Download pdf - Presence of type III collagen in bone from a patient with osteogenesis imperfecta

European Joumal of

Pediatrics Europ. J. Pediat. 125, 29--37 (1977)

�9 by Springer-Verlag 1977

Presence of Type III Collagen in Bone from a Patient with Osteogenesis Imperfecta

P. K. Mtiller l, K. Raisch 1, K. Matzen 2, and S. Gay~

1 Max-Planck-Institut ftir Biochemie, Abt. Bindegewebsforschung (Dir.: Prof. Dr. K. Kiihn), D-8033 Martinsried b. Mtinchen, Federal Republic of Germany

2 Orthopedic Clinic, University of Munich (Dir.: Prof. A. N. Witt), Harlachinger Stral3e, D-8000 Miinchen, Federal Republic of Germany

Abstract. Samples of bone from a patient with osteogenesis imperfecta were found to synthesize and contain type III collagen as well as type I collagen. Normal bone contains only type I collagen except in the lining cells of the bone marrow cavities. In the patient's tissue, type III collagen was localized in non- fibrillar structures in discrete areas of the bone. These and previous studies indicate that certain types of osteogenesis imperfecta may be caused by a failure of normal bone maturation and the sites in which the type III collagen is found appear to be defects in the bone.

Key words: Osteogenesis imperfecta - Collagen types - Bone In vitro study - Immunofluorescence.

Introduction

A marked increase in bone fragility is found in patients with osteogenesis imperfecta. Some patients are born with multiple fractures and with other severe manifestations of the disorder. Such patients are classified as having osteogenesis imperfecta congenita. Other patients, less severely affected, are classified as osteogenesis imperfecta tarda. In both forms of the disease a variety of tissues including sclera, teeth and skin are affected in addition to bone and these disorders are not thought to be just a disease of bone (McKusick, 1972).

Morphological studies on specimens from patients with osteogenesis imperfecta indicate that bone from these patients is less mature than normal. The collagen is observed in fibres of diverse diameter and these have the affinity for silver stains shown by reticulin but not by collagen fibres from mature bone. In addition, the manner in which the collagen fibres in the bone are organized resembles foetal more than adult tissue. These observations point to defects in bone development and in collagen structure (McKusick, 1972).

Recent evidence indicates that skin fibroblasts from some patients with osteogenesis imperfecta synthesize more type III collagen, relative to type I collagen, than normal (Penttinen et al., 1975; Mtfller et al., 1975). Normally, type III collagen is found in higher concentrations in foetal skin than in adult skin although it is retained throughout life in skin, aorta and other tissues (Epstein et

30 P.K. Mtiller et al.

al . , 1975). N o r m a l b o n e appea r s to c o n t a i n p r i m a r i l y type I co l lagen . H e r e we

h a v e s tud ied the o c c u r r e n c e o f t ype I a n d I I I co l l agen in samples o f b o n e f r o m a

pa t i en t w i t h o s t eogenes i s impe r f ec t a .

Materials and Methods

Materials

L-(3,4-3H)proline (24.5 Ci /m tool) and (35S)cysteine (38.99 Ci /m tool) were purchased from the New England Nuclear Corporation (Dreieichenhain, FRG). Dulbecco's modified Eagles medium was obtained from Laborservice (Miinchen, FRG); carboxymethylcellulose (Whatman, micro granular, preswollen) was obtained from Hormuth and Vetter (Heidelberg, FRG) and Agarose A 1.5 ( 2 0 0 ~ 0 0 mesh) was a product of Bio-Rad Laboratories (Mtinchen, FRG). Bovine pepsin (2 x crystallized) was purchased from Serva (Heidelberg, FRG). Aquasol (New England Nuclear Corp.) was used for liquid scintillation counting in a Beckman LS 250. Falcon tissue culture flasks (Becton and Dickinson) were obtained from Laborservice (Mtinchen, FRG).

Case Report

The patient was first hospitalized at the age of 5 years due to impaired growth and reportedly "weak" bones with numerous fractures. Mental development was normal but skeletal development was slow. The clinical manifestations, including fragile bones and blue sclera, justified classification of the disorder as osteogenesis imperfecta tarda (McKusick, 1972). A detailed documentation of the clinical studies will be published elsewherel. As a consequence of fracture, a pseudoarthrosis with subsequent dislocation developed at the middle of the left radius. During surgery, a section of bone about 2 cm in length was removed from the radius and the distal end of the specimen was used for biochemical and immunological studies.

Tissue Culture and Biochemical Procedures

Following removal, bone samples were placed immediately in tissue culture medium. About 5 g of bone was minced into small pieces and incubated in Dulbecco Vogt's modified Eagles medium lacking cysteine at 37 ~ with continuous shaking. The medium was supplemented with sodium ascorbate (50 ~g/ml), penicillin (400 U/ml) and/%aminoproprionitrile (100 ixg/ml) and flushed with a mixture of CO2/air (5%/95%). Radioactive amino acids (250 ~ Ci each of L-(3,4- 3H)proline and L-(3sS)cysteine) were added to a total volume of 20 ml of the medium and incubated for 24 hours with the tissue. At the end of this period, the medium was decanted and protein precipitated with ammonium sulfate (40% saturation at 4~ The bone tissue was demineralized using 0.01 M EDTA and subsequently dialysed against 0.5 M acetic acid. After 4 days, the pieces of bone were homogenized in a Potter Elvehjem homogenizer. The homogenized tissue and the (NH4)2SO4 precipitate were treated with pepsin (Pontz et al., 1973). Since little radioactivity was observed in the medium fraction following this treatment, it was combined with the material solubilized from the tissue with pepsin before chromatography. Aliquots were dialyzed against the appropriate buffers and chromatographed either on CM-cellulose or Agarose A 1.5 under denaturing conditions. Collagen from the skin of lathyritic rats was added as carrier protein and as an internal marker (Pontz et al., 1973; Piez et al., 1963; Piez, 1968).

In order to form segment-long-spacing (SLS) aggregates, the pepsin-solubilized material was dialyzed against 0.4% ATP. The SLS aggregates were stained with uranyl acetate and phosphotungstic acid and examined in a Siemens Elmiskop I electron microscope. Other portions of the insoluble residue were used for amino acid analysis. For immunofluorescent

1 Matzen et al., in preparation

Type III Collagen in Osteogenesis Imperfecta 31

investigation, a small piece of the fresh bone was also demineralized and stained with type- specific antibodies aqainst either type I or type III collagen (Nowack et al., 1976). These samples were examined in a Zeiss standard 19 microscope, equipped with a fluorescent incident light condenser. Similar studies were carried out with bone from healthy individuals of the same age.

R e s u l t s

The bone specimen of the patient with osteogenesis imperfecta incorporated much more (35S) cysteine and L-(3,4-3H)proline into pept ide-bound material than bone f rom healthy individuals. Since it was the aim of this study to show synthesis of type I I I collagen, we treated the newly synthesized material with pepsin, which removes procol lagen peptides containing (asS-)cysteine label. Thus, the presence of this amino acid is a marker for type I I I collagen, which contains cysteine in a pepsin-resistant region of the molecule.

When the pepsin digested material was applied to Agarose A 1.5, a significant p ropor t ion of the radioactivi ty was found as (35S) cysteine-containing material, the size of 7 -componen t s of collagen (Fig. 1). This material, when isolated and reduced, was converted to a-sized chains as judged by polyacrylamide gel

0.6

E

e,l

"0.4

z

rn n~

~o~ <

~ 3 E

o

E N 2

m

i

LRSC A

~ ~ vt

I I I I

B

200 300 400 EFFLUEN f VOLUME,ml

Fig. 1 A and B. Pattern of molecular sieve chromatography on Agarose A 1.5. A; Elution pattern of lathyritic rat skin collagen, added as an internal marker. B: Distribution of radioactivity incorporated into protein resistant to pepsin digestion. ( i__l__i) 35S_cysteine ' (--A--A--) L(3,4-3H)proline. The column was eluted with I M CaC12, 0.05 M Tris pH 7.4 at room temper- ature. 0.5 ml aliquots of the 6 ml fractions were used for determining radioactivity

3-L z, "0 3

3

~n 13

L ~ I

co

32 P.K. Mttller et al.

0.8

E o.6

('-.i

z a~ m

mm 0.2 <

A LRSC ~,I{I) /~12 ~.2

3rodientl

5 L . _ _ _

g

trl

I I I I I

o 200 400 600 800 EFFLUENT VOLUME,ml

Fig.2 A and B. Separation of the components of collagen by CM-cellulose chromatography under denaturing conditions. A: Elution pat tern of lathyritic rat skin collagen, present as an internal marker. B: Distribution of the newly synthesized protein resistant to pepsin digestion. ( I - - l - - l ) 3sS-cystein, ( - - A - - A - - ) L(3,4-3H)proline. The resin (1.5 x 8 cm) was equilibrated with 0.02 M potassium acetate and 1 M urea pH 4.8 and maintained at 44 ~ . The column was developed by a linear gradient from 0 to 0.14 M LiC1 over a total volume of 1 1. 1 ml aliquots of the 10 ml fractions were used for determining radioactivity

electrophoresis (not shown). The Agarose chromatogram also indicated that only a small proportion of the radioactive material had the size of oe-ehain of collagen (Fig. 1). In addition, we applied another aliquot of the pepsin-solubilized material to CM-cellulose and chromatographed it under denaturing conditions (Fig. 2). Both type I and type III collagen chains were obtained from the osteogenesis imperfecta bone.

Since type I collagen lacks cysteine, the distribution of (3sS)-cysteine indicates the occurrence of type III collagen. Usually, the type III collagen trimer chromatographs in a position between /312 components and ~2 chains of authentic lathyritic rat skin collagen. The nature of the (asS) cysteine labelled material eluting with t~ l(I) chains is not yet clear, since further investigation of this material did not render conclusive results due to lack of material. The material eluting in the region of/312 was also isolated and applied to a column of Agarose A 1.5. This material had the size of trimers of collagen (not shown).

In contrast, the little radioactive material synthesized by healthy bone during in vitro incubation could not be identified as either type I or type III collagen when chromatographed on CM-cellulose under denaturing conditions.


Fig, 3. Electronmicrograph of segments-long-spacing aggregates of demineralized and pepsin- solubilized bone. Following dialysis against 0.4% ATP pH 3.0, the precipitate was stained with uranyl-acetate and phosphotungstic acid. Magnification: 90,000x

Fig. 4 a--c. Electron micrographs of segments-long-spacing aggregates after staining with uranyl-acetate and phosphotungstic acid. a: Typical cross striation pattern of segments-long- spacing obtained from type I collagen (in solubilized material from affected bone), b: Type III collagen, e: Typical cross striation pattern of segments-long-spacing type lII collagen from calf skin. The arrows indicate the areas with characteristic differences in the cross striation pattern of type I and type III collagen. Magnification: 180,000x

34 P.K. Mtiller et al.

Fig. 5. Immunofluorescence micrographs of demineralized compacta bone from healthy individuals, stained with antitype I collagen antibodies. Magnification: 160x

Further evidence for the occurrence of type I and type III collagen was obtained by studying the SLS aggregates formed by exposure of the pepsin- solubilized material to ATP. The band patterns of Type I and III collagen are readily distinguished (Fig. 3). The cross striation pattern of SLS crystallites of human type III collagen is identical to that of type III collagen from calf skin (Fig. 4).

Specific antisera against type I or type III collagen were used to establish the distribution of these collagens in normal and osteogenesis imperfeeta bone. Only type I collagen was found in the compacta region of normal bone (Fig. 5), while type III collagen was found exclusively in association with the marrow cavities (not shown). Considerable difference was noted in t he osteogenesis imperfecta bone. The compact region of osteogenesis imperfecta bone contained circular structures ("holes"), in addition to fibrous areas (Fig. 6). The circular structures were found to react with type III collagen antibodies but not with type I antibodies (Fig. 6b). Similar abnormal structures have been observed recently in bone from other patients with this disease (Teitelbaum et al., 1974). The normal

Type III Collagen in Osteogenesis Imperfeeta 35

Fig. 6 a and b. Immunofluorescence micrographs of demineralized bone from the patient with osteogenesis imperfecta tarda, stained with the anti-type III antibodies, a Fibrous structures in the compacta, stained with anti-type I collagen antibodies, b Circular structures ("holes") in the compacta. Magnification: 80x

fibrous tissue surrounding the "holes" stained with type I, but not with type lII, antibodies (Fig. 6b). The circular structures were clearly distinct from marrow cavities which also stained with type I I I collagen antibodies.

Discussion

Previous studies on fibroblasts from certain patients with osteogenesis imperfecta have shown an elevated synthesis of type I I I collagen, relative to type I, in monolayer culture (Penttinen et al., 1975; MUller et al., 1975). The results presented here on bone tissue corroborate the earlier conclusions with skin fibroblasts that the normal metabolism of collagen by bone is altered in these patients. We have shown by biochemical methods that a bone specimen from an affected child synthesized type I I I collagen in organ culture. This was not found with bone f rom healthy individuals, which was metabolically less active. A pattern characteristic of type I and type I I I collagen was observed on CM- cellulose chromatography of the synthesized material. The demonstration of reducible -/-components firmly indicates the synthesis of type III collagen. The presence of type I I I collagen in the bone was further substantiated by studying both electron micrographs of segment-long-spacing aggregates of collagen prepared from the bone and by immunofluorescent identification. Both methods clearly demonstrated the presence of type I I I collagen in bone f rom this patient

36 P.K. MOiler et al.

with osteogenesis imperfecta. Of part icular interest, the staining indicated that there are "holes" filled with type I I I collagen in the compact bone.

It should be ment ioned that the heterogeneity of the (3sS)-cysteine labelled material eluted both with a l chains and i l l2 components of lathyritic rat skin collagen was also found in previous studies in another case of osteogenesis imperfecta (Mtiller et al., 1975). In agreement with the results outlined above, we found a ratio of hydroxyprol ine to proline in the affected bone which was slightly higher than in normal bone. This also indicates the presence of type I I I collagen, which is known to have a higher content of hydroxyprol ine than type I collagen (Chung et al., 1974; Epstein, 1974).

The initial events of the disorder in this patient are obscure. It is possible that "holes" formed in the compac t bone because the osteoblasts synthesized an excess of type I I I collagen during development and that this did not mineralize. Alternatively, the holes could have formed because there was a defect in either the synthesis or the degradat ion of type I collagen. It is conceivable that "repair" cells, which synthesize an excessive p ropor t ion of type I I I collagen, fill-in the holes with type I I I collagen. In any case, these abnormal structures may explain the high fragility of the bones and the slow healing of fractures in patients with this form of osteogenesis imperfecta tarda. However, it should be mentioned that we have studied other patients with osteogenesis imperfecta tarda and found normal synthesis of type I I I collagen. This indicates tha t the disorder does not represent a single entity and it is likely that it can be caused by a number of fundamenta l ly distinct biochemical defects.

Acknowledgements. We are gratefully indebted to Prof. Dr. K. Ktlhn and Prof. Dr. A. N. Witt for supporting our study and for providing us with tissue and bone specimens. We thank Mrs. Helene Kadner for excellent technical assistance. This study was supported by a grant from the Deutsche Forschungsgemeinschaft (Mu 378/4 1975) within the project group "Biopolymere und Biomechanik yon Bindegewebssystemen".

References

Chung, E., Miller, E. J.: Collagen polymorphism: characterization of molecules with the chain composition (al(III)) 3 in human tissues. Science 183, 1200--1201 (1974)

Epstein, E. H. Jr.: (al(III))3 Human skin collagen. J. Biol. Chem. 249, 3225--3231 (1974) Epstein, E. H. Jr. :, Munderloh, N. H.: Isolation and characterization of CNBr peptides of human

(al(III))3 collagen and tissue distribution of (~1(I))2t~2 and (al(III)) 3 collagen. J. Biol. Chem. 250, 9304--9312 (1975)

MeKusick, V. A.: In: "Heritable Disorders of Connective Tissue", pp. 390--441. Saint Louis, USA: The C. V. Mosby Company 1972

MUller, P. K., Lemmen, C., Gay, S., Meigel, W. N.: Disturbance in the regulation of the type of collagen synthesized in a form of osteogenesis imperfecta. Eur. J. Biochem. 59, 97--104 (1975)

Nowack, H., Gay, S., Wick, G., Becker, U., Timpl, R.: Preparation and use in immunohistology of antibodies specific for type I and type III collagen and procollagen. J. Immunol. Methods 12, 117--124 (1976)

Penttinen, R. P., Lichtenstein, J. R., Martin, G. R., McKusick, V. A.: Abnormal collagen metabolism in cultured cells in osteogenesis imperfecta. Proc. Nat. Acad. Sci. USA 72, 586--589 (1975)


Piez, K. A., Eigner, E. A., Lewis, M. S.: The chromatographic separation and amino acid composition of the subunits of several collagens. Biochemistry 2, 58--66 (1963)

Piez, K. A.: Molecular weight determination of random coil polypeptides from collagen by molecular sieve chromatography. Anal. Biochem. 26, 305--312 (1968)

Pontz, B. F., MUller, P. K., Meigel, W. N.: A study on the conversion of procollagen. J. Biol. Chem. 248, 7558--7564 (1973)

Teitelbaum, S. L., Kraft, W. J., Avioli, L. V.: Bone collagen aggregation abnormalities in osteogenesis imperfeeta. Calcif. Tissue Res. 17, 75--79 (1974)

Received October 1, 1976