Document zdKDE6Qz01Yw60XKE0rVbKw3B

hysiology Is With sfcin * * 8S. {X983) Fan. 'teSkin. Phiiaf 55* tiS.J. (T985J d suppression hac reserpinc tian indepen / Med. Delabre M. aJ studies of 'hyum menta<ity T celts. /. enkam D.A. rapotvencs of usilology 89. PATRICK C.H. Iiruination of. rMaUnCeOoUuSs CLiatt- Sr. /. ccp. Path. (1988) 69. 717-737 Comparisons of the pathogenicity of long and short fibres of chrysotile asbestos in rats J.M.G.V Davis* and A.D. Jones " Pathology Branch and f Physics Branch, Institute of Occupational Medicine. S Roxburgh Place. Edinburgh ; P.m $SU. UK Rcceivcifjfor^Vabiicanun 12 February 1988 Accep`tS3'foF`'piiblieatioa ro May 1988 Summary. Long-term inhalation and intraperitoneal injection studies were undertaken with laboratory rat* treated with a specially prepared short-fibre sample of Canadian chiysotile asbestos. This was compared, ac an equal mass dose, to dust generated from the same chrysotile batch so as to contain the highest possible number oflong fibres. The long-fibre cloud contained roughly five times more fibres^ 5/im hi length, as seen by phase contrast optical microscopy (PCOM). For increasing lengths' the ratio between the dust clouds increased progressively, reaching.over 8o:.J. for fibres >30;im in length.,R'ats treated with long-fibre r' dermairospores to ' it invw Derm. A fJ.J 1iTmonthdusting" periodT tfireetUncs moreshorc chrysotile than longhad been retained in the V^atTf mass dose of 2^rm> of dust.'both long miti'siiort'clifvs'otile produced mesotheliomaT in morrtEaiiqo% of rats. At a dose level of 2.5ms of dust the short-fibre chrysotile produced / -v | mesotheliomas in only one-third as many rats as the long-fibre dust which still produced i mesotheliomas ii^more th^n_90% of animals injected. lAt a dose level of o.ismg of dust, the ,} short-fibre chrysotile produced no mesotheliomas' while the long-fibre chrysotile still produced y these tumours in 66% of rats. In the two highest doses, where short-fibre chrysotile produced ) mesotheliomas, the mean tumour induction period was significantly longer than for tumours v -p--r-od_uHced by long ch^.r.y..s.otile. ............. _ ... ........ .. . . Keywords: asbestos, dimensions, carcinogenicity Since the dangers of industrial exposure to asbestos were first realized much experimen tal work has been undertaken to determine which parameters of any dust cloud arc most important in disease production. One of the earliest factors to be examined was fibre length and King at al. (1946) administered chrysotile samples cut on a special micro tome to rabbits by intratracheal injection. Animals injected with dust of fibre length approximately cither r 5 or 2.5 ym. developed nodular peribronchial fibrosis with some Correspondence: Dr f.M.G. Davis. Pathology Bcaneh. Institute of Occupational Medicine. 8 Roxburgh Place. Edinburgh EH8 9SU. CJK. 717 * i}. in f 3 x<i} I jt 8 J.M.G. Davis and A.D. Jones progressive fibrosis of the alveolar walls, but lliese changes were much more marked in the animals receiving the long-fibre dust Similar intratracheal injection studies, using guinea-pigs, were reported by Vorwald el al. (1951). This group injected long' and 'short' samples of the main asbestos varieties, chrysotile, amosite and crocidoiite as well as anthophyllite. tremolite and bracite. Only the short-fibre- `.samples- of chrysotile and amosite wouldShow be-considered really short with almost'all-Cb'res < 3/an in length ' and while these produced little tissue reac tion. long-fibre preparations produced dis tinct pulmonary fibrosis. Klosterkotter (1968) extended these studies by using both \thc intratracheal and intxaperitoneal injec tion of chrysotile and crocidoiite ground to an average fibre length of <s/im. They found that these samples produced little or no fibrosis in either site, in contrast. longer fibres of the same asbestos types resulted in considerable fibrosis in both areas. Almost identical results were obtained by Hilsehcr ec al. (1.970) using very similar techniques. Davis (T97*) administered many finely ground mineral samples to mice by intraperi toneal injection.'Included In this scries of dusts were a normal long-fibre chrysotile sample and two shbrt-6b.re. chrysotile prep arations. These were .synthetic ehrysottie with a maximum length of i/irn and chryso tile fragmented by ultrasonic treatment in liquid until all fibres were below 1 fim in , length. The short-fibre samples produced almost no tissue reaction while the long-fibre chrysotile produced massive peritoneal fibro1 sis. Wright & Kuschncr (1977) injected samples of crocidoiite. synthetic fluoramphibole and glass fibre intratrocheally into guinea-pigs. For each mineral there was a long-fibre preparation with up to 80% of fibres > io/rm in'length and a short-fibre sample with lew fibres >sum in length. All the long-fibre samples produced widespread pulmonary fibrosis while the short-fibre materials produced little tissue reaction. Stanton & Wrench (1972) demonstrated that, in addition to its effect on fibrosis, fibre length was important in carcinogenesis wi the production of mesotheliomas. Carefu sized samples of a number of mineral fifc types were administered to rats by pleui implantation and it was found that mesoti: lioma production was related to the numfc of tong, thin fibres implanted. Also in 197 Smith ct al. reported that hamsters giv intrapleural injections of crocidoiite grou; to a fibre length <l/im did not devel mesotheliomas while those injected with long-fibre preparation of the same mater did. Stanton's work was further develop (Stanton r.t al. 1977. 1981) and it w determined that the most dangerous ttbz were those > 8/xm in length and < 1.5/un diameter. Similar findings were reported Pott & Friedrichs (1972) and Pott et (1976) and the work of this group has be continued until the present time with mai further publications (e-g. Pott (1978.198; Pott & Ziem {1983)). Fibre length also appears to be importau dusts are inhaled and Vorwald ft al. (195 reported that guinea-pigs, rats and mi treated for up to two years with ball-mill chrysotile developed no appreciable pulmo ary reaction and no suggestion of asbestos Complete fibre sizing of the dust clouds w not. undertaken but it was reported th white 90% of the dust was non-fibrous. or third of the fibres present were > iopm length. These amounted to a concentratii of approximately '5 fibres/ml. The importance of long fibres in the pt duction of pulmonary fibrosis followii inhalation was further emphasized by Tii bcell & Skidmore (1968) and Wcbsi (197O). The former authors exposed rats ai guinea-pigs to long and short fibres of am site and obtained a much greater reactii with the long-fibre sample. Webster treat baboons with a finely ground crocidoiite dt with a fibre length below 5/im and obtain only a macrophage reaction in the lungs. Until recently inhalation studies wi short-fibre samples of asbestos had not bei continued for sufficient time to explore tl production of pulmonary tumours and .12/13/96 17:59 g903 877 7558 UTHCT RESEARCH Pathogenicity of long and shorefibres of asbestos in rats 7-19 some cases the dust clouds had been insuffi ciently categorized regarding the numbers of fibres in any size range for definite conclu sions to.be drawn. Davis et al. (1986ft) reported life-time inhalation studies in rats undertaken with a specially prepared shortfibre sample of amosite with almost all fibres < Slim, in length. The effects of this dust were compared to those of a long-fibre dust cloud generated from raw commercial amosite. While the long-fibre dust produced wide spread pulmonary fibrosis and pulmonary tumours or mesotheliomas developed in oncthird of the animals, neither pulmonary tumours nor pulmonary fibrosis developed in animals exposed to the short-fibre dust. Long-term inhalation studies with a shortfibre chrysotile preparation produced by ballmlliing were reported by Platek el al. (xy8 3). j These workers examined rats for up to 24 (months after the start of dust exposure and I monkeys for up to 28 months and reported that the short-fibre chrysotile produced 'no compound related lesions including fibrosis'. The dose level used in these studies was, however, only iing/nri, which is lower than used by most workers examining fibre patho genicity. and it has also been suggested by fencer it al. (1978) that short-fibre chryso tile produced by ball-milling is subject to a level of crystal damage. This is sufficient to make results difficult to interpret in relation to hazards resulting from short fibres pro duced during the manufacture of asbestos products or during the subsequent usage of these materials. lolicoeur et al. (198T) reported on a method of obtaining short-fibre samples of chrysotile by a sedimentation technique which has overcome these difficulties. This preparation contained no fibres >8/zm in length and has been used In a number of short-term studies to examine the early pathogenicity of (lie short chrysotile fibres. This short-fibre chrysotile preparation retained the haemolytic activity normally associated with chrysotile asbestos (Peie Sc Calvert 1983) and was toxic to pulmonary macrophages in short-term in vitro culture (Nadeau el ai 1:986). Lemairc (X985) exam ined cell recruitment into rat lungs following intratracheal injection of the short-fibre chrysotile preparation and UfCC chrysotile (B). While the UTCC material produced a prolonged response with Increased numbers of neutrophils in lavage fluid for up to 14 days following injection, the short-fibre chry sotile produce only an insignificant increase ijrneutrojphjlS; for 1 day following injection, fiothicfij^sotfle samples produced increased numbers of pulmonary macrophages which continued for up to 14 days, but the shortfibre dust produced a much lower level of response. In a separate publication Lemaire et al. (1985) reported on changes in pulmon ary morphology produced in rats following the intratracheal injection of the two chryso tile samples. While the UTCC dust produced fibrosis around distorted and obstructed air ways, the short-fibre dust, although causing an accummuiation of inflammatory ceils, produced ,110 fibrosis at all While these studies have suggested that the harmful potential of very short chrysotile is much, less than that of chrysotile dust eqntetalHg long fibres, intratracheal injec tion tjh an 'abnormal technique which bypasses thc normal pulmonary clearance mechanisms. In order to investigate the fibrogeiiic and carcinogenic potential of any fibre sample in normal situations, inhalation studies ate necessary. This paper reports long-term inhalation and injection studies in rats treated either with dust generated from a specially pre pared short-fibre sample of chrysotile or dust from the same type of raw chrysotile gener ated in. such a way as to contain a large proportion of long fibres. Materials and methods The' short-fibre chrysotile sample used in these Studies was supplied by the Institute for Research '.and Development of Asbestos (IRDA) auii was prepared by the differential sedimentation' techniques reported by Jolicoeur 31. (1981). Chemical, X-ray dlfirac- l 'I : @00. |petc twen 12/13/96 17:59 g903 877 7558 /UTHCT RESEARCH 00 720 f.M.G. Davis and A.D. Jonas tion and differential thermal analyses have confirmed that the chrysotile structure was substantially unchanged by the sedimen tation process. Thejong-fibre dust cloud was generate^lr^ctl^from the same batch ol' chrysotile'(grade ^-30) that had been used to prepard.t^ejshott-fifcresampie. The dust clotids- WCtirgenerated using a Timbrell fibrpui^us.fe'tiispenser (Tlmbrell r.t al. 1970) escribed by Beckett (197s). anc^a flUldiibd bed aerosol (TSI 3400) forre^ecrivity-tbe original long-fibre and short-fibre chrysotile respectively. The fluidized bed aerosol generator was needed to break up particle aggregates in the short- fibre material and to produce a cloud consist ing mainly of separate fibres. The electro static charge carried oh the airborne dust was reduced by exposure to ionizing radia tion from ^-sources. The purpose ofreducing the charge :W3is to. avoid differences in charge level which brtght-havc. arisen from using different types.ofdust dispenser (Johnston et. al. 1987). ThVMSsi of electrostatic charge can substandally.ephnce the deposition of chrysotile flbresTljolicp.^gi.. 1983). The res-pirabTe;dusfjpa.Vs concentrations were measured daily.Jpifug the Caseila MRE XT3A with samples taken for the lull 7 h or dust exposure and on each ofthe 224 days of dusting. The total dust concentrations were measured using open-face somm-diameter filter holders facing downwards with yh- samples collected daily on ten days at a flow rate of 2 l/minute. The fibre number concen trations were assessed-from membrane filter snatch (or very short period) samples col lected using 25inm-diameter Gelman heads (open-faced filters)'on 85 separate days. The samples were examined using phase contrast microscopy (at fioo x magnification) and counts made of fibres With length > 5/^m, aspect ratio greater than 3: r and diameter < 3/mi. Tlic counting procedures are essen tially the same as those used for assessing samples from occupational environments (AlA T979. HSE 1984). The fibre size distributions were assessed also from snatch samples collected on mem brane filters and nuclepore filters for o and scanning electron microscopy re: ively. Phase contrast microscopy was u: produce length distributions for the which constituted the number conce tion, i.e. those fibres longer than 5pm The scanning electron microscope ( was used to produce size distribution fibres longer than 0.4/mi. with mea merits taken from the video screei 10000 x magnification. The mint! length corresponds to a detection Iim 0.13/mi for the diamecers of fibres wit aspect ratio of approximately 3:1. Samples of long and short-fibre chryj used in the intruperitoneal infection sn were collected directly from airborne m: ial in the inhalation chambers by an elu tion process chosen to select the respir fraction of the dust cloud (Bolton n al. 19 For counting and sizing purposes, weis amounts ofdust were resuspended in km volumes of 35% ethanol in distilled w: These suspensions were treated with a 1 period of ultrasonication to disperse f clumps before filtering on nuclepore filt The filters were air dried under contro conditions before being coated with gold examination by scanning electron mic: copy (SEM) aL a magnification of k.) no<; In the inhalation studies two groups oi rats of the AF/HAN strain were exposer clouds of either long or short-fibre chryso at a dose level of iomg/mJ of respirable d for 5 days each week during a period 0 year. The animals were aged 3 months at 1 start of the study. Following the cessation dusting, four animals from each group w< killed to examine early histological chanj in the lung tissue and to determine t pulmonary content of retained chrysoti Four mote animals from each group w< killed 6 months after the end ol' dusting b the remainder were allowed to live out thi Full life span until the experiment was Ism nated when the number of survivors In oi group dropped Co six. This point was reachi at 3T months after the start of dusting win the surviving rats were aged 34 months. Pathogenicity of long and short fibres of asbestos in rats 721 For die injection studies six groups of 24 rats were given a single intraperitoneal injection of either long or short chrysodlc at dose levels of 25. 2.5 or o.25mg. For injec tion the animals were anaesthetized with ether and the dust was suspended in iml of Dulbecco's phosphate buffered saline. All animals in the injection studies were allowed to live out almost all of their full life span and were killed when moribund. Forty-eight rats of the same age as the test groups were maintained in the experimental unit for their full life span as controls. Control animals were kept under identical conditions to the experimental groups except that they were maintained in normal animal holding rooms while animals used in the inhalation studies were in the inhalation chambers. Animals were killed when they showed signs of ill-health until the study was terminated when the seven surviving controls had reached an age of 3 years. The autopsy schedule for animals in the inhalation studies and the control groups included the macroscopic examination of ail major organ systems for signs of abnor mality. and particular attention was paid to the vtscerai and parietal pleural surfaces. Any abnormalities found were taken for histological examination. In addition repre sentative material was routinely taken from the following organs lor histological examin ation whether pathological change was evi dent or not: lungs, mediastinal and hilar lymph nodes, liver, spleen, pancreas, kidney, gastro-intestinal tract, mesenteric lymph nodes, adrenal glands, testis and brain. Tis sue for histological examination was fixed in 10% formal saline and embedded in paraffin wax. The lungs were fixed by instillation prior to excision from the thoracic cavity. Sections of all tissues were routinely stained with H&E. Van Gieseu's collagen stain and Gordon and Sweet's reticuiin stain were used for the study of pulmonary fibrosis. Serial sections were cut from the lungs of all animals autopsied with, groups of sections being mounted at approximately imro-intervals throughout, the block. This procedure ' provided material for detailed examination from six to .eight levels or each lung. Measurement of pulmonary fibrosis was undertaken by similar methods to those previously described by Davis et at. (1,978) except that an electronic image anaiyser (Graphic Information Systems Limited, GDSi) was available for use In conjunction with the light microscope (Davis ct al, 1985). Single lung sections were examined and the sections selected to contain the maximum area of' lung parenchyma. As previously described, interstitial fibrosis was estimated using'a x 2 microscope objective lens and expressed as a percentage of total lung tissue. Feribron.chiolar lesions are more numerous and Smaller and so the lung tissue was scanned with.an eyepiece graticule covering the tissue area of 2.92mm1 and divided into loo squares. A x 4 objective lens was used. Peribronchiolar lesions were recorded as the percentage of squares containing lesions of this type. . The amount of chrysotile retained in the left lung was assayed for rats killed at the end of dust exposure and 6 montlis later. The right lung from these animals was prepared for histology. The lung burdens of chrysotile were recovered by ashing the lungs in oxygeiij-jplasma' in a low-temperature asher (Nanotech Pioti), washing the residue with distilled water and recovering on a filter. The residue material was formed into a potas sium bromide disc and the chrysotile content determined by infrared spectrophotometry (Dodgson & Whittaker f973). Comparisons of the dust contents of left and right lungs have indicated that the ratio of the contents is 0.6: x and this ratio was used to estimate the total lung burdens. For statistical analysis the survival func tions for each experimental group were estimated using the product-limit method (Kaplan & Meier 1958). The survival curves for the different series were tested for homo geneity Using tb.e Gcralized WUeoxon (Breslow) statistic in the statistical package BMDP (Dixon et al. 1983). fn order to compare the overall'mortality of the two groups of ani- 12/13/96 18:01 903 877 7558 UTHCT RESEARCH @00' IT 723 j.M.G. Davis and A.D. Jones mals. the sacrifice of an animat as part, of a planned kill was treated as a censoring event, all other deaths were.treated as responses. Incidence of tumours in particular sites was also compared bywes^acing survival func tions. All death^K^Slh^without tumours (n the site inV^uSdoHJl'ivere recorded as censoring events.' Wfitfeftumours in the specific site Wfire tres^ed;as responses irres pective of whether rhe'jdeath was planned or unplanned." From^.the/in'traperitoiieal injec tion studies,' thciifiihbet oftumours recorded as well as the survival times of the animals were combined using Cox's method to esti mate the relative hazard of the different asbestos types at ail dose levels (Cox 1972. Kalbfleisch & Prentice rggo). Comparisons of levels of pulmonary fibro sis and lung dust burdens were made using the generalized linear models facilities in the statistical package GEN.STAT (Alvey ct al. 19 77). The proportion,: o( lung area with Interstitial flbrosi? wgsjogarithmicaily trans formed before analysis gslng conventional analysis of variance rn^thods. The propor tion of tlie points on the iung with peribron chial fibrosis was,analysts! by logistic regres sion techniques. Lung dust burdens were compared by analysis of variance. Results The target mean respirable dust concentra- tions of TOmg/m* were achieved. The tc dust mass concentrations and fibre num concentrations corresponding to the resj able dust concentration of xomg/m-5 shown in Table r. PCOA/f examination of samples from two dust clouds confirmed chat die `loi fibre' dust cloud did contain many mi fibres of all sizes above 5/im than the 'she fibre' cloud. However, the short-fibre mat ial used had obviously contained more lc fibres than the original preparations repor: on by Jollcoeur et al. (r9Sr) when It m suggested that the sedimentation technlq produced chrysotile samples with all fibi <Sjuu in length. In the present study t ratio of fibre number concentration betwe the tong and short-fibre dust clouds was 5 for all fibres longer than Sfim rising to 20 for fibres longer than zQfim. and So . 1 . fibres longer than 30/rm. The fibre-lens distributions obtained by PCOM for the !o and short-fibre chrysotile clouds arc ilk (rated in Fig. t together with recen; obtained figures for UTCC chrysotile ` which are included for comparison. T. fibre-length and diameter distriburio obtained by SEM for the three chrysot. varieties are illustrated in Figs 2 and 3. The sizings show ChaL the fibre-length disurib tions of the iong-fibre chrysotile cloud gene ated from bulk 4-T-30 material and used the present study were very close to du Table r. Airbonie dust concentrations Type of measurement; * Short-fibre Long-fibre UICC chrysotile chrysotile chrysot Mass of respirable dust mg/m (by Cnsella MRE xt3A) Mass of total dust mg/m `.{by downward-facing open somm-fllter sampling at 21/min) ,' ' Fibrrt aumberobtained.by PCOM fihr&j/ml for fibres longer (^m) than 5 u> zo 30 ro.o X3.9 I I/O 330 33 4 to-o y- xr.81 n. 53m 2360 byo 320 Dirrcivirn 'frurntr 10 /.enow 12/13/98 18:02 Q903 877 7558______________ UTHCT RESEARCH Pathogenicity of long and short fibres of asbestos in rats ' 7*3 99.99 r i 99.9 h 99 98 - 95 go - ccn 80 c33 70 S 60 57a3 50 ai 40 1 eou QV. 30 20 10 O * <0 *s O-i r o.oi L 5 O A _I____I__ I--1-----1---- 1----!------ j10 15 20 25 30 <J0 SO 70 100 150 Length (p.ml 500 F5g. l. Fibre length distributions oflong urul short-fibre chrysorile dust clouds ns welt ys ULCC chrysotile -A`. Fibres sized by phase contrast optical microscopy at a magnification of >; 6ou. Number sized in parentheses: , short fibre (.1.150): *. long fibre (.224S): UlCCA 144-76). 00 12/13/98 18:02 903 877 7558 UTHCT RESEARCH 724 ' it--- J.M.G. Davis and A.D. Jones @00! Length (nm) Fig. 3. Fibre length distributions of long and short-fibre (longer than 0.4.^m; chrysotile dust clouds as w as Ui.CC chrysotile `A'-. Fibres sized by scanning election microscopy at a x to 000. Number sized parentheses: . short fibre (500): 0, long fibre (Goo); A, UTCCA (600). It>i?n?nn?n Tiucncc 1 o corud 12/13/96 18:02 903 877 7558 .,-UTHCT RESEARCH OK Pathogenicity of long and short fibres of asbestos in rats 72 5 l; !H 12/13/96 18:02 0903 877 7558 UTHCT RESEARCH 01 726 J.M.G. Davis and A.D. Jones , clouds recently generated in our laboratory from UICC chrysotile (Davis ctal. 1988). The 6brc diameters ofthe long-fibre mater ial were, however, somewhat lower than those of the -Iji'GG chrysotile. Probably because of this diameter difference the number offibres >jj;tmin length counted by PCOM was higOTi:'`tti;^}ie'long-fibre dust clouds than `fQ^dudsj^fJfiCC chrysotile of the same respfihble dusLjnass. The surviy^of|ntohis'treated with both long and short-fibre rhtysotile and animals in the control group was extremely good with the majority reaching an age of more than 800 days. There were no significant differences between the survival of the two test groups treated with the long and short chrysotile preparations.but the control ani mals did survive' longer on average (P=> 0.002). Both groups of rats dusted with the chi sotile preparations developed the same pi tern of pathological change previou; reported in similar studies from this Instltr (Davis et al. 1:978, 1985, 1986sw:). At t' end of the T 2-month dusting period the ma lesions present were deposits of granulatii tissue around the terminal and respirato bronchioles (pig. 4). This granulation tissi consisted mainly of macrophages and fibr blasts but foreign-body giant cells were al. present. At tz months after the start dusting there was marked reticulin srainir in the peribronchiolar deposits althouf relatively little collagen could be demo: strated by Van Giesen's stain. With increa ing time after dust exposure, however, coll; gen staining became progressively mo: marked and in old animals the lesions cot sisted of mainly acellular fibrous tissue. Bot ` /> V* P''"'' i3r'. CV rr r&esgz mr 2$ VI A Fig. 4. An area ofperibronchiolar fibrosis from the longs ofa rat after ra months exposure to a dust i'Ii.hk of'long chrysotile. X350.- Pathogenicity of long and short fibres of asbestos In rats Table 2. Levels of pulmonary fibrosis produced by long and short-fibre chrysotlle dusts ' 727 ChrymiHle dust Long chrysotlle Short chrysotlle Control group Time after start of exposure (months) Number of rats examined Peribronchiolar fibrosis Interstitial fibrosis 12 4 r4_o (9.4-18.2) 0 18 4 8.9 (6.X-1I.2) Cl 28-30 rx -- 12.6 (6.4-23.9) 3fr~m -`4 U.9-5-7) (x-7-2-3) 0.2 O (0-0.9) 28~30 10 2.4 (O.j-6.71 33 23 0 o.x (0-1.9) The figures in brackets are ranges. Pig. 5. An area of early interstitial fibrosis from the lungs or a rat ra months after the termination of exposure to a dust eloud oflong-fibre chrysotile. The alveolar walls arc thickened anil the epithelial lining appears to consist almost entirely of rounded type II pneuuiocytes. X 35- 12/13/96 18:04 903 877 7558 UTHCT RESEARCH 01 728 J.M.G. Davis and A.D. Jones treatment groups showed deposits of this peribronchiolar fibrosis around many of chc smallest airways at 12 months (Table 2) but levels in the animals treated with long-fibre chrysottle were significantly higher than foe the group treatedXviffitfi^Shdrt-libre mater ial (P< 0.001), By r8 months both treat ment groups^hOWed-a^ifeductioii in peribron.chiolar|fibtosis anaolaken overall, this reduction W$ .afe.^gjatificant (Pco-oor). After 18 month? mom* the start of dusting, widespread alveolar interstitial fibrosis deve loped and this tended to obscure many of the earlier fibrotic deposits. For this reason, estimations of peribronchiolar fibrosis were limited to the first two killing dates. From about 18 months onwards areas of lung tissue in some animals showed a pro gressive thickening of alveolae septa. This thickening of alveolar walls In response 1 treatment with asbestos was described i detail by Davis ct al. (1986c). (n its earlic form it was caused almost, entirely by hype plasia of alveolar lining cells (Fig. 5) but lan there was considerable deposition ofreciculi and eventually collagen in the septal wall As shown in Table 2, areas of alveoli interstitial fibrosis became more widesprea in both treatment groups with increasin time after the end ufthe dusting period. Mea areas of interstitial fibrosis found in thus animals that survived to within 2 mouths ( the filial killing date were 12.6% for tb group treated with long-fibre ebrysotile (1 rats) and 2.4% for those treated with shori fibre chrysotile (10 rats). This difference i significant (Pco'oooi). In many areas c lung tissue the interstitial fibrotic element c [7-v'i F4-ill- ... SC* ^.* ... 4ftl 0 -iT^na Fig. 6. An area of advanced Interstitial fibrosis from the lungs of a rat 17 months after the termination 0 exposure to a dust cloud of'long' fibre chryxotilc. The airspaces which in many cases are enlarged and w longer correspond to the original alveoli an: lined with rounded type II pneumocytes. The airspace wall are greatly thickened with reticulin and collagen which can be demonstrated by special stains, * 350 Pathogenicity of long and short fibres of asbestos in rats 729 these lesions remained predominant throughout the study (Fig. 6) but in others the hyperplasia of alveolar epithelial cells became progressively more marked to pro duce a pattern of adenomatosis (Fig. 7). Some definite adenomas could be seen to have developed from the central regions of these areas and it is likely that this was also the site of origin of some carcinomas although by the time most of these were discovered they were too widespread to be certain. All animals developing pulmonary neoplasms had significant levels of intersti tial fibrosis or adenomatosis in addition. Rats from both the groups treated with cither long-fibre or short-fibre ebrysotile developed some pulmonary neoplasms. However, while animals treated with the long-fibre preparation developed eight ade nomas, eleven carcinomas and cwo pleural mesotheliomas, animals Created with shortfibre chrysotile developed only one adenoma and six carcinomas (Table 3).. These overall tumour numbers are significantly different (P< 0.002). The figures used in Table 3 represent ,, the. most serious pulmonary tumour pSSBgt jit ?my one rat. Two animals treated with tliie long-fibre chrysotile dust Vrai^eniRn^Wjhonary adenomas as well as '^fuiitionajy^arcinoraa. ^M^mSotwo pleural mesotheliomas deyelpped^i^^e present study, many of those rats examined in advanced age that had been treated with either the long or the short-fibre chrysotile preparation showed areas of vehicular pleural metaplasia. This type of lesion was also found in previous long-term inhalation studies with different asbestos preparations (Davis ei ai 1986a). Areas of metaplasia consisted of loose. Fig. 7. An area of advanced interstitial fibrosis from the lungs of a-33-tnonth-old rat treated with long- fibre chrysotile. The eoithcliul lining of the airspaces has become'so prominent that the pattern is one of adenomatosis, x 350. . -t * 12/13/96 18:06____ 0903 877 7558 ________ UTHCT RESEARCH 001 Peritoneal mesotheliomas i l o vTotals li 8 1 Fig, 8, Vesicular hypcrplasia'of the visceral pleural surface of a rat lung x& months after the end i exposure to 'lorigUibre chrysople. Many tissue spaces have been formed in a matrix of loose connecciv tissue and these'are.line^ witfi flattened cells of mesorhellal type, x 350 XT'.-.- fibrous tissue containing" large vesicular spaces lined with flattened cells (Fig. 8) and previous TEM studies had shown that these cells were of mesotheiial type. Occasionally the walls between ^vesicular spaces were so thin that they consisted of two closet; apposed layers of extended and flattened cell with no basement membrane between them Where cells were supported by areas o fibrous tissue a basement membrane wa i DornTwon <i'n/enre- i t < r nnwl 12^13^96^18|06^903 877 75 5 8 UTHCT RESEARCH Pathogenicity, of long and short fibres of asbestos in rats 731 present. While no method has been deve loped for the direct quantification of this pleural metaplasia. Its occurrence is closely related to the presence of advanced intersti tial fibrosis or adenomatosis in the lung tissue and it is particularly common where patches of this type of parenchymal lesion have readied the surface. This means that in cwo chrysotile groups or from the controls. Since most of the animals were allowed to survive until of an advanced age. the major ity of lesions present at autopsy were asso ciated with senescence. /Sparge prediction of the rats in both c_h_ry_^_q_tj_ie_t_re_a_l_h_h_ift_t'_g_ro_up_s_a_n_d__the.._c_o_n_trol the present study it was much more fre quently found in animals created with longfibre ehrysolile than In those treated with the short-fibre preparation. It is uncertain whether or not this vesicu lar pleural metaplasia is a potential precursor of pleural mesotheliomas or not. While almost all the chrysotile'treated animals surviving until the last 6 months of the present study showed some evidence of this type of change, only two definite pleural mesotheliomas were found. Both of these occcciurred in the long-fibre treatment group d neither showed histological patterns liar to the vesicular hyperplasia. One of these mesotheliomas had a papillary struc ture and the other consisted of spindle shaped cells of fibrusarcomalos type. No important differences were noted in the types of non-tieoplasnc disease arising in the extra-thoracic tissues of animals firom the numbers'etf tunrifafs recorded are Illustrated in Table 4 vtfl3*.\in(iividual tumours are recorded although a number of animals had more chan one tumour present ac autopsy. This applied to three animals in the longGbre treatment group, six animals in the short-fibre group and six animals in the control group. There were no significant differences between the overall numbers of tumours found in the three groups of rats. The lung content of chrysotile found in the long and short-fibre treatment groups at the end of(the dustir^ period and 6 months later is illustrated inutile 5. Animals treaced with shorjTfibre.jchryso,tile retained nearly three times^sjjnlcjh dust at the end of the exposure' perin(|as",aQinjJ|ftceated with long fibre. In the subsequent.6^nohths. however. 89% of the short-fibr&i$\fysottlc had been removed from the lung tissue compared to only 54% of Table 4. Non-pulmoiiary tumours found in animals treated with long and short-fibre samples of chrysotile asbestos and in a control group of rats Number of rats examined Organ system Uigestive/poritoncal Urinngenital Endocrine Musculo, skeletal and integumentary KeUeulo-endutl1eii.il/va3cuInr Central nervous system Totals B. benign; M. malignant. Long chrysotile Short chrysotile 4-1 .. 4i B M ,, B M I x; r -- X ; T* s 6 fi *' '-A'J - V . 8 6 3 5" r .-.r -- 6 ^ S*l` -- fi ---- -- 13 tl 10 19 Controls 47 li M. 32 --I 13 4 r5 --- 3 --I L7 r6 Dirrcnjcn tiubnrr 1 r n t\i / @01. 12/13/96 18:07 0903 877 7558 UTHCT RESEARCH 01 J.M.G. Duvis and A.D. font's Toblo Lung dust burdens from the chrysotile inhalation exposures Mean lung burden of asbestos . /yW-'ti'r.it'i v^fy~*^r>vfa*o- rt-fibre chrysotile cxposurV' >Left lung only --"-v 3 days ' (86) 6 months ' Roth lungs** 1019 (-*44) log (26) Long-fibre chrysotile Left lung only Both lungs 135 (75) 62 (15) 35i (195) r6x l 39) * Units are micrograms of dust (u.ri. In brackets) and are mean figures obtained from groups of four rats. ** Figures obtained by calculation from the left lung dust contone. the long-fibre dust. These differences in pul monary deposition-and clearance between the long and short-chrysotile preparations are both significant (P<0.01). Inirapcritoneal irificitdti studies Sizing ofdusts by-mmningelection microscopy. The dust samples of longind short chrysotile used in these injection studieS'wcre collected from die inhalation chmitb'ers used in the inhalation studies by an clutriarion process in order that they would be as comparable as posstble with the dusts inhaled by the rats. It Is recognized, however, that dust samples collected in this way will not be identical in fibre size distribution to airborne dusts and a separate SEM sizing exercise was undertaken with the dusts used for injection. The results obtained are illustrated in Tabic 6 expressed as fibre numbers per milligram of injected dust in each size range. Difficulties in deter mining from these figures an exact number of fibres in any size range needed to produce mesotheliomas are explained in the discus sion section of this paper. The numbers of peritoneal mesotheliomas found following injection of long and shortfibre chrysotile at three dose levels are listed In Table 7 together with the mean tumour Table fi. Numbers of long arid short-fibre chrysotile present in the dust samples used in the Intraperitoneal injection study ' Fibre length Long-fibre chrysotile Short-fibre chrysotile `All fibres >1 >3c.-;:: >)S5r.v. > ; i. 92.7 57-1 21.2 15.Z 10.2 343-0 tbZ.O 24.4 7-4 2.3 . These figures representthe numberofiibres in each size range in each microgram of injected dust. They were obtained using scanning electron microscopy at a magni 4 fication of x 10 000. IPCTPM/Fn TiUKTin'r 10 x.rnpu _12/13/96 18:07 0903 877 7558 UTHCT RESEARCH Pathogenicity of long and short fibres of asbestos in rats Table 7, Mesothelioma production following Injection of long and short chrysotile dust Dost; (mg) 25 2.5 0.25 Long-flbre chrysotile Short-fibre ehrysodle Mean induction $$ Mean induction Animals with period :^prnals-jv'itrir,, .. period mesothelioma (days) mesolh'elfo'mSfA (days) 23 22 361 .. 228 .m'.ljifeM*'r-' 657054 5*1 16 736 Group size 24. rats in ail cases. 733 induction periods for each group. At the highest dose level of *5mg. little difference was found in the numbers of tumours pro duced by the long.' und short-fibre ehrysodle preparations. At dose levels of 2.5 and fo.25mg, however, the short-fibre chrysotile material was significantly less carcinogenic than the long 0.001). The 0.25ms dose of short-fibre chrysotile failed to produce any tumours in the group of 24 rats. Within the experimental groups injected with either the long or short-fibre chrysotile preparation there was evidence of an increased tumour induction period with reducing dose of dust. This phenomenon has previously been reported for both chrysotile and amphibole asbestos (Bolton etai 1984). Even at the 2smg dose level where tumour numbers were similar, the mean tumour induction period was different with the short-fibre preparation taking significantly longer to produce mesotheliomas (Pco.oor). This increased tumour induc tion period with short-fibre chrysotile was also seen at the 2.5mg dose level where it was equally significant and would probably have been seen with the 0.2 smg dose if the short-fibre chrysotile had produced any mesotheliomas at this level. Discussion The present study has confirmed the impor- tancc of.fibre length in the pathogenicity of chryshtireasbeatos although the results are less clear cut than chose recently reported for omosite where. the inhalation of short fibres produced neither pulmonary fibrosis nor tnmours((Davis fft ul. 1986b). The reason for this lies almost certainly in the quality of the short-fibrtficbtysodle preparation that was available for Inhalation studies, bn the origi nal report bfflolicoeur et al. (1981) it was stated thati'the sedimentation techniques which had been developed allowed the pro duction of chrysotile samples with all fibres below 8jan in length. It Was found, however, that while this certainly applied to samples of a few milligrams, it was not practicable for the 1.5kg needed for long-term inhalation studies.' m,the. airborne dust clouds at a respirable mass dose of iomg/m-b the shortfibre cloud had only five times fewer fibres > 5/<in in length as seen by PCOM than the long-fibre sample used for comparison. However.'at greater fibre lengths the differences were mote substantial with 20 times fewer fibres > zoftm in length and 80 times fewer fibres >3ojrm in length. This short-fibre chrysotile sample cer tainly represented the shortest chrysotile that has been available for long-term Inhala tion studies and has permitted a meaningful comparison to be made with long-fibre dust using both inhalation and injection tech niques. Because of fibre levels of over 300 --Dcrcnrcn vn/nrivr' t . rnnui 01 -12/13/96 18:08 Q903 877 75S8 UTHCT RESEARCH @01 Ui 734 .. J.M.G. Davis and A.D. Janes fibres/mi > xo/im in length, it is not surpris ing that the short-fibre dust cloud produced both pulmonary librosis and pulmonary tumours. However, the long-fibre dust with live times more fibres > 5/J.m in length and six times more fibres >iofim in length produced approximately six times more advanced intert|liSal;JJfibrogts in the oldest animals and.|M'b?l|I<;s-Ilore Pulmonary of both long and short inhalation periods'. Ii was suggested by Timbrell (1973) that thi; was due to the curly chrysotile fibres beinf more likely to impact on the upper airway: and to be cleared on the mucociliary escala tor within a few hours. Recent work witi long and short amosite (Davis at al. 1986b and the present study with long and short fibre chrysotile have allowed a better under tumours. Thefe'-figurpJ^aidjindicate that whereas only.the reladl^^png fibres stimu- laSTfiBroSld. Ihe shorle'rTibres (cg/mr-in length) whichjwcre present in much larger numbers in the short-fibre preparation, do have some carcinogenic potential. With an amosild preparation, however, where ex tremely tew fibres were > Sfim in length, neither interstitial fibrosis nor pulmonary tumours developed (Davis et ul. x$86b). It standing of the interrelationships betweei deposition, mechanical clearance from iun tissue, and fibre dissolution. With both amosite and chrysotile. signifi cantly more short-fibre material than Ion Was present in tne lungs at the end of a one year dusting period, confirming that Ion fibres penetrate lts easily mto~rhirrirm parenchyma. The InearTBItures in questio were 3570 and 5640/rg/rac for amosite an has been reported previously from this unit (Davis et al. 1986c) that asbestos-related pulmonary tumours in rats tend to develop from areas ofinterstitial fibrosis via a stage of epithelializahotimf^irspaces which leads to adenomatosisT'/itils Tikelyi therefore, that 350 and 1020/zg/rat for chrysotile. The. figures demonstrate, however, that mac more long-fibre amosite was presenc than tfc very much shorter chrysotile material, ind eating thatfibre penetration was not the on. factor in determining the lung dust burdei Lhese stages of InterstitCaI(fibrogis; and adeno- maipgiJrcpresenc the;dtrecc resp6nsc to the dust, with the progreSS(bn:bf 'precancerous' ISSioffs to definite tumours being a more haphazard process. I f this is so then a fibrous dust that is unable to produce fibrosis, is unlikely to produce tumours. ' An understanding ofthe precise number of During the 6-month period after the end dusting the short-fibre amosite and dny* tile preparations cleared more rapidly chi the long from the lung: at this time the ior and short-fibre amosite burdens were 30S and 4470/jg respectively with the compa able figures for chrysotile being t6i a1 rog/ig. This more rapid clearance of she fibres in any size range uecessary to produce tumours has been an important research goal for many years. The problem is. how ever. beset with technical difficulties, par ticularly where inhalation studies are con cerned. It is relatively easy to determine the numbers and size? offibres'in a dust cloud but the proportioniTreachjng the alveoli and retained in the fung fur long periods will vary with fibre size and the (jhemical durability uf the fibres. A numbef.foY.workers including Wagner et al. (1974), Middleton et al. (1977) and Davis et al (1978) have shown that following exposure to the same respirable dust mass, much .more nmphiboie than chrysotile was present"in rat lungs at the end fibres would be expected ifmechanical rem vai of the short, easily phagoeytosed fibres i pulmonary macrophages was the main pr cess involved. However, the proportion chrysotile removed was much higher th: for the equivalent amosite preparations wi clearance of long and short-fibre amosite 14 and 20% respectively compared to 54 a; 89% for the long and short chrysotile. Tt possible that macrophages phagocytn.se a remove chrysotile fibres more readily th amosite but there is no evidence of this one is more likely that, these findings demi strale that dissolution of chrysotile fibres lung tissue is an important factor in 1 removal of Lhis dust. These observations Aii-Q-g-- g903 377 7558 UTHCT RESEARCH @02 Pathogenicity of long and short fibres of asbestos in rats 7S5 very well with findings from lung tissue obvious that very large numbers of fibres where far more amphibole chan chrysotile is must be injected into the peritoneum of rats found in the lungs of asbestos workers at to produce mesotheliomas. The dose orshortautopsy, even when chrysotile has been fibre chrysotile which produced no tumours known to constitute by far the bulk of dust in 24 rats was calculated to contain inhaled (Pooley 1976. Gylseth ct aL xq83). -8;6 x xo-fibres of all lengths aud 57 x ro6 The combination of variable fibre deposi .fibres > 8fim in length (Stanton ctal. (1977) tion and dissolution makes it impossible at .^calculated chat fibres > 8/mx in length were present to calculate accurately the numbers ^c2$cf|ccarcinogenic). While the number of of fibres in a dust cloud thac will be available was too small for it to be in the lung to cause disease many months certaffi"l|iq^this represents a definite zero later. Injection studies where there are no response, "the injected dose was still very problems of deposition and clearance of dust substantial in terms of fibre number. Even should permit an easier examination of the allowing for the short life span of the rat, this number of fibres necessary to produce gw-m* fn ha stmng "vidftWflJi thnt_flnhnnrLil tumours. With chrysotile, however, the numbersmf fibres must reach the pleura In problem is still complicated by the separation order to make it likely taatmesotHelioma will offibres into individual fibrils which may be a variable process when large masses of dust -arc.iajected. Electron microscope counting of develop. This may be of relevance to current fears of the possible carcinogenic effect of extremely low exposures ofhumans to asbes fibres from a liquid suspension also presents 1technical difficulties. With all asbestos types 1 it is difficult to be certain that the filtration of tos, such as occur in the environments of public buildings and schools. I fibres onto nuclepore filters results in a {Acknowledgements * completely even distribution of the material. With chrysotile this is particularly difficult ' !Jhis studyjwas funded by the Institute for and even In dilute suspensions the fibres tend Researc^knd Development of Asbestos --------to-aggregate- into-dumps,,0n<s method of (IRDA);ICntreal. Canada, avo.id*ing th< i.s i.s ,th1 e. use o. rf a brief period of The .authors -wouldi___- 1*______;------T _h**''****" ml" --L________________t J ll:Tik_e to acknowledge ultrasonlcation of the solution before filt Lhe skilled technical worlrof Mrs K. Niven, ration. a process adopted by most workers Mrs D. Lyster, Mr D. Hay. Mr S. Clarice and examining the fibre content of human lungs. Mlss.G, Liddlc and the secretarial assistauce There is a risk, however, thauhjs .treatment ofNtrs M. Brebner. WILLcause-come splitting of fibres, particu larly chrysotile. and so increase the numbers counted. References In spite of these problems, it was still felt to be a worthwhile exercise to produce SEM "counts from the long and short-fibre chrysotile preparations used In the injection studies At.vev N.Gv, BANftEin C.F.. Baxibr R.I. et at. (1977) GENSTAT. A general statistical pro- ' gramme.-Rothamsted Experimental Station. Asbestos InternationAt. Association (1979) Reference method for the determination of (Table 6). For the reasons stated, they prob ''airtjcTnVe asbestos fibre concentrations at wurJc- ably represent an overestimate of the - plages by light microscopy (membranes filter iTontber of fibresjnjected although with the' " "method).' (Recommended Technical Method separatioaof efirysotile info individual-fibrils. .. ' _no. 1);London: Asbestos International Assoei- in the tissues, they will be an underestimate of the number of active fibre units available adori^ '" ' Beckett S.T. (1975) The generation and evalu ation of OICC asbestos clouds in animal expo to stimulate tumour production. However, sure chambers. Ann. Oa'up. Hyg. tS, 187-198. regardless of che level of accuracy of the Bolton K.R.. Davis Donaloson K. & chrysotile counts obtained by SEM. it is Wright A. (1982) Variations in the carcinoge- ! If 12/13/96 18:09 0903 877 7558 UTHCT RESEARCH 02: 1 736 J.M..G. Davis and A.D. Jones ' ' nfeicy of mineral fibres. Ann. Occup. Hyp. zb. 569-58.*. Bolton R.E., Davis J.M.G.. MillTM b.. Donaldson K. & WRIGHT A. (j^S^ Tha effect of dose of asbestos on'tncsotheiioniH production in the laboratory rar. fa Vt International Pneumoconio sis Conference. Bochum: Bergbau-BeruisgenosNcnschaft. ppjciazH-ito^f^.. Cox D.R. (1972)RegrcssiouillBdels and life tables. /. Stat. Soc. B 34, x87|i:20. Davis J.M.G. j'1972) TK^brogenic effects of mineral dustsifiiectedjtjb the pleural cavity of Hilscher W,, Sethi S. &. Friedrichs KM. (197 2usatnrnenhangc jwischen Asbestnse ui Faserlange. Nulurwissenschajicn 57, 356-55 Johnston A.M.. Vincent MI, & Jones A.D. (198 Electrical charge characteristics of dry aerosi produced by a number of laboratory mechai cal dispensers. Aerosol Scl Tech. 6, rI5-127 Jolicukuk C., Roberge F. & Fortier J. (198 Separation of short fibers from bulk chrysot asbestos Hber materials: analysis and physic chemical characterization. Can. /. C/iem. 5 r 140-2148. mice. 8r.J.j$pfPath.\55, 190-204.. Davis J.M.C..Beckgit, S.X.. Bolton R.E.. Collinus P. & Middleton A.P. (1978) Mass and number of fibres in the pathogenesis of asbestos related lung disease in rats. Br. /. Cuncer 37,673-6S8. Davis f.M.U.. Addison J.. Bolton R.E.. Donaldson K., Jones A.D. & Miller B.G. (1985) Inhalation studies on the affects of tremolite and bmcire dust in rats. Carcinogenesis 6. 66 "-(174. Davis J.M.G.. Addison J.. Bolton R.E., Donaldson K. & Jones A.D. (198601 Inhalation and injec tion studies in .nusmsihg. dust samples from dirysotile asbestos prepared by h wet dispersion process. Br. I. exp. Path. 67. 113-129. Davis I.M.G.. AdDison. f.I.BoijTON R.K.. Donaldson K.. Jones A.D. & SMITH;n^Jiy86b) The patho genicity of long versuiviihorrfibrc samples of amositc asbestos administered to rars hy inha lation and jii.traperitq.rieai injection. Br. /. exp. Putlt. 67. 415-430. vC . Jones A.D.. Johnston A.M. & Vincent J.H. (T98 Static electrification of airborne asbestos du In Aerosols in the Mining and Industrial We. Environments. Eds V.A Msrple & B.Y.H. Li Ann Arbor; Ann Arbor Science, pp. fity-6^ Kaldfleiscii J.D. & Prentice R.L. (rgSo) 7 Statistical Analysis of Failure Time Data. Nt York: Wiley. Karlan EX- & Meier P. (T958) Non-parameo estimation from incomplete observations. Am. Stats. Assoc. 53, 45 7-48 t. King E.T.. Clegg J.W. & Rae V.M. (1946) Effect asbestos and ofasbestos and aluminium, on t lungs of rabbits. Tbnrtix 1, 188-197. KlOSTOUCOTTER W. (X968) Expcriiucnrelle llnti suchungen uber die Bcdeutung der Fawrian fur die Asbest-Fibrose sowin Ilntersuchung uber die Beeinflussung der Fibrose lurch Po vinylpyridln-N-oxid. fn Biologischc Wirkung des Asbestos. Internationale Konferenz Dresde Davis J.M.G.. Bolton R.K-- Brown D. & Tullv H.E. (1986c) Experimental lesions in rats corres ponding to advanced human asbestosis. Exp. Mol. Pathol. 4, 207-221. Davis J.M.G., Bolton R.E.. .Douglas A.N.. Jones A.D. & Smith T. (1988) The effects of electro static charge on tbe pathogenicity of chrysotile asbestos. Br. J. bid. Med- (in press). Dixon W.J.. brown M-O-.-Hnoelman u. Foare J.W.. Hill M.A.. Jbnnwch R.I. & Tovorek J.D. (T983) BMDP Statistical Software. California: University of Califomia'Press. Dodgson J. & Whittaker W, (1973) The determi nation of quartz in respirable dust samples by Infrared spectrophotometry-. Ami. Occup. llyg, 373-387................ ` ` Cylsetii B.. Mowe G. & Wannac A. (1983) Fibre type and concentration-in the lungs of workers Berlin: Deutsches Zentraiinsatut fur Arbeitsu dizin. pp. 47-52. Langer A.M., Wolff M.S.. kohl A.N. & Selikc LJ. (T97.S) Variation of properties of chrysot asbestos subjected to milling. J. Toxicol Envirt Health 4,173-188. Ijimaire I. (1985) Characterization of the bro choalveoiar cellular respouse in experimen asbestonix. Am. Rev. Rcspir. Dis. 131, T34-T4 I.tMAinii 1.. Nadeau D,, Dunnigan J. & Masse (1985) An assessment of the fibrogenio pow dal of very short 4-T-30 chrysodle by intrah cheal instillation in rats. Environ. Res. 36, 31 326. Middleton A.P.. Beckett S.T., Davis JJW (^977) A study of the short-term retention a clearance of inhaled asbestos by rats usl UTCC reference samples, inhaled Particles IV. I In an asbestos cement factory, fir. /. Ind. Med. W.H. Walton. Oxford: Pcrgamon Press, j 40,375-379. ' Health and Safety Executive (1984) Asbestos-- control limits--measurment of airborne dust con centrations and the assessment of control measures. London: Healch and Safety Executive (HSE Guidance note EHtG). 247-2S7Nadeau D., FouQUErre-CotrrtmE L. & Khokami (in press) Mineral fiber toxicity: further e dence for an active involvement of the surfs chemistry. Paper presented at the Intematior Symposium on Respirable Oust in the Mine 12/13/96 18:10 0903 877 7558 ^^-&w?3CTHCT RESEARCH Paihnge.nidty of long and shun fibres of asbestos .in rats 73 7 Industry. Held at Pennsylvania State Univer- Stanton M.F.. Layard M., Tf.ceris A.. Mii.i.v.h fi shy- Symposium report. May M.' & KENT E. (1977) Carcinogenicity of Pesl.teudJ.y.Po., n&.UiCeah ia.veemrtolRyt.ic(aXc9t6-io.3n1) oAfs^hcoomrt apsabreastitvoes 'eXS>tafr*in'b*+trro<o*u`ns\7Mgf l--.aFs..s-.LJa.yNaurtd. C.Ma- ,,nc_Te_ercInesrti,s5A.8.,. .M5.8il'7l-e6r0Ir3S.... fibres on human, rat and sheep erythrocytes. MayM.. Morcan E. & Smith A. (1 98 x) Relation Environ. Res. 31, x64-r75< ,T,, ofj>article dimension to carcinogenicity in Plater s.F.. groth u.h.. Ulrich G.E., Stettt.sr t ' ampKib.iJIe asbestos and ocher fibrous minerals. L.U.. Finnell M.S. St Stall. M. (19a 5) Chronic /. Npjj$Sttr Inst. 67, 965-975. inhalation ofshort asbestos fibres, Fundam Appl. TlMBlKi(jC973) Physical factors as etiological Toxicol. 5, 337-340. " , meahlmTshis. In Biological Effects of Mineral Poolev F.D. (19 76) An examination ofthe fibrousv; , Fiftjft5cientific Publication No. 8. Ed. P, mineral content ofasbescos lung tissue from the'^j -rji-Bo^bvski. Lyon: International Agency for Canadian chrysorile mining industry. Environ?ffZ'Rcs^iuch on Cancer, pp. 295-303. R~es. 12. 281-298. timdrellV. 4 SkidmoreJ.w. (1968) Significance Pott F. & Friedrichs K.H. (1972) Tumoren der of fibre length in experimental axbextoiis. In Rattc nach i.p. Inleknon Faserformiger S'taube. Bialogische Wirkurtgen des Asbestos. Jnteraatio- Nalunoisscnschafte.n 59, 318. imlsKonfsrenz Dresden. Deutsches Zentralin- Potr F.. Friedrichs K.II. & Huth P. (1976) Stitut fur Arbeitsmedian. pp. 52-56. Ergebnisse aus Ticrversuchen sue kanxemge- Timbrell V.: Skidmore J.W.. Hyett A.W. St nen Wirkung faserformiger Scaiibe und Ihre Wagner J.C- (970) Exposure chambers for Denlung irx Hinblick auf die Tumorenstchuiig inhalation experiments with standard refer* beim Menschen. Zentralbl. (Saktcriol. Pimtsil- enw samples of asbestos of the international tent, Infer.rJorixitr. Hyg. Abt. I Oriy. 162, 4.67 . 'Union AgainxtCancer (UICC). ] Aerosol. Sci. 1. 505. 2x5-223. Pott F. (1978) Some aspects on the dosimetry of ^ VonWAinAJ.. DuRKAN T.M. At Pratt P.C. (1951) t_ht_e. c__a__r__c.i_n__o__g___e__n_1i_c___p__o__t_e__n__c___y_or_f _ais__b__e__sat._o__s___a__n__dS o..tkhI__e__r_ ' Expc>$i'bhtal s-4tu..dJiIe__s_-Tof as__b__e__sat_o__saisW. AA MA Arch. fibrous dusK. Staub 38. 486-490. Ind.'jiyrj'.i.Qccup. Med. 3, 1.43. Pott F. ( 98 3) Animal experiments with mineral WAcNisct^p.. Berry G,, Skidmore J.W, & Timbrell fibres, in Short <mtf Thin Mineral Fibres. Proceed V.$04.) The effects of the inhalation of ings from a symposium. Stockholm. Arbeit: ach.": asbaidiin ruts. Br. J. Cancer 29, 252-269. Halsa X9. 133-1651.. -V ^'Webster 1419 70) The pathogenesis ofasbestosis. Pott F. & 2ihm U. (1983) Zur Bedeutung vnn ffjn Pneumoconiosis. .Proceedings of the Eulerna- Lange und Ducchmesser einer Faser fur ihre ' "tionaf Conference, Johannesburg. Ed. H. Sha kunvserogene Potenz. In Fibrous Dusts. Report of piro. Oxford: Oxford University Press, pp. 117- a conference. Strasbourg. Uusscldorf; VDI Ver- 1x9.. lag GmRH. pp. 335-342. Wright G.W. & kuschner M. (1977) The Smith VV.fi,. Hubert D-D. St Badollex M.S. (1972) Biological differences in response to long and influence ofvarying lengths of glass and asbes tos fibres on tissue response in guinea pigs. In short asbestos fibres. Am. ind. Hug. Assoc. ]. 3. Inhaled Particles IV. Ed. W.H. Walton. Oxford: Arfu. Pergamon Press, pp. 455-477. Stanton M.F. & Wrench C. (1972) Mechanisms of mesothelioma induction with asbestos and fibrous glass. /. Nat Cancer Inst. 48, 797-821. @02;