Document 7OgpeggbdbpqR0BBmv8RbRMDE

JOURNAL OF THE AMERICAN COLLEGE OF TONJCOLOCV Volume 12. Number I, 1993 Mir; Ann Lii*r,. Inc., Publish*!* Uterine Changes in Female Mice Following Lifetime Inhalation of Wholly Vaporized Unleaded Gasoline: A Possible Relationship to Observed Liver Tumors? JUDITH A. MACGREGOR.1 WARD R. RICHTER.1 and RENAE I. MAGAW* ABSTRACT An increased incidence of primary hepatocellular tumors was reported in female B6C3Fj mice follow ing lifetime exposure to high levels (2056 ppm) of wholly vaporized unleaded gasoline. This effect was not observed In male mice, nor in females exposed to either 67 or 292 ppm gasoline. No explanation for the sex and dose-specific effect was discussed in the initial report or has been subsequently published. At necropsy, a decreased incidence of enlarged'cystic uteri was also noted among high dose females: however, no correlating histopathologic changes were reported. Because the liver neoplastic response was limited to the females, and because spontaneous liver tumors are known to be influenced by the hormonal environment, we reexamined the uterine tissues microscopically. We observed a dramatic difference among treatment groups in the severity of cystic endometrial hyperplasia. The incidence or moderate cystic endometrial hyperplasia at the terminal sacrifice was 76 Q in the control animils whereas none of the animals at the highest exposure level had lesions that were graded moderate or severe. In addition, uterine atrophy was present in 35fr of mice examined at the high exposure level and was absent in other groups. The changes noted are consistent w ith an altered hormonal influence on the uterus. The finding of a decrease in cystic endometrial hyperplasia and an increase in uterine atrophy in the same dose group in which liver tumors were found suggesLs that these effects may be interrelated. INTRODUCTION asoline is perhaps ere of the most familar substances to people ir. industrialized societies Every year. 100 billion G|uMerts of gasoline are produced in the United States alone Thousands of workers at sen ice stations and gasoline distribution terminals, and millions of consumers create a high potential for human evposure to gasoline. One lifetime study on the health effects of gasoline has been conducted in the mouse. In this study, a compound-related increase in primary hepatocellular tumors was reported for female BoCJF, mice evposed to 2056 ppm wholly vaporized unleaded gasoline (IRDC. 19S3; MjcFjrland et al. I9S2. 19S4j.b). Significant increases were not observed in femj'e mice evposed to two lower treatment levels Among all female mice a: risk, liver tumor incidences were 10. 17. IJ, and 37<1 for the 0. 67. 292. and 2CJ6 ppm treatment groups, respectively. The majority of tumors appeared late in the study without other signs of hepaiotovicity A compound-related increase in hepatocellular tumors was not present in male mice: incidence rates were ,'6. 26. 39. and 3JCe for the 0.67. 292. and 2056 ppm treatment levels, respectively (Magaw et al., 1993) Other treatment-related effects were 'Chevron Rc-carch A. Tc.hni'loyv Co . P O Bov AOS. Richmond. CA. :P-,,--ucw-LSR East VHNi,.ne.*NJ 'ICF kji-er Engineer.. Is:1.' Ham-on Sntet. Tin Floor Oakland. CA. 119 cMA 120069 MACGREGOR ET AL. limited to a decrease in enlarged/cystic uieri observed at gross necropsy in females exposed to the high treatment level. Nororrelating histopatholojic changes were reported for the gross observation and the effect *as not thought to be of significance and was given little attention (IRDC. 1983). No explanations for the increased incidence of liver tumors or the decreased incidence of eslarjed/cystic uteri have been reported- It is unlikely that the carcinogenic response resulted from a direct mutagenic event. Unleaded gasoline has been extensively tested in several mutagenesis assays and the results of these studies support the general conclusion that unleaded gasoline is not mutagenic (Conaway et al.. 1983; EPA, 198": NESCaUM. 1989). In addition, several aspects about the observed tumorigenic effect are unusual and also suggest that the tumors may not have been produced as a result of a direct action of wholly vaporized unleaded gasoline cc hepatocytes. The finding of a hepatotumongenic effect in female, but not male. B6C3F, mice is not common (Haseman and Huff. 1967; Haseman el al.. 1987). The late appearance of the Jiver rumors without any ether signs of hepatotoxicity and the fact that the final incidence observed in the high dose group was similar to that observed in control male mice irTthe same study are also noteworthy. The development of hepatocellular tumors in the mouse is known to be hormonally depended (Agnew and Gardner. 1952; Andervont. 1950; Warwick. 1971). Data and selected tissues from the original study were reviewed to determine whether there was any evidence of compound-related endocrine effects Results of the rev iew are presented in this paper The possibility of a relationship between the observed effects cn die uteri and the development of hepatocellular tumors is also explored. METHODS The Fna! report of the lifetime inhalation study of wholly vaporized unleaded gasoline was thcrauj'.ly reviewed. The International Research and Development Corporation (IRDC. 19S3) conducted a 2-year study ir which croups of tOO male and 100 female B6C3F, mice were exposed to 0. 67. 292. and 2056 ppm gasoline by whole body inhalation. Interim sacrifices were conducted after approximately 3. 6. 12, and 18 months exposure. Terminal sacrifices were conducted after 103-107 weeks for male mice and after 113 weeks for female mice. Selected tissues from the study were obtained and subjected to an independent histopathologic review. The original slides for livers, ovaries, and uten of all female mice were requested from the laboratory. Tie laboratory could not locate the slides and. therefore, paraffin blocks containing tissues collected during the study were obtained instead. Blocks containing ovarian tissue were grossly examined and many were found to be too small to consistently obtain adequate slides. Therefore, effectx on the ovaries could not be assessed. Blocks containing livers and uteri were recut. sections were stained with hematoxylin and eosin. and new slides we;; prepared The liver slides were reviewed to confirm the original diagnoses. The uteri were evaluated histclogcally for the presence of any compound-related effect. Careful attention was paid to the histopathologic observation of uterine endometrial hypepiasia, cystic endometrial hyperplasia, and atrophy because of the decrease in the incidence of enlargel'cystic uteri described as a possible treatment-related effect for female mice in the final study report. These lesions were carefully graded on a four-point scale of severity corresponding to subjective classifications of trace, mild, moderate, and severe. It was not possible to make precise measurements of thickness because the sections were lecuts and because the original embedding wa> not done anticipating the need to make such measurements. Hyperplasia andor cystic hyperplasia were recorded when there was in increase in endom.erral thickness beyond what was considered normal for an adult sexually mature female in anestrus. The grade of Lice was used for the minima! amour.: of hyperplasia that could be subjectively differentiated from normal. A hypeplastif uterus with a grade of severe was approximately four times as large (in cross section) as a uterus with (race hyperplasia. The severe hyperplastic epithelium included multiple cystic glands, varying in size. The difference between a uterus with trace and severe hyperplasia was quite evident by examination of the slides with the naked eye. The grades of nitj and moderate were equally spaced between the two extremes. Examples of uteri w ih trace and severe ystic hyperplasia are prevented in Figures I and 2. respectively. The i%nosi5 of atrophy was used when the uterus was smaller than a normal adult uterus A fc.'-point >cate was als useJ to describe the severity of this lesion The grade of truce represented the slightest decrease in 120 CMA120070 gasoline lterine and liver effects in mice FIG. I. Trace cynic endometrial hyperplasia of the uterus. Mouse exposed to 2056 ppm wholly vaporized unleaded gasoline and sacrificed at terminal sacrifice. Same magnification as Fijure 2 wjth a 20* objective. thickness of the endometrium and myometrium that could be detected. A uterus with moderate atrophy w-aa approximate^ half normal size. Mild was intermediate between trace and moderate. RESULTS Table I indicates that ex sue endometrial hyperplasia wi. presen; in the uterus of most of the female mice examined Cystic hyperplasia was first observed in one animal thai died after 9 months on study and the incidence increased by the J2*mon:h interim sacrifice and remained h;|h throughout the remainder of Ihe study. Figure 3 presents low-power photomicrographs demonstrating cache: the four seventy grades Overall, the incidence of the lesion was slightly decreased only a: the highest treatment level 0056 ppm). However, there was a dramatic trcatmcnt.re.'ated chine: in ieventy of the lesion that w as evident as early as the 12-month sacrifice time and that continued throughout the study. Average severity grades were derived for each treatment group by multiplying the number of animals scored for each seventy by the severity ride, summing the results for ill animals in the treatment group, and dividing by the toial number of animals ir. the group with cystic hyperplasia. Cystic hyperplasia was much more severe in the controls and at the low. treatment level than at the mid and high treatment levels. The severity observed in the control and lew exposure a.-.irrals was representative of the spanianeous picture for "normal" animals of this age, but the altered pattern of seventy observed at the two higher treatment levels was unusual and was considered to be treatment related. When the severity of cystic hyperplasia was evaluated separately for liver tumor-bearing and non-rumor-bearing animals, there was a tender,:-, toward the lesions to be milder :n animals with tumors. Avenge severities were 1.8 FIG. 2. Severe cystic er.dcmrtrij! hyperplasia of the uterus Control mouse sacrificedat terminal sacrifice. Same magnification is Figure I with a 20* objective. 121 CMA 120071 i TaHI-If I. iMUNiM'l! ANI. Sl-VltmV IH 11II-MNI: CyVIH' IIy.'IHII ASIA: AlJ. I;IMAI I: MH1! I:.KAMINU. HV TlMli IN' DlAIII Timt iff ilnith Trvulmcnl level* Nn. etamined -- Cystic hyperplasia tJ Twee Mild Mi .derale .Severe Average scverily* <12 HUHItlll |2 If. 22 t) 1 II II 1) II 34 21 22 II II 0 II II II Ul l) It II II l) II 0 3.11 II It /I-/.I Hmitlit 1 2 .1 II 13 II 1 n II it II 1i4 3 AA 211 3.11 3.11 2.7 "Treatment kveli: I (I ppm; I 67 ppm; J " W? ppm; 4 " 21156 p)>m. `Severity grade*: trace " 1; milil " 2;rnndetalc " 3; KYcrc 4. 4 II 5 1 2 1 II 2.2 /H um, It. Tun'. 1 234 If. 2*3 I'l 1 24 17 23 IH II 322 1 3 f. III (. *1 II . 6 4 2 4 (I It 2.6 2.7 2.2 2.3 Terminal uH rifu e 1 2 34 41 33 37 411 411 32 37 34 11 3 5 If. 6 9 If. IH 31 III 6 0 3 II 0 0 2.9 2.5 2.0 1.5 CMA 120072 GASOLINE UTERINE AND UVER EFFECTS IN MICE FIG. 3. C.V" 'twiron' of uterine hnn> illustratinj the four grades of severiti. All at 30f mjjmfife: an Upper lc.it 'fieri- i-r.-ur ' upper nfht. moderate, group 3 animot: lo^er left. mild, eroup " animal. !ee; nht. trj. e ^roup a join j] 133 CMA 120073 GASOLINE UTERINE AND LIVER EFFECTS IN MICE FIG. 3. C:i"> ^e-horn of uterine hn.-n* illustrating the four grades or severity. All at 20-' wjenififjtron Upper loti. e\ere. group I jnmul. upper rif h:. moderate, group 3 animal, loner left. mild, group ? emmal: loner right. tru.c group a jmrrul. 133 CMA 120074 MACGREGOR ET AL. and ).J in lurnor-bearing animals at the mid and high dose levels, respectively, and 2.3 and 1.9 in non-tumor bearing animals at the same dose levels. Because nearly all of the animals in the study developed cystic h>perplasti. lesions, and the subjective nature of the seventy scoring, it would be difficult to definitively conclude that : difference between the tumor and non-iumo: bearing groups exists. Table 2 presents the incidence and severity of uterine atrophy. Atrophy vus observed almost exclusively in mici exposed to 2Cf6 ppm. The first lesion was observed in an animal exposed to 67ppm that died between 12 and It months on study, but the majority of lesions w.trr observed only at the end of the study. At terminal sacrifice. U/A>' mice exposed to 20i4 ppm were diagnosed with utennc atrophy, while the lesion was absent from control animal and those exposed to 67 or 292 ppm. DISCUSSION Cystic endometrial hyperplasia is a common spontaneous condition in mice (Cosgrove et al.. 1978, Malinin am Malinin. 1972. Sheldon and Greenman. 1979; Ward etal.. 1979). It is generally believed to result from changes ir the relative levels ofesTroger and progesterone (Chandar.d Chauhan. 197J. Lindahl and Willen. 199l;Tang etal. 198-). The inc dence of this lesion in mice vanes by train and tends to increase with age. In many stains, nearly every female older than IS months w-iU have some degree of endometrial hyperplasia (Burek et aj.. I9S2). Exogenous estrogen can induce cystic hyperplasia (Tang et al.. 198-1). Several rodent diets have been shown tr have estrogenic activ ity and this activity has been show-r to increase the size and weight of the mouse uterus (Dram et al.. 19S1: Greenman et al.. 1987; Thigpen et al.. 1987a,b.c). Uterine changes that develop as ar. animal ages may merely be biolcgica! indicators of endocrine function and animals ir. which this aging effect is marked may be mort sensitive indiceors of differences in estrogen levels. Studies in which the uteri are affected to a marked extent may :n effect, serve is biological lest systems that allow any reduction in the normal development of cystic endometria hyperplasia to be easily observed. The decreased severity and incidence of cystic endometrial hyperplasia observed after exposure to wholly vipcr.zed unleaded gasoline are unusual, and indicate that the normal development of the lesions in exposec animals was altered 1; is possible that a component of gasoline, ora group of compounds, interferes with norma hormonal influences on the uterus. Another plausible hypothesis involves modulation of the biotrar.sformation o: endogenous substances at high hydrocarbon exposure levels. Several haloger.ated hydrocarbon* (e.fchlordane, dieldctn. hepiac.nior. polybromir.ated biphenyls) are known to stimulate liver microsomal enzymes and have beer shown to decrease ihe concemration of estrogen in rats and mice and to inhibit estrogen's uterotrophic activity tr exposed animals i Bonhaus et al.. I9S1; Welch et al.. 1979). A review of the toxicologic studies on gasoline produced some supporting evidence for this possibilityincreased liver weights, a gross observation ofien associated with induction of liver enzymes, were observed ir 50-day inhalation studies in rats (1IT. 1985) and monkeys (Kuna and Ulrich. 19SJ. MacFarland. I9S-*). In the IRDC lifetime .nhalation study, liver weights were increased in the high dose female mice as early i*:he 3-month l.-.ier.m sacrifice, mar.v months before effects were observed in the uteri. Table 2. Incidence and Severity of Uterine atrophy: All Female Mice Examined Number examined Alrophy Trace Mild Moderate Severe 0 ppm 105 1 0 1 0 0 Treatment level 67 ppm 292 ppm 97 98 10 00 10 00 00 20S6 ppm 97 16 1 12 1 2 12a CMA 120075 GASOLINE. UTERINE AND LIVER EFFECTS IN MICE E' itfence Tor a possible alteration in the hormonal status of female mice exposed to high levels of unleaded gasoline is interesting because of the possibility of a connection between this effect and the occurrence of liver tumora in these animals. Mouse liver tumors are one of the most controversial endpoints in carcino|enesis testing. Mechanisms through which these tumors develop hive not been fully worked out and remain a topic of considerable debate and research. Many factors can influence the spontaneous incidence or the induced incidence (ECETOC. 1982; Turusov and Taltayama. 1979; Warwick. 1971) of mouse liver tumors and the incidence ti know n to v ary from laboratory to laboratory and even from study to study within the same laboratory (Tarone et al., 1981). The sex of the animal has been shown to have a marked effect. The spontaneous rate of hepatocellular tumors for B6C3F, and other strains of mice is known to be higher in males than females (Agntw and Gardner. 19S2; Andervont. 1950; Haseman et al., 1985; Tarone et al.. 1981; Ward et al., 1979). Haseman et al. (1985) reported 311* spontaneous incidence in B6C3F, males and 85 in females. Reasons for this difference have been explored in many studies and the hormonal environment ofthe animals has been shown to play an important role. Experimentally, ovariectomy and the administration of testosterone have been shown to enhance spontaneous hepatocellular tumor incidences in females, while orchiectomy and estrogen administration reduce the incidence in males. Castrated malts and females generally express similar spontaneous liver tumor rates (Agnew and Gardner, 1952; Andervont. 1950, Warwick, 1971). Ir. :he lifetime inhalation study of wholly vaporiced unleaded gasoline, the liver tumor incidences observed in all groups of female mice other than the high dose group were low and generally within the eipected spontaneous range for female 36C3F, miee (Haseman et al.. 1985: Tarone et al., 1981). The incidence ir. the high dose group was significantly greater than the other groups and was. in fact, similar to incidence rates observed in male miee in the same study (1RDC, 1983. MaeFarland 1982. 1984; MacFarland et al.. 19S~; Magaw et al.. 1993) and to ex peered spontaneous rates for male B6C3F, mice (Haseman et al.. 1985; Tarone et al.. 1981; Ward et al.. 1979). These observations are consistent with the idea that uterine effects may serve as a marker for some endocrine disturbance that altera the incidence of hepatocellular tumors in female miee. REFERENCES ACvW L R C . neGARDNER, W.L'. 11952). The incidence ofspontaneoushepatomas inC5H. C3K(low milk factor), and CSA met and the effect of estrogen and androgen on the occurrence of these tumors in C3H mice Cancer Res 12:757-761. ANC5SV0ST. H B (1950) Studies on the occurrence of spontaneous hepatomas in miee of strains CuH and CBa. J. Natl. Cance-lnsi >1:511-59:. BONHaL'S.D w,. McCORMaCK. K.M.. 8RA5ELTON. w E.. and HOOK. J B. (1981) Effect of poly brominated biphenyls or. hepatic microsomai metabolism of estrogens and uierotrophic action of administered esirogen in rats-J, Toxicol. Environ. Health t:;u 1-150 BL RK. J D . MOLELLO, i A . and W'aRNER. S.D 11982) Selected nonneoplssiic diseases, in' The Muutt in Biomtdieal Reiwrch Vot It Direst) H L. Fuller. JD Small, and J G. Fox (eds.l. Aesdemie Press; New York, pp 415--is. CHAN'O S . and CHaLHA.V H V.S. (I975i Cystic endometrial hyperplasia in Sheep and goau Indian J. Anim. Sei- 4S;71-7S. COhAWAy.CC . SCHREINER. C.A.. and CR ACC. 5.T. 1198:). Mutagenicity eviluaiion of petroleum hydrocarbons, in: ThtTimit'let} ofPr:rvlt*m H'd't/curbpns. Procttthnxi efa Symposium.. Afos, 1957, H.N. MacFarland. C.E. Holdswonh. JA MacGregor. R V> Call, and M L. Kane (eds.): American Petroleum Jnjtiiute. Washington. DC. pp 128-131. COSGROVE, c .. SATTERFIELD. L.C.. BOWLES. N D.. and KLIMA. W C. tl97l>. Diseases of aging uncreated virgin female RFM and B ALB e mice J Gerontol. 33:171-183. DRaNE. H M . HEBERT. C-N . and SAB A. N. (1981) Effect of diet on oesirogen bio-assay in mice Fd Cosmet. Toxicol. 19:93-96. EUROPEAN CHEMICAL INDUSTRY ECOLOGY AND TOXICOLOGY CE.h7R<CSTOCl. (1982). Hepatocarci.iogenesis m Labtnaioty Rodtms Relevance for Man ECETOC Monograph No 4. Brussels. GREENMAN, DL.. FULLERTON. F.R.. SUBER. R.. and FARMER. J. 119)7). EfTects of purified (A1N-76A) and r.jturahif.ffdiem iN'IH-OT) diets on responses of BaLS'c and BbC3F, female mice to estradiol. J. Toxicol Environ Health 22:351-362. H ASM AN". J K . and HL FF. J . (1987] Species correlation in long-term carcinogenicity sruJies Cancer Lett 37:125-132. HASEMAN.) K .HUFF. J E..RAO.G.N .ARNOLD. J E . BOORMAN. C A., and McCONNELL. E E (1985). Neoplasms observed in untreated and com oil gavape control groups of F3au N rats and rC57BL'6N v CuH.HcNiFl lB6C3F() mice. J. Null Cunce-lnsl 75:975-984. 125 CMA120076 MACGREGOR ET AL. HaSEMAN, J K,, HUFF, J E.* ZEICER. ,. and MfCONNELL. E-E (1917). Co^pusii^e tr$ulu of 327 chemical rartinofenieity studies Environ Health Persptci 74:229*235. If7 RSStARCH INSTITUTE. <I9SJ|, Thirteen-week inhalation toaieiiy study of a 0-145" gasoline distillate fraction in rati. Sponsored by Uit American Petroleum Institute. API Doe No 32-32405 February. 1915. INTERNATIONA!, RESEARCH AND DEVELOPMENT CORPORATION (IRDCl- (19S3). Moior Fuel Chronic Inhalation Study. Volt |-6 Sponsored by the American Petroleum Institute Sept. 15, 1913. KUNA. R A . and ULRICH. C E. II9S4) Subchrome inhalation toiicity of two moior fuel* J. Am Coll. Totieol. 3:217-229. LIN'D A HL. B . and WILLEN. R, (1991J Endometrial hyperplasia following eitrojcn utatment without the addition ofgestagen. A follow-up study after withdrawal of estrogens Antioanttr Ret. 11:2071-2074. Mj;FaRLAND. H.N. (1912) Qironie gasoline tot icily. in. The Toutoh1} oj Ftlrohw* Hydroeorbont. Frocet<*ti of 0 Sy-.fOiii.ir, May 1912. H.N. MacFarland. CE- Holdjwonh. J.A. MacGregor, R.W Cali, and M L. Rant (eds.l. American Pt::olcum Institute Washington. DC. pp. 71-16. Mi:FaRLA,ND. H.N. (I9t4). Xenobiotic induced kidney lesions Hydrocarbons. The 90-day and 2-year gasoline studies, in: Adienr/j m Modern Rnvironmniol Toxicology, Vo/. 7. Renal tecu of Petroleum Hydrocarbon!. M.A. MeMman. C.P. Htntttm, J.J Thorpe, and N.X Weaver teds.) Pnnerton Scientific Publishers. Inc,: Princeton. NJ. pp. 31-36MitFARLAND. H.N.. ULRICH. C.R.. HOLDSWORTH. C.E.. KITCHEN. D.N.. HaLLIWELL. W.H.. and BLUM. S.C. t!9$4) A chtonic inhalation Studs with unleaded gasoline sapor. J Am Coll. Toaicol 3:231-241. MaCaw, R i,. RICHTER. w.R . and MacCRIOOR. J.A. (19931. A retvaluaiion of liter tumors in mice eipcted to wholly lapcr.rtd unitided gasoline J Am Coll Totieol 12:195-199. Malinin, c j . and MALININ. LSI. ||972I. Age*relaied ipor.tar.eous uteHne lesions in mice 1 Cerontol 27:193-196. NORTHEAST STaTLS FOR COORDINATED AIR USE MANAGEMENT iNESCaU.M) M9!9j Evaluation of the health tfecu from tiposurt to gasoline and gasoline i spOTS. Air Tone* Commintt, SHELDON. )A C . and OREENMAN. D L. 0979). Spontaneous leuont in control BALB-c female mice. J. Environ Pathol. Teti.-ol 3:1*5-167. TaNC. F.Y . BONF1GLIO. T A., and TANG. L K tl9SJ) Effect of eatie|oen and progesterone on the development of erdenetna.'hyperplas-.a in Lhe Fischer rat Biol Reprod 31:399-413. TAS ONE. R E . CHU. K C . and WaRD. J M. 119SI) 'anabilirv tn the ra:ei of aome common naturally occurring tumors in F,i.-Vct3-- rc-s and iC*`!L6N X C3K HeMFi tBbC.'F,) mice. J Natl Cancer Imi 66:II'S-I1II. THIGPEN. J E LI. L.A . RICHTER. C B-. LEEETK1N. H.. and JAMESON. C.W. t ISETai 7>.t moult b.euiiy for the d: ertior. of es::cgtni; activity in rodent dieu: I A standa'diaed method for conducting the mouse bcassay Lab Anima.' Set. J7.596-60I. THICFEN. J E LI. L A . RICHTER. C B.. LESETKIN. E.H , and JAMESON. C.W i!97fc, pie mouac bioauay forth* d: r::ion of es'.-ojtni: activity in rodent ditti. II Comparative euregtnie activity of purified. certified and standard open and dried ferm.li tevder.: diets Lab Animal Set. 37:602-605. THIG*EN.J E LESETKIN E H..DaAE5.M L.. RICHTER, CB . and CRA* FORD. D (ISJ7c) The mouse bioutay for Ll.t d::rciior.e'eitro|enic activity in rodent dicta 111 Stimulation of uterine weight by deitrovt. lucrcie. and com lurch Lab Am-:l Sci 37.Mt-609. TLRLSOV. V S . and T.AKaYaMa. S (I9r9). Tumors of the live:, in Petholofy o/Tumo'i ofLaboratory Animeb. Vol. II. T.-s'S of tht \luut VS Tvruaov led I International Agency for Research on Cancer. I.ARC Scientific Publication No 23. Lyon pp 193-211. L S ENVIRONMENTAL PROTECTION AGENCY lEPAt. (I9|7i Eveluation of the careinog enieiry of unleaded gaaolinc Cfr.;t of Heal:.', and Environments' Aueiimem. Washington. D C . EP A 6-67-001. U ARD. J.M . GOODMAN. D G . SQUIRE. R A . CHU. K C.. and LINHaRT. M S tl979l Neoplastic and nonneeplaitie leuoni ir. aginf tCS7BL6N * C3H'KeN)F, (BbC3F,| mice J Natl Cancer Imt 63:549-434. W'aR'WCK. C P (19711 Me:ibolum of liver carcinogen! and other factorj influencing liver cancer, in Lntr Cancer. International Agency fe: Reiearch on Cancer. I ARC SCiennfie Publication Vo. I. Lyon, pp 121-137. WELCH. R M . LEVIN. W KUNTZMAN. R.. JaCOSSEN.M . andCOSNEY, A H. D97I) Effect of hileger.aied hydroviton irjertiddcj on -.he tnetibeliim and uieruirophie action oftimogc.il ir. nts and mice Toaicol. Appl, Pharmacol I9:23J-2A6. Addtcai reprint requests to: Or. JuJilh A. S!ccGrcjor Chevron Rcirorch and Ttchnohiy Co. P 0. Sot JOSJ Richmond. CA f-'SOd 126 CMA120077 JOURNAL OF THE AMERICAN COLLEGE OF TOXICOLOGY Volume 11, Number 2,1993 Man Ann Liebcrt, Inc-, Publiihcn A Reexamination of Liver Tumors in Mice Exposed to Wholly Vaporized Unleaded Gasoline RENAE I. MAGAW,' WARD R. RICHTER.1 and JUDiTH A. MACGREGOR5 .ABSTRACT A single lifetime Inhalation study of the effects of wholly vaporized unleaded gasoline has been conducted in the mouse. An elevation of hepatocellular tumors was reported only in females and only at the high treatment level (2056 ppm). Several reports of this study have appeared previously in the scientific literature, Including a comprehensive summary published in this journal. Different incidence rates were reported for mouse liver tumors in these publications without explanations of how they were calculated of why they differed. To clarify the data, we recently examined the final report of the gasoline study, the individual animal data, and slides prepared from liver tissue collected during the study. The previously reported incidence rates did not take into consideration alt animals at risk for tumor development or all animals with hepatocellutar tumors. Revised incidence rates are presented for hepatocellular tumors In all mice on study. These rates are approximately 2(k*25?i lower than previously reported. INTRODUCTION jiftnc.e inhesion study of the effects of wholly vaporized unleaded gasoline on mice was conducted by the AInternational Researeh'and Development Corporation (IRDC. 1983) and sponsored by the American Petroleum Institute. The results of the study indicated that wholly vaporized unleaded gasoline induced a statistically significant compound-related increase in hepatocellular tumors in female mice. No increase wi* observed in male mice. The srudv was originally reported by MaeFarland et al. in this journal (19!4) and elsewhere (1982, 198*), and it was reviewed extensively by the U.S. Environmental Protection Ageney (1957) and other organizations (HE1. 1985; NESCAL'M. 1989). Slightly different incidence figures were presented for mouse liver tumors in the published reports. The differences could n be explained on the basis of information presented in the articles and raised questions about why the published rates differed. The data and the reports in question have been thoroughly rev iewed and the actual liver tumor incidence observed in mice in this study was determined. IRDC STUDY SUMMARY AND PUBLISHED INCIDENCE DATA IRDC (1983) conducted a 2 )ear study in which groups of 100 male and 100 female 86C3F, mice w-ere exposed to 0 67, 292, and 2056 ppm of wholly vaporized unleaded gasoline by whole body inhalation. Interim sacrifices *1CF XA:>er Ergintcft, 1SDO Hi.--ion Sutei. fw, ?P^irT2oLSR. Millstone* Si." *0)twcn Rtbtirch & Tct^notcjy Co , F O. Set Ok. CA. 195 CMA 120078 MACAW ET AL. were conducted after approximately 3,6.12, and 18 month! of exposure. Terminal sacrifices were conducted after 103-10? weeks for male mice and 113 weeks for female mice. The incidence rales for liver tumors have been reportedseveral times, as shown inTable 1 (MaeFarland, l982;MacFarlandetal., I984;NESCAUM. 1989, EPA, 1987). Hepatocellular turnon were defined as adenomas and carcinomas in these publications. LIVER TUMOR INCIDENCE REVIEW In order to determine the cue incidence of liver tumors observed in the study and the reasons why previously reported incidence rates vary, the 1RDC study history and final report were carefully examined. In addition, as independent hisiopathological review of liver sections was conducted- Because the original slides could not be located by IRIX, paraffin blocks containing liver tissue were obtained from the laboratory. The blocks were recut. sections were stained with hematoxylin and eosin, and new slides were prepared. The criteria utilized in the histopathological review for the diagnoses of hepatocellular adenoma and carcinoma were as summarized by Brooks and Roe (1985) and by Popp (1985) for adenomas and carcinomas, respectively. Briefly, adenomas were sharply demarcated and they compressed adjacent liver parenchyma. The cytology of the cells varied, with cytoplasm that was eosinophilic, basophilic, clear, or vacuolated. Carcinomas were chancier* tied by some or all of the following: an increased degree of pleomorphism, greater nuclear/cytoplismic ntio. trabecular plates several cells thick, and increased mitotic activity. Some were large with irregular or invasive growth panerns. The review' of the IRDC final report indicated that the previously reported incidence rates do not take into consideration all mice at risk of tumor development or all animals with hepatocellular tumors. The first hepatocellular tumor was observ ed at the 12-month interim sacrifice and thus all animals alive at that time should be considered at risk for tumor development and should be included in the incidence ntes. The rates repotted by MaeFarland, NESCAUM, and those used by the EPA for estimating the carcinogenic potency of unleaded gasoline correspond to the incidence of hepatocellular turnon in mice that died or were killed during the 18-month to terminal sacrifice period. Animals that were killed or that died on study prior to 18 months were not included. The TaJLE 1. PRIMARY HEfATOCtLLLT_A.lt NEOPLASMS INCIDENCE RATES REPORTED IN Mid TVeermenr Itvtl 0 ppm 67 ppm 292 ppm EPA, pages 5-14* EPA, pages 3-15* MaeFarland* NESCAUM* Revised incidene*f 8/57 (14)* 8/100(8) 8/57 (14) 8/57 (14) 8/79 (10) Female mice 10/52 (19) 10/100(10) 10/52 (19) KV52 (19) 13/76 (17) 13/57 (23) I2'J00(I2) 12/57 (21) 12/57 (21) 12/79 (15) MaeFarland* NESCAUM* Revised ineidencef 23/51 (45) 23/51 (45) 27/75 (36) Male mice 15/42 (36) 15/42 (36) 18/70 (26) 20 44 (45) 20/44 (45) 18/71 (39) `Incidence used to estimate carcinogenic potency in EPA (1987). ^Percentages are reported in parentheses. `Incidence reported in summary section of EPA (1987). ^Incidences reported in MaeFarland (1982, 1984) and MaeFarland et al. (1984), `Incidence reported in NESCAUM (1989). `Corrected incidence determined from Lhis review. 196 20S6 ppm 28/56 (50) 27/100(27) 27/56 (48) 27/56 (48) 29/78 (37) 24/54 (44) 24/J4 (44) 27/77 (35) CMA 120079 UNLEADED GASOLLNE AND MOUSE LIVER TUMORS summary incidence figures repored for female mice by EPA differ. In this ease, the numerators are based on mice that died during the iS-momh to terminal sacrif.ee period, while the denominators represent the total number of mice initially started on test, including those that died prior to 12 months. These animals should not be considered at risk of tumor development and should not be Included in the incidence rates. The study history provides several possible explanations for why the data were reported incorrectly. The histopathologic evaluation of the study was carried out by several pathologists at different laboratories. Separate reports for the different sacrifice groups were prepared and reviewed at different times. The preliminary-draft terminal sacrifice report included data for animals that died on study after the l 8-month interim sacrifice along with the data for animals sacrificed at the end ofthe experiment. It did not include or summarize any data for animals that died prior to these times. This report was reviewed, discussed, and summarized prior to completion of the final IRDC report. When the final report was issued for the entire study many months later, the various reviewers did not go back and incorporate data for animals that died previously. -fr. addition to problems in tabulating the data, there was a change in the terminology for proliferative liver lesions used during the course of the study that resulted is some mice with hepatocellular tumors being excluded from the reported incidence rates. The diagnosis of neoplastic nodule was used at the 12- and 1 S-month interim sacrifices, but not at other time periods. This term was proposed by Squire and Levin (1975) to describe proliferative lesions only in the rat. It was proposed because of the difficulty in differentiating between hyperplastic and neoplastic lesions in the rat on morphologic grounds. The term neoplastic nodule is not typically used to describe mouse tumors, but was used for diagnosing lesions in 'his study for mice at these tw-o interim sacrifices. Animals diagnosed with neoplastic nodules were not included in the incidence figures reported by MacFarland, NE5CAUM. or EPA. Lesions diagnosed as neoplastic nodules by the original pathologist were reevaluated as pan of this review and their morphology was found to be consistent with the diagnosis of hepatocellular adenoma. The morphology of these lesions w as also consistent with that of the lesions diagnosed as hepatocellular adenoma at later times in the study and animals with neoplastic nodules should be included in the tumor incidence figures. REVISED TUMOR INCIDENCE Tables 2 and 3 present the total number of animals with hepatocellular tumors by tumor type and for all tumor types combined for female and male mice, respectively. The overall revised incidence rates presented for all hepatocellular tumors include all animals considered to be at risk for tumor development (i.e., all animals alive at 12 months) and all animals with primary liver tumors, including those originally diagnosed as neoplastic nodules. Table 2. PatMAitr Hepatocellular Neoplasms in Female Mice: Numbers of Animals wtth Tumors and Revised Incidence Treasment level 0 ppm 67 ppm 292 ppm Neoplastic nodule Adenoma Carcinoma Total benign Total benign and malignant With neoplastic nodules Without neoplastic nodules Revised incidence 0 I 7 I 1 1 8/79 (10*) 3 4 6 7 13 10 13/76 (17%) O 4* 9 4 12* 12* 12/79 (15%) *Or.e mouse had an adenoma during the IS- to Ji-monuh study period.' `One mouse had an adenoma and a carcinoma and Is counted only for the most advanced lesion. 2016 ppm 1 8* 21 9 29* 28* 29/78 (J7%) 197 CMA 120080 MACAW ET Al_ Table 3. Palmary Hepatocellular Neoplasms in Male Mice: Numbers op Animals myth Tumors and Revised Incidence Treatment level 0 ppm 67 ppm 792 ppm 2056 ppm Neoplastic nodule Adenoma Carcinoma Total benign Total benign and malignant With neoplastic nobules Without neoplastic nodules Revised incidence - 4 12 12* 16 27* 24 27/75 (36%) 3 4 14 7 18* 15* 18/70 (26%) 5 5 19* 10 28* 23* 28/71 (39%) 2 5 21* 7 27* 25* 27/77 (35%) `One mouse hid a carcinoma at the 12 month sacrifice (0 ppm) or during the 12- to 1!-month period (2056 ppm). * Three mice had a carcinoma during the 12- to 18-month period. `One mouse had a neoplastic nodule and a carcinoma and is counted only for the most advanced lesion. `Three mice had an adenoma and a carcinoma and are counted only for the most advanced lesion. 'One mouse had an adenoma and l carcinoma and is counted only for the most sdvanced lesion. The xv ised incidence rate observed in female mice exposed to 2056 ppm wholly vaporized unleaded gasoline is approximately 25% lower than previously reported. The revised rate is the result of correcting several inconsistencies in tabulating and analyzing data from this study. Inherent problems associated with multiple pathologists conducting the microscopic evaluations further complicated the analysis and should be avoided. The incorrect incidence data have been used to characterize the potential risks associated with human exposure to gasoline by several regulatory groups (EPA. 1987; NE5CAUM, 1989). These risks are based on a treaLT.er.t-relaxd effect observed only at the high dose level and. therefore, the incidence changts identified in this dose gToup could alter the estimated risks. This suggests that these risk assessments should be reevaluated. REFERENCES BROOKS. P.N.. and ROE. FJ C. (1915). D/gesritr System, T.C, Jones, U. Mohr, and R.D. Hunt (cds.). Sprir.ger-Vertag: Berlin, pp. 47-5). HEALTH EFFECTS INSTITUTE. (!983). Gasoline Vapor Exposure and Human Cancer, Evaluation of Eaisting Scientific Information and Recommendations for Future Research. Report of the HEI Health Review Committee. INTERNATIONAL RESEARCH AND DEVELOPMENT CORPORATION. (19S3). Motor Fuel Chronic Inhalation Study. Vots. 1-6. Sponsored by the American Petroleum (ratirule. Sept. 15. 1913. MecFARL-AN'D, H.N. (1912). Chronic gasoline toxicity, in: The Tosieolefy cf Petroleum Hydrocarbons. Proceeding of6 Symposium, May, !9S2. H N. MacFariand, C.E. Holdsworth, J.A. MacGregor, R.W. CaJI, and M.L. Kane (eds.): American Petroleum (nsu'tute; Washington, D C., pp. 75--(6. MaeFARLAND, H N. (1954). Xenobiotie induced kidney lesions: Hydrocarbons. The 9CLday and 2-year gasoline studies, in: Ad*cr.cu in Modern Envicorjntntsl Tciicoicpy. Vo/. 7, Ptruti Effects of Petroleum Hydrocarbons. M.A. MehJmaa, C.F. Hemstreet. J J. Thorpe, and N.K. Weaver (edj.): Princeton Scientific Publishers, Tne.; Princeton, N'J. pp. 51-56. Mj.-FaRL.AND, H.N,, ULRICH. C R.. HOLDSWORTH, C E.. KITCHEN, D.N., HaLLTWELL, W.H.. and BLUM, S.C. (1954). A chronic inhalation study with unleaded gasoline vapor. J. Am. Coll, Toxicol. 3:231 --24*. NORTHEAST STATES FOR COORDINATED AJR USE MANAGEMENT iNESCAL'M) (!9S9). Evaluation of the health effects ftom eiposere to gasoline and gasoline vapors. Ait Toxics Comminee. POPP, j.A, (I9S5) D/'grrr/ir System. T.C. Jones, U. Mohr, and R D. Hunt (eds )* Springer-Vertag Berlin, pp. 39--4J. 198 CMA 120081 unleaded gasoline and mouse liver tumors SQUIRE. R.A., Md LEV/tt. MX (I97SI. Rtpor oft wwtiKop I# ciusificttion of ipooific MpueetHultr kuoot in rut. Cocer Ret. U.S. ENVIRONMENTAL PROTECTION AGENCY. CI9IT). Evtluttion oftht ctrcino^tnicny ofunlttded ttvsGn*. OfEet of HetJtfi end Enviromncncil Attala*nl, Wtifunpon. D C-, PA/60Q/6-t7-0QI. Address reprint requests to: Dr. Judith A. MteCrtfor Chevron Rtttexch and Ttchnelofy Co. r.o. Bex m* Richmond, CA 94SQ4 199 CMA 120082 fundamental and applied toxicology 9,67g-686 (1987) A Two-Generation Reproduction Study with Monochlorobenzene Vapor in Rats1 R. S. NaIR,* J. A. BaRTER, t R- E. SCHROEDER,t A. KNEZEVICH,t AND C. R. STACK *Monsanto Company. 800 North Lindbergh Boulevard, St. Louis. Missouri 63167; tPPG Industries. Inc.. One PPG Place. Pittsburgh. Pennsylvania 152 72;%Bio/dynamics. Inc.. East Millstone, New Jersey 088 73; and ^Chemical Manufacturers Association. 2301M Street, N.W., Washington, D.C. 20037. A Two-Generation Reproduction Study with Monochlorobenzene Vapor in Rats. Nair, R. S.. Barter, J. a., Schroeder, R. E.t Knezevich, a., and Stacx, C. R. (1987). Fundam. Appl Toxicol. 9, 678-686. Groups of 30 male and 30 female Sprague-Dawley CD rats, desig nated as the Fo generation, were exposed to vapor of monochlorobenzene (MCB) at target con centrations of0, 50,1 SO, or 450 ppm for 10 weeks prior to mating and during mating, gestation, and lactation. The progeny of the Ft, generation was designated as the F, generation and groups of 30 male and 30 female F, animals were exposed to the same concentrations of MCB as the F0 parents. Exposure of F, animals was initiated 1 week postweaning and lasted 11 weeks prior to mating and through mating, gestation, and lactation. All F2 pups were observed through weaning at which time they were killed. Observations made during the study included body weights, food consumption, mating and fertility indices, pup and litter survival indices, and histopathology of selected tissues. No mortality was observed during the course of this study. Body weights and food consumption for all treated groups were comparable to controls during the growih period. Maternal body weight data during gestation and lactation were also compara ble between the control and treated groups. Mating and fertility indices for males and females for both generations appeared unaffected by treatment. Pup and litter survival Indices for all treated groups were comparable to those of controls. Hepatocellular hypertrophy and renal changes (tubular dilation with eosinophilic material, interstitial nephritis, and foci of regenera tive epithelium) were observed among Ft, and F, male rats exposed to 150 and 450 ppm MCB. The incidence of bilateral degeneration of the testicular germinal epithelium was increased among F0 adults in the high-concentration group and this lesion was observed only unilaterally in the mid- and high-concentration group F, adults. The relationship ofthese testicular changes to exposure to MCB is unclear because there did not appear to be any increase in intensity and/ or incidence of testicular lesions among F, adults that had longer exposure. In addition, overall mean mating and fertility indices for all groups were comparable for both generations. In sum mary. exposure of rats to MCB at levels of 50. 150, or 450 ppm did not have any adverse effects on reproductive performance or fertility of male and female rats. C is7 Society ofToxmoioiy. Monochlorobenzene (MCB) is used as a sol vent and an intermediate in the chemical in dustry. Under Section 4 of the Toxic Sub stances Control Act, in 1980 the EPA pro posed a test rule for chlorobenzenes which 1 This study was sponsored by the Chlorobenzenes Program Panel (CPP) under the auspices of the Chemical Manufacturers Association. Members of CPP include the Monsanto Co., PPG Industries. Inc,, and Standard Chlorine Chemical Co., l"nc. included reproductive effects testing for MCB. In order to satisfy this need, the indus try has voluntarily conducted a two-genera tion reproduction study with MCB. The re sults of this study are reported below. The re production study was designed to assess the long-term effects of MCB through two gener ations of rats and to determine whether the test substance produces abnormalities in pa rental activities from mating through lacta tion, during pregnancy, or duringgrowth and 0272-0590/87 S3.00 Copvntht ? J 98" b> :He Society of Toxicoioty All njfcts of reproduction in any fonti revr*ed CMA 120083 REPRODUCTION STUDY WITH MONOCHLOROBENZENE 679 development of offspring from conception through maturity. The inhalation route was selected because it is anticipated to be one of the primary routes of potential human expo sure. The dose levels for this study were selected on the basis of the following previously re ported toxicity studies with MCB. Mated Fi scher 344 rats exposed to MCB vapor at 590 ppm daily for 6 hr per day on Days 6 through 15 of gestation exhibited a slight weight loss during the first 2 days of exposure and showed a significant increase in absolute and relative liver weights at study termination. No teratogenic effects were observed up to the highest level tested. 590 ppm (John et al., 1984). In a subchronic inhalation study, rats, rabbits, and guinea pigs were exposed to 200, 475, or 1000 ppm MCB in the air. 7 hr/day, 5 days/week for a total of 32 exposures. Histopathological changes were observed in the liver and kidneys of all species at 475 and 1000 ppm (AIHA. 1964). Sullivan et al. (1983) investigated the excretion of monochlorobenzene following exposure to 100, 300, and 700 ppm of MCB for an 8-hr period. The total amount of MCB excreted (expired air and urine) did not increase dispropor tionately between 100 and 400 ppm, but the total excretion was increased more than two fold when concentration increased from 400 to 700 ppm. These data show that metabolic clearance of MCB from blood becomes satu rated at exposure concentrations of 400 ppm for 8 hr. Sullivan's data also indicated a change in metabolic profile when expo sure concentration increased from 400 to 700 ppm. On the basis of the above data, a concentra tion of 450 ppm was selected as the high-exposure level because it was expected to pro duce liver and kidney changes. In addition, the metabolic profile at this concentration would be similar to that expected for the lower concentration generally experienced in the workplace. This concentration is also six times the current TLV (registered trademark of the American Conference of Governmen tal Industrial Hygienists). The middle and low concentrations selected were 150 and 50 ppm, respectively. METHODS Test material, MCB is a colorless liquid which has a boiling point of 132.1*C. The test material was supplied by Monsanto Company and gas-liquid chromatography analyses of all drums provided to the testing laboratory revealed a punty of 99.9%. Animals and housing conditions Male (129-233 g) and female (131-162 g) Sprague-Dawley-derived rats (CD. registered trademark of Charles River Breeding Laboratories. Portage, MI) were utilized for this study. Animals were 6 weeks of age on the first day of exposure and were acclimated to the housing conditions within the animal quarters (temperature 65-77T. relative humidity 40-70%, 12-hr light/dark cycle) for at least 13 days prior to initiation ofexposure to the test material. Rodent Lab oratory Chow/f002 (registered trademark of Ralston Purina Company. St. Louis. MO) and water were avail able ad libitum during the nonexposure period. Water was also available in the exposure chamber during lacta tion. AH animals were individually housed in suspended stainless steel cages with wvre mesh floors except during lactation. On Day 20 ofgestation, each female's cage was fitted with a stainless steel floor pan and bedding (hard wood shavings) was provided. The stainless steel floor pans were removed on Day 14 of lactation. Fresh bed ding was provided on an as-needed basis. Exposure chamber and operating conditions. For ex posure to the test atmosphere of MCB. animal cages were placed in 6-m3 stainless steel and glass inhalation cham bers. The chambers were operated dynamically at an air flow rate of at least 2140 liters/min This flow rate was calculated to provide one complete air change every 2.8 min. For generating an atmosphere of MCB vapor, the test material was fed into an atomizing nozzle via an FMI fluid metering pump. The vaporized test material was di luted with preconditioned air pnor to entry into the ex posure chamber. For determining analytical concentra tion of MCB, a MIRAN IA organic vapor analyzer wras used. At least four samples were drawn daily at hourly intervals from each exposure chamber A daily nominal concentration was determined by dividing the difference in weight of the generation apparatus and test material before and after exposure by the total volume ofair deliv ered during the day. Experimental design Four groups of 30 male and 30 female rats, designated as the generation, were ex posed to MCB vapor (exposure chamber and operating conditions given above) for 5 hr/dav. 7 days week at tar get concentrations of 0, 50. 150. and 450 ppm MCB for a period of approximately 10 weeks prior to mating. For mating, one male rat. was cohoused wuh one female rat within the same ireatment group at mghi until evidence of mating was observed or until 10 nights of cohabitation elapsed. If mating had noi occurred after 10 days, the CMA 120084 680 NAIR ET AL. female was housed nightly with a different male within the same treatment group (random selection) for a sec ond 10-day inter.al. The presence of copulatory plug and/or the presence of sperm in the vaginal smear was considered to be sufficient evidence that a particular fe male had mated. The day on which such evidence was observed was designated as Day 0 of gestation. Once the female had mated, she was housed individually for the remainder of the gestation penod. Exposure of females and males continued through gestation and lactation. Dams were not exposed from Day 20 of gestation to Day 4 of lactation (day of parturition was considered Day 0 oflactation) in order to reduce the stress to dams at partu rition and to pups resulting from physical removal of the mother for a penod of 6 hr early in lactation. Except for this break, exposure of F0 adults continued daily until termination. The progeny of the F0 generation were designated as the Ft generation. After F, pups were weaned, all F0 ani mals were killed. SomeF, weanlings were used as parents for the succeeding generation. The selection procedure of Ft pups to become the F, adult generation (30 ani mals/sex) was conducted to maximize representation from the number of litters available. The pups selected for the Ft parental generation received the same concen trations of MCB as the F0 parents. Their MCB exposure was initiated 1 week after weaningand was continued for at least 11 weeks prior to mating. Extreme care was taken during F, mating to avoid sibling matings. Exposure of the Ft adults to the test material was continued through mating, gestation, and lactation phases with a break in treatment for the dams late in gestation and early lacta tion similar to that of the F0 dams. Ft adults were killed after the weaning of the F; pups. All F; pups were killed at weaning (Day 21 of lactation) and necropsied. All adult and weanling animals were observed for mor tality and clinical signs of toxicity twice daily Detailed physical examinations were performed weekly. Male body weights were recorded weekly, female body weights were recorded weekly prior to mating, on DaysO, 4, 14. and 20 of gestation, on DaysO. 4. 7, 14, and 21 oflactanon. and weekly again following lactation. Food con sumption was measured weekly during the growth pe nod Litters were examined twice daily for general ap pearance of the pups and for dead pups The number of pups in each litter and pup sex distribution were recorded at birth (Day 0) and Days 4, 7. 14. and 21 of lactation. On Day 4 of lactation all litters with greater than eight pups were culled to that number and the sex distribution within litters was equalized when possible. Individual pup weights were recorded on DaysO. 4. 7, u, and 21 of lactation. Complete gross postmortem examinations were con ducted on all F0 and F, parenis. all F, weanlings not se lected io become parents of the F; generation, and all F; weanlings Liver and brain weights of F0 and F, adults were recorded. Liver, kidneys, pituitary gland, and repro ductive organs (males--tesies. epididymides, seminal vesicle, and prostate, females--vagina, uterus, and ova ries) were examined microscopically for all F0 and F, adult animals in the control and high-dose groups. Liver, kidneys, and testes of male rats in the low and mid-con centration groups were examined histologically, also. For histological evaluation, all tissues were fixed in 10% for malin. Following fixation, the tissues were dehydrated, defatted, and embedded in paraffin. A 6-um section of each tissue (for paired organs one section from the left and right organ was taken) was stained with hematoxylin and eosin and evaluated by a pathologist on a scale of 1 to 5, where 1 is considered to be minimal (51% of tissue on slide affected), 2 is mild (2-10% of tissue on slide affected), 3 is moderate (11-30% of tissue on slide affected), 4 is moderately severe (31-60% of tissue on slide affected), and 5 is severe (>60% of tissue on slide affected). The mating index for males (ratio of number of males for which mating was confirmed in at least one female to total number of males) and females (ratio of number of females showing evidence of mating to total number of females), pregnancy rate (ratio of number pregnant to number mated), and fertility index for males (ratio of number of males impregnating a female to number mat ing) were calculated for each ofthe two matings. Pup sur vival indices (ratio of number of pups alive at a given interval to number of pups alive at the previous interval) at various intervals during lactation were also calculated. Statistical analyses Mean body weights, food con sumption, organ weights, organ to body weight ratios, gestation lengths, and numbers of offspring were evalu ated statistically using the following methods. A Bart lett's test (Snedecor and Cochran. 1972) was performed to determine if groups had equal variances. If variances were equal (p > 0.01), parametric procedures were used to test significance: if not (p S 0.01). nonparametnc pro cedures were used. Parametric methods utilized included one-way analysis of variance (Snedecor and Cochran. 1972) which was followed by a Dunnett test (Dunnett, 1955. 1964) if a significant F-value was obtained. The Kruskal-Wallistest (Hollander and Wolfe. 1973) was the nonparametnc method employed for determining equal ity oftreatment effects, and ifsignificant differences were obtained. Dunn's summed rank test (Hollander and Wolfe. 1973) was used to determine which treatments differed from controls The nonparametnc test for deter mining monoionic trend was the Jonckheere's test (Hol lander and Wolfe. 1973) and the standard linear regres sion was utilized for the parametnc case Pup \iability indices and pup survival indices were analyzed with the litter as the experimental unit and these data were trans formed using the arc sine transformation (Snedecor and Cochran. 1972). Incidence data, which included pregnancy rates, fertil ity indices, mating indices, and Inter survival indices, were analy zed using contingency tables. Firsi, a standard chi-square analysis (Snedecor and Cochran, 1972) was performed to determine if the proportion of incidences CMA 120085 REPRODUCTION STUDY WITH MONOCHLOROBENZENE 681 differed between the groups tested. Next, each treatment group was compared to the control group using a 2 X 2 Fisher exact test (Bradley, 1968), and the significance level was corrected using the Bonrcrroni inequality (Miller. 1966). Also, an Armitage test (Armnage. 1955) for linear trend was performed. RESULTS Cumulative mean ( standard deviation) "analytical concentrations fofthe tow-, mid-, and high-concentration levels were 51 (5), 151 (8), and 451 (25) ppm, respectively, for the F0 generation and 49 (4), 150 ( 11), and 454 (21) ppm, respectively, for the F, generation. These concentrations are compa rable to the target concentration of 50, 150, and 450 ppm. The nominal concentration tended to be approximately 10fc lower than the analytical concentration. No mortality was observed in the treated and control groups during each of the F0 and Ft adult generations. Mean body weights dur ing the growth period for males and females in the F0 and Ft generations were comparable for all groups. Food consumption for all groups was also comparable during the two generations. Reproductive indices for the F0 and F, adult generations are depicted in Table 1. Mating and fertility indices for males and fe males for both generations appeared un affected by treatment. The mean ( standard deviation) number of days for each male and female to mate for the control, low-, mid-, and high-concentration groups in the F0 gen eration was 4.2 4.6, 3.1 2.8, 2.8 2.0, and 3.1 2.9, respectively. In the F, genera tion, the mean ( standard deviation) num ber of days for mating to occur w as 4.9 4.2, 3.2 2.3, 3.1 2.4. and 4.6 4.1 for the control, low-, mid-, and high-concentration groups, respectively. Mean litter and pup survival indices for the Ft and Fj litters are shown in Table 2. In the F, litters, pup and litter survival for all treated groups was comparable to that of controls. In the F; litters, a slight decrease in pup survival index (Days 0-4) was observed in the highconcentration level only. The majority of the decrease in pup survival index was attributed to two dams; one dam lost 12 of 15 pups dur ing lactation Days 0^4, and the other dam lost all 10 pups during the same time period. If this had been a compound-related re sponse, more affected litters would have been expected. Therefore, pup survival indices for control and treated animals for both litters were considered comparable. Table 3 shows the absolute and relative (to body weight) liver weights for the F0 and Fx generations. Significant increases in mean ab solute or relative liver weight were observed in the F0 and F| generations in the 150 and 450 ppm groups. The mean relative liver weight for males (F, generation) in the 50 ppm group was also significantly elevated when compared to that of the control group. At necropsy, an increase in the incidence of small flaccid testes and dilated renal pelvis was observed in the high-dose group for both F0 and Fi adults. The incidence of small flaccid testes among F0 adults was 0, 1, and 3 in the low-, mid-, and high-concentration groups, respectively, while in the F, adults the incidence for this observation was 0. 1, and 5 for the 50, 150, and 450 ppm. respectively. The incidence of dilated renal pelvis ap peared to be elevated in F0 high-dose males and all treated groups among F, adults when compared to that of controls (Table 4). Among F| adults, the increase in incidence was not dose related; therefore the relation ship of this observation to treatment is con sidered to be equivocal. Microscopic changes related to treatment were observed in the liver and kidney. Hepa tocellular hypertrophy was observed in male rats. The incidence of this observation for F0 adults was 0, 0, 5. and 14 in the control, low-, mid-, and high-concentration groups, respec tively, while for F, adults the incidence was 2, 0, 3, and 7 in the 0. 50, 150. and 450 ppm groups, respectively. All affected male rats showedonh minimal to mild hepatocellular hypertrophy. Only one high-concentration- CMA 120086 682 NAIR etal. TABLE 1 A Two-Generation Reproduction Sudy in Rats with Monochlorobenzene Vapor; Mating, Pregnancy, and Fertility' Rates Group (ppm) 0 50 150 450 Mating Females (No. mated '/total) Males (No. mated`/total) No. Percentage No. Percentage Pregnancy, females (No. pregnant/ No. mated) No. Percentage Fertility, males (No. impregnating7 No. mated) No. Percentage 30/30 30/30 29/30 30/30 100.0 100.0 96.7 100.0 F,, mating (forF, liners) 26/30 86.7 27/30 28/30 93.3 30/30 29/30 96.7 27/29 29/30 96.7 26/30 90.0 100.0 93.1 86.7 24/26 28/28 '27/29 26/29 9X3 100.0 93,1 89.7 F, mating (for F2 litters) 0 30/30 50 29/30 150 29/30 450 28/30 100.0 96.7 96.7 93.3 27/30 29/30 29/30 25/30 90.0 96.7 96.7 83.3 24/30 29/29a 23/29 25/28 80.0 100.0 79.3 89.3 21/27 29/29 23/29 22/25 * Number of females showing evidence of mating (plug and/or sperm and/or pregnancy). * Number of males for which mating was confirmed in at least one female. ' Number of males mated with at least one female for which parturition was evident. J Statistically significant at p * 0.05. 77.8 100.0 79.3 88.0 level female rat in the F0 generation showed hepatocellular hypertrophy. Renal degeneration and inflammatory le sions were limited to the male rats. The in cidence of this observation is presented in Table 5, As depicted, the incidence and se verity of renal changes were elevated in the mid- and high-concentration groups. These changes were considered to be related to ex posure to MCB. The incidence of unilateral or bilateral de generation of the germinal epithelium of the testes was increased in the mid- and high-con centration groups in both the F0 and gen erations (Table 6). However, there did not ap pear to be an increase in intensity and/or inci dence of testicular lesions among Fx animals that had longer exposure than F0 animals as a result of their in utero exposure and potential exposure during lactation. Also, all animals in the high-concentration group in the generation showed bilateral lesions while all affected F\ animals showed unilateral lesions only. In order to determine the effect oftestic ular lesions on reproductive performance, the reproductive performance of the animals that showed degeneration of the germinal ep ithelium was reviewed. These data revealed that the affected control animals in both the F0 and Ft generations were successful in sir ing litters, while in the mid-concentration group, all Fq animals and two of three Ft ani mals successfully sired litters. In the highconcentration group, three of six animals affected in each of the two generations were successful in siring litters. DISCUSSION This study was designed to evaluate the effects of monochlorobenzene vapor on the reproductive performance and fertility of rats through two consecutive generations. Target CMA 120087 REPRODUCTION STUDY WITH MONOCHLOROBENZENE 683 Group (ppm) 0 SO ISO 450 TABLE 2 A Two-Generation Reproduction Study in Rats with Monochlorobenzene Vapor: Pup and Litter Survival Indices Pup viability index at birth (live/total bom)* Pup survival indices (days) 0-44 4-21* Litter survival index* No. Percentage 96.9 11.3' 96.1 * 12.0 99.0 2.4 96.4 7.4 F, litters 92.4* 14.6 96.7* 8.3 94.1 * 19.4 90.5 * 20.5 96.7 * 10.2 95.0* 18.5 97.1 * 10.2 93.5 *20.1 27/27 29/30 26/27 24/26 100.0 96.7 96.3 92L3 F; litters 0 98.1 * 4.9 94.4 * 10.9 88.4*28.1 22/24 50 97.2* 5.6 93.2* 19.1 99.1* 3.3 28/29 150 97.9* 3.8 97.6* 5.7 89.1 *28.8 21/23 450 97.4* 4.9 87.7 * 24.6 89.6 * 28.7 22/25 91.7 96.6 91J 88.0 * Total No. oflive pups at Day 0/total No. of pups observed (live plus dead pups) at Day 0 for individual females. 4 Total No. oflive pups at Day 4 (precull)/total No. oflive pups at Day 0 for individual females. ' Total No. oflive pups at Day 21 /total No. oflive pups at Day 4 (postcull) for individual females. 'Total No. oflitters with live pups at weaning (Day 21)/ total No. of litters with live pups at birth. ' Mean * SD. No statistically significant differences from control. TABLE 3 Mean Liv ER WEIGHTS FOR F0 AND F, GENERATIONS OF RaTS EXPOSED TO MONOCHLOROBENZENE VaPOR Males Females Dose level Absolute (g) Relative (g/IOOgbody wt)" Absolute (g) Relative (g/IOOgbody wt)* Control 50 ppm 150 ppm 450 ppm 19.25 2.214 19.03*3.12 21.48 *2.33' 21.84*3.84' F0 generation 3.61 *0.35 3.60 * 0.34 4.06 * 0.30' 4.12*0.61' 11.45* 1.29 12.00* 1.26 12.08* 1.13 13.27 * 1.54' 3.77 * 0.30 3.89*0.23 3.95*0.21* 4.41 *0.33' Control 50 ppm 150 ppm 450 ppm 18.31 2.15 19.46*2.55 21.71 *3.49' 23.35 *4.09' F, generation 3.47*0.32 3.73 *0.36* 4.15 *0.46' 4.44 * 0.40' 12.36*2.27 12.70* 1.58 13.13* 1.55 13.95 * 1.99' Terminal body weight. 4 Mean * SD. ' Significantly different from control p 0.01 using the nonparametric Dunn test. * Significantly different from control p 0.05 using the parametric Dunnett test. ' Significantly different from control p 0.01 using the parametric Dunnett test. 4.16*0.60 4.17*0.35 4.36 0.41 4.59*0.37' CMA 120088 684 F0 adults Ft adults NAIRETAL. table 4 Incidence of Diluted Renal Pelvis in F0 and F, adults Male (MCB. ppm) Female (MCB, ppm) 0 50 150 450 0 50 150 11 2 14 6 5 54 6 4 01 2 450 5 2 " exposure levels were 50, 150, and 450 ppm and the analytical measurements revealed that chamber concentrations were in agree ment with the target exposure concentra tions. No adverse effect of treatment was evident on body weight, food consumption, physical observations, reproductive performance, or fertility in both F0 and Ft generations. Evalu ation of pup delivery, nursing, and weaning data revealed no compound-related changes in either generation. Absolute and/or relative liver weights were significantly elevated in the mid- and highconcentration groups for both Fj and F\ adults. In the low-concentration group, only the relative liver weights of the F male rats were elevated. Histopathological examina tion of the liver revealed hepatocellular hy pertrophy in the centrilobular region in males only in the mid- and high-concentration groups. Hepatocellular hypertrophy may be a reflection of the ability of monochloroben zene to induce hepatic enzymes. In the ab sence of microscopic changes, the biological significance of changes in the liver weight of low-concentration-group males and midand high-concentration-group females is un clear. Exposure of rats to MCB vapor at 150 and 450 ppm also resulted in degenerative and inflammatory lesions in the kidneys. The incidence and severity of the microscopic changes in kidneys increased with dose and are considered to be treatment related. The observations of histological changes in the liver and kidney following exposure to MCB are not unexpected because these observa tions have been reported previously (Hol- table 5 Incidence of Renal Changes among Male Rats Inhaling Monochlorobenzene for Two Generations' Group (ppm) Fa adults Ft adults 0 50 150 450 0 50 150 450 Total No of animals examined U/tubular dilation/eosinophilic matenal B/tubular dilation/eosinophilic matenal L'/chronic interstitial nephntis B/chromc interstitial nephntis B/foci of regenerative epithelium U/foci of regenerative epithelium 30 30 03 01 00 11 01 00 30 2 4 0 7 5 0 30 30 30 30 3 446 15 4 3 8 1 121 9 016 8 1 04 0 00 1 30 6 16 0 11 10 1 S'otc U/, unilateral B/. bilateral. " These lesions were absent in female rats at the same concentrations. CMA 120089 REPRODUCTION STUDY WITH MONOCHLOROBENZENE 685 TABLE 6 Incidence and Severity of Unilateral and Bilateral Degeneration of Germinal Epithelium among Male Rats Exposed to Vapor of Monochlorobenzene Group (ppm) F0 adults F, adults a 50 150 450 0 50 150 450 Total examined JO* 30 Unilateral Minimal 1-- Mild ---- " Moderate --_ Moderately severe -- -- Severe ---- Bilateral Minimal Mild ---- _ Moderate ---- Moderately severe -- -- Severe -- 30 -- 1 -- -- 1 -- -- -- -- -- 30 30 30 30 -- ---- -- -- ---- 1 -- ---- -- -- ---- I -- ---- 1 3 1-- -- _-- ---- 1 ---- -- -- ---- -- 2 ---- -- * Two sections of testes eval-ated histologically and seventy of lesions graded from minimal to severe. 30 1 -- -- 5 -- -- -- -- -- -- Iingsworth et al., 1956: \IHA, 1964; Dalich and Larson, 1985: NTP. 1985). There was also an increase in the incidence of degenerative testicular changes (unilateral or bilateral) in high-dose males (F0 and generations) and the mid-dose males. The relationship between testicular damage and exposure to MCB is unclear because; (i) As shown in Table 1. mean mating, pregnancy, and male fertility indices for both F0 and generations were comparable for all groups, (ii) Overall, the number of males not siring litters in the control, lew-, mid-, and highconcentration groups in the F0 generation was 6, 2, 3. and 4. respectively, while in the F[ generation the incidence was 9.1,7, and 8 for the control, low-, mid-, and high-concen tration groups, respectively. Thus, the num ber of males not siring litters in the F0 and Fi generations was similar. However, if one examines the number of males not siring lit ters among animals with testicular lesions, three of six animals with testicular lesions in the high-dose group in each generation did not sire litters, (iii) The mean number of days for males and females to mate in each group was comparable for all groups in the two gen erations. While there was a dose-related in crease in the incidence of degeneration of the germinal epithelium in both Fq and F\ adults, there was no increase in incidence and/or se verity of lesions in the F\ generation when compared to the F0 generation. Since the length of exposure in the F-, adult is longer as a result of the in utero exposure as well as the potential exposure during lactation, the F[ adults should have been more severely affected. Moreover, previous subchronic studies in rats given MCB at equivalent or higher doses did not show any histologic tes ticular damage (Hollingsworth et al.. 1956; A1HA, 1964; NTP. 1985). In summary, exposure to monochloroben zene at levels of 50, 150. and 450 ppm did not have any adverse effects on reproductive performance or fertility of male and female rats through two consecutive generations. ACKNOWLEDGMENTS The technical support of Dr. Jim Temll and of Cathy Houseman. Ellen Whiting. Gregory McGurr, and Joann B. Brink (all from Bio/dvnamics. Inc.) is gratefully recog- CMA 120090 686 NAIR ET AL. nized. We thank Pat Ploudre (Monsanto Company) for her efforts in typing this manuscript REFERENCES American Industrial Hygiene Association (AIHA) (1964). Hygiene guide series: Chlorobenzene (monochlorobenzene, chlorobenzol, phenyl chloride). Amer. Ind. Hyg. .tisoc J 25,97-99. ArmitaGE, P. (1955). Test for linear trends in propor tions and frequencies. Biometrics 11,375-386. Bradley. J. V. (1968). Distribution-Free Statistical Tests, pp. 195-203. Prentice-Hall, Englewood Cliffs, - NJ. Daljch, G. M., and Larson. R. E. (1985). Temporal and dose-response features of monochlorobenzene hepaioioxicity in rats. Fundam. Appl. Toxicol. 5,105-- 116. Dunnett. C. W. (1955). Multiple comparison proce dure for comparing several treatments with a control. J. Amer. Stat. Assoc. 50,1096-1121. Dunnett, C. W. (1964). New tables for multiple com parisons with a control. Biometrics 20.482-491. Hollander, M., and Wolfe, D. a. (1973). Monparametric Statistical Methods, pp. 114-116 and 131. Wi ley, New York. Hollingsworth, R. L., Rowe, V. K... Oyen, F., Majle, H. R., and Spencer, H. C. (1956). A. M. A. Arch. Ind. Health. 14,138. John, J. A., Hayes, w. c., Hanley, T. R., Jr., John son, K. A., Gushow, T. S., and Rao, K, S. (1984). Inhalation teratology study on monochlorobenzene in rats and rabbits. Toxicol. Appl. Pharmacol. 76, 365373. MILLER, R. G,, Jr. (1966). Simultaneous Statistical In ference. p. 15. McGraw-Hill, New York. National Toxicology Program (1985). Toxicology and Carcinogenicity Studies of Chlorobenzene (CAS Mo. 108-90-7) inF344/N Rats andB6C3F1 Mice (Garage Studies). Technical Report Series, No. 261. Snedecor, G. W., and Cochran, W. G. (1972), Statis tical Methods. 6th ed., pp. 250-253, 277-279, and 296-298. Iowa State Univ. Press, Ames, IA. Sullivan, T. M., Born, G. S., Carlson, G. P.. and Kessler, W. V, (1983). The pharmacokinetics of in haled chlorobenzene in the rat. Toxicol. Appl. Phar macol. 71,194-203. CMA 120091 JOURNAL or THE AMERICA* COLLEGE OF TOXICOLOGY Volume ||. Number 3.199J Mar; Ann Litbrrt, Inc., Publishers A Female Rat Fertility Study with Inhaled Benzene ROBERTA. KUNA.' MARK J. N1COLICH.7 RAYMOND E. SCHROEDER,5 and GEORGE M. RUSCH4 ABSTRACT Pur* benzene was administered to female rats by inhalation at dose levels of 1, JO, 30, and 300 ppm for 6 h'day. The exposure took ptace during a JO-week premating period and during mating, gestation, and lactation periods. There was no effect on female reproductive perfor* mance. A trend toward-reduced body and organ weights was observed in the 21 "day.old pups at the 30 and 300 ppm dose lex els. Except for reduced body and liver weights in the female pups at 300 ppm, these differences were not statistically significant. Kidney to body weight ratios for female pups were statistically higher than control values in the 10, 30, and 300 ppm exposure levels. These organ weights and organ-to-body weight ratio changes appear to have been reflective of differences in body weights. No treatment-related effects were seen in pup survival during lactation or in the gross postmortem evaluation of these pups on Day 21 oflactation. Thus exposure of dams to levels as high as 300 ppm resulted in onlv minimal effects on the pups. INTRODUCTION he potexti al fob reproductive risL due to benzene exposures is not a new concern ofthe occupational health Tprofessional. Hen and .Maak (1938) noted (onadal abnormalities in both male and female rats follow in; an undefined benzene exposure. This earl) research was later confirmed (Wolf et al., 1936) in males of two other species, rabbits and guinea pips. Although much of the original reproductive data were equivocal, the) did stimulate continued interest in and studs of the potential effects of benzene on reproduction. Kuna and Kapp (19SI1 noted fetoioxieity in the form of decreased bod) weights after exposure of dams to 50 ppm benzene during days 6 to 15 of gestation. Similar observations have also been recorded by Green et *1. (1978), Hudak and Ungvary (1978). and Murray et al.( 1979), however, the) were alt associated with much higher chamber concentrations of benzene. The present occupational exposure limits for benzene are: American Council of Governmental Industrial H.v gienists 10 ppm; and Occupational Safety and Health Administration 1 ppm with a short-term exposure limit of f ppm (Gu/dc io Occupational Exposure Limits, 1991). The above data describing reproductive risk were derived from exposures which were markedly in excess of present exposure guidelines. Therefore, the current study was designed to evaluate exposure levels which are routinely encountered in many occupational settings as well as higher exposures which have previously been demonstrated as having a fetotoxic effect. $iudv sponsored b) L S manufacturers of benzene under the auspices of the Chemical Manufacturers Association and American Petroleum Institute (Ref. # BP 3.0). 'National Starch and Chemical Company. Bridgewater, NJ. :mon Biomedical Sciences. Inc,, West Amwell. NJ. 'Bio dv n.mirs. Inc , Eas: Millstone. NJ. 'Allied Signal. Inc., Mom now n. NJ. 275 CMA 120092 KUNA ET AL. MATERIALS AND METHODS . The benzene used w as frcm American Petroleum InsiiiuteOemicil Manufacturers' Association and was shown to be approximately 99 96" pure (ASTM D-1016). Fis e groups of 26 female and 13 pros en fertile male Spragut-Diwlcy rats (Charles River Breeding Laboratories, Wilmington. MA) were used. Animals were housed two per cage during the first week and individually during week 2. In the premating treatment period they were individually caged during exposure and nonexposure periods. For mating, one male and two females were housed per cage. Females were housed individually during gestation, panurition, and with litter djring lactation. All cages were stainless-steel wire mesh. Food (Purina Laboratory Chow) and water were prov-ded ad libitum, except for exposure periods when females were without food. Water analy ses were performed during the premating treatment period and near termination of the study. Assays were performed for benzene, pesticides, metals, and PCBs. All analyses were done according to proposed or approved En\ ironmental Protection Agency methodologies (Federal Register 44:69-*64). Animal rooms were maintained on a 12-h light-dark cycle, and temperature and humidity were monitored. The two laner parameter were also recorded in exposure chambers. Benzene vapor coneemtrioni (0, 1, 10, 30. and 300 ppm) were generated in 1 m' stainless steel and glass chambers using a flash-ev aperation generating system. In this system the benzene was pumped into a heated flask (J0`C for I, 10. and 30 ppm exposures, and 80*C for the 300 ppm exposure), flash-evaporated, and the vapor mixed with the chamber a^ supply. The three-neck flask was heated using a Clas-Col heating mantle and temperature was measured .sing a Glas-Col laboratory pyromeier. Benzene delivery was accomplished using a Sage instrument Model 222 and Ml syringe pump. Modifications in chamber concentration were made by adjusting pump flow rates cr chamber air flow. Air flow- was determined by measuring the change in pressure across a one-inch orifice pla'e using a Dwyer magnehelic pressure gauge. Air flows were maintained between 150 and 250 L'min for the durar :n of the exposures. Chamber atmospheric cc-cemrations were monitored using a Wilks Instrument Co.. Miran Long Pathlength Infrared Mode) 1 A, or an cn-'ine Tracor Gas Chromatograph Model 550. The exposure levels were determined by comparing the observ ed absc-tance ofthese samples to a previously determined standard curve generated using the test material under the same ristrumental sellings. The gas chromatograph was used because of interference in the infrared monitoring procedure (pathlength 21.75 m) for the 10 and 1 ppm exposure groups. Chamber e.posures for a'l study group females were conducted 6 h'day, 5 davs'week during a 10-week premaung and mating period, and daily from days 0 to 20 of gestation and days 5 io 20 of lactation. To determine ifesirous was affected by treatment, daily vaginal smears were made and evaluated for each female beginning 2 weeks prior to initiation of mating. During the mating period, two females were caged with the same male nightly, Females were then examined each morning for evidence d mating (vaginal plug and'or sperm in the vaginal smear). The day on which such ev idenre wax observed was defined as Day Oof gestation. The male was initially caged for E days with the females or uniil both were mated Or;e mated, females were housed indiv idually for the duration of gestation. If a female did not mate, she was placed with a different male for an additional 7-day interval which was followed, if necessary, with yet another "-day interval with a third male. Observations of females for mortality and gross clinical signs were made twice daily. Detailed physical examinations were performed weekly throughout the study. Body weights were recorded once weekly through completion of the mating per.od. Mated females were weighed on Days 0,7, 14. and 21 of gestation ard on Days 0. 4. 14, and 21 of lactation. Pups were counted, weighed, and sexedon Days 0.4.14, and 21 of lactation. Litters were observed twice daily. On Day 4 of lactation, litters :f more than 10 pups were randomly culled to 10 w ith an equal number per gender whetc possible. Pups that died were weighed and sexed by iniemal examination. All dams were given a gross postmortem examination. Method of sacrifice was by overdose of ether. Abnormal tivsues noted during ihese evz'uations were saved in 10T neutral buffered formalin. Uteri were examired for the presence and number of imp'ictaiion sites, and along with ovaries, were saved in 10ft neutral buffered formalin solution. Gross postmortem exam,Yat'ens, including internal gender determinations were performed on all pups sacrificed by ether over dove on Day 21 c'lactation and pups found dead during lactation. The latter were also checked for the presence or absence ofrrvlk ir. 'he stomach Liver, kidney, and in males, testes weights were recorded fc: each pup. 276' CMA 120093 FEMALE RAT FERTILITY WITH INHALED BENZENE Thirty-three organs and tissues along with any abnormal lesions were sued from two pups per sea per litter in 10* buffeted formalin for possible future histopathological examination. Pups found dead prior to Day 4 of lactation were preserved in 701; ethanol. Statistical evaluation of dim body weights and body weight changes during the treatment period, as well as gestation length and the number of pups per liner w as by a standard one-w ay analysis of variance. Data were fust tested with Bartlett's test for equality of variance. The pregnancy percentage, viability index, and lactation index were tested for equality using a Chi-square test. If this test indicated statistical differences, each treated group was compared with the control group using Fisher's Exact test. This was done only after correcting the alpha (Type I probability) level for multiple tests via the Bonferrone correction (Miller 1966). Pup body weights at Days 0,4, 14, and 21, organ weights and organ-body weight ratios at Day 21 were tested using a nested design. Pups are talen as nested within dams, and dams nested within treatment groups. The significance of the difference between groups is determined by the ratio of dose group mean square crTar divided by the dam w ithin group mean square error. If the ratio indicated a statistically significant difference in means, individual group mean differences were determined by the least significant difference (two sample Mesa with the pooled variance) technique (Snedecor and Cochran, 1967). RESULTS Mean chamber benaene concentrations and ranges, as determined analytically, are presented in Table 1. Body weights of dams were comparable between the control and treated groups during the treatment period. Likewise, mean body weigh: gain during this same interval was comparable between these same groups, and no inequality of variance was indicated (Table 2). The type and incidence of in vivo observations noted during weekly physical evalumons were within the limits of that normally observed in the strain of animal used in this investigation. Parturition and lactation data art presented in Table 3. These data reflect a statistically significant decrease in the lactation index at the 10 ppm level, which wu not seen in either the 30 or 300 ppm groups and was considered not to be biologically significant. Mra.*, pup body weights on Days 0, 4. and 14 of lactation did not show a significant difference from corresponding control values for either gender. Pup body and organ weight values are presented in Tables 4 and 5. The only changes of statistical significance in these data were the decreases observed at Day 21 in female body and liver weights associated with the 300 ppm exposure level. Mean organ to body weight ratios (Table 6) showed nonsignificant decreases in male and female pup liver weights, and a significant increase in female pop kidney weigh: a: the 10. ?0. and 300 ppm exposure levels. Cross necropsy observations of benaene-exposed dams and pups as compared w ith control dams and pups did not reveal any treatment-related effects Likewise, evaluation of gross pathology findings in dead pups recovered during the study or pups culled from litters at Day 4 did not reveal effects attributed to treatment. DISCUSSION Exposure of adult female Sprague-Dawley rats to benaene in a or.e-generation reproduction study produced no evider.ee of toxicity. body weight, and'or altered reproductive performance. However, Kuna and Kipp (1981), Take I. Mix's Levels and Ranges of Benzene in Test Chxmbep Atmosphere Phatr Premating Mating Gestation Lactation 1.0 10(0.3-3.5) 0.7 (0.0-1.4) 0 8(0.2-1.4) 1.2(02-1.4) E.rpojure Lev e/i (ppm) 10 JO 9.8 (7.4-12.7) 9.8(8.0-12.1) 10 9 (8.0-14.1) 12.7(9.5-15.2) 30.3 (26.3-35.4) 30.9 (29.6-33.0) 30.9 (29.6-33.0) 30.7 (29 0-32.4) J00 299 (273-334) 300(287-319) 296 (260-319) 298 (260-337) 277 CMA 120094 KUNA ET AL. CMA 120095 i Table 2. Mean Matkbnal floor Wnr.irrs and Door Wikuit Cain (p) Dumno Gestation ano Lactation Group) ppm 0 Ctumgr Onmitmjnayst_____________________________ Wright 7 H 2/ 0-2/ in tlo.lv 0 4 I/O N 11/1 N tll/IO N 1V/30 N V/JOO N 253 + 20* 22 255 + 21 22 261 19 24 234 21 19 254 * 2S 23 273 22 22 273 + 24 22 282 + 20 24 273 23 19 273 21 23 300 23 22 301 26 22 311 23 24 301+25 19 30! + 29 23 375 + 32 22 376 32 22 368 + 30 24 384 30 19 377 * 39 23 122 23 22 I2f 20 22 126 21 24 UK) If 19 126 22 23 284 + 23 21" 282 24 22 288 24 24 289 23 19 289 34 23 289 2.1 21" 291 22 22 298 + 26 24 294 + 23 19 2% 31 22 'Mean standarddeviation. "One ifam initialed delivery, pup in vaginal canal; however no litter found neat day. No data presented for ibia animal. N number of observation*. lunation (Days) 14 311 24 21 319 23 22 325 23 24 318 + 26 19 316 + 28 22 21 322 21 21 320 22 22 323 * 24 24 328 + 25 19 330 * 35 22 Changr Wright 0-2/ 39 13 21 38 + 15 22 37 * 19 24 39* 12 19 41 + IS 22 in ilmiv FEMALE RAT FERTILITY WITH INHALED BENZENE Table 3. Partvwtion and Lactation Data Obsrnaiion Percentage prcgnanl No. pregnant*,no. mated No. of litter* Mean gestation length (day*) Mean no. pups litter Gender ratio (M/F) Mean no. UFC4 Viability index {%) No. pups alive at 4 days pee- cull No pups bora alive Lactation index (%) No. pups alive at 21 days-No, pups alive at 4 days-postcull / 88.3 23/26 21 21.7 i 0.1 11.7 = 0.6 123/120 14 97.1 236/243 99.0 195/197 /; 96.2 25/26 22 21.8 = 0.1 11.8 = 0.6 137/117 1.0 98.S 251/254 100.0 206/206 Croup III 92.3 24/26 24 21.9 $0.1 12.6:0.6 148/145 0.8 99.0 290/293 IV 84.0 21/25 19 21.7*0.1 11.8:0.7 101/122 l.l 99.5* 217/218 94.3* 216/229 99.4 177/178 V 88 J 23/26 23 21.6= 0.1 12.0= 0.6 144/117 1.5 96.9 253/261 98.0 198-202 * Number pregnant * number of females with eiiher linen, uterine implants, or scan. ^ Unaccounted-for-ccncepti derived by subtracting total number of pups at birth from number of implant scar*. ' Significantly different from control, p < 0.Q5. Green et al (1977). and Hudak and Ungvary < 1978) have all observed decreases in dam bod} weights after comparable exposures to benzene during the period of gestation. One explanation for this difference is that the ahov e bod; w eight deleaves were seen at doses in excess of w hat was reported in this paper. When usirg a chronic exposure of 30C ppm. Sy nder et al. (1978) only observed rat bod} weight decreases after 30 weeks. Therefore, the benrene exposure coventrztions used in this study were neither high enough nor of long enough duration to cause a response of maternal weight loss. Investigators fGieen el al., 197S. Hudzk and L'ngvary, 1978. Kuna and Kapp. 1981; Muni; et al.. 1979) stud; ing the teratogenicity of benzene have all observed decreased fetal body weights at Day 20 of gestation with exposure levels to 2230 ppm In this study, a similar fetal body weight response was observed in new boro birth w eights. 300 ppm group. Although male and female pup weights in the 300 ppm gtoup were lower than control pup w eights at all observation times, statistical significance could only be established for 21-day-old female pups. Male and female pups in the 30 ppm exposure group tended to have lower body weights at Days 14 and 21 of lactation. The above pup body weight charges from this investigation and historical references were observed at different time intervals. This is most probably due to the difference in benzene exposure schedules to the dams, but does not diminish the similarity and'or consistency of biological response. The data indicates that poor pup weight gain in male and female rat pjps appears to be an effect of benzene exposure to adult rats. In conjunction with the reported body weight decreases in this study there was also a statistically significant reduction in female pup liv er w eights in the 300 ppm group at Day 21 of lactation. Nonsignificant decreases in liver weights and liver to body weight ratios were seen in male pups in the 30 and 300 ppm exposure levels. Kidney to body weight ratios were statistically higher than control values in the 10, 30, and 300 ppm exposure levels for female pups and non-statistically higher in male pups. Although no other long-term benrene reproduction studies are available for comparison of the data, these differences in liver and kidney to body weight ratios may simply reflect differences in body weights rather than a direct response to benzene exposure. In conclusion, exposure of female rats to 1, 10, 30. and 300 ppm benzene from premating through lactation did not produce any treatment-related effects in the dams. However, pup body and liver weight decreases were noted at the termination of the study in the 30 and 300 ppm groups along with increaved kidney to body weight ratios in the 10. 30. and 3X pgm exposure groups for male (not statistically significant) and female pups. The significance of these changes is unknown at this time. 279 CMA 120096 KUNA ET AL. CMA 120097 Gnwpt jtpm Day 0 Body Wl. I/O N ll/I N lll/IO N 1V/30 N V/300 N 6.27 0.66* 123 6.40 0.33 137 6.36 0.63 146 6.24 0.64 101 6.03 0.36 144 `Mean slamtird de*ialion. N * number ofotaenralMM. Tabu; 4. Malf. Pur Hour Wnnin <p) and Okran Wwuirr (g) Data Day 4 Pialrull Body Wl. 9.69 1.49 102 10.03 1.31 III 9.99 1.74 112 9.60 1.63 60 9.49 1.49 102 Oar 14 Body Wt. 23.29 3.26 101 23.60 3.13 HI 23.32 4.09 no 21.62 3.68 79 21.94 3.01 101 Day 21 Body Wl. 37.56 3.73 101 38.24 7.20 III 37.92 7.68 106 33.92 6.22 79 34.14 3.73 101 Tent* 0.16 0.04 101 0.18 0.03 III 0.17 0.04 108 0.13 0.04 79 0.13 0.03 101 Organ wightt Liver 1.71 0.39 101 1.71 0.44 110 1.69 0.47 108 1.31 0.41 79 1.43 0.35 101 Kidneys 0.44 0.08 101 0.47 *0.10 III 0.47 0.10 108 0.42 0.09 76 0.42 0.09 101 FEMALE RAT FERTILITY WITH INHALED BENZENE Table 5. Female Pup Body W'eights (g) and Organ weight (g) Data Group1 ppm Dor 0 Body Wt. Day 4 Posteull Body Wt. Day 14 Body Wt. VO N 11/1 N 111/10 N 1V/30 N V 300 N 5.99 = 0.54* 120 5.95 = 0.48 117 6.11 = 0.61 145 6.01 = 0.62 122 5.79 = 0.51 117 9.20= 1.28 95 9.25 = 1.54 95 9.73 = 1.58 117 9 43 = 1.50 98 9.18 = 1.47 99 22.20 = 3.03 94 21.94 = 2.87 95 22.80 = 3.97 112 21.77= 3.42 98 20.85 = 2.95 98 `Mean = standard deviation. `Significantlv differeni from control group. p < 0.05. N * number of observations. Day 21 Body Wt. 36.3 = 5.20 94 35.45 = 5.89 95 37.71 = 7.14 101 33.68 = 5.40 98 32.59 = 5.05* 97 Orion Weights Liver 1.69 = 0.36 94 1.59 = 0.41 95 1,78 = 0.43 108 1.56 = 0.33 98 1.46 = 0.34* 97 Kidney 0.44 = 0.08 94 0.44 = 0.08 94 0.49 = 0.10 108 0.45 = 0.08 98 0.44 = 0.08 96 Table 6. Mean Organ/Day 21 Body Weight Ratios x 10-2 Croup! ppm Tenet Molei Liver 1/0 N U'l N IIL10 N JV.-30 N V/300 N 0 426 = 0.068* 101 0.47" = 0.145 111 0.44? = 0.059 108 0.442 = 0.078 79 0 446 = 0.060 101 4.493 = 0.513 101 4.438 = 0.479 110 4.435 = 0.665 108 4 3 80 = 0.536 79 4.216 = 0.470 101 'Mean = standard deviation. `Significant!} different from control. p < 0.01. N = number of observations. Kidney 1.177 = 0.108 101 1.226 = 0.145 111 1.241 =0.125 108 1.239 = 0.093 78 1.229 = 0.142 101 Females Liver Kidney 4 626 = 0.546 94 4 438 = 0.674 95 4,700 = 0.492 108 4 601 = 0.507 98 4.442 = 0.469 97 1.223 = 0.103 94 1.256 = 0.095 94 1.295 = 0.103* 108 1.330 = 0.115b 98 1.333 = 0.136b 96 REFERENCES GREEN, ) D., LEONG, B K., i`d LASKIN, S. (1978) FdOieulcitj of inhaled benaene in na. Toxicol Appl. Pharmacol. 46:9-18. Guide 10 Oeeupoiioiultipoture I'e/ufj--1991. 2nd ed. American Confncnce of Government at! Industrial Hv jicrixu. HETT. J and MAAK. H (1938; Beniol and Keimdrusen. Kliniseht Wochenschr. 17:1376. HL'DAK, A. md LNGVARY. G. (1978). Embryotoaie effects of benaene *nd ill methv! derivatives: toluene. i>Iene. Toaicolog) 11:55-63. KL'S'A, R A and K.APP. R.W. (1981). The emb^otoiie/teratogenie potential of benux vapor in mi. Toiicol, Appl. Pharmacol 57:1-7. MILLER, R J.11554; S.'iuftt-.r.-ua Sioiinieal Inference. MeGra* Hill. Vc York. 281- CMA 120098 KUNA ET AL. MURRAY. FJ., JOHN. J.A.. RA.MPY, L.W., KUNA. R.A., ind SCHWEIZ, B.A. (1979). Embryo toiieitj of inhiled hcniene in mitt xnd nbbiii. Am Ind. H)|iene AstOC. J, 40:993-991. SN'EDECOR, G.W ind COCHRAN, W.C. (1967). Siotiuica!Mrshodi. 6th Ed. Ioi Suit Uni*ersit) Press. Ames !A. SYNDER. C.A..GOLDSTEIN, B.D.. SELLaKUMAN. A,, WOLMAN. S.. BROMBERG. I., ERUCKMAN. M.N.. AND LASK1N, $. (197|). Hemitotoiicity of inhiled berunnc to Spri(u(<Dile) mi ind AKR mice it 300 ppm. J. Toxicol. Environ Hcilth 4.-60S-61I. WOLF. M A.. ROWE, V.K.. MCCOLL1STER, D.D., HOLLINGSWORTH. R.L.. ind OYEN. F. (19S6), Toxicolojicil itudiei of ccrtiioill) tiled benunes ind benxene. Arch. Ind. Hcilth 14:317-391. Address reprint requests to: K. A. Kuna National Starch A Chem Co. Findtrnt Ave. Bridgtu-attr, NJ 08807 282 CMA 120099 TERATOLOGY EVALUATION OF METHYL TERTIARY BUTYL ETHER IN RATS AND MICE C. Clifford Conaway Texaco Inc., Beacon Research Cenler. 8eacon, New York Ra>mond E. Schroeder 8i0'dynamic$ Inc.^East Millstone, New Jersey Neil X. Snyder American Petroleum Institute, Washington, D.C. .`fated CO Sprague-Dawrley rats and CO-1 mice were exposed during the period of organogenesis to target concentrations of 0, 250. 1000, and 2S0C ppm metnyi i-butyl ether (MTSSt, S'one of the control or test-group animals died during the treatment or posttreatment periods, females mere sacrificed on d 20 Iratsl or d 13 imicei. No aduse eh'ects of treatment mere reflected in maternal parameters of body weight. ate' consumption, or Over weight or m physical examination data for either species, food consumption fell ..n the groups of treated rats during d 9-12: simile' but non significant effects were observed for mice during d 12-15. In rats, no treatment-related changes were recorded m the uterine implantation data, fetal site parameters, or *'etal sex d-stribution data. Examination of fetuses for external atmorrr.alities. skeletal malformations. Or ossification vanat.ons did not re1. ?al any cnar.ges caused by MTSc exposure. A shgn: increase m feta.' resorptions was observed m the groups of miCv- c-. ->ed to /o and higr. concent'at-ons. this increase was attributed to two fen.-,:--- , each group that had an unusually high number of resorptions, rather than to itment itstlt .So significant effects were observed in any groups of treated mice- t-n external and soft-tissue examination or evaluation of skeletal abnormalities oi onuication var.at.ons The incidence of fused sternebrae in tne high-concentrat.on -roup in creased slightly, which m.ignt be attributed to fetotoxicrty. Th.s studs was conduced unde' the auspices of ihe American Petroleum Institute for the follow.ng sponsor companies: Area Chemical Compan), Euon Corporaiion, Che-mische Werlte Huls *C. Peirotev Chemical Corporation. Phillips Petroleum Company, Shell Oil Company, and Tevac; Chemical Company. *"e authors acknowledge ihe technical assistance ot Maureen Youmetf, Xathr>n Sloan, and Xare* Van Svser .'Reproduction Te*atoiog\ Sectioni and ot Soger Ben-Dyke tlrthj'ation To*icolocs Section' ot 3.0 dynamics. Inc. s"e-en: jdo'ess tor C CIttiord Csnawas >> Pus/ OttiCe Bos I2SU2, Arlington \i-st.nia TJ109. 5re,ent jcoress tor Xei! K Sn\dC' is Atlantic S.chheld Co Los Angeles, Calilcr-wa 4t'051. \eque''s tor 'epr.nts should he se"t to C E Holdsnonh A--.encan Petroleum Institute, 1220 L 5fee: S k\ V.tih.npsr.. O C IfCOi ( CMA 120100 ,*98 C. C. COV AW AX ET AL INTRODUCTION Methyl f-butyl ether (MTBE) has been approved by the U.S. Envi ronmental Protection Agency as an octane booster in motor fuei at concentrations as high as 11% (Federal Register, 1979, 1981). Human exposure may occur during the processes of manufacture, distribution, and blending of MTBE; the greatest potential for exposure, however, is from the inhalation of gasoline vapor containing MT8E at service sta tions. MTBE Has a characteristic odor but a low degree of acute toxicity. It is slightly irritating to the eyes and mucous membranes and minimally irritating to intact skin (Reynolds et al., 1974). The 4-h LC50 for MTBE in Sprague-Dawley rats is approximately 35,000 ppm. When 3000 ppm MTBE was administered 6 h/d, 3 d/wk for 9 d, signs of anesthesia--i.e., a lack of response to stimuli-- were occasionally observed. In addition, nasal irritation and liver enlargement were observed when the animals were sacrificed (American Petroleum Institute, 198-4)- Irritation conges tion, and localized inflammation of the nasal cavity and respiramr. tract were also reported when Sprague-Dawley rats were exposed to 1000 ppm MTBE for 13 wk; signs of light anesthesia occurred during the periods of exposure (Huls, 1980). MTBE has been tested for toxic ef fects at concentrations of 10 or 15% (v/v) in high-octane premium gaso line. Addition of MTBE did not increase acute toxicity of motor gaso line, but it did lengthen the barbiturate-induced sleep time, reduce spontaneous motor activity, and cause slight disturbances in motor co ordination (Snamprogette S.P.A., 1972). Methyl ;-butyl ether was not mutagenic using the Ames Salmonella micrcsome assay; it did not induce significant sister chromatid ex changes or chromosomal aberrations in Chinese hamster ovary cells in vitro. Although MT8E was not mutagenic in the L317SY mouse lym phoma assay without microsomal activation, it did induce significant mutations in the presence of a microsomal S9 fraction. It was surmised that a metabolite, possibly formaldehyde, was the mutagenic agent, rather than MTBE itself (Arco Chemical Company, 1980). The present study was undertaken to determine the potential embryotoxic or teratogenic effects of MTBE in Sprague-Dawley rats and CD-I mice. METHODS Mated Sprague-Dawley-derived rats (CD', Charles River Breeding Laboratories, Inc., Kingston, N.Y.), 57 d old and weighing approxi mate^ 203 g at d 0 of pregnancy, were randomlv distributed into 4 groups of 25 each. Pregnancy was confirmed by the presence c; sperm in a vaginal' smear or a copulatory plug. Each mated female was identi fied with a metal ear tag and housed individual!) in 3 stairless-steel. CMA 120101 TERATOLOGY OF MT8E IN RATS AND MICE 799 suspended wire mesh cage, with tap water provided. Certified Purina laboratory chow (5002) was provided ad libitum, except during chamber exposures w-hen the food and water were removed. Artificial light was cycled 12 h on, 12 h off. The temperature was maintained at 65-75'F, with relative humidity ranging between 43 and 71%. Male rats used for breeding were selected from an in-house colony. Mating was effected by overnight cohabitation with two females per male. Mated female strain CD*1 mice, 72 d old at the initiation of mating and weighing approximately 26 g, were randomly distributed into 4 groups or 30. The mating and husbandry procedures were identical to those used with the rats, except that the temperature fluctuated be tween 67 and 76'F, with a relative humidity of 35-72%. Pregnancy was confirmed by a copulatory plug. The methyl f-butyl ether (lots 0-28496 and AN-1993) w-as provided by the American Petroleum Institute. The purity of the MTBE was deter mined to be 95.03-98.93% using gas chromatography, with less than 0.80% t-buty! alcohol, 0.43% methanol, and 0.25% diisobutylene as contaminants. The MTBE was transported via Teflon tubes to FM! fluid metering pumps (mode! PP6-20), which de'ivered the test material directly to air atomizing nozzles (14 in, j$S Spraying Systems) mounted on the inlet portals of 10-m3 stainless-steel and glass chambers. The chambers were operated at an airflow- rate of 3C30 l/min, with 1 complete air change e'.ery 3.3 min (f,, = 15.3 min). Exposures were conducted 6 h.'d during d 6-15 of gestation for both species, with target concentrations of 0 (group I), 250 (group 11 j, 1000 ig'oup III), and 2500 (group IV) ppm. Exposure concentrations of MTBE were monitored every 30 min using a Miran 1-A infrared analyzer and a calibration curve. Nominal concen trations for each exposure chamber were calculated daily. The aerosol content w as determined at least cnee per week using a Rcyco portable particle monitor (mode! 218). Maternal body weights were recorded for both species on d 0, 6, 12, 15, and 18 and on a 20 for rats. Detailed physical examinations for signs of toxicity were performed at the same time that weights were recorded, after exposures, during d 6-15. Food and water consump tion were recorded for the following intervals of gestation: d 6-9, d 9-12, d 12-15, and d 15-18 for both rats and mice, and d 18-20 for rats. Dams were sacrificed on d 20 (rats) or d 18 (mice! by CO: inhalation. Laparotomies were performed, and dams and pups were examined carefully for gross abnormalities. Each fetus was weighed, the crownrump distance was recorded, and the sex of each rat fetus was deter mined from the anogenital distance. Mice were sexed internally at evisce'vatlon or during the soft-tissue examination. Late and early resorp tions were scored When .no uterine implantation sites were observed, the uterus was-stained with a 10% solution of-a.mmonium sulfide (Salewski, Iv-M) to vsuahze the foci of implantation. One-third of the fe CMA 120102 800 c. c, CON'*' AY ET AL. tuses in each litter were preserved intact in Bouin's solution and pro cessed for the evaluation of soft tissue using Wilson's (1965) free-hand razor-blade sectioning technique. The remaining two-thirds of the fe tuses were eviscerated, fixed, cleared, and stained with Alizarin red S for skeletal examination (Crary, 1962). The equality of means was evaluated statistically using a one-way analysis of variance (ANOVA) technique, followed by a multiple-com parison procedure if necessary. Bartlett's test (Snedecor and Cochran, 1957) was performed to determine if groups had equal degrees of var iance. If the variances were equal, parametric procedures were used; if not, nonparametric procedures were utilized. A one-way ANOVA, using the F distribution to assess the significance level (Gill, 1973), was fc lowed by Dunnett's (1955 test if significant differences among the means were indicated. If n: .-^parametric methods were required, the Kruskal-Wallis test xvas used, it significant differences were indicated, D.nn's summed rank test was performed to determine which treat ment groups differed from the control group (Hollander and Wolfe, 19"3). Methods for analyzing trends were also utilized, i.e., regression a-alvsis or jonckheere's test for monotcnic trend (Hollander and Wolfe, 1973). Incidence data for treatment versus control groups -- i.e., preg nancy rates, the percentage of fetuses with malformations, the perce*tage of litters containing malformed fetuses, and ossification varia tions-- were analyzed using standard chi-square analysis (Snedecor and Cochran, 1967). This analysis was followed by a 2 x 2 Hsrer exact test (Bradley, 1968), using the Bonferroni inequality correction 'Miller, 1955) to insure the stated significance level. Armitage's 11955' test was pevormed to determine if there was a linear dosage-respcr.se rela tionship. RESULTS Experimental Exposures: Rats and Mice The cumulative mean chamber concentrations = SD for rat expo sures as determined by infrared (Miran) were 0, 250 z 5, 1CC0 = 12, a.-o 2430 o 91 ppm, for group I (0 ppm), group II (250 ppm), g'cup III ('720 ppm), and group IV (2500 ppm), respectively. Corresponding moan nominal concentrations xvere 0, 250 = 10, 1100 n and 3320 r 127 ppm. For mice, the Miran cumulative mean concentrations were 0, 260 = 26, 1110 = 119. and 2710 r 233 ppm, respect:-.e'y; cor responding cumulative mean nominal concentrations were 0, 2o0 n 19,. 1220 n 94, and 3500 - 153 ppm, rcspcctix ;'>, for group I. g'oup II, -g'ttup 111-, and group IV. CMA 120103 TERATOIOC1 OF MTBE IN RATS AND MIC! 801 A leak was discovered in the sampling manifold late in the study and may have caused an underestimation of the chamber concentra tions from Miran data by as much as 10%. The possibility of error was further indicated by higher nominal concentration data. No correction factor, however, was applied to the Miran data to account for the leak. A substantial amount of aerosol was measured in the mid- and highconcentration chambers; the particle diameter was determined to be less than 1 -2 y.m for"a substantial proportion of the aerosol, which is in the respirable range. The presence of aerosol would also explain some of the discrepancies observed between the mean analytical and mean nominal concentrations. Experimental Results--Rats Maternal changes. All dams survived the treatment and observa tion period. The mean body weight and the mean weight gain during the pretreatment !c 0-6), treatment (d 6-15), and posttreatment (d 13-20) intervals were comparable in the control and treatment groups. The mean food consumption data were unremarkable, except that during the d 9-12 interval there was a significant reduction in mean food consumption in all three treatment groups. Although there was considerable variability in water consumption and a possible dose-re lated increase during d 18-20, no concentration-related charges in water consumption were observed during the period of MTBE expo sure. No treatment-related effects were noted during the evaluations of the live test animals. Upon sacrifice, the mean liver weights, both ab solute and relative to the d 20 body weight, were not significantly af fected. No gross postmortem effects attributed to the exposure were recorded. The pregnancy rates for the control and treatment groups were comparable. The mean numbers of corpora lutea, uterine implantations, resorp tions, and live fetuses (Table 1) were not significantly different in the treatment and control groups. Complete resorption of litters occurred in 1 low-concentration-group female and 1 mid-concentration-group female, while 1 control-group female had an in utero litter consisting of 12 resorption sites and a single li\e fetus. Only 1 dead fetus was recov ered during the d 20 sacrifices; it was from the litter of a low-ccmcent.-ation-g-oup female. Fetal morphologic changes. The mean fetal weights and crownrump distances were statistically comparable in all groups. A prepon derance of male fetuses (58%) in the mid-concentration group v.as att-'butec to bidcg'Cal xariabihiy rather than to a treatment effect. Ex ternal malformations noted in the control and treated fetuses are presented in Table 2. The incidence of external malformations was CMA 120104 802 C C. CONAWAY ET AL. TABLE 1. Corpora lutea and Uterine Implantation, Rats' Treatment group (ppm) Number of pregnant females Mean per pregnant females = SD Corpora lutea Implants Resorptions Live fetuses Mean Mean percent Mean Mean percent number of implants number of implants = SO = SD r SO = SD 1 (0) II 050) 23 14.8 = 1.3 15.1. = 2.1 13.8 = 14 13.9 - 14 1.4 = 2.5 1.5 : 2.5 10.6 = 19.6 11.3 = 204 12.4 = 3.0 12.4 = 3.2 89.4 = 19.6 88.4 = 20.4 III 25 14.0 12.7 0.9 7.2 11.8 92.8 (1000) = 2.4 = 1.8 = 2.6 s 19.8 - 3.1 = 19.8 IV 24 14.3 13.0 1.2 10.0 11.8 90.0 i2S00> _ 2.2 r 2.1 = 1.6 = 14.4 - 3.1 = 14,4 ' No statistical.'* significant differences from the control group. comparable in the control and treatment groups. Pale fetuses ;3.6% in the control g'Oup, 0.7% in the low-concentration group, 0.3% in the mid-concentration group, and 1.4% in the high-concentration group) were not regarded as biologically significant. The incidence of soft-tissue malformations, on both a per fetus and TABLE 2. Su^.mar- of fetal External Malformations, Rats* Treatment group fppm) Malformation Fetuses Number4 % Litters Number* % 1 (0) Absence of cranium and multiple c-amo'facial defects Totai 1.309 0.3 1r25 4,0 1,309 0.3 1.25 4.0 II (250) No malformations 0286 0/22 -- III Exe-cephaly with open eye. protuding nooo) tc.-gue, and fleshy outgrowth above snout 1295 0.3 1/24 4.2 Total 1.295 0.3 1/24 4.2 IV (2500) Darkened area on tail 1/284 0.4 1/24 4.2 Ede--atous with domed cranium and e-e defects Totit 1.284 0.4 124 4.2 1284 0.7 2/24 8 3 So siai.j: cj'!v s.gnificam difference* from the control group. ie`-ses with the malfefmat.Gn/pumber of 'tf'uscs examined per group. f Number c! !it:e*s containing fetuses with a mafformaticn/number cf litters evaluated. CMA 120105 TERATOLOGY Of MTBE IV RATS AND MICE 803 a per litter basis, was not statistically different in the control and treat ment groups (Table 3). Fetal skeletal abnormalities. The incidence of skeletal malforma tions, on both a per fetus and a per litter basis, were statistically com parable in the control and treatment groups (Table 4). A cervical rib and reduced cranial ossification were found in the one fetus with a cranial/ facial malformation. In the mid-concentration group, reduced cranial ossification was recorded for the fetus with exencephaly, and facial malformations were noted upon external examination. In the highconcentration group, fused ribs and fused cervical vertebral transverse processes were observed in the edematous fetus. Experimental Results--Mice Maternal changes. A slight increase in the incidence of lacrimation was seen among females during the treatment period. The food con sumption data were not remarkable except for a slight nonsignificant, dose-related decrease in mean food consumption during d 12-15. A slight, dose-related decrease in mean water consumption, nonsignifi cant, was observed during the d 9-12 gestation interval. No mortality occurred among the treated females that were exposed to MTBE. One low-concentration-group female died on d 0 of gestation, and an addi- TXBLE 3. Feu: ;cr-7nt_* v.ait'ormations. Rats* Treatment group tppmi laitormarion Peruses Number* % Litters Number* % i (0) Sera) pe,. . intended, ureter distended or both 4/102 4.0 3/25 2-0 Edem.i-p.i subcutaneous) --thoracic anp ibcum.nal region 1/1C2 1.0 1/23 4.0 Total Sn02 4.9 4,IS 16.0 II (ISO) Ureter distended Portion or .-.ght cerebral hemisphere outside c-anial cavity (between Cranium and skin layer) Total 1/95 1.-9S 195 1.1 102 1.1 1/22 2.1 2/22 4.S 4.5 9.1 III noooi Ureter d `tended Total 1/96 1.0 114 1/96 1.0 1/24 4.2 42 IV OSOOI Sena.' both ; and ureter distended, or 3-93 393 3.2 2*14 3,2 224 8.3 8J ' Vo tuti.-'Cj'V Sig- 'M-' i.rterences ncm the control group. : O' 'Cuses the maltor-ij'icn number 01 te'uses ewmnned per group. ' V,-;c p: Utters';;-":,. - -g tetuses "Oh the maliormot.on number or M:e*s equated. CMA 120106 804 C. C. CONAWAY FT AL TABLE 4, Fetal Skeletal Malformations. Rats' Treatment group (ppm) Malformation Fetuses Number4 % Litters Number* % 1 (0) II (2501 III (1000) IV 0500) Cervical rib and reduced crania) ossification: Total Wavy ribs (bilateral) -Total Reduced cranial ossifications Total Angulated rib (unilateral) Fused ribs (bilaterall and fused cervical vertebral transverse process Total 1/207 1/207 1/191 1/191 1/199 1/199 1/191 1/191 2/191 05 0.5 0.S 05 05 05 05 05 1.0 1/25 1/M 1/22 1/22 1/24 1/24 1.24 1/24 2/2* 4.0 4.0 45 5 42 42 42 42 85 ' No statistically significant differences from the control group. i Sumter ot fetuses with the malformation number of tetuses examined per g-cup. ' Sumter or 'itters containing tetuses with a malformation.number ot litters evaluated. tional mated female was added to the group to provide a total of 30 mated females. The mean body weight and the mean body weight gain for d 0-6, d 6-15, and d 15-18 were comparable in the control and treatment groups. Mean liver weight and liver weight as a proportion of the cor rected d-13 body weight were not affected by MTBE exposures. No treatment-related effects were recorded during the gross postmortem examination of the mice. The implantation data for the control and treatment groups (Table 5) were net significantly different. The mean number of resorption sites and the mean percentage of resorptions/implants were slightly higher in the lew- and high-concentration groups than in the control group, a difference that could be attributed to a high number of resorptions in two litters in each of these treatment groups. In the low-concentration group, 2 females had 10/11 (90.9%) and 719 (77.3%) resorptions/im plants, respectively. In the high-concentration group, two females had litters comprised entirely of resorptions. Excluding the data for females in the low- and high-concentration groups with a large number of re sorption sites, resorption data for these groups were similar to those for the control group. The mean number of live fetuses per litter was comparable in the control and treatment groups. Fetal morphologic changes. There were no significant differences in the mean fetal weights and crown-rump distances of either sex be- CMA 120107 i TABLE S. Uterine Implantation Data, Day 111 CeMilmn Sjcrificcv Mite* Treatment group Ippml 1 (0| It I2S0| III 11000) tv (isnti) Number of pregnant females* 21. 2BC 24 21 Mcmii numlirr of impUnltilKms ; SI) IIS 111 12.1 1.7 11.2 20 i i.n i l ft Kesurplnins Mean number f. Ml Percentage of implants Ml t 11 I 12 1.0 2.1 1.1 1-1 10 i.1 Ml 100 Ifl.b 21.9 11.0 t 9.1 17.1 2S.6 Live fetuses Mean number SI) Percent age of implanls M) 10.) 2.0 10.2 3.2 9.9 2.0 *1 9 3.3 90.4 *1.7 111.1 22.0 SB.2 9.6 82.1 2S.4 Total number dead felines 12 1 2 2 * No ttalrclH all) rliffrirm e\ trim! the imilrul group 11 [idudci lenulrs lli.it rl.-lu CMI prrm.ilurriy lone control and one group Ml femalel. ` Includes Iwo li iiules who .. .it. n were stained willi ammonium sulfide and in which loci were Idenlilied but lhal had no fetuses. Data for these females were not included in the calculations of mean uterine implan tation data for this group. CMA 120108 806 CC. CONAWAY ETAL TABLE 6. Summary of Fetal Eiiernal Elimination Dm. Mice1 Treatment group (ppm) Malformation Fetuses Number* % litter* Number4 % 1 (0) II (250) 111 (1000) IV (2500) Cleft palate TotaJ Spini bifida. filamentous ~ tail. imperforate anus Total Kinked tail Porr.cn of tail dark purple Total Kinked tail Hi.-.c!imb flexure Port-.cn of tail dark purple Tc:ai 1/281 1/281 1.266 1.266 1/251 1251 3-251 1290 1290 1290 3290 0.4 1/27 0.4 1/27 0.4 1/26 0.4 1/26 0.4 . 1,25 0.8 225 1.2 3-25 0.3 127 0.3 127 0.3 127 1.0 3-27 3.7 3.7 3.8 3.8 4.0 8.0 12.0 3.7 3.7 3.7 n.i ' No statist;*.tv i.j'iiicant differences from the control group. * Number etuis* delivered pups) with the malformation number of fetuses delivered pups) eiamined pe* group. r Number ot itters containing fetuses (delivered pups) with a malformation number ci litters evaluated. tween the control and treatment groups. No treatment-related effects on the fetal sex distribution were reported. The incidence of malforma tions observed upon external examination (Table 6) was comparable in the control and treatment groups. Malformations recorded during the soft-tissue examinations and the number of fetuses and litters affected were also comparable in the control, low-, mid-, and high-concentra tion groups (Tazle 7). Fetal skeletal abnormalities. The incidence of skeletal malforma tions in the mouse, on both a per fetus and a per litter basis, was com parable in the control, lowand mid-concentration groups (Table 3). In the high-concentration group, the incidence of fetuses with skeletal malformations and of litters containing such fetuses increased, but the increase was rot significant. Cleft palates, not noted during the ex ternal equations and suggested by irregularities in the ossification pattern of the palatir.e'maxillary bone, were observed in one controlgroup fetus and two high-concentration-group fetuses. Thus, the total incidence of cleft palate observed in the external, soft-tissue, and skel etal evaluations in this study was 0.7% (2/281), 0% (0 263), 0.4% (1.251), and 0.7% .2 292'' for the control, low-, mid-, and high-concentration groups, respectively. A low incidence of stemebral abnormalities (i.e., fused ste'-ebrae) was observed in the treatment groups, while none were obsev ed n the control group. CMA 120109 TERATOLOGY OF MTBE IN RATS ANO MICE 807 TAJti 7. :;:ii Sen-Issue Malformations, Mice* Treatment group (ppm) Malformation Fetuses Number* % Utters Number* % 1 (01 Raul pelvis distended 2/91 U Brain tissue (cerebral hemisphere-- >ft s>de) present between ca.nium aod overlying skin layer 1/91 1.1 Total _ 3/91 3J It C30I No malformations cm -- III ItOQOI >e--collection of blood between !e*s and eyelid Ce" palate* Total 1/83 U 1/83 1J 2/83 2.4 rv C3001 Mean defect --aortic and pulmonary arches exist as a common vessel Cfsoral hemisphere -- lateral ventricles distended Clef palate* Total 1/93 1.1 1/95 1.1 1/93 1.1 3/9S 3J 2/27 1/27 3/27 0/27 1/25 1/25 2/25 1/27 1/27 1(27 3/27 7.4 3.7 11.1 -- 4.0 4.0 8.0 3* 3.7 3-7 11.1 ' Nc s:a: s: cj h significant differences from the control group. * N.mce- j- -et.ses idelive'ed pups: with the malformation number of fetuses 'delivered pups) eiammeo ; g'OuO. c Sum;*' o' i.tte's containing fetuses (delivered pups) with the malformation number of litter* evaiuatec. 'Na: n::*c dur ng the fetal external examination. DISCISSION Although there is some indication that food consumption declined in both rats and mice during portions of the gestation period, this did not hs-.e any significant effect on the terminal maternal body weight, maternal weight gain, or fetal body weight in either species. The mean number of resorptions per litter in the control group of Sprag-.e-OswIey rats was 1.4, somewhat elevated from the laboratory histones1 mean of 0.7 but within the range of 0.3-1.8 recorded for 24 studies. The incidence of resorptions in control mice was within the normal narge; mice typically resorb 10-30% of fetuses, but the per centage depends on the strain of mouse and the laboratory. The inci dence sct't-tissue malformations and major skeletal abnormalities in rats a*d mice exposed to MT8E did not increase significantly. A slight c'cse-'e'ated Increase in minor skeletal abnormalities in mice (Table 8) was "ct stat'stically significant. It was concluded that MTBE adminis- CMA 120110 SOS C. C. CONAWAY ET AL. TABlf I. re-ai SVeetal Malformations. Mice4 Treatment group (ppm/ Malformation Fetuses Number* % Litters Number4 % 1 (0) II (250) III (10001 IV '2500: Cle-t palate* Fused ribs (unilateral) and misaligned thoracic centra Total fused stemebrae _Scimblec sternebrae Fused ribs 'unilateral) and ver.eoral defects: Total Fused sternebrae Fused ribs .unilateral) Anrulated lbs (unilateral) Toui Clem palate* Fused sternebrae i CUl " 2/190 1/190 3/190 1.180 nso nSO 3180 2168 1168 M68 a-168 119$ 4/195 6/1 95 1.1 1/27 0L5 1/27 u 2/27 flj 1/26 Oi 1/26 US 1/26 17 3/26 n * . 225 15 1/23 15 1/25 la a/25 U 2/27 Z-* 4,17 11 6/27 3.7 3.7 7.4 3.8 3.8 3.8 11.5 8.0 4.0 4.0 16.0 7.4 14.8 22.2 4 No sta: snca.lv s.gnu'icmt differences from the control g-oup. 1 Nu~oe' st te'.ses (de'Wered pups) with the malior-nation numbers fetuses (delivered pups) examine; ;e- grc-S. 1 N'umse- sf liners containing fetuses with the maltormatien/numbet'of litters evaluated. 4 Ot;e-.it:ons st the palatine/maxillary ossifications suggest a clen-salate malformation: one of these etuses s noted during the external examination to have i cleft palate. The second fetus wmalformations sf the palatine'mav'larv cmifiratmns was not -cted to have a dert palate. 4 Gbse'wf'On c the pa'atine/maxillary ossifications suggest a cleft-palate malformation; how ever. a :a atai de*ect was not noted during the fetal external examination of these same fetuses. tered bv inhalation at the concentrations tested was not ma:ernally toxic, erbrycroxic, or teratogenic. R=?HESCS Americsn :t'cle.m Institute. 1984. A nine day inhalation study of m"E in the rat. Sus.mitted by B.s d.'amic! inc.. East Millstone, N.|.. to the American Petroleum Institute. Afco C*e_ cal Cjmpanv. 1980. Methyl tertiary butvl ethei Acute 'sucologtcal studies. Arco Chi~ cil Co Newton Square. PA 19073. Armitau ? i?5S ~ests for linear trends in proportions and '"requeue e< 3<omttncs ft'.373-386. Sradlev.V S9& Fisher Exact Test. In Distribution free Statisfical 'esa. pp. 195-203 Englewood C' 'ts. n J.. ;-en!ice-1"iall. Chem.s;-e .`.erxe -'uls AC IHuls). 1980. Methyl tert.ars butvl ether ;d~veron): Three month inha- laton 'cue r. n rats Submitted by Inveresk Rf-eirch 'nternatio-al, Edinburgh, Scotland to C'"i" :Che .`-erlte hiuls AC, Crarv, 2 - f96I vtcd.t.ed bencyl alcohol clearing ui jlirai'n-stainec ipec mens without loss of fie. : tv S'; - Tecnnof. 3?: 124-12S. Cunne", C '.V :~~5. A -- _11ipIe comparison procedure for compjr-' `eve'll treatments with a c;--: / a- $u: 5oc. 5C 1096-1121. CMA 120111 TERATOtOCY Of MTBE IN RATS AND MICE 809 Federal Register. 1979. 44:12242-12249 (March 6). Federal Register. '981. 46:18582-385*6 (July 28). CHI. J. L 1978. Design and Analysis of Experiments in the Animal and Medical Sciences. Ames: low* State University Press. Hollander. M., and Wolfe, D. A. 1973. Chapter 6 In Nonparametnc Statistical Methods. New York: Wiley. Miller. R. C., Ir. 1966. Simultaneous Statistical Inference. New York: McCr*w+lilI. Reynolds. R. W.. Smith. ]. $.. and Steinmeu. I. 1974. Methyl ethers as motor fuel components. Presented as 168th National Meeting of Division of Petroleum Chemistry, American Chemical Society, Atlantic Cfy, Nrf. Salewski. E. 1964. farbemethode turn makroskopischen nachweis von implantations stellen am uterus der ratie, Arch. Pathol. Exp. Pharmakol. 247:367 Snamprogette S. ?. A. 1972. Research on "solvent SB" (C1SB). alone or as a component of various mixtures. Resort submitted by the Institute di farmacologi* della facofca di Medicine e Chirurgi* dell Lnversitita Cattoiica del Sacro Cuore di Roma to Snamprogette 5.PA. Snedecor, C. w.. and Cochran. W. C. 1967. One Way Classifications. In Statistical Methods, 6th ed., Chapter110. Ames: Iowa State University Press. Wilson, J. C. 19S3- Methods for administering agents and detecting malformations in experi mental animals. In Teratology: Principles and Techniques, eds. J, C. Wilson and I. Workany, pp. 262-2" Chicago: University of Chicago Press. Received January 28, 79SS Accepted March U. 798S CMA 120112 Toxicology and Industrial Health, Vol. 3, No. 4, 1987 519 METHYL TERTIARY BUTYL ETHER INHALATION IN RATS: A SINGLE GENERATION REPRODUCTION STUDY ROBERT W. BILES,* RAYMOND E. SCHROEDERf AND CHARLES E. HOLDSWORTH* 'Exxon Biomedical Sciences, Incorporated Exxon Research & Engineering Company East Millstone, New Jersey fBio/dynamics Inc. East Millstone, New Jersey $American Petroleum Institute Washington, D.C. Male rats exposed to target concentrations of methyl tertiary butyl ether (MtBE) at 300, 1300 and 3400 ppmfor 6 hoursjday, 5 days/ week for 12 weeks were mated to female rats exposed to the same concentrations for a 3-week period. Exposures continued through the mating period and the females continued exposures during ges tation andfrom days 5-21 lactation of the litters (Fix) (no expo sures days 0-4 lactation). A second litter (Fib) was produced under the same mating and post mating exposure regimen. No adverse effect of treatment was observed with the adult animals (Fo) throughout the in-life portion of the study. The only remarkable . finding was an increased incidence of dilated renal pelves in the low- and high-dosefemales (Fo). All gonad weights, male accessory reproductive organ weights, organ-to-body weight ratios and reproductive organ histopathology were unremarkable upon com parison of treated animals with air sham controls. The mating indi ces andfertility indices in exposed animalsfor both mating intervals (F\t and Fib) were not significantly different from con trols. Pregnancy rates w ere comparable between treated and con trolfemales for the first litter interval (Fix) but were slightly lower (not statistically significant) than control on the second litter inter-1 2 3 1. Address all technical comments questions to Robert W. Biles, Exxon Biomedical Sciences, Inc., Exxon Research U Engineering Co., P.O. Box 235, East Millstone, NJ 08873 Address reprint requests to Charles E. Holdsuorth, American Petroleum Institute, 1220 L Street, NW, Washington, DC 20005. 2. Key words' inhalation, methyl tertiary butyl ether, oxygenated hydrocarbons, reproduction study, single generation. 3. Abbreviations* CHO, Chinese hamster ovary, MtBE, methyl ternary butyl ether CMA 120113 520 Biles, Schroeder and Holdsworth val (Fib)- Treated animal mean gestation length and the mean number ofpups at birth were not statistically different from con trols. The pup viability indices at birth were comparable for control and treated groups for the F\, generation, but the mid- and highdose groups displayed a slight statistically significant decrease in the Fu> generation; the decrease was not considered to be biologi cally significant and perhaps not treatment-related. Litter survival indices were comparable between control and treated groupsfor both litter intervals. Pups of mid- and high-dose females had slightly lower (not statistically significant) mean weights at days 14 and 21 of lactation but this was not considered treatment-related. The most frequent post-mortem observationfor pups sacrificed at day 21 of lactation was dilated renal pelves. This did not appear to be related to treatment. It is concluded that MtBE inhalation in rats results in little adverse reproductive toxicity as shown in a two litter, one generation reproduction assay in rats. INTRODUCTION In the late 1970s, considerable attention was placed on the development and use of oxygenated hydrocarbons as either high octane motor fuels or as octane improvers. This research effort was primarily driven by the anticipation of an EPA-mandated decrease in lead antiknock usage (Reynolds et al., 1974). Methyl tertiary butyl ether (MtBE) emerged as one of the leading oxygenated compounds for use in gasoline blending. Currently, MtBE is used as an octane improver for motor gasoline, and the use has increased as leaded gasoline phasedown continues as promulgated by the EPA (Ember, 1984). The primary route of exposure to workers is through inhalation which occurs during the production and transportation processes, as well as when MtBE is blended into gasoline. The primary source of potential exposure of the general public is from gasoline vapors from those gasolines blended using varying amounts of MtBE (Rey nolds et al., 1974). Animal studies have been used to assess the acute and subchronic toxicities of MtBE. As well, studies have been completed to assess the genotoxic and teratogenic potential of MtBE. The acute inhalation LCjo to rats has been reported to be 35,000 ppm-4 hour (Arco Chemical Company, 1980) and the acute rat oral LD50 has been reported to be *= 4 g/kg (API, 1980). One subchronic inhalation study (Huls, 1980) conducted in rats for 6 hours/day, 5 days, week for 13 weeks produced minimal to no toxic effects through the dose range tested (250, 500 and 1000 ppm). Another subchronic inhalation study (API, 1984) (6 CMA 120114 Toxicology and Industrial Health, Vol. 3, No. 4, 1987 521 hours/day for 9 days) in rats, conducted essentially as a probe for the present study, produced a reduced response to auditory stimuli, increased absolute and relative liver weights and some minor deviations in clinical parameters at the highest exposure level of 3000 ppm. These findings of relatively low toxicity suggested the selection of the maximum exposure concentration for the current study in order to assure some toxicity in the parental generation. Other toxicity findings relevant to the present study have been derived from short term mutagenicity assays and a teratology study in rats and mice. MtBE was not mutagenic when tested in Salmonella and Saccharomyces or in tests for sister chro matid exchange and chromosomal aberrations in Chinese hamster ovary (CHO) cells in vitro (API, 1980). MtBE was mutagenic in the L5178Y mouse lymphoma assay only in the presence of an S-9 metabolic activation system (API, 1980). An inhalation teratology study (Conaway et al., 1985) in rats and mice exposed to MtBE at nominal concentrations of 260, 1100 and 3300 ppm produced no treatment-related effects in the maternal rats or their offspring and only slight increases in fetal resorptions and fused sternebrae in mice. The work reported here was conducted to evaluate whether exposure to MtBE would (1) affect the fertility, reproductive systems and performance of male and/or female rats, and (2) be detrimental to the development of the resulting offspring. METHODS Test Material. MtBE was provided through the American Petroleum Institute and was shown by GC analysis to be 95-99% pure. Other prominent materials identified in these analyses were (1) secondary-butyl methyl ether, < 0.45%, (2) methanol, < 0.43%, (3) tertiary-butyl alcohol, < 0.79% and (4) diisobutylene, < 0.22%. Inhalation Chamber Design, Vapor Generation and Analyses. The 10 cubic meter stainless steel and glass exposure chambers were operated dynamically at an airflow rate of 3000 liters per minute. The liquid was metered from reservoirs via Teflon tubing directly to air atomizing nozzles. The MtBE was atomized into the chamber air stream which, in turn, vaporized the test material. Target concentrations were 0 (control), 250, 1000 and 2500 parts per million MtBE (ppm v v in air). The concentra tion of MtBE was determined in each chamber (including control) every 30 minutes during the exposure period with a MIRAN I-A General Organic Vapor Analyzer. Backup analyses included daily checks of chamber concentrations and weekly verifi cation of the MIRAN using on-line gas chromatography. Animals. Male and female CD (Sprague-Dawley derived) rats, approximately four weeks old at receipt, were obtained from Charles River. Kingston, NY, and accli mated for 30 days. Animals were fed Purina Rat Chow (certified diet), and water was available ad libitum-, however, neither w'ere available during exposure periods. Hard CMA 120115 522 Biles, Schroeder and Holdsworth wood shavings were placed in each female's cage on day 20 of gestation and fresh litter was provided as needed through day 14 of lactation. All animals (male and female) were sorted into treatment groups with a computerized program which also sorted among the groups to equalize mean body weights between the groups. The animals were housed individually in stainless steel mesh cages except during the mating period and during lactation. Temperatures were maintained between 62-72 F; relative humidity was kept between 30-70%; and there was a 12/12 hour light/dark cycle. Experimental Design. Four exposure groups were established, each with 15 males and 30 females: Group 1-0 ppm (sham control), Group II - 250 ppm, Group III -1000 ppm and Group IV - 2500 ppm. Exposures were conducted for 6 hours/day, 5 days/week during the premating interval, which was 12 weeks for males and 3 weeks for females. Two mating intervals were conducted, and in each interval two females were caged nightly with one male from the same treatment group. Once mated, females were removed to litter. After a Five-day interval, all unmated females were randomly reassigned to a different mating unit and exposed to a male from the same treatment group for five additional days. Females unmated after 10 days of mating were randomly assigned to a different mating unit for a third Five-day interval. Males continued to be exposed 6 hours/day, 5 days/week during mating intervals and mating rest intervals, while mated females were exposed daily 7 days/week, 6 hours/ day from day 0-20 of gestation. Females were not treated from day 21 of gestation to day 4 of lactation. On day 5 of lactation, daily exposure for the females was resumed for 6 hours/day until day 20 of lactation. However, the litters (Fu) were not exposed. The F|j litters were weaned, and the Fo males and Fo females underwent a two-week mating rest period, after which a second mating period began in order to produce a second litter (Fib) of animals. (Fo) males were sacriFiced at the end of this mating period while (Fo) females underwent the same exposure regimen described for the First mating period. Fo females were sacriFiced at the end of the (Fib) weaning. During the First and second mating periods, females that did not mate were treated 7 days,' week, 6 hours/day until all Ft, (Fib) litters were weaned. Females that mated but did not deliver a litter resumed daily treatment at completion of a 26-day postmating interval. In-life observations for Fo males and females were made for mortality and gross signs of toxicologic and pharmacologic effects. Body weights were obtained initially and weekly thereafter for males, while females were weighed weekly until mated and then again on days 0, 7, 13 and 20 of gestation and days 0, 7, 14 and 21 of lactation. At sacrifice, each Fo animal was examined grossly with particular attention given to the reproductive organs. The testes and epididymides from each male and ovaries from each female were evaluated microscopically. CMA 120116 Toxicology and Industrial Health, Vol. 3, No. 4, 1987 523 The newborn pups from both litters were inspected and counted shortly after birth and weighed on days 0, 7, 14 and 21 of lactation. External sexing was confirmed by internal inspection of gonads at sacrifice. On day 4 of lactation, each litter with greater than 10 pups was culled to that number using a randomization technique. Pups found dead at parturition and during lactation were grossly examined both internally and externally for malformations. Pups were sacrificed on day 21 of lactation and given a gross postmortem examination. The gonads and abnormal tissue were saved for histopathological evaluation. Statistical procedures employed for parametric data included Bartlett's test for equal variance and standard one-way ANOVA with F distribution (Snedecor and Cochran, 1967). If significant differences among means occurred, then Dunnett's test (Dunnett, 1955) was used to determine which treatment means were different from control. For nonparametric data, the Kruskal-Wallis test (Gill, 1978) for equality of means was used. If differences among means occurred, then Dunn's summed rank test (Hol lander and Wolff, 1973) was used to determine which treatment means were different from the control. All statistical tests, except Bartlett's test, were conducted at the 0.059t and O.OlCc two-sided risk level. RESULTS Fo`Adults. General. No control or treated animals died during the treatment or mating periods. Mean weekly body weights and mean weight gain for males during the pretreatment period, mating and post mating periods were comparable between all groups (treatment and control); no adverse effect of treatment was evident. Likewise, no treatment-related effect was evident in body weight data for the female animals during pretreatment and post mating periods. Gestation and lactation body weights were also comparable between control and treated groups for both litter intervals. The types and incidences of in-life observations were similar between TABLE 1 Analysis of MtBE Exposure Concentrations Concentration Target (ppm) Actual* Exposure* (ppm) Animals Male & Female Male & Female Group I II III IV 0 250 1000 2500 0 290 22 1180 107 2860 254 0 300 x7 1240 37 2980 132 `Cumulative mean = SD. _*M[RANS * a miran leak was discovered and correct;ed du ring male exposures Actual final ar'.alytical exposure values are shown in above table. CMA 120117 524 Biles, Schroeder and Holdsworth control and treated animals; no adverse effect of treatment was noted. The most common types of in-life observations noted were nasal discharge, rales, alopecia and lacrimation. Necropsy, Organ Weights and Histopathology. The only finding of note from the gross post-mortem examinations was an increased incidence of dilated renal pelves in females from Groups II (250 ppm) and IV (2500 ppm). The incidence of dilated renal pelves for females in the control and treated groups II, III and IV were 1/30, 4/30, 0 30 and 5/30. Fo-Adults. General. No control or treated animals died during the treatment or mating periods. Mean weekly body weights and mean weight gain for males during the pretreatment period, mating and post mating periods were comparable between all groups (treatment and control); no adverse effect of treatment was evident. Likewise, no treatment-related effect was evident in body weight data for the female animals during pretreatment and post mating periods. Gestation and lactation body weights were also comparable between control and treated groups for both litter intervals. The types and incidences of in-life observations were similar between control and treated animals; no adverse effect of treatment was noted. The most common types of in-life observations noted were nasal discharge, rales, alopecia and lacrimation. Necropsy, Organ Weights and Histopathology. The only finding of note from the gross post-mortem examinations was an increased incidence of dilated renal pelvis in females from Groups II (250 ppm) and IV (2500 ppm). The incidence of dilated renal pelves for females in the control and treated groups II, III and IV were 1 30, 4/30, 0 30 and 5, 30. No treatment-related effects were evident in adult gonad weight data. Organ to body weight ratio data for gonads from both adult male and females were not different between treated groups and controls. Mean ovary weight and mean ovary to body weight ratio for the Fo females were comparable between the control and treated groups (data not shown). In the Fo males, mean weight data (absolute and relative to terminal body weight) for the testes, epididymides, seminal vesicles and prostate were comparable between the control and treated groups (data not shown). Microscopic evaluations of the gonads of both male and female animals revealed no treatmentrelated effects. Mating, Fertility and Pregnancy Indices. For both mating intervals (Fi, and Fib), the mating indices for males and females were not statistically different between the control and treated groups (Table 2). The male fertility indices were also comparable between the control and treated groups for both litter intervals (Table 3). No adverse effect of treatment was evident in male fertility data. During the first litter interval (F:i), pregnancy rates were comparable between the control, low- and high-dose groups and slightly lower (not statistically significant) than controls in the mid-dose (Group III) group (Table 3). In the second litter interval, pregnancy rates were slightly CMA 120118 Toxicology and Industrial Health, Vol. 3, No. 4, 1987 CMA 120119 TABLE 2 Mating Indices for Male and Female Rats Exposed to MtBE for Two Mating Intervals Fi. Litter Interval Group [Target ( )[ _ Males' No.e % Females* No." % l (0 ppm) 12 80.0 27 90.0 [I (250 ppm) 14 93.3 30 100.0 III (1000 ppm) 14 93.3 27 90.0 IV (2500 ppm) 15 100.0 29 96.7 -N = 15. `N = JO. 'Number of males mating with a female, '`Number of females mating. Fib Litter Interval Males Females No. % No. % 12 80.0 25 83.3 14 93.3 29 96.7 14 93.3 29 96.7 12 80.0 25 83.3 cro/i LA Biles, Schroeder and Holdsworth CMA 120120 I \ TABLE 3 Male Fertility Indices and Pregnancy Rates for Rats Exposed to MtBe for Two Mating Intervals (Target ( )( 1 (0 ppm) 11 (250 ppm) III (1000 ppm) IV (2500 ppm) Fi. Litter Interval Male Fertility Indices Pregnancy Rale No.* % No." % 11/12 91.7 25/27 92.6 14/14 100.0 28/30 93.3 13/14 92.9 20/27 74.1 14/15 93.3 24/29 82.8 Fib Utter Interval Male Fertility' Indices Pregnancy Rate No. % No. % 11/12 91.7 22/25 88.0 13/14 92.9 23/29 79.3 13/14 92.9 23/29 79.3 11/12 91.7 19/25 76.0 `Number of males impregnating a female/number of males mating a female. kNumber of pregnant females/number of mated females. Vtooi Toxicology and Industrial Health, Vol. 3, No. 4, 1987 527 lower (not statistically significant) in each treated group. No clear dose relationship was evident in either litter interval. In the control group, two females did not mate with a male during either mating interval; all treated females mated at least once during these same intervals. During both mating periods, 28 control (93.3%), 28 low-dose (93.3%), 25 mid-dose (83.3%) and 26 high-dose females (86.7%) delivered at least one litter. Gestation Length, Parturition Data and Litter Survival Indices. Mean gestation length data (range 22.1 to 22.3 days) were comparable between the control and treated groups for both litter intervals. The mean number of pups at birth (live, dead and total) was also comparable between control and treated groups during both litter intervals (Table 4). Pup viability indices at birth were comparable for control and treated groups for the first litter interval, but in the second litter interval, the mid- and high-dose groups displayed a slight, but significant, decrease in the pup variability index (Table 5). The pup viability indices seen in the mid- and high-dose groups during the second litter interval (Fib) were similar to the pup viability index observed for the control group during the first litter interval (Fu). Litter survival indices (ratio of litters weaned to litters with live pups at birth) were comparable between control and treated groups for both litter intervals (Table 5). Pups. Mean pup body weights (distinguished by sex) throughout the lactation period for both litter intervals were comparable in the control and low-dose groups. The mean pup weights for the mid-and high-dose groups were comparable to controls at days 0 and 7 of lactation, but on days 14 and 21 of lactation they were slightly (but not statistically significant) lower than controls (Table 6). Pup survival indices during the day 0-4 and 4-21 lactation intervals are summarized for both litter intervals in Table 7. During the first litter interval, pup survival indices for the day 0-4 period were comparable between the control and high-dose group and significantly lower than control in the low-and mid-dose groups. Pup survival indices for the day 4-21 period were comparable between the control and treated groups. In the mid-dose group, the decrease in pup survival index during the day 0-4 interval was, in part, attributed to the loss of one entire litter containing 12 pups at birth; excluding data for this litter, the day 0-4 pup survival index for the mid-dose w ould be 94.0%. In the second litter interval, pup survival indices during the day 0-4 interval were comparable between the control and each of the treated groups. Likewise, during the day 4-21 interval, pup survival indices were comparable between the control and treated groups (Table 7). Pup sex distribution indices at day 0 (pre-cull) and day 21 (post-cull) were similar between the control and treated groups for both litter intervals (data not shown). The most frequent post-mortem observation for pups sacrificed at day 21 of lactation was dilated renal pelves. The incidence of this observation was comparable in the CMA 120121 t Biles, Schroeder and Holdsworth CMA 120122 I TABLE 4 Number of Litters and Parturition Data for Rats Exposed to MlBE for Two Mating Intervals Group {Target ( )] 1 (0 ppm) II (250 ppm) III (1000 ppm) IV (2500 ppm) n of Litters 25 28 20 24 F i. Litter Interval Mean # or Pups at Birth'l Live Dead Total 13.2 0.3 13.5 12.4 0.3 12.7 11.5 0.7 12.2 13.3 0.4 13.7 U of Litters 22 23 23 19 Fit, Litter Interval Mean It of Pups at Birth Live Dead Total 13.3 0.1 13.4 12.8 0.3 13.1 12.1 0.6 12.7 12.3 0.6 12.9 'Data collected at the interval litter examination. LKn> 00 Toxicology and Industrial Health, Vol. 3, No. 4, 1987 CMA 120123 i TABLE 5 Pup Viability Indices and Litter Survival Indices for Two Litter Intervals Pi, Utter Interval Fit, Litter Interval Group (Target ( )] 1 (0 ppm) It (250 ppm) 111 (1000 ppm) Pup Viability Indices No.* 330/338 347/354 % 97.6 98.0 230/243 94.7 Litter Survival Indices No." 25/25 % too.o 27/27 100.0 18/19 94.7 Pup Viability Indices No. 292/295 % 99.0 295/302 97.7 278/291* 95.5 Litter Survival Indices No. 22/22 % 100.0 22/23 95.7 23/23 100.0 IV (2500 ppm) 320/329 97.3 24/24 too.o 234/245* 95.5 19/19 100.0 `Total number of live pups at birlh/lotal number of pups (live and dead) at birth. ^Number of females that wean litters (day 21)/number of females with live pups ai birth (day 0). P< ,05. LSI to NO Biles, Schroeder and H oldsw orth CMA 120124 [Target ( )1 1 (0 ppm) II (250 ppm) III (1000 ppm) IV (2500 ppm) TABLE 6 Mean Flip Body Weight Data During Lactation--Both Litter Intervals Male 01 Female 6.2 5.9 6.0 5.6 3.8 5.5 6.0 5.6 Male 13.2 13.1 12.5 12.8 Mean Pup Weight1 (grams) Irelation Days 1t 14 Female Male Female Fi. Utter Interval 13.0 23.6 23.0 12.4 24.1 22.8 12.2 21.6 21.0 12.6 22.4 21.9 Male 36.2 36.9 32.9 33.8 21 Female 35.1 34.7 32.5 33.1 1 (0 ppm) II (250 ppm) III (1000 ppm) 6.0 6.1 6.0 IV (2500 ppm) 6.1 `Litler as (tic experimental unit. 5.7 5.7 5.6 5.7 Fih Litter Interval 13.2 12.5 23.8 22.7 36.9 35.1 13.3 12.6 23.6 22.6 36.8 34.4 12.8 12.2 22.7 21.7 34.7 33.1 13.1 12.5 22.8 22.1 34.9 33.6 u* O Toxicology and Industrial Health, Vol. 3, No. 4, 1987 531 Group [Target ( )] I (0 ppm) II (250 ppm) III (1000 ppm) IV (2500 ppm) TABLE 7 Pup Survival Indices for Both Litter Intervals Pup Survival Index Lactation Days 0-4 (pre-cull) 4-21 (post-cuU) No.* % No." Fl Litter Interval % 324/330 98.2 247/250 98.8 317/347** 91.4 246/254 96.9 205/230** 89.1 162/165 98.2 305/320 95.3 235/236 99.6 Fib Litter Interval I (0 ppm) 282/292 96.6 211/218 II (250 ppm) 275/295 93.2 210/213 III (1000 ppm) 273/278 98.2 214/215 IV (2500 ppm) 231/234 98.7 182/182** `Number of pups on day 4 (pre-cull)/ number of live pups on day 0. ''Number of pups on day 21 number of pups on day 4 (post-cull). "P< .01. 96.8 98.6 99.5 100.0 control and high-dose groups for both litter intervals, but was slightly (not statistically significant) elevated in the low- and mid-dose groups for both litter intervals (data not shown). Other observations noted for the weanling day 21 pups during the gross post-mortem examination occurred infrequently and did not appear to indicate a relationship with treatment. DISCUSSION The present study was designed as an evaluation of the potential reproductive system toxicity of methyl-t-butyl ether. No adverse effects of treatment were seen in either male or female adult rats exposed to MtBE for approximately 28 weeks (males) and 16 weeks (females) at the approximate concentrations of 300, 1200 and 3000 ppm. Although a 9 day probe study at 3000 ppm indicated some toxic potential (irritability, possible CNS effects) the rats at all doses in the present study gained weight commen surate with control animals and displayed no unusual toxicological signs. The type and incidence-of in-life signs such as rales, lacrimation, nasal discharge and alopecia CMA 120125 532 Biles, Schroeder and Holdsworth were also similar to the controls. In gross post-mortem, the only finding of note was a nonsignificant increase in dilated renal pelves in females of the low- and high-dose groups. The effect was not considered to be treatment-related since it was apparently not dose-related. Gonad weights and gross and histopathological observations of the gonads from both male and female rats exposed to MtBE were uniformaly consistent with control animals; no treatment-related effects were evident. Also, weights and histopathology results of organs of the male reproductive tract (testes, epididymides, seminal vesicles and prostate)- were indistinguishable between control and treated animals. Conse quently, MtBE displays little to no structural effect on the reproductive system of the adult rat. With respect to reproductive function, no adverse effects of treatment were seen in male and female mating indices, male fertility indices, litter survival indices, litter size, gestation length, or sex ratio of offspring. No treatment-related effects w ith respect to pregnancy rates were evident during the first litter interval; however, pregnancy rates from the second litter interval were slightly, but not statistically significantly lower than the controls and no clear dose-response relationship was evident. No adverse effect on female fertility was indicated from this study. In summary, no significant effects on reproductive performance were observed in this study at concentrations of MtBE as high as 3000 ppm. The effects of parental MtBE exposures were minimal to the offspring. The significant decrease in pup viability in mid and high dose groups of the F;!, generation is borderline with respect to biological significance. None of the pup viability indices of either generation are below approximately 95%. The outstanding viability (99%) of the F;s, control group contributes to the statistical significance seen in the Fib mid and high dose groups. Pup survival indices during the day 0-4 lactation period were significantly (P < .05) lower than controls in the low- and mid-dose groups during the first litter interval. However, the Fi, high-dose group displayed no reduction in pup survival when compared to controls, and also no reduction in pup survival was seen in any treated group during the second litter interval. Moreover, pup survival indices for the day 4-21 lactation period (post-cull) were comparable between the control and treated groups for both litter intervals. Consequently, the reduced pup surv ival in the Fi, low- and mid-dose groups is not believed to be a treatment-related effect. Pups nursing to mid- and high-dose females had slightly lower body w eights at days 14 and 21 of lactation for both litter intervals. The reduction in mean pup weights did not demonstrate a clear dose relationship. Moreover, the differences from ccntrols were not statistically significant, and the mid- and high-dose groups weights were compar able on days 0 and 7 of lactation. Finally, gross external and internal examination of pups delivered and subsequently weaned to MtBE treated females did no: reveal any adverse effects on any organ system due to treatment. The current scientific literature is unexpectedly silent with respect to one or more CMA 120126 Toxicology and Industrial Health. Vol. 3. No. 4. 1987 533 generation reproduction studies with aliphatic, aromatic or oxygenated aliphatic/ aromatic hydrocarbons. Pesticides, drugs, and food additives, and certain chlorinated hydrocarbons make up the majority of these studies. One multi-generation dietary study of 1-3 butanediol in rats has been published which did report a decrease in pregnancy rates of Fu rats and some effects in fetal growth; however no terata, dominant lethal or chromosomal aberrations were noted (Hess et al., 1981). Finally, with respect to reproductive outcome, a report of a two species (rat and mouse) inhalation teratology study on MtBE at doses similar to the present study concluded that MtBE is not maternally toxic, embryotoxic or teratogenic (Conaway et al., 1985). This present study indicates that an oxygenated hydrocarbon, MtBE, possesses little adverse reproductive toxicity potential as defined in a two litter, one generation reproduction assay in rats. ACKNOWLEDGEMENT The authors would like to acknowledge the invaluable assistance of the MtBE Ad Hoc Task Force Participants including C. C. Conaway, K. Hoover, C. Kirwin, S. Ridlon, R. N. Roth, G. S. Simon, N. K. Snyder, L. Stewart and F. B. Thomas. REFERENCES API (1980). Rationale and Program for Human and Environmental Health Effects of the Gasoline Additive Methyl-t-butyl Ether. API (1984). A Nine Day Inhalation Study of MtBE in the Rat. Submitted by Bio dynamics Inc., East Millstone, NJ to the American Petroleum Institute. ARCO CHEMICAL COMPANY. (1980). Methyl Ternary Butyl/Ether: Acute Toxicolog ical Studies. CONAWAY, C.C.,SCHROEDER, R.E. and SNYDER, N.K.( 1985). Teratology-Evaluation of Methyl Tertiary Butyl Ether in Rats and Mice. J. Tox. and Environ. Hlth. 16:797-809. DUNNETT, C.W. (1955). A Multiple Comparison Procedure for Comparing Several Treat ments with a Control. J. Amer. Statist. Soc. 50:1096-1121. EMBER, L. (1984). EPA Study Backs Cut in Lead Use in Gas. Chemical and Engineering News, April 9. GILL, J.L. (1978). Design and Analysis of Experiments in the Animal and Medical Sciences, Iowa State University Press. HESS, F.G., JR., COX, G.E., BAILEY, D.E., PARENT, R.A. and BECCI, P.J. (1981). Reproduction and Teratology Study of 1,3-butanediol in Rats. Journal of Applied Toxi cology 1:202-209. HOLLANDER, M. and WOLFF, D.A. (1973). Nonparametric Statistical Methods, Ch. 6. New York: John Wiley and Sons. HULS (1980). Methyl Tertiary Butyl Ether (Driveron): Three Month Inhalation Toxicity in Rats. Submitted by Ineresk Research International, Edinburgh, Scotland to Chemische -Werke Huls AG. CMA 120127 534 Biles, Schroeder and Holdsworth REYNOLDS, R.W., SMITH, J.S. and STEINMETZ, I. (1974). Methyl Ethers as Motor Fuel Components. Presented at 168 National Meeting of Division of Petroleum Chemistry, American Chemical Society, Atlantic City, NJ. SNEDECOR, G.W. and COCHRAN, W.G. (1967). Chapter 10, One Way Classifications in Statistical Methods, 6th Edition. Ames: Iowa State University Press. Received June 12, 1987 Accepted August 4, 1987 CMA 120128 SUBCHRONIC INHALATION TOXICITY AND REPRODUCTIVE ASSESSMENT IN RATS OF THREE CHLORINATED PROPENES F. R. Johannsen, G. J. Levinskas Monsanto Company, St. Louis, Missouri G. M. Rusch, R. E. Schroeder Bio/dynamic5, Incorporated, East Millstone, New Jersey Croups of 75 male and 75 female Sprague-Dawley rats were exposed to 7 of 3 chloropropene (2,3-Di - DCP: 1,2,3-Tri - TRCP; and 1,1,2,3-Tetra - TECP) vapors to provide information on repeated exposures and the potential for reproductive impairment by the most likely route of occupational exposure. Target exposure concentrations were 0, 7, 5, and 15 ppm, 6 h/d, 5 d/wk for 13 wk. The following parameters were evaluated: pharmacotoxic signs, survival, body weights, hematology, clinical blood chemistry, urine analysis, gross and histopathology (over 40 tissues/rat), organ weights, and se lected weight ratios. Signs of nasal irritation were noted in rats exposed to 75 ppm of either DCP or TRCP but not TECP. Small decreases in overall body weight were ob served in female rats exposed to 15 ppm TCP. An increase 15%) in spleen weight, with no corresponding histopathological or clinical findings, was observed in 15 ppm DCP-treated male rats. No other effects considered related to treatment were observed following exposure to any of the three chlorinated propenes. Additional groups of 10 male and 20 female Sprague-Dawley rats were exposed to DCP, TRCP, or TECP vapors at target concentrations of 0, 1, or 5 ppm for 6 h/d, 5 d/wk for a 10-wk premating period, a mating period, and the first 14 d (females only) of gestation. Females were allowed to deliver litters and the offspring were evaluated during a 21-d lactation period. Mating, pregnancy, and fertility indices were generally comparable among all test groups, although female mating and pregnancy indices of both DCP-treated females were lower than expected in the regular and postrecovery reprdduction phase. No effects were seen on pup survival, sex distribution, body weights, organ weights, and ratios. A modest reduction in pup body weights was observed following TECP exposure but was attributed to large litter size. No treatment-related ejects were seen following necropsy of adults or weanlings, nor were such effects noted following microscopic evaluation of gonads from parental animals. Presented in part at the 68th Annual Meeting of the Federation of American Societies for Experimental Biology, St. Louis, Missouri, April 1-6, 1984. The authors wish to express their gratitude to Mrs. Diana Jones, Monsanto Co., for assistance in preparation of this manuscript. Present address for C. M. Rusch is Allied-Signal, Inc., Morristown, NJ, Requests for reprints should be sent to Dr. Frederick R. Johannsen, Monsanto Company, 800 N. Lindbergh Blvd., St. Louis, MO 63167. 291 Journal of Toxicology and Environmental Health, 33:291-302, 1991 Copyright ; 1991 by Hemisphere Publishing Corporation CMA 120129 292 F. R. IOHANNSEN ET AL INTRODUCTION Chlorinated propenes are important chemical intermediates used in industrial and agricultural chemical processes. Because of the volatility of this chemical class, inhalation of vapors in the workplace is considered to be one of the principal potential occupational exposure routes. A paucity of toxicity data by the inhalation route exists in the literature for chloropropene isomers other than the 1,3-dichloropropene isomer. A threshold limit value (TLV) of 1 ppm has been established for dichloropropene (isomers undefined) (ACGIH, 1986). Inhalation toxicity data for 1,3-dichloropropene include single 4-h LC50s of 729-900 ppnrfor Fischer 344 rats (Parker et al., 1982; Stott et al., 1988) and subchronic inhalation studies both with 1,3-dichloropropene (Torkelson and Oyen, 1977; Torkelson and Rowe, 1981; Stott et al., 1988; Hanley, 1987) and with a mixture of 1,3-dichloropropene and 1,2dichloropropane (Parker et al., 1982; Linnett et al., 1988). Lomax et al. (1989) also have reported on the chronic inhalation toxicity of technicalgrade 1,3-dichloropropene in rats and mice. The present study was designed to provide a comparative toxicologic profile of three chloropropene isomers not previously evaluated toxicologically, including the isomer of 2,3-dichloropropene (DCP), 1,2,3trichloropropene (TRCP), and 1,1,2,3-tetrachloropropene (TECP). For com parative purposes, each chlorinated propene was tested under a similar exposure regimen following subchronic inhalation exposure. Additional information was also derived on the ability of these chlorinated pro penes to affect reproduction/fertility in the rat at exposure levels that were anticipated to produce no more than minimal irritation. In preparation for this study, a series of range-finding studies were used for subsequent dose level selection. In 1-mo inhalation pilot stud ies, groups of 5 male and 5 female Sprague-Dawley rats were exposed 6 h/d, 5 d/wk to vapors of either 2,3-DCP, 1,2,3-TRCP, or 1,1,2,3,-TECP (Mon santo Company, unpublished observations, 1976). Each chlorinated pro pene was evaluated at target levels of 0, 5, 20, or 100 ppm. Deaths related to treatment resulted in groups of rats exposed to the high dose (100 ppm) of both TECP (1M, 2F) and TRCP (5M, 5F). Reduced mean body weights were observed primarily in female rats, in treated groups ex posed to mid and high levels of DCP and TRCP and all levels of TECP. Signs of irritation (droopy eyelids, alopecia) generally correlated to those test levels with observed weight loss. Exposure to TRCP produced the most significant irritation (droopy lids, lacrimation, red nasal discharge). Only gross pathologic, not histopathologic, examinations were per formed. Noteworthy findings included red nasal and anal discharge on 2 high-dose (100 ppm) TRCP-treated males and the failure of lungs to col lapse, likely due to pulmonary edema, in those rats that died on test after exposure to 100 ppm TECP. CMA 120130 SUBCHRONIC/REPRODUCTIVE EFFECTS OF CHLOROPROPENES 293 METHODS Test Material and Atmosphere Generation Each of the chlorinated propenes used in these studies were ana lyzed by gas chromatography to verify purity, as follows: DCP, >99%; TRCP, 95%; TECP, >99%. Each test article was placed in a 30-ml midget impinger while nitro gen gas was passed through the vessel. The resultant vapor was directed into a 760-1 dynamic airflow inhalation chamber system and diluted with room air to produce-the desired exposure concentrations. Chambers were operated at average flow rates of 130 l/min for DCP and TECP expo sures and at 142 l/min for TRCP. This provided the theoretical 99% equili bration times of 27 and 25 min, respectively (Silver, 1946). After 2 wk of exposure the delivery system for TRCP was modified to one using a me tering pump to assure uniform generation of the cis and trans isomers. In this system, the test article was placed in a glass syringe and metered into the side tube of a 250-ml filtering flask using a Sage model 341 syringe pump. This flask was warmed on a heating mantel with an autotransformer. Nitrogen gas was then passed through the flask and the resultant vapor was directed into the inhalation chamber. In developing the infrared chamber monitoring conditions, scans were made on each material, and then a wavelength most sensitive for each test agent was selected. The cell pathlength was determined as the one giving the best response over the range of exposure levels. The conditions selected are given in Table 1. Chamber concentrations were determined intermittently at least three times daily by direct analysis of chamber samples using a Miran model IIA infrared (IR) analyzer (Wilks Instrument Co., Darien, Conn.). Exposure levels were determined by comparison of the observed absor bance of these samples to a previously determined calibration curve gen erated using the same test material under the similar instrumental set tings. Estimation of the consistency of the IR was obtained by periodic comparison of daily nominal concentrations throughout the course of each study. No significant differences were observed that indicated a change in the characteristics of the IR monitor. The IR measurements for the low-dose (1 ppm) TRCP concentrations appeared to be subject to water-vapor interference resulting in an over- TABLE 1. Exposure Conditions Compound 2,3-DCP 1,2,3-TRCP 1,1,2,3-TECP Wavelength (^m) 11.2 12.6 10.9 Pathlength (dial reading) 13.0 14.0 9.0 CMA 120131 294 F. R. JOHANNSEN ET AL estimation of actual exposure concentrations. A colorimetric technique using 4-p-nitrobenzyl pyridine reagent (Rusch et al., 1976) was used to independently determine chamber concentrations at this test level. From these results it was concluded that some water-vapor interference oc curred and that actual concentrations were even closer to target levels than indicated by IR. No such problems were encountered at other TRCP levels. Animals Charles River Sprague-Dawley-derived rats were acclimatized and as signed to study according to a random-number table procedure. Rats for the subchronic phase were 7 wk of age at study inception, while those used in the reproduction phase were 62 d old. Rats were pair-housed in stainless steel wire caging during exposure and singly housed at all other times, except during mating periods. Water and rodent feed (Purina Lab oratory Chow, St. Louis, Mo.) were available ad libitum during nonexpo sure periods. Experimental Evaluation of Animals from 13-Wfeek Toxicity Studies Groups of 15 male and 15 female rats were exposed to target concen trations of 0 (concomitant control), 1, 5, or 15 ppm of the respective chloropropene for 6 h/d, 5 d/wk for 13 wk. All animals were observed for survival twice daily and given a de tailed physical examination on a weekly basis throughout the study pe riod. Individual body weights were recorded weekly from 10 d prior to exposure through termination. Clinical parameters were determined for all rats in the control and the highest dose groups (15 ppm) exposed to each chloropropene. Measurements were taken at the study midpoint (d 44-45) and for all surviving animals in all test groups just prior (d 85-88) to term. These parameters included hematology (hemoglobin, hematocrit, erythrocytes, clotting time, and total and differential leukocytes) and serum chemistry (serum glutamic pyruvic transaminase, alkaline phosphatase, urea nitro gen, and glucose). Urinalysis (gross appearance, specific gravity, pH, pro tein, bilirubin, ketones, glucose, occult blood, and sediment) was per formed on all animals in each of the 15-ppm chloropropene groups and the control group on study d 47-48 and again on d 89. Complete necrop sies were performed on all animals sacrificed at study term. Organ weights and organ/body weight ratios were recorded/calculated for brain, gonads, heart, kidneys, liver, lungs, pituitary and spleen. Micro scopic examination of the following tissues were performed for all con trol and high-dose (15 ppm DCP, TCP and TECP) animals: adrenals, bone marrow, brain, eye, gonads, heart, colon, duodenum, ileum, kidneys, CMA 120132 SUBCHRONICREPRODUCTIVE EFFECTS OF CHIOROPROPENE5 295 liver, lung, lymph nodes, mammary gland, pancreas, pituitary, salivary gland, skeletal muscle, skin, spinal cord, spleen, stomach, thyroid, uri nary bladder, uterus, prostate, and any gross lesions or tissue masses possibly attributable to treatment. Experimental Evaluation of Animals from Reproduction Study Groups of 10 male and 20 female rats were exposed to nominal con centrations of 0 (concomitant control), 1, or 5 ppm of either DCP, TRCP, or TECP. Test concentrations were selected such that no more than mini mal irritation would result, inasmuch as this toxicological parameter would, in itself, significantly limit the potential level of repeated expo sure occupationally. Exposures occurred in the same chambers with rats in the subchronic phase. Rats in the reproduction phase, however, were exposed for a 10-wk premating period, a mating period (not in excess of 30 d), and from d 0 through d 14 (day exposure stopped) of gestation (females only). All exposures were 6 h/d, 5 d/wk. During the mating period, male and female rats from the same treat ment group were cohoused nightly, initially one male per two females. Males were cohoused with females up to 10 consecutive days. If mating had not occurred following this interval, females were paired with a dif ferent male for a second or if necessary even a third 10-d interval. Fe males were examined each morning for evidence of mating (vaginal plug or sperm). Gestation d 1 was defined as the day mating was confirmed. Mated females were housed individually for the duration of gestation. On d 19 of gestation, dams were provided nesting material and were examined twice daily for signs of parturition. The day all pups were deliv ered was defined as day 0 of lactation. All litters were weaned on d 21. Both DCP-treated groups contained a higher frequency of female rats that had not mated or did not bear litters after the full 30-d mating cycle. These rats (6 per group) were held for a 60-d period during which they incurred no exposures to DCP. Then each female was caged with a proven male rat of the same strain from an in-house breeding colony. Each female was caged with the same male nightly for 10 consecutive days. Females unmated after this interval were caged with a different male nightly for an additional 10-consecutive-day period. Mating and pregnancy were determined, and if pregnant the animal entered the study phase as outlined previously. All adult rats were observed twice daily for overt signs of toxicity and survival. Detailed physical examinations were made weekly. Individual body weights for both adult male and female rats were recorded weekly throughout the study, beginning at initiation of the 10-wk premating pe riod. Mated females were weighed on d 0, 6, 15, and 20 of gestation and on d 0, 4,14, and 21 of lactation. Litters were observed daily for the presence of dead pups, which CMA 120133 2% F, R. (OHANNSEN ET AL were recorded. Surviving pups of each sex were weighed on d 0,1,4,14, and 21 of lactation. All adult male rats on study were sacrificed after completion of the 30-d mating period; chloropropene exposures were terminated after mat ing. Mated females were sacrificed at weaning or 26 d postmating if pregnancy was not observed. Female rats from the DCP-treated groups were retained for a posttreatment period described earlier. Complete necropsy was performed on each adult animal with particular attention given to the reproductive tract. The following organ weights were re corded and body-weight ratios calculated: brain, gonads, epididymides, spleen, heart, lungs, kidneys, and liver. Microscopic evaluations of the ovaries, testes, and/or epididymides were made for each adult animal. Offspring were killed at weaning (d 21) and a thorough gross necropsy was performed. Sex was determined by internal inspection of the go nads. Statistical Methods Body weights, organ weights, and weight ratios from compoundtreated groups were compared to controls using Dunnett's multiple comparison test (Dunnett, 1964). Hematology, clinical chemistry parame ters, and gestation length, offspring weight, and number of offspring from treated groups were compared to controls using appropriate tests of variance (F~test), and when significant, the Student's f-test was em ployed as modified by Cochran (Snedecor and Cochran, 1967). Survival indices (offspring, litter, mortality), mating indices, pregnancy rates, and fertility indices were compared using the chi-square method of analysis (Snedecor and Cochran, 1967). RESULTS Analytical Concentrations Chamber sample analyses indicated that rats in the 1, 5, and IS ppm test groups were actually exposed to the following cumulative mean ana lytical concentrations (SD): DCP: 1.3 (0.4), 4.9 (0.7), and 15.0 (1.4) ppm TRCP: 2.8 (0.9), 5.5 (1.8), and 14.9 (2.4) ppm TECP: 1.0 (0.3), 5.0 (1.4), and 14.3 (2.3) ppm The vast majority of daily analytical values were within 5%, and no overlap between test levels was observed. CMA 120134 SUBCHRONIC/REPRODUCTIVE EFFECTS OF CHLOROPROPENES 297 Subchronic Study Phase Daily Observations and Mortality With the exception of a single acci dental death occurring in the TECP high-dose group during blood draw ing on test d 45, all other test animals survived the duration of the study. No abnormal behavioral reactions were noted in any of the treated ani mals. Red nasal discharge and yellow staining of the anogenital fur were noted with increased frequency in the 15 ppm DCP and TRCP groups, and both the 5 and 15 ppm TECP test groups. A 9% decrease in body weight gain for female rats in the 15 ppm TRCP group was observed at the conclusion of the study. No other effects on animal body weight gain related to chloropropene treatment were seen. Hematology, Serum Chemistry, Urinalysis A few scattered observa tions of statistically significant differences between control and individ ual exposed groups were observed in hematology, serum chemistry, and urinalysis parameters measured at study term. However, none appeared to be outside of normal biological limits, nor did any appear to be expo sure related. Organ Weights and Weight Ratios Elevated mean, absolute (12%) and relative (15%), spleen weights were observed in terminally killed male rats exposed to 15 ppm DCP. Similar findings were not evident in groups of female rats at this exposure level, nor were similar findings observed in any lower DCP-treated groups. The lack of any significant changes in the hematological profile and the lack of corresponding histopathological changes in these 15 ppm DCP-treated male rats suggest these find ings are unrelated to treatment. Analysis of remaining absolute and rela tive organ weights from other DCP-, TRCP-, and TECP-treated groups resulted in only sporadic, statistically significant differences between var ious treated and control groups, with no apparent dose- or treatmentrelated trend evident. Cross and Microscopic Pathology No differences were observed in gross pathology findings between control rats and those treated with DCP, TRCP, or TECP. Microscopic examination of animals of both chloropropene-treated and control groups revealed no changes that could be correlated to 13-wk exposure to DCP, TRCP, or TECP. Reproduction Phase Parental Generation Observations and Mating Indices A single male rat in the 1 ppm DCP group was killed in a moribund state prior to scheduled sacrifice. No treatment-related differences were observed in group body weights between chloropropene-treated (DCP, TRCP, and TECP) and control animals during the premating phase or in females during gestation and lactation. Small differences in mating indices (male and female) were observed in both DCP-treated groups when compared to control values (Table 2). CMA 120135 298 F. R. IOHANNSEN ET AL TABLE 2. Mating and Fertility Indices of Rats Exposed to 2,3-Dichloropropene (OCP), 1,2,3Trichloropropene (TRCP), or 1,1,2,3-Tetrachloropropene (TECP) Vapors Mating Fertility Vapor ppm Female* Male' Female6 Male' Control 0 20/20 9/10 19/20 DCP 1 5 17/20 6/10 14/17 18/19* 8/10 14/18 TRCP 1 - 19/20 5 20/20 8/10 17/19 8/10 20/20 TECP 1 5 20/20 20/20 9/10 17/20 9/10 20/20 9/9 6/6 8/8 8/8 8/8 9/9 9/9 Note. No significant differences between treated and control groups. 'Number mated vs. group total. ^Number pregnant vs. number mated. 'Number of males impregnating vs. number mated. ^Excludes consideration of one hermaphrodite. However, none of these differences were statistically significant. No ap parent effects in similar indices were seen in TRCP- or TECP-treated rats. With the exception of the DCP-treated groups, pregnancy and impregna tion indices appeared normal in all treated groups (Table 2). Posttreatment Recovery Croup Six female rats, all from the 1 and 5 ppm DCP groups, that either did not mate or conceive during the 30-d mating period were retained for a 60-d posttreatment recovery period. Several of these individuals mated, became pregnant, and bore normal litters. However, both the mating index and the pregnancy rate for each group remained somewhat lower than expected (Table 3). No clear evi dence of a dose response was seen, nor were there any physical findings at necropsy that were indicative of a treatment-related effect. Litter Data Gestational length, fetal viability at birth, litter size (Table 4), and litter survival (Table 5) did not appear adversely affected by DCP, TRCP, or TECP exposures. TABLE 3. Mating and Pregnancy Indices of 2,3-Dichloropropene (DCP) Treated Rats Cohabitated with Proven Males in the Postre covery Mating Phase Number of females DCP (ppm) Cohabitated Mated Pregnant 16 A2 S 5' 3 1 'Excludes hermaphrodite. CMA 120136 SUBCHRONIC/REPRODUCTIVE EFFECTS OF CHLOROPROPENES 299 TABLE 4. Gestation Length, Fetal Viability at Birth, and Litter Size of Offspring from Rats Exposed to 2,3-Dichloropropene (DCP), 1,2,3-Trichloropropene (TRCP), or 1,1,2,3Tetrachloropropene (TECP) Vapors Vapor ppm Gestation length (d) Fetal viability Alive Dead Control DCP TRCP TECP 0 1 5 1S 1 5 22.2 21.B 22.1 22.1 22.1 21.9 21.9 10.2 10.9 10.6 10.8 10.9 11.1 13.8' 0.3 0.2 0.4 0.1 0.2 0.4 0.2 'Significantly (p ,01) different from control. Necropsy and Organ Weights The only remarkable gross lesion ob served after necropsy of control and chloropropene-treated adult rats was that of a hermaphrodite in the 5 ppm DCP-treated female group. This animal had a fully formed uterus with primordial epididymis, testes, and seminal vesicles. Gross lesions were confirmed microscopically. This observation is considered representative of a congenital malformation and not treatment related. This animal did not mate throughout the course of both the full and postrecovery portion of the study. Histopathology Microscopic examination of the gonads of DCP-, TRCP-, and TECP-treated rats was generally unremarkable. Sporadic, sub tle microscopic findings (mild focal testicular atrophy) were found in the testes of two males from the low-dose 2,3-DCP group in the reproduction phase. No such findings were observed at the next higher (5 ppm) test level in this study phase or even at 15 ppm in the subchronic phase. Thus, no dose-response relationship was established. This type of lesion TABLE 5. Survival and Growth Indices of Offspring from Rats Exposed to 1,1,2,3Tetrachloropropene (TECP) Vapors TECP (ppm) Mean number (SD) of pups alive at Birth d 21 0 Mean pup weight (g SD) at d 4 14 21 0 10.2 + 3.4 10.2 2.1 6.5 0.7 10.7 1.5 27.7 3.7 45.6 7.4 1 11.1 3.1 10.6 2.9 6.3 0.5 10.3 1.5 26.8 4.9 43.2 7.7 5 13.8 2.1' 12.6 2.5' 6.2 0.7 9.4 1.2' 24.4 4.1* 39.7 6.2* 'Significantly different from control, p < .01. ^Significantly different from control, p < .05. CMA 120137 300 F. R. JOHAN NSEN T AL occurs spontaneously with increasing frequency in rats in an age-related fashion (Heywood and James, 1985). Due to the protocol design, these males were of an age (-6 mo) before termination when sucn lesions begin to manifest themselves. An evaluation of several studies of >6 mo duration at this laboratory confirmed the age change of this lesion in the CD rat. Thus, this lesion is considered to be spontaneous in nature and unrelated to compound administration. No such effects were noted in other TRCP- or TECP-treated groups. Weanling Generation Offspring Data and Necropsy Mean pup survival of chloropropenetreated groups was comparable to control levels, with one exception. Pup survival from dams administered 5.ppm DCP was significantly lower than control values after 1 and 4 d of lactation. These differences were attributable to the loss of a single litter before d 1 of lactation in a dam for which d 0 of gestation had not been confirmed. This single incident was judged to be unrelated to treatment. Significant reductions in group mean body weights were observed on lactation d 4, 14, and 21 from pups whose dams were exposed to 5 ppm TECP (Table 5). This study group also exhibited a significantly (p < .01) increased number of pups delivered at d 0 and remaining alive at weaning (lactation d 21). Inasmuch as litters were not culled, it is possible that reduced pup weights at the 5 ppm TECP level were reflective of a reduced nutritional status due to the increased survivorship in this study group. Mean pup weights have been shown to decrease with increasing litter size. Khera et al. (1989) reported a difference of approximately 5 g between mean litter weights of dams nursing 10 pups per litter versus 14 pups per litter. This 5-g disparity would account for the entire weight differential observed in the present study observed through the 21-d lactation period. No other effects on pup weight or weight gain were observed in other chloropropene-treated groups. Sex distribution ob served in this study was judged to have exhibited normal distribution. Necropsy of pups found dead during lactation and killed on d 21 of lactation revealed no treatment-related effects. DISCUSSION Lethal concentrations of the 1,3-DCP isomer in laboratory animals can produce lung, liver, and kidney injury (Torkelson and Oyen, 1977). However, multiple exposures to 1,3-DCP or a DCP-containing mixture at lower exposure levels have resulted only in slight toxicological effects in several mammalian species (Highman and Heppel, 1946; Torkelson and Oyen, 1977; Torkelson and Rowe, 1981). Rats, guinea pigs, rabbits, and dogs exposed to up to 3 ppm 1,3-DCP for 6 mo exhibited no histopatho- CMA 120138 DUE TO CLERICAL ERROR IN THE BATE LABELING PROCESS. THIS BLANK PAGE HAS BEEN INSERTED TO MAINTAIN BATES PAGE ORDER. THIS PAGE IS NOT A DOCUMENT. CMA 120139 SLBCHRONIC/REPRODUCTIVE EFFECTS OF CHLOROPROPENES 301 logic effects, except fqr slight, apparently reversible changes in the kid neys of male rats at the highest test level (Torkelson and Oyen, 1977). Subsequent data from the same laboratory (Torkelson and Rowe, 1981) reported no gross or microscopic changes in either the kidney or liver (or any other tissue or organ) in mice or rats exposed to 1,3-DCP concen trations of 12-93 ppm for 13 wk. More recent data from that facility, using Fischer 344 rats and 86C3F1 mice, have shown depressed growth rates and degenerative change of the respiratory tract in animals exposed to 1,3-DCP levels at and above 90 ppm for 13 wk. Hyperplasia of the urinary bladder epithelium was also observed in female mice. The results observed in the present comparative study with three chloropropene derivatives are consistent with subchronic inhalation tox icity results previously reported for 1,3-DCP. In the present study, no specific target organ toxicity was observed with 2,3-DCP, TRCP, or TECP when administered at atmospheric concentrations up to 15 ppm for 3 mo. Specifically, no changes in organ weights, gross or microscopic pa thology, or clinical chemistry findings indicative of a toxicologic effect were observed in any of the tissues or organs for which toxicologic le sions had been reported after high-level acute exposure with 1,3-DCP. The slight splenic weight increase seen with 2,3-DCP in the present study is of questionable significance since it was not accompanied by corre sponding hematologic or pathologic changes. The fact that this organ has not been reported as a target site in previous studies or with other chlorinated propenes in this study supports a contention that this find ing is unrelated to exposure. Findings in the reproductive phase of this study have shown no re productive and target tissue (testes) effects similar to that of a halogenated alkane, dibromochloropropane (DBCP). In the case of DBCP, ef fects were noted in male germinal tissue that affected male fertility (Rao et al., 1983). This lack of reproductive target tissue toxicity is consistent with previous subchronic reports for 1,3-DCP (Torkelson and Oyen, 1977; Stott et al., 1988). The lack of effects on male fertility observed after treatment with DCP, TRCP, or TECP is also consistent with results of a two-generation inhalation reproduction study in rats with the 1,3-DCP isomer (Breslin et al,, 1989). The lower-than-expected mating frequency seen with DCP-treated females, while not significantly different from controls, is of possible relationship to treatment in light of the reduced mating/pregnancy frequencies also seen during the postrecovery period. Still, no apparent physical aberrations related to treatment were ob served at necropsy to resolve this issue. No such effects were apparent following either TRCP or TECP exposure at either exposure level evalu ated. The results of the present study extend our knowledge regarding the subchronic toxicologic effects of chlorinated propenes. The lack of sig nificant toxicity observed at levels greater than those expected to be CMA 120140 302 F. R. JOHANNSEN FT AL tolerated because of irritation from occupational exposure thus supports the present dichloropropene TLV and suggests that this occupational standard would possess an acceptable margin of safety for TECP and TRCP as well. REFERENCES American Conference of Governmental Industrial Hygienists. 1986. Documentation of the Thresh old Limit Values and Biological Exposure Indices, 5th ed. Washington, D.C.: ACGIH. Bresiin, W. |.t Kirk, H. D., Streeter, C M., Quast, J. F., and Szabo, J. R. 1989. 1,3-Dichloropropene: Two generation, inhalation reproduction study in Fischer 344 rats. Fundam. Appl. Toxicol. 12:129-143. Dunnett, C. W. 1964. New tables for multiple comparisons with a control. Biometrics 20:482-491. Hanley, T. R., Jr,, John-Green, J. A., Young, J. T., Calhoun, L. L, and Rao, K. S. 1987. Evaluation of the effects of inhalation exposure to 1,3-dichloropropene on fetal development in rats and rab bits. Fundam. Appl. Toxicol. 8:562-570. Heywood, R., and James, R. W. 1985. Current laboratory approaches for assessing male reproduc tive toxicity: Testicular toxicity in laboratory animals. In Reproductive Toxicology, ed. R. L Dixon, pp. 147-160. New York: Raven Press. Highman, B., and Heppel, L A. 1946. Toxicology of 1,2-dichloropropane (propylene dichloride). III. Pathologic changes produced by a short series of daily exposures. Arch. Pathol 42:525-534. Khera, K. 5., Grice, H. C., and Clegg, D. J., eds. 1989. Chapter IV. Multigeneration studies. In Current Issues in Toxicology, pp. 21-40. New York: Springer-Verlag. Linnett, S. L., Clark, D. C., Blair, D., and Cassidy, S. L. 1988. Effects of subchronic inhalation of D-D (1,3*dichloropropene/1 2-dichloropropane) on reproduction in male and female rats. Fundam. Appl. Toxicol. 10:214-223. Lomax, L. G., Stott, W. T., Johnson, K. A., Calhoun. L. L-, Yano, B. L., and Quast, ). F. 1989. The chronic toxicity and oncogenicity of inhaled technical grade 1,3-dichloropropene in rats and mice. Fundam. Appl. Toxicol. 12:418-431. Parker, C. M., Coate, W. B., and Voelker, R. W. 1982. Subchronic inhalation toxicity of 1,3dichloropropene/1,2-dichloropropane (D-D) in mice and rats. J. Toxicol. Environ. Health 9:899910. Rao, K. S., Burek, J. D., Murray, F. J,, John, J. A., Schwetz, B. A., Bell, T. J., Potts, W, J,, and Parker, C. M. 1983. Toxicologic and reproductive effects of inhaled 1,2-dibromo-3-ch!oropropane in rats. Fundam. Appl. Toxicol. 3:104-110. Rusch, C. M., La Mendola, S. L., Katz, C. V., and Laskin, S. 1976. Determination of low levels of dimethylcarbamoyl chloride in air. Anal. Chem. 4B:2259-2261. Silver, S. D. 1946. Constant flow gassing chambers: Principles influencing design and operation. /. Lab. Chn. Med. 3T.11S3-1161. Snedecor, G. W,, and Cochran, W, G. 1967. Statistical Methods, 6th ed. Ames: Iowa State University Press. Stott, W. T., Young, J. T., Calhoun. L. L., and Battjes, J. E. 1988. Subchronic toxicity of inhaled technical grade 1,3-dichloropropene in rats and mice. Fundam. Appl. Toxicol. 11:207-220. Torkelson, T. R., and Oyen, F. 1977. The toxicity of 1,3-dichloropropene as determined by repeated exposure of laboratory animals. Am. Ind. Hyg. Assoc, ]. 38:217-223. Torkelson, T. R., and Rowe, V. K. 1981. Halogenated aliphatic hydrocarbons containing chlorine, bromine and iodine. In Patty's Industrial Hygiene and Toxicology, eds. G. D. Clayton and F. . Clayton, Vbl. IIB, pp, 3433-3601. New York: Wiley. Received lune 14, 7990 Accepted February 77, 7997 JOURNAL OF THE AMERICAN COLLEGE OF TOXICOLOGY Volume 12, Number 2, 1993 Muj Ann LJcbcrt, Inc., Publisher* A Reexamination of Liver Tumors in Mice Exposed to Wholly Vaporized Unleaded Gasoline RENAE I. MAGAW.' WARD R. RICHTER,2 and JUDITH A. MACGREGOR3 ABSTRACT A single lifetime inhalation study of the effects of wholly vaporized unleaded gasoline has been conducted in the mouse. An elevation of hepatocellular tumors was reported only in females and only at the high treatment level (2056 ppm). Several reports of this study have appeared previously in the scientific literature, including a comprehensive summary published in this journal. Different incidence rates were reported for mouse liver tumors in these publications without explanations of how they were calculated of why they differed. To clarify the data, we recently examined the final report of the gasoline study, the individual animal data, and slides prepared from liver tissue collected during the study. The previously reported incidence rates did not take into consideration all animals at risk for tumor development or all animals with hepatocellular tumors. Revised incidence rates are presented for hepatocellular tumors in all mice on study. These rates are approximately 20-25% lower than previously reported. INTRODUCTION lifetime inhalation study of the effects of wholly vaporized unleaded gasoline on mice was conducted by the International Research and Development Corporation (IRDC, 1983) and sponsored by the American Petroleum Institute. The results of the study indicated that wholly vaporized unleaded gasoline induced a statistically significant compound-related increase in hepatocellular tumors in female mice. No increase was observed in male mice. The study was originally reported by MacFarland et al. m this journal (1984) and elsewhere (1982. 1984), and it was reviewed extensively by the U.S. Environmental Protection Agency (1987) and other organizations (HEI. 1985; NESCAUM, 1989). Slightly different incidence figures were presented for mouse liver tumors in the published reports. The differences could not be explained on the basis of information presented in the articles and raised questions about why the published rates differed. The data and the reports in question have been thoroughly reviewed and the actual liver tumor incidence observed in mice in this study was determined. IRDC STUDY SUMMARY AND PUBLISHED INCIDENCE DATA IRDC (1983) conducted a 2-year study in which groups of 100 male and 100 female B6C3F, mice were exposed to 0, 67, 292, and 2056 ppm of wholly vaporized unleaded gasoline by whole body inhalation. Interim sacrifices '1CF Kaiser Engineers, 1800 Hamson Sireei. 7th Floor, Oakland. CA. IPharmaco-L5R. East MiUstone, NJ 'Chevron Research A Technology Co , P O Box 4054. Richmond, CA 195 CMA A 20A 42 MAGAW ET AL. were conducted after approximately 3.6, 12, and IS months of exposure. Terminal sacrifices were conducted after 103-107 weeks for maie mice and 113 weeks for female mice. The incidence rates for liver tumors have been reported sevdral times, as shown in Table 1 (MacFarland. 1982;MacFarlandetal., 1984; NESCAUM, 1989; EPA. 1987). Hepatocellular tumors were defined as adenomas and carcinomas in these publications. LIVER TUMOR INCIDENCE REVIEW In order to determine the true incidence of liver tumors observed in the study and the reasons why previously reported incidence rates vary, the 1RDC study history and final report were carefully examined. In addition, an independent histopathological review of liver sections was conducted. Because the original slides could not be located by IRDC, paraffin blockscontaintng liver tissue were obtained from the laboratory. The blocks were rectn, sections were stained with hematoxylin and eosin. and new slides were prepared. The criteria utilized in the histopathological review for the diagnoses of hepatocellular adenoma and carcinoma were as summarized by Brooks and Roe (1983) and by Popp (1983) for adenomas and carcinomas, respectively. Briefly, adenomas were sharply demarcated and they compressed adjacent liver parenchyma. The cytology of the cells varied, with cytoplasm that was eosinophilic, besophilic, clear, or vacuolated. Carcinomas were character ized by some or all of the following; an increased degree of pleomorphism. greater nuclear/cytoplasmic ratio, trabecular plates several cells thick, and increased mitotic activity. Some were large with irregular or invasive growth patterns. The review of the IRDC final report indicated that the previously reported incidence rates do not take into consideration all mice at risk of tumor development or all animals with hepatocellular tumors. The first hepatocellular tumor was observed at the 12-month interim sacrifice and thus all animals alive at that time should be considered at risk for tumor development and should be included in the incidence rates. The rates reported by MacFarland, NESCAUM. and those used by the EPA for estimating the carcinogenic potency of unleaded gasoline correspond to the incidence of hepatocellular turnon in mice that died or were killed during the 18-month to terminal sacrifice penod. Animals that were killed or that died on study prior to 18 months were not included. The Table 1. Primary Hepatocellular Neoplasms Incidence Rates Reported in Mice Treatment level 0 ppm 67 ppm 292 ppm EPA, pages 5-14* EPA. pages 3--15* MacFarland* NESCAUM* Revised incidence' 8/57 (14)*' 8/100(8) 8/57 (14) 8/57 (14) 8/79 (10) Female mice 10/52 (19) 10/100(10) 10/52 (19) 10/52 (19) 13/76 (17) 13/57 (23) 12/100(12) 12/57 (21) 12/57 (21) 12/79 (15) MacFarland* NESCAUM* Revised-incidence' 23/51 (45) 23/51 (45) 27/75 (36) Male mice 15/42 (36) 15/42 (36) 18/70 (26) 20/44 (45) 20/44 (45) 28/71 (39) Incidence used lo estimale carcinogenic potency in EPA (1987). ''Percentages are reponed in parentheses. `Incidence reported in summary section of EPA (1987). "Incidences reported in MacFarland (1982, 1984) and MacFarland el al (1984), 'Incidence reported in NESCAUM (1989) 'Corrected incidence determined from this review. 2056 ppm 28/56 (50) 27/100(27) 27/56 (48) 27/56 (48) 29/78 (37) 24/54 (44) 24/54 (44) 27/77 (35) UNLEADED GASOLINE AND MOUSE LIVER TUMORS summary incidence figures reported for female mice by EPA differ. In this case, the numerators are based on mice that died during the 18-month to terminal sacrifice period, while the denominators represent the total number of mice initially started on test, including those that died prior to 12 months. These animals should not be considered at risk of tumor development and should not be included in the incidence rates. The study history provides several possible explanations for why the data were reported incorrectly. The histopathologic evaluation of the study was carried out by several pathologists at different laboratories. Separate reports for the different sacrifice groups were prepared and reviewed at different times. The preliminary draft terminal sacrifice report included data for animals that died on study after the 18-month interim sacrifice along with the data for animals sacrificed at the end of the experiment. It did not include or summarise any data for animals that died prior to these times. This report was reviewedI discussed, and summarized prior to completion of the final IRDC report. When the final report was issued for the entire study many months later, the various reviewers did not go back and incorporate data for animals that died previously. In addition to problems in tabulating the data, there was a change in the terminology for proliferative liver lesions used during the course of the study that resulted in some mice with hepatocellular tumors being excluded from the reported incidence rates. The diagnosis of neoplastic nodule was used at the 12- and 18-month interim sacrifices, but not at other time periods. This term was proposed by Squire and Levin (1975) to describe projiferative lesions only in the rat. It was proposed because of the difficulty in differentiating between hyperplastic and neoplastic lesions in the rat on morphologic grounds. The term neoplastic nodule is not typically used to describe mouse tumors, but was used for diagnosing lesions in this study for mice at these two interim sacrifices. Animals diagnosed with neoplastic nodules were not included in the incidence figures reported by MacFartand. NESCAUM, or EPA. Lesions diagnosed as neoplastic nodules by the original pathologist were reevaluated as pan of this review and their morphology was found to be consistent with the diagnosis of hepatocellular adenoma. The morphology of these lesions was also consistent with that of the lesions diagnosed as hepatocellular adenoma at later times in the study and animals with neoplastic nodules should be included in the tumor incidence figures. REVISED TUMOR INCIDENCE Tables 2 and 3 present the total number of animals with hepatocellular tumors by tumor type and for all tumor types combined for female and male mice, respectively. The overall revised incidence rates presented for all hepatocellular tumors include all animals considered to be at risk for tumor development (i.e., all animals alive at 12 months) and all animals with primary liver tumors, including those originally diagnosed as neoplastic nodules. Table 2. Primary Hepatocellular Neoplasms in Female Mice: Numbers of animals with Tumors and Revised Incidence Treatment level 0 ppm 67 ppm 292 ppm Neoplastic nodule Adenoma Carcinoma Total benign Total benign and malignant With neoplastic nodules Without neoplastic nodules Revised incidence 0 1 7 1 8 8 8/79 (10%) 3 4 6 7 13 10 13/76 (17%) 0 4* 9 4 12b i:b 12/79 (15%) "One mouse had an adenoma during the 18- to 24-month study period. bOne mouse had an adenoma and acarcinoma and is counted only f6r the most advanced lesion. 2056 ppm 1 8* 21 9 29" 28b 29/78 (37%) 197 CMA 120144 MAGAW ET AL. Table 3. Primary Hepatocellular Neoplasms in Male Mice; Numbers of Animals wrm Tumors and Revised Incidence Treatment level 0 ppm 67 ppm 292 ppm 2056 ppm Neoplastic nodule Adenoma Carcinoma Total benign Total benign and malignant With neoplastic nobules Without neoplastic nodules Revised incidence 4 12 12* 16 2r 24 27/75 (36%) 3 4 14 7 Iff1 15* 18/70 (26%) 5 5 19* 10 28* 23* 28/71 (39%) 2 5 21* 7 27* 25* 27/77 (35%) `One mouse had a carcinoma at the 12 month sacrifice (0 ppm) or during the 12- to 18-month period (2056 ppm). `'Three mice had a carcinoma during the 12- to 18-month period. 'One mouse had a neoplastic nodule and a carcinoma and is counted only for the most advanced lesion. ``Three mice had an adenoma and a carcinoma and are counted only for the most advanced lesion. 'One mouse had an adenoma and a carcinoma and is counted only for the most advanced lesion. The revised incidence rate observed in female mice exposed to 2056 ppm wholly vaporized unleaded gasoline is approximately 25% lower than previously reported. The revised rate is the result of correcting several inconsistencies in tabulating and analyzing data from this study. Inherent problems associated with multiple pathologists conducting the microscopic evaluations further complicated the analysis and should be avoided. The incorrect incidence data have been used to characterize the potential risks associated with human exposure to gasoline by several regulatory groups (EPA. 1987; NESCAUM, 1989). These risks are based on a treatment-related effect observed only at the high dose level and, therefore, the incidence changes identified in this dose group could alter the estimated risks. This suggests that these risk assessments should be reevaluated. REFERENCES BROOKS. P.N., and ROE. FJ C. (1985). Digestive System. T.C. Jones. U Mohr, and R.D, Hum teds ). Spnnger-Veriag: Berlin, pp. 47-5 1. HEALTH EFFECTS INSTITUTE. (1985). Gesoline Vapor Exposure and Human Cancer. Evaluation of Existing Scientific Information and Recommendations for Future Research. Report of the HE1 Health Review Committee. INTERNATIONAL RESEARCH AND DEVELOPMENT CORPORATION. (1983). Motor Fuel Chrome Inhalanon Study. Vote. 1-0. Sponsored by the American Petroleum Institute. Sept IS. 1983. MacFARLAND, H N. (1982). Chronic gasoline tonicity, in: The Toxicology of Petroleum Hydrocarbons. Proceedings of a Symposium. May. 1982. H.N MaeFarland. C E. Holdswonh, J A. MacGregor. R.W. Call, and M L Kane(eds.): American Petroleum Institute: Washington, D C.. pp. 78-80. MacFARLAND. H.N. (1984). Xcnobiotic induced kidney lesions: Hydrocarbons The 90-day and 2-year gasoline studies, in: Advances in Modern Environmental Toxicology. Voi. 7, Penal Ejects of Petroleum Hydrocarbons. M.A. Mchlman. C.P. Hemslreet. J.J. Thorpe, and N K Weaver (eds ): Princeton Scientific Publishers, Inc.. Princeton. NJ, pp. 51-56 MacFARLAND. H N . ULRICH. C.R.. HOLDSWORTH, C.E., KITCHEN, D.N., HALLIWELL, W.H.. and BLUM. S.C. (1984). A chronic inhalation study with unleaded gasoline vapor J Am. Coll. Toxicol. 3:231-248. NORTHEAST STATES FOR COORDINATED AIR USE MANAGEMENT (NESCAUM) (1989) Evaluation of the health effects from exposure lo gasoline and gasoline vapors. Air Toxics Committee. POPP. J.A (1985) Digestive Systtm TC Jones. U. Mohr, and R.D. Hunt (eds ): Spnnger-Veriag' Berlin, pp 39--*5, 198 CMA120145 UNLEADED GASOLINE AND MOUSE LIVER TUMORS SQUIRE. R.A., and LEVITT. M.H. (1973). Repeat of a workshop in classification of specific hepatocellular lesnts in rati Cancer Rea. 35:3114-3223. US ENVIRONMENTAL PROTECTION AGENCY. (1917) Evaluation of the carcinogenicity Of unleaded gasoline Office of Health and Environmental Assessment WasJuapon, D C.. EPA/600/6-87-001. Address reprint requests to: Dr. Judith A. MacGregor Chevron Research and Technology Co. P.O. Box 4054 Richmond. CA 94804 199 CMA 120146 THE TOXICOLOGIST An Official Publication of the Society of Toxicology and Abstract Issue of FUNDAMENTAL AND APPLIED TOXICOLOGY An Official Journal of the Society of Toxicology Published by Academic Press, Inc. 978 AN INHALATION ONCOGENICITY STUDY IN RATS OF METHYLETHYLKETOXIME PE Newton. W L Wooding, W E Rinehart. Pharmaco LSR, East Millstone, NJ; Industrial Health Foundation, Pittsburgh, PA Fischer 344 rats (80/sex/group) were exposed 6 hrs/day, 5 days/week for 26 months via whole-body inhalation to MEKO (methylethylketoxime) vapor concentrations of 0,15 1 (X SD), 75 2 and 374 10 ppm. Survivorship at 26 months (34% in the males; 60% in the females) was unaffected by the MEKO exposures. Microscopically, MEKO related findings included: conges tion of the spleen with pigment in reticuloendothelial cells and extramedullary hematopoiesis, hepatocellular carcinoma in the male 374 ppm rats, a dose re lated increase in hepatocellular adenomas in the males at 75 and 374 ppm, an increase incidence of basophilic foci in males and females at 374 ppm (with increased severity in the males in the 15,75 and 374 ppm group males and 374 ppm group females). In the nasal turbinates, primarily the dorsal meatus, de generation of the olfactory epithelium was seen in both males and females at 374 ppm with less severe findings at 75 and 15 ppm. These changes consisted of thinning of the olfactory epithelium with some loss of apical cytoplasm and fewer layers present. In conclusion, under the conditions of this study, MEKO produced changes in the olfactory epithelium in all MEKO exposed groups and was a liver onco gen in male rats at 75 ppm. Abstracts of the 34th Annual Meeting Vol. 15, No. 1, March 95 CMA 120147 THE TOXICOLOGIST CHRONIC INHALATION STUDY OP COMMERCIAL HEXANE IN MICE. W 0 D&ughtrev. Exxon Biomedical Sciences, Inc., Ease Millstone, NJ. J S Duffy. Texaco, Inc., Beacon, NY. L 5 KaddocJc. Unocal Corp., Loa Angeles, CA. P W Kelly. Phillips Petroleum Co.,-Bartlesville, OK. T H Keenan. Ashland Chemical Co., Columbus, OH. W R Richter. Phamaco LSR, East Millstone, NJ. R A Rhoden. American Petroleua Institute, Washington, DC. The oncogenic potential of commercial hexane (CHJ was evaluated in 3tc,F. mice. The CH tested consisted primarily of six-carbon isomers in the following liquid volume percents: n-bexane (S2%), methylcyclcpentaaa (16%), 3-methylpentane (16%), 2-methylpentane (13%) and cyclohexane (3%) . CH was administered to four groups of 50 animals per sex for 6 hours/day, 5 days/wee)c for two years at target concen trations of 0, 900, 3000, and 9000 ppm in air. There were no significant differences in survivorship between control and CH-exposed group3. Body weight gain was reduced in bighdcse females, but not in females of the aidand low-dose groups nor in males. Microscopic examination revealed a treatment-related increase in hepatocellular neoplasms (adenomas and carcinomas) in females of the 9000 ppm group. There was also an increase in the incidence of pituitary proliferative changes in all groups of CK-exposed females compared to controls. A decrease in the severity and a slight decrease in the incidence of cystic uterine endometrial hyperplasia were observed in the high-dose females. No neoplastic changes were found in CK-exposed males. Abstracts of the 33rd Annual Meeting Vol. 14, No. 1, March 1994 THE TOXICOLOGIST CHRONIC INHALATION STUDY OP COMMERCIAL HSXAK3 IN RATS. D H Kelly. Phillips Petroleum Co., Bartlesville, OK. J S Duffy. Texaco, lac.. Beacon, NY. L S Haddock. Unocal Corp., Los Angeles, CA. w D Dauchtrev. Zxxoa Biomedical Sciences, Inc.,- Zast Millstone, NJ. T H Keenan. Ashland Chemical Co., Columbus, OH. ? S Nevton. Pharmaco LSR, Zast Millstone, NJ. R A Rhoden. American Petroleum Institute, Washington, DC. The oncogenic potential of commercial hexane CCH) was evaluated in Pischer 344 rats. The test sample of CH consisted primarily of the following six-carbon isomers: n-hexane (52%), nethylcyclcpentamc (16%), 3-methylpentaae (15%), 2-netbylpentane (13%), and cyclohexane (3%). CK was administered to four groups of 50 animals per sex for 6 hours/day, 5 days /week for two years at target concentrations of 0, 900, 3000, and 3000 ppm in air. There were no significant differences in survivorship between control and CH-exposed groups. 3ody weight gain was significantly reduced in both males and females from the 3000 and 9000 ppm groups. Histological evidence of mucosal irritation was observed in the nasal turbinates and larynx of CH-exposed males and females. The incidence of this finding was generally greatest in animals from the 9000 ppm group. No statistically sig nificant differences in overall or individual tumor incidences between control and exposed animals were observed. Thus, lifetime exposures to CH at levels up to 9000 ppm were not carcinogenic in Pischer 344 rats. Abstracts of the 33rd Annual Meeting Vol. 14, No. 1, March 1994 CMA 120149 INHALATION TOXICITY OF PHOSPHINE IN THE RAT: ACUTE, SUBCHRONIC, AND DEVELOPMENTAL Paul E. Newton Raymond E. Schroeder Pharmaco-LSR Inc. East Millstone, New Jersey Jeremiah B. Sullivan DECESCH AMERICA, Inc. Weyers Cave, Virginia William M. Busey Deborah A. Banas Experimental Pathology Labs, Inc. Herndon, Virginia Lethality is the primary hazard of phosphine exposures. All phosphine-related effects seen at sublethal exposure levels were relatively small and completely reversible ei ther during the exposure or during a recovery period. Acutely, phosphine exposures were lethal to female Fischer 344 rats at a cumulative concentration-time product of about 180 ppm-hr if the concentration were greater than 5-7 ppm. For duly 6-hr expo sures, the median lethal times were 3 days at 10 ppm and 4 days at 7.5 ppm. Thirteen daily 6-hr exposures to 5 ppm were not lethal. Decreased erythrocytes, lung conges tion, and increased kidney weights with coagulative necrosis of the tubular epithe lium in the outer cortex were seen in the 10 ppm rats only. The effects were more severe in females than in males. Subchronic exposures to 0.37, 1, or 3.1 ppm of phos phine were conducted. Ten animals per sex per group were sacrificed after 4 and 13 weeks of exposure and 4 weeks of recovery. These exposure produced a dose-related decrease in body weight gain at 1 and 3 ppm. Food consumption was decreased at 1 and 3 ppm and transiently in the 0.37 ppm group. Five percent decreases in erythro cytes, hemoglobin, and hematocrit were seen in the 3 ppm group after 13 weeks of exposure. All effects seen in the subchronic study were completely reversible either dunng the 13-week exposure or the 4-week recovery period. Exposure of pregnant CD* rats (24 per group) to 0.03, 0.33, 2.8, or 4.9 ppm of phosphine for 6 hr/day over the davs 6-15 gestation interval was not maternally or developmentally toxic This research was supported by the Metal Phosphide Task Force. The authors gratefully ac knowledge the technical assistance of Doris Bowden, Ellen Whiting, Brian Pillsbury, and Albert Couchman. Requests for reprints should be sent to Dr. Paul E. Newton, Bio/Dynamics, Inc., Box 2360, East Millstone, NJ 08875-2360. 223 Inhalation Toxicology, 5(21:223-239, 1993 Copyright 2 1993 Taylor & Francis 0895-8378/93 S1Q.00 + .00 CMA 120150 224 P. L NEWTON ET AL INTRODUCTION Phosphine, or hydrogen phosphide (CAS: 7803-51-2), is a highly toxic colorless gas. Phosphine is flammable, explosive in air, and can autoignite at room temperature. Because of impurities, the technical product has a disagreeable garlic-like or fishy odor. Phosphine has several different industrial uses. Aluminum or magne sium phosphides are used as fumigants because pellets of these pow ders can be rapidly hydrolyzed to phosphine in moist conditions. Zinc phosphide is used as a rodenticide. Phosphine is also used in the synthe sis of organophosphines and as a dopant in semiconductor production. Many studies have reported the toxicity of phosphine in both hu mans and animals. Studies have been reported pn this highly toxic gas as early as 1829 (Klimmer, 1969). However, the toxicity of phosphine, and in particular the low-level subchronic or chronic toxicity of phosphine, re mains unclear. Divergent opinions about the observed effects, exposure levels that produce effects, and the mechanism of action are partially attributable to flaws in earlier experimental designs. These problems in cluded insufficient numbers of animals, the purity of the test material, and the use of estimated or nominal exposure levels. The high toxicity of phosphine has resulted in low permissible expo sure levels. The current threshold limit value (TLV) for phosphine estab lished by the American Conference of Governmental Industrial Hygien ists (ACGIH) is 0.3 ppm. The ACGIH has also established a short-term exposure level (STEL) of 1 ppm for exposures time-weighted over any 15min period. However, a recent occupational exposure survey reported that during phosphine fumigation, worker exposure levels frequently ex ceed this STEL (Zaebst et al., 1989). To characterize the inhalation toxicity of phosphine, a series of stud ies was conducted. The results of an acute, subchronic, and develop mental toxicity study are reported here. MATERIALS AND METHODS Exposure Levels For the single 6-hr acute exposures, 0, 2.5, 5, and 10 ppm were se lected based on a published 4-hr LC50 of 10 ppm (Waritz and Brown, 1965, Muthu et al., 1980). Fifteen animals per sex per group were exposed, with 5 per sex per group scheduled for sacrifice after the exposure and 10 per sex per group after a 2-week observation period. Based on both minimal effects seen in the acute exposure results and reports in the literature, which indicated that exposures to 3 ppm would be tolerable for a 13week study, the subchronic exposure levels were set at 0, 0.3, 1, and 3 ppm. Thirtv animals per sex per group were exposed with scheduled CMA 120151 INHALATION TOXICITY OF PHOSPHINE 225 sacrifices of 10 animals per sex per group after 4 and 13 weeks of expo sure and 4 weeks of recovery. However, because there was no effect seen in the interim sacrifice group after 4 weeks of exposure, concern was raised over whether the high exposure level was close to a maximum tolerated dose for the 13-week study. Therefore, satellite groups of 10 animals per sex per group were added at target concentrations of 0 and 10 ppm. Four of 10 females died after the third exposure. As a result, the 10 ppm exposures were stopped and satellite groups at target concentra tions of 0 and 5 ppm were added. All of these animals survived until the scheduled sacrifice 13 days later. For the subsequent developmental study, 24 pregnant females per group were exposed to target concentra tions of Or 0.03, 0.3, 3, 5, or 7.5 ppm. Because of deaths at the high exposure level, this group was terminated. Test Material In order not to exceed the lower explosive limit during exposure atmosphere generation, the phosphine was supplied as 1% in nitrogen (AIRCO, Riverton, NJ). Impurities identified in the phosphine used to generate the 1% phosphine in nitrogen mixture were as follows (PPM): nitrogen (0.72), oxygen and argon (0.22), carbon monoxide and dioxide (3.61), arsine (<2), total hydrocarbons (<0.5), other phosphines (< SO), and moisture (0.6). Chemical similarity of the phosphine in the test cylin der with phosphine generated from a metal phosphide was confirmed by gas chromatography. Exposure System All exposures were conducted for 6 hr/day in 10,000-L chambers. The stainless steel and glass chambers (Harford Metal Products, Inc.. Aber deen, MD) had pyramidal tops and bottoms. Chamber airflow entered tangentially into the turret at the top of the chambers to facilitate mixing. The chambers were operated dynamically at flow rates of 2000 L/min. This provided one complete air change every 12 min and a 99% equilibra tion time of 23 min. All chambers were operated at a slight negative pressure (-0.5 cm H20) relative to the surrounding area. Chamber air flow, temperature, and relative humidity were recorded hourly during the exposures. In the subchronic study, each test animal's location was rotated on a daily basis throughout the chamber. For the exposures, the 1% phosphine in nitrogen mixture was deliv ered to the chambers from the cylinder through a stainless steel regula tor equipped with a vent and purge valve. The flow of the mixture to each chamber was monitored by a digital mass flowmeter (Sierra Instru ments Inc., Monterey, CA). The animals remained in the chambers for 30 min following each exposure. During this time the chamber was cleared CMA 120152 IV, P. L NEWTON ET Al. using room air at the same airflow rate used during the exposure. Con trol animals were subjected to the same procedures as the test-exposed animals, except that they were exposed to clean air only during the 6-h in-chamber period. As a safety precaution, from the time the phosphine cylinder was opened until it was closed and the chambers flushed out, ail technicians wore fresh-air-supplied positive-pressure suits. Exposure Atmosphere Analysis Samples for GC determination of the phosphine exposure levels were withdrawn hourly through a septum in a line being flushed with chamber air from the breathing zone of the animals. Analyses were per formed on a Hewlett-Packard model 5890A GC/NPD. 100/120 Chromosorb 102 packing was used in a 6-ft x 2 mm ID glass column. The nitro gen carrier gas flow rate was 20 ml/min. The air and hydrogen detector gai pressures were SO and 20 psi, respectively. The temperature settings of the CC were injector 150C, detector 300C, and column 120C. Each dav the CC was calibrated before exposure initiation and checked after each series of hourly measurements using certified gas standards (AIRCO, Riverton, NJ). Animals Male and female Fischer 344 rats, 4 weeks of age at receipt, were obtained from Charles River Breeding Laboratory (Kingston, NY) and used for the acute and subchronic studies. Sprague-Dawlev derived CDR rats. 6 weeks of age at receipt, were obtained from the Portage, Michigan facility of this supplier and used for the developmental toxicity study. The animals were acclimated for at least 12 days during which time they were examined by a veterinarian. Animals were doubly housed in ele vated, stainless steel, wire mesh cages during the first week of acclima tion and were individually housed thereafter. Tap water (via automatic watering system) and Purina Laboratory Chow (Ralston Purina Co., St. Louis, MO) were provided ad libitum except during the actual exposures. The no. 5001 diet was used for the acute and subchronic studies and no. 5002 diet was used for the developmental toxicity study. The animals were maintained on a 12:12 hr light/dark cycle. The environment in the animal's quarters was maintained at 68-74F and 40-60% relative humidin to the maximum extent possible. The animals were individually eartagged for identification and randomly sorted for equal body weights among ail the exposure groups. To provide mated animals for the devel opmental toxicity study, females were cohoused nightly with males (1:1) of the same strain from an in-house breeding colony. Females were ob served each morning for evidence of a vaginal plug and/or microscopic observations of sperm in the vaginal rinse. The day evidence of mating CMA120153 INHALATION TOXICITY OF PHOSPHINE 227 was seen was identified as day zero of gestation. Females were at least 9 weeks of age at initiation of mating (186-281 g). Clinical Laboratory Tests In the subchronic study, hematological and clinical chemistry indices were evaluated after 4 and 13 weeks of exposure and after a 4-week recovery period. Blood samples were obtained from fasted rats via veni puncture of the orbital sinus under light ethyl ether anesthesia. Hemato logical indices including hemoglobin, hematocrit, erythrocyte count, mean corpuscular hemoglobin, mean corpuscular volume, mean corpus cular hemoglobin concentration, and total leukocyte count were deter mined on an ELT-8 Hematology Analyzer (Ortho Diagnostic Systems, Westwood, MA). Differential leukocyte counts and erythrocyte morphol ogy were determined manually. Serum biochemical evaluations includ ing aspartate aminotransferase, alanine aminotransferase, alkaline phos phatase, blood urea nitrogen (BUN), fasting glucose, total protein, albumin, and chloride were determined using a CentrifiCHem System 500 (Union Carbide, Rye, NY). Sodium and potassion determinations were performed using an Electrolyte E2A Analyzer (Beckman Instru ments, Brea, CA). Animal Observations In both the acute and subchronic studies, animals were observed during each exposure for viability and overt signs of toxicity. Detailed observations were conducted weekly. In the subchronic study body weights and food consumption were measured twice pretest and weekly thereafter. Fasting body weights were measured at termination. Ophthal moscopic examinations were performed on all animals pretest and on the day before their scheduled sacrifice. In the developmental toxicity study, each animal was given a detailed physical examination on days 0, 6-15, and 20 of gestation. Body weights were recorded on davs 0, 6, 10, 12,16, and 20 of gestation. Food consumptions were recorded over the day 0-6, 6-10. 10-16, and 16-20 gestation intervals. Gross Necropsy and Histopathoiogy In the acute and subchronic studies, complete gross postmortem ex aminations were performed on all animals. The external surface, all ori fices, the cranial cavity, carcass, the external and sectioned surfaces of the brain and spinal cord, nasal cavity and paranasal sinuses, the tho racic, abdominal, and pelvic cavities and their viscera, and the cervical tissues and organs were examined. The following tissues were preserved in neutral bunered 10% formalin and following routine processing and hematcxylin-eosin staining were examined microscopically: adrenal CMA120154 ns P. E. NEWTON ET AL gland, aorta, sternum, brain, esophagus, eyes, heart, cecum, colon, duo denum. ileum, jejunum, kidneys, liver, right lung lobes, lymph nodes, mammary gland, larynx, nasal turbinates, nerve, ovaries, pancreas, pitui tary, prostate salivary glands, seminal vesicles, skin, spinal cord, spleen, stomach, testes with epididymides, thymus, thyroid/parathyroid glands, trachea, urinary bladder, uterus, vagina, and any tissues with gross le sions. Maternal Evaluations (Developmental Toxicity Study) On day 20 of gestation, all surviving dams were sacrificed by exsanguination under ethyl ether anesthesia. Immediately following sacrifice, the ovaries and uterus were excised intact and weighed. Along with re cording the number of corpora lutea on each, ovary, the numbers and reiative locations of live and dead fetuses, early and late resorptions were recorded for each uterine horn. If no uterine implants were grossly ap parent, the uterus was stained with ammonium sulfide (Salewski, 1964). \\hen foci were visualized following staining, the females were consid ered pregnant in the calculation of the pregnancy rate; however, the number of foci was not included in the calculation of uterine implanta tion data. Fetai Evaluations Each fetus was individually identified, weighed, sexed, and examined tor external malformations/variations to include observation for palatal defects. Approximately one half of the fetuses in each litter were evalu ated for visceral malformations/variations using a microdissection proce dure similar to that described by Staples (1974). The heads of these fe tuses vvere fixed in Bouin's solution and processed for evaluation using a razor biade sectioning technique. The remaining fetuses in each litter were processed for staining of the ossified skeletal structures with al izarin red S. These fetuses were then evaluated for skeletal malforma tions and ossification variations. Fetal visceral and skeletal evaluations were performed under a dissecting microscope. Statistical Analyses Statistical evaluation of equality of means was made by the appropri ate one-way analysis of variance (ANOVA) technique, followed by a multi ple comparison procedure if needed. First Bartlett's test was performed to determine if groups had equal variance. If the variances were equal, parametric procedures were used; if not, -nonparametric procedures were used. The parametric procedures were the standard one-way ANOVA using the F distribution to assess significance. If significant dif ferences among the means were indicated, Dunnett's test was used to CMA 120155 INHALATION TOXICITY OF PHOSPHINE 229 determine which means were significantly different from the control. If a nonparametric procedure for testing equality of means was needed, the Kruskal-Wallis test was used. If differences were indicated, a summed rank test (Dunn) was used to determine which treatments differ from control. A statistical test for trend in the dose levels was also performed. In the parametric case, standard regression techniques with a test for trend and lack of fit were used. In the nonparametric case, Jonckheere's test for monotonic trend was used. Statistical analysis of incidence data was performed using contin gency tables. First, a standard x1 analysis was performed to determine if the proportion of incidences differed between the groups tested. In keeping with standard statistical practice, if any one cell had an expected value of less than 5, this step was not reported. Second, each treatment group was compared to the control group using a 2 x 2 Fisher exact test; the significance level was corrected via the Bonferroni inequality to as sure an overall test of the stated significance level. Third, Armitage's test for linear trend in the dosage groups was performed. All ratios were transformed via the arc sine transformation prior.to analysis. The test for equal variance (Bartlett's) was conducted at the 1% twosided risk level. All other statistical tests were conducted at the 5% and 1%, two-sided risk levels. RESULTS Acute Exposures Single acute 6-hr exposures to 0, 2.4, 4.9, and 10 ppm were con ducted. All animals survived these exposures. During the exposures, a few animals showed red or mucoidal nasal discharge. During the 14-day recovery period, these observations abated. There was no effect on body weight. Gross postmortem and microscopic examinations revealed no treatment-related findings. Subchronic Exposures The exposure concentration results for the subchronic study (Table 1) show excellent agreement between the target, analytical, and nominal results. Distribution measurements performed throughout the exposure chambers showed no significant gradient. Aerosol measurements showed similar particle size and concentration among all the groups (in cluding control), indicating that there was no measurable test substance present as an aerosol. Temperature and relative humidity levels were acceptable. CMA120156 230 P. E. NEWTON ET AL Table 1. Target, Analytical, and Nominal Exposure Levels (mean SD) During the Subchronic and Developmental Inhalation Toxicity Studies of Phosphine Crouo Target level (ppm) Number exposures AnaMical level (ppm) Nominal level (ppm) Subchronic 1 11 III IV V* VI* VII0 Vlllb Developmental 1 II III IV V VI 0 03 1.0 3.0 0 10 0 3.0 0 0.03 03 3 3 7-5 65 0 65 637 038 033 0.0S 65 12) 0.24 0.99 0.11 65 3.1 1.0 3.3 0.6 30 -- 3 10 0.7 9.1 0.7 13 0 -- 13 3.1 0.6 3.1 1.2 10 0 0 10 0.03 X 0.01 0.03 0.01 10 033 X 0.07 030 0.02 10 13 0.6 2.7 0.1 10 4.9 X 0.6 4.6 X 0.2 10 7.0 X 0.9 6.8 0.7 `Satellite grouo. started midstudy and terminated after 3 aavs of exoosure due to excessive mortality 'Satellite group, started arter termination of the 10 ppm group and sacrificed with groups l-IV animals. Morality As described earlier, 4 of 10 females died after 3 days of exposure to 10 ppm of phosphine. This resulted in the premature termi nation of that group. The concentration-time product (C x T) that pro duced death was 180 ppm-hr. Females were more sensitive than males to the lethal effect of phosphine in this study. Phvsical Observations Weekly detailed physical observations showed only sporadic effects or effects that were seen equally in all groups. There was no indication of any exposure-related effects. There was no indication of any compound-related ocular disease. Boev Weight and Food Consumption There was a statistically signif icant decrease in body weight gains in both the males and females at 1 ppm and higher (Fig. 1 and 2). However, these body weight effects dif fered between the sexes. In the males, the decrease in body weight was not statistically significant until the last 2 weeks of exposure. The body weights then recovered toward control weights after termination of the exposures. In the females, the body weights were transiently decreased during only the first few weeks of exposure. The difference in the weight gained during the first 3 weeks between the control and phosphineexposed females showed a dose-related decrease and is presented in Table 2. After the initial decrease, there was recovery. The significant decrease in body weight was associated with the sig- CMA120157 INHALATION TOXICITY OF PHOSPHINE 231 nificantly decreased food consumption seen in both males and females (Figs. 3 and 4). The food consumption effect was more prolonged than the body weight effect and also transiently included the 0.37 ppm group. However, the food consumption effect was also reversible either during the exposures or the recovery period. Hematology No hematological effects were seen in the females. The males had a 5% statistically significant increase in platelets in the 3.1 ppm group after 4 weeks of exposure. The males also had a 5% statistically significant decrease in hemoglobin, hematocrit, and erythrocyte counts in the 3.1 ppm group after 13 weeks of exposure. These erythrocyte effects appear to be exposure-related as the satellite 10 ppm group, sacri ficed aftet 3 days due to high mortality, also showed decreased erythro cyte counts. The 5.1 ppm group in this study did not show any effects possibly because they only had 13 days of exposure prior to sacrifice. The observed hematological effects were reversible. After a 4-week re covery period all parameters were normal. Clinical Chemistry The males also showed a few changes in the clin ical chemistry parameters. Increased BUN levels were seen after 4 weeks at 3.1 ppm (13%), after 3 weeks at 5.1 ppm (32%), and after 3 days at 10 co z <i o: CD * H H tii 3 O m --nessuexr -- 0 3 6 9 12 15 18 WEEK FIGURE 1. Body weight gain of male Fischer 344 rats during a 13-week inhalation toxicity study of phosphine followed by a 4-week recovery period. CMA120158 0 3 S 9 12 15 18 WEEK FIGURE 2. Body weight gain of female Fischer 344 rats during a 13-week inhalation toxicity study of phosphine followed by a 4-week recovery period. ppm (23%). Therefore, these appear to be exposure-related. However, this effect was transient as BUN was not elevated at 3.1 ppm after 13 weeks. A 17% increase in alkaline phosphatase was also seen in the 10 ppm group. The only other change that was possibly exposure-related was a decreased serum glutamic pyruvic transaminase (SGPT) seen in the 3.1 ppm males and females after 13 weeks of exposure. However, there is no known toxicological significance for a decreased SGPT and the effect may have been produced by two unusually high control animal values. Table 2. Percent Difference in Body Weight Gain Between Control and Exposed Females During a Subchronic Inhalation Toxicity Study of Phosphine Exposure level (ppm) Week 1 Week 2 Week 3 Week4 0.37 -1.8 -1.8 -2.9 -1.9 1.0 -4.1" -1.4 -47* -3.1 3.1 -3.6" -- 4.8** -5.4" -3.6* 5.1 -2.5 -8.1*' -7.1* -3.0* First recovery week, `p s .05; "p s .01. CMA120159 INHALATION TOXICITY OF PHOSPHINE 233 WEEK FIGURE 3. Food consumption of male Fischer 344 rats during a 13-week inhalation tonocity study of phosphine followed by a 4-week recovery period. Other changes occurred sporadically and were not consistent over sex or time. Terminal Organ and Body Weights and Organ/Body Weight Ratios Significant changes in absolute or relative kidney and liver weights occurred. The 12% increase in kidney weights in the 10 ppm satellite group correlated with the observed microscopic changes. There were also exposure-related, but not dose-related, decreases in male liver weights and ratios in the 0.37,1, 3.1, and 10 ppm groups. The toxicologi cal significance of decreased liver weights is unclear. Pathology Cross postmortem observations were random except for an increased incidence of small seminal vesicles that were observed in the 1 and 3.1 ppm groups (6/10 and 5/10) after 13 weeks but not after the 4-week recovery period. This finding may be related to the test substance but because of a lack of a microscopic correlate its significance is equivo cal. Microscopically, treatment-related lesions occurred in the kidneys of animals of both sexes exposed at 10 ppm and consisted of coagulative necrosis of the tubular epithelium of tubules in the outer cortex, pre sumably the proximal convoluted tubules. These lesions were more $e- CMA120160 234 0 3 6 9 12 IS 18 WEEK FIGURE 4. Food consumption of female Fischer 344 rats during a 13-week inhalation toxicity study of phosphine followed by a 4-week recovery period. vere in females than males. Renal lesions were not noted in animals receiving 5.1 or 3.1 ppm of phosphine under the conditions of this study. Congestion was noted in the lungs of the female animals receiving 10 ppm of phosphine and dying during the study. Following a 4*week recovery period, no treatment-related lesions were noted in any of the tissues examined from animals exposed to 3.1, 5.1, or 10 ppm phosphine. Spontaneous disease lesions and incidental findings were of the usual number and type commonly seen in Fischer 344 rats. They were essentially comparable between control- and treated animals. Developmental Toxicity--Maternal Parameters Mortality Rates Exposure concentrations for the developmental study are presented in Table 1. No mortality occurred in the control or in the 0.03, 0.33, 2.8, or 4.9 ppm exposure groups. In the 7.0 ppm exposure group, the first 14 females treated died after 3-10 days of exposure (days 8-15 of gestation). Due to this mortality, this group was terminated. Pregnancy Rates Pregnancy rates for the 0.03, 0.33, 2.8, and 4.9 ppm groups were comparable to control (Table 3). CMA120161 INHALATION TOXICITY OF PHOSPHINE 23S Body Weight and Food Consumption No adverse effect of treat ment at an exposure level to 4.9 ppm was evident from body weights or weight change data over the gestation period (Table 3). Likewise, food consumption data throughout gestation were not adversely affected by treatment to the 4.9 ppm exposure level (Table 3). Physical Observations No adverse effect of treatment with phos phine at an exposure level to 4.9 ppm was evident from the detailed physical evaluations. Even at the 7.0 ppm exposure level, which was ter minated early due to increased mortality, no adverse effect of treatment was evident from physical observation data. Uterine .Implantation Data No adverse effect of treatment with phosphine, at an exposure level to 4.9 ppm was evident from uterine implantation data (Table 3). In the low-exposure group (0.03 ppm) the mean number of resorption sites, the mean resorption/implant ratio, and the incidence of females with resorptions were higher than control and these differences were statistically significant. The reason for this in crease in resorption data at the 0.03 ppm exposure level was unclear; however, in the absence of similar responses in resorption data at the higher exposure levels evaluated, no adverse effect of treatment was indicated. Gross Postmortem Evaluations No adverse effect of treatment up to an exposure level of 4.9 ppm was evident from the maternal gross post mortem examinations. Discoloration (reddening) of the lungs and liver was observed in some of the animals that died at the 7.0 ppm exposure level. This was not considered unusual in animals that die and are not exsanguinated before postmortem examination. Developmental Toxicity--Fetal Parameters Weight and Sex Distribution Data Mean fetal weights and the ratio of male/femaie fetuses per group for the phosphine-treated groups were comparable to that of control. Malformation/Variation Data No adverse effect of treatment with phosphine to an exposure level of 4.9 ppm was evident from the fetal external, visceral, or skeletal evaluations (Tables 4 and 5). In the phosphine-treated groups, low incidences of malformations/variations on a per-fetus and per-litter basis were seen among all group. However, these data did not differ statistically from the control data and, due to the lack of similarity in the types of malformations seen among the phosphine-treated groups, no adverse effect of treatment was indicated. DISCUSSION Lethality was the primary hazard of phosphine exposure (7.0 ppm and above). The C x T product of 180 ppm-hr is similar to the results of CMA120162 236 P. E. NEWTON ET AL Klimmer (1969) who found a C x T product of 213-hr for maJe Wistar rats. Klimmer also found similar C x T products for cats, guinea pigs, rabbits, turkeys, and chickens. This similar interspecies C x T product may also hold for humans as Flury (cited in Klimmer, 1969) indicated that a single inhalation of 5-10 min duration should be lethal to humans at 1000 ppm (83-166 ppm-hr). Table 3. Maternal Body Weight and Food Consumption Data, Uterine Implantation Data, and Fetal Weight Data in Rats Exposed Prenatally to Phosphine Exposure groups (actual levels, ppm) Data 1 II III IV V (0) (0.03) (0.33) (2-S) (4.9) No. females: Mated 24 Pregnant 22 (91.7%) Maternal body weight data (g): Gestation day (GD) 0 234 17* CD 20 tactual) 360 21 CD 20 iCorrected)b 283 16 Weight gain (g) CD 6-16 44 7 GD 6-20s 20 11 Maternal food consumption (g/kg/dav): GD 0-6 84 19* CD 6-10 74 8 CD 10-16 78 5 CD 16-20 83 6 Per pregnant female: Corpora lutea (CL) 16.0 2.0* Implants (1) 15.4 2.0 Preimpiant loss index (CL-LCL) 0.04 0.05 Live fetuses 14.9 2.1 Resorptions (R) 0.5 0.9 R/l ratio 0.04 0.05 Litters with one or more resporption (X) 8 (36.4%) Fetal weights <g) Males 'Ml 3.30 0.23* Females (F) 3.20 0.21 Both sexes 3.24 0.21 Ratio or total M/F: Fetuses 1.0 24 21 (87.5%) 236 15 361 21 288 18 44 * 8 21 9 86 r 13 72 6 79 6 86 i 9 16J 1.9 15.7 1.8 0.04 0.06 14.1 * 1.6 1.6 1.2" 0.10 0.07* 16* (76.2%) 3.28 0.31 3.11 0.22 3.18 * 0.25 0.9 24 24 (100%) 236 19 360 20 287 18 44 21 25 18 81 21 75 11 80 12 85 8 16.0 2.0 14.7 2.4 0.08 0.13 14.0 2.3 0.7 1.0 0.05 0.07 9 (37.5%) 3.31 0.17 3.22 0.19 3.25 0.17 1.1 24 23 (95-3%) 237 19 364 24 289 = 21 42 c 6 22 -- 9 87 s 13 72 = 6 78 = 7 86 = 6 15.9 = 2-5 15.0 . 2-S 0.05 = 0.11 1422 - 2J 0.9 =: 0.9 0.06 = 0.06 14 (60.9%) 3.35 = 0.25 3.20 = 0.22 3.28 = 023 1.1 24 23 (95.8%) 236 18 368 25 290 21 43 15 23 9 89 22 69 10 84 18 91 8* 16.1 2.3 15.4 2.3 0.04 0.07 14.7 2.6 0.7 1.0 0.05 0.07 10 (43.5%) 3.30 0.21 3.13 0.22 3.22 0.20 1.0 'Mean = standard deviation. ''Corrected day 20 gestation body weight - actual day 20 gestation weight gravid uterus. p s .05: **p s .01. - the weight of CMA 120163 INHALATION TOXICITY OF PHOSPHINE 237 Table 4. Sun--lary oi Malformation Data in an Inhalation Developmental Toxicity Study of Phosphine n ?als Exposure group (actual levels, ppm) Data 1 II III IV V (0) (0.03) (0.33) (2.8) (4.9) No. refuses litters) evaluated: External Viscera Skeletal _ External exarsmatiorc Filamentous tail Microenania anc small tongue Edematous with airly tail Visceral exanjiatiorc Microoisrsalmia Distencec lateral ventricles of brain Foldee *ecna Cert zaire Persisted truncus arteriosus Abserce ;r ductus arteriosus \5D Skeletal exa,-.nations: Absence :t verteorae (L, S, O Cervical _a Cramai ocne detects \ertebrai zerects alone) \ertebra. -b defects 3 L verts* ae 327(22) 161 (21) 158(22) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 296 (21) 131 (21) 146 (21) 335(24) 174(24) 163 04) 326(23) 167(23) 159(23) No. fetuses (litters) affected 1*(1) 0(0) 0(0) 1d (1) 1d (1) 1(1) 0(0) 0(0) 0(0) 0(0) 1*(1) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) '1** (1) 0(0) 0(0) 0(0) 0(0) 0(01 0(0) 0(0) 0(0) 0(0) 1 (1) 1b(is 0(0) 0(01 0(01 0(0) 0(0) 0(0) 1* (1) 0(0) 0(0) i* m 0(0> 0(0) 0(0) 0(0) 0(0) 0(0) 1(1) 1(1) 1(1) 338(23) 175(23) 163(23) 0(0) 0(0) 1c (1) 0(0) 0(0) 1 (D 0(0) 1c (D 1c (D 1c (1) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) Total %) fetuses affected* External --aifortrations Mscera, --aJformaaons Skeletai --alformaoons Tota: %) i.tte-s affected* External -alformations Uscerai --alformations Skeletal -aiformaoons 0(0) 1 (0.3) 1 (0_3i 0(0) 0(0) : (i.3) 0(0) 1 (0.6) 0(0) 1(0.7) 2 (Id! 3(1.9) 0(0) 1 (4.8) 1 (4d' 0(0) 0(0) 2(9.5) 0(01 1 (4.3) 0(0) 1 (4.8) 2 (8.31 3 (13.0) 1 (0.3) 2(1.1) 0(0) 1 (4.3) 2(8.7) 0(0) 'Same re".; saving multiple malformations. ''Same retus saving multiple malformations. 'Same refus saving multiple malformations. dSame 'ecus saving multiple malformations. 'Same tetus saving multiple malformations. 'No stat nc2.Iv significant differences in comparison to control data. V5D. ve-t-cular seotal defect; L, lumbar region; S, sacral region; C, caLdal region. CMA120164 238 P. E. NEWTON ET AL Table 5. Summary of Variation Data in an Inhalation Developmental Toxicity Study of Phosphine in Rats Exposure group (actual levels, ppm) Data 1 II III (0) (0.03) (033) IV V (2.8) (4.9) No. fetuses (litters) affected External examination: Shiny appearance Visceral examination: Absence of innominate artery (aortic arch) Ureters tortuous Ureters distended Skeletal examination'; Cranial bones: interparietal 1 Supraocapital 1 Hyoid U Parietal 1 Vertebral observations: Cr TP-I Th C-l S TP-I C TP-I CTP-U Stemebra 4th 1 5th U 6th U Ribs 1st L rud 0(0) 0(0) 4(3) 0(0) 29(14) 32(14) 28(10) 5(4) 5(4) 22(10) 30(16) 65 (20) 13(7) 12 (6) 59 (18) 7(5) 2(2) 1(1) 1(1) 0(0) 0(0) 0(0) 2(2) 0-(0) 1(1) 27(14). 29(13) 16(7) 2(1) 19 (10) 9(9) 14(9) 1 (1) 12(71 25(131 29(10) 50(17) 15(7) 12 (8) 19 (10) 13 (10) 43(17) 12(8) 11 (5) 54(19) 13(71 3(3) 8(6) 50(17) 12 (6) 5(4) 0(0) 1(1) 2(2) 0(0) 27(13) 23(10) 21(13) 4(2) 8(5) 23(14) 22(11) 38(15) 12(6) 6(3) 51 (IS) 6(4) 4(4) 0(0) 0(0) 2(1) 0(0) 28 (15) 27 (13) 19 (10) 1(1) 16(9) 21 (13) 27 (14) 51 (21) 14(7) 5(4) 61 (22) 19(11) 2(2) Total <*) fetuses with variations19 External Visceral Skeletal Total (%) litters containing affected fetusesb External Visceral Skeletal 0(0) 1(03) 1 (03) 0(0) 0(0) 4(2.5) 0(0) 2(1.1) 2(13) 2(1.1) 115 (72.8) 112 (76.71 112 (68.7) 104 (65.4) 115 (70.6) 0(0) 3 (14.3) 22(100) 1 (4.8) 0(0) 21 (100) 1 (4.2) 2(83) 23(100) 0(0) 2 .8.7) 22 195.7) 0(0) 1 (43) 22 (95.7) *This is not a presentation of all ossification variation data seen during the study but summarizes those variations seen with greatest frequency among the control and/or treated group fetuses. b\o statistically significant differences relative to control. I, incompletely ossified; U, unossified; Cr, cervical; Th, thoracic; L lumbar; S, sacral; C, caudal; TP, transverse process(es); C, centrum; Rud, rudimentary structurets). cMA 120165 INHALATION TOXICITY OF PHOSPHINE 239 The results from this study also confirm Klimmer's conclusion that a threshold or nonlinear lethality effect exists. If Haber's law of C x T - constant were true, the 5.1 ppm group should have died after approximately six 6-hr exposures. Instead, all animals in the 5.1 ppm group survived and were unremarkable at the scheduled sacrifice after 13 days of exposure. In the developmental study, 7.0 ppm was fatal to pregnant rats after 3-10 days of exposure. The median time to death, 4 days, also gives a C x T product of approximately 180 ppm-hr. Therefore the lethality threshold for repeated daily 6-hr exposures to phosphine is 5.1-7.0 ppm. All other phosphine-related effects seen at sublethal exposure levels were relatively small and completely reversible. This included the effect on body weight gain, as also previously reported by Waritz and Brown 1975), and hematological effects, as also previously reported by Klimmer 1969) and Muller (1940). The congestion seen in the lungs and kidneys of animais that died was also similar to that previously reported in major organs by Muller (1940) and in the lungs by Muthu et al. (1980). The results of this study indicate lethality to be the primary hazard from phosphine exposures. The current TLV of 0.3 ppm gives a safety factor of approximately 20. REFERENCES C'immer. O. R. 1969. Contribution to the study of the action of phosphine (PHj): the question of the so-called chronic phosphine poisoning. Springer-Verlag reprinted translation of "Beitraezur wirkung des phosphorwasserstorfes (PHj)." Arch. Toxicol. 24:164-187. Muller. '.V. 1940. Uber Phosphorwasserstaffvergiftung (Tierversuche). I. Mitt. Akute and Subakute Verziftung. Naunyn-Schmiedebergs Arch. Exp. Pith. Pharmak. 195:184-193. Muthu, st. M., Krishnakuman, M. K_, Muraiidhara, and Majuntder, S. K. 1980. A study on the acute inhalation toxicity of phosphine to albino rats. Bull. Environ. Contam. Toxicol. 24:404--110. ialewski E. 1964. Farbemethode zum Makroskopischen Machweis von Implantationsstellen am Uteus der Ratte. Arch. Pathol. Exp. Pharmakol. 247:367. 5;aples. 1. E. 1974. Detection of visceral alterations in mammalian fetuses. Teratology 9-J7A. Aaritz. R. S., and Brown, R. M. 197S. Acute and subacute inhalation toxicities of phosphine, pheny- Iphcsphine and triphenylphosphine. Am. Ind. Hyg. ). 36:452-458. 2aebst, D. D., Blade, L. M., Burroughs, C. E.. Morrelli-Schroth, P., and Wbodfin, W. J. 1988. Phos phine exposure in grain elevators during fumigation with aluminum phosphide. Appl. Ind. Hyg I. 3:145-154. CMA 120166 presents o ii NH CH H, CH INDUSTRIAL HEALTH FOUNDATION 34 Penn Circle Wes* Pittsburgh, Pa. 15206 CMM20167 INHALATION TEBATOLOGY STUDIES ON CYCLOHEXANONE Raymond E. Schroeder, M.S.,* and Elton R. Homan, Ph.D.** This inhalation teratology study of cyclohexanone was undertaken pursuant to a negotiated testing agreement with the Environmental Protection Agency and was performed by Bio/Dynamics Incorporated. The study was conducted in rats and mice. Findings with respect to rats are based on the draft report. Results of the mouse study are currently incompletely evaluated and only qualitative results will be presented at this time. Slide 1 Cyclohexanone was provided by Allied Corporation and was subjected to 'appropriate analytical and stability evaluation. Vapor was generated using a Laskin generator and diluted with preconditioned room air. Chamber concentrations of cyclohexanone vapor ware monitored hourly by infrared absorption. Experimental animals were female Sprague-Dawley derived CD rats from Charles River Breeding Laboratories. This strain was chosen on the basis of previous experience, the existence of a broad data base and known sensitivity to acetylsalicylic acic, a known rat teratogen. The females were 69 days of age upon receipt and 83 days old at first mating to proven males, two females to one male. Successful mating was determined by vaginal smears. Microscopic finding of sperm, permitted definition o: Day 0 of gestation. Females were sacrificed on Day 20 of gestation. Females were exposed to cyclohexanone vapor on gestational Days 6-19, inclusive, for six hours/day. Concentrations of 0, 300, 650 and 1400 ppm. were selected based on previous experiments. Animals were weighed on Day 0, 6, 15 and 20. Animals were examined twice daily for mortality and gross signs of toxicologic or pharmacologic effect anc subjected to detailed physical examination cr. Days 0, 6, 10, 15 and 20. Early in the study, following observations of lethargy, routine evaluation of startle response was initiated while animals were in the chambers before, during and after exposure three times weekly. Following sacrifice by lethal exposure to diethyl ether vapor on~Day 20, all females were subjected to a complete necropsy. The intact uterus and ovaries were removed and weighed. Number anc location were recorded for live ar.d dead fetuses, early and late resorptions and implantation sites. The number of corpora iutea were recorded for each ovary. * Manager, Reproducticn/Teratolcgy Section, Bio/Dynamics, Inc. , P.O. Box 43, East Millstone, NJ 08573 ** Senior Toxicologist, Bushy Run Research Center, Union Caroide Corporation, R.D. ff4, Mellon Road, Export, PA 15632 206 CMA12168 Each fetus was given a gross examination, weighed and sexed. Approximately one half the fetuses in each litter were examined for soft tissue malformations using the Staples microdissection technique. Heads were sectioned serially following fixation in Bouin's solution. The other half of the fetuses in each litter were stained with Alizarin Red S for skeletal malformations and ossification variations. Study results were analyzed statistically using appropriate parametric or nonparametric tests for interval data and incidence data. Turning to the results, clinical observations included the finding of lacrimation initially on Day 6 in high level females only. The incidence anc severity of this finding decreased over the course of the study. Also noted in this group were nasal discharge, lethargy and, on Day 15 only, vaginal discharge in several females. Observations of startle responses made during exposure disclosed a concentration and time related pattern. At 300 ppm there -were a few instances of sluggish response. At 650 ppm the incidence was higher and_ became apparent as early as one hour after beginning of exposure. At 1400 ppm there was a high incidence of lethargy at one hour, and many rats were essentially non-responsive for the duration of daily exposure. Slide 2 In this slide, anc most of those which follow, cyclohexanone concentrations are shown as column headings, evaluation criteria shown on the left. Statistically significant differences from control are indicated oy asterisks. There were 26 females mated in each treatment group. None died during the course of the study and 23 or 24 became pregnant in each group. Kone aborted or delivered prematurely. At necropsy, the pregnant or non-pregnant status of each rat was confirmed and all pregnant females were found to oe carrying viable fetuses. Slide 3 There were no significant differences between maternal Dody weights at any time except at the HOC ppm level. At 1400 ppm the weight differences with respect to controls were statistically significant at Day 15 and Day 2t. Slide 4 In this presentation of maternal body weigh; changes, the toxic effect of treatment at the 1400 ppm level is apparent from Dey 6 on. During the critical period of organogenesis, Days 6-15 in the rat, the weight gain in this group is 44% less than the comparable control group value. For the period of Days 15--20, the high level weight gain is 30% less than control value. Side 5 The first row of this table compares the mean weight of the gravid uterus in each group. The reduction in the high level group is significant. In the next row, the corrected maternal body weight was calculated by subtracting the uterine weight from the Day 20 body weight shown on the previous table. Finally, the corrected body weight change for the interval from Day 6 to Day 20 was calculated by suotraction. Again the 1400 ppm group mean is significantly less than the control mean weight gain. Thus 207 CN\A 120^69 the reduction in weight gain associated with the high level of cyclohexanone, which can be considered a maximally tolerated dose, is the result of reduced gains in both dams and fetuses. Slide 6 Reproductive data are summarized in this table. No significant differences between treated groups are apparent with respect to numbers of corpora lutea, implantation sites, pups or resorptions. Highly significant reductions are apparent in mean body weights of pups of both sexes ir. the high level litters. There was no effect on sex ratio, reported conventionally as percent males. Slide 7 Fetal effects are summarized in this table for the three types of examination used,-showing incidences for both fetuses and litters. External malformations were absent or negligibie in all groups. Visceral malformations were significantly lower with respect to fetal incidence in the 650 and 1400 ppm groups than in the control group. The predominant finding was slight distention of the renal pelvis. This is regarded as a minor malformation and has been commonly found in low incidence historically in control rats of this strain. These findings indicate no adverse treatment effect from fetal visceral evaluation. The most commonly notec visceral variation was tortuous ureter. Visceral variations were less frequent in the mid level concentration of cyclohexanone and were absent in the high level group, thus nc adverse treatment effect was found ir. this category. Skeletal malformations are represented oy one fetus with unilateral wavy rizs, one with unilateral fused ribs and thoracic \ertebral malformations and one with a misaligned thoracic vertebral centrum. These three fetuses were in different litters and all were from the control exposure group. There were no skeletal malformations in fetuses from any treatment group. The incidence of fetuses with various ossification variations was considered similar in the control, lew and middle concentration groups. In the high concentration group, the incidence o: fetuses with incomplete ossification of the cranial bones, and/or ossification irregularities of the hyoid was notably increased from control data. Likewise, at the high concentration, the incidence of incomplete or unossified sternebrae and unossified metatarsals and foreiimb phalanges was also increased. Thus, at tne high concentration, there was a generalized retardation of ossification.. With respect to fetal examinations, tnen, evaluation of fetuses recovered from females treated at 300 or 650 ppm of cyclohexanone for external, visceral or skeletal malformations did not reveal an increase in malformations. At the 1400 ppm level, no increase in malformations was evident during fetal external visceral or skeletal examinations. However, the incidence of fetuses with at least one ossification variation was increased at this concentration and some retardation in ossification was indicated, particularly in cranial ossifications, sternebrae and phalanges. These are regarded as evidence of fetotoxicity rather than teratogenicity. 208 CMA 120170 To summarize the results of this study of the teratologic potential of cyclo hexanone vapor in rats, there were no treatment effects at 300 or 65C ppm other than transient lethargy. Detrimental effects seen at 1400 ppm included maternal toxicity reflected in significantly reduced body weight gain, and transient lacrimation, lethargy and nasal discharge. Fetotoxicity was indicated by reduced fetal weight gain and an increased incidence of variations in and retardation of ossification processes. There was no evidence of teratogenicity. Slide 9 This synopsis of an as yet incompletely evaluated mouse teratology study of cyclohexanone shows a brief outline of the experimental design and preliminary findings. CD-I mice, 30 mated females per group were exposed to control air or 1400 ppm of cyclohexanone for 6_hours/day during gestation Days 6-17 and sacrificed on Day 18. Preliminary findings include lacrimation, nasal discharge, marked lethargy and a white material, possibly a discharge, on the eyes c: the females. Maternal weight was reduced on gestation Day 18, and weight gain in the interval from Day 6 to Day 18. The numoer of live fetuses per litter was reduced and the number c: resorptions per litter increased. Fetal body weight was reduced. There were r.o treatment related external or soft tissue abnormalities. Although fused riss were noted in two fetuses from treated dams, changes c: similar severity were seen in control fetuses. There was an increased incidence of ossification variations, representing retarded development. There was no evidence of terato genicity. In closing we would like to acknowledge the assistance of Dr. Rochelle W.Tyl, Manager of Teratology, Bushy Run Research Center, ir. the preparation of this material. 209 CMA 120171 Slide 1 Cyclohexanone Teratology Study in Rats Experimental Design Route: Inhalation Test Sys en: CD Rats (26 females/group) 83 days old at mating "Sacrificed on gestational iay 20 Regimen: 07 300, 650, 1400 ppm 6 hours/dav daily on gestational days. 6-1 Obserra; tons: In lid - physical/behavioral signs Maternal body weight and body weight change Post"::' ten - Maternal necropsy Reproductive system exam: Live fetuses Dead fetuses Resorptions Implantation sites Corpora lutea Fetal exam: Body weight Sex External Soft tissues (Staples) Skeleton 210 CMA 120172 Slide 2 gyclohexanone Teratology Study In Rats Maternal Survival and Pregnancy Status Females mated Died Pregnant Aborted Delivered premature Examined at scheduled necropsy Nonpregnant Pregnant With viable fetuses Cyclohexanone Concentration 0 ppm 300 ppm 650 ppm 1400 ppm 26 26 00 24 23 00 00 26 26 26 0 24 0 0 26 26 0 23 0 0 26 3 24 23 24 23 2 24 24 3 23 23 211 CMA 120173 Slide 3 Cyclohexanone Teratology Study in Rats Mean Maternal Body Weight During Gestation Day 0 Day 6 Day 15 Day 20 * z < .05 ** p C .01 Cyclohexanone Concentration 0 ppm 300 ppm 650 ppm 1400 ppm 273 g 307 350 420 274 g 310 357 432 273 g 309 346 418 274 g 310 333 * 383 ** 212 CMA120174 Slide 4 Cyclohexanone Teratology Study In Rats Mean Maternal Body Weight Change During Gestation . Days 0-6 ?reexposure Days 6-15 Organogenesis Days 15-20 Days 6-20 Cyclohexanone Concentration 0 ppm 300 ppm 650 ppm 1400 ppm 34 g 36 g 36 g 43 47 37 70 75 113 122 71 108 36 g 2 4 ** 49 ** 73 ** 213 CMA120175 Slide 5 Cyclohexanone Teratology Study 'in Rats Mean Uterine Weight and Net Maternal Body Weight Change Gravid Uterine Weight (GUW) Corr BW (BW - GUW) Corrected BW Change D6-20 ** p < .01 Cyclohexanone Concentration 0 ppm 300 ppm 650 ppm 1400 ppm 80.2 g 87.0 g 80.0 g 59.9 g ' 339.5 346.6 338.9 323.0 32.8 35.5 29.3 13.2 ** 214 CMA 120176 SUde 6 Cyclohexanone Teratology Study In Rats Summary of Reproductive Data Corpora lutea per dam Implant sites per dam Preimplantation loss Mean litter size Resorptions per dam Litters with 1 or more resorptions Mean body weight (g) Viable fetuses Male fetuses Female fetuses Percent males .01 Cyclohexanone Concentration 0 ppm 300 ppm 650 ppm 1400 ppm 15.9 17.4 16.3 15.8 14.8 15.8 14.7- 14.3 7.12 13.8 1.0 9.02 15.1 0.7 10.02 13.6 1.1 9.62 13.3 1.0 10 10 16 15 3.67 3.77 3.56 48.6 3.66 3.80 3.53 48.3 3.68 3.76 3.58 53.4 :.73 ** 2.33 ** 2.64 ** 52.9 215 CMA 120177 Slide 7 Cyclohexanone Teratology Study In Rats Sternary of Fetal Effects Cyclohexanone Concentration 0 ppm 300 ppm 650 ppm 1400 ppm External malformations (2- affected) Fetuses 0.0 Litters 0.0 Visceral malformations (2 affected) Fetuses Litters 6.4 29.2 Visceral variations (* affected) Fetuses 5.6 Litters 29.2 Skeletal malformations (2 affected) Fetuses 1.9 Litters 12.5 Skeletal variations (" affected) Fetuses 54.2 Litters 100.0 0.6 8.7 4.4 26.1 5.0 17.4 0.0 0.0 89.3' ~ 100.0 0.0 0.0 0.0 0.0 . 0.6 ** 4.2 0.6 ** 4.3 0.6 0.0 ** 4.2 0.0 ** 0.0 0.0 0.0 0.0 87.3 100.0 97.9 **i 100.0 * p < .05 ** p < .01 (a) Incompletely ossified cranial benes, unossified or incompletely ossified sternebrae and unossified metatarsals and forelirab phalanges. 216 CMA120178 Slide 8 Cyclohexanone Teratology Study in Rats Summary of Effects No significant treatment effects at 300 or 650 ppm Detrimental effects seen at 1400 ppm: Maternal toxicity Reduced body weight gain Lacrination, lethargy and nasal discharge Fetotoxicity Reduced fetal weight gain Increased incidence of ossification variations No evidence of embryotoxicity or teratogenicity 217 CMA 120179 Slide 9 Cyclohexanone Teratology Stud? In Mice Experimental Design: CD-I mice Inhalation concentrations 0, 1400 ppm 6 hr/.day gd 6-17 Sacrifice gd 18 Preliminary findings: Maternal lacrimation, nasal discharge, marked lethargy, white material on eyes Maternal body weight reduced gd 18 only Maternal weight gain reduced gd 5-18 Number live fetuses/litter reduced Number and 7, resorptions/lizter increased Fetal body weight reduced No treatment related external or soft tissue abnormalities Skeletal abnormalities - fused ribs in 2 fetuses Increased incidence of ossification variations No evidence of teratogenicity ERH/vlt/2243 06/20/84 218 CMA120180