Pergamon

PII: S0003-^878(98)00066-0

Ann. oi-cup. Hyg.. Vol. 42. No. 8. pp. 521-530. 1998c 1998 British Occupational Hygiene Society

Published by Elsevier Science Ltd. All rights reservedPrinted in Great Britain.

0003 4878,98 SI9.00+ 0.00

Surface Chemistry and Quartz HazardB. FUBINI*

[I'niiersita di Torino, Facolta di Farmacia. Dipartimento di Chimica InoraanicajChimica Fisica eChimica del Maleriali. Via Pielro Giuria 7. \Q\25{Torino. Itafv] .

I The variability of quartz hazard is related to the characteristics of particulate toxicants. Althoughthese have the same chemical composition, they exist in various forms and surface states, eachone eliciting different biological responses. On the basis of data from the literature, surfacechemical properties are associated to the subsequent stages reported by Donaldson and Borm(1998) in the mechanistic model proposed for quartz carcinogenicity. Surface radicals and iron-derived reactive oxygen species (ROS) are implicated in oxidative stress, considered to be the keyevent in the development of librosis and lung cancer. Other chemical functionalities related tocytotoxicity, however, modulate the overall pathogenicity by regulating transport and clearance.The chemical features deriving from the intrinsic characteristics of a silica dust—e.g. its origin—as well as those generated by external factors—e.g. contaminants, associated minerals—arediscussed in relation to their possible role in the pathogenic mechanism^ i 1998 British Occu-pational Hygiene Society. Published by Elsevier Science Ltd.

Keywords: crystalline silica: quartz; diatomite earth: fibrogenicity; carcinogenicity: surface chemistry

INTRODUCTION

The paper "Quartz hazard: a variable entity' (Don-aldson and Borm, 1998) well reflects what were thedifficulties met by the IARC working group in eva-luating the carcinogenicity of crystalline silica (IARC,1997) and focuses on the differences existing betweenmolecular and paniculate toxicants. Particulates neveract as a constant entity, their reactivity in a biologicalmedium depending on the micromorphology at theatomic level and on the mechanical, thermal andchemical history of a given dust, as well as the frequentpresence of surface contaminants (Fubini et al., 1998in press). This is particularly true with silica: becauseof the partially covalent silicon-oxygen bond of thismaterial, several crystalline forms are found in nature,with different biological activity (reviewed by Guthrie,1995) and more than ten different chemical func-tionalities may be stabilized at the surface (recentlyreviewed by Fubini, 1998). Therefore, in contrast toe.g. benzene, which is a toxicant, with only one formeach source of crystalline silica dust has its own car-cinogenic potential, which can be largely modified byeven slight alterations of the state of the surface. Well-targeted acellular and/or cellular in vitro tests, using

Received 1 June 1998; in final form 6 June 1998.* Author to whom correspondence should be addressed.Tel: (39) 11 670 7566; Fax: (39) 11 670 7855; E-mail[fubinifa.ch.unito.in

silica samples with controlled properties, should beemployed to investigate the molecular basis of themechanism(s) of action of particulate silica, and helpin the interpretation of the apparently contradictoryepidemiological results.

The aim of the present communication is to clarifythe chemical basis of the questions raised by Don-aldson and Borm and to point out a few additionalpoints on the variability of quartz hazard.

The evaluation by the IARC working group statedthat carcinogenicity may be dependent on inherentcharacteristics of the crystalline silica or on externalfactors affecting its biological activity.

The 'inherent characteristics' of the silica areaccounted for by the state of the external surface(defects, chemical functionalities etc.) determined bythe origin of the sample, while the 'external factors'suggest that contact, association or contamination bysubstances other than silica might activate (or blunt)silica carcinogenicity. In both cases we will havedifferent chemical/toxicant entities having the samenominal composition, SiO2, but with different behav-iour towards living matter.

An additional point raised in the present com-munication is that both inherent characteristics andexternal factors may act in different stages of the devel-opment of the disease, i.e. play a role in more thanone of the events leading from deposition of quartz tosilicosis and lung cancer. Different surface func-tionalities may be implicated in each step. Recent stud-

521

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from

522 B. Fubini

ies examining the intensity of the various biologicalresponses elicited by a series of different silica dustshave in fact shown that some of the endpoints sug-gested by Donaldson and Borm, e.g. cytotoxity.inflammation, transforming potency and DNA dam-age are not related to the same property of the silicaparticle (Daniel et al., 1995; Elias et al., 1995; Elias etal., submitted). The carcinogenic potency of a givendust is the result ofhow and to what extent each of theparticle characteristics plays a role in one (or more)of the subsequent cellular and molecular events takingplace in the carcinogenic mechanism.

The present communication is an attempt to extractfrom the vast literature on silica related biologicaleffects the following information:

1. which are the physico-chemical properties impli-cated in the various stages of the proposed patho-genic mechanism

2. how the 'origin" of a silica dust determines thepathogenic potential

3. which are the possible chemical roles played by'external factors'

CHEMICAL PROPERTIES INVOLVED IN THESUBSEQUENT STEPS OF THE PATHOGENIC

RESPONSE TO SILICA

Figure 1 shows a modified version of the mechanismreported by Donaldson and Borm, in which surfacecharacteristics have been tentatively associated withthe various steps in the mechanism yielding the devel-opment of cancer, on the basis of the results obtainedin animal and cellular tests on variously modified silicasamples. The pathway involving inflammation as anintermediate state, from which factors inducing trans-formation and proliferation of epithelial cells orig-

inhibition of clearance

silica particle in the lung

• form,• crystal structure'SiOH•SiO-• surface radicals, charges

a

contact with alveolar macrophages:stimulation, cell damage andphagocytosis

ROS:particle-derived

cell- derived

**t * -

clearance

oxidative stress

aluminium ionsother metal ionshydrophobic surfacePVPNO coating

ironparticle derived ROSfree radicalssurface radicals

g

cancer

effect on epithelial cells:proliferation and mutation

in'L I- r ° ,, v P ySiC° C h e m i C a l f a C t ° r S in l h e S e q u e n c e o f e v e m s l e a d i nS t 0 t h e Pathologies associated toinhalation or crystalline s,l,ca reported by Donaldson and Borm (1998). The form of the particle, crystall.nity and the

f v oLrmLronh rA8M^PSH(S^H) " ^ *"?*" ^ " ^ ^ e X t e m ° f Ce" da™^ a c l l v a t i o n a n d stimulation ofalveolar macrophages AM) and polymorphonuclea.ed cells (PMN) Also dissociated silanols (SiO -) and features related tomechamcal activate (surface radicals and charges) may play a role at this stage (a). Phagocytosis ends up with rupture ofthe phagolysosomal membrane, cell death, inhibition of clearance and accumulation of silica particles in the lung (b) If the

pavnp°NnTPS 3 r e ( l°,n 'C b°nudS l ° m C t a l iOnS- h y d r O g e " b o n d i n 8 t 0 P°'ymers s u c h a s Polyvinyl-pyridin-N-oxide(PVPNO) or converse mto siloxane bndges (Si-O-Si) by thermal treatments imparting hydrophobicity to the surface) themembrano ytic and cytotoxic potency of the dust decreases and the particles can be cleared out from the lung (c) DuringovZTT" °LVnt P M N - Ce""deriVed 3 n d ^ M ™ ^ Reactive Oxygen Species (ROS) both contribute "oev^ntuan () ' n f l a n T l : ° n <e ; y'eldln§ s i l i c o s i s O- proliferate and mutagenic effects on epithelial cells (g) andeventually lung cancer. Several other cell derived factors (nitric oxide, cytokines. arach.donic acid metabolites and Jrowthfactors) are also released. N.tnc oxide may react with fibre-released superoxide anion. with formation of peroxonitriterad cak

OIe f ^ i a l l c a P ^ i c l e s a n d e P i t h e l i a ' «=Us: particle derived ROS (peroxobridges. superoxide L s , hvdroToradicals generated via Fenton chemistry by traces of iron) induce transformations in these cells (h). enhancing the'effects

caused by the oxidative stress (g)

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from

Surface chemistry and quartz hazard 523

inate, appears the most plausible (IARC, 1997). It hasbeen reported, in fact, that mutations were caused byinflammatory leucocytes from quartz exposed rats,but quartz itself had no mutagenic effects on epithelialcells in culture (Driscoll et al, 1995). However, anadditional role of direct particle/target cell interaction,taking place on proliferating cells stimulated andtransformed by inflammation, cannot be fullydiscarded. This latter path has therefore been addedto the scheme of Donaldson and Borm, consideringthat various studies have shown morphological trans-formations following direct contact of V79 cell lines(Lin et al, 1996; Zhong et al, 1997) and of someembryo cells with quartz (Hesterberg etai, 1986; Eliaset al., 1995; Saffiotti and Ahmed, 1995; Elias el al.,submitted). Moreover, silicosis and cancer do notnecessarily follow a fully common pathway, even iffibrosis appear to be a prerequisite for the devel-opment of quartz-associated cancer. Several cases ofnon-tumorigenic quartz sources have been reported(e.g. coal mine dusts, some gold, tungsten and zincmines) while in very few cases (mostly confined to fewlaboratory samples) crystalline silica dusts were foundto be non-fibrogenic. This suggests that some specificfeatures impart tumorigenicity to a fibrogenic silicadust or alternatively that some surface modificationsmay affect tumorigenicity without affecting the fib-rogenic potential.

Figure 1 reports the following steps:Step a: Once deposited in the alveolar space the

silica particle will cause cell damage and stimulation;silica induced membranolysis, originated by strongadsorption of membrane components onto the silicaparticle (Nash et al., 1966) is related to the distributionand abundance of silanols (SiOH) groups at the sur-face (Hemenway et al., 1993; Fubini, 1997) and tosilanols groups dissociated in water (Nolan et al.,1981); cytotoxicity, as measured by inhibition of cellgrowth or cytosolic enzyme leaking, is also relatedto silanols (Pandurangi et al., 1990; Fubini et al.,submitted). If the particle is coated with polymers(Mao et al., 1995), has been chemically modified(Wiessner el al., 1990), is hydrophobic (Fubini el al.,submitted) or has been treated with aluminium salts(Brown and Donaldson, 1996) the effect of silanols ismuch reduced or even blunted. Under these cir-cumstances the particle will follow path c in Fig. 1,i.e. will be cleared from the lung to the upper airwaysor to lymphonodes by macrophages. Alternatively,following path b, clearance will be inhibited andphagocytosis will eventually end up with cell deathfollowing disruption of the phagolysosomemembrane. A continuous ingestion-reingestion cycle,with accumulation of the free particles in the lungand persistent inflammation with release of cytokines,Reactive Oxygen Species (ROS), arachidonic acidmetabolites and growth factors, will be established.

Step d and e: Particle derived ROS (Fubini et al.,1989a; Giamello et al., 1990; Dalai et al., 1990; Shi etal., 1995) and cell derived ROS (Vallyathan el al.,

1992) will both contribute to a state of oxidative stress(d), persisting as long as the inflammation (e) persists.Cells will also release nitric oxide which contributesto the oxidative stress and in the presence of the super-oxide ion forms the dangerous compound peroxon-itrite.

At this stage species differences in the response tosilica may show up. Rats are more susceptible thanmice to the fibrotic action of silica, and silica derivedlung cancers have been found in rats but not in miceor hamsters (IARC, 1997). It has been recentlyreported that the amount of macrophage derivednitric oxide released varies remarkably between ani-mal species, e.g. rat and hamster (Dorger et al., 1997).Nitric oxide will contribute substantially to the oxi-dative stress. Morphometric analysis of AM fromhumans and several animal species suggested thatnumber and size range of particles that can be phag-ocytosized and cleared differs among species, as aconsequence of AM cell size (Krombach el al., 1997).Therefore, with the same silica dust, inhaled particlesfollowing path b and c will vary between species. Therelease of some cytokines by AM has also beenrecently reported to be species specific. TNF-aresponse to silica was in fact downregulated in micebut upregulated in rats cell culture from bron-choalveolar fluids following silica exposure (Huaux,1998), which may account for the larger fibrotic actionof silica on rats.

Step h. Particle derived ROS, such as free radicalsor peroxides, are implicated in direct damage to theepithelial cells. Several particle derived ROS have beenreported, such as hydroxyl radical, superoxide anionand peroxides (Shi et al., 1995). The production ofsilica-derived free radicals is much higher on freshlyground materials, where surface peroxide or hyd-roperoxides are formed (Fubini et al., 1990; Giamelloet al., 1990; Volante et al, 1994); therefore this step ismore relevant in the case of freshly ground than agedsilicas. If some iron, even a trace, is present at thesilica surface—which is a very common situation withmineral samples, and even with so-called pure samplesas Min-U-Sil (Saffiotti and Ahmed, 1995)—Fentonchemistry may be activated, with consequent pro-longed release of radicals, which may cause DNAdamage and transformation in target cells. Free rad-ical generation does not usually relate to the actualamount of iron but to small fractions of iron with aparticular redox and coordination state (Fubini et al,1995b; Gilmour el al, 1995). The availability of ironsites at the surface will therefore also depend on sur-face micromorphology, history etc.

As a consequence of step g and h, mutations andproliferation in epithelial cells may initiate a neo-plastic transformation. Therefore the potential of theinhaled particles to catalyse ROS release and to per-sistently activate macrophages would determine thecarcinogenicity of a given dust. However, any surfaceproperty favouring path c instead of b, by loweringthe extent of accumulation of the dust in the lungs

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from

524 B. Fubini

and consequent inflammation, will also lower the car-cinogenic potential.

ORIGIN OF THE DUST

A 'native, uncontaminated quartz surface', i.e. asurface made up of real crystal faces, is very rarely indirect contact with biological matter. The only casereported concerns in vitro and in vivo experiments onquartz microcrystals exhibiting perfect habit, whichwere obtained in micrometre size by crystal growth inhydrothermal conditions (Czernikowski et ai, 1991).They were more fibrogenic to rats than other quartzdusts but no data were reported which could giveindications on their potential carcinogenicity. Crys-talline silica dusts of respirable size are usually gen-erated either by grinding macroscopic crystals ofquartz —or of other polymorphs—of mineral originor by heating/calcining biogenic silicas, mainly diato-maceous earths.

In the former case, mechanical fracture does notcommonly follow crystal planes. The shape of theparticles generated is very irregular, with sharp edgesand spikes. Small particles (smaller than 50 nm indiameter) stick to bigger ones, firmly held by the sur-face charges produced by grinding (Fubini, 1998).Freshly ground dusts are more fibrogenic than agedones (Goethe et at., 1971; Vallyathan et ai, 1995). Thesurface produced by mechanical cleavage of chemicalbonds is usually very reactive, and the state of thesurface depends markedly on the grinding procedureand the components of the environment in which thegrinding took place (Fubini et ai, 1989a; Costa et ai,1991). A dry oxygen atmosphere favours formationof surface radicals and ROS, while a wet one assistsfull surface hydration at broken bonds, with virtuallyno yield in surface reactive forms. If ROS, as hypo-thesized, are implicated in some stages of the patho-genic mechanism, mining and processing the same orewith different procedures may generate dusts differingin their pathogenic potential. Prolonged milling pro-gressively converts the outer parts of the particlesfrom crystalline to amorphous, which lowers the dusttoxicity. Removal of this amorphous external layer(Beilby layer) from quartz by chemical etching with

hydrofluoric acid yields an increase in the fibrogenicityof the dust (Engelbrecht et ai, 1958).

The other common source of respirable crystallinesilica dusts is the various 'biogenic silicas' such asdiatomaceous earths, rice husks, etc. These dusts areusually made up of amorphous silica particles, whichstill retain the form of the living matter from whichthey were generated. Upon mild heating, however,they are converted directly into cristobalite, withoutlarge variation in their micromorphology, nor sin-tering with increase in particle size (Fubini et ai,1995a). Because of their origin they retain a high levelof impurities, often including alkali metals and alka-line earth oxides and iron ions (IARC, 1997).

Fly ashes and fuel ashes are other sources of par-ticulate silica which are potentially toxic. Usuallythese dusts have experienced a very high temperaturewhich modifies the surface state. A review on thehealth effects related to fuel ashes reports minimaleffects on experimental animals (Raask and Schilling,1980). Quartz in these ashes was found in the form ofspheres or rounded particles, produced by the actionof the surface tension on the near-to-melt particles. Infly ashes too, quartz is in the form of smooth roundspheres. Form, surface micromorphology (smooth vs.scratched) and composition are all modified at hightemperature. Heating in fact progressively converts ahydrophilic surface into a hydrophobic one (Hemen-wayetai, 1994; Fubini et ai, 1995a) and anneals mostsurface radicals (Fubini, 1994), causing a remarkabledecrease both in cytotoxicity (Fubini et ai, submitted)and in the transforming potency of the dust (Elias etai, submitted). Coal fly ashes have been found muchless active than quartz dust in TNF release, probablybecause of the modifications which had taken place atthe surface of the silica particles exposed to a hightemperature. No surface radicals and very few freeradicals were detected with this material; these wereon the other hand abundant in the pure quartz dust(Min-U-Sil)used for comparison (Borm, private com-munication).

The characteristics of silica dusts of different originare summarized and compared in Table I. Table 2reports the consequences of some of the inherentcharacteristics of a silica dust on the chemical status

Table 1. Characteristics of the participates determined by their origin

Comminution of crystals e.g.grinding, ball milling

Combustion e.g. coal ashes,fly ashes

Biogenic e.g.diatomaceous earth,rice husks

sharp edges and spikes

irregular surface, chargessurface radicalshydrophiliccontamination by components of thegrinding chamber

spherical particles

smooth surfaceno surface radicalhydrophobiccontamination by carbon and othercomponents

retention of shape from livingorganisms.indented irregular surfaceno surface radicalsvery hydrophilicalkaline, alkaline earth, and iron ionsfrom original material

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from

Table 2. Influence of the inherent characteristics of a silica dust on surface properties and on some biological responses, considered possible end-points in the mechanistic model proposed(Figure 1)

Inherentcharacteristics

Physico-chemicaleffects'1

In vitro cytotoxicitymembranolysish

In vivo inflammation,Iibrosish

Cell-freeDNA damagch

DNA damage and/ortransformation in cells'1

crvst-.illinitv

form, micromorphology

grinding, ball milling

thermal treatments

etching HF

HCI

etching K.OH

crystal habit determines thedensity of Si and O atoms atthe surface

edges and surfaces irregularand indented at the atomic-level (reactive) vs smooth (lessreactive)

generates fresh surfaces withradicals and charges

Convert silanols SiOH insiloxane bridges Si-O -Siwith increase inhydrophobicity

Removes amorphous layersand metal contaminants. Si-F at the surface

Removes metal contaminants

Removes external layers. K +

at the surface

smaller size moremembranolytic Wiessner eltil.. 1989

most crystalline (quartz,tridymite and cristobalitc)and some amorphous silicasare membranolytic andcytotoxic. Reviewed byDriscoll (1995)

| Pandurangi el al.. 1990;Hemenway el al., 1993

I Nolan <•/«/.. 1981; | Danieleial.. 1995

Nolan eial.. 1981

smaller size less fibrogenicWiessner el al., 1988b

only crystalline silicas arefibrogenic. fibrogenicityvaries from one to the otherpolymorph Wiessner el al..1988a

| Vallyathan el al.. 1995

modify translocation tolymphnodes and clearanceHemenway el al., 1994

t Engelbrecht el al.. 1958

quartz, cristobalite andtridymite Daniel el al., 1993

| Daniel el al.. 1995

| Miles el al.. 1994

crystalline silicas anddiatomitc earth Safliotti el al.,1993; Elias <-/«/., 1994 andElias ei al.. submitted; Hartand Hersterberg, 1998

| elongated shape Hart andHersterberg, 1998

t Elias <'/«/., 1994 and Eliasel al., submitted

| Elias eial., 1994

Safliotti and Ahmed. 1995

[ indicates that the intensity ol the biological response is lower; f indicates that the intensity of the biological response is higher" reference to chemical features from Her (1979) and Fubini (1998);b effects also discussed in IARC. 1997 and in the review by Fubini (1998)

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from

526 B. Fubini

of the surface and on some biological responses whichcan be considered relevant end-points on the basis ofthe mechanistic model proposed in Fig. 1.

THE ROLE OF 'EXTERNAL FACTORS' METALCONTAMINANTS, ADSORPTION AND COATING

A vast literature exist on the effects of contaminantsin silica related health effects, which has been recentlyreviewed (Fubini, 1998). The major facts are sum-marized in Table 3.

The most common and relevant contaminantassociated to quartz is coal dust. When in intimatecontact with it (in coal mine dusts or in ground mix-tures), quartz looses its pathogenic potential (Don-aldson and Borm, 1998). Some metals or metal oxides(see Table 3) appear to act in similar way. The sourcesof quartz found to be non-carcinogenic in epi-demiological studies (IARC, 1997) are all from met-allic mines, gold, zinc and tungsten. Intimate contactbetween quartz and carbon or metals—all reducingagents—probably modifies the nature of the surfacesites involved in the carcinogenic mechanism. A poss-ible explanation is that carbon or metals assist theannealing of surface radicals, thus eliminating particlegenerated ROS, reducing the oxidative stress andhence damage to epithelial cells. Alternatively metalions, by binding to silanols, will reduce membranolysisand cytotoxicity, favouring elimination of the par-ticles via path c in Fig. 1, instead of accumulation andinflammation occurring in path b. This latter hypoth-esis suggests that these contaminants may act similarlyto the inhibitors of silica fibrogenicity. The two mostcommon ways found to inhibit the fibrogenic responsein experimental animals, and occasionally employedas prevention in humans, are treatment with alu-minium salts (Le Bouffant el al, 1977; Begin el al,1987) or with the polymer polyvinyl-pyridin-N-oxide(PVPNO) (reviewed by Castranova, 1996). Both actby blunting the cytotoxicity of silanols: metal ionsreplace hydrogen in silanols, the — N = O groups inthe polymer are strongly hydrogen-bonded to silanols,thus preventing adsorption of cell membrane com-ponents onto the silica surface. The reported Table 3reduction in cytotoxicity of chemically modified silicascoated by various polymers, surfactants, etc (Wiessnerel al, 1990) is likely to be due to similar effects.

Metal contaminants, and particularly iron, will alsocause adverse effects. The role of traces of iron inthe generation of ROS, causing DNA damage, celltransformation and pulmonary reactions, is welldocumented (Castranova el al., 1997). The highpathogenicity found upon exposure to crystalline sil-icas of biogenic origin, mostly ex diatomite earth,(IARC, 1997), can probably be ascribed to the pres-ence of iron ions, derived from the original livingmatter, always present in these kind of materials. Asreported with other mineral dusts (Gilmour el al.1995). not all iron is active, iron in oxides-haematiteand magnetite—having been found inactive in ROS

generation (Fubini el al, 1995b). Conversely a bulkoxide (haematite Fe2O3) mixed with quartz decreasedboth cytotoxicity and transforming potency (Saffiottiand Ahmed, 1995).

The effect of chemical etching deserves some com-ments, as it can cause apparently conflicting effects.Very rarely have people been exposed to chemicallyetched dusts. Several studies however, report in vitroand in vivo experiments performed on untreated andacid treated (HF, HC1) silica dusts. This procedure,which removes iron (Saffiotti and Ahmed, 1995)—and probably other metal impurities— with HF alsoremoves the outmost amorphous layer, leaving a freshcrystalline face exposed. Cytotoxicity was enhanced(Kriegseis et al, 1987), probably because of removalof metal ions, and fibrogenicity too (Engelbrecht elal, 1958), because of exposure of crystalline faces tocells. The surface, however, may retain some featuresoriginated by the etching chemical employed. It isnoteworthy that membranolysis was decreased byhydrofluoric etching but increased by basic (K.OH)etching (Nolan et al, 1981). Small traces of fluorideor potassium ions are thus sufficient to modulate par-ticle/cell interaction.

Removal of iron, on the other hand, eliminates oneof the major sources of free radicals and consequentDNA damage: when removed by acids both effectswere in fact dramatically decreased (Daniel et al,1995), confirming a crucial role played by metals, evenin trace amounts, associated to specific surface siteslocated at some crystal faces of quartz.

CONCLUSIONS

Several chemical factors contribute to the devel-opment of chronic inflammation and subsequent sili-cosis following inhalation of silica particles. In thehypothesis of a common pathway for fibrosis andlung cancer, modification of any of these factors, bymodulating the fibrogenic response, will also affect thecarcinogenicity of a given dust. This is sufficient toexplain the extreme variability in carcinogenic poten-tial among different sources of silica containing dusts.

However, fibrosis and lung cancer need not proceedvia common molecular pathways, as has been sug-gested by Donaldson and Borm, even if fibrosis is aprerequisite for the development of silica-associatedlung cancer. By the same token the chemical proper-ties involved in the fibrotic mechanism may be differ-ent from those causing lung cancer in silica exposedpopulations, even if, by blunting the inflammatoryresponse, the carcinogenic potential decreases. Car-cinogenicity may stem from surface sites (e.g. surfaceROS generated upon grinding, iron ions in appro-priate coordination, surface defects which can acco-modate endogenous iron) able to catalyse free radicalgeneration. This latter process, on the one hand, willcontribute to the onset of oxidative stress, and on theother hand, will cause direct damage to epithelial cells,already damaged and/or proliferating because of

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from

Table 3. Effect of external factors on the surface characteristics of a silica dust and related influence on some of the biological responses, considered possible end-points in the mechanistic modelproposed (ligure I)

Externalfactors

surfacemodifications'

in vitro cylotoxicitymembranolysisb

in vivo inflammation,fibrosis"

cell-freeDNA damage11

DNA damage, and/orIransformation in cells'1

ContaminantsCarbon(intimate mixture)

Aluminium ions

iron ions

Alkaline ions: Ca ; '

iron oxide (haematite)

Metallic iron

Kaolin, illyte

Chemical modificationsOrganosilanes linked to thesurface (covalent bonding)

Coatiiu/sPVPNO

Phospholipids

Reduction of surfacesites'.'Annealing of radicals?

SiOAl replace SiOH

SiO Fe ' . free radical generation

SiO Ca'

Reduction of surface sites?Annealing of radicals?

Release of aluminium ions?

i Kriegseis ei al.. 1987;

| Nolan ct al. 1981

| Kriegseis ct al., 1987

| Saffiotti and Ahmed. 1995

1 Lc Bouffant ct al.. 1977

Le Bouffant ct al., 1977

Reduction of free silanols. changes j Wiessner ct al.. 1990:of hydrophilicity vs hydrophobicity Castranova el al.. 1996

| Cullen eta/., 1997

| Le Bouffant el al.. 1977

i Wiessner el al.. 1990

Binds to silanols (H-bonding)

Adsorbs onto the surface

| Nolan et al.. 1981; Klockars J reviewed by Castranovaei al.. 1990: Mao el al.. 1995 (1996)

1 Wallace ct al., 1985;Wiessner el al.. 1990;Antonini and Rcasor. 1994

unaffected Wiessner el al..1990J. Antonini and Reasor,1994

I Saffiotti and Ahmed,1995

t Safliolli and Ahmed,1995

I Sufliolti and Ahmed,1995

3.

I indicates that the intensity of the biological response is lower; | indicates that the intensity of the biological response is higher" reference to chemical features from Her (1979) and Fubini (1998);b effects discussed in IARC. 1997 and in the review by Fubini (1998)

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from

528 B. Fubini

chronic inflammation. The abundance, stability andreactivity of these kinds of surface sites is extremelyvariable, their nature being strictly dependent on thehistory of the dust and readily modified by ambientconditions, chemical reactions and presence of con-taminants. The variability of'quartz hazard' is there-fore a direct consequence of the physico-chemicalproperties of silica. Accurate analysis of the charac-teristics of the dusts to which the cohorts of the variousepidemiological studies were exposed might help inthe clarification of the factors which favour the devel-opment of silica-associated lung cancer. Once thepathogenic mechanism(s) are fully elucidated at themolecular level, the hazard associated to a givensource of crystalline silica should be predictable.

REFERENCES

Antonini, J. M. and Reasor, M. J. (1994) Effect of short-term exogenous pulmonary surfactant treatment on acutelung ddamage associated with the inlratracheal instillationof silica. J. Toxicol. Environ. Health 43, 85-101.

Begin, R., Masse, S., Sebastien, P., Martel, M., Bosse, J..Dubois, F., Geoffrey, M. and Labbe, J. (1987) Sustainedefficacy of aluminum to reduce quartz toxicity in the lung.Exp. Lung. Res. 13, 205-222.

Brown, G. M. and Donaldson, K. (1996) Modulation ofquartz toxicity by aluminium. In: Silica anil Silica-InducedLung Diseases, pp. 299-304. CRC Press, Boca Raton. F.USA.

Castranova, V. (1996) Suppression of the cytotoxicity andfibrogenicity of silica with PVPNO. In: Silica and Silica-Induced Lung Diseases (Edited by Castranova, V., Vally-athan, V. and Wallace W. E.), pp. 293-298. CRC Press.Boca Raton, F. USA.

Castranova, V., van Dyke, K., Wu, L., Dalai, N.S. andVallyathan, V. (1996) Suppression of silica-induced tox-icity with organosilane surface coating. In: Silica andSilica-Induced Lung Diseases (Edited by Castranova. V..Vallyathan, V., and Wallace W. E.), pp. 283-292. eds CRCPress, Boca Raton, F. USA.

Castranova. V., Vallyathan, V., Ramsey. D. M.. McLaurin.J. L., Pack, D.. Leonard, S., Barger, M. W.. Ma. J. Y.C , Dalai, N. S. and Teass, A. (1997) Augmentation ofpulmonary reactions to quartz inhalation by traceamounts of iron-containing particles. EnvironmentalHealth Perspectives 105(5), 1319-1324.

Costa, D., Fubini, B., Giamello, E. and Volante, M. (1991)A novel type of active site at the surface of quartz and itspossible impact on pathogenicity. Can. J. Chem. 69, 1427-1434.

Cullen, R. T., Vallyathan, V.. Hagen, S. and Donaldson. K.(1997) Protection by iron against the toxic effects of quartz.Ann. Occup. Hyg. 41(1), 420-425.

Czernikowski, M., Van Damme. H. and Erre. R. (1991)Caracterization des etats des surface de quartz et desdepots aluminiques en relation avec la silicose. InMechanisms in occupational lung diseases (Sebastien P. ed.)Colloque INSERM. 203. 271-303.

Dalai. N. S.. Shi. X. and Vallyathan. V. (1990) Role of thefree radicals in the mechanisms of hemolysis and lipidperoxidation by silica: comparative ESR and cytotoxicitystudies. J. Toxicol. Emir. Health 29, 307-316.

Daniel. N. L.. Mao. Y. and Safiiotti. U. (1993) OxidativeDNA damage by crystalline silica. Free Rad. Biol. Med.14, 463-^72.

Daniel. L. N.. Mao. Y.. Wang. T.-C. L.. Markey. C. J..Markey. S. P.. Shi. X. and Saffiotti. U. (1995) DNA strandbreakage, thymine glycol production, and hydroxyl radical

generation induced by different samples of crystalline silicain vitro. Environ. Res. 71, 60-73.

Donaldson. K. and Borm. P. J. A. (1998) The quartz hazard:a variable entity. Ann. occup. Hyg. in press.

Dorger, M., Jesh, N. K., Rieder, G., Hirvonen, M.-R., Sav-olainen, K., Krombach, F. and Messmer, K. (1997) Spec-ies differences in nitric oxide formation by rat and hamsteralveolar macrophages in vitro. Am. J. Respir Cell Mol.Biol. 16,413-420.

Driscoll, K. E. (1995) The toxicology of crystalline silicastudied in vitro. Appl. Occup. Environ. Hvg. 10, 1118-1125.

Driscoll, K. E., Deyo, L. C , Howard, B. W., Poynter, J. andCarter, J. M. (1995) Characterizing mutagenesis in thehrpt gene of alveolar epithelial-cells. Experimental LungResearch 21,941-956.

Elias, Z., Poirot, O., Daniere, M. C , Terzetti, F., Marande,A. M., Dzwigaj, S., Pezerat, H. and Fubini, B. (1995)Study of genotoxicity in Chinese hamster V79 cells andmorphological transformation of Syrian hamster embryocells treated with physico-chemically different types ofsilica. In Proceedings of the International Seminar onAssessment of Carcinogenic Risk from OccupationalExposure to Inorganic Substances, October 1995, Luxem-bourg, A8.

Elias, Z., Poirot, O., Daniere, M. C , Terzetti, F., Marande,A. M., Dzwigaj, S., Pezerat, H. Zanetti, G. and Fubini,B. Comparative study of cytotoxicity and morphologicaltransformation of Syrian hamster embryo cells inducedby silica with different surface properties. Submitted toCarcinogenesis.

Engelbrecht, F. M., Yoganathan, M., King, E. J. and Nagel-schmidt, G. (1958) Fibrosis and collagen in rat's lungproduced by etched and unetched free silica dusts. Arch.Ind. Health 17,287-294.

Fubini, B., Giamello, E. and Volante, M. (1989) The possiblerole of surface oxygen species in quartz pathogenicity.Inorg. Chim. Ada 163, 187-189.

Fubini, B., Giamello, E., Pugliese, L. and Volante, M. (1989)Mechanically induced defects in quartz and their impacton pathogenicity. Solid. State Ionics 32-33, 334-343.

Fubini, B., Giamello, E., Volante, M. and Bolis, V. (1990)Chemical functionalities at the silica surface determiningits reactivity when inhaled: formation and reactivity ofsurface radicals. Toxicol. Ind. Health 6, 571-594.

Fubini. B. (1994) Which surface functionalities are impliedin dust toxicity? In: Cellular and Molecular Effects of Min-eral and Synthetic Dusts and Fibres. (Edited by Davis. J.M. G. and Jaurand. M. C) , NATO ASI Series, Vol..H85,pp. 347-358. Springer-Verlag, New-York.

Fubini. B.. Bolis. V.. Cavenago, A. and Volante, M. (1995)Physico.chemical properties of crystalline silica dusts andtheir possible implication in various biological responses.Scand. J. Work Environ Health 21(1), 9-15.

Fubini. B.. Mollo. L. and Giamello. E. (1995) Free radicalgeneration at the solid/liquid interface in iron containingminerals. Free Rad. Res. 23, 593-614.

Fubini. B. (1997) Surface reactivity in the pathogenicresponse to particulates. Environmental Health Per-spectives 105(5), 1013-1020.

Fubini. B. (1998) Health effect of silica in The surface proper-ties of silicas (Edited by Legrand J. P.). pp. 415-464. JWiley and Sons.

Fubini. B..Aust. A. E.. Bolton. R. E.. Borm. P. J. A.. Bruch.J.. Ciapetti, G.. Donaldson. K.. Elias. Z.. Gold. J.. Jaur-and. M. C . Kane. A. B.. Lison. D. and Muhle. H. (1998)Non-animal (Alternative) Tests For Evaluating The Tox-icity Of Solid Xenobiotics ATLA. in press.

Fubini. B.. Zanetti, G.. Altilia, S.. Tiozzo. R.. Lison. D. andSaffiotti U. Cytotoxicity of crystalline silica: cell responseto variously heated cristobalite. Chem Res Tox.. submit-ted.

Giamello. E.. Fubini. B.. Volante. M. and Costa. D. (1990)

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from

Surface chemistry and quartz hazard 529

Surface oxygen radicals originating via redox reactionsduring the mechanical activation of crystalline SiO2 inhydrogen peroxide. Colloids and surfaces 45, 155-165.

Gilmour. P. S.. Beswick. P. H.. Brown. D. M. and Donald-son, K. (1995) Detection of surface free radical activity ofrespirable industrial fibres using supercoiled X 174RFIplasmid DNA. Carcinogenesis 16, 2973-2979.

Goethe. C. J., Lidstrom. L. and Swensson. A. (1971) Influ-ence of mode of disintegration on the fibrogenelic powerof quartz particles. Me<l. Lav. 62, 375-377.

Guthrie. G. D. (1995) Mineralogical factors affect the bio-logical activity of crystalline silica. Appl. Occup. Environ.Hyg. 10, 1126-1131.

Hart. G. A. and Hersteberg. T. W. (1998) In vitro toxicityof respirable-size particles of diatomaceous earth and crys-talline silica compared with asbestos and titanium dioxide.J. Occup. Envir. Med. 40, 29^2.

Hemenway. D.. Absher, M.. Fubini. B. and Bolis. V. (1993)What is the relationship between hemolytic potential andfibrogenicity of mineral dusts? Arch. Environ. Health 48,343-347.

Hemenway, D. R., Absher, M., Landesman, M.. Trombley,L. and Emerson, R. J. (1986) Differential lung responsefollowing silicon dioxide polymorph aerosol exposure. In:Silica, Silicosis and Cancer. (Edited by Goldsmith, D. F..Win D. M., and Shy, C. M.), pp. 105-116. PraegerPublisher, New-York.

Hemenway, D. R., Absher, M. P.. Fubini. B.. Trombley, L.,Vacek, P.. Volante, M. and Cavenago. A. (1994) Surfacefunctionalities are related to biological response and trans-port of crystalline silica. Inhaled Particles VII. Ann. Occup.Hyg. 38, 447^54.

Hesterberg, T. W., Oshimura, M.. Brody, A. R. and Barrett,J. C. (1986) Asbestos and silica induce morphologicaltransformation of mammalian cells in culture: a possiblemechanism. In: Silica, Silicosis and Cancer. (Edited byGoldsmith, D. F., Winn, D. M. and Shy, C. M.), Contro-versy in Occupational Medicine, pp. 177-190. Praeger,New-York.

Huaux, F. (1998) Contribution of macrophage-derived cyto-kines in the pathogenesis of lung fibrosis induced byinhaled mineral particles. Thesis. Universite Catholique deLouvain. Belgium.

Her, R. K. (1979) The chemistry of silica. John Wiley & Sons.New York.

International Agency for Research on Cancer. World HealthOrganization (1997) IARC Monographs on the Evalu-ation of Carcinogenic Risks to Humans. Silica, some sili-cates, coal dust and para-aramid fibrils. Vol. 68. IARCed., Lyon. France.

Klockars. M.. Hedenborg, M. and Vanhala. E. (1990) Effectof two particle surface-modifying agents, polyvinyl-pyridine-N-oxide and carboxymethilcellulose, on thequartz and asbestos mineral fiber-induced production ofreactive metabolites by human polymorphonuclear leu-kocytes. Archil. Envir. Health 45, 8-14.

Kriegseis, W., Scharmann, A. and Serafin. J. (1987) Inves-tigations of surface properties of silica dusts with regardto their cytotoxicity. Ann. occup. Hyg. 31, 417-427.

Krombach. F., Munzing Allmeling, A.-M.. Gerlach. J. T..Behr, J. and Dorger. M. (1997) Cell size and alveolarmacrophages: an interspecies comparison. Envir HealthPersp. 105(5), 1261-1263.

Le Bouffant. L., Daniel, H. and Martin. J. C. (1977) Thetherapeuthic action of aluminium compounds on thedevelopment of experimental lesions produced by purequartz or mixed dust. Inhaled Particles IV, 389^00.

Lin, X.. Keane, M. J., Zhong, B. Z.. Ong, T. and Wallace.W. E. (1996) Micronucleus formation in V79 cells treatedwith respirable silica dispersed in medium and in simulatedpulmonary surfactant. Mutat. Res. 361, 89-94.

Mao, Y., Daniel, L. N., Knapton, A. D., Shi, X. and Saffiotti.U. (1995) Protective effects of silanol group binding agents

on quartz toxicity to rat lung alveolar cells. Appl. Occup.Environ. Hyg. 10, 1132-1137.

Miles, P. R., Bowman, L.. Jones, W. G., Berry, D. S. andVallyathan, V. (1994) Changes in alveolar lavage materialsand lung microsomal xenobiotic metabolism followingexposures to HCl-washed or unwashed crystalline silica.Toxicol. and Appl. Pharmacol. 129, 235-242.

Nash. T.. Allison. A. C. and Harington, J. S. (1966) Physico-chemical properties of silica in relation to its toxicity.Nature 210, 259-261.

Nolan, R. P., Langer, A. M.. Harington, J. S., Oster, G. andSelikoff, I. J. (1981) Quartz haemolysis as related to itssurface functionalities. Environ. Res. 26, 503-520.

Pandurangi, R. S.. Seehra, M. S., Razzaboni, B. L. andBolsaitis, P. (1990) Surface and bulk infrared modes ofcrystalline and amorphous silica particles: a study of therelation of surface structure to cytotoxicity of respirablesilica. Environ. Health Perspect. 86, 327-336.

Raask, E. and Schilling, C. J. (1980) Research findings onthe toxicity od quartz particles relevant to pulverized fuelash. Ann occup. Hyg. 23, 147-157.

Saffiotti, U. and Ahmed, N. (1995) Neoplastic trans-formation by quartz in the BALB/3T3/A31-1-1 cell lineand the effects of associated minerals. Teralogen. Carcino-gen. Mutagen. 15, 339-356.

Saffiotti, U., Daniel, L. N., Mao, Y, Olufemi Williams, A.,Edward Kaighn, M., Ahmed, N. and Knapton, A.D.(1993) Biological studies on the carcinogenic mechanismsof quartz In: Health Effects of Mineral Dusts, (Editedby Guthrie, G. D. Jr, and Mossman, B. T.) Reviews inMineralogy, Vol. 28, pp. 523-544. Mineralogical Societyof America Washington D.C.

Shi, X., Nao, Y., Daniel, L. N., Saffiotti, U., Dalai, N. S.and Vallyathan. V. (1995) Generation of reactive oxygenspecies by quartz particles and its implication for cellulardamage. Appl. Occup. Environ. Hyg. 10, 1138-1144.

Vallyathan, V., Shi, X., Dalai, N. S., Irr, W. and Castranova,V. (1988) Generation of free radicals from freshly frac-tured silica dust: potential role in acute silica-induced lunginjury. Am. Rev. Respir. Dis. 138, 1213-1219.

Vallyathan, V., Mega. J. F., Shi. X. and Dalai, N. S. (1992)Enhanced generation of free radicals from phagocytesinduced by mineral dusts. Am. J. Respir. Cell Mol. Biol. 6,404-413.

Vallyathan. V., Castranova, V., Pack, D., Leonard, S., Shu-maker, J., Hubbs, A. F.. Shoemaker, D. A., Ramsay, D.M.. Pretty, J. R.. McLaurin, J. L., Khan, A. and Teass.A. (1995) Freshly fractured quartz inhalation leads toenhanced lung injury and inflammation in rats. Am J.Respir. Crit. Care Med. 152, 1003-1009.

Volante. M., Giamello, E.. Merlo. E., Mollo. L. and andFubini, B. (1994) Enhanced surface reactivity on mech-anically activated covalent solids and its relationship withthe toxicity of freshly ground dusts. In: An EPR studyProceedings of the First International Conference on Mech-anochemistry, (edited by K.Tkachova), pp. 125-130. Cam-bridge Intersci Pub.

Wallace, W. E., Vallyathan, V., Keane, M. J. and Robinson,V. (1985) In vitro biologic toxicity of native and surface-modified silica and kaolin../. Toxicol. Env. Health 16,415-424.

Wiessner, J. H.. Henderson. J. D , Jr., Sohnle, P. G., Mandel.S. N. and Mandel, G. S. (1988a) The effect of crystalstructure on mouse lung inflammation and fibrosis. Am.Rev. Respir. Dis. 138, 445-450.

Wiessner. J. H., Mandel, S. N., Sohnle, P. G. and Mandel,G. S. (1988b) Effect of particle size on quartz-inducedhemolysis and on lung inflammation and fibrosis. Exp.Lung Res. 15,801-812.

Wiessner, J. H., Mandel, N. S., Sohnle, P. G., Hasegawa, A.and Mandel. G. S. (1990) The effect of chemical modi-fication of quartz surfaces on particulate-induced pul-monary inflammation and fibrosis in the mouse. Am. Rev.Respir. Dis. 141, 111-116.

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from

530 B. Fubini

Yamamo, Y., Kagawa, J., Hanaoka, T., Takahashi, T., Zhong, B. Z., Ong, T. and Whong, W. Z. (1997) Studies onKasai. H.. Tsugane. S. and Watanabe, S. (1995) Oxidative the relationship between treatment condition and micron-DNA damage induced by silica in vivo. Environ. Res. 69, ucleus induction in V79 cells exposed to silica and glass102-107. fibers. Mulai. Res. 391, 111-116.

by guest on June 9, 2016http://annhyg.oxfordjournals.org/

Dow

nloaded from