Document pg0297J9mYLK9vo3my2r9JJw

FUNDAMENTALANDAPPI IEDTOXICOLOGY 10,221-232 (1988) Mice Exposed in Utero to 20 ppm Benzene Exhibit Altered Numbers of Recognizable Hematopoietic Cells Up to Seven Weeks after Exposure U T A. QLLER' AND CARROLL A. SNYDER' New York University Medical Center, Insticure ofEnvironmenml Medicine. 550 Firsr A venue. ,Veu Yurk.New York 10016 Recei red .tfarch19. I987: accepted .4 u w t 4. I987 Mice Exposed in Uiero to 20 ppm Benzene Exhibit Altered Numbersof Recognizable Hema- topoietic Cells Up to Seben Weeks after Exposure. KELLER, K.A,, AND SNYDER. C. A. (1988). Fundam. Appl. To.ucoi 10, 224-232. Pregnant Swiss Webster mice were exposed from Day 6 rhrough Day 15 of gestation to either air, 5 ppm. I O ppm, or 20 ppm benzene. On Dab 16 of gestation. 2 days after birth. and b weeks after birth. progeny of the exposed dams were assayed for the amount and type of hemoglobin produced and for recognizable hematopoietic cellsin the peripheral blood and hematopoietic organs. None of the benzene exposures induced significant changes in the indices assayed from the 164ay fetuses. In contrast. 2day neonates exposed in utero to all concentrations of benzene exhibited reduced numbers of circulating erythroid precursor cells. In addition. those '-day neonates exposed In utero to 20 ppm benzene exhibited increased numbers of hepatic hematopoietic blast cells and granulopoietic precursor cells accompanied by decreased numbers of erythropoietic precursor cells.Six-weekadult mice exposed in utero to 20 ppm benzene exhibited a similar pattern of enhanced granulopoiesis. These animals exhibited elevated numbers of splenic hematopoietic blast cells and granulopoietic precursor cells accompanied by decreased numbers of marrow erythropoietic precursor cells. These results suggest that in uiero exposures to low concentrations of benzene can induce persistent enhanced production of recognizable granulopoietic elements in the hematopoietic systems of mice. c 1988sooayori-oxlcoiogy. Although benzene is a hematotoxicant. it is one of the most widely used industrial chemicals(Chem. and Eng. News. 1984).Estimates of US.worker populations exposed to ben- zene range from 600.000 to 2.000.000 (U.S. Department of HEW. 1974; U.S. OSH.4, 1978;Luken and Miller. 1981). Occupational exposures to benzene are currently regulated at a threshold limit value (TLV) of I ppm. In addition to occupational exposures, it is estimated that U.S. urban dwellers are exposed to nearly 1 mg ofbenzene daily (National Research Council, 1980). ' Current address: International Research and Devel- opment Corporation. Mattawan. MI. To whom all correspondence should be addressed, Several studies concerning the developmental toxicity of' benzene have been performed using a variety of exposure routes (see Murray ef al.. 1979. and Mehlman e[ ai.. 1980. for reviews).The results ofthese studies indicate that benzene is not a teratogen although it may be embryotoxic. Surprisingly, although benzene is a potent hematotoxicant, there have been very few studies concerning the effectsof benzene on the developing hematopoietic system (Pollini er af.. 1965; Koizumi and Suzuki. 1966: Murray et ai.. 1979; Mizens. 1982). Benzene may be particularly toxic to fetal hematopoietic cells since the populations of these cells are actively expanding(Pau1 e[ai., 1969)and since benzene is particularly toxic to actively replicating cells (Snyder. 1987). 0272-0590/88 $3.00 Copyright 0 1988by the Society ofToxicolcgy All nghv,of reproduction in any form reserved. 224 HEMATOTOXICITY OF /I! C'TERO BE:NZENE EXPOSURE 225 Recently we reported marked effects on the erythrocytic and granulocytic colony forming ceils of mice exposed in utero to 5 , 10. or '0 ppm benzene (Keller and Snyder. 1986). Moreover, when mice previously exposed in iller0 to 10 ppm benzene were reexposed to 10 ppm benzene as adults. marked depressions in the numbers of granulocytic and erythrocytic colony forming cells were observed. We now report our findings concerning the effects of in utero exposure of mice to 5. 10. or 20 ppm benzene on recognizable hematopoietic precursor cells. peripheral blood cells, and hemoglobin production. Exposures were conducted from Days 6 to 15 of gestation in order to coincide with the time of rapid expansion of the embryonic and fetal hematopoietic systems. Assays were performed on 16-day-old fetal, 2-day-old neonatal, and 6-week-old progeny of air or benzene-exposed dams. These motocois allowed for the assessment ofboth the immediate and long-term effects of the exposures. METHODS (Drew and Laskin. 1971). Animals were exposed in wire mesh cages and were not allowed food and water during exposures. Exposures were conducted for 6 hriday over Dahs 6 through 15 of gestation. Air sham-treated pregnant micc were exposed in duplicate chambers to filtered, conditioned air using the same exposure and animal handling regmes. Benzene atmospheres were generated by a flow-dilution system (Snyder et ai.. 1978). The apparatus consisted of a 1000-mi flask and reservoir tank containing glass distilled benzene of chromatographic quality (Burdick and Jackson Laboratones. Inc., Muskegon, MI). Benzene vapor was produced by passing an airstream over the surface of liquid benzene at a flow rate controlled by an external rotometer. The benzene-ladened air was then mixed with filtered air leading into the exposure chamber. The chamber concentrations of benzene were determined evey 0.5 hr by uv spectrophotometry ( 5 and 10 ppm) or by an infrared gas analyzer (20 ppm). For uv spectrophotometry, chamber air was bubbled through a glass midget impinger. The resultant alcohol-benzene solution was read for absorbance at 254.5 nm and compared to a predetermined calibration curve. which had been corrected for bubbler etfciency. The infrared analyzer (Miran- CVF. Foxboro Analytical. Norwalk. CT) was precalibrated with known concentrations of benzene. A wavelength of 9.75pm was used for optimum absorption of benzene, with an effective pathlength of 20.15 m. I . Animals 3. E-rposureProtocois Male and female Swiss Webster mice (Crl: CFW(SW)Br, Charles River) were used throughout the experiments. Animals were purchased at 6 weeks of age 3nd allowed to acclimate IO a Iz-hr lightidark cycle for 2 weeks in a climate-controlled room. Except during exposure penods animals were allowed food (Purina lab chow)and water ad librrum.Timed pregnancies were obtained by placing two normal cycling proestrus. virgin females (age 8-12 weeks) with one male during the last 2 hr of the dark cycle. Females were then checked for presence of a vaginal plug or sperm in a vaginal smear with the day ofmating considered as Day Oof pregnancy. Randomly selected pregnant females were then placed into individual control and test groups of 5-10 animals each. Female body weights. as well as general appearance and behavior. were monitored throughout pregnancy. 2. Inhalation Espasures Whole body inhalation exposures were conducted in 1.O m' stainless-steel and plexiglass dynamic chambers Esprrrment I . The responses to in uiero exposure to 5. 10.or 20 ppm benzene were determined in fetuses on the 16th day ofgestation. Five benzene-exposed and five airexposed pregnant mice were sacrificed by cervical dislocation. their uteri removed. and the number of live. dead. and resorbed fetuses recorded. Two male and two female fetuses were then randomly selected. weighed. and examined for any external gross morphological malformation. Peripheral blood samples were taken from trunk blood of decapitated fetuses for red and white cell counts and for hemoglobin analysis. Livers were removed aseptically for enumeration of recognizable cells in the hematopoietic differentiating. proliferating pool (DPP). E\perimmi 2. The responses to rn utero exposure to 5. 10. or 20 ppm benzene were determined in 2day-old neonates. Five benzene-exposed and five airexposed pregnant females were allowed to proceed through normal parturition. Two male and tu0 female neonateswere then randomly selected at 2 davs of age and subjected to the same protocol as that described above with 16dayold fetuses. 226 KELLER AND SNYDER Experimeni 3. Responses due to in utero exposure to measuring the pH on a Beckman Model 3560digital pH 5, 10. or 20 ppm benzene were investigated in 6-week- meter (Beckman Instruments, Inc., IrVine. CA). Addi- old progeny. Five benzeneexposed and five air-exposed tional control gels were stained for hemoglobin with ben- pregnant dams were again allowed t o proceed through zidine to ensure correct identification of the hemoglobin normal parturition. At 6 weeks of age, one male and one bands. female were randomly selected from each litter. Periph- Gels were fixed in 10%trichloroacetic acid and stained eral blood samples were obtained from tail veins for red with Coomassie blue. Stained gels were scanned at 550 and white cell counts and for hemoglobin analysis. These nm on a Gilford Model 250 spectrophotometer (Gilford animalswere then terminated by ceMcal dislocationand Instrument Laboratories, Inc., Oberlin, OH) and the rel- their spleen and femurs removed aseptically for enumer- ative amounts of Hemoglobin A major and the Hem+ ation of recognizable cells in the DPP. globin A minor quantitated by measuring the areas un- der the curves. DPP quaontitalion.Livers from 16day-old fetuses and 4. Blood Cell Parameters 2-day-old neonates were placed in - 5ml icecold, ster- ile-supplemented a medium (SAM-Supplemented Q Peripheral blood. Free flowing venous tail blood from 6-week-oldadults and trunk blood from fetusesand new nates were used for red blood cell (RBC), white blood cell (WBC), and blood cell diKerential counts. Peripheral RBC and WBC counts were determined with a Coulter Counter Model ZB-1 (Coulter Electronics. Hileah. FL). Differential counts were determined by counting 100nucleated cells in Wright/Giemsa stained blood smears. Hemoglobin quaniiiation. Peripheral blood was o b tained as described above. Mean corpuscuiar hemoglobin was determined using a standard spectrophotometric cyanmethemoglobin technique (Boehringer Mannheim Diagnostics, Inc.. Houston, TX). Briefly, 5-10 pl of blood was placed in Drabkin's reagent to lyse the red cells and release hemoglobin. The hemoglobin was convened to cyanmethemoglobin with K3Fe(CN)6and KCN. Ab sorbance of the cyanmethemoglobin solutions were read at 540 nm. The average hemoglobin content per cell was then calculated by dividing total hemoglobin by the re- media: IOOX u media, 1% nonessential amino acid solution. 1% sodium pyruvate solution. 1% L-glutamine, 2% penicillin/streptomycin sulfate solution, sodium bicarbonate, pH 7.2). Single cell suspensions were prepared by repeated aspirations of the tissue through a siliconized fine-tipped Pasteur pipet. Suspensions of splenic cells were obtained by pressing the spleen tissue through a stainless steel wire mesh grid ( 1 50 divisions per square inch) and by repeated aspirations in SAM as described for liver cell suspensions. To obtain marrow cells, the marrow cavities of femurs were repeatedly flushed with cold SAM and separated into single cell suspensions by repeated aspirations. All three types of cell suspensions were then gently pelleted by centrifugation and resuspended in a solution of 3: I fetal calf serum:H1O. Smears ofthese cell suspensionswere made and stained with benzidine and Wright/Giemsa (LoBue et al.. 1963). Slides were coded and read blindly. Five hundred nucleated cells per slide were categorized. spective red cell count. Hemoglobin typing. Peripheral blood obtained as described above was washed in isotonic saline and lysed 5. Dora Evaluation with distilled water. After centrifugation. the supernate was removed and stored in glass vials at -2o'C. Prior to Peripheral and organ blood cell counts were deter- separation on polyacrylamide gels. samples were thawed mined on the four animals (two males and two females) and assayed for protein content by the Pierce BCA Pro- randomly selected from each litter. Cell counts from ani- tein Assay (Pierce Chemical Co., Rockford. IL).All he- mals randomly selected from matched air control litters moglobin samples were diluted to 250 p d m l with dis- were determined each time cell counts were determined tilled water containing 0.01% potassium cyanide and from a benzene-treated litter. Thus. each benzene-ex- stored at 4'C until the followingday. posed animal had its own age-matched air control. In al- The hemoglobins were separated by isoelectric focus- most all cases. there were five litters per exposure. per age ing on polyacrylamide tube gels. Solutions of0.6% etha- group. Differences in the cell counts were evaluated by nolamine and 0.01 M phosphoric acid were used as ca- the Studentf test using the litter as the experimental unit. tholyte and anolyte, respectively. Approximately 25 p g Differencesgreater than the two-tailed, p < 0.05, Student total protein in a volume of I 0 0 pl was applied onto the f value were considered significant. top of the gels in a carrier solution of 8 M urea containing Differential counts of peripheral blood cells and DPP 5% ampholyte. A voltage of 400 V was maintained for 5 cells were evaluated as follows. Peripheral and organ hr and thereafter increased to 800 V for I hr to increase slides made from one male and one female progeny were the resolution of the bands. The pH gradient was assessed randomly selected from each litter. Slidesfrom five litters by sectioninga gel into 20 or more 5-mm pieces, eluting per age group per benzene treatment were so selected. the ampholy& into IO mM potassium chloride, and Slides from five age-matched air control litten were se- HEMATOTOXICITY OF IN LITER0 BENZENE EXPOSURE 227 TABLE I PERIPHERAL BLOOD CELL INDICES (Mean & SE)" RBC/mm3 X IO6 MCH Nucleated cells per mm' Ratio of HbA majordto HbA minor 16-Day fetuses 5 ppm Air 10 ppm Air 5.5 f0.4 4.9 f0.2 5.4 2 1.2 3.4 2 0.3 2.4 k 0. I 2.8k0.1 2.3 f0.3 3.1 k 0 . 3 47.677 k 5.615 35.952 k 3.374 *63,363 17,394 44,145 k 9.342 1.1 f0.1 - 0.8aO.I - 20 ppm Air 2-Day neonates 5 ppm Air 4.3 f0.4 5.2 f0.2 '7.0 2 0.1 6.4 a 0. I 2.1 2 0.3 2. I f0.03 I .8 .+- 0.03 2.0 k 0.07 75,749 2 3.330 99,332 f 10,910 29.151 k 10.003 28.024 k 5,302 1.0 f0. I 0.9 f0.1 1.o k 0.1 - 10 ppm Air 6.4 t 0.4 5.8 2 0.2 2.0 f0. I 2.3 .+- 0.2 4.487 f 419 5,793 2 847 1.0?0.1 - 20 PPm Air 6.9 lr 0.2 6.4 f0.3 2.1 k 0 . l 2.0 k 0. I 5.059 f 560 5,075 ? 658 *1.1 f 0 . 2 1.o 0.2 6-Week adults 5 ppm Air 10 ppm Air 7.6 2 O.Zb 6.6 f0.2 7.1 20.2 6.9 f0.2 1.9 f0.2 2.2 2 0.2 2.2 fr 0.1 2.3 2 0.1 8,983 k 655 9,174* 753 *9,494 649 9.691 k 648 1.1 k0.I - 1.5 k0.1 - 20 ppm Air 7.9 2 O.2b 7.2 2 0.2 2.0 f0. I 2.1 20.1 8.476k 573 8.142 2 722 1.5 f0.2 1.2 ? 0.2 Values determined from fiveindividual litters (two males and two females per litter) per treatment. per age group. Statistically different from control values, p < 0.05. Mean corpuscular hemoglobin. For all indices except ratios of HbA major to HbA minor. there were age-matched air control litters for each litter exposed to a given benzene concentration. For ratios of HbA major 10 HbA minor. one set of age-matched litters served as controls for all benzene-exuosed litters within a given age group. Methods of statistical analyses for all indices are given under Methods. lected in a similar fashion. Unlike the cell count data discussed above, one set of air control slides served as control for all three benzene treatments within a given age group. For example, slides from 16day-oId fetuses exposed in uiero to 5. 10, or 20 ppm benzene were compared to slides from 16day-old fetuses exposed in ufero to air. Other age groupswere handled analogously. These data were evaluated first using a two-way analysisof variance with treatment and gender as covariates. F ratios were significant for differences in the responses of the genders in only 3 of the 42 two-way analyses of variance performed.In Contrast, F ratios were significant for the responses due to benzene treatment in 15 of the 42 twoway analysesof variance performed.For this reason total litter responses (male and female) vs treatment were as- sessed by one-way analysisof variance followed by Dunnett's tests. Ratiosofhemoglobin A major to hernoglobin A minor were also analyzed by one-way analyses of variance followed by Dunnett's tests. Differences between treatmen8 were considered significant when they were greater than the two-tailed, p < 0.05 Dunnett's critical value(Dunnett, 1964). RESULTS The mean daily benzene exposure concentrations k standard deviations for the three exposures employed were 5.1 f 0.5, 9.9 k 0.6, and 20.4 k 0.9 ppm. 228 KELLER AND SNYDER TABLE 2 PERIPHERAL BLOODCELL DIFFERENTIAU" ~ ~~ Late ~~~ Primitive Dividing Nondividing nucleated nucleated nucleated Blasts granulocytes granulocytes red cells red cells red cells Lymphocytes 16-Day fetuses Air 5 PPm 10 ppm 20ppm 0.00rO.00 0.50+_0.16 0.00r0.00 2.lOr0.67 0.10&0.00 0.9020.28 O.lO20.00 1.50k0.50 1.60+0.50 3 . 6 0 2 1.57 1.3020.33 2.2020.63 5.102 1.34 5 . 8 0 2 1.88 4.0020.60 3.90r0.79 0.4020.22 0.80r0.25 l.20-cO.39 1.5020.34 92.421.95 87.4k4.11 92.42 1.20 90.72 1.48 2-Day neonates Air 0.00&0.00 3.80 20.66 5 ppm 0.2020.14 3.10r0.57 IOppm 0.10&0.10 5.902 1.04 20 ppm 0.10+.0.10 2 . l O r 0 . 6 2 67.60 2 2.44 72.3023.09 67.9022.88 80.4022.67' 7.30 r 1.36 6.20k 1.79 1.70r0.62' 3.6020.88 0.50&0.22b 7.30k0.83 O.OO+O.OOb 1.60+-0.45b - -- - 6-Week adults Air 5 PPm lOppm 20 ppm 0.00~0.002 . 7 0 k 0 . 4 7 0.00+0.00 1.20+0.47 O.OO?O.00 0.6020.22 0.10+0.10 2.2020.63 19.32228 22.052.47 24.222.59 16.722.27 O.OOrO.OO 0.2020.14 0.102O.IO 0.2020.13 0.00~0.000.10~0.10 0.10~0.10 0.20?0.13 - - - - 14.0 r 3. I 17.9 2 2.4 16.9 -+_ 2.0 14.2 2 2.5 75.0 2 3.0 12.3 -+ 3.1 75.1 2 2.9 +77.6 2.4 a One hundred cells counted from one male and one female from each of five individual litters per treatment, per 'age group. Numbers arc means 2 SE. Denotes statisticallysignificant from corresponding control values. p < 0.05. I . Maternal To,uicity There was no evidence of maternal toxicity among dams exposed to any concentration of benzene tested as determined by maternal morbidity, mortality, or weight loss during the exposures. 2. Fetal and h'eonatal To-vicity There was no evidence of nonhematopoietic toxicity among any of the fetal or neonatal progeny exposed in ulero to any concentration of benzene studied. Litter sizes, male/ female ratios, and body weights, as well as the numbers of dead, resorbed. or malformed fetuses, were all within control limits. 3. Hematopoietic Toxicity among Progeny F ratios indicated significant differences in the responses of the genders to the exposures in only 3 of the 42 analyses of variance performed. These 3 cases were peripheral blood late nucleated red cellsof 6-week adults, bone marrow dividing granulocytes of 6-week adults, and bone marrow lymphocytes of 6week adults. When the total litter responses were evaluated. none of the values for these three indices among benzene-treated animals were statistically different from control values (Tables 2 and 3). a. Peripheral blood indices. No pattern of statistically significant differences between benzene-exposed and airexposed progeny was observed among any of the peripheral blood counts or hemoglobin indices (Table 1). A small number of statistical differences were observed but these occurred in no discernible pattern and probably have little toxicological significance (Table 1). b. Peripheral blood differentials. There were marked statistically significant depressions in the numbers of early nucleated red HEMATOTOXICITY OF IN UTERO BENZENE EXPOSURE TABLE 3 ~ Blasts CELL D ~ ~ R E N T ~OFAHLEMSATOPOIETIC ORGANS' Dividing Nondividing granulocytes granulocytes Early nucleated red cells Late nucleated red cells 229 Lymphocytes 3.40 = 0.64 4.70 = 0.83 3.20 I1.00 4.70 I0.65 13.2 f 3.1 12.5r2.l 12.7 -t 2.8 5.7 -t 1.4 L4.9 2 2.7 12.1 f 1.9 14.6 f2.5 11.5 f 2.8 19.8 t 2.5 21.7 k 2.3 25.3 k 5.9 20.2 2 2.0 396.5 rc_ 8.5 395.5 -t 6.5 407.4 f 5.5 418.1 -t 5.3 51.6 f7.3 52.9 f4.0 35.2 rt 6. 1 38.5 rt 4.4 0.80 5 0.33 2.80 -t 0.70' 0.80 -t 0.29 2.60 f0.56' 6.90 2 1.1 6.60 t 0.9 7.50 -t 0.9 l7.30f 1.4' 27.024.5 35.0 2 2.1 36.6 f2.7 52.0 f 3.2b 15.0k 1.33 14.9 jr 1.93 10.4 2 0.97 12.5 5 1.47 286.0-t 13.9 *282.0 k 9.0 289.9 4.8 199.4 ? 10.6' 155.5-t 13.6 156.1 f 10.4 150.2 ? 5.6 213.8 f 11.9' 2.22 z 0.89 3.30 k 0.97 1.80 -t 0.63 3.60 -t 0.97 6.30 f 1.0 9 . 5 0 t 1.2 7.60 -t 1.3 8.90 t 1.3 83.128.6 113.5 f 10.5 93.5 f 5.8 78.2f 12.3 18.8~2.6 14.3 5 1.9 12.5 2 1.7 10.45 I S b 123.0f5.2 134.6" 7.5 104.6 f6.8 115.6f9.1 257.6211.7 215.6 -t 15.7' 272.7f 9.9 271.5f 17.7 0.20+0.13 0.30 -t 0.15 '0.40 rt 0.22 I .30 f0.45 I .30 t 0.4 2. IO f0.5 1.30 f0.4 3.80 f0.7' 4.80k1.3 9.40 2 1.4' 3.40rt0.8 15.2 2 2.0' 2.3020.4 4.20 t 0.8 3.6020.8 5.50 k 1.1 38.429.2 50.1 f9.0 44.253.1 41.7 f 5.3 450.9f9.1 430.9 f9.9 445.3lr4.0 430.7 2 8.2 'Five hundred cells counted from one male and one female offspring from each of five individual litters per treat- ment. per age group. Numbers are means fSE. cells (basophilic norrnoblasts) in the peripheral blood of 2-day neonates exposed in utero to all concentrations of benzene studied (Table 2). These depressionsof erythroid precursors occurred in a dose/response pattern. Those neonates exposed in utero to 20 ppm benzene also exhibited depressed numbers of late nucleated red cells (polychromatic normoblastsand their nucleated progeny)and elevated numbers of nondividing granulocytes (granulocytes differentiated to at least the metamyelocyte stage). c. Cell differentials of hematopoietic organs. There were several statistically signifi- Z 30 KELLER A N D SNYDER ) x e d in wferoto 20 pprn benzene (Table 3). throid precursor cells did not affect the levels I'hese ani rnals exhibited elevated numbers of of circulating red cells in these animals. The qdenic: hlasts, dividing granulocytes, and erythron has a large capacity to deal with an ,mndi~idinggranulocytes accompanied by erythropoietic insult. It is known, for exam- depressed levels of marrow early nucleated ple, that during phases of acute anemia, the led cells (basophilic normoblasts). cell cycle time of recognizable erythroid pre- cursor cells shortens (Hanna el al., 1969). DISCUSSION This can lead to a lower steady-state number oferythroid precursors while at the same time producing normal or near normal numbers Exposure of pregnant mice to 20 ppm ben- of mature functioning red cells. Such a com- zene clearly induced hematotoxic responses pensatory mechanism may be in effect in the offspring. Toxicity was evident by de- among the in utero benzeneexposed mice in pressions in the numbers oferythroid precur- this study. sor cells and elevations in the numbers of It is interesting to note that the 2 d a y neo- granulocytic precursor cells in both neonatal nates and 6-week a.iults exposed in utero to and 6-week-old offspring. It is noteworthy benzene exhibited changes in recognizable that the alterations in the numbers ofthe pre- hematopoietic precursor cells but the 16day cursor cell types persisted in the adult mice 7 fetuses did not. This is especially interesting weeks after their in uferoexposures. Decre- because the fetuses were assayed 1 day after ments of erythroid precursor numbers ac- their last exposure while the neonates and companied by increases of granulocytic pre- adults were assayed 1 week and 7 weeks, re- cursor numbers have been observed pre- spectively,after their last exposures. The rap viously in this laboratory among mice idly expanding fetal erythron (Chui 2f ~ 1 . . undergoing benzene exposures (Snyder ef ai., 1971) may be able to temporarily mask any 1980; Baarson er af.. 1982; Rosenthal and damage inflicted by exposure to benzene. Snyder, 1984). However, in the previous However, it appears that some persistent studies, mice were exposed as adults and the damage is induced by the exposures which exposure concentrations were an order of becomes evident when expansion of the ery- magnitude higher than those employed in the thron ceases as the animals age. present study with in iuero-exposed mice. Exposures to benzene in iiiero induced no Neonates exposed to all concentrations of observable changes in the HbA major/HbA benzene studied exhibited depressed levels of minor ratio. Such changes are indicative of peripheral early nucleated red cells (Table 2). functional defects in the normal matura- During the hepatic phase of fetal hematopoi- tional process of erythroid precursor cells. In- esis, erythroid precursor cells compose 50- creasesin HbA minor have been shown to ac- 70% of the nucleated hematopoietic cells company erythropoietic stress in mice (Alter found in the liver (Paul er al., 1969; Tarbutt et ai., 1982; Pappas and Cauchi, 1982) and and Cole, 1970). Our exposures were there- are analogous to increases in fetal hemoglo- fore conducted during a time of preponder- bin seen in some erythropoietically stressed ance of erythroid cells and it is perhaps likely humans. that the erythroid cells would show the great- Although benzene is apparently not a te- est toxic response. Although there were ratogen for common laboratory animals marked depressions in the numbers of recog- (Murray ef af.. 1979; Mehlman et ~ 1 . 1. 980), nizable erythroid precursor cells among 2- our data indicate that maternal exposures day neonates and 6-week adult mice exposed to low concentrations of benzene induce in urero to benzene, these depressions in ery- changes in the developing murine fetal hema- HEMATOTOXICITY OF IN UTERO BENZENE EXPOSURE 23 1 topoietic system. Although many aspects of CHUI, D., DJALDE~MTI.,,MARKS, P.,AND RIFKIND,R. human and murine hematopoietic ontogeny are similar, extrapolation of our findings to the possible risk for humans is difficult for two reasons. First, resting mice respire about ( 197I). Erythropoietineffectson fetal mouse erythroid cells. I. Cell populations and hemoglobin synthesis. J. Cell. Bioi. 51,585-595. DREW, R.. AND LASKIN, s. (1973). Environmental inha- lation chambers. In Mcrhods oJAnimal Experimenra- 10 times more air per minute per gram body tion(W.[.Gay, Ed.), Vol.4.p~. I-41.AcademicPms. weight than resting humans (Guyton, 1947). NY. Thus, for a given air concentration of ben- zene, a mouse would be expected to inhale a larger dose of benzene than a human. On the other hand, a human fetus has a greater ca- DUNNETTC.. (1964). New tabla for multiple comparisonswithacontrol. Biometr&stO,482-491. GILL- I., AND STRIPP. B.( I 975). Pre- and postnatal enzyme capacity for drug metabolite production. Fed. PTOC.34,172-178. pacity to metabolize xenobiotics than a GUYTON,A. (1947). Rapintory volumes of laboratory mouse fetus (Gillette and Stnpp, 1975; Net- animals. Amer. J. Physid. 150.70-77. ter, 1976). Thus, for a given dose of benzene reaching fetal circulations, a human fetus may form greater amounts of the toxic ben- zene metabolites believed responsible for benzene's toxicity (Rushmore et af., 1984) HANNA, 1.. TARBUIT. R.. AND hMERTON. L (1969). Shortening of the cell cycle time of erythroid p m r - -sonin mponw to anemia. Brit.1.Haemamol. l6,38 I 387. U U E R . K. A.. AND SNYDER. C. A. (1986). Mice ex- posed in utero to low concentrations of benzene ex- than a mouse fetus. -hibit enduring changes in their colony forming hema- topoieticcells. Toxicology.4 x 1 7 l 18l . KOIZUMI,A.. AND SUZUKI, T. (1966). A P r e h i n a ~ R e - ACKNOWLEDGMENTS port on the Influence oJBenzeneand theBlood on Mice Congenilally A f l i e d by Benzene. Presented at the 36th Annual Mming oflap. Soc. of Hyg., 216. Thismearchwassupported byGrantOH01713 from the National Institute for Occupational Safety and Health and by Grant 15-54 from the March of Dimes and in part by Center Grant Esoo260 from the National Institute of Environmental Health Sciences. Center Grant CA13343from the National Cancer Institute, and LOBUE.J.. DORNFEST. B.. GORDON, A.. H o w , J.. AND QUASTLER. H. (1963). Marrow distribution of rat fe- murs determined by cell enumeration and FeS9label- ing. Proc. Soc.Exp. Bioi. Med. 112,1058-1062. LUKEN, R.. AND MILLER. s. (1981). The benefits and costs of regulating benzene. J. Air Pollut. Control Special Institutional Grant 00009 from the American .-lSsOC. 31,l254-1259. c.. c.Cancer Society. This work was performed in partial ful- MEHLMAN, M.. SCHREINER. AND MACKERER. fillment of the requirements for the Ph.D. degree by (1980). Current status of benzene teratology: A brief K.A.K. under the tutelage of C.A.S. The authors thank review. J.Environ. Parh. To-ricol.4, 123-131. Mr. Michael Barbieri for operating the inhalation cham- MIZENS. M. (1982). The effects of benzene on e n t h r e benand Ms. Maureen Freitag for typing the manuscript. poicsis in the fetal mouse. Diss. Abst. Int. 42. 145. MURRAY. F. I.. JOHN. J. A,. RAMPY. L. W.. KUNA.R.. AND S C H W E f L B. (1979). Embryotoxicity of inhaled REFERENCES benzene in mice and rabbits. Amer. Industr. Hyg. .~SSOC.J.40.933-998. National Research Council (1980). Drinking Ukter and ALTER. B. P., CAMPBELL A. S.. HOLLAND. J. A.. AND FRIEND. G. (1982). Increased mouse minor hemogle bin during erythroid stms. A model for hernoglobin regulation. Exp. Hemarol. 10,754-760. Health. Vol. 3. pp. 80-86. Academy Pms. Washington. DC. NETTER. K. (1976). Developmental aspects ofdrug metabolism. I n ,\khanisms of T0.ricif.rand Metabolism BAARsON. K.. SNYDER.~A. .. GREEN. J.. SELLAKUMAR.(Karki. N.. Ed.). Vol. 6. pp. 3-16. Pergamon. New A.. GOLDSTEIN. B.. AND ALBERT. R. (1982). The he- York. matotoxic effects of inhaled benzene on peripheral PAPPAS.s.. AND CAUCHI. M. N. ( I 982). Factors a f f i i n g blood. bone marrow, and spleen cellsafe incnased by hemoglobin synthesisin the mouse. Exp. Hematol. 10. ingested ethanol. Toxicol.Appl. Pharmacal. 64,393- 701-708. 404. PAUL. t.,CONKIE. D..AND FRESHNEY. R. (1969).Eryth- Chemical and Engineering Ne% lune I I. 1984. ropoietic cell population changes during the hepatic SNYDER C.A.. ch.DSTEIN, 8. D.. SELLAKUMAA.RR... BROMBERG1., LASKIN,S..AND ALBERT,R. E. (1980). The inhalationtoxicology ofbenzene. Incidence of he- matopoietic neoplasmsand hematotoxicityin AKR/J and C5701/6 mice. Toxicd. Appl. PharmmoI. 54, 323-33 I. SNYDER, c. A,. G o u > s m N , B., SELUKIJMARA. .. WOHLMANS, BROMBERG1., ERLICHMANM. .. AND LAS-, S. (1978). Hematotoxicityof inhaled benzene to Spraguc-Dawley ratsand AKR mice at 300 ppm. J. Toxid. Environ. Health 4,605-6 18. TARSUIT.R.AND COLE, R.(1970). Cell population kinetics oferythroid tissue in the liver of fetal mice. J. Embryd. Exp. Morphd. 24,429-446. U.S. Department of HEW (1974). HEW Publication (NIOSH)74- 137. pp. 20-22. U.S.OSHA (1978). Occupational exposure to benzene. Fed. Reg. 43(52), February.