Document VGaO8amGrpM9OmoBN09BJEKY4

1 DRUG AND CHEMICAL TOXICOLOGY, 10(142), 49-86 (1987) CYTOTOXICITY OF RESPIRABLE DUSTS FROM INDUSTRIAL MINERALS: COMPARISON OF TWO NATURALLY OCCURRING AND TWO MAN-MADE SILICATES DENIS NADEAU1, LOUISE FOUQUETTE-COUTURE1, DANIEL PARADIS1, JALEH KHORAMI2, DENIS LANE1 and JACQUES DUNNIGAN3. laboratoire de Caractgrisatidr. de l'Amiante, Departements de biologie1 et 2 de chirnie , Faculty des sciences. University de Sherbrooke, and L'Institut de Varniante, Division de la recherche , Pavilion Marie-Victorin, Uni versits de Sherbrooke, Sherbrooke, Quebec, CANADA OIK 2R1 INTRODUCTION Foreword. Because no single material has ever been found which could replace chrysotile asbestos in its numerous diversified technological applications, many different materials have been proposed or are currently used extensively as substitutes. For technological reasons, the majority of the substitute materials are intended for those applications where cshorts chrysotile fibers are traditionally used, i.e. in situations where insulating properties or ainert* inorganic filler materials are desired. fi 50 NADFAU F.T AL,, proposed for asbestos substitution, numerous man-made materials have on the other hand been selected (e.g. ((mineral woolsn). Related to this is the fact that comparatively few studies have been carried out that could provide indications as to the relevant merits of these newly introduced materials from a biological and health stand point. In view of the irrefutable evidences on the hazards to human health after exposure to ((respirable# particles (for examples, see: ((Biological effects of mineral fibers)), 1980), many ip vitro assays have been developed and used to study the response of target cells to mineral dusts (for examples, see: The in vitro effects of mineral dusts, 1980). Therefore, because naturally occurring and man-made fibrous materials are often proposed as allegedly safe or safer substitutes for chrysotile asbestos, this fact prompted us to evaluate the relative in vitro cytotoxic response of rat. pulmonary alveolar macrophages toward four industrial silicates. In order to establish a comparison between two in vitro assays, the membranolytic profiles of the mineral dusts were also evaluated. Description of the Silicates Tested. 1. Naturally Occurring Silicates. Chrysotile asbestos is the well known hydrated magnesium silicate which has been in use for many decades. In order to comply with its different commercial uses, several grades of this natural silicate have been fashioned by the mining industry. For example, the so-called ((Quebec Standard)) (QS) grade 3 fibers are very long and are used primarily in the manufacture of different types of yarns and rovings. In turn, these products are used jn the manufacture of various textiles, such as protective gloves, clothings and fire blankets, woven-type clutch facings and gaskets, and electrical insulating tapes. Of increasingly shorter length, QS grades 4 and 5 fibers are typically used as reinforcing A ,i .:ij| m ' - 4 l * fc. * i 0 DEAIJ RT AL. have on the s have been meri ts of tand point, ealth after cal effects veloped and >r examples, re, because proposed as this fact s> e of rat -fcates. In ssays, the wel1 known iecades. In ?.des of this For example, long and are and rovings. js textiles, type clutch increasingly reinforcing CYTOTOXICITY OF RESPIRABLE DirSTS agents in the manufacture of fibre-cement products (pipes, sheets, roofing tiles, etc.). The shorter# fibers representative of QS grades 6 to 9 are used primarily as fibrous inorganic filler in manufactured products such as thermoplastics compounds, flooring tiles, jointing cements, and asphalt roofing papers. As found in these commercial grades, the dimensions (length and diameter) of the chrysotile fibrers are generally far out of the icrespirables range and therefore, are not readily suitable for biological experimentation. For this reason, tirespirabies or biologically pertinent fractions must be prepared from these commercial grades. The most well known are the UICC preparations, which have been obtained by various comminution and extraction methods applied to a mixture of different grades and sources (Timbrell andRendall, 1971-1972). While these particular preparations represented a valuable effort to provide the investigators with standardized biologically pertinent asbestos samples, they may not be entirely representative of the actual crespirables fibers which may be emitted from real life situations. Consequently, milder extraction procedures must then be devised to obtain unaltered fiber fractions danger et al., 1978). Attapulgite, a naturally occurring magnesium aluminium silicate, was named from Attapulgus, Georgia, USA, the source of the first sample studied in the mid-thirties. Attapulgite clay derives its non-swelling, needlelike morphology from'its three-dimensional structure, and the shape and size of these needles result in remarkable properties such as a very high surface area (Haden, 1963). The commercial availability of attapulgite in large tonnage and at relatively low costs, has made it an important, source of raw rate-rial for a variety of industries. 52 NADEAU KT AL. Originally developed for the petroleum industry (drilling muds), attapulgite finds its current largest uses in many unrelated applications such as a carrier, a diluent and a conditionner for agriculture chemicals. It is also used in adhesives, sealants, mastics, paints, pharmaceuticals, and many other applications such as animal bedding, polishes, wax emulsions, laundry washing powders and tape joint cements. Recently, it was even proposed as a component of ar. asbestos-free friction material. Contrary to many commercial grades of chrysotile, the dimensions of the American attapulgite fibers commercially available are already in the respirable range. For this reason, no further communiticn and extraction are required for biological experimentation. Man-Made Silicates. The term man-made mineral fibers (MMHF) embraces a variety of fibrous inorganic materials typically obtained from amorphous molten slags, certain types of rocks, and various glassy composites. Inevitably, the elemental composition vary from one type to another. But, in general, these fibers are either mainly aluminium or calcium silicates; minor components sbch as iron, boron or titanium oxides are also found associated with these silicates. The KMMF are known under a variety of different names such as ((ceramic fibersn, ((processed mineral fibers#, vkaowooli), ((fibrous ceramic aluminium silciate glass)), rock wool# ((slag wool)), ((mineral wool)), xglass fibers#, and so on. In the present study, we selected a commercial grade of an aluminium silicate material: Fiberfrax. This type of silicate is proposed as a substitute fcr chrysotile asbestos in applications where thyxotropic properties are desired; it is also used as an additive to epoxies, plastic mixes, caulking compounds, fine, paper production, and injection molding compounds. In general, mineral and glass wools are formed by blowing a thin stream of che molten mass by means of high pressure air or steam jets. NADEAU ET AI.. rilling muds), d applications ture chemicals, harmaceuticals, polishes, wax Recently, it ,tion material, ansions of the already in the and extraction .Jir) embraces a 'ram amorphous -i'j composites, another. But, cium silicates; are also found r a variety of ineral fibersn, .ck wool)) slag esent study, we i I; Fiberfrax. 'sotile asbestos it is also used nds, fine paper ig a thin stream jr steam jets. CYTOTOXICITY OF RESPIRABLE DUSTS S3 Centrifugal spinning is also used, where the melt is allowed to fall onto a high speed rotating disk. As a consequence, the diameters of these MFWF vary considerably. For example, slag wools may countnin up to 60% of fibers with a diameter between 0-3 um, while rock wools and glass wools may respectively contain 36% and 26% of such small diameter fibers. For this reason, extraction methods must also be used to obtain fibrous fractions biologically pertinent. In contrast to the above-mentioned MMMF, a relatively new material has been introduced on the market for substitution of asbestos in temperature insulation products. In carefully controlled pressure and temperature conditions, the addition of CaO to SiOj will cause the production of fine calcium silicate fibers known as xorotlite (Funk, 1965). The dimensions of the fibers, being essentially 100" in the ttresplrable ranges, allow the use of this material as such in biological experimentations. MATERIALS AND METHODS Source of the Fibrous Silicates. The separation of very short chrysotile fibers follows the method outlined by Jolicoeur et al. (1981). Briefly, 60 g of Johns-Manville grade 4T30 chrysotile fibers, slightly opened by two passes through a mechanical conditionner (B.O.P.), were mixed with 6 L of demineralized water, and Inverted 20 times. After sedimentation, the fine suspension in the supernatant was concentrated by filtration, and the remaining water was removed by a freeze-drying process. The complete procedure will be outlined elsewhere (Kimnerle, Roberge and Nadeau, unpublished results). The aluminium -silicate sample used in this study is of the HSA (High Specific Area) grade of Fiberfrax , obtained from The Carborundum Company, d. 54 NADEAU ET AL. `Niagara Falls, NY, USA. A ashortu fraction was extracted as follows: 20 g of the material was suspended in 7.5 L of demineralized water and inverted 10 times. After 24 h of sedimentation, 5.5 L of the supernatant were aspirated, care being taken not to remove the material floating at the surface. The colloidal suspension was then filtered on 0.8p filters (Klllipore Ltd., Mississauga, Ontario, Canada), and dried to a cake under an infrared lamp. A typical extraction yields 120-130 mg/L of solid particles. The clay mineral attapulgite was obtained from the Quincy deposits (Florida, USA) and the calcium silicate xonotlite was provided by a European supplier. The American attapulgite and the calcium silicate dusts were tested as supplied. Physico-Chemical Characterization of the Silicates. 1. Surface Area and Zeta Potential. Specific surface area measurements were determined by measuring the quantity of nitrogen adsorbed, using the BET theory (Brunaueur et al., 1938). Before the measurements the samples were conditioned overnight at 110C, under an helium gas stream (Quantasorb, model QS-10; Quantachrome Corp. Syosset, NY). Zeta potentials were determined in 1 mM KC1 (pH7.0) on a Zeta Reader Komiine-Sanderson (model ZR11, Noram, Pointe-Claire, Quebec, Canada). 2. Size Distribution. For the length and diameter distributions, the following procedure was used: 0.5 mg of each fiber sample was resuspended (25 full strokes of a glass Dounce homogenizer; Dunnigan et al., 1980) in 2 ml of saline (0.19J, P/V) containing 25 ug/ml of Tween 80 (Sigma Chemical Co., St. Louis, MO). After serial dilutions, the final concentration of Tween 80"was-reduced to 1 pg/ml and, before the filtration step, the fiber \ NADEAU ET At. follows: 20 g er and inverted ipematant were loating at the i 0.8y filters to a cate under mg/L of solid Quincy deposits provided by a silicate dusts ea measurements jrbed, using the nts the samples ium gas stream Zeta potentials .otnline-Sanderson stributions, the was resuspended et al., 1980) in 1 (Sigma Chemical concentration of . step, the fiber ! CYTOTOXICITY OF RESPIRABLE DUSTS 55 suspensions were adjusted to yield on the membrane 0.25 yg/cm^ for chrysotile, 1.0 pg/cm for attapulgile and xonotlite, and 3.0 pg/cm'- for Fiberfrax. For the transmission electron microscopy (TEH) study, the carbon-coated Nuclepore technique was used (Anderson and Long, 1976; Chatfield et al., 1978), Fifty ml of each suspension were filtered cither on a 0.4 y or on a 0.1 y Nuclepore membrane (47 mm; Maynard Scientific, Ste-Foy, Quebec), with a usable surface of ~ 10 cm . Sections of each O membrane (-10 mm ) were selected arbitrarily for the TEM study. Basically, after being deposited on 200-mesh copper grids (J.B. EM Services, Pointe-Claire, Qufibec), the membranes were first coated with carbon, then solubilized with chloroform. Representative pictures of each sample were taken or. a Philips 200 TEM, and the length and diameter distributions were determined by computer-assisted image analysis (Bioquant, R & M Biometrics, Inc., Nashville, TN). According to the current convention, any particle presenting an aspect ratio (length/diameter) loss than three to one was not counted as a fiber. tn Vitro Bioloqical Assays. I. Hemolysis. The hemolytic activities of the four silicates were evaluated as follows: red blood cells (RBC) were collected from whole blood drawn from the vena cava of the rats used for the isolation of the macrophages. The heparinized RBC were washed three times in a modified (330 mOsra instead of 290 mOsm) veronal buffer (Harlngton et al., 1971) and then resuspended at 4% (V/V) in the same buffer (Dunnigan et al., 1980; Pele et al., 1983). With a Oounce homogenizer, the fibers were resuspended in the veronal buffer (supplemented with 2 yg/ml of Tween 80) at a concentration twice the final concentration desired. For each.assay, 1 ml of RBC was mixed with 1 ml of fibers in a siliconized scintillation glass vial capped with a gas permeable membrane I 1 :: i .-1 |' 1: * :i 56 NADEAU ET AL. closu-re (P.H. Cap, type 20; Kimble/Owens-lllinois, Toledo, OH). Blank assays (no fibers) were also included to evaluate the non-specific hemolysis due to the mechanical stresses of the procedure (usually 1-2?). The vials were then incubated at 37C in a reciprocal water bath set at 96 strokes/min. At any given time, the assays were stopped by the addition of 0.2 ml of 13 (V/V) glutaraldehyde. Following a centrifugation at 500 XG/10 min, the clear supernatants were immediately diluted 1;4(V:V) with the veronal buffer, and the absorbance (0.0.) values were determined at 541 nm. The percentage of release of hemoglobin was evaluated by comparison with a fully lysed sample (1003 release), that is by sonicating for 20 sec (Heat Systems Sonicator Cell Disruptor, model W-220F set at level one; Bionetic Ltee/Ltd., St-Laurent, Quebec) a blank assay before processing it as above. The percentage of hemolysis was calculated as follows: % Hemolysis 0.D541 sample - 0,0541 blank 0.0. lysed sample 541 X 100 (4.1) 2- Cytotoxicity. The target cells were obtained from 200-250g Long-Evans black hooded rats [Onale Crl:(LE)BR] ; Charles River Canada Inc., St-Constant, Qu&bec), delivered in filtered cages, and immediately housed under a laminar flow enclosure (BIOCLEAN PC 580, Hazel ton Systems, Aberdeen, HO). The animals v/ere fed autoclaved rat chow, had access to water ad_libitum, and were maintained under an alternating period of artificial light and darkness (LD 1Z:12), with light from 07.00 to 19.00 h. The rats were kept as such until the day of the experiment, which was usually within one or two weeks from their shipment. The day before the experiment, the rats were fasted overnight. / Nadeau et al. !o, OH). Blank he ner-specific (usually 1-2X). r bath set at 96 the addition of ntrifugation at ii luted 1:4(V:V) were detemined as evaluated by is by sonicating 1 W-220F set at ;rk assay before r.s calculated as X 100 (4.1) i0-250g Long-Evans ar Canada Inc., rmnediately housed iazeiton Systems, .iw, had access to nating period of 07.00 to 19.00 h. riment, which was he day before the OTOTOXICITY OF RESPIRABLE DUSTS 57 Isolation of pulmonary alveolar macrophages (PAH). The animals were sacrificed with an overdose of sodium pentobarbital (250 mg/kg, i.p.), exsanguinated via the vena cava, and cannulated at the upper trachea level. The bronchoalveolar lavages (BAL), were carried out in situ (4 X 10 ml) with a modified Hanks' balanced salt solution (BSS) developed for that purpose. The composition of the (Ca.Mg)-free BSS is given in Table l. Then, the BAL were centrifuged at 240 XG/10 min. The individual cell pellets were submitted to a brief osmotic shock to eliminate contaminating RBC. After pooling, the lung free cells (LFC) were washed twice with the BSS solution. The BAL usually yield -10^ LFC/rat, of which ^95% are viable PAH (Trypan blue dye exclusion test). For the establishment of the PAM monolayers, two protocols were used. O Fresh monolayers. In Linbro multi-well plate (24 wells, -2cm /well; Flow Laboratories, Inc., McLean, VA), the LFC were seeded at 4 X 105 cells/well and incubated for 2 h at 37C in water-saturated atmospheric air. The composition of our incubation medium is shown in Table 2. After this pre-incubation period, the cell monolayers were rinced with fresh medium and then challenged for 18 h with the mineral fibers. One da.v-old cultured monolayers. The cell monolayers were established as described above, with the exception that before being challenged with the fibrous silicates, PAH were allowed to incubate further for 24 h following the 2 h pre-incubation period. As with fresh PAH, cultured cells were rinced with fresh medium and then exposed to the mineral fibers for 18 h. Each fibrous material was autoclaved for 10 min at 121C prior to its dispersion. The silicates were then resuspended in a saline-Tween BO solution (0.19% and 25 ug/ml, respectively) with 25 full strokes of a Dounce homogenize? in order to have dispersions of the testing materials -wf 58 NADEAU ET AL. TABLE 1. Composition of the (Ca,Mg)-Free BSS Developed for the BAL Components Concentrations KCla KaCla NaHC03a KH2P04a Na2HP04.7H?0a Na?S04a Hepes (Ultrol)k Glucose3 Penicillin (base)c Streptomycin (base)c Fungizonec Tetracycline (HCL)d Tylocine0 0.40 mg/ml 8.12 mg/ml 0.35 mg/ml 0.06 mg/ml 0.09 mg/ml 0.07 mg/ml 2.38 mg/ml 1.00 mg/ml 100 U/ml 100 pg/ml 250 ng/ml 5 pg/ml 60 pg/ml (a) Fisher Scientific Co., Fair Lawn, NO. (b) Calblochem-Behring, San Diego, CA. (c) GIBCO Canada Ltd., Burlington, Ontario. (d) Sigma Chemical Co. Conditions (37C) gH: adjusted to 6.80 + 0. 05 Osmolality: 298 t 2 mOsm similar to those used for the size distribution study. As with the hemolysis assay, the final concentration of Tween 80 in the cytotoxicity assay was 1 pg/ml. Two doses, 50 pg (33.3 pg/ml; 25 pg/cm ) and 250 pg o (166.7 pg/ml; 125 pg/cm ) were selected for the evaluation of the cytotoxic effects of the silicates on PAH. * ^- :adeau f.t al. or the BAL s (37C) to 6.80 0.05 '98 ?. mOsm As wi th the cytotoxicity ) and 250 ug the cytotoxic CYTOTOXICITY OF RESPIRABLE DUSTS 59 TABLE 2. Composition of the Incubation Medium Developed for the PAM Cytotoxicity Assay Components Concentrations KCla NaCla CaCl?.2H70a HgCl2.6H20d MgS04.7H20a NaHC0,a J KH2P04a Na2HP04./H20a Hepes (Ultrol)^ Fetal bovine serum0 (heat-inactivated) Glucose4 Amino acidsc Non-essential amino iacids0 Vitamins'" L-Glutaminc Penicillin (base)0 Streptomycin (base)0 Fungizone0 Tetracycline {HCL)^ 0.40 mg/ml 6.56 mg/ml 0.74 mg/ml 0.51 mg/ml 0.62 mg/ml 0.35 mg/ml 0.06 mg/ml 0.09 mg/ml 2.38 mg/ml 4? (V/V) 1.00 mg/ml MEM (IX) MEM (IX) MEM (IX) 292 ug/ml 50 U/ml 50 p g/ml 125 ng/ml 5 pg/ml (a) Fisher Scientific Co. (b) Cal b i ocher.-Behri ng. (c) GIBCO Canada Ltd. Conditions (37C) H: adjusted to 6.80 0.05 Osmolality: 300 t 5 mOsm Incubation: water-saturated air 60 3ioch.emical Analyses. NADEAU ET AL. For each type of fiber and every dose studied, duplicate cell incubations were made; incubation blanks (no fiber) were also included. After being centrifuged at Z40 XG/IO mir. to obtain cell-free supernatants, the medium of one of the duplicate incubation was used for the determination of the extracellular enzyme leakages and the production of lactic acid. For the quantification of ATP, the very few cells recovered from the previous centrifugation were combined to their corresponding cell monolayer. The counterpart of the duplicate cell incubations was treated with Triton X-100 (final concentration: 0.1%), then sonicated briefly for 15 sec (Heat Systems Sonicator Cell disruptor, model W-10 set at level one; Bionetics Ltfie/Ltd.) to determine the total enzymatic activities (cells + medium) and DNA contents. 1. Extracellular Enzyme Leakages. cytotoxicity were measured as follows: The two enzymatic markers of LPH. L-lactate dehydrogenase (EC 1.1.1.27) was determined by a modification of the procedure of Hacker et al. (1960): NAD (10 mM) and lactic acid (70 mM) in TAPS buffer (85 mM), pH 8.5 (Beckman Canada Inc., Montreal, Quebec). The assay is based on the rate of formation of NADH at 340 nm, and the activity is measured as milli- International Units (mIU)/ml. 3-GAL. The 6-galactosidase (EC 3.2.1.23) assay is based on the procedures of Levy and HcAllan (1963) and Christomanou et al. (1977): p-nitrophenyi- -D-galactoside (4 mM) in citrate buffer (50 mM), pH 4.4. The reaction is stopped by the addition of 0.5 M glycine buffer (pH 10.5), and the activity is measured as nmoles of p-nitrophenyl released/min/ml. NADEAU ET AL. duplicate cell also included, -o supernatants, used for the e production of cells recovered responding cell ons was treated ted briefly for t at level one; tic activities ic markers of .e mined by a NAD (10 mM) and ian Canada Inc., tion of NADH at rnational Units n the procedures p-nitrophenylThe reaction is and the activity CYTOTOXICITY OF RESPIRABI.F. DUSTS 61 2. Production of Lactic Acid. Lactate synthesis by PAM was evaluated by measuring the amount of the metabolite released into the incubation medium (Paradis and Nadeau, 1982). 3. Cellular ATP. PAM (cell pellet + monolayer) were successively treated with 1 ml of dimethyl sulfoxide (DMSO; Sigma Chemical Co.) and 3 ml of demineralized water to release the nucleotides. Following the extraction, the diluted DMSO extracts were immediately frozen in liquid nitrogen and kept at -60C. The ATP was measured later by a bioluminescence assay, using the LKB Luciferin-Luciferase Monitoring Reagent (Fisher Scientific Co.). Before being assayed, freshly thawed samples were properly diluted with an Hepes buffer (25 mM, pH 7.75) supplemented with MgSO^ (10 mM) and NaN3 (0.02", P/V). For the assay, 150-pl aliquots were mixed with 40 pi of ATP reagent, and the bioluminescence was recorded for 10 sec at room temperature with the integral mode of the luminometer (Berthold Biolumat LB 9500, Labserco Ltd., Oakville, Ontario). An internal standard of ATP (10 pi) was added to evaluate the presence of quenching factors (Turner Designs, 1981). The ATP concentration was evaluated as follows: ATP (ng/ml) Volume STD Volume S LJ BLU x Concentration STD BLS - BLU -- "J (4.2) where : STD S BLU BLS ........... .. OIL internal standard of ATP sample recorded bioluminescence of the sample recorded bioluminescence of the sample after the addition of the internal standard ditution of the sample On average, 10 PAM contain -1500 ng of ATP. <y. 62 NADEAU ET AL. 4. .Cellular DNA. DMA contents were measured with a fluorescence enhancement technique, by modifying the method of Adams and Storrie (1981); their procedure is based on the lysis of cells with Triton X-100 and the use of the Hoechst dye 33258 (Calbiochem, La Jolla, CA). From a stock solution of the bisbenzimide fluorochrome (200 ug/ml) kept refrigerated in the dark, a working solution is made freshly the day of the experiment. The assay buffer (100 mi! NaCI, 10 mM EDTA, 10 mM Trfs, pH 7.0; Brunk et al., 1979) was set at a final dye concentration of 100 ng/ml, and kept protected from the light at all time. The DNA standard stock solution, made with highly polymerised calf thymus UNA (Sigma Chemical Co.), was adjusted spectrophotometrically (O.D. at 260 nm for a 50 ug/ml DNA solution = 1.0) and stored at 4C. Daily, a standard curve (0-500 ng) was made by mixing one volume of the DNA standard stock solution with two volumes of a concentrate (1.6X) of the culture medium and 0.2 volume of Triton X-100 1.6% (P/V). As with the cell monolayers, each point of the standard curve was sonicated briefly for 15 sec (Brunk et al., 1979). For the DNA determinations, 100-pl aliquots (triplicate) of the sonicated DNA standards or samples were mixed With 3 ml of the assay buffer. The fluorescence intensity was measured on a Perkin-Elmer Luminescence Spectrometer (model LS-5; Perkin-Elmer Canada Ltd., Montreal, Quebec), with the excitation and emission wavelengths respectively set at 350 and 455 nm. As recommended by Downs and Nilfinger (1983), 10 nm slit widths were used. Although the DNA determinations were usually done on freshly sonicated cell samples, it was found that the measurements could be postponed up to 48 h later if EDTA (100 mM, final concentration) was added to the sonicated samples to Inhibit DNAses (Junowicz and Spencer, 1973). Moreover? -sifi.ee we noted an interference cf the attapulgite fibers on the NADEAU ET AL, a fluorescence id Storrie (1981); Lon X-100 and the ,). From a stock it refrigerated in f the experiment. ., pH 7.0; Brunk 0 ng/ml, and kept polymerised calf ometrically (O.D. at 4C. Daily, a the ONA standard Jof the culture ,s with the cell ed briefly for 15 iplicats) of the -- ,i trtl of the assay j i a Perkin-Elmer a Ltd., Montreal, spectively set at 1983), 10 nm slit usually done on .urements could be tration) was added id Spencer, 1973). ite fibers on the CYTOTOXICITY OF RESPIRABLE DUSTS 63 intensity of the fluorescence reaction, probably by adsorption of the fluorochrome, it might be necessary to spin down some cellular extracts before taking the 100-pl aliquots for the DNA determinations. On average, 10 PAH contain -10 ug of DNA. Statistical analyses. The experimental data where evaluated first with a one-way analysis of variance (Tallarida and Murray, 1981). Then, when the F-valuas were significant, the fiber-exposed cell monolayers were compared to their respective control incubations by performing a Dunnett's multiple comparison test (Tallarida and Hurray, 1981). This test was selected for its ability to compare several treated groups to a single control group (Dunnett, 1955). The level of significance was set at p sQ.05 and, for each parameter evaluated, no statistical difference was ever observed between the two control groups (one for each fiber concentration) of either incubation condition (fresh or one day-old cultured cell monolayers). RESULTS Physico-Chemical Properties of the Industrial Silicates. As seen in Fig. 1 to 4, the TEM micrographs show that three of the four silicates are in a close dimensional range, that is chrysotile, attapulgite and xonotlite. Quantitative size analysis (Table 3) revealed that the. average fiber length of these three minerals was -1.0 pm. The diameters were also very similar, with the exception of the xonotlite sample which had an aspect ratio about half the value of the chrysotile and attapulgite fibers. All the short chrysotile fibers had a length $5.0 pm and a diameter ^0V2 pm., The attapulgite fibers were IQOS^l.Opm in length S i :1 | I > I NADEAU ET AL. E . n_! <a CYTOTOXICITY OF RESPIRABLE DUSTS 0) 10 E 3. CO t to 0) V> 4-> Er tlc/s1; ae o J-- X cn . cn f 6B NADEAU ET AL TABLE 3. Physico-Chemical P roperties o f the Fibrous S ilic a te s ns L> a +- c OS +J.-- V) O > V a- e 4-> CO V - 0) 4; a. fsl o ?a_ nj* CO CO CO +1 in an CO CM CM M" + 1I u ai in u a (O <*- u t- o 4- c irt ot r^-. 0 ^ fH U OifM u in cl o OS E fr r o<U 4-> lO CO CM u> CO eo *3" a --* os Cu tn Otc-xo: CEw cn </) Oj *> uo Hi -r0.4J vt in ns 4-* <C Of c= 4) E cs s. 3 si/> a> (O as <U E <. 13 as cn S 3 s. <u > c .c +j --. tn g C 3. aiw- CO CO CO US CM LT> in CO p 1 ro r-- CM CM oOOo m oo CM CO o o --t CO <a a> in +- as ns a. <4 C? > O o- h- co ^ <u os 4- XJ a> Cn 4-> X to 4J u o3 V* Q. H fc. >* (0 o as s~ 4- c xa +J o T-- o <c X Lu O-TS U (ti 4-> c(O Lq;- ns m o -- 4J as h- x+>: o t0-) in 4-* * >> s v* t- xos xor*o l_ u> *o. ns 4- 4-> CU OJ ns XC3 l<-U &V <$- c: o(0 f= S- 3- Q. &- OuJ xaas Vi <u II k. b. idn: a3Cs %a; O) CO r-**0 tw- +5- JhZ- JOQS o cr- ns mt3u *.urc-- --R9 xX: E OJ o x: s- i (c) The values In parantheses represent the properties o f the s ta rtin g m aterial from which each Kshort# fib e r fra c tio n was iso la te d . NADEAU ET At. CYTOTOXICITY OF RESPIRABLE DUSTS 69 and^O.l pm in diameter. For the xonotlite sample, 100% of the fibers had a length $3.0 pm and a diameter -SQ.7Q pm. The Fiherfrax sample was more heterogenous in size and, contrary to the other three silicates which did not contain any fiber longer than 5 pm, only 505 of the aluminium silicate fibers were in this size range (data not shown). The diameters of the Fiberfrax fibers ranged from 0.2 to 1.0 pm, and the mean aspect ratio was -39 (Table 3). As expected, the chrysotile sample has a positive surface charge due to its brucite layer (Table 3). The other silicates have a typical negative charge, the highest value being for the aluminium silicate sample, [n decreasing order of magnitude, the specific surface areas were: attapulgite > xonotlite > chrysotile > Fiberfrax (Table 3). One can also observe that the isolation of dshorta fractions of chrysotile and Fiberfrax caused a two-to fourfold increase in the specific surface area of these two samples. Biological Effects of the Fibrous Silicates. As seen in Table 4, .the chrysotile, attapulgite and xonotlite samples were markedly hemolytic to the rat erythrocytes. Comparatively, the Fiberfrax particles did not show any significant hemolytic activity, even at the highest dose tested {1000 pg/ml). Chrysotile hemolysis was the strongest, followed very closely by xonotlite; attapulgite fibers were consistently slightly less hemolytic than the other two silicate fibers. In view of the different surface charges between these mineral dusts (Table 3), it is interesting to note that both negatively and positively charged minerals are able to induce hemolysis. Similarly, there appears to be no straightforward relationship between the fiber surface areas (Table 3) and the hemolytic activities. Nonetheless, it is well known 70 NADEAU ET AL. C/I \ I 1; i; . i. 4O- S_ a. u 4>-> O c ns nr QJ __I <CCD 1 K (80J Lar)t <*o-> </) c\j r- vo ro CM r-*. CM o +1 -H -H -U co o o o Nco Ur-J- cCoM N O co f->. CM I-- rH sM -H O O O in Csl CM CO U3 r-. : CM H -H +1 cO CO O t-v o> <nnr ua ro vr 41 41 00 onr C-J o cr\ CM CM -H 4C 4< o If) CO CO CCMD ir---*i 0i-> \ nass)- xt>:> ( > (b) The data represent the mean values S.O,t 4 ), a fte r 60 min o f incubation. {c) p sO.05 when comprared to c h ry s o tile (Dunnett's m u ltip le comparison te s t, two-sided, using c h ry s o tile as the control group. HADEAU ET AL. CYTOTOXICITY OF RESPIRABLE DUSTS 71 that .both factors can participate in the membranolytic effect of mineral dusts (Harington et al., 1971; Light and Wei, 1977; Schnitzer and Pundsack, 1970; Morgan et al., 1977; daurand et al., 1979s). The in vitro cytotoxic effects of the silicates are illustrated in Fig. 5 and 6. Using fresh PAM monolayers (Figure S), one can observe that at the highest dose (250 ug), all four mineral dusts were highly cytotoxic. The extracellular releases of the enzymatic markers LDH and S-GAL were respectively -60-70S and -40-50T (Fig. 56 and 5D). The resulting low viability of the cells was confirmed by the 80 decrease in ATP cell contents (Fig. 5F). When exposed to a lower dose of mineral dust (50 pg), only chrysotile, attapulgite and the aluminium silicate fibers presented a significant cytotoxic effect, as judged by the enzyme releases and the ATP cell contents (Fig. 5A, 5C and 50). Nonetheless, the xonotlite fibers can also be considered slightly toxic at this dose, since the ATP contents of the exposed-cell monolayers were significantly reduced by -205?. It is interesting to note that at either dose of the attapulgite, the calcium and the aluminium silicate fibers, the LDH releases were always higher than for the lysosomal enzyme. For the chrysotile fibers, the magnitude of the enzyme releases were roughly, similar. Using a more common experimental protocol, the cytotoxicity of the silicate dusts was then evaluated on one day-old cultured PAM monolayers (Fig. 6). Again, one can observe that all four fibers were cytotoxic at 250 pg (Figures 6B, 6D and 6F). However, it is striking to see that, with older cells, the cytotoxicity of the chrysotile fibers was greatly reduced (Fig. 5 vs. Fig. 6). Moreover, at 50 pg (Fig. 6A, 6C and 6D), the chrysotile dust induced only a aselectiven release of the lysosomal marker of (3-GAL (Davies et al., 1974), while both the attapulgite 8 i w:--t !1 )?+! 1 !> i -4 ia 111 $ ] :t n NADEAU ET AL. RAT PULMONARY MACROPHAGE RESPONSE (FRESH MONOLAYER) 5 80 i t CHRYSOTILE 2 ATTAPULGITE 2 60 3 XONOTLITE o 4 AL-SILSCATE Ui Q -p UJ H (f) > <H 111 o UJ - DC 2 o > ** h-- it >; h- - o< LLl 5 > M Z Ui * SOug of fibres PAU 1 23 ALONE --------* 250 ug - i of fibres P-GAL 70 %3 \` S\ ki m .fl.lLi PAM 1 ALONE ___ 23 + 250 ug of fibres ( J NADEAU ET At. SPONSE 3 50 UQ fibre* 4 3 '50 ug -- fibres 4 CYTOTOXICITY OF RESPIRABLE DUSTS 73 and aluminium silicate fibers were still highly cytotoxic. At this low dose, and in these conditions, the calcium silicate fibers were not very toxic to the cells. These results are in agreement with our earlier observations to the : effect that tissue culture conditions can modulate the in vitro response of j PAM toward mineral or man-made fibers (Nadeau et al., 1983a, Dunnigan et al., 1984). For the evaluation of the viability of the cell after the challenge with the mineral dusts, two parameters were used: the ATP cell content (Fig. 5 E.F and 6 E,F), and the production of lactic acid (Table 5). Although usefull for an estimation of the metabolic | I activity of PAM, one can see that the production of lactate is not as sensitive as ATP, especially with older cells. With a low dose of fibers, slight stimulations of the cell metabolism can even be observed (Table 5). Finally, no significant effect of the silicate fibers on total AON cell contents was noted (data not shown). DISCUSSION The significance of in vitro cytotoxicity data on mfneral dusts cannot be easily extrapolated to their in vivo pathogenic potential. However, several authors have examined the correlation that exists between the in vitro and in vivo biological reactivity of many mineral dusts (Morgan et al., 1977; Brown et al., 1978; Chamberlain et al., 1979; Richards et al., 1980; Davis et al., 1985). For the assessment of the biological effects of industrial dusts, it is generally recognized that one should always use Fig. 5 Cytotoxic Effects of Chrysotile, Attapulgite, Xonotlite and Fiberfrax (Al-Silicate) Fibers on Rat PAM. Releases of lactate Dehydrogenase (LDH; A and B) and B-Galactosidase (3-GAL; C and D), and_ATP Contents (E and F) of Freshly Cultured Cells Exposed to"50 yg (A, C and E) and 250 ug (B, D and F) of Mineral Fibers. IslSir 74 NADEAU ET AL. RAT PULMONARY MACROPHAGE RESPONSE (MONOI.AVER CULTURED FOR 24 MRS) : S' L 'H i S 100 1 CHRYSOTtLE 3- 2 ATTAPULGITE 80 3 XONOTLITE O 4 AL-SILICATE LU 60 5 Z 40 O LU P>* 20 CO < LU O _l s PAM 1 2 3 ALONE ---------- + 50 y0 of fibres LU T <t o > U. O >K c O < LU s > N Z LU ALONE---------- + 50 Iig of fibre? * 4 flTfrimi. PAM ALONE 23 + 250 ug of fibres f< ( : A DEAD ET al. SPONSE 7 34 250 ug ------fibres 23 4 250 ug ------ -- .< fibres F 'll 234 > 250 ug --------------of libras CYTOTOXICITY OF RESPIRABLE D'JSTS 75 multiple assay systems and parameters. However, the utilisation of short-term in vitro assays like hemolysis and PAH cultures are of scientific relevance, because cytotoxicity tests can provide practical informations on the pathogenic events linked to the exposure of an organism to respirable particles (Tilkes, 1985). Three of the four silicates tested certainly fall into that dimensional range: the chrysotile, -xonotlite and attapulgite samples are 100% respirable. For the Fiberfrax sample, the classification is less well defined; nevertheless, our short fraction can be considered highly respirable. If we look first at the aluminium silicate fibers, we found that this sample was non-hemolytic, but highly cytotoxic. This duality of the responses between the two in vitro assays is not uncommon: for example, the amphibole asbestos fibers are very well known to be poorly hemolytic while being highly cytotoxic to macrophages (Harington, 1976; Davies, 1980). While hemolytic dusts cannot be automatically classified as fibrogenlc (Manyai et al., 1969), inversely, non-hemolytic or poorly hemolytic minerals dusts are not necessarily non-fibrogenic (Richards et al., 1980). On the other hand, dusts cytotoxic to macrophages in vitro usually cause fibrosis in vivo (Harks and Nagelschmidt, 1959; Styles and Wilson, 1973; Davies et al., 1980). It is not known at this time if the presence of nun-fibrous particles in the Fiberfrax sample can account for the observed cytotoxic effect. However, the presence of long and thin fibers in the aluminium silicate sample is certainly an important factor to consider (Chamberlain et al., 1979). notwithstanding the morphology of the Fiberfrax sample, it is interesting to Fig. 6 Cytotoxic Effects of Chrysotile, Attapulgite, Xonotlite and Fiberfrax (Al-Si 1icate) Fibers on Rat PAH. Releases of Lactate Dehydrogenase (LDH; A and B) and 3-Galactosidase (6-GAL; C and D), and ATP Contents (E and F) of One Dav-Old Cultured Cells Exposed to 50 jg (A, C and E) and 250 pg (B, D and F) of Hineral Fibers.-.- -  ;: `ti It \ ' 76 NADEAU ET AI.. 1i * 1< |;i if; | ! 14 1! rr it ,? v* 'f iIsI j ! TABLE 5. Production of Lactic Acid by PAH Exposed to the Fibrous Silicates Controls Culture Conditions Fresh Monolayer X Day-Old Monolayer Percentage of Production3 100 14.7b 100 t 7.3C Chrysotile 50 pg 250 ug Attapulgite 50 pg 250 pg 78.7 t 20.7 39.9c 31.4d 62.7 i 20.ld 28.9 t 12.6d 106.6 12.3 102.0 t 12.6 81.2 5.5 57.6 i 11.7 Xonotl i te 50 pg 250 pg 95.5 i 13.9 36.2 17.6d 113.7 7.6 87.5 * 9.1 Fiberfrax 50 pg 250 pg 71.7 * 16.8 55.4 28.0d 94.0 17.4 82.0 20.0 (a) The data represent the mean value t S.D. (n = 5). (b) For this set of conditions, the mean production of lactic acid was 121.7 pg. (c) For this set of conditions, the mean production of lactic acid was 184.2 pg. (d) p<0.05 when compared to the untreated cells (controls). hi ?3 f.j If A IP *i iff i rwA-.vf-vv. I3EAU ET AT.. is Silicates id Monolayer i t 7.3 i 12.3 i i 12.6 t 5.5 / 11.7d ' i 7.6 . i 9.1 i t 17.4 i i 20.0 CYTOTOXICITY OF RESPIRABLE DUSTS 77 ' note that aluminium silicate fibers have recently been reported to be both j fibrogenic and carcinogenic to rats (Davis et al., 1983). The direct comparison of the other three fibrous silicates, namely attapulgite, xonotlite and chrysotile, is facilitated considering their average dimensions. As shown before, very short* chrysotile fibers were found to be both hemolytic and cytotoxic (Dunnigan et al., 1980; Nadeau, 1981; Pele et al., 1983). Similarly, our results confirm that attapulgite fibers from various sources are highly hemolytic and cytotoxic (Jaurand et al., 1979*3; Nadeau, 1981; Nadeau et al., 1983**; Harvey et al., 1984), and can even stimulate squamous metaplasia in vitro (Woodworth et al., 1983). Calcium silicates particles are also known to be biological active in vitro (Hunt et al., 1981; Skaug et al., 1984). However, contrary to chrysotile and attapulgite, which have a uneauivocally fibrous nature, the morphology of this type of mineral dust vary greatly with its origin (Skaug et al., 1984). Nonetheless, these authors showed that a highly fibrous synthetic tobermorite containing xonotlite was the most cytotoxic sample among the five types of calcium silicates tested. Although the biological relevancy of sshorts fibers is very often questioned, recent reports state clearly that shorter fibers should not be considered innocuous, but rather that the biological reactivity of fibers is diminishing with fiber length (Gormley et al., 1985; Davis et al., 1985). In fact, the very short* chrysotile fibers used in this study have been shown to produce some degree of alveolitis both in the rat (Lemaire et al., 1985) and in the sheep (Bfigin et al., in press). In the rat model, although no fibrotic lesions like those induced by the UICC B chrysotile fibers were apparent, the alveolitis persisted during the 60 days of the experiment; 9 months after the single 5-mg exposure, the alveolitis reaction was still present in some animals (I. Lemaire, personal communication). In the sheep 78 XADEAU ET AL. model,- similar results were seen 60 days after the exposure to a single 100-mg dose of the same short chrysotile fibers; yet, some areas of early fibrosis were even observed, although to a lesser degree when compared to the longer UICC B fibers. In their study. Begin et al. also examined the lung reactivity to the American attapulgite. They concluded that attapulgite fibers can induce lesions comparable to those of short chrysotile fibers, except that early peribronchiolar lesions were seen more often with the attapulgite sample. Although of lower intensity than those of the highly fibrogenic UICC B chrysotile sample, they nonetheless predicted that the lung lesions induced by the American attapulgite fibers should have a fibrotic evolution (Lee et al., 1981). So, at least as far as pulmonary fibrosis is concerned, there is a growing interest to the effect that various forms of attapulgites could present in humans a risk similar to asbestos (Sors et al., 1979; Bignon et al., 1980; Begin et al., in press). These in vivo findings are particularly important in view of our in o vitro data. The density of the mineral dusts are: chrysotile, 2.5 g/cm 3 (Jolicoeur et al., 1981); attapulgite, =2.5 g/cm (Haden, 1963); Fiberfrax, 33 2.7 g/cm (The Carborundum Company, 1984); xonotlite, 2.7 g/cm (Kudoh and Tak6uchi, 1979). So, in combination with the data in Table 4.3, on a basis of fiber number rather than weight, one should expect to have a cytotoxicity scale like: attapulgite = chrysotile > xonotlite>> Fiberfrax. However, with the exception of the aluminium silicate sample, which contains tong and thin fibers potentially more cytotoxic (Brown et al., 1978; Beck and Tilkes, 1980; Gormley et al.,, 1985), the decreasing order of cytotoxicity observed for the other three short silicates was: attapulgite >> xonotlite chrysotile. This ranking, based on the relative potency of each dust with the two PAM assays, correlate very well with the available in vivo data on the short chrysotile and American attapulgite fibers (Lemaire et al., 1985; B6gin et al., in press). \OEAU ET AL. Lo a single as of early ared to the ed the lung attapulgite lie fibers, n with the the highly at the lung a fibrotic fibrosis is us forms of ors et al., of our ui 2.5 g/cm3 Fiberfrax, (Kudoh and on a basis to have a > Fiberfrax. ch contains 8; Beck and ytotoxicity xonotlite h dust with Ivo data on al.. 1985; CYTOTOXICITY OF RESPIRABLE DUSTS 79 , In conclusion, it is believed that the combined use of fresh and i cultured PAM for the evaluation of the in vitro biological activity of ! mineral dusts may be usefull to extrapolate their in vivo pathogenecity. The direct comparison of different respirable dusts can then be achieved, more so if the particles tested have similar morphologies. SUMMARY The membranolytic and cytotoxic properties of two naturally occurring (chrysotile asbestos; attapulgite clay) and two man-made (Fiberfrax^, an . aluminium-silicate, and xonotlite, a calcium silicate) industrial minerals were compared. Short fiber fractions of chrysotile and Fiberfrax were obtained by sedimentation in demineralized water, while the attapulgite and xonotlite samples were used as obtained. The aluminium silicate fibers were found to be non- hemolytic, while for the other three silicates, chrysotile . had the strongest hemolysis potential, followed very closely by xonotlite; attapulgite was less hemolytic than the former two silicates, but was , nevertheless highly hemolytic to the rat erythrocytes. Using rat pulmonary alveolar macrophages, the In vitro cytotoxicity assays . showed that with fresh cell monolayers, all four silicates were equivalent in ` causing cell damages at a dose of 250 ug; at a lower dose (50 ug), the intensity of the cytotoxic effect was in the decreasing order: Fiberfrax >.attapulgite > chrysotile > xonotlite. With one day-old cultured cell monolayers, a dose of 250 pg of the silicates fibers was less cytotoxic, with the exception of the attapulgite fibers which remained essentially as cytotoxic as with the fresh cell monolayers. The reduced cytotoxic response was especially noticeable with the chrysotile fibers. At 50 jig, the cytotoxicity scale of the mineral dusts with one day-old cell monolayers was essentially the same as the one obtained with the fresh cell monolayers, that , is: Fiberfrax s-attapnlgite-> chrysotile ^.xonotlite. I 1 .: i i ! ! 80 NADEAU ET AI.. Overall, these in' tests imply: 1) that all four industrial silicates tested can be considered to be ((biologically actives; Z) that on the basis of their different reactivities with the two types of cell culture conditions used, their biological reactivity in vivo might be quite distinct. This might be especially true for at least the chrysotile, attapulgite and xonotlite short fibers, considering that these three types of silicate dusts have very similar dimensions. Moreover, for the chrysotile and attapulgite samples*, fiber numbers is probably not an important factor, since the density of the Jtwo silicates is roughly the same. The unaltered cytotoxic responses of the' American attapulgite fibers-in the two macrophage assays correlate well with the fact that short attapulgite fibers seems to have a stronger in vivo reactivity than short chrysotile fibers. In fact, it might be the strongest of the four Industrial silicates tested. On that basis, it is concluded that the combined use of fresh and cultured PAM for an in vitro assay may be useful! to extrapolate the pathogenicity of (trespirables dusts, more so if the comparison is done with particles with similar morphologies. | ACKNOWLEDGMENTS I 'I J He wish to thank Mr. Bruno Ga gne 1, Mr. Alai2n Lemieux , Ms. Irene LSvesque 1 , Mr. Daniel Ouellet3 and Ms. Martine Senneville1 for their able technical assistance, and Ms. Ginette Prince for typing this manuscript. This work was supported by L'Institut de I'amiante, Division de la recherche (Grant ?AL9). i: 1 REFERENCES >3 l Adams, C.J. and B. Storrie (1981). A simple DNA-dependent fluorescence enhancement, assay for cell numbdr. J. Histochem. Cytochem. 29: 3Z6-3Z8. WkiYvK >5? J .DSAU ET AL. 1 silicates he basis of conditions inct. This ulgite and icate dusts attapulgite the density c responses s correlate stronger iri ght be the sis, it is an in vitro dusts, hoiogies. LSvesque*, technical This work rche (Grant -cence 9; 326-328. ii - 1 CYTOTOXICITY OF RESPIRABLE DUSTS 81 I l ; Anderson, C.H. and J.M. Long (July, 1976). aPreliminary interim procedure for fibrous asbestos*. U.S. Environmental Protection Agency (E.P.A.). . Analytical Chemistry Branch, Environmental Research Laboratory, Athens, Georgia, 30601, USA. Beck, E.G. and F. Tilkes (1980). *In vitro* effects of defined mineral ' fibres. In: The in vitro effects of mineral dusts. R.C. Brown, I.P. Sormley, K. Chamberlain and R. Davies (Ed.) p. 339-343. Academic Press, London, New York, Toronto, Sydney, San Francisco. Begin, R., S. MassS, M. Rola-Pleszczynski, M. Geoffroy, M. Martel, Y. Desmarais and P. SSbastien. The lung biological activity of american attapulgite. Environ. Res. (in press). Bignon, J., P. SGbastien, A. Gaudichet and M.C. Jaurand (1980). Biological effects of attapulgite. In: Biological effects of mineral fibres. Vol. 1, J.C. Wagner (Ed.) p. 163-181. International Agency for Research ' on Cancer (1ARC), Scientific publication No. 30, Lyon, France. Biological effects of mineral fibres (1980). Vol. 1 and 2, J.C. Wagner . (Ed.). International Agency for Research on Cancer (IARC), Scientific publication No. 30, Lyon, France. ; Brown, R.C., M. Chamberlain,*D.M. Griffiths and V. Timbrell (1978). The . effect of fibre size on the in vitrobiological activity of three types _ of amphibole asbestos. Int. J. Cancer 22:721-727. 8runaueur, S., P.H. Emmett and E. Teller (1938). Adsorption of gases in multimolecular layers. J. Amer. Chem. Soc. 60: 309-319. Brunk, C.F., K.C. Jones and T.W. James (1979). Assay of nanogram ' quantities of DNA in cellular homogenates. Anal. Biochem. 92: 497-500. Chamberlain, M., R.C. Brown, R. Davies and D.H. Griffiths (1979). In vitro . prediction of the pathogenicity of mineral dusts. Br. J. Exp. Pathol. . SO: 320-327. Chatfield, E.J., R.W. Glass and M.J. Dillon (January 1978). Preparation of water samples for asbestos fiber counting by electron microscopy* EPA Report EPA-600/4-78-011. National Technical Information Service, Springfield. Virginia, 22161, USA. Christomanou, H., C. Cap and K. Sandhoff (1977). Isoelectric focusing pattern of acid hydrolases in cultured fibroblasts, leucocytes and cell-free amniotic fluid. Neuropadiatrie 8: 238-252. i.J ' i e t Dayies, P., A.C. Allison, J. Akerman, A. Butterfield and S. Williams (1974). Asbestos induces selective release of lysosomal enzymes from mononuclear phagocytes. Nature (Lond.) 251: 423-425. : Davies, R. (1980). The effect of mineral fibres on macrophages. In: Biological effects of mineral fibres. Vol. 1, J.C. Wagner (Ed.) p. 419-425. International Agency for Research on Cancer (IARC), Scientific publication No. 30, Lyon, France. ; :* Davies, R., H. Chamberlain, R.C. Brown and D.M. Griffiths (1980). Identification of toxic mineral dusts using manjaalian cells. Transactions of the Royal Society of Edinburg Earth Sciences 71: 181-184. Davis, J.M.G., J. Addison, R.E. Bolton, X. Donaldson, A.O. Jones and A. r Wright (1983). The pathogenic effects of fibrous ceramic aluminium silicate glass on' rats. In: Biological effects of man-made mineral fibres. WHO/IARC, EURO Reports and Studies 81: 124. Davis, J.M.G., R.E. Bolton, H. Cowie, K. Donaldson, I.P. Gormley, A.O. Jones and A. Wright (1985). Comparisons of the biological effects of mineral fibre samples using in vitro and in vivo assay systems. In: In ' vitro effects of mineral dusts. E.G. Beck and J. Bignon (Ed.;. NATO :l . ASI Series G: Ecological Sciences, Vo. 3. p. 405-411, Springer-Verlag, Berlin, Heidelberg. Downs, T.R. and W.W. Wilfinger (1983). .Fluorometric quantification of DNA in cells and tissue. Anal. Biochem. 131: 538-547. f Ounnett, C.W. (1955). A multiple comparison procedure for comparing several treatments with a control. J. Amer. Stat. Assoc. 50: 1096-1121. >! NADEAU ET AL. ion Service, cusing -ucocytes and lams enzymes from In: Wagner (Ed.) ncer (IARC), .lian cells, fences 71: and A. .ic aluminium made mineral , A.D. -1 effects of ems. In: In (Ed.). NATO inger-Verlag, ion of DNA ring .0: 1096-1121. CYTOTOXICITY OP RESPIRABLE DO STS 83 Dunnigan, J,, 0. Nadeau and 0. Paradis (1984). Cytotoxic effects of aramid fibres on rat pulmonary macrophages: comparison with chrysotile asbestos. Toxicol. Let. 20: 277-282. Dunnigan, J., D. Nadeau, D. Paradis, J.-P. Pele, R. Calvert, J.-M. Lalancette and M. Cossette (1980). Cytotoxic and haemolytic effect of' native and chemically modified chrysotile. Proceedings of the 4th International Conference on Asbestos, Torino, Italy, Hay 26-30, Vol. II, p. 747-772. Funk, H. (1965). Chemical alteration of poorly crystallised calcium silicate hydrates by hydrothermal treatment. Symposium on sAutoclaved calcium silicate building products::. University of London, School of Pharmacy, London, Hay 18-21, p. 122-123. Gormley, I.P., G.H. Brown, H. Cowie, A. Wright and J.M.G. Davis (1985). The effects of fiber length on the in vitro cytotoxicity of asbestos samples in three different assay systems. In: In vitro effects of mineral dusts. E.G. Seek and J. Bignon (Ed.). NATO ASI Series G: Ecological Sciences, Vol. 3, p. 397-404, Springer-Verlag, 8erlin, Heidelberg. Haden, W.L. Jr. (1963). Attapulgite: properties and uses. Clays and Clay Minerals 10: 284-290. Harington, J.S. (1976). The biological effects of mineral fibers, especially asbestos, as seen from in vitro and in vivo studies. Ann. Anat. Pathol. 21: 155-198. Hgrington, J.S., K. Hiller and G. Hacnab (1971). Hemolysis by asbestos. Environ. Res. 4: 95-117. Harvey, G., H. PagS- and L. Dumas (1984). Binding of environmental 84 NADEAU ET AL. Oaurand, H.C., L. Magne and J. Bignon (1979a). Inhibition by phospholipids of haemolytic action of asbestos. Brit. J. Industr. Med. 36: 113-116. Jaurand, M.C., J. Bignon, L. Magne, A. Renier and J. Lafuma (1979^). Interaction des fibres avec les globules rouges et les macrophages alyeolaires in vitro. Rev. Fr. Hal. Resp. 7: 717-722. Jolicoeur, C-, P. Roberge and ,).L. Fortier (1981). Separation of short fibers from bulk chrysotile asbestos fiber materials: analysis and physico-chemical characterization. Can. J. Chem. 59: 1140-1148, Ounowicz, E. and O.H. Spencer (1973). Studies on bovine pancreatic deoxyribonuclease A.I. General properties and activation with different bivalent metals. Biochim. Biophys. Acta 312: 72-84. Kudoh, V. and Y. Takfiuchi (1979). Polytypism of xonotlite: (I) Structure of an A-l polytype. Miner. J. 9: 349-373. Langer, A.M., M.S. Wolff, A.N. Rohland and I.J. Selikoff (1978). Variation of properties of chrysotile asbestos subjected to milling. J. Tox. Env. Health 4: 173-188. Lee, K.P., C.E. Barras, F.D. Griffith, R.S. Waritz and C.A. Lapin (1981). Comparative pulmonary responses to inhaled inorganic fibers with asbestos and fiberglass. Environ. Res. 24: 167-191. Lemaire, I., 0. Nadeau, J. Dunnigan and S. Masse (1985). An assessment of the fibrogenic potential of very short 4T30 chrysotile by intratracheal instillation in rats. Environ. Res. 36: 314-326. Levy, G.A. and A. McAllen (1963). Mammalian fucosidases. Biochem. J. 87: 361-367. Light, H.C. and E.T. Wei (1977). Surface charge and hemolytic activity of asbestos. Environ. Res. 13: 135-145. Hanyai, 3., J. Kabai, J. Kis, E. Suveges and H. Timar (1969). The in vitro hemolytic effect of various clay minerals. Med, Lav. 60: 331-342. I Harks, J. and G. Nagelschmidt (1959). Study of the toxicity of dust with use of the in vitro dehydrogenase technique. AHA Archives of Industrial Heal th *20:' 383-389. ADEAU ET AL. SpholipidS 113-116. >b). macrophages short analysis and 1148. .ic ith different tructure Variation Tox. Env. , (1981). fibers with issment of intratracheal cm. J. 87: ctivity of he in vitro 31-342. dust with of Industrial CYTOTOXICITY OF RESPIRABLE DUSTS 85 Morgan, A., A. Holmes and R.J. Talbot (1977). The haemolytic activity of some fibrous amphiboles and its relation to their specific surface areas. Ann. Occup. Hyg. 20: 39-48. Nadeau, D. (1981). Cytotoxfcitfi des fibres naturelles et des fibres artlcicielles. Rapport final presents au Ministere de I'Energie et des Ressources, Gouvernemant du Quebec, Quebec (QC), Canada, G1R 4X7, Nadeau, D., D. Paradis, A. Gaudreau, L. Geoffrion and L. Fouquette- Couture (1983). Modulation of the in vitro reactivity of pulmonary alveolar macrophages to chrysotile asbestos as a function of time in culture. Environ. Health Perspec. 51: 374. Nadeau, D,, D. Paradis, A. Gaudreau, J.P. Pel6 and R. Calvert (1983*). Biological evaluation of various natural and man-made mineral fibers: cytotoxicity, hemolytic activity and chemiluminescence study. Environ. Health Perspec. 51: 374. Paradis, D. and D. Hadeau (1982). Quantitative determination of lactic acid in cell culture media and biological fluids: optimization by automatic rate analysis. J. Tis. Cult. Methods 7: 175-180. Pele, J.P., R. Calvert and J. Dunnigan (1983). Hemolytic activity of chrysotile asbestos fibers: a freeze-fracture study. Environ. Res. 31: 152-163 Richards, R.J., G. George, J. Hust and T.O. Tetley (1980). The relationship between the hemolytic potential of certain particulates and their reactivity at the lung surface sin vivo*. In: The in vitro . effects of mineral dusts. R.C. Brown, I.P. Gormley, M. Chamberlain and R. Davies (Ed.). p. 324-332. Academic Press, London, New York, Toronto, Sydney, San Francisco. Schnitzer, R.J. and F.L. Pundsack (1970). Asbestos hemolysis. Environ. Res. 3: 1-13. Sfcaug, V., R. Davies and B. Gylseth (1984). In vitro macrophage cytotoxicity of five calcium silicates. Brit. J. Ind. Med. 41: 116-121. l htf/jfti&kii i ^ ij *> { "5; 86 NADEAU ET AL. Sorsi H., A. Gaudichet, P. SSbastien, J. Bignon and P. Even (1979). Lung fibrosis after inhalation of fibrous attapulgite. Thorax 34: 695-696. Styles, J.A. and J. Wilson (1973). Comparison between in vitro toxicity of polymer and mineral dusts and their fibrogenicity. Ann. Occup. Hyg. 16: 241-250. Tallarlda, R.J. and R.B. Hurray (1981). In: Manual of pharmacologic calculations with computer programs. Springer-Verlag, N.Y. . The Carborundum Company (1984). Form C733-1, p 1. The in vitro effects of mineral dusts (1980). R.C. Brown, I.P. Gormley, M. Chamberlain and R. Davies (Ed.). Academic Press, London, New York, Toronto, Sydney, San Francisco. Tilkes, F. (1985). In vitro cytotoxicity systems for mineral dusts. State of the art. In: In vitro effects of mineral dusts, p 389-396. E.G. Beck and J, Bignon (Ed.). NATO ASI Series G: Ecological Sciences, Vol. 3, p. 389-396. Springer-Yerlag, Berlin, Heidelberg. Timbrel!, V. and R.E.G. Rendall (1971/72). Preparation of the UICC standard reference samples of asbestos. Powder Technol. 5: 279-287. Turner Designs, Mountain View, CA, USA (1981). ATP measurements: Theory and practice. Luminescence Review, Bulletin #202. Hacker, W.E.C. and P.J. Snodgrass (1960). Serum LDH activity in pulmonary embolism diagnosis. J. Am. Med. Assoc. 174: 2142-2145. Woodworth, C.D., B.T. Mossman and J.E. Craighead (1983). Induction of squamous metaplasia in organ cultures of hamster trachea by naturally occurring and synthetic fibers. Cancer Res. 43: 4906-4912. 1 I