TERATOLOGY 45603-610 (1992)
Valproic Acid-induced Placental and Teratogenic Effects in Rats K.S. KHERA Toxicology Research Division, Bureau of Chemical Safety, Food Directorate, Health Protection Branch, Health & Welfare Canada, Ottawa, Ontario KIA OL2,Canada
ABSTRACT Studies on teratogenicity and pathology of the cenceptus were conducted in Sprague-Dawley rats treated with 600, 800, and 1,000 mgkg valproic acid PO on day 13 of pregnancy. Each of the three doses was maternotoxic and caused (1)resorptions and/or abortions, reduction in the number of live fetuses per litter and mean fetal weight, and defects of the tail, rib and phalanx; and (2) degenerative changes in the labyrinth (thrombosis, angiectasis in the maternal lacunar network, necrosis of cytotrophoblasts and suppressed proliferation of fetal capillaries), reduced diameter nearing obliteration of umbilical vessels, with or without karyorrhexis of embryonic tissues. The lesions in the placental labyrinth were specific but, in the embryonic tissues, they were generalized. It was postulated that the vascular lesions in the labyrinth and umbilicus may have influenced embryonic development by reducing maternoembryonic gaseous and nutritional exchange. o 1992 Wiley-Liss, Inc. The initial morphologic site of action of a chemical teratogen is generally believed to be in the embryo itself and in the tissues that are directly or indirectly (or inductively) involved in fetal malformations. Therefore, a vast majority of studies on teratogenic mechanisms are conducted on the embryo and its associated tissues. The results obtained leave a lingering doubt whether the embryonic changes are primary or, rather, due to a secondary consequence of a chemical's initial action on maternal and extraembryonic (placental and paraplacental membranal) tissues. The extraembryonic, and some of the maternal, tissues (i.e., blood and vascular network in the placental labyrinth) are responsible for maternoembryonic exchange of 0 2 / C 0 2and nutritiodmetabolite requirements during embryonic development. It is therefore plausible that the functioning of this exchange, if jeopardized by chemically induced alterations in the maternal and extraembryonic tissues, could by itself cause adverse effects in the embryo. This concept, already known, has recently been revived (Khera, '91) since early physiological and histological changes induced by chemicals in the maternal and extraembryonic tissues were found to be related to adverse effects in 0 1992 WILEY-LISS, INC.
term fetuses. Inadequate embryonic nutrition may influence the morphogenesis either on its own or by modulating the expression of homeobox genes and proto-oncogenes involved in prenatal development (Adamson, '87; Astigiano et al., '89). In the present report, changes observed in extraembryonic and embryonic tissues are related to adverse effects on embryonic development following valproic acid dosing of rats. The embryonic response consisted of diverse anomalies (including defects of the kidney, heart, neuraxis, ribs, and phalanFed, resorptions (Ong et al., '83; Vorhees, 87; Ritter et al., '87; Binkerd et al., '88; Klug et al., ,911, and abortions (this study), hence nonspecific or less restricted; however, lesions in the extraembryonic tissue involved the labyrinth and were more specific. MATERIALS AND METHODS
Animals Virgin female Sprague Dawley rats (weighing 225-275 g) were obtained from Charles River, St. Constant, Quebec. They
Received July 9, 1991; accepted December 10,1991
604
K.S. KHERA
were paired overnight with proven breeders. The morning of finding a positive vaginal smear was defined as day 1 of pregnancy. All animals had free access to tap water and Purina Lab Chow. They were housed in a thermostatically maintained room at 22 2 1°C and 50 2 5%relative humidity, illuminated with a 12 h/day light cycle regulated by a time switch.
Chemical and doses Valproic acid (2-propylpentanoic acid) was obtained from Sigma (St. Louis, MO). It was dissolved in propylene glycol and administered as a single dose of 0,600,800, or 1,000 mgtkg body weight by gavage on day 13 of pregnancy. The day of pregnancy and dose combination that gave maximum adverse effects on term fetuses with minimum maternal lethality was determined in a preliminary study. Teratology The toxicity in dams and structural changes in term fetuses were evaluated following routine teratology procedures. Twothirds of fetuses from a litter were examined for skeletal defects and the remainder for visceral anomalies. Histology of the conceptus Six rat dams from each of the 600-, 800-, and l,OOO-mg/kg test groups were killed by CO, asphyxiation a t 24 h after dosing and 6 additional rat dams from the 600-mgkg group at 48 h after dosing. Appropriate propylene glycol treated controls were included for each test group. The entire uterus, with conceptuses in situ was removed and stored in 10% neutral buffered formalin for 3 weeks. Each conceptus, still in the uterine capsule, was dehydrated and embedded in paraffin. Serial 5-pm thick sagittal sections were cut from each conceptus and stained with hematoxylin and eosin (H&E). Observations of serial sections examined by an optic microscope aided in defining the extent of lesions and interrelation of tissues changes in the conceptus. Statistical analysis of teratology data The arithmetic mean (M) and standard error (SE) were calculated for each treatment and control group. Student’s t-test was used to compare the results between treatment and control groups. All data were analysed on a Peach IV Executive computer
using a commercial program prepared for the F- and unpaired t-tests for unequal variances. Only test and control differences at P 5 0.05 are reported. RESULTS
Teratology Valproic acid administered PO on day 13 of pregnancy caused maternal body weight reduction in rats at the 600-, 800-, and l,OOO-mg/kg doses (P < 0.01) and vaginal bleeding within 48 h of dosing at the 1,000mgkg dose (Table 1).At each of the three doses, valproic acid reduced the number of live fetuses (dose-related), decreased the mean fetal body weight and increased the incidences of resorptions and abortions (P < 0.05). Also, a low incidence of shortened tail, forepaw syndactyly or ectrodactyly and hindpaw polydactyly was observed at the 600 and 800 mg/kg doses (Table 1). Histology Rat conceptus In control rats, the “basal venous receptacle” is a flattened cavity lined with cytotrophoblasts, located at the deepest part of the labyrinth just exterior to the chorionic disc (Khera, ’92). It apparently received venous blood from the maternal lacunar network of the labyrinth, which is drained out of the basal venous receptacle via the “major venous channel.” 600 mgkg: 24 and 48 h postdosing Changes in two of the 12 conceptuses a t 24 h postdosing and 6 of the 20 conceptuses a t 48 h postdosing were similar and are described together. The “basal venous receptacle” and maternal vascular lacunae in the labyrinth of seven of the eight valproic acid-treated conceptuses showed the presence of variably sized thrombi (Fig. 1). In two conceptuses, the thrombus, chorionic disc, and the enclosed labyrinthine tissue were all disintegrated; thus, the visceral yolk sac membrane was directly exposed to the degenerating labyrinth. In the third conceptus, the degenerating chorionic disc and the parietal and visceral yolk sac layers all showed a wide gap and, as a consequence, the amnion was in contact with the decidua basalis; the amniotic cavity contained embryonic blood. In the latter and the two preceding conceptuses, no umbilical vessels could be detected. Maternal vascular lacu-
PLACENTAL EFFECTS OF VALPROIC ACID
605
TABLE 1 . Teratology study in Sprague Dawley rats: Valproic acid-induced maternel and fetal effects following maternal PO treatment on day 13 of pregnancy Control Dose, mg valproic acidkg (propylene glycol) 600 800 1,000 1. Data on dams: i. Number: at tenddead 2010 1710 2010 1510 ii. Number showing vaginal bleeding 0 0 0 8 iii. Body weight gain in 2 days after dosing (g) 9.0 2 2.7 -3.0 f 2.3** -1.3 f 1.7** -11.9 2 2.0** iv. Consumption in two days postdosing, meaddam (g) Food 36.1 2 2.4 26.1 f 2.1** 25.1 f 7.9 ND Water 92.3 k 4.6 113.0 f 7.4 88.2 f 20.1 ND 11. Data on fetuses (mean k SE): Live, numberll 13.8 rfr 0.8 6.1f 1.6** 3.2 % 1.2*** 0 Aborted, number& 0 0 8.8 f 1.6*** 7.2 f 1.5*** Resorbed + dead, numberll 0.7 f 0.3 6.7 f 1.6** 4.7 k 1.4** 7.6 rfr 1.5*** 4.4 2 0.1** Body weight, g 5.1 f 0.1 4.3f 0.2** 111. Fetal anomalies, %: Short tail 0 9 13th rib rudimentary 14 21 Forepaw (left or bilateral) Syndactyly 4 0 Ectrodactyly 21 0 Hindpaw (left or bilateral) Polydactyly 0 0
*P < 0.05. **P< 0.01. ***P< 0.001.
nae in the labyrinth were dilated and contained clear eosinophilic fluid in two conceptuses and, in the other two, the basal placenta was reduced in size. No valproic acid-related histologic changes were observed in embryonic tissues. 800 mgkg: 24 h postdosing Of the 16 conceptuses examined, six showed no lesions and the remaining 10 had degenerative changes in the embryonic as well as in the extraembryonic tissues. In the labyrinth, the “basal venous receptacle” and maternal vascular lacunae were markedly dilated (Figs. 2, 3) and, in some instances, with plasma. The cytotrophoblasts lining the maternal lacunae frequently showed widely dispersed foci of karyorrhexis and thrombi anchored to the degenerating cytotrophoblast. In the labyrinth, embryonic capillaries were reduced in size and number, and the major embryonic vessels exterior to the chorionic plate collected into um-
C M T
Abbreviations Fetal capillaries Maternal lacuna Cytotrophoblast
bilical vessels were conspicuously decreased in diameter. Quite often erythroblasts with fragmented nuclei were observed in embryonic capillaries. In addition, “hemangiomatous” lesions of maternal vascular lacunae in the labyrinth, similar to those observed at the 1,000-mgkg dose of valproic acid (Figs. 4,5), were noticed in two conceptuses. The “hemangiomatous” lesion is defined as dysplasia of maternal vascular lacunae into a cluster of microhematomas or blood-containing intercommunicating caverns lined with trophoblastic stroma that contained fewer embryonic capillaries than in the control labyrinth. The umbilical cord was fibrotic and reduced in size with lumens of umbilical vessels reduced or stenotic. Major lesions are listed in Table 2. 1,000 mgkg: 24 h postdosing All five conceptuses studied had lesions in both extraembryonic and embryonic tissues that were similar yet more pronounced than those observed a t the 800-mg/kg dose (Table 2). The labyrinthine changes such as trophoblastic necrosis and reduction in size and number of embryonic capillaries and vessels were far more pronounced at the 1,000-mgkg dose. In two conceptuses, large
606
K.S. KHERA
Fig. 1. Rat labyrinth, 600 mg valproic acidkg. The “basal venous receptable” of the maternal lacunar system occluded with thrombus (arrows), which com-
prises fibrin strands and entrapped maternal erythrocytes. Chorionic plate enclosing fetal vessels is at the bottom. x 80.
DISCUSSION areas of labyrinth were occupied by hemanSignificance of pathologic changes giomatous lesions (see Figs. 4 , 5 ) and, in the other two conceptuses, the parietal and visThe development ofthe labyrinth depends ceral Yolk Sac layers were undergoing karY- on the coherent growth of three simultaorrhexis. In one ConcePtus, maternal laby- neously differentiating components: (1) inrinthine lacunae communicated with the tercommunicatingmaternal vascular lacuyolk sac cavity as the result of thrombotic nae, (2) branching embryonic capillaries disintegration of the intervening labyrinth extended from the allantoic villi, and (3) the tissue and the parietal yolk sac membrane. trophoblast that separate the maternal and In one conceptus, there was a widespread pyknosis of granulated metrial gland cells in the uterine tissue. The umbilical cord was absent in one conFig. 2. Control rat labyrinth, day 14 of pregnancy. ceptus, rudimentary ending blindly in the exocoelomic cavity in two, and degenerated Maternal vascular lacunae containing erythrocytes, fecapillaries containing nucleated erythroblasts and but still in contactwith fetal vascular sys- tal interfaced cytotrophoblast are shown. Maternal lacuna tem in the remaining two conceptuses. containing a fetal erythroblast among erythrocytes (arAll embryos showed generalized karyor- row), although less frequent, is observed in controls. rhexis with disrupted embryonic architec- X 320. Fig. 3. Rat labyrinth, day 14 of pregnancy and 24 h and the presence Of primitive erythro- after 800 mg valproic acidkg dosing. Dilated maternal suggestive Of vascular lacunae with concomitant reductions in fetal cfiic cells in the great capillaries and cytotrophoblasts. a possible hematopoietic disturbance.
PLACENTAL EFFECTS OF VALPROIC ACID
TABLE 2. Valproic acid-induced histologic changes in the conceptal tissues of rats Treatment groups 600 800 Dose (mgikg) Control 016 8/32 10116 Number: affecteaexamined 1. Labyrinth 1. Isolated thrombi surrounded by degenerating 0 7 3 labvrinthine tissue ii. Maternal vascular lacunae: dilated with plasma 0 2 10 iii. Maternal vascular lacunae: hemangiomatouswith 0 1 2 suppressed villigenesis iv. Thrombosis of maternal vascular lacunae associated 0 2 0 with disintegrated Reichert's membrane, and the intervening labyrinthine tissue; exocelomic or yolk sac cavity directly exposed to degenerating labyrinth V. Large embryonic vessels: absent or reduced in 0 0 10 diameter 2. Umbilical cord Absent, or fibrotic with decreased or stenotic lumen of 0 3 10 umbilical vessels 3. Embryo Generalized karyorrhexis 0 0 10
embryonic blood and through which the maternoembryonic exchange is believed to take place. These three components are integrated into interconnected trabeculae that form a spongelike network. The adaptive development of the labyrinth is closely synchronized with a rapid enlargement of the blood vessels in the umbilical cord (or its forerunner, the allantoic stalk) and of the vascular network in the embryo. A vectorial maternoembryonic exchange system thus develops by an increasing diameter of umbilical vessels that link the expanding matternoembryonic vascular beds in the labyrinth with the growing vascular system in the embryo. Valproic acid at the 800- and 1,000-mgkg doses seemed to disrupt the synchrony in the development of the cytotrophoblast, maternal vascular lacunae, and embryonal capillaries in the labyrinth. The surface area at the maternoembryonic vascular interface in the labyrinth and the size of the umbilical vessels, fundamental to a continued embryonic development, were reduced. These changes are likely to modulate nutritional and homeostatic conditions in which the embryo develops and thus to determine its viability and growth. Serious placental
609
1,000 515
5 3 2 3
5
5 5
changes are known to cause adverse effects on embryonic developments. Failure of villous formation at the endometrium-chorioallantoic border and the absence of endometrial invasion by chorionic cells have been reported to result in embryonic death and abortion in donkey-in-horse pregnancy established by embryo transfer (Allen, '82). A reduction in surface exchange area (maternal intervillous space volume/embryonic capillary area) and decreased parenchymal tissue in human and sheep placentas has been reported as the major determinant of embryonic growth (Aherne and Dunnill, '66; Kulhanek et al., '74; Molteni et al., '78; Teasdale, '84). Thrombosis at the three doses of valproic acid was observed in maternal vascular lacunae. The intravascular thrombosis seemed specific to the labyrinthine lacunae since the rest of the maternal or embryonic vascular tree of the conceptus appeared unaffected. Embryonic changes subsequent to thrombosis would largely depend upon the fate of the thrombus, degree of vascular occlusion, alterations in the tissues vascularized by the thrombosed vessel, magnitude of decrease in maternal blood flow and a number of other factors. In two conceptuses from the 600-mgkg group, the consequences of disintegration in labyrinthine tissue subjacent to the thrombus remained unknown Figs. 4 and 5. Rat labyrinth on day 14 of pregnancy, but would seem sufficient to induce serious 24 h after dosing 1,000 mg valproic acidkg. Dysplasia of embryonic effects. Uteroplacental embolizamaternal vascular lacunae into hemangiomatous lesions, reduction in the number of cytotrophoblasts,and tion in pregnant sheep caused intrauterine fewer fetal capillaries. x 80 (Fig. 4),and x 320 (Fig. 5). growth retardation (IUGR) that resulted
610
K.S. KHERA
from decreased oxidative metabolic rate in the embryo [Clapp et al., ’81; Block et al., ’90).Placental lesions such as thrombus formation, infarction, chorioamnionitis (Ornoy et al., ’81),avascularity and multiple cysts (“hydatiform degeneration”) (Huber et al., ’57; Sadovsky and Laufer, ’61; Eckman and Carrow, ’62) have been reported to play an important role in spontaneous abortions of humans. Relationship of lesions with teratogenicity The 600,800, and 1,000 mg valproic acid/ kg caused reductions in the number of live fetuses, mean fetal weight, resorptions andl or abortions, and fetal anomalies (see Table 1).These three doses produced histopathological changes in the extraembryonic tissues; however, generalized embryonic necrosis occurred only at the 800 and 1,000 mgkg. At the last two doses, the presence and absence of embryonic necrosis were always found associated with the respective presence and absence of extraembryonic lesions. Whether a causal interrelationship exists between the extraembryonic lesions and embryonic necrosis a t these doses needs further analysis. The 600-mg/kg dose caused, in term fetuses, the adverse effects (see Table 1) and, at 48 h postdosing, extraembryonic lesions with no detectable change in the embryonic morphology. These findings, tend to suggest that it would be imprudent to dismiss the extraembryonic lesions as being inconsequential in fetal development or to hold a direct embryonal effect of valproic acid as the sole mechanism for adverse fetal effects. LITERATURE CITED Adamson, E.D. (1987) Oncogenes in development. Development, 99t449-471. Aherne, W., and M.S. Dunnill (1966) Quantitative aspects of placental structure. J. Pathol. 91:123-139. Allen, W.R. (1982) Immunological aspects of the equine endometrial cup reaction and the effect of xenogeneic extraspecies pregnancy in horses and donkeys. In: Placenta, Structure and Function. R.B. Heap, J.S. Perry and B.J. Weir, eds. J. Reprod. Fertil., Suppl., 31t57-94. Astigiano, S., M.I. Sherman, and P. Abarzua (1989) Regulation and patterns of endogenous and exogenous
genes expression during differentiation of embryonal carcinoma cells. Environ. Health Perspect., 8Ot25-38. Binkerd, P.E., J.M. Rowland, H. Nau, and A.G. Hendrickx (1988) Evaluation of valproic acid (VPA) developmental toxicity and pharmacokinetics in SpragueDawley rats. Fund. Appl. Toxicol., 11t485-493. Block, S.B., D.H. Schlafer, R.A. Wentworth, and L.A. Kreitzer (1990) Regional blood flow distribution in fetal sheep with intrauterine growth retardation produced by decreased umbilical placental perfusion. J. Dev. Physiol. 13:81-85. Clapp, J.F. 111, H.H. Czeto, R. Larrow, J. Hewitt, and L.I. Mann (1981) Fetal metabolic response to experimental placental damage. Am. J. Obstet. Gynecol., 140:446 -45 1. Eckman, T.R., and L.A. Carrow (1962)Placental lesions in spontaneous abortion. Am. J. Obstet. Gynecol., 84: 222-228. Huber, C.P., J.R. Melin, and F. Vellios (1957) Changes in the chorionic tissue of aborted pregnancy. Am. J . Obstet. Gynecol., 73.569-578. Khera, K.S. (1991) Chemically induced alterations in maternal homeostasis and histology of conceptus: Their etiologic significance in rat fetal anomalies. Teratology, 44~259-297. Khera, K.S. (1992) The “maternal arterial space” in the center of rats’ placenta may be venous, not arterial. Teratology, 45t125-129. Klug, S., C. Lewandowski, F. Zappel, H.-J. Merker, H. Nau, and D. Neubert (1990) Effect of valproic acid, some of its metabolites and analogues on prenatal development of rats in vitro and comparison with effects in vivo. Arch. Toxicol., 64:545-553. Kulhanek, J.F., G. Meschia, E.L. Makowski, and F.C. Battaglia (1974) Changes in DNA content and urea permeability of the sheep placenta. Am. J. Physiol., 226:1257-1263. Molteni, R.A., S.J. Stys and F.C. Battaglia (1978) h l a tionship of fetal and placental weight in human beings: fetallplacental and placental weight ratios at various gestational ages and birth weight distributions. J . Reprod. Med., 21t327. Ong, L.L., J.L. Schardein, J.A. Petrere, R. Sakowski, H. Jordan, R.R. Humphrey, J.E. Fitzgerald, and F.A. de la Iglesia (1983) Teratogenesis of calcium valproate in rats. Fund. Appl. Toxicol., 3t121-126. Ornoy, A., J . Salamon-Arnon, Z. Ben-Zur, and G. Kohn (1981) Placental findings in spontaneous abortion and stillbirths. Teratology, 24:243-252. Ritter, E.J., W.J. Scott, Jr., J.L. Randall, and J.M. Ritter (1987) Teratogenicity of di(2-ethylhexy1)phthalate, 2-ethylhexanol, 2-ethylhexanoic acid, and valproic acid, and potentiation by caffeine. Teratology, 35:41-46. Sadovsky, A., and A. Laufer (1961) Placental changes in early spontaneous abortion. Obstet. Gynecol., 17:678683. Teasdale, F. (1984) Idiopathic intrauterine growth retardation: histomorphometry of human placenta. Placenta, 533-92. Vorhees, C.V. (1987) Teratogenicity and developmental toxicity of valproic acid in rats. Teratology, 35:195202.
Valproic Acid-induced Placental and Teratogenic Effects in Rats K.S. KHERA Toxicology Research Division, Bureau of Chemical Safety, Food Directorate, Health Protection Branch, Health & Welfare Canada, Ottawa, Ontario KIA OL2,Canada
ABSTRACT Studies on teratogenicity and pathology of the cenceptus were conducted in Sprague-Dawley rats treated with 600, 800, and 1,000 mgkg valproic acid PO on day 13 of pregnancy. Each of the three doses was maternotoxic and caused (1)resorptions and/or abortions, reduction in the number of live fetuses per litter and mean fetal weight, and defects of the tail, rib and phalanx; and (2) degenerative changes in the labyrinth (thrombosis, angiectasis in the maternal lacunar network, necrosis of cytotrophoblasts and suppressed proliferation of fetal capillaries), reduced diameter nearing obliteration of umbilical vessels, with or without karyorrhexis of embryonic tissues. The lesions in the placental labyrinth were specific but, in the embryonic tissues, they were generalized. It was postulated that the vascular lesions in the labyrinth and umbilicus may have influenced embryonic development by reducing maternoembryonic gaseous and nutritional exchange. o 1992 Wiley-Liss, Inc. The initial morphologic site of action of a chemical teratogen is generally believed to be in the embryo itself and in the tissues that are directly or indirectly (or inductively) involved in fetal malformations. Therefore, a vast majority of studies on teratogenic mechanisms are conducted on the embryo and its associated tissues. The results obtained leave a lingering doubt whether the embryonic changes are primary or, rather, due to a secondary consequence of a chemical's initial action on maternal and extraembryonic (placental and paraplacental membranal) tissues. The extraembryonic, and some of the maternal, tissues (i.e., blood and vascular network in the placental labyrinth) are responsible for maternoembryonic exchange of 0 2 / C 0 2and nutritiodmetabolite requirements during embryonic development. It is therefore plausible that the functioning of this exchange, if jeopardized by chemically induced alterations in the maternal and extraembryonic tissues, could by itself cause adverse effects in the embryo. This concept, already known, has recently been revived (Khera, '91) since early physiological and histological changes induced by chemicals in the maternal and extraembryonic tissues were found to be related to adverse effects in 0 1992 WILEY-LISS, INC.
term fetuses. Inadequate embryonic nutrition may influence the morphogenesis either on its own or by modulating the expression of homeobox genes and proto-oncogenes involved in prenatal development (Adamson, '87; Astigiano et al., '89). In the present report, changes observed in extraembryonic and embryonic tissues are related to adverse effects on embryonic development following valproic acid dosing of rats. The embryonic response consisted of diverse anomalies (including defects of the kidney, heart, neuraxis, ribs, and phalanFed, resorptions (Ong et al., '83; Vorhees, 87; Ritter et al., '87; Binkerd et al., '88; Klug et al., ,911, and abortions (this study), hence nonspecific or less restricted; however, lesions in the extraembryonic tissue involved the labyrinth and were more specific. MATERIALS AND METHODS
Animals Virgin female Sprague Dawley rats (weighing 225-275 g) were obtained from Charles River, St. Constant, Quebec. They
Received July 9, 1991; accepted December 10,1991
604
K.S. KHERA
were paired overnight with proven breeders. The morning of finding a positive vaginal smear was defined as day 1 of pregnancy. All animals had free access to tap water and Purina Lab Chow. They were housed in a thermostatically maintained room at 22 2 1°C and 50 2 5%relative humidity, illuminated with a 12 h/day light cycle regulated by a time switch.
Chemical and doses Valproic acid (2-propylpentanoic acid) was obtained from Sigma (St. Louis, MO). It was dissolved in propylene glycol and administered as a single dose of 0,600,800, or 1,000 mgtkg body weight by gavage on day 13 of pregnancy. The day of pregnancy and dose combination that gave maximum adverse effects on term fetuses with minimum maternal lethality was determined in a preliminary study. Teratology The toxicity in dams and structural changes in term fetuses were evaluated following routine teratology procedures. Twothirds of fetuses from a litter were examined for skeletal defects and the remainder for visceral anomalies. Histology of the conceptus Six rat dams from each of the 600-, 800-, and l,OOO-mg/kg test groups were killed by CO, asphyxiation a t 24 h after dosing and 6 additional rat dams from the 600-mgkg group at 48 h after dosing. Appropriate propylene glycol treated controls were included for each test group. The entire uterus, with conceptuses in situ was removed and stored in 10% neutral buffered formalin for 3 weeks. Each conceptus, still in the uterine capsule, was dehydrated and embedded in paraffin. Serial 5-pm thick sagittal sections were cut from each conceptus and stained with hematoxylin and eosin (H&E). Observations of serial sections examined by an optic microscope aided in defining the extent of lesions and interrelation of tissues changes in the conceptus. Statistical analysis of teratology data The arithmetic mean (M) and standard error (SE) were calculated for each treatment and control group. Student’s t-test was used to compare the results between treatment and control groups. All data were analysed on a Peach IV Executive computer
using a commercial program prepared for the F- and unpaired t-tests for unequal variances. Only test and control differences at P 5 0.05 are reported. RESULTS
Teratology Valproic acid administered PO on day 13 of pregnancy caused maternal body weight reduction in rats at the 600-, 800-, and l,OOO-mg/kg doses (P < 0.01) and vaginal bleeding within 48 h of dosing at the 1,000mgkg dose (Table 1).At each of the three doses, valproic acid reduced the number of live fetuses (dose-related), decreased the mean fetal body weight and increased the incidences of resorptions and abortions (P < 0.05). Also, a low incidence of shortened tail, forepaw syndactyly or ectrodactyly and hindpaw polydactyly was observed at the 600 and 800 mg/kg doses (Table 1). Histology Rat conceptus In control rats, the “basal venous receptacle” is a flattened cavity lined with cytotrophoblasts, located at the deepest part of the labyrinth just exterior to the chorionic disc (Khera, ’92). It apparently received venous blood from the maternal lacunar network of the labyrinth, which is drained out of the basal venous receptacle via the “major venous channel.” 600 mgkg: 24 and 48 h postdosing Changes in two of the 12 conceptuses a t 24 h postdosing and 6 of the 20 conceptuses a t 48 h postdosing were similar and are described together. The “basal venous receptacle” and maternal vascular lacunae in the labyrinth of seven of the eight valproic acid-treated conceptuses showed the presence of variably sized thrombi (Fig. 1). In two conceptuses, the thrombus, chorionic disc, and the enclosed labyrinthine tissue were all disintegrated; thus, the visceral yolk sac membrane was directly exposed to the degenerating labyrinth. In the third conceptus, the degenerating chorionic disc and the parietal and visceral yolk sac layers all showed a wide gap and, as a consequence, the amnion was in contact with the decidua basalis; the amniotic cavity contained embryonic blood. In the latter and the two preceding conceptuses, no umbilical vessels could be detected. Maternal vascular lacu-
PLACENTAL EFFECTS OF VALPROIC ACID
605
TABLE 1 . Teratology study in Sprague Dawley rats: Valproic acid-induced maternel and fetal effects following maternal PO treatment on day 13 of pregnancy Control Dose, mg valproic acidkg (propylene glycol) 600 800 1,000 1. Data on dams: i. Number: at tenddead 2010 1710 2010 1510 ii. Number showing vaginal bleeding 0 0 0 8 iii. Body weight gain in 2 days after dosing (g) 9.0 2 2.7 -3.0 f 2.3** -1.3 f 1.7** -11.9 2 2.0** iv. Consumption in two days postdosing, meaddam (g) Food 36.1 2 2.4 26.1 f 2.1** 25.1 f 7.9 ND Water 92.3 k 4.6 113.0 f 7.4 88.2 f 20.1 ND 11. Data on fetuses (mean k SE): Live, numberll 13.8 rfr 0.8 6.1f 1.6** 3.2 % 1.2*** 0 Aborted, number& 0 0 8.8 f 1.6*** 7.2 f 1.5*** Resorbed + dead, numberll 0.7 f 0.3 6.7 f 1.6** 4.7 k 1.4** 7.6 rfr 1.5*** 4.4 2 0.1** Body weight, g 5.1 f 0.1 4.3f 0.2** 111. Fetal anomalies, %: Short tail 0 9 13th rib rudimentary 14 21 Forepaw (left or bilateral) Syndactyly 4 0 Ectrodactyly 21 0 Hindpaw (left or bilateral) Polydactyly 0 0
*P < 0.05. **P< 0.01. ***P< 0.001.
nae in the labyrinth were dilated and contained clear eosinophilic fluid in two conceptuses and, in the other two, the basal placenta was reduced in size. No valproic acid-related histologic changes were observed in embryonic tissues. 800 mgkg: 24 h postdosing Of the 16 conceptuses examined, six showed no lesions and the remaining 10 had degenerative changes in the embryonic as well as in the extraembryonic tissues. In the labyrinth, the “basal venous receptacle” and maternal vascular lacunae were markedly dilated (Figs. 2, 3) and, in some instances, with plasma. The cytotrophoblasts lining the maternal lacunae frequently showed widely dispersed foci of karyorrhexis and thrombi anchored to the degenerating cytotrophoblast. In the labyrinth, embryonic capillaries were reduced in size and number, and the major embryonic vessels exterior to the chorionic plate collected into um-
C M T
Abbreviations Fetal capillaries Maternal lacuna Cytotrophoblast
bilical vessels were conspicuously decreased in diameter. Quite often erythroblasts with fragmented nuclei were observed in embryonic capillaries. In addition, “hemangiomatous” lesions of maternal vascular lacunae in the labyrinth, similar to those observed at the 1,000-mgkg dose of valproic acid (Figs. 4,5), were noticed in two conceptuses. The “hemangiomatous” lesion is defined as dysplasia of maternal vascular lacunae into a cluster of microhematomas or blood-containing intercommunicating caverns lined with trophoblastic stroma that contained fewer embryonic capillaries than in the control labyrinth. The umbilical cord was fibrotic and reduced in size with lumens of umbilical vessels reduced or stenotic. Major lesions are listed in Table 2. 1,000 mgkg: 24 h postdosing All five conceptuses studied had lesions in both extraembryonic and embryonic tissues that were similar yet more pronounced than those observed a t the 800-mg/kg dose (Table 2). The labyrinthine changes such as trophoblastic necrosis and reduction in size and number of embryonic capillaries and vessels were far more pronounced at the 1,000-mgkg dose. In two conceptuses, large
606
K.S. KHERA
Fig. 1. Rat labyrinth, 600 mg valproic acidkg. The “basal venous receptable” of the maternal lacunar system occluded with thrombus (arrows), which com-
prises fibrin strands and entrapped maternal erythrocytes. Chorionic plate enclosing fetal vessels is at the bottom. x 80.
DISCUSSION areas of labyrinth were occupied by hemanSignificance of pathologic changes giomatous lesions (see Figs. 4 , 5 ) and, in the other two conceptuses, the parietal and visThe development ofthe labyrinth depends ceral Yolk Sac layers were undergoing karY- on the coherent growth of three simultaorrhexis. In one ConcePtus, maternal laby- neously differentiating components: (1) inrinthine lacunae communicated with the tercommunicatingmaternal vascular lacuyolk sac cavity as the result of thrombotic nae, (2) branching embryonic capillaries disintegration of the intervening labyrinth extended from the allantoic villi, and (3) the tissue and the parietal yolk sac membrane. trophoblast that separate the maternal and In one conceptus, there was a widespread pyknosis of granulated metrial gland cells in the uterine tissue. The umbilical cord was absent in one conFig. 2. Control rat labyrinth, day 14 of pregnancy. ceptus, rudimentary ending blindly in the exocoelomic cavity in two, and degenerated Maternal vascular lacunae containing erythrocytes, fecapillaries containing nucleated erythroblasts and but still in contactwith fetal vascular sys- tal interfaced cytotrophoblast are shown. Maternal lacuna tem in the remaining two conceptuses. containing a fetal erythroblast among erythrocytes (arAll embryos showed generalized karyor- row), although less frequent, is observed in controls. rhexis with disrupted embryonic architec- X 320. Fig. 3. Rat labyrinth, day 14 of pregnancy and 24 h and the presence Of primitive erythro- after 800 mg valproic acidkg dosing. Dilated maternal suggestive Of vascular lacunae with concomitant reductions in fetal cfiic cells in the great capillaries and cytotrophoblasts. a possible hematopoietic disturbance.
PLACENTAL EFFECTS OF VALPROIC ACID
TABLE 2. Valproic acid-induced histologic changes in the conceptal tissues of rats Treatment groups 600 800 Dose (mgikg) Control 016 8/32 10116 Number: affecteaexamined 1. Labyrinth 1. Isolated thrombi surrounded by degenerating 0 7 3 labvrinthine tissue ii. Maternal vascular lacunae: dilated with plasma 0 2 10 iii. Maternal vascular lacunae: hemangiomatouswith 0 1 2 suppressed villigenesis iv. Thrombosis of maternal vascular lacunae associated 0 2 0 with disintegrated Reichert's membrane, and the intervening labyrinthine tissue; exocelomic or yolk sac cavity directly exposed to degenerating labyrinth V. Large embryonic vessels: absent or reduced in 0 0 10 diameter 2. Umbilical cord Absent, or fibrotic with decreased or stenotic lumen of 0 3 10 umbilical vessels 3. Embryo Generalized karyorrhexis 0 0 10
embryonic blood and through which the maternoembryonic exchange is believed to take place. These three components are integrated into interconnected trabeculae that form a spongelike network. The adaptive development of the labyrinth is closely synchronized with a rapid enlargement of the blood vessels in the umbilical cord (or its forerunner, the allantoic stalk) and of the vascular network in the embryo. A vectorial maternoembryonic exchange system thus develops by an increasing diameter of umbilical vessels that link the expanding matternoembryonic vascular beds in the labyrinth with the growing vascular system in the embryo. Valproic acid at the 800- and 1,000-mgkg doses seemed to disrupt the synchrony in the development of the cytotrophoblast, maternal vascular lacunae, and embryonal capillaries in the labyrinth. The surface area at the maternoembryonic vascular interface in the labyrinth and the size of the umbilical vessels, fundamental to a continued embryonic development, were reduced. These changes are likely to modulate nutritional and homeostatic conditions in which the embryo develops and thus to determine its viability and growth. Serious placental
609
1,000 515
5 3 2 3
5
5 5
changes are known to cause adverse effects on embryonic developments. Failure of villous formation at the endometrium-chorioallantoic border and the absence of endometrial invasion by chorionic cells have been reported to result in embryonic death and abortion in donkey-in-horse pregnancy established by embryo transfer (Allen, '82). A reduction in surface exchange area (maternal intervillous space volume/embryonic capillary area) and decreased parenchymal tissue in human and sheep placentas has been reported as the major determinant of embryonic growth (Aherne and Dunnill, '66; Kulhanek et al., '74; Molteni et al., '78; Teasdale, '84). Thrombosis at the three doses of valproic acid was observed in maternal vascular lacunae. The intravascular thrombosis seemed specific to the labyrinthine lacunae since the rest of the maternal or embryonic vascular tree of the conceptus appeared unaffected. Embryonic changes subsequent to thrombosis would largely depend upon the fate of the thrombus, degree of vascular occlusion, alterations in the tissues vascularized by the thrombosed vessel, magnitude of decrease in maternal blood flow and a number of other factors. In two conceptuses from the 600-mgkg group, the consequences of disintegration in labyrinthine tissue subjacent to the thrombus remained unknown Figs. 4 and 5. Rat labyrinth on day 14 of pregnancy, but would seem sufficient to induce serious 24 h after dosing 1,000 mg valproic acidkg. Dysplasia of embryonic effects. Uteroplacental embolizamaternal vascular lacunae into hemangiomatous lesions, reduction in the number of cytotrophoblasts,and tion in pregnant sheep caused intrauterine fewer fetal capillaries. x 80 (Fig. 4),and x 320 (Fig. 5). growth retardation (IUGR) that resulted
610
K.S. KHERA
from decreased oxidative metabolic rate in the embryo [Clapp et al., ’81; Block et al., ’90).Placental lesions such as thrombus formation, infarction, chorioamnionitis (Ornoy et al., ’81),avascularity and multiple cysts (“hydatiform degeneration”) (Huber et al., ’57; Sadovsky and Laufer, ’61; Eckman and Carrow, ’62) have been reported to play an important role in spontaneous abortions of humans. Relationship of lesions with teratogenicity The 600,800, and 1,000 mg valproic acid/ kg caused reductions in the number of live fetuses, mean fetal weight, resorptions andl or abortions, and fetal anomalies (see Table 1).These three doses produced histopathological changes in the extraembryonic tissues; however, generalized embryonic necrosis occurred only at the 800 and 1,000 mgkg. At the last two doses, the presence and absence of embryonic necrosis were always found associated with the respective presence and absence of extraembryonic lesions. Whether a causal interrelationship exists between the extraembryonic lesions and embryonic necrosis a t these doses needs further analysis. The 600-mg/kg dose caused, in term fetuses, the adverse effects (see Table 1) and, at 48 h postdosing, extraembryonic lesions with no detectable change in the embryonic morphology. These findings, tend to suggest that it would be imprudent to dismiss the extraembryonic lesions as being inconsequential in fetal development or to hold a direct embryonal effect of valproic acid as the sole mechanism for adverse fetal effects. LITERATURE CITED Adamson, E.D. (1987) Oncogenes in development. Development, 99t449-471. Aherne, W., and M.S. Dunnill (1966) Quantitative aspects of placental structure. J. Pathol. 91:123-139. Allen, W.R. (1982) Immunological aspects of the equine endometrial cup reaction and the effect of xenogeneic extraspecies pregnancy in horses and donkeys. In: Placenta, Structure and Function. R.B. Heap, J.S. Perry and B.J. Weir, eds. J. Reprod. Fertil., Suppl., 31t57-94. Astigiano, S., M.I. Sherman, and P. Abarzua (1989) Regulation and patterns of endogenous and exogenous
genes expression during differentiation of embryonal carcinoma cells. Environ. Health Perspect., 8Ot25-38. Binkerd, P.E., J.M. Rowland, H. Nau, and A.G. Hendrickx (1988) Evaluation of valproic acid (VPA) developmental toxicity and pharmacokinetics in SpragueDawley rats. Fund. Appl. Toxicol., 11t485-493. Block, S.B., D.H. Schlafer, R.A. Wentworth, and L.A. Kreitzer (1990) Regional blood flow distribution in fetal sheep with intrauterine growth retardation produced by decreased umbilical placental perfusion. J. Dev. Physiol. 13:81-85. Clapp, J.F. 111, H.H. Czeto, R. Larrow, J. Hewitt, and L.I. Mann (1981) Fetal metabolic response to experimental placental damage. Am. J. Obstet. Gynecol., 140:446 -45 1. Eckman, T.R., and L.A. Carrow (1962)Placental lesions in spontaneous abortion. Am. J. Obstet. Gynecol., 84: 222-228. Huber, C.P., J.R. Melin, and F. Vellios (1957) Changes in the chorionic tissue of aborted pregnancy. Am. J . Obstet. Gynecol., 73.569-578. Khera, K.S. (1991) Chemically induced alterations in maternal homeostasis and histology of conceptus: Their etiologic significance in rat fetal anomalies. Teratology, 44~259-297. Khera, K.S. (1992) The “maternal arterial space” in the center of rats’ placenta may be venous, not arterial. Teratology, 45t125-129. Klug, S., C. Lewandowski, F. Zappel, H.-J. Merker, H. Nau, and D. Neubert (1990) Effect of valproic acid, some of its metabolites and analogues on prenatal development of rats in vitro and comparison with effects in vivo. Arch. Toxicol., 64:545-553. Kulhanek, J.F., G. Meschia, E.L. Makowski, and F.C. Battaglia (1974) Changes in DNA content and urea permeability of the sheep placenta. Am. J. Physiol., 226:1257-1263. Molteni, R.A., S.J. Stys and F.C. Battaglia (1978) h l a tionship of fetal and placental weight in human beings: fetallplacental and placental weight ratios at various gestational ages and birth weight distributions. J . Reprod. Med., 21t327. Ong, L.L., J.L. Schardein, J.A. Petrere, R. Sakowski, H. Jordan, R.R. Humphrey, J.E. Fitzgerald, and F.A. de la Iglesia (1983) Teratogenesis of calcium valproate in rats. Fund. Appl. Toxicol., 3t121-126. Ornoy, A., J . Salamon-Arnon, Z. Ben-Zur, and G. Kohn (1981) Placental findings in spontaneous abortion and stillbirths. Teratology, 24:243-252. Ritter, E.J., W.J. Scott, Jr., J.L. Randall, and J.M. Ritter (1987) Teratogenicity of di(2-ethylhexy1)phthalate, 2-ethylhexanol, 2-ethylhexanoic acid, and valproic acid, and potentiation by caffeine. Teratology, 35:41-46. Sadovsky, A., and A. Laufer (1961) Placental changes in early spontaneous abortion. Obstet. Gynecol., 17:678683. Teasdale, F. (1984) Idiopathic intrauterine growth retardation: histomorphometry of human placenta. Placenta, 533-92. Vorhees, C.V. (1987) Teratogenicity and developmental toxicity of valproic acid in rats. Teratology, 35:195202.
