American Journal of Medical Genetics 37:412-421 (1990)
Homozygous Achondroplasia: Morphologic and Biochemical Study of Cartilage Ritta Stanescu, Victor Stanescu, and Pierre Maroteaux CNRS URA.584, Hbpital des Enfants-Malades, Paris, France We have performed histochemical, immunohistochemical, electron microscopic, and biochemical studies on the upper tibia1 cartilage from a case of homozygous achondroplasia. The growth zone was narrow and disorganized. Columnization was absent except for a few areas with short rows of cells. Hypertrophy was reduced to scattered clusters of cells. The provisional calcification was patchy and primary trabeculae were thick and irregularly arranged. Islands of fibrous or fibrocartilagineous tissue were found along the growth zone. The matrix did not stain with safranin 0 and lacked metachromasia, except for pericellular rims around the hypertrophic cell clusters. Staining with antibodies against the large proteoglycan monomers and chondroitin-4-sulfatewas weakly positive. Electron microscopic examination showed that only a few cells had degenerative signs. In most areas of the matrix, proteoglycan granules were absent. Areas with dense collagen fibers were seen. In contrast to the growth zone, the cartilage of the remaining epiphyses had normal histochemical, immunohistochemical, and electron microscopic appearance. The large proteoglycan monomers had a normal composition and hydrodynamic size. '&pe I1 and XI collagen, pepsin fragments of type IX collagen, and several noncollagenous proteins extracted from cartilage had a normal electrophoretic migration. It is suggested that a mutation affecting a matrix component or a regulatory pathway present only or predominantly in the growth area of the chondroepiphysismight explain the findings.
KEY WORDS: proteoglycans, collagen
Received for publication October 13,1989;revision received February 19, 1990. Address reprint requests to Dr. Ritta Stanescu, CNRS-URA 584, Clinique Maurice Lamy, HBpital des Enfants-Malades, 149 Rue de Sevres, 75743 Paris Cedex 15, France.
0 1990 Wiley-Liss, Inc.
INTRODUCTION Achondroplasia is one of the most common osteochondrodysplasias, affecting about one in every 26,000 live-born infants [Oberklaid et al., 19791.Achondroplasia has been known for centuries, but only for the past 30 years has it been precisely separated from other types of short limb dwarfism. Despite several morphological [Stanescu et al., 1966, 1970, 1972, 197713; Ponseti, 1970, 1988; Rimoin et al., 1970;Maynard et al., 1981;Ippolito et al., 1988; Horton et al., 19881 and biochemical studies [Stanescu et al., 1972, 1976, 1977; Pedrini-Mille and Pedrini, 1971; Horton et al., 19851 of cartilage of heterozygous achondroplastic patients, the precise mechanism of disruption of bone development in this condition remains unknown. Homozygous achondroplasia is a rare and severe condition, expected theoretically to affect 25% of the offspring of mating between heterozygous achondroplasts. The condition is usually lethal in the neonatal period, but some patients have survived beyond early infancy [Pauli et al., 19831.Clinical, radiographic, and in some instances histopathological data of a few cases with homozygous achondroplasia have been reported [Hall et al., 1969;Rogovits et al., 1972; Rimoin et al., 1973;Yang et al., 1976; Kozlowski et al., 1977; Sillence et al., 1979; Horton et al., 1988; Aterman et al., 1983; Pauli et al., 1983; Mozkowitz et al., 19891. A more detailed study of the cartilage of homozygous cases might give hints concerning the pathogenetic mechanism of achondroplasia, We had the opportunity to perform histochemical, immunohistochemical, ultrastructural, and biochemical studies on the epiphyseal cartilage of a case of homozygous achondroplasia in which it was possible to obtain the tissue within a short period from death. This male newborn infant was delivered by cesarean section at a gestational age of 37 weeks. The father, aged 41, has achondroplasia. The mother, aged 41, has achondroplasia also (120 cm height). The newborn had the appearance of severe achondroplasia with a large skull, bossed forehead, depressed nasal ridge, narrow thorax, and severe micromelia. The length was 41 cm, the weight 2,600 g, and the head circumference 37 cm. Since his birth, respiratory troubles appeared and assisted ventilation was necessary. The ultrasound examination showed marked cerebral ventricular dilatation. The respiratory distress worsened despite treatment. No
Cartilage of Homozygous Achondroplasia compression of the upper spinal cord was found, but the ventricular dilatation progressed rapidly. The respiratory distress became severe and the patient died when he was 2 months old. At that age, the length was 42 cm, the weight 3,160 g, and the head circumference 42 cm. The radiologic findings were compatible with the diagnosis of homozygous achondroplasia (Fig. 1).The bones of the limbs were short and their shape had similarities to that found in thanatophoric dysplasia. The iliums were square shaped and the acetabular roofs horizontal. The height of the vertebral bodies was decreased and the skull was large with bossed frontal and occipital regions. Post-mortem examination showed no visceral malformations.
Fig. 1. Radiograph of the patient at the age of8 days. The appearance is similar to that found in heterozygous achondroplasia but more marked: severe shortening of long bones and square ilia with horizontal, flat, acetabular roofs.
413
STUDY OF CARTILAGE Methods Tissue preparation. The inferior femoral and the upper tibial epiphyses were obtained within one hour of death. Small fragments from the tibial epiphysis taken across the growth zone and from the epiphyseal cartilage proper a t a distance from the bone were fixed for light microscopy and for ultrastructural studies. The remaining cartilage was quickly frozen with solid carbon dioxide and stored a t - 40°C. Undecalcified frozen sections were cut in a cryostat a t - 25°C from blocks ofcartilage containing both the narrow growth zone and the epiphyseal cartilage proper. Groups of 30 pm-thick sections for microchemistry were alternated with sections 7 and 4 pm thick to be used for histochemical and immunohistochemical studies. The 30 pm-thick sections were freeze-dried and the cartilage was separated by microdissection from bone, perichondrium, and vascular canals as described before [Stanescu et al., 19721. The separated cartilage was used for microchemical analysis. The biochemical studies requiring larger amounts of tissue were performed on frozen 40 pm-thick sections obtained from larger blocks of cartilage separated macroscopically from bone and perichondrium. Control specimens were obtained from a term newborn with apparently normal growth, dead from acute disease (obstetrical accident). The inferior femoral and upper tibial epiphyses were obtained within 2 hours of death. Histochemical methods. The 7 pm-thick undecalcified frozen sections were stained with azure A at pH 1.75 and 7; with safranin 0, 0.5% in 0.1M sodium acetate pH 4.6, after fixation in 4% neutral formalin containing 0.1% safranin 0 and washing with 0.05% safranin 0 in PBS; and with Mallory, Van Gieson, and Von Kossa stains after 10 min fixation in 4% neutral formalin. The 2 pm-thick glycolmethacrylate sections were stained with azure 11-methylene blue and with the azure 11-methylene blue-Von Kossa method. Immunohistochemical methods. Purified polyclonal antibodies against the hyaluronic acid binding region of human cartilage large monomers and against link protein were gifts from Drs. T.E. Hardingham and M.T. Bayliss (KennedyInstitute, London, GB). The monoclonal antibody 2-B-6 against chondroitin-4-sulfate was a gift from Dr. Bruce Caterson (West Virginia University, Morgantown, USA). The tests were performed on 4 pm-thick fresh frozen sections after 30 min digestion with chondroitinase ABC, with a second antibody labeled with peroxydase. For control slides the first antibody was replaced with normal serum or omitted. Electron microscopy methods. Small fragments of tissue were prepared as already described [Stanescu et al., 19841and ultrathin sections were examined with a Zeiss 902 electron microscope. Pro teoglycan (PG) analysis Extraction. The cartilage sections were extracted as already described [Stanescu et al., 19771. The extracts contained 86% and 91% of total tissue hexuronate of the patient and control cartilage respectively. PG purification. Two procedures were used. a) Equilibrium-density centrifugation in a CsCl gradient was
414
Stanescu et al.
performed under associative conditions as described before [Stanescu et al., 1977al. The bottom third of the gradient (A1 fraction) represented 94-95% of the total hexuronic acid and was submitted to equilibrium-density gradient centrifugation under dissociative conditions as described before [Stanescu et al., 1977bl. The bottom third fraction (AID1)contained 96-97% of the hexuronic acid. b) Ion-exchange chromatography. Guanidinium chloride extracts were dialyzed against 8M urea, 0.05M lkist HCl buffer pH 6.8 and the solutions were chromatographed on DEAE-cellulose columns as described before [Stanescu et al., 1977al. Gel chromatography of PG. This was performed on Sepharose CL 2B columns under dissociative conditions as already described [Stanescu et al., 19861. In order to test aggregation of PG, Ai fractions were applied to Sepharose CL 2B columns eluted with 0.5M sodium acetate pH 6.8, and the surfaces of excluded and included peaks were measured. In other experiments the AIDl fractions were chromatographed after overnight incubation with 2% wlw hyaluronic acid (Healon, Pharmacia) at 4°C. Gel chromatography of glycosaminoglycan chains. Purified PG (AID1fraction) were treated with 1M NaBH4, 0.05M NaOH for 48 h at 45°C. After adjusting the pH, the glycosaminoglycans were precipitated with 4 vol. of ethanol, 1.2% sodium acetate and then analyzed on Sepharose CL 6B columns eluted with 4M guanidinium chloride, 0.05M sodium acetate pH 6.8. Gel electrophoresis of PG monomers on composite gels. Electrophoresis was performed in 0.7%agarose, 1.2 acrylamide rod gels [Stanescu et al., 19771and in composite submerged slab gels [Stanescu and Chaminade, 19871. Electrophoretic transfer and immunoblotting of PG monomers. These were performed as already described [Stanescu et al. 19881 by using a polyclonal antibody against the human hyaluronic acid binding region of proteoglycan core. Electrophoresis of glycoprotein cores of PG monomers. PG monomers (AIDl fraction) were digested with chondroitinase ACII lyase and keratinase in the presence of proteolysis inhibitors [Oike et al., 19821 as already described [Stanescu and Chaminade, 19871. In other experiments ovomucoid was used as a proteolysis inhibitor [Heinegard and Sommarin, 19871. The digests and digestion mixtures were analyzed by 4% SDS-PAGE as already described [Stanescu and Chaminade, 19871. Collagen analysis Collagen solubilization. Collagen was obtained from washed cartilage residues by pepsin digestion as described before [Stanescu et al., 19761. Collagen salt fractionation. The extracted collagen was submitted to an initial purification and was salt fractionated into 0.9N, 1.2M, and 2M NaCl precipitates in order to obtain fractions enriched in various collagen types [Miller and Rhodes, 19821. Gel electrophoresis of collagens. This was performed in 6% SDS polyacrylamide slab gels according to a modification of the method of Mechanic [19791. Control samples of the same fractions digested with clostridial collagenase (Worthington CLSPA, 707 uimg purified according to Peterkovsky [1982]) were run on the same gels.
Gel electrophoresis of cyunogen-bromide-derivedpeptides from type 11 collagen. The peptides were prepared from the 0.9M NaCl as described before [Stanescu et al., 19761 and analyzed by electrophoresis in 8% SDS-polyacrylamide gels. Noncollagenous proteins analysis. The noncollagenous proteins extracted with 4M guanidinium chloride were separated from proteoglycans by ion-exchange chromatography in 8M urea, reduced, and analyzed on 7% SDS-polyacrylamide gels as already described [Chaminade et al., 19791. Analytical procedures. Hexuronate was measured by the method of Bitter and Muir [1962], hydroxyproline according to Woessner [19611. Amino acids and hexosamines were determined on a Biotronics amino-acid analyzer after hydrolysis under nitrogen of samples in 6M HC1 for 24 h at 100°C or in 4M HC1 for 4 h a t 100°C for amino acids and hexosamines, respectively. RESULTS Histologic, Histochemical, and Immunohistochemical Findings (Figs. 2,3) The histologic examination showed that the growth zone of the upper tibia1 cartilage was very narrow. The cells were irregularly arranged without column formation, except for rare areas where a few cells were grouped in a row. In many areas the hypertrophy of chondrocytes was absent. In some areas clusters of irregularly arranged cells with a partial hypertrophy were observed. The primary trabeculae were rare, short, and irregularly disposed, and some of them were thick and horizontally arranged. In some areas the growth cartilage was replaced by islands of fibrous or fibrocartilagineous tissue bordered towards the metaphysis by a bone bar. The Ranvier groove was prolonged towards the center of the growth zone by a band of young fibrous tissue bordered by a narrow bone rim. The epiphyseal cartilage proper had a n apparently normal cell density and distribution for age. The cartilage vascular canals had a normal appearance. The results of the histochemical and immunohistochemical tests are summarized in Table I. The most striking change observed in the histochemical study was a n abnormal distribution of metachromasia and safranin 0 staining. These stainings had a normal appearance in the epiphyseal cartilage and decreased gradually close to the growth area. In the growth area itself there were nonmetachromatic areas and areas with patchy metachromasia. In the zones with clusters of partially hypertrophied cells, metachromatic rims were present around the cells. The distribution of stainings contrasted with the findings in newborns with normal growth where the growth area was strongly metachromatic and safranin 0 positive (control in this study and newborns studied by R. Stanescu et al., [19731).Loss of proteoglycans during tissue processing was avoided in our study by using freshfrozen sections unfixed and stained with azure A [Szirmai, 19631 and frozen sections fixed in fixatives containing safranin 0 [Shepard and Mitchell, 19761. In contrast to the growth area, the epiphyseal cartilage proper was strongly metachromatic and was stained with safranin 0. The stainings were similar to those found in the control.
Cartilage of Homozygous Achondroplasia
Fig. 2. A: Upper tibial cartilage. Fresh-frozen section, Azure A, pH 1.75.The cartilage of the epiphysis is metachromatic but metachromasia decreases towards the narrow growth area (G). The latter is not metachromatic except for several groups of hypertrophic cells which have metachromatic rims (arrows). x 20.5. B: Upper tibial cartilage. Freshfrozen section, Van Gieson staining. The narrow growth zone (GI is more densely stained than the rest of the cartilage. In the middle of the growth area there is a fibro-cartilaginous zone (arrows). To the left of the figure, the Ranvier grove is prolonged inwards (arrowhead). x 20.5. C: Upper tibial cartilage. Fresh-frozen section, Safranin 0. Lack of columnization and of staining and small cells in the growth area. D: Upper tibial cartilage control. Fresh-frozen section, Safranin 0. Columns and stained matrix in the growth area. x 65.
Electron Microscopic Findings (Figs. 4,5) Electron microscopic examination showed that both in the growth area and in the epiphyseal cartilage proper the cells had an active appearance with well-developed rough endoplasmic reticulum, Golgi systems, and many mitochondria. Degenerated cells were rare. In most areas, the hypertrophy of cells was lacking or incomplete. In the growth zone, the granules formed by proteoglycan precipi-
415
Fig. 3. A Upper tibial cartilage. Growth zone. Fresh-frozen section, Azure A, pH 1.75. The matrix is not metachromatic. The chondrocytesin the upper part of the area have metachromatic rims (arrows); in the lower part the metachromatic pericellular rims are lacking (arrowheads). ~ 2 2 0 .B Upper tibial cartilage. Growth zone. Undecalcified, glycolmethacrylate embedding, 2 km-thick section. Lack of metachromasia and of provisional calcification. Small, irregularly arranged chondrocytes. Columnization is lacking; a few cells are arranged in a row (arrows). In the lower part of the figure is a bone trabecula (TIhorizontally arranged and above it a vessel with perivascular cells. Osteoblasts with an active appearance (arrowhead). x 620.
tation were absent from the pericellular and the interterritorial matrix. The matrix contained many collagen fibers irregularly arranged, some of them thick, banded, and/or with angular trajects. The matrix of the epiphyseal cartilage proper contained proteoglycan granules and fine collagen fibers. The calcified deposits are patchy and irregularly arranged.
Biochemical Data The growth zone was too narrow to be properly separated from the epiphyseal cartilage proper. The data were obtained from whole epiphyseal cartilage.
416
Stanescu et al. TABLE I. Histochemical and Immunohistochemical Tests* Method ~~
Detects ~~~~~
Metachromasia, azure A, pH 1 7 5
Sulfated glycosaminoglycans
Metachromasia, azure A, PH 7
Glycosaminoglycans
Metachromasia, azure 11methylene blue
Glycosaminoglycans
Safranin 0
Glycosaminoglycans
Van Gieson
Collagen
Mallory trichrome
Proteins
Von Kossa
Insoluble phosphate and carbonates indicating calcium deposits
Immunohistochemical
Core of large proglycans
Immunohistochemical Immunohistochemical
Link-protein Chondroitin-4-sulfate
Epiphysis Growth area ~ _ _ _ ~~-~ - . _ Nonmetachromatic, rare areas with some patchy metachromasia around clusters of partially hypertrophied cells Similar to control Strongly diminished, except patchy areas around rare clusters of urobablv hypertrophied c d l s " Similar to control Nonmetachromatic, except patchy areas around clusters of partially hypertrophied cells SImilar to control Not stained, except areas mentioned above Similar to control More intense than in control Similar to control More intense than in control Provisional calcification: absent in some areas, reduced to small patches of mineralized deposits in other areas Similar to control Positive but less intense than in control Similar to control Similar to control Similar to control Positive but less intense than in control ~
Similar to control
All tests were performed on undecalcified, frozen sections except for azure 11-methylene blue staining, which was performed on glycolmethacrylate sections. The results were compared with the control and with the newborns studied by R. Stanescu et al. 119731.
Composition. The composition of cartilage is shown in Table 11. The galactosamine and glucosamine contents were moderately decreased compared with the control. Collagens. Gel electrophoresis of the fraction of pepsin-extracted collagen precipitated with 0.9M NaCl showed that the a band of type I1 collagen had similar migrations in achondroplasia and in the control (Fig. 6A). In both cases a faint a2 band was visible, originating very probably from type I collagen present in the vascular cartilage canals, which are very well developed in newborns [Stanescu et al., 19731. The bands corresponding to type XI collagen from the 1.2M NaCl fraction also had similar migrations (Fig. 6A). In this fraction and especially in 2M NaCl fractions, bands representing fragments of type IX collagen were shown to have similar migration in both the achondroplastic and the normal cartilage (Fig. 6B). Identification of collagen bands was made by digestion with purified clostridial collagenase (Fig. 6). Gel electrophoretic analysis of CB-derived peptides of type I1 collagen showed a similar pattern in the patient and control (Fig. 7D). Proteoglycans. The parameters of the large proteoglycan monomers extracted from epiphyseal cartilage were similar to those found in the control: electrophoretic migration (Fig. 7B), gel chromatography peaks (Fig. 7A), gel chromatography peak of glycosaminoglycan chains, and gel electrophoresis of glycoprotein cores (Fig. 7E). The composition of the AIDl fractions (galactosamine, glucosamine, amino acids) was similar (not shown) with galactosamineiglucosamine ratios of 11.5 and 11.8, respec-
tively. Electrophoresis, transfer, and electroblotting showed that both PG monomers of the achondroplastic and of the control contain epitopes of the hyaluronic acid binding region (Fig. 7 0 . The percentage of aggregates in A1 fractions was higher in the control (40%) than in the achondroplastic (27%). After incubation with hyaluronic acid the percentage of aggregates increased but was still higher in the control (69%) than in the achondroplastic (54%). Noncollageneous proteins. The proteins extracted with 4M guanidinium chloride had a similar electrophoretic control in the patient and in the control (Fig. 7F).
DISCUSSION Growth cartilage histopathology has been studied in several cases of homozygous achondroplasia. Hall et al. [19691noted a generalized absence of regular column formation with a short growth zone and Yang et al. [1976] a narrow growth zone with minimal proliferation and hypertrophy without column formation. In a more detailed histological study Aterman et al. [1983] found a severe disorder of the growth cartilage of long bones with irregularity of the epiphyseal line, lack of orderly column formation, distortion of the vasculature, and focal distribution of rather small and irregularly arranged hypertrophic cells separated by tongue-like projections of resting cartilage and fibrous or fibro-cartilaginous tissue. Horton et al. [19881 described in the iliac crest and foramen magnum cartilage a reduction in the thickness of the growth plate most marked in the proliferative zone with a "quali-
Cartilage of Homozygous Achondroplasia
Fig. 4. A Chondrocytes ofthe growth zone. Electron micrograph showing rather small cells without signs of degeneration and with well-developed organites. The bar is 1 pm. B Provisional calcification zone. Patchy calcification (arrow). Lack or incomplete hypertrophy of chondrocytes (arrowheads). The bar is 1 pm.
tatively intact" endochondral ossification process.Many of the cells had a normal appearance; others had degenerating signs with excessive aggregation of collagen fibrils near the latter. In our case the upper tibia1 growth cartilage was severely disorganized: very narrow growth area and lack of columnization except for a few areas with short rows of cells, hypertrophy reduced to a few scattered clusters of cells; patchy provisional calcification, thick and irregularly arranged primary trabeculae, and islands of fibrous or fibro-cartilaginous tissue. In many areas of the growth zone the matrix was not metachromatic and did not stain with safranin 0. However, immunohistochemical tests for core proteins of large
417
Fig. 5. A. Matrix of the epiphyseal cartilage proper. Electron micrograph showing fine collagen fibers and proteoglycan granules (arrows). The bar is 0.5 pm. B: Matrix of the growth area with thicker-banded collagen fibers and lack of proteoglycan granules. The bar is 0.5 pm. C: Pericellular matrix of a chondrocyte from the epiphyseal cartilage proper. Fine collagen fibers, numerous proteoglycan granules (arrows). GLY, glycogen. The bar is 0.5 pm. D: Pericellular matrix of a chondrocyte from the growth zone. Dense, relatively thick, banded collagen fibers without proteoglycan granules. The bar is 0.5 pm.
proteoglycan monomers and for chondroitin-4-sulfate were faintly positive. Intensity of safranin 0 staining was shown to be proportional to cartilage glycosaminoglycan content [Rosenberg, 19711. However, the immunohistochemical tests are more sensitive than the histochemical stainings and were found to be positive in osteoarthritic tissue with severe proteoglycan depletion and nondetectable safranin 0 staining [Camplejohn and Allard, 19881. These data showed that in our case, most areas of the growth zone had a very low proteoglycan concentration. These results were confirmed by the electron microscopic examination. Dense granules of precipitated proteoglycan monomers [Hascall and Hascall, 19841were absent from most areas of the growth zone. The growth zone was stained stronger than in normals with the Van Gieson technique and contained areas of dense and in some places wide or angular collagen fibers.
418
Stanescu et al. TABLE 11. Composition of Cartilage
Dry weight 92 of wet weight Hydroxyproline % of dry weight Galactosamine nMimg dry weight Glucosamine nMimg dry weight Extraction yield of hexuronic acid % of the total Extraction yield of hydroxyproline % of the total
Homozygous achondroplasia
Control
16
14
4.4
3.8
335
444
36
44
86
91
78
82
Abnormal collagen organization was observed in the growth plate of heterozygous achondroplasia IStanescu et al., 1966,1970,1972; Ponseti, 1970,1988;Maynard et al., 19811and the concentration of hydroxyproline in a microdissected area of the growth zone was found increased [Stanescu et al., 19721. The presence of several scattered groups ofhypertrophic cells with metachromatic rims and of fibrous zones along the growth zone showed that the cartilage lesions of the growth plate were not homogenous. This was already noted in heterozygous achondroplasia by Stanescu et al. [19721 and by Maynard et al. 119811. In a case of homozygous achondroplasia Aterman et al. [19831found zones of resting cartilage and of fibrous tissue interspersed between foci of hypertrophic cells. It is not known if this heterogeneity can be related to the cell heterogeneity of the proliferative chondrocytes described in some studies [Wilsman et al., 1981; Simon and Cooke, 1988; Hwang et al. 19781. In contrast to the growth area, the remaining epiphyseal cartilage had no apparent morphological or histochemical abnormalities. Several parameters of extracted proteoglycan monomers as well as the electrophoretic pattern of type I1 collagen a band and of CBderived peptides were similar to those found in the control. The electrophoretic pattern of type I1 collagen a chains and of CB peptides was found to be normal in heterozygous achondroplasia IStanescu et al., 1976; Horton et al. 19851 and linkage studies demonstrated discordant inheritance of achondroplasia and COL2Al alleles [Francomano and Pyeritz, 1988; Ogilvie et al., 1986; Wordsworth et al., 19881. Our study showed that a chains of type XI collagen and pepsin fragments from type IX collagen were present and gave a normal electrophoretic pattern in cartilage of homozygous achondroplasia. Several noncollagenous proteins extracted with guanidinium chloride including link proteins showed a normal electrophoretic pattern on 7% SDS-PAGE. Our study also showed that the abnormalities of cartilage in homozygousachondroplasia seem to be limited to the growth area. The severe alterations of the growth area are not due to cell degeneration since the electron microscopic study of our case, in which tissue fixation was performed shortly after death, showed that most chondro-
Fig. 6. A: Gel electrophoresis of pepsin-extracted cartilage collagens partially separated by salt precipitations. 1: 0.9M NaCl fraction, homozygous achondroplasia (AH). 2: 0.9M NaCl fraction, Control (N). 3 0.9M NaCl fraction, AH, digested with collagenase. 4:0.9M NaCl fraction, N, digested with collagenase. 5 1.2M NaCl fraction, AH. 6 1.2M NaCl fraction, N. 7 and 8: The same as in 5 and 6, digested with collagenase. B: Gel electrophoresis of pepsin-extracted cartilage collagens partially separated by salt precipitations. 1: 1.2M NaCl fraction, AH. 2 1.2MNaCl fraction, N. 3 and 4 the same as in 1 and 2, after collagenase digestion. 5 2M NaCl fraction, AH. 6 2M NaCl fraction, N. 7 and 8: The same as in 5 and 6, after collagenase digestion.
Cartilage of Homozygous Achondroplasia
Fig. 7. A Gel chromatography of proteoglycan monomers (AlDd of cartilage on Sepharose CL 2B in 4M guanidinium chloride, 0.5%Triton x 100.The profile of proteoglycans from homozygous achondroplasia (1) is similar to that of proteoglycans from a newborn with apparent normal growth (2). B: Gel electrophoresis of proteoglycan monomers from homozygous achondroplasia (1) and from control (2). Similar electrophoretic migration. C: Immunoblotting of proteoglycan monomers from homozygous achondroplasia and from control. Both contain the hyaluronic acid binding region. D: Gel electrophoresis of CNBr-derived peptides of type I1 collagen from homozygous achondroplasia (arrowheads) and from control (arrows). Identical pattern. E: Gel electrophoresis of core proteins of proteoglycan monomers from homozygous achondroplasia (lane 4) and from control (lane 3). Similar electrophoretic migration. Lane 2: Digestion medium; Lane 1:Calibration kit for molecular weight. F: Gel electrophoresis of noncollagenous proteins extracted from homozygous achondroplasia (lane 2) and from control (lane3).Similar pattern. Lane 1: Calibration kit for molecular weight.
cytes did not present degeneration signs and had wellpreserved cell organelles. A metabolic defect was suggested in achondroplasia. Mackler et al. [1986]found abnormal oxidative phosphorylation and decreased cytochrome levels in cultured fibroblasts from homozygouspatients, suggesting that this might impair the growth of cells at low oxygen concentrations such as found in the resting cartilage. However, oxygen tension determinations in the epiphyseal plate showed low values in the resting and hypertrophic zones and highest values in the proliferative area [Brighton and Heppenstall, 19711. In our case the resting zone had a normal appearance and the proliferative area was severely affected. On the other hand, an initial energy deficiency would probably impair the production of other matrix components. In our case collagen and link proteins
419
were well represented in the growth area. The aclac rabbit with chondrodystrophy and altered oxidative energy production [Bargman et al., 19721 is different in many respects from human achondroplasia. Hereditary defects in which the synthesis of cartilage proteoglycansis impaired have been recorded in mice and chicken chondrodysplasias[Kimata et al., 1979;Stirpe et al., 1987;Quinterer and Goetinck, 19811.In cmdkmdmice which failed to synthesize cartilage-characteristic proteoglycans, lack of columnization and wide collagen fibers were found in the growth area [Kobayakawa et al., 19851. However, these conditions have a recessive type of transmission and the proteoglycan defect is not limited to the growth area. Growth factors were isolated from cartilage and bone but the physiologic or pathologic roles of the locally produced growth factors are still poorly known [Kato et al., 1981; Klagsbrun and Smith, 1980; Nilsson et al., 1986; Rosen et al., 1988; Hamerman et al., 19861. A possible abnormality of a growth factor or of its chondrocytereceptor cannot be ruled out in achondroplasia. Our study of cartilage of homozygous achondroplasia shows that such an abnormality should produce both a decrease in the cell proliferation and hypertrophy and changes in the matrix: severe decrease of cartilage proteglycans limited to the growth area with well-represented collagen and link proteins in the same area. A role for local IGF-1or its receptor seems unlikely since the growth cartilage of achondroplasia is different from that of pituitary dwarfism [Stanescu et al., 1970, 19721. A more likely pathogenetic mechanism might be a mutation that affects precursors of growth-zone chondrocytes. This mutation might produce alterations of a cartilage matrix component secreted only in the growth area or disturb a regulatory pathway affecting predominantly this area. Severalpopulationsof chondrocytesexist within the normal chondroepiphysesduring development, devoted to articular cartilage, growth of the chondroepiphysis itself, growth of a secondary center of ossification, and longitudinal growth of the bone. Differences in the morphology and histochemistry of the cells, of the matrix, in lectin binding by cells, and in the behaviour of cells in culture were described [Hincliffe and Johnson, 1983; Glaser and Conrad, 1981; Farnum and Wilsman, 1988; Boyan et al., 19881. More recently it was shown that very early in the development there is heterogeneity of the chondroepiphyseal regions as regards cell kinetic and proteoglycan synthesis [Diao et al., 19891. Chondrocytes isolated from cartilage tissue with different developmental destinies differ qualitatively and quantitatively in total collagen synthesis, procollagen processing, and distribution of collagen types [Gerstenfeld et al., 19891. A better understanding of differences that exist between chondrocytes and their precursors in various areas of the chondroepiphysis will help to elucidate the defect in certain chondrodysplasias including achondroplasia. Another point to be emphasized is the resemblance between thanatophoric dysplasia and homozygousachondroplasia already remarked by Hall et al. [1969] and Rogovits et al. [19721 for the clinical and radiographic features and by Aterman et al. [19831for the histopathology of growth plate. Lack of metachromasia at low pH
420
Stanescu et al.
limited to the growth area was also observed in thanatophoric dysplasia [Stanescu et al., 19771.
ACKNOWLEDGMENTS We would like to thank: Drs. 0. de Bethman and J.P. Relier for communication of their observation, Drs. T. Hardingham, M.T. Bylis, and B. Caterson for providing the antibodies used in this study; Mrs. F. Chaminade, T. Do, and M.P. Muriel for technical assistance; and Miss A. L’Horisson for typing the manuscript. REFERENCES Aterman K, Welch JP, Taylor PG (1983): Presumed homozygous achondroplasia. A review and report of a further case. Path01 Res Pract 178:27-39. Bargman GL, Mackler B, Shepard TH (1972):Studies of oxidative energy deficiency I. Achondroplasia in the rabbit. Arch Biochem Biophys 150:137-146. Bitter T, Muir H (1962): A modified uronic acid carbazole reaction. Anal Biochem 4:330-334. Boyan BD, Schwartz Z, Swain CD, Carnes DL, Zishs T 11988):Differential expression of phenotype by resting zone and growth region costochondral chondrocytes in vitro. Bone 9:185-194. Brighton CT, Heppenstall RB (1971): Oxygen tension in zones of the epiphyseal plate, the metaphysis and diaphysis. J Bone Joint Surg 53A:719-728. Camplejohn KL, Allard SA (1988):Limitations of Safranin 0 staining in proteoglycan depleted cartilage demonstrated with monoclonal antibodies. Histochemistry 89:185-188. Chaminade F, Stanescu V, Stanescu R, Maroteaux P (1979): Link-proteins and non collagenous proteins from normal and chondrodysplastic cartilages. Eur J Pediatr 131:237-245. Diao E, Zaleske DJ, Arella D, Trahan C, Armstrong A, Ehrlich MG, Mankin HJ (1989): Kinetic and biochemical heterogeneity in vertebrate chondroepiphyseal regions during development. J Orthop Res 7:502-510. Farnum CE, Wilsman NJ (1988):Lectin-binding histochemistry of intracellular and extracellular glycoconjugates of the reserve cell zone of growth plate cartilage. J Orthop Res 6:166-179. Francomano CA, Pyeritz RE (1988): Achondroplasia is not caused by mutation in the gene for type I1 collagen. Am J Med Genet 29:955-961. Gerstenfeld LC, Finer MH, Boedtker H (1989): Quantitative analysis of collagen expression in embryonic chick chondrocytes having different developmental fates. J Biol Chem 2645112-5120. Glaser JE, Conrad HE (1981): Formation of matrix vesicles by cultured chick embryo chondrocytes. J Biol Chem 256:6757-6745. Hall JG, Dorst JP, Taybi H, Scott CI, Langer LO, Jr., McKusick VA (1969): Two probable cases of homozygosity for the achondroplasia gene. Birth Defects 5(4):24-34. Hamerman D, Sasse J, Klagsbrun MA (1986):A cartilage derived growth factor enhanced hyaluronate synthesis and diminishes sulfated glycosaminoglycan synthesis in chondrocytes. J Cell Physiol 127: 317-322. Hascall VC, Hascall GK (1984): Proteoglycans. In Hay ED (ed): “Cell Biology of Extracellular Matrix.” NY: Plenum Press, pp.39-60. Heinegard D, Sommarin Y (1987):Isolation and characterization of proteoglycan. In Cunningham LW (ed): “Methods in Enzymology,” vol. 144. NY: Academic Press, pp 319-372. Hincliffe JR, Johnson DR (1983): Growth of Cartilage. In Hall BK (ed): “Cartilage,” vol. 2. NY: Academic Press, pp 255-296. Horton WA, Chou JW, Machado MA (1985): Cartilage collagen analysis in the chondrodystrophies. Coll Relat Res 5:349-354. Horton WA, Hoid OJ,Machado MA, Campbell D (1988): Growth plate cartilage studies in achondroplasia. In Nicoletti B, Kopits SE, Ascani E, McKusick VA (eds): “Human Achondroplasia.” NY: Plenum Press, pp 81-90. Hwang WS (1978): Ultrastructure of human fetal and neonatal hyaline cartilage. J Path01 126:209-218. Ippolitc E, Maynard JA, Mickelson MR, Ponseti IV (1988): Histochemical and ultrastructural study of the growth plate in achondroplasia. In
Nicoletti B, Kopits SE, Ascani E, McKusick VA (eds):“Human Achondroplasia.” NY: Plenum, pp 61-72. Kato Y, Nomura Y, Tsuji M, Kinoshita M, Ohmal H, Suzuki F (1981): Somatomedin-like peptide(s) isolated from fetal bovine cartilage (cartilage-derived factor): isolation and some properties. Proc Natl Acad Sci USA 78:6831-6835. Kimata K, Barach HJ, Brown K, Pennypacker J P (1979): Absence of proteoglycan core protein in cartilage from CMD (cartilage deficient matrix) mouse. J Biol Chem 2565961-6968, Klagsbrun M, Smith S (1980): Purification of cartilage-derived growth factor. J Biol Chem 255:10859-10866. Kobayakawa M, Wata H, Brown KS, Kimata K (1985): Abnormal collagen fibrillogenesis in epiphyseal cartilage of CMD (cartilagematrix deficiency) mouse. Coll Relat Res 5:137-147. Kozlowski K, Masel J , Morris L, Kyan JH, Collins E, Van Vliet P, Woolnough H (1977): Neonatal death dwarfism (report of 17 cases). Australas Radiol 21:164-183. Mackler B, Grace R, Davis KA, Shepard TH, Hall J G (1986):Studies of human achondroplasia: oxidative metabolism in tissue culture cells. Teratology 33:9-13. Maynard J A , Ippolito EG, Ponseti IV,Mickelson MR (1981):Histochemistry and ultrastructure of the growth plate in achondroplasia. J Bone Joint Surg 63-A969-979. Mechanic G (1979): Collagen biochemistry. In Simmons DJ, Kunin AS (eds): “Skeletal Research. An Experimental Approach.” NY: Academic Press, pp 227-240. Miller EJ, Rhodes PK (1982): Preparation and characterization of the different types of collagen. In Cunningham LW, Frederiksen DW (eds):“Methods in Enzymology, vol. 2 Part A. NY: Academic Press, pp 33-64. Moskowitz N, Carson B, Kopits S, Levitt R, Hart G (1989): Foramen magnum decompression in an infant with homozygous achondroplasia. J Neurosurg 70:126-128. Nilsson A, Isgaard J, Lindahl A, Dahlstrom A, Skoltner A, Isaksson OGP (1986): Regulation by growth hormone of number of chondrocytes containing IFG-I in rat growth plate. Science 233571-574. Oberklaid F, Danks DM, Jensen F, Stace I, Rosshandler S (1979):Achondroplasia and hypochondroplasia: comments on frequency, mutation rate and radiological features in skull and spine. J Med Genet 16:140-146. Ogilvie D, Wordsworth P, Thompson E, Sykes B (1986):Evidence against the structural gene encoding type I1 collagen (COL2A1)as the mutant locus in achondroplasia. J Med Genet 23:19-22. Oike Y, Kimata K, Shinomura T, Suzuki S (1982):A mapping technique for probing the structure of proteoglycan core molecules. J Biol Chem 257:9751-9758. Pauli RM, Conroy MM, Langer LO, Mc Lone DG, Naidich T, Franciosi R, Ratner IM, Copps SC (1983): Homozygous achondroplasia with survival beyond infancy. Am J Med Genet 16459-473. Pedrini-Milk A, Pedrini V (1971):Studies of human iliac crest cartilage. I11 Proteinpolysaccharides in human achondroplasia. Calcif ‘Ilssue Res 8106-113. Peterkovsky B (1982): Bacterial collagenese. In Cunningham LW, Frederiksen DW (eds): “Methods in Enzymology.” NY: Academic Press, pp 453-471. Ponseti IV (1970): Skeletal growth in achondroplasia. J Bone Joint Surg 52:701-716. Ponseti IV (1988): Bone formation in achondroplasia. In Nicoletti B, Kopits SE, Ascani E, McKusick VA (eds): “Human Achondroplasia.” NY: Plenum, pp 109-122. Quinterer MI, Goetinck P F (1981): A biochemical analysis of cartilage proteoglycan in the avian mutant micromelia. Abbott Dev Genet 2:35-48. Rimoin DL, Hughes GNF, Kaufman RL, Rosenthal RE, Mc Alister WH, Silberberg R (1970): Endochondral ossification in achondroplastic dwarfism. N Engl J Med 283:728-735. Rimoin DL, Mc Alister WH, Saldino RM, Hall JG (1973): Histologic appearance of some types of congenital dwarfism. Prog Pediatr Radiol 4:68-92. Rogovits N, Weissenbacher G, Zweymiiller E (1972): Homozygote Achondroplasie und thanatophore Zwerwuchs-pranatal diagnostizierbare Skeletstorungen. Geburtshilfe Frauenheilkd 3 2 184-191.
Cartilage of Homozygous Achondroplasia Rosen DM, Stempren SA, Thompson AY, Seyedin SM (1988):‘Ikansforming growth factor-beta modulates the expression of osteoblast and chondroblast phenotypes in vitro. J Cell Physiol 134337-346. Rosenberg LC (1971):Chemical basis for the histologicaluse of Safranin 0 in the study of articular cartilage. J Bone Joint Surg 53-A:69-82. Shepard N, Mitchell N (1976): The localization of proteoglycan by light and electron microscopy using Safranin 0. J Ultrastruct Res 54:451-460. Sillence DO, Horton WA, Rimoin DL (1979): Morphologic studies in the skeletal dysplasias. Am J Pathol 96:813-860. Simon MR, Cooke PS (1988): Cellular heterogeneity and insulin-like growth factor I immunoreactivity among epiphyseal growth plate chondrocyte in the pig. Acta Anat (Basel) 133:66-69. Stanescu R, Stanescu V, Maroteaux P (1973): Histological and histochemical studies on the human growth cartilage in fetuses and newborns. Biol Neonate 23:414-431. Stanescu V, Bona C, Ionescu V, Gottlieb F (1966):Etudes histochimiques et histoenzymologiques sur le cartilage de croissance tibial dans l’achondroplasie humaine. Rev Fr Et Clin Biol 11:600-610. Stanescu V, Bona C, Ionescu V (1970):The tibial growing cartilage biopsy in the study of growth disturbances. Acta Endocrinol (Copenh) 64577-601. Stanescu V, Stanescu R, Szirmai JA (1972):Microchemical analysis ofthe human tibial growth cartilage in various forms of dwarfism. Acta Endocrinol (Copenh) 69:659-688. Stanescu V, Maroteaux P, Stanescu R (1976): Etude par electrophorese sur gel des chaines OL et des CNBr peptides de collagene du cartilage de croissance dans les chondrodysplasies. Ann Genet (Paris) 19:119-125. Stanescu V, Maroteaux P, Sobczak E (1977a): Proteoglycan populations of baboon (Papio papio) articular cartilage. Biochem J 163:103-109. Stanescu V, Stanescu R, Maroteaux P (197713):Etude morphologique
421
et biochimique du cartilage de croissance dans les ost60chondrodysplasies. Arch Fr Pediatr 34 (Suppl 3):1-80. Stanescu V, Stanescu R, Maroteaux P (1984):Pathogenic mechanisms in osteochondrodysplasias. J Bone Joint Surg 66-A817-836. Stanescu V, Chaminade F, Muriel MP (1986):The protein core ofthe layer proteoglycan monomer of articular cartilage. Connect Tissue Res 16:377-384. Stanescu V, Chaminade F (1987): Proteoglycan electrophoresis on horizontal submerged polyacrylamide-agarose gels. Connect Tissue Res 163-77. Stanescu V, Chaminade F, Muriel MP (1988): Age-related changes in small proteoglycans of low buoyant density of human articular cartilage. Connect T.lssue Res 17:239-252. Stirpe NS, Argraves WS, Goetinck PF (1987): Chondrocytes from the cartilage proteoglycan-deficient mutant, nanomelia, synthesize greatly reduced levels of the proteoglycan core protein transcript. Dev Biol 124:77-81. Szirmai JA (1963): Quantitative approaches in the histochemistry of mucopolysaccharides. J Histochem 11:24-34. Wilsman NJ, Farnum CE, Hilley HD (1981):Ultrastructural evidence ofa functional heterogeneity among physeal chondrocytes in growing swine. Am J Vet Res 42:188-195. Wordsworth P, Ogilvie D, Priestley L, Smith R, Wynne-Davis R, Bryan S (1988):Structural and segregation analysis of the type I collagen gene (COL2 Al) in some heritable chondrdysplasias. J Med Genet 251521-527. Woesner J F (1961): The determination of hydroxyproline in tissue and protein samples containing small proportions of this amino-acid.Arch Biochem Biophys 93:440-447. Yang SS,Heidelberger KP, Brough AJ, Corbett DP, Bernstein J (1976): Lethal short-limbed chondrodysplasia in early infancy. In Rosenberg HS, Bolarde RP (eds): “Perspectives in Pediatric Pathology,” vol. 3. Chicago: Year Book Medical Publisher, pp 1-40.