Document 91rEOD0pQJYL9mzmLV4R3vZGV

Lsf.i*r F--'8' ^3 Br. J. Cancer (1982) 45, 352 BIOLOGICAL EFFECTS OF TREMOLITE J. C. WAGNER, M. CHAMBERLAIN*, R. C. BROWN, G. BERRY, F. D. POOLEYf, R. DAVIES and D. M. GRIFFITHS From the MRC Pneumoconiosis Unit, Llandough Hospital, Penarth. S. Glam. CFG lA'II" and the fDepartment of Mineral Exploitation. University College. Cardiff Received 12 August 1981 Accepted 10 November 1981 Summary.--Tremolite is an amphibole which has been implicated in a variety of disease patterns in different parts of the world. It occurs in a number of phases, which are chemically identical but have specific physical characteristics. In an attempt to clarify the epidemiological findings, tremolite fibres of 3 specific forms-- A, B and C--were characterized and studied for biological activity by: (i) in vivo intrapleural injection of rats (2 separate experiments--1 with poor survival). (ii) in vitro enzyme release from mouse peritoneal macrophages (ill) in vitro giant-cell formation in A549 cultures (iv) in vitro cytotoxicity for V79-4 cells. Sample C, which contained more long thin fibres than A and B, was alone in producing mesotheliomas. G, but not A or B, induced LDH and B-glucuronidase enzyme release, and induced giant cells. A was not cytotoxic, B moderately cytotoxic and C as highly cytotoxic as UICC crocidolite. The in vivo studies were marred by being split between 2 experiments, of which the second had poor survival. We are aware of the weakness of our in vivo data, but as Tremolite C was being considered for commercial use on the European market we felt it timely to submit our findings for publication. Tremolite is an amphibole mineral (a chain silicate similar to asbestos) found in several countries. It has limited industrial value, but is used for stuccoing the exterior of buildings in the Middle East. It is frequently found as a contaminant of other minerals that are being exploited commercially. Data on the health hazards of tremolite are currently being collected because it is an amphibole mineral and may be capable of causing diseases similar to those in duced bv amphibole asbestos (Wagner, 19S0, 1982). A flake-like tremolite is found as a contaminant of talc in California but so far there is no evidence of disease which may be attributed to this tremolite. In the massive ehrvsotile ore bodies in Quebec Province in Canada, there are irregular deposits of a coarse-fibred tre molite. This material is found in the lungs of miners with pulmonary fibrosis and pleural plaques, but there is no correlation with mesotheliomas (Pooley, 1970). Further south in the northern part of New York State, a finer tremolite occurs as a contaminant of the talc deposits which are being exploited. From these mines there is evidence of pulmonary fibrosis, excess carcinoma of lung and pleura and a peritoneal mesothelioma (Kleinfeld et al., 1967,1974). Pleural plaques have occurred in the agricultural areas of Czechoslovakia and Yugoslavia and in the tobacco growing regions of Bulgaria and Greece, all areas in which coarse tremolite fibres * Present address: Unilever Research, Cohvorth House, Sliambrook, Bedford. U.K. 10003034 BIOLOGICAL EFFECTS OF TREMOLITE 353 > 1 pm in diameter are found in the soil. In Cyprus, a few mesotheliomas hare occurred among the people living in the vicinity of the asbestos mine. This mine has large chrysotile ore bodies associated with tremolite, some of which is finefibred. These people use the tremolite, which is separated from the chrysotile during the milling process, for stuccoing their houses. A similar situation exists in Turkey, where in some areas the tremolite used for stuccoing is associated with the incidence of pleural calcification, meso theliomas and bronchial cancers (Yazicioglu et ah, 1980). A pure tremolite consisting of fine fibres is mined in South Korea. The carcinogenic potential of mineral dusts can be assessed by depositing the dusts in the pleural cavities of rats, either by injection (Wagner & Berry, 1969) or implantation (Stanton & Wrench, 1972). Previous work has demonstrated that the ability of a dust to induce mesotheliomas is related to the content of fibres over ~ 8 /xm long and thinner than ~l-5 /xm (Stanton et ah, 1977). Mammalian cells in tissue culture have been shown to be sensitive to the toxic effects of fibres of similar dimensions to those Avhieh are carcinogenic in vivo (Brown et ah, 1978; Chamberlain el ah, 1979; Wade et ah, 1980). It was considered important to test 3 types of tremolite which would cover the spectrum of particle types found throughout the world, using both in vivo and in vitro tests as the first stage in assessing the potential health hazards. MATERIALS AXD METHODS Dust samples The three samples of tremolite available were the flake-like material from the Cali fornian talc deposits, a medium-sized fibrous mineral from Greenland and the fine-fibred material from South Korea. The samples used in this investigation were selected because upon degradation they were found to form fibrous particles with very different size distributions both in length and dia meter ranges. All samples were prepared by milling in a small agate mill and ultra sonic dispersion, large particles being re moved by sedimentation in water. Tremolite Sample A was prepared from a sample of Californian tremolitic talc which originally contained 62% talc and 38% Fig. 1.--Length and diameter distribution of fibres in Tremolite Sample A. 10003035 354 J. C. WAGXER ET AL. Sample A B C Table I.--Particles!fig tremolite samples Xo. of non-fibrous particles (x 10<) 6-9 20-7 3-3 Total no. of fibres (x 10<) 51 4-8 15-5 Xo. of fibres > 8 fim long (x 103) and <1-5 >m diameter 1-7 0 56-1 tremolite. The talc in the sample was re duced by froth flotation to produce a tremolite sample >95% pure, the remaining material consisting mainly of talc together with minor magnesium and calcium carbonate material. The length and diameter distribution of the fibrous particles in this sample are illustrated by Fig. 1; most fibres were < 6 fim long and <0-8 fim diameter. The number of particles/ Hg is shown in Table I. The chemical analysis is contained in Table II, from which it can be seen that the iron content of this tre molite is relative!}' low, but its distinguishing feature was that it contained significantly more potassium and sodium than the other samples. Sample B was prepared from a tremolite rock specimen which originated from Green land. On comminution the specimen was found to break down into fibrous particles, most of which were <3 fim long and <1-2 fim diameter (Fig. 2). The number of particles/ fig is shown in Table I. In comparison with Tremolite A, this sample contained fibres which were on average both shorter and thicker. Sample B contained larger propor tions of calcium and iron than Sample A and little potassium (Table II). Sample C was prepared from a rock specimen which originated from South Korea. In appearance, the hand specimen contained no visual impurities, and on size reduction produced a dust which contained fibres up to 140 fim long, of which most were <0-6 /ini diameter (see Fig. 3). The fibres in sample C were very much longer and finer than those in samples A and B. The number of particles/ fig is presented in Table I. The iron content of Sample C was lower than that of Sample B but higher than that of Sample A (Table II). The chemical compositions of the 3 samples are very similar, the exceptions being the K and Xa in Sample A. and the Fe in Sample B. Analysis offibre size distributions The methods used for the preparation of mineral dust samples for viewing in the transmission electronmicroscope (TEM) have been described fully elsewhere (Brown et al., 1978). Briefly, an appropriately diluted sus- 10003036 %-t BIOLOGICAL EFFECTS OF TRE.MOLITE 355 included) Sample Na,0 -MgO AUOi SiOt KtO CaO FeO A 30 24-5 1-2 69-8 1-1 9-6 0-3 *? B 0-6 23-9 0-3 59-3 0-1 13-6 2-0 C 0-6 24-9 0-4 58-8 0-2 13-9 0-9 pension of each dust was filtered on to a 0-1/mi-pore-size Millipore membrane filter. The filters were coated with carbon and plaed over parallel-bar EM grids which were on an acetone-soaked sponge. As the membrane filters dissolved, the carbon coat, and the dust particles, were deposited on the grids which were then viewed by TEM. Overlapping fields were photographed and large mosaics constructed. The length and diameter of at least 300 fibres were measured where possible, and the numbers of fibres and non-fibrous particles determined. The rules advocated by Cooper et al. (1978) were, used for counting the fibres on the photo graphs. Typical photographs of each dust are shown in Fig. 4. In vivo carcinogenicity The experimental animals used in this investigation were barrier-protected SPF rats of the Sprague-Dawley and Wistar strains. Each dust sample was prepared in physiological saline (0-9% w/v NaCl in distilled water) at 50 mg/ml and sterilized by autoclaving (15 lb/in2 for 20 min). The dose of 20 mg of experimental material per rat was injected into the right pleural cavity (Wagner 4 Berry, 1969); animals receiving 0-4 ml saline served as controls. Equal numbers of males and females were used in each experimental group. Sample A was injected into 32 Wistar rats 2 j-ears before the rest of the investigation. 10003037 356 J. C. W'AGXER ET AL. TREMOLITE 'C' TREMOLITE 'A' - .y- -?/ 'vi.-v ... .... . ii i 10 ym TREMOLITE'B' Fic. 4.--Electron micrographs of the 3 samples of tremolite. Rats receiving SFA chrysotile served as positive controls in this experiment. Samples B and C were injected into groups of 48 Sprague-Dnwley rats: a group of 32 animals receiving UICC crocidolite served as positive controls. Rats were 8-10 weeks old when injected and were allowed to live out their lives. In vitro toxicity Enzyme release from mouse peritoneal macrophages.--Mouse peritoneal macrophages were obtained from 22-27g Swiss TO mice (Tuck and Son Ltd, Battlebridge, Essex) by peritoneal lavage using 3-5 ml Medium 199 containing 5 i.u. heparin, 100 i.u. benzyl- 10003038 BIOLOGICAL EFFECTS OF TREMOLITE 357 penicillin and 100 pg streptomycin per ml. About 1-5 xlO6 cells in 2 ml of the above medium Mere placed in albumin-coated 35mm-diameter Petri dishes (Davies, 1980) and allowed to attach for 1 h at 37C. The non-adherent cells were then removed by washing with phosphate-buffered saline. The remaining cells were cultured in 2 ml Medium 199 containing antibiotics and 10% heatinactivated acid-treated foetal calf serum (Chamberlain et al., 1979) in an atmosphere of 5% CO2 >n a'r at 37;C. The medium of macrophage cultures prepared 24 h previously was replaced by 2 ml of medium containing the test dust at 50, 100 or 150 /ig/ml; control cultures received medium without dust. After 18h incubation the medium was collected and the adherent cells disrupted by the addition of 2 ml saline containing 0-1% Triton X-100 and 0-1% bovine serum albumin and rubbing the Petri-dish surface with a silicone-rubber bung. Both the medium and the cell lysates were centrifuged at 800 g for 10 min and the supernatants assayed for lactic dehydro genase (LDH) and /1-glucuronidase (BGL) by the continuous-flow fluorimetric method of Morgan et al. (1978) using a Perkin Elmer Model 3000 fluorescence spectrophotometer. Giant-cell formation in A-549 cultures.-- Type II alveolar cells (A549) derived from a human tumour (Leiber et al.. 1976) were obtained from Dr G. Todaro. XCI. Bethesda, Maryland, U.S.A. The cells were grown in Dulbecco's modification of Eagle's minimal essential medium supplemented with 10% heat-inactivated FCS and antibiotics in an atmosphere of 10% CO in air at 37C. A standard inoculum of 10s cells was added to each of a series of 25cm2 culture flasks along with an appropriate amount of dust suspension. Four flasks were used for each dust. 2 at a dust concentration of 100 pg/ml and 2 at 200 pg/ml. Flasks with no dust served as controls. All the cultures were incubated for 5 days, the cells detached using trypsin-EDTA and suspended in an appropriate volume of medium and photo graphed on a haemacytometer. The dia meters of 200 cells from each treatment were measured as described in Chamberlain & Brown (1978). Cytotoxicity to V79-4 cells.--Chinese ham ster lung cells (V79-4) described by Chu & Mailing (1968) were obtained from Dr C. F. Arlett, MRC Cell Mutation Unit. Brighton, and cultured in MEM supplemented with 15% FCS and antibiotics at 37C in an atmosphere of 5% CO2 in air. This method has been reported in detail elsewhere (Chamberlain & Brown, 1978). Briefly, the survival of V79-4 cells in the absence or presence of a series of concentra tions of each dust was determined by adding the appropriate amount of dust to a suspen sion of single cells. The cell/dust mixtures were then placed on 60mm-diameter Petri dishes (--200 cells/dish) and incubated for 6 days. After incubation the medium was removed, the cells fixed with 10% formal saline and stained with 1% methylene blue. The colonies on each dish were counted in an automatic colony counter (Micro Measure ments Ltd. Cambridge). RESULTS Physical characteristics of the dusts All of the dusts contained fibres; representative photographs are shown in Fig. 4. The size distributions of the fibres in each dust are shown in Figs 1-3 and the numbers of particles per fig are presented in Table I. Samples A and B contained relatively few fibres and Sample C contained many very long thin fibres. Induction of mesotheliomas The percentages of rats developing a mesothelioma following the various treat ments are shown in Table III. The survival of the animals in Experiment II was poor because of infection, and is discussed later. In vitro toxicity Enzyme release from mouse peritoneal macrophages.--The release of both LDH and BGL from mouse peritoneal macro phages after 18 h incubation with each of the dusts is shown in Table IV. Samples A and B had little effect on the cells, Sample C induced the release of 30% LDH and over 60% BGL. Giant-cell formation in A549 cells.--The percentage of giant cells induced by each dust in cultures of A549 cells is shown in Table V. UICC crocidolite induced a significant percentage of giant 24 10003039 358 J. C. WAGXER ET AL. Table III.--Carcinogenic activities of the dusts in experimental animals Eipt I Saline control Sample A SFA chrysotile ( + ve control) Xo. of rats examined 32 31 32 Mean survival after injection (days) 717 644 612 mesotheliomas <o> 0 (0) 0 (0) 20 (62) Exptll Saline control Sample B Sample C UICC crocidolite (+ ve control) 23 48 47 31 532 549 541 537 0 (0) 0 (0) 14 (30) 2 (6) f ?e Table IV.--Activity of dusts against mouse peritoneal macrophages, measured by release of enzymes (mean of 4 cultures 950 confidence limits) Dust at 100 fig/ml Control Tremolite A Tremolite B Tremolite C L'ICC crocidolite % LDH release 5-6 + 0-4 10-4+1-2 14-9 + 0 8 28-3 + 0-9 39-1 + 1* 0 BGL-release 3-6 + 0-3 9-9 + 0-8 14-92-6 62-8 + 1-0 48-3 + 4-3 Table V.--Activity of dusts against .<4549 cells (% of giant cells, with 95% confidence limits; giant cells defined as those > 25 pm diameter) Treatment Control 1-47 (0 5-4 2] Dose 100 pg/ml 200 pg'ml Tremolite A Tremolite B Tremolite C UICC crocidolite 10(0-3-3-6) 5-3 (3 0-9-2) 19-8 (14-9-23-S) 14 4 (10-2-19-9) 4-5 (2-4-8-31 3-0 (1 -3-0 9) 24-3 (19-1-30-9) 26-3 (20-1-32-7) l i Table VI.--Cytotoxicity to V79-4 cells Dust Tremolite A Tremolite B Tremolite C UICC Crocidolite % survival at 50 jig/ml ( 95% confidence limits) 101 -0 11 -5 36-7 + 6-7 3-5+1-2 2-9+1-2 cells, as reported by Chamberlain & Brown (197S). Of the test dusts, only Sample C induced giant cells, and it was as active as UICC crocidolite. Cytotoxicity to T'79-4 cells.--The cyto toxic potentials of the dusts towards V79-4 cells are shown in Table VI. Sample A was inert, B was moderately toxic, but C was as toxic as UICC croci dolite. DISCUSSION As indicated in the introduction, data on the human health hazards of tremolite are currently being collected. We report here experimental studies on both the carcinogenic effects in vivo and the cytotoxic effect in vitro, of 3 samples of tremolite. Many inorganic dusts have been shown to be carcinogenic in experimental animals (for a review see IARC, 1977). Stanton et al. (1977) and Stanton & Layard (1978) demonstrated that the carcinogenic poten tial of a dust correlates with the number of fibres longer than ~ 8 jtm and thinner than ~l-5 pm per unit mass. We have reported previously that fibres of very similar size are responsible for cytotoxic 10003040 BIOLOGICAL EFFECTS OF TRE.MOLITE 359 effects in 3 types of mammalian cells the same period, 6 Sprague-Dawley rats (Brown ft ah, 1978; Chamberlain et ah, out of 4S developed mesotheliomas after 1979) . Wade et al. (1980) have made injections with UICC African chrysotile similar observations. In view of the fact (Wagner et ah, 1980b) with a mean that fibres of similar size are both carcino survival only one month longer than with genic in vivo and cytotoxic in vitro, the crocidolite in the experiment reported use of certain mammalian cells for the here. In Wistar rats UICC crocidolite detection of potentially pathogenic dusts produces more mesotheliomas than UICC has been proposed (Chamberlain et ah, chrysotile (Wagner et ah, 1973). Thus a 1979; Wade et ah, 1980; Brown et ah, low mesothelioma rate is not characteristic 1980) . of the Sprague-Dawley strain that we Only one of the tremolite samples, C, used. A second possibility is that the low was carcinogenic. This sample was also rate with crocidolite was a chance finding, consistently very active in the 3 in vitro and comparison of the present experiment systems. Sample C contained 5-6 xlO4 with the earlier ones indicates that this fibres >8 /an long and <T5 pm in possibility cannot be excluded (P>01). diameter per fig. UICC crocidolite, used The present experiment is imprecise due as a carcinogen-positive control dust, to the poor survival; in terms of mesothe contained 6-4 x 104 fibres of this size per lioma rate, there is an efficiency of only fig. UICC crocidolite and Tremolite Sample 40% of the earlier ones, i.e. the 31 rats C were found to be very similar in their with low survival are equivalent to only activities in the in vitro systems (Tables 12 rats with the longer survival previously IV, V & VI). However, Tremolite Sample obtained with Wistar rats. C seemed to be more carcinogenic than Unsatisfactory though the experiment UICC crocidolite (Table III). on carcinogenesis of Sample C may be, It is a weakness that the animal data owing to the near failure of the positive reported here though from 2 separate controls, the fact remains that Sample C experiments, using 2 strains of rat, were produced 14 mesotheliomas in 47 rats, impaired by the poor survival due to whereas Samples A and B produced none. infection in the second experiment. In In view of the foregoing remarks we this experiment only 2 Sprague-Dawley think that it is wiser to interpret the data rats (6%) injected with the UICC eorci- presented here in a qualitative rather than dolite positive control developed meso a quantitative manner. By analogy with theliomas. This is much lower than other members of the amphibole asbestos obtained previously with Wistar rats minerals we suspect that Tremolite Sample (Wagner et ah, 1973; Berry & Wagner, C, originating from South Korea, would 1976; Wagner et ah, 1980a) which gave be a human health hazard if present in 46% mesotheliomas on average. The sufficient airborne concentration. An ex mean survival in these earlier experi periment exposing rats to airborne clouds ments was over 4 months longer than in of this tremolite by inhalation is now the experiment reported here, which being planned as the next stage in assess partly explains the difference in meso ing the potential health hazard. thelioma rate. However, after allowing for survival, the mesothelioma rate in the second experiment reported here was only between 1/4 and 1/2 of the previous rates. The reason for this low rate is unknown, but an obvious possibility' is that it is characteristic of the SpragueDawley strain that we used. However, in another experiment carried out during REFERENCES Berry, G. & Wagner, J. C. (1976) Effect of age at inoculation of asbestos on occurrence of meso theliomas in rats. Int. J. Cancer, 17, 477. Brown, R. C., Chamberlain, M., Griffiths, M. & Timbrell, V. (1978) The effect of fibre size on the in vitro biological activity of three types of amphibole asbestos. Int.J. Cancer, 22, 721. . Brown, R. C., Chamberlain, M.. Davies, R. <fc Sutton, G. T. (1980) The in vitro activities of pathogenic mineral dusts. Toxicology, 17, 143. 10003041 V .usury o r ohiuauo lium akv "1 360 J. C. WAGNER ET AL. Chamberlain, M. 4 Brown, R. C. (1978) The cytotoxic effects of asbestos and other mineral dust in tissue culture cell lines. Br. J. Exp. Pathol., 59, 183. Chamberlain, M., Brown-, R. C., Davies, R. 4 Griffiths, D. M. (1979) In vitro prediction of the pathogenicitv of mineral dusts. Br. J. Exp. Pathol., 60, 320. Chi-, E. H. V. 4 Mallino, H. V. (1988) Mammalian cell genetics: Chemical induction of specific locus mutations in Chinese Hamster cells <n et'Iro. Proe. Eatl Acad. Sci., 61, 1306. Cooper, D. W,, Feldman, H. A. 4 Chase, G. R. (1978) Fiber counting: A source of error corrected. Am. Ind. Hyg. Assoc. J., 39, 362. Davies, R. (1980) The effect of dusts on enzyme release from macrophages. In The In Vitro Effects of Mineral Dusts, (Eds Brown et al.) London: Academic Press, p. 67. 1ARC (1977) The Evaluation of Carcinogenic Risk of Chemicals to Man. 14, Asbestos. Lyon: IARC. Kleinfeld, M., Messite, J., Kooyman, O. 4 Zaki, M. H. (1967) Mortality among talc miners ,and millers in New York State. Arch. Environ. Hlth, 14 663. Kleinfeld. M., Messite, J. 4 Zaki. M. H. (1974) Mortality experiences among talc workers: A follow-up study. J. Occup. Med., 16, 345. Leiber, M., Smith, S.. Szakal, A., Nelson-Rees, W. 4 Todaro, G. (1976) A continuous tumour- cell line from a human lung carcinoma with properties of Type II alveolar epithelial cells. Int. J. Cancer, 17, 62. Morgan, D. M. L., Vint, S. 4 Rideoft, J. M. (1978) Continuous flow fluorimetric assay of lysosomal enzymes. Med. Lab. Sci., 35, 335. PooleY, F. D. (1976) An examination of the fibrous mineral content of asbestos lung tissue from the ,Canadian chrvsotile mining industrv. Environ. Res., 12 281. ` Stanton, M. F. 4 Wrench, C. (1972) Mechanisms ,of mesothelioma induction with asbestos and fibrous glass. J. Wall Cancer Inst., 48 797. Stanton, M. F., La yard, M., Teoeris. A., Miller, E., May, M. 4 Kent, E. (1977) Carcinogenicity of fibrous glass: pleural response in the rat in rela tion to fiber dimensions. J. Satl Cancer Inst., 58, 587. Stanton, M. F. 4 Layard, M. (1978) The carcino genicity of fibrous minerals. In Proceedings of a Workshop on Asbestos: Definitions and Measure ment Methods. (Eds Gravatt et al.) National Bureau of Standards (U.S.A.) Special Publication, 506, 143. Wade, M. J., Lipkin, L. E., Stanton, M. F. 4 Frank, A. L. (1980) P38SDi in vitro cytotoxicity assay- as applied to asbestos and other minerals: Its possible relevance to carcinogenicity. In The in Vitro Effects of Mineral Dusts. (Eds Brown et al.) London: Academic Press, p. 351. Wagner, J. C. (1980) Environmental and occupa tional exposure to natural mineral fibres: In Biological Effects of Mineral Fibres. (Ed. Wagner.) Lyon: IARC. p. 995. Wagner. J. C. (1982) Mineral Fibre Carcinogenesis. Am. Chem. Soc. Monogr. (in the press). Wagner, J. C. 4 Berry. G. (1969) Mesotheliomas in rats following inoculation with asbestos, fir. J. Cancer, 23, 567. Wagner, J. C., Berry. G. 4 Timbrell, V. (1973) Mesotheliomata in rats after inoculation with asbestos and other materials, fir. J. Cancer. 28, 173. Wagner, J. C., Hill, R. J.. Berry. G. 4 Waoner, M. M. F. (1980a) Treatments affecting the rate ,of asbestos-induced mesotheliomas, fir. J. Cancer, 41 918. Wacner. J. C., Berry, G.. Hill, R. J., Mi-nday, D. E. 4 Skidmore, J. W. (19806) Animal experi ments with man-made mineral fibres. In The Biological Effects of Mineral Fibres. (Ed. Wagner). Lyon: IARC. p. 361. Yazicioglu, S.. Ikayto, R., Balci. K., Sayli, B. S. 4 Yorulmaz, B. (1980) Pleural calcification, pleural mesotheliomas and bronchial cancers caused by tremolite dust. Thorax, 35, 564. 10003042 APR 5 1985 yy ^ y\p- NOTICE: WARNING CONCERNING COPYRIGHT RESTRICTIONS The copyright law of the United States (Title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material. Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specified conditions is that the photocopy or other repro duction is not to be "used for any purpose other than private.study, scholarship or research." If a user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of "fair use," that user may be liable for copyright infringement. This institution reserves the right to refuse to accept a copying order if, in its judgement, fulfillment of the order would involve violation of copyright. 10003033