The Morphogenesis of P-Aminopropionitrile-induced Rib Malformation in Fetal Golden Syrian Hamsters M. J. WILEY AND M. G. JONEJA Department of Anatomy, University of Toronto, Toronto, Ontario, M5S 1A8, Canada and Department of Anatomy, Queen's Uniuersity, Kingston, Ontario K7L 3N6, Canada
ABSTRACT In order to provide information on the mechanism of @-aminopropionitrile (papn) induced teratogenesis, the pathogenesis of a fetal rib abnormality was studied a t relatively short time intervals following maternal treatment with 2,500 m g k g aqueous Papn on day 11 of gestation. Histochemical tests of ribs from papn-exposed fetuses indicated a slight decrease in the level of glycosaminoglycans but a t a time when the defect was already morphologically established. Ultrastructural observations on the chondrocytes of ribs from papn-exposed fetuses revealed alterations in mitochondrial structure indicative of a slight cytotoxic effect for the teratogen. The mitochondrial changes were transient, occurring initially a t three hours after treatment and lasting for nine hours. Alterations in the size of collagen fibres in the cartilage of the fetal rib were also observed in the offspring of Papn treated females. The mean diameter of collagen fibres in the ribs of control fetuses increased throughout the course of the study. The mean diameter of fibres in the fetuses of Bapn-exposed females failed to show any increase and was found to be significantly less than controls as early as three hours following maternal administration. The results suggested that the principal factor in the production of the fetal rib deformity was fundamentally the same as t h a t known to affect the adult; namely a defect in the extracellular maturation of collagen. Maternal administration of P-aminoproprionitrile (papn) has been shown to produce a broad spectrum of fetal abnormalities including weight loss (Herd and Orbison, '661, cleft palate (Wirtschafter and Williams, '57; Abramovich and De Voto, '67, '68; Steffek et al., '71; Pratt and King, '72; Mato et al., '75; Walker, '75) ectocardia and gastroschisis (Barrow, '71; Barrow and Steffek, '74) and bony abnormalities such as exostoses and curvatures of the long bones and spine (Ferm, '60; Barrow and Steffek, '74). The majority of studies on the mechanisms of Papn-induced teratogenesis have utilized histochemical and biochemical techniques; detailed ultrastructural studies are lacking. In addition, many of these studies have been carried out several days following maternal administration and have failed to consider events occurring soon after the initial treatment. Several authors have ascribed the effects of Papn to a n inhibition of both interand intramolecular cross linking in collagen TERATOLOCY (1978) 28: 173-186.
fibres (Pratt and King, '72; Barrow e t al., '74). Others (Clemmons and Angevine, '57; Dasler and Milliser, '57; Wirtschafter, '57; Levene et al., '66; Clemmons, '66a) have provided evidence supporting a cytotoxic effect of Papn on fetal tissues. The mucopolysaccharides of the connective tissue matrix have also been implicated as the primary target of Papn activity, however, the data are ambiguous (Pyorala et al., '57; Hall, '72a,b; Levene et al., '66). Recently, Wiley and Joneja ('76) reported that Papn produced rib malformations in 100%of the offspring of hamsters exposed to this teratogen on day 11 of gestation. In order to provide additional information on the mechanisms of papn-induced teratogenicity, the purpose of the present study was to examine the pathogenesis of the rib malformation with special attention to the earliest stages, by means of both light and electron microsCOPY. Received Feb. 22, '78. Accepted Mar. 14, '78.
173
174
M . J. WILEY A N D M. G. J O N E J A
MATERIALS AND METHODS
’
Virgin Golden Syrian Hamsters weighing 90 t 5 gm were mated from 10 to 12 P.M. and the following day was designated as day 0 of gestation. Mated animals were kept singly in wire cages a t ambient temperature in a room with controlled light and dark periods of 12 hours each. Purina Lab Chow and t a p water were provided ad libitum. Each treated animal received a single injection of 2,500 mgikg aqueous /3-aminopropionitrile fumarate (Sigma) by gavage. Controls received a single dose of distilled water. Observations for each subsequent procedure were made on nine treated and nine control samples representing three fetuses from three different litters for each timed interval. In order to study the gross morphology of the defect, term fetuses were stained with alizarin red (Dawson, ‘26) following exposure on one of days 10, 11 or 12 of gestation. To examine t h e morphogenesis of the abnormality, matched treated and control dams were killed by prolonged chloroform anesthesia at 1, 3, 6, 12, 24, 48 and 72 hours following administration of the teratogen on day 11 of gestation. Live fetuses were removed from the uterus and dissected to remove t h e distal half of ribs 6 to 8. The ribs were fixed in either 10%neutral buffered formalin plus 10%cetylpyridinium chloride. or cold 2.5X) glutaraldehyde in 0.067 m phosphate buffer. Following routine paraffin embedding, t h e following histochemical procedures were carried out simultaneously on formalin-fixed treated and control samples: toluidine blue for metachromasia, P.A.S. for neutral mucosubstances, alcian blue pH 2.5 for glycosaminoglycans, and alcian blue pH 1.0 for sulfated glycosaminoglycans. All of t h e procedures were done with diastase or hyaluronidase digested controls according to Spicer et al. (’67). For ultrastructural studies, the glutaraldehyde fixed material was post-fixed a t room temperature in 1%OSO, in veronal acetate buffer, pH 7.4. The tissue was then stained en bloc in 2% aqueous uranyl acetate for three hours, dehydrated, and embedded in Epon 812. Thin sections were stained for 20 minutes in saturated uranyl acetate in 1:1, 70% ethanol: absolute methanol followed by lead citrate (Reynolds, ’63). The sections were examined with an Hitachi HU-11-E electron microscope operating a t 75 kv. Collagen fibril diameters in the rib cartilage matrix were measured a t
magnifications of 175,000 x and mean values a t various intervals, after maternal treatment were obtained by pooling approximately 200 measurements from nine fetuses fixed a t each of the times shown in figure 6. RESULTS
Maternal treatment with Papn on days 10, 11 or 12 of gestation resulted in offspring displaying abnormal rib morphology (fig. 1). Alizarin staining revealed distortions which took the form of “s”-shaped deviations from the normal rib structure. The abnormality was always bilateral and ribs 6 to 8 were invariably the most severely affected. The position of t h e defect however, shifted with the time of maternal treatment. Fetuses from females treated on day 10 of gestation (fig. l b ) most often displayed t h e abnormality a t the midpoint of t h e ossified rib tissue, while fetuses from females exposed on day 11 (fig. l c ) tended to have t h e defect at the junction of t h e proximal two-thirds and distal one-third of t h e ossified tissue. The majority of fetuses exposed to papn on day 12 of gestation (fig. Id) had the defect in t h e cartilage distal to the alizarin-stained tissue.
Light microscopy The abnormality consistently appeared within the first 24 hours of maternal Papn administration. From 1 hour to 12 hours following treatment on day 11 of gestation, the rib tissue in Papn-exposed fetuses was indistinguishable from controls by light microscopy (figs. 2a,b). By 12 days of gestation, however, the cartilage rib models of Papn-treated fetuses had undergone a marked disorganization (figs. 3a,b). The defect occupied the zone of hypertrophic and degenerating chondrocytes. There was considerable thickening of t h e perichondrial connective tissue and a marked dilatation of the perichondrial capillaries (fig. 3b). This was particularly evident in t h e area of greatest distortion. By term, development of the ribs had proceeded normally, however ossification in the perichondrial connective tissue resulted in a n abnormally thickened bone collar a t these points. An examination of t h e affinity of treated and control specimens for toluidine blue, alcian blue pH 2.5 and alcian blue pH 1.0 revealed no difference prior to 24 hours following maternal treatment. At t h a t time, how’ High Oak Ranch Ltd , R R No l, Goodwood. Ontario
MORPHOGENESIS OF PAPN-INDUCED RIB MALFORMATION
ever, there was a slight but reproducible decrease in the degree of toluidine blue-induced metachromasia, and in alcianophilia both a t pH 2.5 and 1.0,in the rib matrix of the treated specimens. This slight decrease was evident again a t 48 and 72 hours following maternal administration. Electron microscopy
The ultrastructure of chondrocytes in the distal half of ribs from the papn-exposed fetuses was not strikingly different from control ribs of a comparable age. The nuclei, rough endoplasmic reticulum, Golgi and glycogen deposits appeared to be normal in both quantity and morphology. There was, however, some evidence of a transient alteration in chondrocyte mitochondrial structure following maternal administration of papn. As early as three hours following treatment, many of the mitochondria in the cells of papnexposed samples had a vesicular and often swollen appearance (fig. 4). They had fewer cristae than those of controls and these were frequently collapsed. In addition, there appeared to be less mitochondrial matrix in the rib chondrocytes of papn-exposed fetuses. The abnormal organelle structure was evident a t 6 (fig. 5) and again at 1 2 hours following treatment, however, by 12 hours the proportion of affected mitochondria was not as large as in the earlier samples. By 12 days of gestation, mitochondrial structure in rib chondrocytes of papn-exposed fetuses appeared to be comparable to controls of the same age. The ultrastructure of the cartilage matrix in both treated and control samples, consisted primarily of a meshwork of thin, unbanded fibrils embedded in a relatively homogeneous ground substance. There was no apparent difference between treated and control samples in either the number or structure of the fibrils. A close examination of fibril diameters over a 24-hour period revealed that from day 11to day 12 of gestation, the mean fibril diameter in control ribs increased from 134 A to 150 A (fig. 6). Although slight, the increase was significant (p < 0.05). The mean fibril diameter in the ribs of papn-exposed fetuses failed to show any such increase and was significantly lower than controls of the same age a t 3, 6, 12 and 24 hours following maternal treatment (fig. 6 ) . DISCUSSION
In an attempt to provide information on the
175
mechanism of papn-induced skeletal malformation, the histological and ultrastructural characteristics of fetal hamster ribs were studied a t relatively short, timed intervals following maternal administration of 2,500 mg/ kg Papn on day 11 of gestation. The rib abnormality was chosen since it occurred in 100% of fetuses exposed a t this time (Wiley and Joneja, '76). The results of the present study show that, in hamsters, the onset of the defect is relatively rapid, occurring between 12 and 24 hours of maternal treatment. The histochemical properties of the rib matrix of papn-exposed fetuses indicated a slight decrease in the level of glycosaminoglycans a t 24 hours following papn-exposure, however, because the decrease was initially observed a t a time when the defect was already well established morphologically, i t is reasonable to conclude that this was not the principal factor contributing to the formation of the abnormality. Several authors have reported decreased activity in a number of enzymes associated with glycosaminoglycan metabolism in papn-ex posed tissues (Elders et al., '73; Del Balso and Kauffman, '751, however, Rosmus et al. ('66) has suggested that such changes are a nonspecific manifestation of Papn activity. The only significant alteration in cell ultrastructure in the ribs of papn-exposed fetuses involved chondrocyte mitochondria. Similar observations of altered mitochondria have been made in several other tissues (Ham, '62; Mato and Uchiyama, '74). These observations are consistent with reports by many authors that Papn causes depressed oxygen uptake in several systems (Juva et al., '59; Clemmons, '66a,b; Elders et al., '73) and support the contention of Clemmons ('66a,b) t h a t Papn liberates cyanide in vivo. Mitochondria are among the most sensitive indicators of cell injury responding to anoxia by swelling and other abnormal structural changes (Moore et al., '56; Okada and Peachy, '57; Bryant et al., '58; Bissi et al., '60). The failure to observe any increase in mean fibril diameter in the affected tissues of the present study is in keeping with the hypothesis of Pratt and King ('72) that Papn acts to interfere with collagen cross-linking in the fetus. Olson and Low ('71) and others (Godman and Porter, '60) have reported that the diameter of the collagen fibril component of embryonic cartilage increases with age, presumably by the addition of tropocollagen from
176
M. J . WILEY AND M. G. JONEJA
the ground substance. Since Papn is known to have no effect on collagen synthesis and secretion (Ale0 e t al., '74; P r a t t and King, '72), this suggests that the decreased diameter of collagen fibrils in treated ribs is due to a failure in their extracellular maturation. The maturation of collagen requires t h e formation of both inter- and intramolecular cross links which give the fibre its tensile strength (Piez, '68). Interference with t h e cross-linking process, mediated by Papn, would be expected to result in a failure of t h e fibrils to grow by addition of new tropocollagen molecules. This has been observed both in t h e present study, and in epiphyses of young rats treated with Papn (Follis and Tousimis, '58). A weakened collagen fibril component would lead to abnormally labile connective tissue structure which may not be able to withstand normal growth pressures. The position of t h e defect along the fetal rib depends on t h e day of maternal exposure. During development of t h e rib, maturing and early hypertrophic chondrocytes a r e located further distally with each day of development as the more proximal cells give way to stages of late hypertrophy and degeneration. The position of t h e rib deformity shifts in a similar fashion depending on t h e time of teratogen insult (fig. 1). This association suggests t h a t t h e defect will occur a t t h e point where tropocollagen secretion and cross-linking are most active a t t h e time t h e Papn reaches t h e tissue. The rapidity with which Bapn-associated changes occur ( 3 hours) in fetal hamster rib is consistent with the data of several authors who have registered significant levels of Papn in fetal r a t tissue within three hours of maternal administration by gavage (Pratt and King, '72; Wilk et al., '72). Pratt and King ('72) have provided convincing evidence of a failure in tropocoliagen cross-linking in t h e fetus as a result of maternal treatment with papn. The results of t h e present study provide morphological evidence that a specific Bapn-induced fetal abnormality may be linked to altered collagen fibril morphology. The extent to which mitochondrial alterations contribute to t h e final abnormality remains to be determined. ACKNOWLEDGMENTS
The authors wish to thank Mr. H. Verstappen for his help in preparing t h e figures and Miss N. Dunsmore for secretarial assistance.
LITERATURE CITED Abramovich, A,, and F. C. DeVoto 1967 Comparcion entre fisuras palatinas producidas por hypervitaminosos "A' y latirismo. Rev. SOC.Argent. Biol., 43: 1968 Anomalous maxillofacial patterns produced by maternal lathyrism in rat fetuses. Arch. Oral. Biol., 13: 523-826. Aleo, J . J . . R. Novack and E. Levy 1974 Collagen synthesis by lathyrogen-treated 3T6 fibroblasts. Conn. Tiss. Res., 2: 91-93. Barrow, M. V . 1971 Beta-aminopropionitrile @apn)-induced ectocardia in fetal rats. Teratology.4: 227-228 (Abstract). Barrow, M. V., and A. J. Steffek 1974 Teratologic and other embryotoxic effects of 6-aminopropionitrile in r a t s Teratology, 10; 165-172. Barrow, M. V., C. F. Simpson and E. Miller 1974 Lathyrism: A review. Quart. J. Biol., 49: 101-128. Bissi, M.. A. Bernelli-Zaggera and E. Cassi 1960 Electron microscopy of rat liver cells in hypoxia. J. Path. Bact., 79: 179.183. Bryant, R. E., W. A. Thomas and R. M. ONeal 1958 An electron microscopic study of myocardial ischemia in the rat. Circ. Res., 6: 699-709. Clemmons, J. J. 1966a The effect of lathyrogenic compounds on oxygen consumption of developing chick embryos. Fed. Proc., 25: 663 (Abstract). 1966b Proline metabolism, collagen formation, and lathyrism. Fed. Proc., 25: 1010-1015. Clemmons, J. J., and D.Angevine 1957 The occurrence of multiple fractures in suckling rats injected with P-amino-propionitrile (Lathyrus factor). Am. J. Pathol.. 33: 175-187. Dasler, W., and R. V. Milliser 1957 Experimental lathyrism in mice fed diets containing sweet peas or P-aminopropioExp. Biol. Med.. 96: 171-174. nitrile. Proc. SOC. Dawson. A. B. 1926 Note on the staining of the skeleton of cleared specimens with alizarin red S. Stain Technol.. 1: 123-125. Del Balso, A. M., and F. C. Kauffman 1975 The effect of P-aminopropionitrile on acid hydrolases and selected metabolites in oro-faclal structures of fetal rats. Arch. Oral. Biol., 20: 251-255. Elders, M. J., J. D. Smith, W. G. Smith and E. R. Hughes 1973 Alterations in glycosaminoglycan metabolism in paminopropionitrile-treatedchick embryos. Biochem. J.. 136: 985-992. Ferm, V. H. 1960 Osteolathyrogenic effects on the d e ~ veloping r a t fetus. J . Embryol. Exp. Morph.. 8: 94-97. Follis, R. H., and A. J. Tousimis 1958 Experimental lathyrism in t h e rat. Nature of t h e defect in epiphyseal cartilage. Proc. SOC. Exp. Biol. Med., 98: 843-848. Godman, G. C., and K. R. Porter 1960 Chondrogenesis studied with the electron microscope. J. Biophys. Biochem. Cytol., 8: 719-760. Hall, B. K. 1972a Skeletal defects in embryonic chicks induced by administration of beta-aminopropionitrile. Teratology, 5: 81-88. 1972b Differentiation of cartilage and bone from common germinal cells. 11. Inhibition of chondrogenesis induced by P-aminopropionitrile. Calc. Tiss. Res.. 8: 276-286. Ham, K. 1962 The fine structure of rat aorta in experimental lathyrism. Austral. J. Exp. Biol. Med. Sci., 40: 353-366. Herd, K.. and J. L. Orbison 1966 Lathyrism in the fetal rat. Arch. Path., 81: 60-66. Juva. K.. T. Nikkari and E. Kulonen 1959 Respiration of
MORPHOGENESIS OF UAPN-INDUCED RIB MALFORMATION tissue slices in the presence of P-aminopropionitrile. Acta. Chem. Scand., 132: 2122 (Abstract). Levene, C. I., J. Kranzler and S. Franco-Browder 1966 Acid mucopolysaccharides of chick embryo cartilage in osteolathyrism. Biochem. J., 101: 435-440. Mato, M., and Y. Uchiyama 1974 Appearance of lamellar structures in Purkinje cells of rat cerebellum after administration of (3-aminopropionitrile. Experientia. 30: 1030-1031. Mato, M., Y. Uchiyama, E. Aikawa and G. R. Smiley 1975 Ultrastructural changes in rat palatal epithelium after /%aminopropionitrile. Teratology, 1I : 153-168. Moore, D. H., H. Ruska and W. M. Copenhaver 1956 Electron microscopic and histochemical observation of muscle degeneration after tourniquet. J . Biophys. Biochem. Cytol., 2: 755-763. Okada, S.,and D. Peachy 1957 Effect of gamma irradiation on the deoxyribonuclease. 11. Activity of isolated mitochondria. J. Biophys. Biochem. Cytol., 3: 239-248. Olson, M., and F. Low 1971 The fine structure of developing cartilage in the chick embryo. Am. J. Anat., 131: 197-215. Piez. K. A. 1968 Cross-linking of collagen and elastin. Ann. Rev. Biochem., 37: 547-569. Pratt, R. M., and C. T. King 1972 Inhibition of collagen cross-linking associated with P-aminopropionitrile-induced cleft palate in the rat. Dev. Biol.. 27: 327-328. Pyorala, K., S. Punsar, T. Seppala and K. Karlsson 1957 Mucopolysaccharides of t h e aorta and epiphyseal car-
177
tilage in lathyritic growing rat fetuses. Acta Path. Microbiol. Scand., 41: 497-510. Reynolds, E. 1963 The use of lead citrate a t high pH. as an electron-opaque stain in electron microscopy. J. Cell Biol., 17: 208.211. Rosmus, J., K. Trnavsky and Z. Deyl 1966 Editorial-Experimental lathyrism-A molecular disease. Biochem. Pharmacol., 15: 1405-1410. Spicer, S. S., R. C. Horn and J. T. Leppi 1967 Histochemistry of connective tissue mucopolysaccharides. In: The Connective Tissue. B. M. Wagner and D. E. Smith, eds. Williams and Wilkins Co., Baltimore. Steffek, A. J., A. C. Verrusio and C. A. Watkins 1971 Cleft palate in rodents after maternal treatment with various lathyrogenic agents. Teratology, 5: 33-40. Walker, B. E. 1975 The mechanism of papn-induced cleft palate in mice. Teratology, 11: 36 (Abstract). Wiley. M. J.,and M. G. Joneja 1976 The teratogenic effects of a-aminopropionitrile in hamsters. Teratology. 14: 43-52. Wilk. A. L., C. T. G. King, E. A. Horigan and A. J. Steffek 1972 Metabolism of P-aminopropionitrile and its teratogenic activity in rats. Teratology, 5: 41-48. Wirtschafter, Z. T. 1957 Acid mucopolysaccharides in t h e histopathogenesis of aortic aneurisms in the Lathyrus-fed rat. A.M.A. Archs. Path., 64: 577-584. Wirtschafter, Z. T., and D. Williams 1957 The dynamics of protein changes in the amniotic fluid of normal and abnormal rat embryos. Am. J. Obst. Gyn., 74: 1022-1028.
PLATE I EXPLANATION OF FIGURES
la-d Alizarin.stained skeletons of fetuses from females treated with 2,500 m g k g Fapn a t different times in gestation and killed on day 14. a Control fetus.
X
13
b
Fetus exposed on day 10. The defect (arrow) is located approximately halfway along the ossified portion of t h e rib. X 11.
c
Fetus exposed on day 11. The defect is located a t the junction of the proximal two-thirds and distal one-third of the ossified portion of the rib. x 12.
d Fetus exposed on day 12. The defect ossified portion of the rib. x 11.
178
IS
located in the cartilage just distal to the
MORPHOGENESIS OF IjAPN INDUCED RIB MALFORMATION M J Wiley and M G doneja
F'LA'I'E
179
1
PLATE 2 EXPLANATION OF FIGURES
2a, b
Longitudinal sections of the distal two-thirds of the seventh rib of fetuses fixed 12 hours following maternal treatment on day 11 of gestation. Toluidine blue.
a
From control fetus.
b
From fetus exposed to 2,500 m g k g papn.
X
180 X
180
3a. b Longitudinal sections of the distal two-thirds of the seventh rib of fetuses fixed 24 hours following maternal treatment on day 11 of gestation. Toluidine blue.
180
145
a
From control fetus.
b
From fetus exposed t o 2,500 m g k g papn. The perichondrial connective tissue (P) is thickened and the perichondrial capillaries are abnormally dilated (arrow). x 145.
X
MORPWOGENESIS OF OAF"-INDUCED RIB MALFORMATION M J Wiley and M. G . J o n e ~ a
PLATE 2
181
PLATE 3 EXPLANATION OF FIGURES
4a, b Chondrocytes in the ribs of fetuses fixed three hours following maternal treatment on day 11 of gestation showing appearance of mitochondria (M)at this time. a
From control fetus. x 6,000.
b From fetus exposed to 2,500 m g k g papn.
182
X
6,000
MORPHOGENESIS OF S A P N INDUCED RIB MALFORMATION M J Wiley and M G Joneja
PLATE 3
183
PLATE 4 EXPLANATION O F FIGURES
5a, b Chondrocytes in the ribs of fetuses fixed six hours following treatment on day 11 of gestation showing the appearance of the mitochondria (MI a t this time.
a From control fetus. x 8,530. b From fetus exposed to 2,500 m g k g Papn
184
X
8,800.
MORPHOGENESIS OF UAPN INDUCED RIB MALFORMATION M J Wiley and M G Joneja
PLATE 4
PLATE 5
Vehicle Control P-ominopropionitrite
I
T
f
T
Y
6
I
3
12
24
Time( hours after injection) EXPLANATION OF FIGURE
6 The mean diameter ( 2 S.D.) of collagen fibrils in the distal half of ribs of hamster fetuses, fixed a t various times following maternal treatment on day 11 of gestation. The asterisk indicates significantly lower diameters as determined by Student’s “t” test, p < 0.05.
186
ABSTRACT In order to provide information on the mechanism of @-aminopropionitrile (papn) induced teratogenesis, the pathogenesis of a fetal rib abnormality was studied a t relatively short time intervals following maternal treatment with 2,500 m g k g aqueous Papn on day 11 of gestation. Histochemical tests of ribs from papn-exposed fetuses indicated a slight decrease in the level of glycosaminoglycans but a t a time when the defect was already morphologically established. Ultrastructural observations on the chondrocytes of ribs from papn-exposed fetuses revealed alterations in mitochondrial structure indicative of a slight cytotoxic effect for the teratogen. The mitochondrial changes were transient, occurring initially a t three hours after treatment and lasting for nine hours. Alterations in the size of collagen fibres in the cartilage of the fetal rib were also observed in the offspring of Papn treated females. The mean diameter of collagen fibres in the ribs of control fetuses increased throughout the course of the study. The mean diameter of fibres in the fetuses of Bapn-exposed females failed to show any increase and was found to be significantly less than controls as early as three hours following maternal administration. The results suggested that the principal factor in the production of the fetal rib deformity was fundamentally the same as t h a t known to affect the adult; namely a defect in the extracellular maturation of collagen. Maternal administration of P-aminoproprionitrile (papn) has been shown to produce a broad spectrum of fetal abnormalities including weight loss (Herd and Orbison, '661, cleft palate (Wirtschafter and Williams, '57; Abramovich and De Voto, '67, '68; Steffek et al., '71; Pratt and King, '72; Mato et al., '75; Walker, '75) ectocardia and gastroschisis (Barrow, '71; Barrow and Steffek, '74) and bony abnormalities such as exostoses and curvatures of the long bones and spine (Ferm, '60; Barrow and Steffek, '74). The majority of studies on the mechanisms of Papn-induced teratogenesis have utilized histochemical and biochemical techniques; detailed ultrastructural studies are lacking. In addition, many of these studies have been carried out several days following maternal administration and have failed to consider events occurring soon after the initial treatment. Several authors have ascribed the effects of Papn to a n inhibition of both interand intramolecular cross linking in collagen TERATOLOCY (1978) 28: 173-186.
fibres (Pratt and King, '72; Barrow e t al., '74). Others (Clemmons and Angevine, '57; Dasler and Milliser, '57; Wirtschafter, '57; Levene et al., '66; Clemmons, '66a) have provided evidence supporting a cytotoxic effect of Papn on fetal tissues. The mucopolysaccharides of the connective tissue matrix have also been implicated as the primary target of Papn activity, however, the data are ambiguous (Pyorala et al., '57; Hall, '72a,b; Levene et al., '66). Recently, Wiley and Joneja ('76) reported that Papn produced rib malformations in 100%of the offspring of hamsters exposed to this teratogen on day 11 of gestation. In order to provide additional information on the mechanisms of papn-induced teratogenicity, the purpose of the present study was to examine the pathogenesis of the rib malformation with special attention to the earliest stages, by means of both light and electron microsCOPY. Received Feb. 22, '78. Accepted Mar. 14, '78.
173
174
M . J. WILEY A N D M. G. J O N E J A
MATERIALS AND METHODS
’
Virgin Golden Syrian Hamsters weighing 90 t 5 gm were mated from 10 to 12 P.M. and the following day was designated as day 0 of gestation. Mated animals were kept singly in wire cages a t ambient temperature in a room with controlled light and dark periods of 12 hours each. Purina Lab Chow and t a p water were provided ad libitum. Each treated animal received a single injection of 2,500 mgikg aqueous /3-aminopropionitrile fumarate (Sigma) by gavage. Controls received a single dose of distilled water. Observations for each subsequent procedure were made on nine treated and nine control samples representing three fetuses from three different litters for each timed interval. In order to study the gross morphology of the defect, term fetuses were stained with alizarin red (Dawson, ‘26) following exposure on one of days 10, 11 or 12 of gestation. To examine t h e morphogenesis of the abnormality, matched treated and control dams were killed by prolonged chloroform anesthesia at 1, 3, 6, 12, 24, 48 and 72 hours following administration of the teratogen on day 11 of gestation. Live fetuses were removed from the uterus and dissected to remove t h e distal half of ribs 6 to 8. The ribs were fixed in either 10%neutral buffered formalin plus 10%cetylpyridinium chloride. or cold 2.5X) glutaraldehyde in 0.067 m phosphate buffer. Following routine paraffin embedding, t h e following histochemical procedures were carried out simultaneously on formalin-fixed treated and control samples: toluidine blue for metachromasia, P.A.S. for neutral mucosubstances, alcian blue pH 2.5 for glycosaminoglycans, and alcian blue pH 1.0 for sulfated glycosaminoglycans. All of t h e procedures were done with diastase or hyaluronidase digested controls according to Spicer et al. (’67). For ultrastructural studies, the glutaraldehyde fixed material was post-fixed a t room temperature in 1%OSO, in veronal acetate buffer, pH 7.4. The tissue was then stained en bloc in 2% aqueous uranyl acetate for three hours, dehydrated, and embedded in Epon 812. Thin sections were stained for 20 minutes in saturated uranyl acetate in 1:1, 70% ethanol: absolute methanol followed by lead citrate (Reynolds, ’63). The sections were examined with an Hitachi HU-11-E electron microscope operating a t 75 kv. Collagen fibril diameters in the rib cartilage matrix were measured a t
magnifications of 175,000 x and mean values a t various intervals, after maternal treatment were obtained by pooling approximately 200 measurements from nine fetuses fixed a t each of the times shown in figure 6. RESULTS
Maternal treatment with Papn on days 10, 11 or 12 of gestation resulted in offspring displaying abnormal rib morphology (fig. 1). Alizarin staining revealed distortions which took the form of “s”-shaped deviations from the normal rib structure. The abnormality was always bilateral and ribs 6 to 8 were invariably the most severely affected. The position of t h e defect however, shifted with the time of maternal treatment. Fetuses from females treated on day 10 of gestation (fig. l b ) most often displayed t h e abnormality a t the midpoint of t h e ossified rib tissue, while fetuses from females exposed on day 11 (fig. l c ) tended to have t h e defect at the junction of t h e proximal two-thirds and distal one-third of t h e ossified tissue. The majority of fetuses exposed to papn on day 12 of gestation (fig. Id) had the defect in t h e cartilage distal to the alizarin-stained tissue.
Light microscopy The abnormality consistently appeared within the first 24 hours of maternal Papn administration. From 1 hour to 12 hours following treatment on day 11 of gestation, the rib tissue in Papn-exposed fetuses was indistinguishable from controls by light microscopy (figs. 2a,b). By 12 days of gestation, however, the cartilage rib models of Papn-treated fetuses had undergone a marked disorganization (figs. 3a,b). The defect occupied the zone of hypertrophic and degenerating chondrocytes. There was considerable thickening of t h e perichondrial connective tissue and a marked dilatation of the perichondrial capillaries (fig. 3b). This was particularly evident in t h e area of greatest distortion. By term, development of the ribs had proceeded normally, however ossification in the perichondrial connective tissue resulted in a n abnormally thickened bone collar a t these points. An examination of t h e affinity of treated and control specimens for toluidine blue, alcian blue pH 2.5 and alcian blue pH 1.0 revealed no difference prior to 24 hours following maternal treatment. At t h a t time, how’ High Oak Ranch Ltd , R R No l, Goodwood. Ontario
MORPHOGENESIS OF PAPN-INDUCED RIB MALFORMATION
ever, there was a slight but reproducible decrease in the degree of toluidine blue-induced metachromasia, and in alcianophilia both a t pH 2.5 and 1.0,in the rib matrix of the treated specimens. This slight decrease was evident again a t 48 and 72 hours following maternal administration. Electron microscopy
The ultrastructure of chondrocytes in the distal half of ribs from the papn-exposed fetuses was not strikingly different from control ribs of a comparable age. The nuclei, rough endoplasmic reticulum, Golgi and glycogen deposits appeared to be normal in both quantity and morphology. There was, however, some evidence of a transient alteration in chondrocyte mitochondrial structure following maternal administration of papn. As early as three hours following treatment, many of the mitochondria in the cells of papnexposed samples had a vesicular and often swollen appearance (fig. 4). They had fewer cristae than those of controls and these were frequently collapsed. In addition, there appeared to be less mitochondrial matrix in the rib chondrocytes of papn-exposed fetuses. The abnormal organelle structure was evident a t 6 (fig. 5) and again at 1 2 hours following treatment, however, by 12 hours the proportion of affected mitochondria was not as large as in the earlier samples. By 12 days of gestation, mitochondrial structure in rib chondrocytes of papn-exposed fetuses appeared to be comparable to controls of the same age. The ultrastructure of the cartilage matrix in both treated and control samples, consisted primarily of a meshwork of thin, unbanded fibrils embedded in a relatively homogeneous ground substance. There was no apparent difference between treated and control samples in either the number or structure of the fibrils. A close examination of fibril diameters over a 24-hour period revealed that from day 11to day 12 of gestation, the mean fibril diameter in control ribs increased from 134 A to 150 A (fig. 6). Although slight, the increase was significant (p < 0.05). The mean fibril diameter in the ribs of papn-exposed fetuses failed to show any such increase and was significantly lower than controls of the same age a t 3, 6, 12 and 24 hours following maternal treatment (fig. 6 ) . DISCUSSION
In an attempt to provide information on the
175
mechanism of papn-induced skeletal malformation, the histological and ultrastructural characteristics of fetal hamster ribs were studied a t relatively short, timed intervals following maternal administration of 2,500 mg/ kg Papn on day 11 of gestation. The rib abnormality was chosen since it occurred in 100% of fetuses exposed a t this time (Wiley and Joneja, '76). The results of the present study show that, in hamsters, the onset of the defect is relatively rapid, occurring between 12 and 24 hours of maternal treatment. The histochemical properties of the rib matrix of papn-exposed fetuses indicated a slight decrease in the level of glycosaminoglycans a t 24 hours following papn-exposure, however, because the decrease was initially observed a t a time when the defect was already well established morphologically, i t is reasonable to conclude that this was not the principal factor contributing to the formation of the abnormality. Several authors have reported decreased activity in a number of enzymes associated with glycosaminoglycan metabolism in papn-ex posed tissues (Elders et al., '73; Del Balso and Kauffman, '751, however, Rosmus et al. ('66) has suggested that such changes are a nonspecific manifestation of Papn activity. The only significant alteration in cell ultrastructure in the ribs of papn-exposed fetuses involved chondrocyte mitochondria. Similar observations of altered mitochondria have been made in several other tissues (Ham, '62; Mato and Uchiyama, '74). These observations are consistent with reports by many authors that Papn causes depressed oxygen uptake in several systems (Juva et al., '59; Clemmons, '66a,b; Elders et al., '73) and support the contention of Clemmons ('66a,b) t h a t Papn liberates cyanide in vivo. Mitochondria are among the most sensitive indicators of cell injury responding to anoxia by swelling and other abnormal structural changes (Moore et al., '56; Okada and Peachy, '57; Bryant et al., '58; Bissi et al., '60). The failure to observe any increase in mean fibril diameter in the affected tissues of the present study is in keeping with the hypothesis of Pratt and King ('72) that Papn acts to interfere with collagen cross-linking in the fetus. Olson and Low ('71) and others (Godman and Porter, '60) have reported that the diameter of the collagen fibril component of embryonic cartilage increases with age, presumably by the addition of tropocollagen from
176
M. J . WILEY AND M. G. JONEJA
the ground substance. Since Papn is known to have no effect on collagen synthesis and secretion (Ale0 e t al., '74; P r a t t and King, '72), this suggests that the decreased diameter of collagen fibrils in treated ribs is due to a failure in their extracellular maturation. The maturation of collagen requires t h e formation of both inter- and intramolecular cross links which give the fibre its tensile strength (Piez, '68). Interference with t h e cross-linking process, mediated by Papn, would be expected to result in a failure of t h e fibrils to grow by addition of new tropocollagen molecules. This has been observed both in t h e present study, and in epiphyses of young rats treated with Papn (Follis and Tousimis, '58). A weakened collagen fibril component would lead to abnormally labile connective tissue structure which may not be able to withstand normal growth pressures. The position of t h e defect along the fetal rib depends on t h e day of maternal exposure. During development of t h e rib, maturing and early hypertrophic chondrocytes a r e located further distally with each day of development as the more proximal cells give way to stages of late hypertrophy and degeneration. The position of t h e rib deformity shifts in a similar fashion depending on t h e time of teratogen insult (fig. 1). This association suggests t h a t t h e defect will occur a t t h e point where tropocollagen secretion and cross-linking are most active a t t h e time t h e Papn reaches t h e tissue. The rapidity with which Bapn-associated changes occur ( 3 hours) in fetal hamster rib is consistent with the data of several authors who have registered significant levels of Papn in fetal r a t tissue within three hours of maternal administration by gavage (Pratt and King, '72; Wilk et al., '72). Pratt and King ('72) have provided convincing evidence of a failure in tropocoliagen cross-linking in t h e fetus as a result of maternal treatment with papn. The results of t h e present study provide morphological evidence that a specific Bapn-induced fetal abnormality may be linked to altered collagen fibril morphology. The extent to which mitochondrial alterations contribute to t h e final abnormality remains to be determined. ACKNOWLEDGMENTS
The authors wish to thank Mr. H. Verstappen for his help in preparing t h e figures and Miss N. Dunsmore for secretarial assistance.
LITERATURE CITED Abramovich, A,, and F. C. DeVoto 1967 Comparcion entre fisuras palatinas producidas por hypervitaminosos "A' y latirismo. Rev. SOC.Argent. Biol., 43: 1968 Anomalous maxillofacial patterns produced by maternal lathyrism in rat fetuses. Arch. Oral. Biol., 13: 523-826. Aleo, J . J . . R. Novack and E. Levy 1974 Collagen synthesis by lathyrogen-treated 3T6 fibroblasts. Conn. Tiss. Res., 2: 91-93. Barrow, M. V . 1971 Beta-aminopropionitrile @apn)-induced ectocardia in fetal rats. Teratology.4: 227-228 (Abstract). Barrow, M. V., and A. J. Steffek 1974 Teratologic and other embryotoxic effects of 6-aminopropionitrile in r a t s Teratology, 10; 165-172. Barrow, M. V., C. F. Simpson and E. Miller 1974 Lathyrism: A review. Quart. J. Biol., 49: 101-128. Bissi, M.. A. Bernelli-Zaggera and E. Cassi 1960 Electron microscopy of rat liver cells in hypoxia. J. Path. Bact., 79: 179.183. Bryant, R. E., W. A. Thomas and R. M. ONeal 1958 An electron microscopic study of myocardial ischemia in the rat. Circ. Res., 6: 699-709. Clemmons, J. J. 1966a The effect of lathyrogenic compounds on oxygen consumption of developing chick embryos. Fed. Proc., 25: 663 (Abstract). 1966b Proline metabolism, collagen formation, and lathyrism. Fed. Proc., 25: 1010-1015. Clemmons, J. J., and D.Angevine 1957 The occurrence of multiple fractures in suckling rats injected with P-amino-propionitrile (Lathyrus factor). Am. J. Pathol.. 33: 175-187. Dasler, W., and R. V. Milliser 1957 Experimental lathyrism in mice fed diets containing sweet peas or P-aminopropioExp. Biol. Med.. 96: 171-174. nitrile. Proc. SOC. Dawson. A. B. 1926 Note on the staining of the skeleton of cleared specimens with alizarin red S. Stain Technol.. 1: 123-125. Del Balso, A. M., and F. C. Kauffman 1975 The effect of P-aminopropionitrile on acid hydrolases and selected metabolites in oro-faclal structures of fetal rats. Arch. Oral. Biol., 20: 251-255. Elders, M. J., J. D. Smith, W. G. Smith and E. R. Hughes 1973 Alterations in glycosaminoglycan metabolism in paminopropionitrile-treatedchick embryos. Biochem. J.. 136: 985-992. Ferm, V. H. 1960 Osteolathyrogenic effects on the d e ~ veloping r a t fetus. J . Embryol. Exp. Morph.. 8: 94-97. Follis, R. H., and A. J. Tousimis 1958 Experimental lathyrism in t h e rat. Nature of t h e defect in epiphyseal cartilage. Proc. SOC. Exp. Biol. Med., 98: 843-848. Godman, G. C., and K. R. Porter 1960 Chondrogenesis studied with the electron microscope. J. Biophys. Biochem. Cytol., 8: 719-760. Hall, B. K. 1972a Skeletal defects in embryonic chicks induced by administration of beta-aminopropionitrile. Teratology, 5: 81-88. 1972b Differentiation of cartilage and bone from common germinal cells. 11. Inhibition of chondrogenesis induced by P-aminopropionitrile. Calc. Tiss. Res.. 8: 276-286. Ham, K. 1962 The fine structure of rat aorta in experimental lathyrism. Austral. J. Exp. Biol. Med. Sci., 40: 353-366. Herd, K.. and J. L. Orbison 1966 Lathyrism in the fetal rat. Arch. Path., 81: 60-66. Juva. K.. T. Nikkari and E. Kulonen 1959 Respiration of
MORPHOGENESIS OF UAPN-INDUCED RIB MALFORMATION tissue slices in the presence of P-aminopropionitrile. Acta. Chem. Scand., 132: 2122 (Abstract). Levene, C. I., J. Kranzler and S. Franco-Browder 1966 Acid mucopolysaccharides of chick embryo cartilage in osteolathyrism. Biochem. J., 101: 435-440. Mato, M., and Y. Uchiyama 1974 Appearance of lamellar structures in Purkinje cells of rat cerebellum after administration of (3-aminopropionitrile. Experientia. 30: 1030-1031. Mato, M., Y. Uchiyama, E. Aikawa and G. R. Smiley 1975 Ultrastructural changes in rat palatal epithelium after /%aminopropionitrile. Teratology, 1I : 153-168. Moore, D. H., H. Ruska and W. M. Copenhaver 1956 Electron microscopic and histochemical observation of muscle degeneration after tourniquet. J . Biophys. Biochem. Cytol., 2: 755-763. Okada, S.,and D. Peachy 1957 Effect of gamma irradiation on the deoxyribonuclease. 11. Activity of isolated mitochondria. J. Biophys. Biochem. Cytol., 3: 239-248. Olson, M., and F. Low 1971 The fine structure of developing cartilage in the chick embryo. Am. J. Anat., 131: 197-215. Piez. K. A. 1968 Cross-linking of collagen and elastin. Ann. Rev. Biochem., 37: 547-569. Pratt, R. M., and C. T. King 1972 Inhibition of collagen cross-linking associated with P-aminopropionitrile-induced cleft palate in the rat. Dev. Biol.. 27: 327-328. Pyorala, K., S. Punsar, T. Seppala and K. Karlsson 1957 Mucopolysaccharides of t h e aorta and epiphyseal car-
177
tilage in lathyritic growing rat fetuses. Acta Path. Microbiol. Scand., 41: 497-510. Reynolds, E. 1963 The use of lead citrate a t high pH. as an electron-opaque stain in electron microscopy. J. Cell Biol., 17: 208.211. Rosmus, J., K. Trnavsky and Z. Deyl 1966 Editorial-Experimental lathyrism-A molecular disease. Biochem. Pharmacol., 15: 1405-1410. Spicer, S. S., R. C. Horn and J. T. Leppi 1967 Histochemistry of connective tissue mucopolysaccharides. In: The Connective Tissue. B. M. Wagner and D. E. Smith, eds. Williams and Wilkins Co., Baltimore. Steffek, A. J., A. C. Verrusio and C. A. Watkins 1971 Cleft palate in rodents after maternal treatment with various lathyrogenic agents. Teratology, 5: 33-40. Walker, B. E. 1975 The mechanism of papn-induced cleft palate in mice. Teratology, 11: 36 (Abstract). Wiley. M. J.,and M. G. Joneja 1976 The teratogenic effects of a-aminopropionitrile in hamsters. Teratology. 14: 43-52. Wilk. A. L., C. T. G. King, E. A. Horigan and A. J. Steffek 1972 Metabolism of P-aminopropionitrile and its teratogenic activity in rats. Teratology, 5: 41-48. Wirtschafter, Z. T. 1957 Acid mucopolysaccharides in t h e histopathogenesis of aortic aneurisms in the Lathyrus-fed rat. A.M.A. Archs. Path., 64: 577-584. Wirtschafter, Z. T., and D. Williams 1957 The dynamics of protein changes in the amniotic fluid of normal and abnormal rat embryos. Am. J. Obst. Gyn., 74: 1022-1028.
PLATE I EXPLANATION OF FIGURES
la-d Alizarin.stained skeletons of fetuses from females treated with 2,500 m g k g Fapn a t different times in gestation and killed on day 14. a Control fetus.
X
13
b
Fetus exposed on day 10. The defect (arrow) is located approximately halfway along the ossified portion of t h e rib. X 11.
c
Fetus exposed on day 11. The defect is located a t the junction of the proximal two-thirds and distal one-third of the ossified portion of the rib. x 12.
d Fetus exposed on day 12. The defect ossified portion of the rib. x 11.
178
IS
located in the cartilage just distal to the
MORPHOGENESIS OF IjAPN INDUCED RIB MALFORMATION M J Wiley and M G doneja
F'LA'I'E
179
1
PLATE 2 EXPLANATION OF FIGURES
2a, b
Longitudinal sections of the distal two-thirds of the seventh rib of fetuses fixed 12 hours following maternal treatment on day 11 of gestation. Toluidine blue.
a
From control fetus.
b
From fetus exposed to 2,500 m g k g papn.
X
180 X
180
3a. b Longitudinal sections of the distal two-thirds of the seventh rib of fetuses fixed 24 hours following maternal treatment on day 11 of gestation. Toluidine blue.
180
145
a
From control fetus.
b
From fetus exposed t o 2,500 m g k g papn. The perichondrial connective tissue (P) is thickened and the perichondrial capillaries are abnormally dilated (arrow). x 145.
X
MORPWOGENESIS OF OAF"-INDUCED RIB MALFORMATION M J Wiley and M. G . J o n e ~ a
PLATE 2
181
PLATE 3 EXPLANATION OF FIGURES
4a, b Chondrocytes in the ribs of fetuses fixed three hours following maternal treatment on day 11 of gestation showing appearance of mitochondria (M)at this time. a
From control fetus. x 6,000.
b From fetus exposed to 2,500 m g k g papn.
182
X
6,000
MORPHOGENESIS OF S A P N INDUCED RIB MALFORMATION M J Wiley and M G Joneja
PLATE 3
183
PLATE 4 EXPLANATION O F FIGURES
5a, b Chondrocytes in the ribs of fetuses fixed six hours following treatment on day 11 of gestation showing the appearance of the mitochondria (MI a t this time.
a From control fetus. x 8,530. b From fetus exposed to 2,500 m g k g Papn
184
X
8,800.
MORPHOGENESIS OF UAPN INDUCED RIB MALFORMATION M J Wiley and M G Joneja
PLATE 4
PLATE 5
Vehicle Control P-ominopropionitrite
I
T
f
T
Y
6
I
3
12
24
Time( hours after injection) EXPLANATION OF FIGURE
6 The mean diameter ( 2 S.D.) of collagen fibrils in the distal half of ribs of hamster fetuses, fixed a t various times following maternal treatment on day 11 of gestation. The asterisk indicates significantly lower diameters as determined by Student’s “t” test, p < 0.05.
186
