Morphological changes in long bone development in fetal akinesia deformation sequence: An experimental study in curarized rat fetuses

TERATOLOGY 45:213-221 (1992)

Morphological Changes in Long Bone Development in Fetal Akinesia Deformation Sequence: An Experimental Study in Curarized Rat Fetuses JOSfi I. RODRfGUEZ, JOSR PALACIOS, ANTONIO RUIZ, MIGUEL SANCHEZ, IGNACIO ALVAREZ, AND ENRIQUE DEMIGUEL Departments of Pathology (J.I.R.,J P . , M.S.) and Experimental Surgery (I.A., E.D.), Hospital La Paz, Madrid, and Department of Pathology (A.R.), Hospital Prlncipe de Asturias, Alcala de Henares, Madrid, Spain

ABSTRACT In order to investigate the transverse growth of the long bones during intrauterine development in the fetal akinesia deformation sequence (FADS), we studied curarized rat fetuses. Curarization was performed by daily subcutaneous administration of D-Tubocurarine from day 17 of gestation until term. Experimental fetuses were compared with a sham-operated control group. The total area and perimeter, the absolute and relative amount of periosteum and bone trabeculae, the major and minor axes, and the elongation factor were measured from histological cross-sections of the femoral metaphysis and diaphysis using an IBAS 1 image analysis system. Curarized rat fetuses showed growth retardation, a short umbilical cord, and multiple articular contractures, a phenotype consistent with FADS. Alterations in femoral shape and transverse growth that affected the diaphysis were noted in these fetuses. These included a decrease of total cross-section area and reduction of the absolute and relative amounts of bone trabeculae with marked thinning of the periosteum. Femoral cross-sections was rounder than controls. These results evidenced an impairment of the membranous (periosteal) ossification of long bones produced by immobilization andlor decrease of muscular strength, and support our previous clinical findings of bone hypoplasia and osteopenia in FADS. Decreased intrauterine fetal motility of intrinsic origin produces a phenotype characterized by secondary anomalies such as growth retardation, craniofacial alterations, arthrogryposis, pulmonary hypoplasia, short umbilical cord, and polyhydramnios (Moessinger, '83; Hall, '86). These anomalies, which were previously designated as Pena-Shokair syndrome (Pena and Shokeir, '841,are now termed as fetal akinesia-hypokinesia deformation sequence (FADS) (Moessinger, '83; Hall, '86). This sequence may be caused by a heterogeneous group of neuromuscular or connective tissue diseases (Hall, '86; Graham, '88; Rodriguez et al., '88a; Rodriguez and Palacios, '91). The recognition that those anomalies are produced by intrauterine immobilization are largely based on Moessinger's experimental studies on curarized rat fetuses (Moessinger, '83). Our previous clinical studies demon0 1992 WILEY-LISS, INC.

strated long bone changes in fetuses and newborns with congenital neuromuscular diseases (Rodriguez et al., '88a) and, based on radiological data, we suggested that long bone hypoplasia should be considered a major anomaly in FADS (Rodriguez et al., '88b; Rodriguez and Palacios, '91). However, since Moessinger's experiments did not evaluate skeletal development, no experimental data on osseous prenatal growth are available t o confirm our hypothesis. The aim of the present study was to determine if long bone changes occur during intrauterine development in curarized rat fetuses, the animal model of FADS proposed by Moessinger ('83).Since our previous clin-

Received March 13, 1991; accepted August 27,1991. Address reprint requests to Dr. Jose I. Rodriguez, Department of Pathology, Hospital La Paz, Paseo Castellana 261, 28046Madrid, Spain.

214

J.I. RODRfGUEZ ET AL

ical data suggest that transverse but not longitudinal long bone growth was affected in newborns with FADS, the present study has been focused on the quantitative analysis of the transverse growth of the femur in curarized rat fetuses. MATERIALS AND METHODS

Wistar rats were mated overnight and the presence of sperm in the vaginal smear was taken to indicate day “0” of gestation. On day 17 and daily thereafter until day 20, five pregnant rats were subjected to laparotomy under ether anesthesia. Both uterine horns were exposed and the fetuses randomly distributed into two groups: experimental and sham operated control groups. The experimental fetuses received daily subcutaneous injections of 10 pg D-Tubocurarine diluted in a total volume of 0.05 ml. The sham operated control fetuses received daily subcutaneous injections of 0.05 ml of saline. Subcutaneous injections in all fetuses were performed with a gauge number 29 via a transmembranous route. No escape of amniotic fluid was noted in any experiment. At term (day 2l), the fetuses were delivered abdominally. They were immediately killed by ether inhalation, weighed, examined, and photographed. The umbilical cord was measured from placental plate to the abdominal wall using a direct-reading caliper, and lungs were weighed. After fixation in 10% buffered formalin, right femora were obtained, radiographed, and measured on amplified x-rays films. After decalcification in 5% nitric acid and routine paraffin embedment, 5 pm thick histological serial cross-sections from the distal epiphysis to the mid-diaphysis were obtained and stained with Hematoxylin and Eosin. Moreover, some histological longitudinal sections (Fig. 1) in several left femora from control rat fetuses in order to select the two levels for quantitative study in the distal metaphysis and mid-diaphysis were performed. Three consecutive histological cross-sections from both selected regions (Figs. 2,3) were measured with an IBAS 1 image analyzer. This image analysis computer system is designed for data acquisition and computation of geometric characteristics by tracing the structures of images placed on a measuring tablet. The basic components of this image analysis system are a digitalized tablet, a computer, a keyboard, and a monitor.

Fig. 1. Histological longitudinal section of the femur of a control rat at term showing the selected levels of study: distal metaphysis and mid-diaphysis (arrows). H&E, original magnification: x 10.

The diaphyseal parameters measured were total area of the cross-section; periosteal area; percentage of periosteal area (a marker of the relative amount of periosteum obtained by calculating the ratio, expressed as percentage, between periosteal area and total area of the cross-section);trabecular area (the space occupied by the trabeculae themselves); percentage of trabecular area (an indicator of real bone mass obtained by calculating the ratio, expressed as percentage, between trabecular area and total area of cross-section excluding periosteal area); perimeter of total cross-section; major and minor axis of the total cross-section (considering the cross-section of the diaphysis as an elliptical-type structure and calculating its axes by the moment of inertia); and elongation factor of the total cross-

BONE DEVELOPMENT IN FETAL AKINESIA DEFORMATION SEQUENCE

215

Fig. 2. Histological cross-section appearance of the distal metaphysis of a control rat at term. Note the existence of two well-differentiated components: medular trabeculae in the center (M) corresponding to second spongiosa, and cortical trabeculae (C) arising from the periosteum. H&E.

Fig. 3. Histological cross-section appearance of the mid-diaphysis of a control rat at term. Note absence of secondary spongiosa. Bone trabeculae arise from the periosteum. H&E.

section (form factor of the elliptical structures, which is the ratio between the minor and the major axis, whereby the circle = 1). The metaphyseal parameters measured were total area of the cross-section; periosteal area; cortical area; cortical trabecular area; percentage of trabecular cortical area

(an indicator of real cortical bone mass obtained by calculating the ratio, expressed as percentage, between trabecular cortical area and cortical area); medular area; medular trabecular area; percentage of medular trabecular area (an indicator of real medular bone mass obtained by calculat-

216

J.I. RODRfGUEZ ET AL.

Fig. 4. Control (left)and curarized (right) rat at term. Paralyzed fetus is smaller and shows joint contractures affecting lower and upper limbs. The neck is kept in a flexed position.

ing the ratio, expressed as a percentage, beTABLE 1. Data obtained from curarized and control rat fetuses' tween trabecular medular area and meduControl Curare lar area); perimeter of total cross-section; n=16 n=22 P major and minor axis of the total cross-sec4.66 * 0.45 3.96 f 0.51 0.0087 tion; elongation factor of the total cross-sec- Weight (g) tion; medular perimeter; medular major and Crown-rump length (cm) 4.65 ? 0.2 4.20 t 0.25 0.0009 minor axis; and medular elongation factor. Umbilical cord In order to assess the differences, if any, length (cm) 3.81 f 0.4 2.61 ? 0.57 0.0002 between the experimental and control Lungs weight (mg) 106.1 f 17.1 89.6 t 11.0 0.0161 group, a one-way analysis of variance Femoral length (km) 3.31 ir 0.12 3.04 t 0.2 0.0068 (ANOVA) was performed. The differences 'Data are means k standard deviations. P , ANOVA test. were judged significant if there was at least a 95% certainty ( R 0 . 0 5 ) that the experimental and control group means were different. femur at the metaphyseal level showed two different and well-defined areas. The cortiRESULTS cal or outer area consisted in the periosteum External examination revealed that par- (in which both a fibrous and a osteogenic alyzed fetuses were smaller with a thinner layer could be distinguished) and by the corskin than sham-operated control fetuses. tical bone (thin trabecula in the external Moreover, they showed joint contractures zone and a continuous bony layer in the inaffecting lower and upper limbs, and necks ternal zone). The medular or inner area conwere usually kept in a flexed position (Fig. sisted in osseous trabecula of endochondral origin and hematopoietic tissue. No differ4). Body measurements and femoral lengths ences could be histologically found between are listed in Table 1. Experimental fetuses, the experimental and the control group in as opposed to sham operated control fetuses, the cortical or medular areas (Figs. 5 , 6 ) . At showed a significant reduction in body the diaphyseal level, the femur was comweight, lung weight, and umbilical cord and posed of periosteum and osseous trabecula crown-rump lengths. Bone length was also without a differentiation between cortical significantly reduced. layer and medular area. There were no subThe histological cross-section study of the stantial histological differences between the

BONE DEVELOPMENT IN FETAL AKINESIA DEFORMATION SEQUENCE

217

Fig. 5. Histological cross-section of the femur at the metaphyseal level, of a control rat fetus at term showing two different and well-defined areas. The cortical or outer area is composed of the periosteum and of the cortical bone. Periosteum consists in two differentiated layers, the outer fibrous layer and the inner osteogenic layer, which contains plump osteoblastic cells. Cortical bone is composed of thin trabecula in the external zone

and a continuous bony layer in the internal zone. The medular or inner area shows osseous trabecula of endochondral origin and hematopoietic tissue.

experimental and the control group (Figs. 7,

aphysis in the experimental fetuses were thinner, with a lower total, periosteal, and trabecular area than those of the sham operated control fetuses. Impairment of periosteal bone growth and reduction of bone mass were also demonstrated by the signif-

8).

The results of the measurements assessed with the histological cross-sections of the femoral diaphysis and metaphysis are presented in Tables 2 and 3. The femoral di-

Fig. 6. Cross-section of the metaphysis of a curarized rat fetus at term depicting a similar histological appearance that the observed in Figure 5.

218

J.I. RODRfGUEZ ET AL.

Fig. 7. Histological cross-section of the femur, a t the diaphyseal level, of a control rat fetus at term showing periosteum and bone trabecula without secondary spongiosa.

Fig. 8. Cross-section of the diaphysis of a curarized rat fetus at term depicting a similar histological appearance that the observed in Figure 7.

icant decrease of the percentage of periosteal and trabecular areas. Diaphysis of the experimental fetuses also showed alterations of their geometry. Owing to a disproportionate reduction of their axes, experimental femora were more circular (a significantly increased elongation factor) than femora from sham operated control fetuses.

The femoral metaphysis of the experimental fetuses were also significantly thinner, more circular, and had lower total areas than sham operated control fetuses. These metaphyseal changes were related to alterations in the periosteum and cortical bone. Thus, periosteal, cortical, and trabecular areas showed significant decreases in curarized fetuses. However, no significant

BONE DEVELOPMENT IN FETAL AKINESIA DEFORMATION SEQUENCE

219

TABLE 2 . Data obtained from histological cross-sections of the femoral diaphysis in curarized and control rat fetuses' Control Curare n = 22 P n = 16 Total area of cross-section (mm') 0.655 f 0.06 0.543 5 0.05 0.0014 Periosteal area (mm? 0.113 I 0.01 0.062 5 0.01 0.0001 Percentage of periosteal area (%) 17.3 f 1.4 0.0001 11.4 5 1.4 Trabecular area (mm') 0.142 f 0.01 0.106 5 0.01 0.0001 Percentage of trabecular area (%) 26.5 f 3.2 0.0046 22.1 f 2.5 Perimeter of cross-section (mm) 2.92 f 0.16 2.68 -t 0.14 0.0032 Major axis of cross-section (mm) 0.99 t 0.05 0.87 5 0.05 0.0002 Minor axis of cross-section (mm) 0.84 f 0.05 0.79 5 0.04 0.0338 Elongation factor 0.860 ? 0.04 0.909 5 0.03 0.0116 ~

~

~~~~~

'Data are means i standard deviations. P , ANOVA test.

TABLE 3. Data obtained from histological cross-sections of the femoral metaphysis in curarued and control rat fetuses' Control Curare n = 16 n=22 P Total area of cross-section (mm') 0.749 I 0.10 0.0064 0.621 f 0.08 Periosteal area (mm') 0.124 f 0.02 0.095 f 0.01 0.007 Cortical area (mm') 0.281 2 0.04 0.205 f 0.04 0.0031 Cortical trabecular area (mm') 0.079 f 0.008 0.060 f 0.009 0.0002 Percentage of cortical trabecular area (%) 28.6 f 3.7 30.0 f 4.2 NS Medular area (mm? 0.343 I 0.07 0.320 f 0.06 NS Medular trabecular area (mm') 0.050 f 0.01 0.042 t 0.01 NS Percentage of medular trabecular area (%) 14.6 f 3.2 13.4 t 3.2 NS Perimeter of total cross-section (mm) 3.21 f 0.37 2.83 I O . 1 8 0.0069 Major axis of total cross-section (mm) 1.06 t 0.08 0.003 0.95 f 0.06 Minor axis of total cross-section (mm) 0.89 f 0.06 0.82 f 0.05 0.0277 Elongation factor of total cross-section 0.844 f 0.02 0.874 2 0.03 0.0488 Medular perimeter (mm) 2.13 t 0.23 2.05 f 0.19 NS Medular major axis (mm) 0.75 I 0.08 0.72 f 0.06 NS Medular minor axis (mm) 0.56 f 0.06 0.56 I 0.05 NS Medular elongation factor 0.751 f 0.02 0.781 t 0.03 0.0343 'Data are means

&

standard deviations. P , ANOVA test. NS, not statistically significant (PP0.05).

differences were noted between either group in any of the medular parameters measured, except for the elongation factor. DISCUSSION

The influence of physical forces on skeletal development have been extensively studied in postnatal life, and it has been demonstrated that a hypokinetic activity level leads to reduction of bone mass and size (Arnaud et al., '86). However, there are few studies on the influence of mechanical forces on prenatal bone (Carter et al., '86). Appendicular avian or mammalian bone anlages will develop to a considerable extent in the absence of mechanical influences, but as growth and morphogenesis proceed their morphology becomes aberrant. Thus, long bones will develop recognizable diaphyseal and epiphyseal regions but will fail to develop a marrow cavity (Thorogood,'83). The crucial role of the neuromuscular system in the normal development of the fetal skele-

ton is stressed by the observation of long bone changes in fetuses and newborns infants with neuromuscular diseases of intrauterine onset. In these patients, thinning of the ribs and limb long bones are frequent findings along with multiple diaphyseal and metaphyseal fractures (Chassevent et al., '78; Burke et al., '86; Rodriguez et al., '88a). Moreover, there is histological evidence that growth plate fractures may occur in utero (Rodriguez et al., '88a). Our previous radiogrammetric study demonstrated that fetus and newborns with congenital neuromuscular diseases had hypoplastic (decreased diaphyseal diameter) and osteopenic (decreased cortical bone mass) long bones (Rodriguez et al., '88b). This study reproduces the animal model of FADS described by Moessinger ('83). Experimental fetuses showed growth retardation, pulmonary hypoplasia, short umbilical cord, and multiple articular contractures. Moessinger in his original paper suggested

220

J.I. RODRfGUEZ ET AL

that prolonged immobilization could result in reduced bone mass, but skeletal studies were not performed. Since intrauterine curarization produces growth retardation, the analysis of bone changes in these fetuses is difficult. It can be speculated that because experimental fetuses are smaller they would also have smaller long bones. Indeed, after taking fetal size into account, many of the values for the absolute measured parameters were similar in the two groups; however, at the diaphysis the periosteal area and the relative amount of the periosteum and trabeculae (percentages of periosteal and trabecular areas) were still significantly lower in curarized rat fetuses. These data suggest that curarization in late gestation could produce an impairment of diaphyseal periosteal ossification with reduced bone mass as a result of immobilization andlor reduction of muscular strength. Periosteum is an osteoprogenitor cell-containing bone envelope which is responsible for normal transversal bone growth. It consists of two main layers: a fibrous limiting membrane and a osteogenic layer known as a cambium layer (Wlodarski, '89) The cambium layer contains preosteoblasts and functional osteoblasts, which produce metaphyseal cortical trabeculae, and diaphyseal trabeculae in rat fetuses. It could be suggested in hypokinesia and/or muscular hypotonia that curarized rat fetuses decrease mechanical usage of developing bone and delay periosteal growth and deposition of bone trabeculae a t the periosteal surface. The suggestion that the changes observed in bone were due to a reduced mechanical usage is also supported by the existence of cross-section geometry alterations. Bone shape is determined by the interactive process between the intrinsic properties of bone tissue and extrinsic mechanical forces related to bone use (Keller and Spengler, '89). Bone cells are sensitive to strain and adjust not only bone mass but also bone shape in relation to its functional activity. Under normal mechanical conditions a diaphyseal cross-section of long bones would tend to be elliptical (Rubin, '84).Therefore, if muscular strength applied to the bone were reduced a more rounded shape would result, as has occurred in these curarized rats. Since this is a study focused on the quantitative analysis of transverse bone growth, a more detailed histological and histochemical study has not been performed. However,

the present experimental model may be useful to analyze, for example, the role that cellular and matrix fibril orientation play in shaping developing long bones. Changes a t the metaphysis were less striking than those observed a t the diaphysis. The cross-sections of the metaphysis of the experimental fetuses were less elliptical and had lower total areas than control fetuses. This reduction of the total area was due to periosteal and cortical thinning; however, no substantial differences were observed after taking fetal size into account. All medular parameters except for the elongation factor remained unchanged in experimental fetuses indicating that there were no alterations in the amount of secondary spongiosa. These results contrast with those observed in growing postnatal rats by Weinreb et al. ('89), who reported a reduction of 50% in the metaphyseal secondary spongiosa following 10 days of immobilization. Our results indicate that in the prenatal period the metaphysis is less responsive to mechanical forces than the diaphysis. If not simply due to intrauterine growth retardation, the bone shortening observed in the experimental fetuses could be related to different pathogenetic mechanisms. Bone has a basic longitudinal growth potential, which can be modulated by endocrine, nutritional, and mechanical factors (Carter et al., '87; Frost, '87). The activity of muscles influences blood flow to the bone and neighboring tissues. A hypoactivity state secondary to muscular hypotonia could reduce nutritional and hormonal supply to the epiphyseal growth plate. On the other hand, it has been suggested that periosteum, which is hypoplastic in curarized rat fetuses, plays a role in the local regulation of the epiphyseal growth plate, probably by affecting the rate of mitosis of the germinal cells (Kuijpers-Jagtman et al., '87). Further histologic and morphometric studies on the epiphysis on the long bones of curarized rat fetuses would be helpful to establish the effect of mechanical influences on longitudinal bone growth. The relative influence of immobilization and muscular hypotonia in the genesis of long bone changes still must be elucidated in the present animal model. Previous clinical studies have suggested that muscular strength is more important than movement in the regulation of fetal long bone development (Rodriguez and Palacios, '91). This

BONE DEVELOPMENT IN FETAL AKINESIA DEFORMATION SEQUENCE

22 1

human deformation. W.B. Saunders, Philadelphia, pp. suggestion is based in the observation that 102-107. long bone hypoplasia is a usual finding in Hall, J.G. (1986) Invited editorial comment: Analysis of newborns with congenital neuromuscular Pena-Sokeir phenotype. Am. J . Med. Genet., 25.99diseases (Rodriguez et al., '88a,b) but not in 117. newborns with oligohydramnios sequence, Keller, T.S.,and D.M. Spengler (1989) Regulation of bone stress and strain in the immature and mature who also had intrauterine limitation of morat femur. J . Biomechanics, 22:1115-1127. tion but normal muscular activity (Palacios KuijpersJagtman, A.M., J.C. Maltha, J.H.M. Bex, and and Rodriguez, '90). Since skeletal differJ.G. Daggers (1987) The influence of vascular and periosteal interferences on the histological structures ences between both groups of patients could of the growth plates of long bones. Anat. Anz., 164: be explained, at least in part, by the timing 245. of reduced motility, an experimental study Moessinger, A.C. (1983) Fetal akinesia deformation seis currently being conducted in order to quence: An animal model. Pediatrics, 72.857-863. evaluate skeletal development in rat fetuses Palacios, J., and J.I. Rodriguez (1990) Extrinsic fetal akinesia and skeletal development. A study in oligosubjected to oligohydramnios from day 17 of hydramnios sequence. Teratology, 42.1-5. gestation until term. Pena, S.D.J., and M.H.K. Shokeir (1974) Syndrome of ACKNOWLEDGMENTS

This study was partially supported by Fondo de Investigaciones Sanitarias de la Seguridad Social of Spain. REFERENCES Arnaud, S.B., V.S. Schneider, and E. Morey-Molton (1986) Effects of inactivity on bone and calcium metabolism. In: Inactivity: Physiological Effects. H. Sandler and J. Vernikos, eds. Academic Press, Orlando, pp. 49-76. Burke, S.W., V.P. Jameson, J.M. Roberts, C.E. Johnston, and J . Willis (1986) Birth fractures in spinal muscular atrophy. J . Pediatr. Orthop., 6t34-36. Carter, D.R., T.E. Orr, D.P. Fyhrie, and D.J. Schurman (1987) Influences of mechanical stress on prenatal and postnatal skeletal development. Clin. Orthop., 219:237-250. Chassevent, J., J . Sauvegrain,M. Besson-Leaud, and G. Kalifa (1978) Myotonic dystrophy (Steinert's disease) in the neonate. Radiology, 127.747-749. Frost, H.M. (1987) The mechanostat: A proposed mechanism of osteoporoses and the bone mass effects of mechanical and non-mechanical agents. Bone Mineral, 2t73-85. Graham, J.M. (1988) Smith's recognizable patterns of

camptodactyly, multiple ankyloses, facial anomalies, and pulmonary hypoplasia: A lethal condition. J . Pediatr., 85.373-375. Rodriguez, J.I., A. Garcia-Alix, J. Palacios, and R. Paniagua (1988a) Changes in the long bones due to immobility caused by neuromuscular diseases: A radiographic and histologic study. J. Bone Joint Surg., 70-A1052-1060. Rodriguez, J.I., J. Palacios, A. Garcia-Alix, I. Pastor, and R. Paniagua (198913) Effects of immobilization on fetal bone development: A morphometric study in newborns with congenital neuromuscular diseases with intrauterine onset. Calcif. Tissue Int., 43:335339. Rodriguez, J.I., and J. Palacios (1991) Pathogenetic mechanisms of fetal akinesia deformation sequence and oligohydramnios sequence. Am. J . Med. Genet., 40:284-289. Rubin, C.T. (1984) Skeletal strain and the functional significance of bone architecture. Calcif. Tissue Int., 36:SllS18. Thorogood, P. (1983) Morphogenesis of cartilage. In: Cartilage Development Differentiation and Growth" B.K. Hall, ed. Academic Press, New York, Vol 2, pp. 223. Weinreb, M., Rodan, G.A., and D.D. Thompson (1989) Osteopenia in the immobilized rat hind limb is associated with increased bone resoption and decreased bone formation. Bone, 10:187-194. Wlodarski, K.H. (1989) Normal and heterotopic periosteum. Clin. Orthop., 241:265-277.