Document gDmagOorwNmxkor88KrjJ6zBJ

EXPERIMENTAI. AND MOI.ECUI.AR PATHOI.OGY 19, 339-353 (1973) Experimental Studies on the Effects of Heated Chrysotile Asbestos and Automobile Brake Lining Dust Injected into the Body Cavities of Mice1 J. M. G. Davis and S. W. Coniam Institute of Occupational Medicine, Roxburgh Place, Edinburgh EHS 9SU, Scotland Received February 20, 1973 A series of experiments with mice involving the intrapleural injection of chrysotile asbestos dust heated to varying temperatures up to 1000'C suggests that the physical shape of the dust particles is more important than the altered chemistry of the dust. Chrysotile heated to no more than 400C produced dust samples containing many long fibers, and these samples resulted in large pleural granulomas. Chrysotile heated to more than 400C produced dust containing mainly particles of low aspect ratio, and these produced very small granulomas. The initial size of the granulomas was very closely related to the final degree of fibrosis found in the animals and, therefore, chrysotile heated to 400C or less was much more fibrogenic than when it was heated to higher temperatures. A sample of automobile brake lining dust was found to con tain very little recognizable chrysotile, and consisted mainly of particles with a low aspect ratio. When this dust was injected into the pleural cavities of mice it produced very small granulomas and little fibrosis. The discovery by Wagner et al. (1960) that exposure to asbestos dust was often associated with the production of mesothelial tumors caused concern about the possible pollution of the normal urban atmosphere by asbestos dust. Wagner had shown that very small doses were apparently sufficient to produce tumors in some cases and it was suggested that the use of asbestos products from ironing boards to brake linings might be liberating sufficient dust into the atmosphere to cause a health hazard. Since automobile brake linings are constantly being abraded in use and since many tons of asbestos are consumed in this manner each year, the possibility of atmosphere pollution from brake lining dust has received particular attention. In order to test the level of atmospheric pollution by asbestos- a number of studies have been undertaken. The lungs from unselected autopsy cases from different parts of the world have been examined and initially, since uncoated asbestos dust is difficult to find in tissues, the presence of asbestos bodies was used as a likely index of the presence of asbestos. The figures obtained showed an increase in the percentage of cases containing bodies with the increasing sophisti cation of the techniques used. The first of these studies was that undertaken by 1 The dust samples used in this study were prepared in the laboratories of T.B.A. Industrial Products Ltd. of Rochdale, England. This work was undertaken as part of the research programe organised by the British Asbastosis Research Council. Cooyrleht (ft 1973 by Academic Press, Inc. All rights ol reproduction ia any form reserved. 339 PA-00312.05 SCF-FA-6704 340 DAVIS AND CONIAM Thomson et al. (1963) in Capetown and these workers reported bodies in mo; than 30'/ of cases. Later, Thomson and Graves (1966) found a similar prevalent! in Miami; Cauna, Totten, and Gross (1965) found bodies in 40% of a series of l ings from Pittsburg and Anjilvel and Thurlbeck (1966) found them in 45/ of their cases in Montreal. Meurman (1966) reported finding bodies in 70% of a group of lungs from Finland and finally, Utidjian, Gross, and De Treville (1968) using a new technique, reported that the lungs of 97 out of 100 consecutive autopsies from Pittsburg contained bodies. However, Davis, Gross, and De Treville (1970) showed that the production of ferruginous bodies was not spe cific for asbestos as had been previously believed and it therefore became neces sary to undertake new studies by which asbestos could be identified by precise methods. This is still a very difficult problem, although Langer et al. (1970) and Pooley et al. (1970) have shown that chrysotile asbestos can be positively identi fied in a large proportion of the lungs of town dwellers. The initial concern over generalized atmospheric pollution by asbestos involved the possibility of tumor production but the problem was given a new dimension by the work of Jagatic (1967). Jagatic reported that chrysotile heated to high temperatures was acutely toxic to mice and found that after the injection of heated chrysotile into the peritoneal cavity, 60/ of mice died within 48 hr. Since chryso tile is the type of asbestos usually used in brake linings and since, during braking, the linings are heated to very high temperatures, it became necessary to consider the possible acute hazards of the inhalation of brake lining dust. In order to study the biological effects of heated chrysotile and brake lining dust in more detail it was decided to undertake a series of experiments in which samples of these dusts were injected into the pleural and peritoneal cavities of mice. In later studies it is hoped to undertake inhalation experiments with these minerals to see whether they should be considered as dangerous atmosphereic pollutants. MATERIALS AND METHODS The mineral samples used in these studies were prepared as follows: Samples of high grade Cassiar chrysotile asbestos were heated in an electric tube furnace for 4 hr at temperatures of either 400C, 600C, 800C, or 1,000C. After cooling, the chrysotile was ground in a mechanical mortar and pestle and half of each sample was sieved through a 250-mesh copper sieve. With normal chrysotile and chrysotile heated to no more than 400 C many long fibers remained in the dust after grinding and since most of these were removed by sieving it was possible to produce samples both with and without long fibers. With chrysotile heated to more than 800C, however, the dust fibers became extremely brittle and grnding was found to break most of these into relatively short lengths. With these samples, therefore, sieving made little difference to the size distribution of the dust particles although experiments were conducted with both sieved and unsieved samples to test their effects on tissues. The thermal decomposition of chrysotile follows a two-stage sequence of dehydroxvlation and breakdown, and the mechanism has been studied by several authors including Ball and Taylor (1963) and Brindley and Hayami (1965). The dehydroxylation of chrysotile takes place in the temperature range 600-780C and CHRYSOTILE ASBESTOS AND AUTOMOBILE BRAKE LINING DUST 341 at 800-850 C the dehydroxylated noncrystalline residue is said to recrystallize to give forsterite and silica. The total decomposition corresponds to Mg, Si* O10 (OH), = 3Mgg SiO* + SiOa + 4H* O . The forsterite persists together with silica up to 1000C but above this temper ature some enstatite is formed 3Mg2 SiO* + SiOa = 2Mgs SiO* + Mg2 Sij O0 . Since the highest temperature used in these experiments was 1000C it was the decomposition of chrysotile to forsterite and silica that formed the basis of the present study. Throughout this paper the dust samples are referred to as heated chrysotile for simplicity. Readers interested in the exact chemistry of any partic ular sample are referred to the information in this paragraph. The sample of brake lining dust used in these experiments was supplied by a commercial producer of brake linings and had been taken from the rear brake drums of two of their test vehicles. These cars had been put through a test routine designed to simulate a long period of normal usage. The original brake lining composition was similar to that described by Lynch in 1968 and in addition to chrysotile asbestos and binding resins the linings had contained both chromium oxide and metallic aluminium granules. It would normally be expected that brake lining dust would contain metallic iron from the brake drum, but an X-ray analysis of the dust sample performed by A. L. Rickards failed to detect crystalline iron on this occasion. With these samples of heated chrysotile and brake lining dusts the following experiments were undertaken. In the first series, in order to confirm the results of Jagatic et al. (1967), doses of 250 mg of chrysotile heated to 1000 C were injected into the peritoneal cavities of 25 Balb/C mice. In the second series of experiments single 10 mg doses of normal chrysotile and chrysotile heated to either 400C, 600C, 800C, or 1000C were injected into the pleural cavities of similar mice. Batches of 25 mice were used for each dust sample and both sieved and unsieved dust specimens were used in each case. In the final series of experiments single doses of 10 mg of automobile brake lining dust were injected into the pleural cavities of another batch of 25 Balb/C mice. In all cases the dust samples were suspended in 1 ml of sterile distilled water before injection. Animals from all these studies were killed at intervals from 7 days to 1 yr after injection and samples of tissue from each were fixed in either formol saline for light microscopy examination or buffered osmium tetroxide for electron microscopy. For light microscopy, sections were stained with either hematoxylin and eosin, E.P.S., or Peris' stain for iron. For electron microscopy, the tissues were embedded in araldite and after sectioning were stained with lead citrate. OBSERVATIONS In the first series of experiments those animals given an intraabdominal injec tion of 250 mg of chrysotile heated to 1000C showed signs of toxic effects within 24 hr and by 48 hr seven had died. The remainder showed continuing signs of disability for several days but all eventually recovered and none died within 6 mo of dust injection. When those mice that had died within 48 hr of injection were M2 DAVIS AND CONIAM examined it was found that little or no cellular response had developed around the dust which remained scattered and loose in the peritoneal cavity. A histological examination of the viscera of those animals showed no signs of damage with the exception of some patchy hydropic change in the kidney tubules. When mice from this series were killed 7 days after dust injection it was found that the dust in the peritoneal cavity had by now become compacted into large gran ulomas. Most of these remained free in the peritoneal cavity, supported by strands of connective tissues, but some became attached to the surface of the omentum. It was only occasionally however that adhesions were formed between the loops of the intestine. Within the granulomas the cellular reaction to the dust consisted largely of macrophages with a few giant cells, fibroblasts, lymphocytes, and plasma cells and with the smaller lesions cells were able to penetrate throughout the dust mass. With a dose of 250 mg, however, some of the dust lesions were so large that complete penetration did not occur and the central regions of these granulomas remained completely acellular. By 2 wk after dust injection most of the granulomas were surrounded by a thin but clearly defined capsule of fibrous tissue and some collagen had been produced among the cells of the granuloma itself. A gradual increase in collagen was seen within the lesions between 7 days and 1 yr after injection but within this period the granulomas still contained many cells. In the experiments using 10 mg doses of cither heated chrysotile or brake lining dust no animals died within the first 2 wk after injection and none showed any toxic effects. Granulomas were produced in response to all the injected samples but cellular reaction to the dust was found to vary according to the dust type, the temperature to which it had been heated, and its method of preparation. Animals injected with either normal chrysotile or chrysotile heated to 400C were found after 2 wk to have cellular granulomas with complete cellular penetration throughout the dust masses. In these lesions cells masked the dust to such an extent that it was extremely difficult to see with the light microscope. In the animals injected with chrysotile ground and sieved the granulomas were relatively small and formed few adhesions, but in those animals injected with unsieved dust the granulomas were much larger for the same dust dose and frequently formed adhesions between the contents of the chest cavity. In animals injected with dust heated to 600C cellular granulomas were still produced but there appeared to be less difference between the sieved and the unsieved dust groups, and both produced only small granulomas without adhesions. In the groups of animals injected with chrysotile heated to either 800 or 1000C the resulting granulomas were always small compared with the dust dose used and no adhesions were found in either the sieved or the unsieved group. In these experiments, however, it was found that the chrysotile dust was now clearly visible in the tissues and in the light microscope appeared to consist of cylindrical or rectangular particles which ranged from 1 to 10^m in diameter and which could be up to 50/im in length. Normally cells were still able to penetrate throughout the dust masses but fewer cells were present than in granulomas produced by normal chrysotile. By 2 wk after dust injection a thin capsule of fibrous tissue was usually formed around those granulomas remaining free in the pleural cavity. Those that formed adhesions, however, did not develop any specific layer of fibrous tissue between the main granuloma and the organ to which they were attached. By 2 wk, also CHRYSOTILE ASBESTOS AND AUTOMOBILE BRAKE LINING DUST 343 Fig. 1. A section of a macrophage from a mouse granuloma 2 wk after an injection of clirysotilo asbestos heated to 0OOC. Croups of chrysotile crystals are seen in two phagosomes (F) within the cell, and the structure of these crystals appears normal. In both cases, the dust appears loosely arranged in the phagosomes and the phagosome membrane has not at this stage contracted closely around the dust. X 65,000. some collagen was usually present among the plagocytic cells that made up the main part of the lesions. During the first few weeks after injection collagen pro duction was more rapid in those granulomas induced by chrysotile dust heated to 800C or more, than in granulomas resulting from either unheated chrysotile or chrysotile heated to no more than 600C. However, by 6 mo after injection this distinction was no longer apparent, and the long-term production of collagen in any given area of granulation tissue did not appear to differ much between similarly prepared chrysotile samples that had been heated to different temper atures. The physical shape of the dust particles did, however, have considerable effect on fibrosis. The total amount of collagen produced in response to a 10-mg dose was closely related to the size of the resulting granulomas and since unsieved samples heated to no more than 400C produced large lesions these samples were more fibrogenic than sieved samples heated to 400C or to samples heated to higher temperatures. All these produced very small granulomas. Electron microscope studies showed that almost all the injected dust from samples heated at 600 C or less was phagocytosed by macrophages and occasional giant cells and few extracellular dust particles wore found. The process of :U4 DAVIS AND CONIAM Fig. 2. A section of a mouse granulomas 4 wfc after the intrapleural injection of chrysotile dust heated to 800C A large bundle of fused chrysotile crystals has been phagocytosed by a macrophage {M), and the phagosome membrane is in close contact with the dust (arrowed). Although the chrysotile crystals are partially fused with one another, the original crystal outline is still discernible. X 65,000. phagocytosis and the subsequent behavior of the dust within cells appeared identical to that reported previously (Davis, 1968, 1970a). Dust particles were initially taken up into large vacuolated phagosomes (Fig. 1), but the membranes of these eventually contracted and became closely opposed to the dust particles. Usually the dust remained on its own inside the phagosome membrane, but occasionally it was found mixed with ferruginous granules and membrane debris in the dense structures that are known as phagosome residual bodies. Most of the dust remained in those structures, but occasionally an odd fiber was found that appeared to have escaped into the cell cytoplasm. Tire macrophages from these experimental groups showed no signs of damage even when they contained large amounts of dust. No differences could be detected in the structural appearance of crystals of normal chrysotile and chrysotile crystals that had been heated to 400C. With dust heated to 600C, however, it was noticeable that the phagocytosed dust, whether sieved or unsieved, contained more individual chrysotile crystals than crystal bundles, and many phagosomes were found closely packed with indi vidual crystals. Electron microscope examination of lesions produced by chrysotile heated to 800C or 1000C showed why the dust was clearly visible in those lesions even with the light microscope. During the heating process the individual CHRYSOTILE ASBESTOS AND AUTOMOBILE BRAKE LINING DUST 345 chrysotile crystals had lost their normal tubular structure and fused with the other crystals in the bundle (Rickards, 1967). This fusion was complete enough to hold the crystals together during grinding so that the dust was broken up into relatively large masses rather than fine fibrous bundles often containing only a few crystals each. When the tissues were examined only a few weeks after dust injection the original outline of the chrysotile crystals within the fused masses could still be detected (Fig. 2), but several months after dust injection this crystalline structure was no longer visible. It nppeared that the dust had undergone some degree of dissolution within the tissues, and the neat crystal arrangement had been replaced by an irregular honeycomb structure (Fig. 3). Although the fused chrysotile masses were much thicker than normal chrysotile dust particles, most of them had no greater length, and the majority could still be phagocytoscd by single macrophages. Very large particles were occasionally seen inside giant cells although these cells are not common in mouse granulomas (Davis, 1970b), and more frequently the larger particles were closely surrounded by unfused macrophages. Inside the cells there were some notable differences between the behavior of the heated chrysotile masses and normal chrysotile. Particles of normal chrysotile are taken up into large phagosomes which presumably contain fluid and the phagosome membrane is often well clear of the contained dust, at I`T<i. 3. A section of a macrophage from a mouse granuloma 3 mo after the injection of chrys otile dust heated to 800C. The macrophage is surrounded by collagen fibers (F), but a large particle of fused chrysotile (C) is present within the cytoplasm. In this case, however, the normal crystal outline has been lost and has been replaced by an irregular honeycomh pattern. X 70,000. ' 340 DAVIS AND CONIAM Fic. 4. Particles of chrysotile dust within a macrophage from a mouse pleural granuloma, 3 mo after dust injection. The dust had been heated to 80Q'C before injection, and by this stage the chrysotile crystal pattern has been replaced by an irregular honeycomb structure. The largest dust particle (P) is associated with masses of dense granules (F) that probably represent ferritin, or haemosiderin, and in some areas these have penetrated deeply into the dust. X 35,000. least in the early stages. With the masses of heated fused chrysotile however, al though they were taken up into membrane bounded vesicles the membrane was always seen in close contact with the dust. Many of the small dust masses could still be taken up by single macrophages which usually showed no signs of damage. In these granulomas, however, it was not uncommon to find dead macrophages containing dust although such dead macrophages have only rarely been found in lesions produced by normal chrysotile. In animals injected with either unsieved normal chrysotile, or chrysotile heated to 400oC some typical ferruginous bodies (Davis, 1970b) were found in the gran ulomas from about 2 wk after dust inj'ection. Sieved samples of normal chrysotile and chrysotile heated to 400C did not, however, result in the production of any ferruginous bodies. This was also true of both sieved and unsieved samples of chrysotile heated to 600C. In animals treated with chrysotile heated to either 800C or I000C no true asbestos bodies were found in which the dust particles were surrounded by a distinct coating of ferruginous material. However, with some of the larger chrysotile fragments, especially those showing the greatest degree of dissolution, it appeared that dense granules similar to ferritin or haemosiderin had penetrated deep into the dust (Fig. 4). Brake lining dust when injected into the pleural cavity of mice produced smaller granulomas than any of the asbestos types used in this study and these CIlliV.SOTJI.K A.Sni'.STOS AMI :U. TOMOIULI'. IUUKK T.IMN'C DUST IT granulomas never formed adhesions, being mainly suspended in (he pleural cavity by thin strands of connective tissue. In the light microscope the brake lining (lust could ho seen In consist mainly of irregularly shaped particles 5-25 /on in diameter, but a number of elongated particles were also present which ranged Ironi 2 to 25 pin in diameter and which could he up to 50 jiin in length. Cells were able to penetrate deeply into tile dust masses hut the number of cells was always comparatively lew compared to the dust load. The granulomas were <|iiiekl\ surrounded by a thin capsule of fibrous tissue within a few days ol inject inn. and some collagen bad been produced among the dust containing cells within 2 wk. Alter this time collagen production increased quite rapidly and the' lesions became' less cellular so that 1 yv alter injection the dost particles had been bound together by a network ol old acellular collagen. Since tin: initial granulomas were so small however, the amount oi collagen produced in rosjxmse t'i the dust was never very great. 1 it:. \ iiarlicle of automobile brake lining dust within a inacmphagt' lVum a mouse pleural Liannlniiia 2 ol. after chisl inii'dimi. Tin- ilrtsl i,. separated from llu: cell oytoplasma l>y a i ii'iuiaata . lint (ilis is taosrlv < ip| a .si 'il to 11)(' (Inst at al omuls (ammeil). 'I lie (Inst consists i .ainlv .1 ili'iisc amorphous materi.>!. am! cnnlaius no reeiiuiii/.uhic ehrywtlile asheslns. There a-e. hev.evr. within I la- tin.: .ume ilense cryslolliue p.ulielrs llial ptnbalib represctil mctallie .1. ;w8 DAVIS AND CONfAM Electron microscope studies showed that the brake lining dust consisted main:, of dark masses showing no internal structure and containing no recognizable chrysotile. Ln most cases the only contaminants within the dust masses were smooth crystalline particles usually about 0.05 of a micron in diameter (Figs. 5 and 6). Since metallic iron was not present in the original dust sample these wore probably particles of metallic aluminum added to the original brake lining mix ture as an abrasive. Very occasionally dust particles were found that consisted of a lighter and more homogenous material which contained in addition to the crystalline particles already mentioned elongated crystals of apparently normal chrysotile (Fig. 7). This type of dust, however, made up much less than 1%- of the total. Small dust particles were phagocytosed in large numbers by single macrophages but as with chrysotile heated to high temperatures normal phago some vacuoles were not seen even in specimens examined only a few days after dust injection. The phagosome membrane contracted quickly around these smaller dust particles to form phagosome residual bodies, and these often contained som. ferruginous granules and membrane debris as well as dust (Fig. 6). The Iargi. dust fragments were sometimes found within small giant cells, but as with heate d Fig. 6. A macrophage from a mouse pleural granuloma 4 wk after the injection of brake lining dust. Dust particles (D) are present, with membrane elements and some fine gran ular material, in irregular structures that probably represent phagosome residual bodies. X 46,000. -ommctl mainly > recognizable st nmscs were uneter (Figs. 5 nple these were ake lining mix!iat consisted of icklition to tin* iare.nt.lv normal ess than 1of hers by sincle normal phag.few days after d these smaller :ontained some 0). The larger as with heated CHRYSOTILE ASBESTOS AND AUTOMOBILE BRAKE LINING DUST JJ49 etiun of brake mm fine wun.'sidiml bodies. Fro. 7. A particle of brake lining dust within a mouse macrophage 2 wk after the injection of this dust. In this instance the dust consists of low density amorphous material within which are embedded recognizable crystals of chrysotile asbestos (C) and larger nonfibrous crystalline particles that probably represent metallic aluminum (A). X 63,000. chrysotile many of the larger particles although surrounded by cells, remained extracellular. Dead macrophages containing brake lining dust were found in the granulomas at all stages of the experiment, but they were less frequent than in granulomas produced by chrysotile heated to either 800C or 1000C. DISCUSSION The present study has confirmed previous suggestions that the physical shape of asbestos particles is important in determining the degree of fibrosis produced in t ssues. Several workers including King et ul. (1946), Vorwald et a]. (1951), and Webster (1965) have suggested that long fiber asbestos is more fibrogenic to lung tissue than short fiber dust and Davis in 1972 showed that this also applied to dust injected into the pleural cavity. It was found in this work that long fiber samples of a number of mineral dusts produced much more fibrosis than short fiber dust. In the studies using heated chrysotile it was found that if the dust was heated to no more than 400"C mechanical grinding still produced dust contain ing a large proportion of long fibers. When normal chrysotile is finely ground the original crystal bundles tend to split longitudinally rather than transversely and therefore very long thin fibers are produced. Sieving this dust removes most 350 DAVIS AND CONIAM of these long fibers and it was therefore possible by using both sieved and unsieved dust samples to gain some idea of the effect of fiber length on the tissues. It was found that the long fiber samples produced larger granulomas than the short fiber dust and since the amount of fibrous tissues produced depended on the initial size of the lesions (Davis 1972) long fiber dust was more fibrogenic than short. When dust was heated to 600C the structure of the chrysotile crystals appeared unchanged in the electron microscope but the ground dust samples contained few long fibers. It appeared that the crystals were more brittle than normal chrysotile and that in most cases the bonding between the crystals had been weakened. This was probably associated with the gradual dehydration that occurs when crysotile is heated from between 100C and 600C. Grinding this type of dust tended to break it into individual short crystals and there was little differ ence between sieved and unsieved dusts. Since few long fibers were present both samples resulted in the production of small granulomas and little fibrous tissue was produced compared to the size of the dust dose. When chrysotile was heated to 800C or 1000C the individual crystals fused with one another and the fused bundles remained brittle. When these samples were ground the original largo bundles of chrysotile crystals fractured transversely as one unit and the resulting dust consisted of thick, but short particles. Sieving this dust did not change the particle size distribution and both samples produced very small granulomas resulting in relatively little fibrosis. With the brake lining dust the particle size distribution was very similar to chrysotile heated to 800C or more. Some enlongated particles were present but these always had a very low aspect ratio. Since very little of this dust could be recognized as chrysotile even with the elec tron microscope, the results are not directly comparable with heated chiysotile but very small granulomas were produced in response to brake lining dust and little fibrosis resulted from its injection. The first group of experiments in this study confirmed the suggestions of Jagatie et at (1967) that chrysotile asbestos heated to high temperatures is far more toxic than normal chysotile. It would appear, however, that very high doses are needed to produce a lethal effect and after a 10 mg dose none of the mice showed any signs of disability. This 10 mg dose for mice corresponds to one of approximately 30 g for a 12 stone man and this is an impossible figure for anyone to take in by inhalation over a short period even under the worst industrial conditions. It is very unlikely, therefore, that this toxic effect of heated chrysotile need be considered seriously when dealing with the problems of general atmos phere pollution by this type of asbestos. From the experimental point of view, however, the reasons for this toxicity are of considerable interest. It has been suggested that when chrysotile is heated to more than 800C free silica is liberated in the amorphous state, and several workers have shown that this can be lethal. Gye and Purdy (1922) demonstrated that colloidal silica was rapidly fatal to experimental animals when injected intravenously and Dale and King (1953) produced similar results. These workers found that colloidal silica was ten times more toxic than crystalline quartz in these conditions. Klostcrkotter and Jotten (1953) also showed that molecularly dispersed silica, whether administered intracheally or intraperitoneally, was lethal to rats. It therefore seems possible that the d acute Regan heated i< toxicity l using thi 1970a) j| With he. quite fvi percent i distinct > otilc. \V for some tore coni (Dav* phag macvopli granlntm and brak membrm it would siblc for residual to the ra Marks est degn to maert cause thi able to reduce [ The fi normal 1964 fint from sili in these resulti initial leng ulom dust the to The is cn bake rneso' tumo lining respo CHRYSOTILE ASBESTOS AND AUTOMOBILE BRAKE LINING DUST 351 ( the deaths of animals given large injections of heated chrysotile were due to acute silica poisoning. Regardless of general systemic toxicity, however, it was found that chrysotile heated to 800 C or more and brake lining dust both showed much more cyto toxicity for individual macrophages than normal chrysotile. In a previous study using the intrapleural injection of chrysotile in experimental animals (Davis, 1970a) it was noted that it was very rare to find dead macrophages in the lesions. With heated chrysotile or brake lining dust, however, macrophage death was quite frequent although dead macrophages never made up more than a few percent of the total. Even the living macrophages from these granulomas showed distinct differences from those in lesions produced in response to normal chrysotile. When normal chrysotile is phagocytosed by macrophages it is retained for some time in a large phagosome vacuole before the membrane of this struc ture contracts around the dust particle. It was shown in organ culture experiments (Davis, 1968) that this process could be completed in as little as 12 hr after phagocytosis but this must be rare and after intrapleural injection of chrysotile macrophages with vacuolated dust containing phagosomes could be found in the granluomas for several weeks. With chrysotile heated to 800C or more, however, and brake lining dust, vacuolated phagosomes were never seen and the phagosome membrane was always closely opposed to the dust particle. From these findings it would appear that chemical differences between the dust samples are respon sible for quicker contraction of the phagosome membrane and the formation of residual bodies, and it is interesting to consider whether this property was related to the raised content of amorphus silica in the heated samples. Marks et al. (1956) showed that whereas amorphous silica produces the great est degree of systemic toxicity, the crystalline varieties are more rapidly cytotoxic to macrophages. It may be, however, that while noncrystalline silica is unable to cause the rupture of phagosome membranes (Allison et al., 1965) it is nonetheless able to increase the permeability of most phagosome membranes enough to reduce phagosome contraction time. The finding that chrysotile heated to more than 600C is more cytotoxic than normal chrysotile yet results in less fibrosis appears to contradict Heppleston's 1964 finding that macrophage death is an essential precursor to fibrosis resulting from silica dust. In fact, however, two conflicting factors are probably involved in these findings. It has been found (Davis, 1972) that the amount of fibrosis resulting from the injection of mineral dusts closely corresponds to the size of the initial cellular response to the dust and this in turn depends largely on the fiber length. Since the heated chrysotile contained no long fibers the resulting gran ulomas were very small and although macrophage death resulting from increased dust cytotoxicity probably accelerated the production of collagen in these lesions the total collagen production was never very large. The finding that automobile brake lining dust has a very low fibrogenicity is encouraging, but the initial concern over the pollution of the atmosphere by brake lining dust concerned its possible association with the production of mesotheliomas, and the present study has not settled this point with certainty. No tumors were found in animals injected with cither heated chrysotile or brake lining dust but the experiments were mainly concerned with the initial tissue response to the dust and were terminated after 1 yr. ft is now known, however, 352 DAVIS AND CONIAM that whereas intrapleural injection of asbestos in rats produces large numbers ot mesotheliomas (Wagner, 1960) few are produced by similar injection in mice (Davis, unpublished) and so no significant results would have been produced from prolonged experiments of this type using brake lining dust. It is planned to undertake new series of experiments that will test the long-term effects of brake lining dust in conditions where normal chrysotile is known to produce tumors. At this point, however, it may be pointed out that electron microscope studies of lesions produced in response to brake lining dust have shown that this dust contains extremely little recognizable chrysotile and what there is is mostly embedded in the plastic bonding material. No computation of urban atmospheric pollution by chrysotile from this source can therefore be made by calculating brake drum wear. REFERENCES Allison, A. C., Harincton, J. S. Birbeck, M., and Nash, T. (1985). Observations on iia> cytotoxic action of silica on macrophages. Inhaled Particles and Vapours, II. In "Proceedings of the British Occupational Hygiene Society Symposium, Cambridge" (Davies, C. N., ed.), pp. 121-131. Published by Pergamon Press. Anjilvel, L., and Thurlbeck, W. M. (1960). The incidence of asbestos bodies in the lungs at random necropsies in Montreal. Can. Med. Asso. J, 95, 1179-1182. Ball, M. C., and Taylob, H. F. W. (1963). The dehydration of chrysotile in air and under hydrothermal conditions. Mineralog. Mag. 33, 467-482. Brindley, G. W., and Hayami, R, (1905). Mechanism of formation of forsterite and enstatite from serpentine. Minetelog. Mag. 35, 189-195, Catjna, D., Totten, R. S., and Gross, P. (1965). Asbestos bodies in human lungs at autopsy. JAMA 192. 371-373. Dale, J. C., and King, E. J. (1953). Acute toxicity of mineral dusts. AMA Arch. Industr. Hyg. Occup. Med. 7, 478-483. Davis, J. M. G. (1968). The effects of chrysotile asbestos dust on lung macrophages main tained in organ culture. Brit. J. Exp. Pathol. 48, 379-385. Davis, J. M. G. (1970a). The long term fibrogenic effects of chrysotile and crocidolite asbestos dust injected into the pleural cavity of experimental animals. Brit. J. Exp. Pathol. 51, 617-627. Davis, J. M. G. (1970b). Further observations on the ultrastructure and chemistry of the formation of asbestos bodies. Exp. Mol. Pathol. 13, 346-358. Davis, J. M. G. (1972). The fibrogenic effects of mineral dusts injected into the pleural cavity of mice. Brit. ]. Exp. Pathol S3, 190-201. Davis, J. M. G., Gross, P., and De Trevillb, B. T. P. (1970). Ferruginous bodies in guinea pigs. Arch. Pathol. 89, 364-373. Ericsson, J. L. E., and Trump, B. F. (1964). Electron microscopic studies of the epithelium of the proximal tubule of the rat kidney. Lab. Invest. 18, 1427-1458. Gye, W. E., and Purdy, W. J. (1924). Poisonous properties of colloidal silica. Brit. J. Exp, Pathol. 5, 238-250. Heppleston, A. G. (1967). Activity of a macrophage factor in collagen formation by silica. Nature (London) 214, 521-522. Jacatic, J., Hubnitz, M. E., Godwin, M. C., and Weiskopf, R. W. (1967), Tissue response to intraperitoneal asbestos with preliminary report of acute toxicity of heat treated asbestos in mice. Emrfron. Res. 1, 217-230. Kinc, E. J., Clecg, J. W., and Rae, V. M. (1946). Effect of asbestos, and of asbestos ami aluminum on the lungs of rabbits. Thorax 1, 188-197. Klosterxotteh, W., and Jotten, 1C. W. (1953). Die Wirkung Verschiedener Kiesselsaureformen in Tierexperiment. Arch. Jff/g. 137, 625-638. CHRYSOTILE ASBESTOS AND AUTOMOBILE BRAKE LINING DUST 353 Lancer, A. M., Rubin, I., and Selikoff, I. (1970), Electron-microscope analysis of asbestos 1 bodies. In "Pneumoconiosis" (Shapiro, H. A., Ed.), pp. 57-89. Proceedings of the Inter national Conference, Johannesburg, Oxford University Press. Lynch, J. R, (1968). Brake lining decomposition products. ]. Air Pollution Control Asso. 18, 824-826. Marks, J. Mason, M,, and N'acelschmidt, G. (1856). A study of dust toxicity using a quanti tative tissue culture technique. Brit. J. Ind. Med. 13, 187-191. Meurma.v, L. (1966). Asbestos bodies and pleural plaques in a Finnish series of autopsy cases. Aeta. Pathol. Microbiol. Scand. Suppl. 181. Pooley, F. D., Oldham, P. D., Chanc Hyum Um, and Wacneh, J. C. (1970). The detection of asbestos in tissues. In "Pneumoconiosis" (Shapiro, A. A., ed.), pp. 108-116. Proceedings of the International Conference, Johannesburg, by Oxford University Press. Rickahdx, A. L. (1967). An investigation of the tubular structure of chrysotile asbestos by electron-microscopy and electron diffraction. Paper presented at a conference on "The Physics and Chemistry of Asbestos Minerals," Oxford 1967. Conference report circulated to delegates only. Thomson, J. G., Kaschula, R. O. C., and MacDonald, R. R. (1963). Asbestos as a modern urban hazard. S. Afr. Med. ]. 37, 77-81. Thomson, J. G., and Graves, W. M. (1966). Asbestos as an Urban Air Contaminant. Arch. Pathol. 81, 458-464. Utidjian, M. D., Gross, P., and De ThevjLLe, R. T. P. (1968). Ferruginous bodies in human lungs: Prevalence at random autopsies. Arch. Environ. Health 17, 327-333. Vohwald, A. J., Duhkan, T. M., and Pratt, P. V. (1951). Experimental studies of Asbestosis. AMA Arch. Ind. Hyg. Occ. Med. 3, 1--43. Waoner, C., Sleccs, C. A., and Mahchand, P.'(I960). Brit. J. Ind. Med. 25, 293. Wacner, J. C., and Berry, C. (1969). Mesotheliomas in rats following inoculation with asbestos. Brit. ]. Cancer 89, 567-581. Webster, I. (1965). Asbestosis Research Project, South African. Pneumoconiosis Review, pp. 25-27.