiol. Plant. 120, 2004

Northern blot analysis of ODC and SAMDC expres-sion levels was consistent with the patterns of the enzymeactivities. It gave the highest expression of the twoenzymes in developing plant samples (Fig. 6), indicatinga direct correlation with the enzyme activities.

In conclusion, the present study contributes to theunderstanding of events leading to somatic embryogen-esis in grapevines, to embryo morphogenesis and tofurther conversion into plantlets. Our data show thatthe polyamines, as well as other plant growth factors,are involved in somatic embryogenesis in grapevine.These compounds could be involved in the active celldivision taking place during embryo differentiation, ratherthan in the cell extension predominant in developingplants, as also found for other plant species (Mengoliet al. 1989).

Acknowledgements – This work was partially supported by the ex60% funds from MURST (Ministry of University, Science andTechnological Research) of Italy to N.B. and by funds of theMinistry of Agricultural Policies, National Project ‘BiotecnologieVegetali’ to L.M. We thank Dr Ivana Gribaudo (Istituto di Viro-logia Vegetale, CNR, Grugliasco, Italy) for providing us with theembryogenic callus cultures, Mr Valentino Poletti (Istituto Agrariodi San Michele all’Adige) for his excellent technical assistance dur-ing grapevine in vitro culture, Mr N. Mele (University of Bologna)for editing of the colour illustrations, Dr Marina Franceschetti(University of Bologna) for collaboration in the design of oligo-nucleotides and Dr Mark Collier (Institute of Food Research,Norwich, UK) and Dr Richard Cirami for critically discussing themanuscript.

References

Altamura MM, Torrigiani P, Falasca G, Rossini P, Bagni N (1993)Morpho-functional gradients in superficial and deep tissuesalong tobacco stem: polyamine levels, biosynthesis and oxida-tion, and organogenesis in vitro. J Plant Physiol 142: 543–551

Bagni N (1989) Polyamines in plant growth and development. In:Bachrach U, Heimer YM (eds) The Physiology of Polyamines.CRC Press, Boca Raton, FL, Vol. 2, pp. 107–120

Bagni N, Scaramagli S, Bueno M, Della Mea M, Torrigiani P(1994) Which is the active form of polyamines in plants?. In:Caldarera CM, Clo C, Moruzzi MS (eds) Polyamines: Biologicaland Clinical Aspects. CLUEB, Bologna, pp. 131–137

Borrell A, Culianez-Macia A, Altabella T, Besford RT, Flores D,Tiburcio AF (1995) Arginine decarboxylase is localized inchloroplasts. Plant Physiol 109: 771–776

Chang S, Puryear J, Cairney J (1993) A simple and efficient methodfor isolating RNA from pine trees. Mol Biol Rep 11: 114–117

Choi JH, Sung ZR (1989) Induction, commitment, and progressionof plant embryogenesis. In: Kung SD, Arntzen CJ (eds) PlantBiotechnology. Butterworths, Boston, MA, pp. 141–159

Cvikrova M, Mala J, Eder J, Hrubcova M, Vagner M (1998)Abscisic acid, polyamines and phenolic acids in sessile oaksomatic embryos in relation to their conversion potential.Plant Physiol Biochem 36: 247–255

El Hadrami I, Michaux-Ferriere N, Carron MP, D’Auzac J (1989)Les polyamines, facteur limitant possible de l’embryogenesesomatique chez l’Hevea brasiliensis. CR Acad Sci Paris, SerieIII: 205–211

Faure O, Mengoli M, Nougarede A, Bagni N (1991) Polyaminepattern and biosynthesis in zygotic and somatic embryo stagesof Vitis vinifera. J Plant Physiol 138: 545–549

Federico R, Angelini R (1991) Polyamine catabolism. In: Slocum RD,Flores HE (eds) Biochemistry and Physiology of Polyamines inPlants. CRC Press, Boca Raton, FL, pp. 41–56

Feirer RP (1995) The biochemistry of conifer embryo development:amino acids, polyamines and storage proteins. In: Jain SM,

Gupta PK, Newton RJ (eds) Somatic Embryogenesis inWoody Plants. Kluwer Academic Publishers, Dordrecht, Vol. 1,pp. 317–336

Feirer RP, Mignon G, Litvay JD (1984) Arginine decarboxylaseand polyamines required for embryogenesis in the wild carrot.Science 223: 1433–1435

Fienberg AA, Choi JH, Lubich WP, Sung ZR (1984) Developmen-tal regulation of polyamine metabolism in growth and differen-tiation of carrot culture. Planta 162: 532–539

Fobert PR, Webb DT (1988) Effects of polyamines, polyamineprecursors, and polyamine biosynthetic inhibitors on somaticembryogenesis from eggplant (Solanum melongena) cotyledons.Can J Bot 66: 1734–1742

Gamborg OL, Miller LA, Ojiama K (1968) Nutrient requirements ofsuspension cultures of soybean root cells. Exp Cell Res 50: 151–158

Gianazza E, De Ponti P, Scienza A, Villa P, Martinelli L (1992)Monitoring by two-dimensional electrophoresis somatic embryo-genesis in leaf and petiole explants from Vitis. Electrophoresis 13:203–209

Helleboid S, Couillerot JP, Hilbert JL, Vasseur J (1995) Inhibitionof direct somatic embryogenesis by a a-difluoromethylargininein a Cichorium hybrid: effect of polyamine content and proteinpatterns. Planta 196: 571–576

Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Proteinmeasurement with the Folin phenol reagent. J Biol Chem 193:265–275

Martinelli L, Gribaudo I (2001) Somatic embryogenesis in grape-vine (Vitis spp.). In: Roublekais-Angelakis K (ed) MolecularBiology and Biotechnology of Grapevine. Kluwer AcademicPublishers, Dordrecht, pp. 327–352

Martinelli L, Gribaudo I, Bertoldi D, Candioli E, Poletti V (2001)High efficiency somatic embryogenesis and plant germination ingrapevine cultivars Chardonnay and Brachetto a grappololungo. Vitis 40: 111–115

Martinelli L, Scienza A, Villa P, De Ponti P, Gianazza E (1993)Enzyme markers for somatic embryogenesis in Vitis. J PlantPhysiol 141: 476–481

MengoliM,BagniN,LuccariniG,NutiRonchiV,Serafini-FracassiniD(1989)Daucus carota cell cultures: polyamines and effect of poly-amine biosynthesis inhibitors in the pre-embryogenic phase anddifferent embryo stages. J Plant Physiol 134: 389–394

Minocha R, Smith DR, Reeves C, Steele KD, Minocha SC (1999)Polyamine levels during the development of zygotic and somaticembryos of Pinus radiata. Physiol Plant 105: 155–164

Murashige T, Skoog F (1962) A revised medium for rapid growth andbio assays with tobacco tissue cultures. Physiol Plant 15: 473–497

Nabha S, Lamblin F, Gillet F, Laurain D, Fliniaux M, David A,Jacquin A (1999) Polyamine content and somatic embryogenesisin Papaver somniferum cells transformed with sam-1 gene. J PlantPhysiol 154: 729–734

Nagl W (1990) Translocation of putrescine in the ovule, suspensorand embryo of Phaseolus coccineus. J Plant Physiol 136: 587–591

Nitsch JP, Nitsch C (1969) Haploid plants from pollen grains.Science 163: 85–87

Panagiotidis CA, Georgatsos JG, Kyriakidis DA (1982) Superin-duction of cytosolic and chromatin-bound ornithine decarboxyl-ase activities of germinating barley seeds by actinomycinD. FEBS Lett 146: 193–196

Primikirios NI, Roubelakis-Angelakis KA (1999) Cloning andexpression of an arginine decarboxylase cDNA from Vitis vini-fera L. cell-suspension cultures. Planta 208: 574–582

Robie CA, Minocha SC (1989) Polyamines and somatic embryo-genesis in carrot. I. The effects of difluoromethylornithine anddifluoromethylarginine. Plant Sci 65: 45–54

Sambrook J, Fritisch EF, Maniatis T (1989) Molecular Cloning: ALaboratory Manual. Cold Spring Harbor Laboratory Press,Cold Spring Harbor, NY

Santanen A, Simola LK (1992) Changes in polyamine metabolismduring somatic embryogenesis in Picea abies. J Plant Physiol 140:475–480

Sass-Kiss A, Hajos Gy, Brezovcsik Gy, Szerdahelyi E, Bardocz S(2001) Biogenically active amines of grapevines, winesand aszu-wines of Tokaj. In: Morgan DML, Milovic V,Krizek M, White A (eds) Biogenically Active Amines inFood. Vol. V. Polyamines and Tumour Growth – Biogenically

Physiol. Plant. 120, 2004 665

Active Amines in Food Processing and Amines Producedby Bacteria. COST Action 917. European Communities,Luxembourg, pp. 84–89

Tassoni A, van Buuren M, Franceschetti M, Fornale S, Bagni N(2000) Polyamine content and metabolism in Arabidopsis

thaliana and effect of spermidine on plant development. PlantPhysiol Biochem 38: 383–393

Torrigiani P, Serafini-Fracassini D, Biondi S, Bagni N (1986)Evidence for the subcellular localization of polyamines and theirbiosynthetic enzymes in plant cells. J Plant Physiol 124: 23–26

Edited by P. Nissen

666 Physiol. Plant. 120, 2004