- Email: [email protected]
Adverse effect of Tripterygium wilfordii extract on mouse embryonic development
Adverse Effect of Tripterygium vdfovdii Extract on Mouse Embryonic Development W.Y. Chan’ and T.B. Ng* Tripterygium wilfordii Hook f. is a medicinal plant which possesses anti-fertility activity in males and has been used in China for the treatment of rheumatoid arthritis, chronic nephritis and other pulmonary diseases for years. The effect of Tripterygium wilfordii extract on in vitro development in day 9 and day 10 mouse embryos was examined. The embryotoxicity of T. wilfordii extract was obvious on day 9 and day 10 embryos at doses of 50 and 100 kg/ml, respectively. The embryonic abnormalities produced included abnormal branchial apparatus, open cranial neural tube and absence of forelimb bud in day 9 embryos. In day 10 embryos, the abnormality appeared first in the branchial apparatus. At h&her doses, additional abnormalities appeared: abnormal optic and otic vesicles and twisted body axis in day 9 embryos; microcephaly, hematoma in cranial region and absence of hindlimb buds in day 10 embryos. There were reductions in final somite number, yolk sac surface area and axial length. Dead cells were present all over the embryo. The blood circulation of the visceral yolk sac of the abnormal embryos became defective. Ultrastructural analysis of the visceral yolk sac revealed abnormalities including decrease in the amount of microvilli, reduction i,n the number of storage vesicles, appearance of large vacant vacuoles and decrease in the number of pinocytotic invaginations on the apical surface of the yolk sac epithelium. The embryotoxic effect of T. wilfordii extract was, however, destroyed by heat treatment of the extract at 80°C for 6 hours. CONTRACEPTION 1995;51:65-71 Tripterygium wilfordii, bryonic development, mouse
KEY WORDS:
aqueous extract, em-
Introduction ripterygium wilfordii Hook f. (TW) is a vinelike plant that grows mainly in southern China. It has been used as a medicinal herb for hundreds of years in traditional Chinese medicine in the treatment of rheumatoid arthritis, chronic nephritis,
T
Departments of ‘Anatomy and %iochemistry. Faculty of Medicine, The Chinese University of Hong Kong, Shagin, N.T., Hong Kong Submitted for publication J ne 16, 1994 Accepted for publication S tember 14, 1994 Correspondence: Dr. W.Y. han, Department of Anatomy, Faculty of Medicine, The Chinese University of$ Hong Kong, Shatin, N.T., Hong Kong, TEL: (852) 2609 6895; FAX: (852) 2603 5031.
Q 1995 Elsevier Science Inc. 655 Avenue of the Americas, New York,
NY 10010
tuberculosis and other pulmonary diseases without serious side effects.’ However, long-term administration of TW leads to a decrease in testicular weight or menstrual disturbance in some patients.2 Evidence showed that extracts of TW possess anti-fertility activity in male rats.3-5 The viability of cultured rat Leydig and Sertoli cells was reduced by compounds purified from the extract of TW.6 In rats, the motility of spermatozoa was lost and fertility was reduced after treatment with TW in vivo.6 Evidence also showed that TW was able to exert a powerful suppressive effect on human immune response by inhibiting antigen- and mitogen-stimulated proliferation of T-cells and B-cells in vitro.’ Pharmacological studies using animals suggested TW might be potentially useful for diseases such as systemic lupus erythematosus or nephrotic syndrome.8 Recently, TW has been employed in some Chinese clinics to treat cancer, rheumatoid arthritis and nephritis.‘-l2 In view of the clinical efficacy of TW and the potentiality of TW being developed into a new drug, it is of paramount importance to investigate if TW is embryotoxic or teratogenic. It has been demonstrated that some plant proteins designated as ribosome inactivating proteins manifest anti-fertility, immunomodulatory and adverse effects on the female reproductive system, 13-i5 and these proteins also possess embryotoxic activity. 13,i4 In the present study, an aqueous extract of TW was obtained according to the traditional method of preparation for use by patients. An in vitro whole embryo culture system using mouse as the animal model was employed for the examination of the effect of the extract on embryonic development.
Materials and Methods Preparation
of the aqueous
extract
of
Tripterygium wilfordii Dried roots of T. wilfordii were bought from a herbal store and were authenticated according to the morphological criteria set in the Encyclopedia of Chinese Medicines.’ The outer two layers of the roots were peeled off and the core containing vascular tissue was extracted with distilled water as described in the ISSN OOlO-7824/95/$9.50 SSDI 0010-7824(94)00001-D
66
Chan and Ng
Table
1. Effect
Contraception 1995;51:65-71
of Tripterygium
wilfordii
extract on development
of day-9 mouse embryos in vitro before and after
heat treatment Day 9 embryos (embryos with 6-S somites)
Concentration (w/ml)
Final somite number3
Surface area4 of yolk sac3 (mm’)
Axial length3 (mm)
40 (100) 42( 100)
22.5 k 1.2 24.2 * 1.6
16.7 k 1.9 17.9 k 4.3
8.1 of: 1.0 8.3 zt 1.2
10 13 26
2 (20)t 0 (W 0 (W
19.4 -c 1.7 14.9 IL 1.8’ -5
11.9 + 2.1 8.1 2 2.1’ 5.5 * 1.3’
6.6 k 1.0 4.5 2 1.0* 3.6 * 0.7’
13 11
13 (100) 10 1911
21.9 k 1.4 23.5 k 1.2
11.7 k 2.5 16.9 * 3.0
6.7 ? 1.1 8.7 * 1.2
No. of embryos treated
No. of embryos alive’ (%)2
40 42
Control In vivo In vitro Experimental, before heat treatment 50 100 200
Experimental, after heat treatment at 80°C for 6 hours 200 300 ‘Embryos with strong heartbeat and active yolk sac circulation. 2Number in parentheses represents percentage. %alues are expressed as mean c S.E. 4Yolk sac is assumed to be spherical. %omites are not discernible in all the embryos examined. ‘p < 0.05, significantly different from the in viva control values tp < 0.05, significantly different from the in viva control values
by Student’s t-test. by Chi-square test.
Encyclopedia of Chinese Medicines’ with modifications. After one hour of extraction in water, the extract was lyophilized and reconstituted with phosphate-buffered saline (pH 7.2) before use. Heattreatment of the reconstituted extract was carried out in a glass vessel at 80°C for 6 hours. In vitro Study Randomly bred ICR female mice were paired with a male mouse overnight. The presence of vaginal plugs the following morning was designated as day 1 of pregnancy. Embryos were explanted and cultured using a procedure modified from Sadler’s whole embryo culture method.16 Embryos with 6-8 and 20-22 somites were, respectively, collected from day 9 and day 10 pregnant mice and dissected out of the decidua in PBl medium.” The visceral yolk sac and ectoplacental cone were left intact after removal of Reichert’s membrane. Culture of embryos in groups of five was carried out in 5 ml culture medium containing 2.5 ml Dubecco’s modified Eagle’s medium (DMEM, Gibco) and 2.5 ml heat-inactivated (56°C for 30 min) rat serum which had been centrifuged immediately after collection. The culture medium had been sterilized by filtration through Millipore filters (0.22 pm pore size) and equilibrated overnight with 5% COZ in air. The aqueous extract of T. wilfordii in a volume of less than 20 ~1 of phosphate-buffered saline was added to the culture medium contained in a 50ml serum bottle (Wheaton). The cultures were maintained on a roller (20-40 rpm, BTC Engineering) at
37°C and regassed every 8 hours with 5% CO, in air for day 9 embryos and 40% 02, 5% CO, and 45% N, gas mixture for day 10 embryos. of Cultured Embryos After 24 h in culture, the yolk sac circulation and heartbeat of the embryos were examined following transfer to PBl medium. They were then dissected free of embryonic membranes and washed twice with phosphate-buffered saline. Observations were made on the gross morphology, somite number and various organ primordia. Special attention was paid to any abnormalities formed. The axial lengths of embryos were determined by planimetric computation of camera lucida drawings of the embryos on a digitizer (Grafbar GP-7, Science Accessories Corporation, USA). Embryos were fixed either in Bouin’s fixative and processed for light microscopic examination, or in half-strength Karnovsky’s fixative’s for 1.5 hours. Embryos for scanning electron microscopy were then post-fixed with 1% osmium tetroxide, and criticalpoint dried with Freon. Gold-palladium-coated specimens were examined with a Jeol JSM-35 GF microscope operated at 15 KV. The yolk sacs of the embryo were fixed for transmission electron microscopy in half-strength Karnovsky’s fixative for 1.5 hours, postfixed in 1% osmium tetroxide, dehydrated in a graded ethanol series and embedded in Spurr’s plastics. SecExamination
tions of 0.5-1.0
toluidine uranyl
km thickness
blue. Ultrathin acetate
were stained with
1%
sections were stained with
and lead citrate
and then
examined
Contraception 1995;51:65-71
Table
T. wilfordii Extract and Mouse Embryonic Development
67
1. Continued
Abnormality No. of abnormal embryos W)
Abnormal optic vesicle
Abnormal otic vesicle
Absence of forelimb bud
3 (81 5 112) 8 (80)t 13 (100)t 26 (lOO)t
with a Jeol JEM-100 KV.
0 PI
2 (5’
(%)’ Twisted body axis
Open cranial neural tube
1 (3) 0 (0’
3 (81 5 (121
0 (0’
Abnormal branchial apparatus
cl (01
0 (01
4 (31)t 23 (88)t
4 (31)t 25 (96)t
8 (80)t 13 (100)t 26 (lOO)t
ll(85)t
23 (88)t
4 (40)t 5 (38)t 26 (lOO)t
17 (65)t
0 (0) 0 PI
5 (38)t 1 191
0 (0’ 0 (0’
0 PI 0 (0’
cl PI 0 (0)
CXII microscope
operating
at 80
Results Effect on in vitro Development of Embryos During Organogenesis Twenty-four hours after explant, all day 9 control embryos with an initial somite number of 6-8 developed normally in the culture (Table 1). The extent of morphogenesis in vitro paralleled that in vivo. Both in vitro and in vivo control embryos showed a low background of abnormahty (Table 1). All morphological parameters measured including the size of the the yolk sac, axial length of the embryo and somite number are similar in these two control groups. T. tilfordii extract (TWE) at a dose of 50 kg/ml induced a dramatic increase in the number and percentage (80%) of abnormal embryos formed. All abnormal embryos exhibited a slow heartbeat and a sluggish yolk sac circulation. Abnormalities observed in the abnormal embryos included absence of forelimb buds, open cranial neural tube and abnormal branchial arches which were either fewer in number as compared with the control or shorter than normal. At 100 pg/ml, all embryos examined did not show a strong heartbeat or an active yolk sac circulation. The yolk sac became smaller than the control and was opaque. Both somite number and axial length were reduced. Abnormalities in otic and optic vesicles, forelimb bud, body axis, cranial neural tube a d branchial apparatus appeared. When the treated em t: ryos were serially sectioned and examined with a mjcroscrope, it was found that in most of the regions examined, more dead cells were found in the treated iembryo than in the control embryo (Fig. 1). At 200 Hg/ml, the size of the yolk sac and the axial length of tlz embryos were further reduced.
2 (20)t
1 (81
None of the embryos showed any heartbeat. The yolk sac became very opaque and blood cells accumulated inside blood vessels of the yolk sac. All the embryos examined were grossly abnormal. Most of the organs examined showed abnormalities and the boundaries of the somites were not identifiable. Dead cells were found everywhere in the embryo. However, when the TWE was heat-treated at 80°C for 6 hours before adding to the culture, the embryotoxic effect was considerably attenuated (Table 1 J. Heat-treated TWE at a concentration as high as 300 Fg/ml did not produce any abnormalities in cultured embryos. Day 10 control embryos with an initial somite number of 20-22 developed normally in vitro 24 hours after culture and the development of the embryos was comparable to that of embryos developed in vivo (Table 2). Culture of day 10 mouse embryos for 24 hours in the presence of TWE resulted in abnormalities only in the branchial apparatus when the dose of the extract was 100 p&ml. When the dose was raised to 200 kg/ml and 300 kg/ml, additional abnormalities including absence of forelimb bud, hematoma in cranial region and microcephaly were evident. Furthermore, there were reductions in the final somite number, axial length and yolk sac surface area (Table 2). All of the abnormal embryos exhibited slow heartbeat and the yolk sac was very opaque. In most of developing organs, more dead cells were found in the treated embryos than in the control embryo (Fig. 2). The embryotoxic effect disappeared after the extract was heat-treated at 80°C for 6 hours (Table 2). Effect on the Morphology of the Visceral Yolk Sac The visceral yolk sac of the control embryos during organogenesis consisted of a layer of endodermal cells and a layer of mesodermal tissues (Fig. 3A) containing
68
Chan and Ng
Contraception 1995;51:65-71
2. Effects of Tripterygium treatment
Table
wilfordii
extract on development of day-10 mouse embryos in vitro before and after heat
Day-10 embryos (embryos with 20-22 somites)
Concentration Wml) Control ln vivo In vitro Experimental, before heat treatment 100
No. of Embryos Treated
No. of Embryos Alive* (%)’
Final Somite Number3
Surface Area4 of Yolk Sac3 (mm’)
39 48
39 (100) 48 (100)
31.2 ZL0.9 32.6 k 1.1
51.7 2 5.5
‘02 K’ 0 PI
30.2 + 0.9 27.2 2 1.7* 26.8 2 2.0’
37.7 rt 9.6 31.7 k 7.6’ 26.2 k 5.7*
33.0 2 1.5 32.0 k 1.5
46.9 k 6.1 48.7 + 6.7
13
200 300
14
10
53.8 -c 3.9
Experimental, after heat treatment at 80°C for 6 hours 10
200 300
15
‘Number in parentheses represents percentage. 2Embryos with strong heartbeat and active yolk sac circulation. 3Values are expressed as mean ? S.E. 9olk sac is assumed to be spherical. ‘p < 0.05 significantly different from the in viva control values tp < 0.05 significantly different from the in vivo control values
by Student’s t-test. by Chi-square test.
vitelline vessels. Numerous long microvilli with blunt tip were found extending from the apical surface of the endodermal cells. At the base of the microvilli immediately beneath the cell surface, pinocytotic invaginations of the plasma membrane were observed, indicating that active pinocytosis was taking place. In the subapical regions between endoder-
ma1 cells, junctional
complexes
with electron-dense
1. Cross-sections of (A) day 9 embryo from the in vitro control group and (B) day 9 experimental embryo treated with 100 kg/ml TWE. The cross-sections were taken at the hindbrain level and the embryos were cultured for 24 hours. Note that in the experimental embryo, more dead cells (arrowheads), as compared with the control embryo, are found in the neural tube (N), branchial arch (AR), pharyngeal epithelium (P) and all other regions of the embryo. H and E staining, bar = 80 p.m. Figure
deposits were found. Organelles such as mitochondria, rough endoplasmic reticulum, clusters of free ribosomes and large storage vesicles were observed within the endodermal cells (Fig. 4A). The visceral yolk sac of the embryos treated with an embryotoxic dose of the TWE appeared opaque under a dissecting microscope. The endodermal layer was deranged and large intracellular vacuoles were found in the subapical area (Fig. 3B). The vitelline vessels of the mesodermal layer were congested with blood cells. Ultrastructural examination of the endodermal cells revealed that there were very few number of the microvilli on the apical surface (Fig. 4B). The pinocytotic invaginations of the plasma membrane were seldom observed, suggesting that pinocytototic activity was suppressed in the TWE-treated yolk sac. The number of large storage vesicles was dramatically decreased and large vacant vacuoles with occasional appearance debris inside were formed. Fragof membrane-like mentation of nucleus was observed in some of the endodermal cells. The ultrastructure of organelles was also affected by the TWE treatment as gross swelling of mitochondria and rough endoplasmic reticulum was frequently found within the endodermal cells (Fig. 4B).
Discussion The results of the present study showed that an aqueous extract of T. wilfordii Hook f. which was obtained according to the traditional method of preparation was embryotoxic or even lethal at high doses to mouse embryos during organogenesis. The results,
T. wilfordii Extract and Mouse
Contraception 1995:51:65-71
Embryonic
Development
69
Table 2. Continued Abnormality Axial Length3 (mm)
16.7 15.6 2 1.8 1.4 11.8 1.9* 13.4 2+ 2.3 10.7 * 1.5’
No. of Abnormal Embryos (%)
(3) 01 (0’ 14 3 (100)t (23)t 10 (looIt
Absence of Hindlimb Bud
0 (01 (0’ (14)t 02 (0’ 3 (30)t
16.9 f 1.4 16.6 ? 1.3
therefore, caution against the use of Tripterygium wilfordii as a medicinal herb during pregnancy. The abnormalities observed were widespread: the overall growth of the embryo was retarded; cell death was induced throughout the body; body axis was twisted; somites number was reduced; and abnormalities were found in the developing eyes, ears, head, neck, trunk and limbs. These adverse effects on embryonic development appeared to be attributed to heat-labile component(s) in the T. wilfordii extract. These component(s) may be chemically different from the glycosides which are responsible for the suppressive effects on the immune system’ and the anti-fertility activity on spermatogenesis4T6 since glycosides should be heat-stable.” The mechanism of embryotoxic action of T. wl’lfordii extract and the chemical nature of the embryotoxic factor(s) await elucidation. Morphological changes were also induced in the exvisceral yolk sac epithelium by the T. wilfordii tract: epithelium was deranged; pinocytotic activity was reduced; microvilli were decreased in number; nuclei were fragmented; and storage vesicles were reduced in number with a concomitant increase in the number of large vacant vacuoles. During organogenesis, all embryonic nutrient uptake takes place mainly in the visceral yolk sac epithelium. The yolk sac epithelium, besides acting as a maternalembryonic barrier, is the principal site for the processing of embryonic nutrients and maintaining histiotrophic nutritional function.” It is believed that the absorptive functiqns are located on the apical surface, whereas secretiion of nutrients after processing takes place via the basal surface. The interruption of normal histiotrophicl status results in teratogenesis, as has been shown iwith trypan b1ue,21,22 momorcharins,14 ethanol23 and anti-yolk sac sera.24 The ultrastructural changes and the decrease in yolk sac cir-
Hematoma in Cranial Region
i/ii 14 0 (100)t (0’ 10 (100)t 0 (0’ 0 (01
(%)’
Microcephaly
Abnormal Branchial Apparatus
(31 01 (0’
:/ii
80 (57)t (0’ 6 (60)t
: vt 8 (80)t
0 (OJ 0 (01
culation after treatment of the embryo with the T. wilfordii extract suggests that the normal functions of the yolk sac may be disrupted. The decrease in pinocytosis also indicates that absorptive activity may be suppressed. The malfunctioning of the yolk sac epithelium could be one of the factors inducing abnormal development of the embryo.
Acknowledgments The authors thank I. L. Lu from the Department of Epidemiology, Schdol of Public Health, Shanghai
Figure 2. Cross-sections of (A) day 10 embryo from the in vitro control group and (B) day 10 experimental embryo treated with 300 Fg/ml TWE. The cross-sections were taken at the level of somites 9 to 10, and the embryos were cultured for 24 hours. The forelimb (L) of the experimental embryo is shorter than that of the control. Also note that the neural epithelium (N) of the experimental embryo becomes disorganised and a lot more dead cells (arrowheads) are present throughout the section. H and E staining, bar = 100 pm.
70
Chan and Ng
Contraception 1995:51:65-71
Figure 3. Cross-sections of the visceral yolk sac of (A) day 9 embryo from the in vitro control group; (B) day 9 experimental embryo treated with 100 +g/ml TWEj and (C) day 10 experimental embryo treated with 300 kg/ml TWE. All of the embryos were cultured for 24 hours. Note that in the yolk sac of abnormal embryos (B) and (C), the endodermal and the mesodermal layers are deranged. The large storage vesicles in the experimental yolk sac are less frequently found while vacant vacuoles (arrowheads) appear in the subapical areas of the endodermal layers. Toluidine blue staining, bar = 20 km.
Figure 4. Transmission electron micrographs of (A) normal visceral yolk sac endodermal epithelium of day 10 embryo from the in vitro control group and (B) visceral yolk sac endodermal epithelium of day 10 embryo treated with 300 kg/ml TWE. In the control (A), a lot of microvilli on the apical surface, pinocytotic invaginations of the apical plasma membrane (arrows), junction complexes (arrowhead) between cells, large storage vesicles (l ), rough endoplasmic reticulum, free ribosomes and normal nuclei are found, while in the treated epithelium (B), fewer microvilli are found on the apical surface, the membranous invaginations are seldomly seen, large storage vesicles disappear, mitochondria and rough endoplasmic reticulum are swollen, and large vacuoles with cell debris (V) are present. Bar = 2 km.
Medical University, People’s Republic of China, for his excellent technical assistance and for his provision of information relating to the preparation of the TW extract. The expert secretarial assistance of Ms. Doris Yeung is also gratefully acknowledged.
References 1. Anonymous. Encyclopedia of Chinese Medicines. Hong Kong and Shanghai: Commercial Press. Jiang Su New Medical School, ed; ~01 2, 1978,2460-70. ‘2. Jia L. The chemistry, pharmacology and clinical application of Tripterigium wilfordii Hook f. Clin Pharmac Bull 1985;20: 101-5. 3. Ding JH, Tao JZ, Ji L, Qian SZ. Effects of Teripterygium wilfordii and compounds isolated from it on epididyma1 function. In: Wong PYD and Cooper TG, eds. Proceeding of the Symposium on Epididymis and Male Fertility. Hong Kong: Springer-Verlag, 1992:3.10. 4. Feng J, Yu N, Xu S, Qian SZ, Wang Y. Effect of T4 from Tripterygium wilfordii on the morphology of testis and epididymis in rats. In: Wong PYD and Cooper TG, eds.
5. 6.
7.
8.
9.
10.
Proceeding of the Symposium on Epididymis and Male Fertility. Hong Kong: Springer-Verlag, 1992:3.9. Qian SZ, Zhong CQ, Xu Y. Effect of Teripterigium wilfordii Hook. F. on the fertility of rats. Contraception 1986j33: 105-10. Lan ZJ, Gu ZP, Lu RF, Zhuang LZ. Effects of multiglycosides of Tripterygium wilfordii (GTW) on rat fertility and Leydig and Sertoli cells. Contraception 1992345: 249-6 1. Tao X, Davis LS, Lipsky PE. Effect of an extract of the Chinese herbal remedy Tripterygium wilfordii Hook F on human immune responsiveness. Arthritis Rheum 1991;34:1274-81. Zhang XY, Tsuchiya N, Dohi M, Yamamoto K, Ishihara K, Okudaira H, Ito K, Miyamoto T. Prolonged survival of MRL-lpr/lpr mice treated with Tripterygium wilfordii Hook F. Clin Immunol Immunopathol 1992362366 71. Shi YM, Wu QZ. Long-term effect of Schonlein-Henoch nephritis with nephritic-nephrotic syndrome in children by traditional Chinese medicine and Western medicine. Chung Kuo Chung Hsi I Chieh Ho Hsa Chih 1992; 12:340-2. Su D, Song Y, Li R. Comparative clinical study of rheu-
Contraception 1995;51:65-71
11.
12. 13.
14.
15.
16.
matoid arthritis treated by triptolide and an ethyl acetate extract of Tripterygium wilfordii. Chung Hsi I Chieh HO Tsa Chih 1990310: 144-6. Tao XL, Sun Y, Dong Y, Xiao YL, Hu DW, Shi YP, Zhu QL, Dai H, Zhang NZ. A prospective, controlled, double-blind, cross-over study of Tripterygium wilfordii Hook F. in treatment of rheumatoid arthritis. Chin Med J (Eng] 1989; 102:327-32. Xu JY, Yang J, Li LZ. Antitumor effect of Tripterygium wilfordii. Chung Kuo Chung Hsi I Chieh Ho Tsa Chih 1992;12:161-4. Chan WY, Ng TB, Wu PJ, Yeung HW. Developmental toxicity and teratogenicity of trichosanthin, a ribosome-inactivating protein, in mice. Teratogen Carcinogen Mut 1993313:47-57. Chan WY, Tam PPL, Choi HL, Ng TB, Yeung HW. Effects of momorcharins on the mouse embryo at the early organogenesis stage. Contraception 1986j34: 537-44. Ng TB, Chan WY, Yeung HW. Proteins with abortifacient, ribosome inactivating, immunomodulatory, antitumor and anti-AIDS activities from Cucurbitaceae plants. Gen Pharmacol 1992323:575-90. Sadler TW. Culture of early somite mouse embryos during organogenesis. J Embryo1 Exp Morph 1979j49: 17-25.
T. wilfordii Extract and Mouse Embryonic Development
71
17. Whittingham DG, Wales RG. Storage of 2-cell mouse embryo in vitro. Austr J Biol Sci 1969j22:1065-8. 18. Kamovsky MJ. A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J Cell Biol 1965327: 135. 19. Ng TB, Wong CM, Li WW, Yeung HW. A steryl glycoside fraction with hemolytic activity from tubers of Momordica cochinchinensis (Family Cucurbitaceae). J Ethnopharm 1986;18:55-61. 20. Beck F, Lloyd JB, Griffiths A. A histochemical and biochemical study of some aspects of placental function in the rat using maternal injection of horseradish peroxidase. J Anatomy 1967;103:461-78. 21. Rogers JM, Daston GP, Ebron MT, Carver B, Stefanadis JG, Grabowski CT. Studies on the mechanism of trypan blue teratogenicity in the rat developing in vivo and in vitro. Teratology 1985;31:389-99. 22. Williams KE, Roberts G, Kidston EM, Beck F, Lloyd JB. Inhibition of pinocytosis in rat yolk sac by trypan blue. Teratology 1976; 14:343-54. 23. Steventon GB, Williams KE. Ethanol-induced inhibition of pinocytosis and proteolysis in rat yolk sac in vitro. Development 1987;99:247-53. 24. Freeman SJ, Brown NA. An in vitro study of teratogenicity in the rat due to antibody-induced yolk sac dysfunction. Identification of the yolk sac antigen involved. Roux’s Arch Dev Biol 1986;195:236-42.
KEY WORDS:
aqueous extract, em-
Introduction ripterygium wilfordii Hook f. (TW) is a vinelike plant that grows mainly in southern China. It has been used as a medicinal herb for hundreds of years in traditional Chinese medicine in the treatment of rheumatoid arthritis, chronic nephritis,
T
Departments of ‘Anatomy and %iochemistry. Faculty of Medicine, The Chinese University of Hong Kong, Shagin, N.T., Hong Kong Submitted for publication J ne 16, 1994 Accepted for publication S tember 14, 1994 Correspondence: Dr. W.Y. han, Department of Anatomy, Faculty of Medicine, The Chinese University of$ Hong Kong, Shatin, N.T., Hong Kong, TEL: (852) 2609 6895; FAX: (852) 2603 5031.
Q 1995 Elsevier Science Inc. 655 Avenue of the Americas, New York,
NY 10010
tuberculosis and other pulmonary diseases without serious side effects.’ However, long-term administration of TW leads to a decrease in testicular weight or menstrual disturbance in some patients.2 Evidence showed that extracts of TW possess anti-fertility activity in male rats.3-5 The viability of cultured rat Leydig and Sertoli cells was reduced by compounds purified from the extract of TW.6 In rats, the motility of spermatozoa was lost and fertility was reduced after treatment with TW in vivo.6 Evidence also showed that TW was able to exert a powerful suppressive effect on human immune response by inhibiting antigen- and mitogen-stimulated proliferation of T-cells and B-cells in vitro.’ Pharmacological studies using animals suggested TW might be potentially useful for diseases such as systemic lupus erythematosus or nephrotic syndrome.8 Recently, TW has been employed in some Chinese clinics to treat cancer, rheumatoid arthritis and nephritis.‘-l2 In view of the clinical efficacy of TW and the potentiality of TW being developed into a new drug, it is of paramount importance to investigate if TW is embryotoxic or teratogenic. It has been demonstrated that some plant proteins designated as ribosome inactivating proteins manifest anti-fertility, immunomodulatory and adverse effects on the female reproductive system, 13-i5 and these proteins also possess embryotoxic activity. 13,i4 In the present study, an aqueous extract of TW was obtained according to the traditional method of preparation for use by patients. An in vitro whole embryo culture system using mouse as the animal model was employed for the examination of the effect of the extract on embryonic development.
Materials and Methods Preparation
of the aqueous
extract
of
Tripterygium wilfordii Dried roots of T. wilfordii were bought from a herbal store and were authenticated according to the morphological criteria set in the Encyclopedia of Chinese Medicines.’ The outer two layers of the roots were peeled off and the core containing vascular tissue was extracted with distilled water as described in the ISSN OOlO-7824/95/$9.50 SSDI 0010-7824(94)00001-D
66
Chan and Ng
Table
1. Effect
Contraception 1995;51:65-71
of Tripterygium
wilfordii
extract on development
of day-9 mouse embryos in vitro before and after
heat treatment Day 9 embryos (embryos with 6-S somites)
Concentration (w/ml)
Final somite number3
Surface area4 of yolk sac3 (mm’)
Axial length3 (mm)
40 (100) 42( 100)
22.5 k 1.2 24.2 * 1.6
16.7 k 1.9 17.9 k 4.3
8.1 of: 1.0 8.3 zt 1.2
10 13 26
2 (20)t 0 (W 0 (W
19.4 -c 1.7 14.9 IL 1.8’ -5
11.9 + 2.1 8.1 2 2.1’ 5.5 * 1.3’
6.6 k 1.0 4.5 2 1.0* 3.6 * 0.7’
13 11
13 (100) 10 1911
21.9 k 1.4 23.5 k 1.2
11.7 k 2.5 16.9 * 3.0
6.7 ? 1.1 8.7 * 1.2
No. of embryos treated
No. of embryos alive’ (%)2
40 42
Control In vivo In vitro Experimental, before heat treatment 50 100 200
Experimental, after heat treatment at 80°C for 6 hours 200 300 ‘Embryos with strong heartbeat and active yolk sac circulation. 2Number in parentheses represents percentage. %alues are expressed as mean c S.E. 4Yolk sac is assumed to be spherical. %omites are not discernible in all the embryos examined. ‘p < 0.05, significantly different from the in viva control values tp < 0.05, significantly different from the in viva control values
by Student’s t-test. by Chi-square test.
Encyclopedia of Chinese Medicines’ with modifications. After one hour of extraction in water, the extract was lyophilized and reconstituted with phosphate-buffered saline (pH 7.2) before use. Heattreatment of the reconstituted extract was carried out in a glass vessel at 80°C for 6 hours. In vitro Study Randomly bred ICR female mice were paired with a male mouse overnight. The presence of vaginal plugs the following morning was designated as day 1 of pregnancy. Embryos were explanted and cultured using a procedure modified from Sadler’s whole embryo culture method.16 Embryos with 6-8 and 20-22 somites were, respectively, collected from day 9 and day 10 pregnant mice and dissected out of the decidua in PBl medium.” The visceral yolk sac and ectoplacental cone were left intact after removal of Reichert’s membrane. Culture of embryos in groups of five was carried out in 5 ml culture medium containing 2.5 ml Dubecco’s modified Eagle’s medium (DMEM, Gibco) and 2.5 ml heat-inactivated (56°C for 30 min) rat serum which had been centrifuged immediately after collection. The culture medium had been sterilized by filtration through Millipore filters (0.22 pm pore size) and equilibrated overnight with 5% COZ in air. The aqueous extract of T. wilfordii in a volume of less than 20 ~1 of phosphate-buffered saline was added to the culture medium contained in a 50ml serum bottle (Wheaton). The cultures were maintained on a roller (20-40 rpm, BTC Engineering) at
37°C and regassed every 8 hours with 5% CO, in air for day 9 embryos and 40% 02, 5% CO, and 45% N, gas mixture for day 10 embryos. of Cultured Embryos After 24 h in culture, the yolk sac circulation and heartbeat of the embryos were examined following transfer to PBl medium. They were then dissected free of embryonic membranes and washed twice with phosphate-buffered saline. Observations were made on the gross morphology, somite number and various organ primordia. Special attention was paid to any abnormalities formed. The axial lengths of embryos were determined by planimetric computation of camera lucida drawings of the embryos on a digitizer (Grafbar GP-7, Science Accessories Corporation, USA). Embryos were fixed either in Bouin’s fixative and processed for light microscopic examination, or in half-strength Karnovsky’s fixative’s for 1.5 hours. Embryos for scanning electron microscopy were then post-fixed with 1% osmium tetroxide, and criticalpoint dried with Freon. Gold-palladium-coated specimens were examined with a Jeol JSM-35 GF microscope operated at 15 KV. The yolk sacs of the embryo were fixed for transmission electron microscopy in half-strength Karnovsky’s fixative for 1.5 hours, postfixed in 1% osmium tetroxide, dehydrated in a graded ethanol series and embedded in Spurr’s plastics. SecExamination
tions of 0.5-1.0
toluidine uranyl
km thickness
blue. Ultrathin acetate
were stained with
1%
sections were stained with
and lead citrate
and then
examined
Contraception 1995;51:65-71
Table
T. wilfordii Extract and Mouse Embryonic Development
67
1. Continued
Abnormality No. of abnormal embryos W)
Abnormal optic vesicle
Abnormal otic vesicle
Absence of forelimb bud
3 (81 5 112) 8 (80)t 13 (100)t 26 (lOO)t
with a Jeol JEM-100 KV.
0 PI
2 (5’
(%)’ Twisted body axis
Open cranial neural tube
1 (3) 0 (0’
3 (81 5 (121
0 (0’
Abnormal branchial apparatus
cl (01
0 (01
4 (31)t 23 (88)t
4 (31)t 25 (96)t
8 (80)t 13 (100)t 26 (lOO)t
ll(85)t
23 (88)t
4 (40)t 5 (38)t 26 (lOO)t
17 (65)t
0 (0) 0 PI
5 (38)t 1 191
0 (0’ 0 (0’
0 PI 0 (0’
cl PI 0 (0)
CXII microscope
operating
at 80
Results Effect on in vitro Development of Embryos During Organogenesis Twenty-four hours after explant, all day 9 control embryos with an initial somite number of 6-8 developed normally in the culture (Table 1). The extent of morphogenesis in vitro paralleled that in vivo. Both in vitro and in vivo control embryos showed a low background of abnormahty (Table 1). All morphological parameters measured including the size of the the yolk sac, axial length of the embryo and somite number are similar in these two control groups. T. tilfordii extract (TWE) at a dose of 50 kg/ml induced a dramatic increase in the number and percentage (80%) of abnormal embryos formed. All abnormal embryos exhibited a slow heartbeat and a sluggish yolk sac circulation. Abnormalities observed in the abnormal embryos included absence of forelimb buds, open cranial neural tube and abnormal branchial arches which were either fewer in number as compared with the control or shorter than normal. At 100 pg/ml, all embryos examined did not show a strong heartbeat or an active yolk sac circulation. The yolk sac became smaller than the control and was opaque. Both somite number and axial length were reduced. Abnormalities in otic and optic vesicles, forelimb bud, body axis, cranial neural tube a d branchial apparatus appeared. When the treated em t: ryos were serially sectioned and examined with a mjcroscrope, it was found that in most of the regions examined, more dead cells were found in the treated iembryo than in the control embryo (Fig. 1). At 200 Hg/ml, the size of the yolk sac and the axial length of tlz embryos were further reduced.
2 (20)t
1 (81
None of the embryos showed any heartbeat. The yolk sac became very opaque and blood cells accumulated inside blood vessels of the yolk sac. All the embryos examined were grossly abnormal. Most of the organs examined showed abnormalities and the boundaries of the somites were not identifiable. Dead cells were found everywhere in the embryo. However, when the TWE was heat-treated at 80°C for 6 hours before adding to the culture, the embryotoxic effect was considerably attenuated (Table 1 J. Heat-treated TWE at a concentration as high as 300 Fg/ml did not produce any abnormalities in cultured embryos. Day 10 control embryos with an initial somite number of 20-22 developed normally in vitro 24 hours after culture and the development of the embryos was comparable to that of embryos developed in vivo (Table 2). Culture of day 10 mouse embryos for 24 hours in the presence of TWE resulted in abnormalities only in the branchial apparatus when the dose of the extract was 100 p&ml. When the dose was raised to 200 kg/ml and 300 kg/ml, additional abnormalities including absence of forelimb bud, hematoma in cranial region and microcephaly were evident. Furthermore, there were reductions in the final somite number, axial length and yolk sac surface area (Table 2). All of the abnormal embryos exhibited slow heartbeat and the yolk sac was very opaque. In most of developing organs, more dead cells were found in the treated embryos than in the control embryo (Fig. 2). The embryotoxic effect disappeared after the extract was heat-treated at 80°C for 6 hours (Table 2). Effect on the Morphology of the Visceral Yolk Sac The visceral yolk sac of the control embryos during organogenesis consisted of a layer of endodermal cells and a layer of mesodermal tissues (Fig. 3A) containing
68
Chan and Ng
Contraception 1995;51:65-71
2. Effects of Tripterygium treatment
Table
wilfordii
extract on development of day-10 mouse embryos in vitro before and after heat
Day-10 embryos (embryos with 20-22 somites)
Concentration Wml) Control ln vivo In vitro Experimental, before heat treatment 100
No. of Embryos Treated
No. of Embryos Alive* (%)’
Final Somite Number3
Surface Area4 of Yolk Sac3 (mm’)
39 48
39 (100) 48 (100)
31.2 ZL0.9 32.6 k 1.1
51.7 2 5.5
‘02 K’ 0 PI
30.2 + 0.9 27.2 2 1.7* 26.8 2 2.0’
37.7 rt 9.6 31.7 k 7.6’ 26.2 k 5.7*
33.0 2 1.5 32.0 k 1.5
46.9 k 6.1 48.7 + 6.7
13
200 300
14
10
53.8 -c 3.9
Experimental, after heat treatment at 80°C for 6 hours 10
200 300
15
‘Number in parentheses represents percentage. 2Embryos with strong heartbeat and active yolk sac circulation. 3Values are expressed as mean ? S.E. 9olk sac is assumed to be spherical. ‘p < 0.05 significantly different from the in viva control values tp < 0.05 significantly different from the in vivo control values
by Student’s t-test. by Chi-square test.
vitelline vessels. Numerous long microvilli with blunt tip were found extending from the apical surface of the endodermal cells. At the base of the microvilli immediately beneath the cell surface, pinocytotic invaginations of the plasma membrane were observed, indicating that active pinocytosis was taking place. In the subapical regions between endoder-
ma1 cells, junctional
complexes
with electron-dense
1. Cross-sections of (A) day 9 embryo from the in vitro control group and (B) day 9 experimental embryo treated with 100 kg/ml TWE. The cross-sections were taken at the hindbrain level and the embryos were cultured for 24 hours. Note that in the experimental embryo, more dead cells (arrowheads), as compared with the control embryo, are found in the neural tube (N), branchial arch (AR), pharyngeal epithelium (P) and all other regions of the embryo. H and E staining, bar = 80 p.m. Figure
deposits were found. Organelles such as mitochondria, rough endoplasmic reticulum, clusters of free ribosomes and large storage vesicles were observed within the endodermal cells (Fig. 4A). The visceral yolk sac of the embryos treated with an embryotoxic dose of the TWE appeared opaque under a dissecting microscope. The endodermal layer was deranged and large intracellular vacuoles were found in the subapical area (Fig. 3B). The vitelline vessels of the mesodermal layer were congested with blood cells. Ultrastructural examination of the endodermal cells revealed that there were very few number of the microvilli on the apical surface (Fig. 4B). The pinocytotic invaginations of the plasma membrane were seldom observed, suggesting that pinocytototic activity was suppressed in the TWE-treated yolk sac. The number of large storage vesicles was dramatically decreased and large vacant vacuoles with occasional appearance debris inside were formed. Fragof membrane-like mentation of nucleus was observed in some of the endodermal cells. The ultrastructure of organelles was also affected by the TWE treatment as gross swelling of mitochondria and rough endoplasmic reticulum was frequently found within the endodermal cells (Fig. 4B).
Discussion The results of the present study showed that an aqueous extract of T. wilfordii Hook f. which was obtained according to the traditional method of preparation was embryotoxic or even lethal at high doses to mouse embryos during organogenesis. The results,
T. wilfordii Extract and Mouse
Contraception 1995:51:65-71
Embryonic
Development
69
Table 2. Continued Abnormality Axial Length3 (mm)
16.7 15.6 2 1.8 1.4 11.8 1.9* 13.4 2+ 2.3 10.7 * 1.5’
No. of Abnormal Embryos (%)
(3) 01 (0’ 14 3 (100)t (23)t 10 (looIt
Absence of Hindlimb Bud
0 (01 (0’ (14)t 02 (0’ 3 (30)t
16.9 f 1.4 16.6 ? 1.3
therefore, caution against the use of Tripterygium wilfordii as a medicinal herb during pregnancy. The abnormalities observed were widespread: the overall growth of the embryo was retarded; cell death was induced throughout the body; body axis was twisted; somites number was reduced; and abnormalities were found in the developing eyes, ears, head, neck, trunk and limbs. These adverse effects on embryonic development appeared to be attributed to heat-labile component(s) in the T. wilfordii extract. These component(s) may be chemically different from the glycosides which are responsible for the suppressive effects on the immune system’ and the anti-fertility activity on spermatogenesis4T6 since glycosides should be heat-stable.” The mechanism of embryotoxic action of T. wl’lfordii extract and the chemical nature of the embryotoxic factor(s) await elucidation. Morphological changes were also induced in the exvisceral yolk sac epithelium by the T. wilfordii tract: epithelium was deranged; pinocytotic activity was reduced; microvilli were decreased in number; nuclei were fragmented; and storage vesicles were reduced in number with a concomitant increase in the number of large vacant vacuoles. During organogenesis, all embryonic nutrient uptake takes place mainly in the visceral yolk sac epithelium. The yolk sac epithelium, besides acting as a maternalembryonic barrier, is the principal site for the processing of embryonic nutrients and maintaining histiotrophic nutritional function.” It is believed that the absorptive functiqns are located on the apical surface, whereas secretiion of nutrients after processing takes place via the basal surface. The interruption of normal histiotrophicl status results in teratogenesis, as has been shown iwith trypan b1ue,21,22 momorcharins,14 ethanol23 and anti-yolk sac sera.24 The ultrastructural changes and the decrease in yolk sac cir-
Hematoma in Cranial Region
i/ii 14 0 (100)t (0’ 10 (100)t 0 (0’ 0 (01
(%)’
Microcephaly
Abnormal Branchial Apparatus
(31 01 (0’
:/ii
80 (57)t (0’ 6 (60)t
: vt 8 (80)t
0 (OJ 0 (01
culation after treatment of the embryo with the T. wilfordii extract suggests that the normal functions of the yolk sac may be disrupted. The decrease in pinocytosis also indicates that absorptive activity may be suppressed. The malfunctioning of the yolk sac epithelium could be one of the factors inducing abnormal development of the embryo.
Acknowledgments The authors thank I. L. Lu from the Department of Epidemiology, Schdol of Public Health, Shanghai
Figure 2. Cross-sections of (A) day 10 embryo from the in vitro control group and (B) day 10 experimental embryo treated with 300 Fg/ml TWE. The cross-sections were taken at the level of somites 9 to 10, and the embryos were cultured for 24 hours. The forelimb (L) of the experimental embryo is shorter than that of the control. Also note that the neural epithelium (N) of the experimental embryo becomes disorganised and a lot more dead cells (arrowheads) are present throughout the section. H and E staining, bar = 100 pm.
70
Chan and Ng
Contraception 1995:51:65-71
Figure 3. Cross-sections of the visceral yolk sac of (A) day 9 embryo from the in vitro control group; (B) day 9 experimental embryo treated with 100 +g/ml TWEj and (C) day 10 experimental embryo treated with 300 kg/ml TWE. All of the embryos were cultured for 24 hours. Note that in the yolk sac of abnormal embryos (B) and (C), the endodermal and the mesodermal layers are deranged. The large storage vesicles in the experimental yolk sac are less frequently found while vacant vacuoles (arrowheads) appear in the subapical areas of the endodermal layers. Toluidine blue staining, bar = 20 km.
Figure 4. Transmission electron micrographs of (A) normal visceral yolk sac endodermal epithelium of day 10 embryo from the in vitro control group and (B) visceral yolk sac endodermal epithelium of day 10 embryo treated with 300 kg/ml TWE. In the control (A), a lot of microvilli on the apical surface, pinocytotic invaginations of the apical plasma membrane (arrows), junction complexes (arrowhead) between cells, large storage vesicles (l ), rough endoplasmic reticulum, free ribosomes and normal nuclei are found, while in the treated epithelium (B), fewer microvilli are found on the apical surface, the membranous invaginations are seldomly seen, large storage vesicles disappear, mitochondria and rough endoplasmic reticulum are swollen, and large vacuoles with cell debris (V) are present. Bar = 2 km.
Medical University, People’s Republic of China, for his excellent technical assistance and for his provision of information relating to the preparation of the TW extract. The expert secretarial assistance of Ms. Doris Yeung is also gratefully acknowledged.
References 1. Anonymous. Encyclopedia of Chinese Medicines. Hong Kong and Shanghai: Commercial Press. Jiang Su New Medical School, ed; ~01 2, 1978,2460-70. ‘2. Jia L. The chemistry, pharmacology and clinical application of Tripterigium wilfordii Hook f. Clin Pharmac Bull 1985;20: 101-5. 3. Ding JH, Tao JZ, Ji L, Qian SZ. Effects of Teripterygium wilfordii and compounds isolated from it on epididyma1 function. In: Wong PYD and Cooper TG, eds. Proceeding of the Symposium on Epididymis and Male Fertility. Hong Kong: Springer-Verlag, 1992:3.10. 4. Feng J, Yu N, Xu S, Qian SZ, Wang Y. Effect of T4 from Tripterygium wilfordii on the morphology of testis and epididymis in rats. In: Wong PYD and Cooper TG, eds.
5. 6.
7.
8.
9.
10.
Proceeding of the Symposium on Epididymis and Male Fertility. Hong Kong: Springer-Verlag, 1992:3.9. Qian SZ, Zhong CQ, Xu Y. Effect of Teripterigium wilfordii Hook. F. on the fertility of rats. Contraception 1986j33: 105-10. Lan ZJ, Gu ZP, Lu RF, Zhuang LZ. Effects of multiglycosides of Tripterygium wilfordii (GTW) on rat fertility and Leydig and Sertoli cells. Contraception 1992345: 249-6 1. Tao X, Davis LS, Lipsky PE. Effect of an extract of the Chinese herbal remedy Tripterygium wilfordii Hook F on human immune responsiveness. Arthritis Rheum 1991;34:1274-81. Zhang XY, Tsuchiya N, Dohi M, Yamamoto K, Ishihara K, Okudaira H, Ito K, Miyamoto T. Prolonged survival of MRL-lpr/lpr mice treated with Tripterygium wilfordii Hook F. Clin Immunol Immunopathol 1992362366 71. Shi YM, Wu QZ. Long-term effect of Schonlein-Henoch nephritis with nephritic-nephrotic syndrome in children by traditional Chinese medicine and Western medicine. Chung Kuo Chung Hsi I Chieh Ho Hsa Chih 1992; 12:340-2. Su D, Song Y, Li R. Comparative clinical study of rheu-
Contraception 1995;51:65-71
11.
12. 13.
14.
15.
16.
matoid arthritis treated by triptolide and an ethyl acetate extract of Tripterygium wilfordii. Chung Hsi I Chieh HO Tsa Chih 1990310: 144-6. Tao XL, Sun Y, Dong Y, Xiao YL, Hu DW, Shi YP, Zhu QL, Dai H, Zhang NZ. A prospective, controlled, double-blind, cross-over study of Tripterygium wilfordii Hook F. in treatment of rheumatoid arthritis. Chin Med J (Eng] 1989; 102:327-32. Xu JY, Yang J, Li LZ. Antitumor effect of Tripterygium wilfordii. Chung Kuo Chung Hsi I Chieh Ho Tsa Chih 1992;12:161-4. Chan WY, Ng TB, Wu PJ, Yeung HW. Developmental toxicity and teratogenicity of trichosanthin, a ribosome-inactivating protein, in mice. Teratogen Carcinogen Mut 1993313:47-57. Chan WY, Tam PPL, Choi HL, Ng TB, Yeung HW. Effects of momorcharins on the mouse embryo at the early organogenesis stage. Contraception 1986j34: 537-44. Ng TB, Chan WY, Yeung HW. Proteins with abortifacient, ribosome inactivating, immunomodulatory, antitumor and anti-AIDS activities from Cucurbitaceae plants. Gen Pharmacol 1992323:575-90. Sadler TW. Culture of early somite mouse embryos during organogenesis. J Embryo1 Exp Morph 1979j49: 17-25.
T. wilfordii Extract and Mouse Embryonic Development
71
17. Whittingham DG, Wales RG. Storage of 2-cell mouse embryo in vitro. Austr J Biol Sci 1969j22:1065-8. 18. Kamovsky MJ. A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J Cell Biol 1965327: 135. 19. Ng TB, Wong CM, Li WW, Yeung HW. A steryl glycoside fraction with hemolytic activity from tubers of Momordica cochinchinensis (Family Cucurbitaceae). J Ethnopharm 1986;18:55-61. 20. Beck F, Lloyd JB, Griffiths A. A histochemical and biochemical study of some aspects of placental function in the rat using maternal injection of horseradish peroxidase. J Anatomy 1967;103:461-78. 21. Rogers JM, Daston GP, Ebron MT, Carver B, Stefanadis JG, Grabowski CT. Studies on the mechanism of trypan blue teratogenicity in the rat developing in vivo and in vitro. Teratology 1985;31:389-99. 22. Williams KE, Roberts G, Kidston EM, Beck F, Lloyd JB. Inhibition of pinocytosis in rat yolk sac by trypan blue. Teratology 1976; 14:343-54. 23. Steventon GB, Williams KE. Ethanol-induced inhibition of pinocytosis and proteolysis in rat yolk sac in vitro. Development 1987;99:247-53. 24. Freeman SJ, Brown NA. An in vitro study of teratogenicity in the rat due to antibody-induced yolk sac dysfunction. Identification of the yolk sac antigen involved. Roux’s Arch Dev Biol 1986;195:236-42.