The C-proteinase that processes procollagens to fibrillar collagens is identical to the protein previously identified as bone morphogenic protein-1.

Bone morphogenic protein-1 (BMP-1) was originally identified as one of several BMPs that induced new bone formation when implanted into ectopic sites in rodents. BMP-1, however, differed from other BMPs in that it its structure was not similar to transforming growth factor beta. Instead, it had a large domain homologous to a metalloendopeptidase isolated from crayfish, an epidermal growth-factor-like domain, and three regions of internal sequence homology referred to as CUB domains. Therefore, BMP-1 was a member of the "astacin families" of zinc-requiring endopeptidases. Many astacins have been shown to play critical roles in embryonic hatching, dorsal/ventral patterning, and early developmental decisions. Here, we have obtained amino acid sequences and isolated cDNA clones for procollagen C-proteinase (EC 3.4.24.19), an enzyme that is essential for the processing of procollagens to fibrillar collagens. The results demonstrate that procollagen C-proteinase is identical to BMP-1.

Cleavage of type I procollagen by C- and N-proteinases is more rapid if the substrate is aggregated with dextran sulfate or polyethylene glycol.

The enzymes procollagen C- and N-proteinases specifically cleave carboxyl- and amino-terminal propeptides of procollagens. After cleavage of the propeptides, the resulting collagens self-assemble into fibrils. In most previous experiments with the enzymes, the substrate was monomeric type I procollagen. Here we have prepared aggregates of type I procollagen from chick embryo tendons by using 1 to 100 micrograms/ml of 500-kDa dextran sulfate or 3 to 5% (w/v) polyethylene glycol (M(r) 3350). Aggregation of the substrate with dextran sulfate increased its rate of cleavage by purified or crude C-proteinase from chick embryo tendons 10- to 15-fold. Aggregation of the substrate with 25 to 100 microgram/ml of dextran sulfate increased the rate of cleavage by purified N-proteinase about 4-fold. The rate of cleavage by crude N-proteinase was enhanced only about 2-fold, apparently because of partial precipitation of the enzyme by dextran sulfate. Using polyethylene glycol to aggregate the substrate increased the rate of cleavage by procollagen C-proteinases 5- to 20-fold. Aggregation with polyethylene glycol also increased the rate of cleavage by purified procollagen N-proteinases 2- to 5-fold. With crude N-proteinase, the rate of cleavage was increased only 1.5-fold. The results suggest that the rate of cleavage of the substrate by both enzymes is increased by the aggregation of the substrate itself by dextran sulfate or polyethylene glycol. The increased rates of cleavage seen after aggregation of substrate can be used to develop more sensitive assays for the enzymic activities.

Characterization of type I procollagen N-proteinase from fetal bovine tendon and skin. Purification of the 500-kilodalton form of the enzyme from bovine tendon.

Procollagen N-proteinase (EC 3.4.24.14) is the enzyme that specifically cleaves the NH2-terminal propeptides from type I procollagen. Two forms of N-proteinase with apparent molecular sizes of 300 and 500 kDa were found in partially purified preparations from fetal bovine tendon extracts. The 500-kDa form of enzyme was purified 16,000-fold with a recovery of 8% from the extracts of the tendons by six purification steps. The purified enzyme was a neutral, Ca(2+)-dependent proteinase (5-10 mM) that was inhibited by metal chelators. The 500-kDa enzyme contained unreduced polypeptides of 58, 125, 170, and 190 kDa which were separated by polyacrylamide gel electrophoresis in sodium dodecyl sulfate. Electron microscopic study indicated that the enzyme molecules were generally globular and had diameters of 33 +/- 4 nm. Other properties of the 500-kDa enzyme were: 1) the Km for type I procollagen is 35 nM at pH 7.5 and 35 degrees C, and the kcat is 290 h-1; 2) the activation energy for reaction with type I procollagen is 10,050 cal mol-1; 3) the isoelectric point is 3.8; 4) the enzyme cleaves the NH2-terminal propeptides of type II procollagen as well as type I procollagen but not of type III procollagen; and 5) the enzyme specifically cleaves a -Pro-Gln- bond in the pro-alpha 1(I) chain and an -Ala-Gln- bond in the pro-alpha 2(I) chain. The bovine N-proteinase with a mass of 300 kDa was found to be similar to the 500-kDa enzyme and appeared to be a degraded form of the 500-kDa enzyme generated during purification. The N-proteinase from fetal bovine skin extracts also contained 300-kDa and 500-kDa enzyme forms.

Self-assembly into fibrils of collagen II by enzymic cleavage of recombinant procollagen II. Lag period, critical concentration, and morphology of fibrils differ from collagen I.

A recently developed recombinant system for synthesis of human procollagen II by stably transfected host cells was used to prepare adequate amounts of protein to study the self-assembly of collagen II into fibrils. The procollagen II was cleaved to pCcollagen II by procollagen N-proteinase (EC 3.4.24.14), the pCcollagen II was chromatographically purified, and the pCcollagen II was then used as a substrate to generate collagen II fibrils by cleavage with procollagen C-proteinase. The kinetics for assembly of collagen II fibrils were similar to those observed previously for the self-assembly of collagen I in that a distinct lag phase was observed followed by a sigmoidal propagation phase. However, under the same experimental conditions, the lag time for assembly of collagen II fibrils was 5-6-fold longer, and the propagation rate for collagen II fibrils was about 30-fold lower than for collagen I fibrils. The relatively long lag time for the assembly of collagen II into fibrils made it possible to demonstrate that most of the conversion of pCcollagen II to collagen II occurred in the solution phase. The critical concentration at 37 degrees C for collagen II was about 50-fold greater than the critical concentration for collagen I. The Gibbs free energy change for the assembly of collagen II into fibrils was -40 kJ/mol, a value that was about 14 kJ/mol less than the free energy change for collagen I and about the same as the free energy change for the homotrimer of collagen I. Dark-field light microscopy and negative-staining electron microscopy demonstrated that the collagen II fibrils were thin and formed network-like structures. The results demonstrated, therefore, that the structural information of the monomer is sufficient to explain the characteristically small diameters and arcade-like geometry of collagen II fibrils found in cartilage and other tissues.

Self-assembly of collagen I from a proband homozygous for a mutation that substituted serine for glycine at position 661 in the alpha 2(I) chain. Possible relationship between the effects of mutations on critical concentration and the severity of the phenotype.

Procollagen I was isolated from cultured skin fibroblasts from a proband who was homozygous for a mutation in the COL1A2 gene that substituted a serine codon for a glycine codon at position 661 of the alpha 2(I) chain. The procollagen I was cleaved to pCcollagen I by procollagen N-proteinase and the pCcollagen I was used as a substrate for assay of self-assembly of collagen I into fibrils. The mutated pCcollagen I was cleaved to collagen I by procollagen C-proteinase at the same rate as control pCcollagen I. However, self-assembly of the mutated collagen I had a lag period that was 15-fold greater than the lag period observed with normal collagen I under the same conditions. Also, self-assembly of the mutated collagen I had a propagation rate of about one-fourth of the propagation rate of normal collagen I. In addition, the critical concentration for fibril assembly was slightly increased. Rotary shadowing electron microscopy of the mutated procollagen I did not reveal any increased flexibility of the triple helix as was seen previously with two mutated procollagens I in which there were substitutions of cysteine for glycine residues in the alpha 1(I) chain (Vogel, B. E., Doelz, R., Kadler, K. E., Hojima, Y., Engel, J., and Prockop, D. J. (1988) J. Biol. Chem. 263, 19249-19255; Lightfoot, S. J., Holmes, D. F., Brass, A., Grant, M. E., Byers, P. H., and Kadler, K. E. (1992) J. Biol. Chem. 267, 25521-25528). However, morphometric analysis by dark-field light microscopy and electron microscopy showed that the fibrils formed from the mutated collagen I appeared thicker in diameter than the fibrils formed from the normal collagen I. Comparison of the results with similar data on four mutated procollagens previously studied raised the possibility that mutations which markedly increase the critical concentration of fibril assembly produce more severe phenotypes than mutations which change other parameters of fibril assembly.

Cadmium ions inhibit procollagen C-proteinase and cupric ions inhibit procollagen N-proteinase.

Procollagen C- and N-proteinases specifically cleave the C- and N-terminal extension propeptides of type I, II and III procollagen molecules. The collagen molecules generated by the enzymes self-assemble into collagen fibrils. We previously observed the inhibition of these enzymes purified from chick tendons by several divalent metals. Here the inhibitory effects of CdCl2, CuCl2, ZnCl2, NiCl2, CoCl2 and Hg(C2H3O2)2 have been studied in detail using crude or purified C- and N-proteinases from chick tendons and sterna. CdCl2 was a strong inhibitor of C-proteinases from both sources, and the inhibition was independent of enzyme purity (I50 = 10-16 microM). In contrast, CuCl2 and ZnCl2 were inhibitory only of purified C-proteinase. With the N-proteinase, CuCl2 was a strong inhibitor, and the inhibition was independent of the purity of the enzyme preparation used (I50 = 14-40 microM). On the other hand, CdCl2 was a moderate inhibitor, and ZnCl2 was a strong inhibitor only of the purified N-proteinase (I50 = 8-17 microM). NiCl2 inhibited crude and purified N-proteinase from sternum (I50 = 23-29 microM) but not from tendon. These results suggest, therefore, that the accumulation of some of these metals in the body may cause suppression of collagen fibril formation in tissues.

Polymerization of pNcollagen I and copolymerization of pNcollagen I with collagen I. A kinetic, thermodynamic, and morphologic study.

Previous observations established that pNcollagen III copolymerized with collagen I and decreased the diameter of the fibrils formed (Romanic, A.M., Adachi, E., Kadler, K.E., Hojima, Y., and Prockop, D.J. (1991) J. Biol. Chem. 266, 12703-12709). Here, procollagen I alone or mixtures of procollagen I and pCcollagen I were incubated with procollagen C-proteinase to generate pNcollagen I or mixtures of pNcollagen I and collagen I. The results confirmed previous reports that pNcollagen I assembles into sheet-like structures. They also demonstrated that polymerization of pNcollagen I exhibits a lag period and propagation phase similar to those seen with other protein self-assembly systems. In addition, the results demonstrated that pNcollagen I formed true copolymers with collagen I in that the presence of pNcollagen I increased the lag time, decreased the propagation rate, and increased the concentration of collagen I in solution at equilibrium. Copolymerization of pNcollagen I with collagen I, however, differed in two features from copolymerization of pNcollagen III with collagen I. One was that, in confirmation of previous work, copolymerization of pNcollagen I with collagen I markedly altered the circularity of the fibrils formed. The second difference was that the copolymerization increased the concentration in solution at equilibrium of pNcollagen I whereas copolymerization with collagen I was previously shown to decrease the concentration in solution of pNcollagen III. The increase in concentration in solution of pNcollagen I was explicable either by the assembly of soluble oligomers of pNcollagen I and collagen I, or by subtle changes in the activities of pNcollagen I and collagen I in the solid-phase. Comparison with previous data with pNcollagen III indicated that although pNcollagen I and pNcollagen III copolymerize with collagen I, there are marked differences in the two kinds of copolymers.

Self-assembly into fibrils of a homotrimer of type I collagen.

Type I collagen, the most abundant structural protein in vertebrates, is comprised of two alpha 1(I) chains and one alpha 2(I) chain. Fibroblasts from a proband with osteogenesis imperfecta, however, were shown to synthesize a type I procollagen that was a homotrimer of pro alpha 1(I) chains. The absence of pro alpha 2(I) chains in the procollagen provided a unique opportunity to assess the role of the alpha 2(I) chain in collagen fibrillogenesis by examining the self-assembly de novo of the homotrimeric collagen generated in vitro. The results demonstrated that the fibrils formed by the homotrimeric collagen had an asymmetric banding pattern similar to fibrils of normal heterotrimeric type I collagen. However, the efficiency for self-assembly of the homotrimer into fibrils was markedly reduced in that the critical concentration at 37 degrees C was 40-fold greater than for self-assembly of the heterotrimeric molecule. A van't Hoff-type plot of the data was used to determine values for delta G, delta H and delta S. The values indicated the self-assembly of the homotrimer is similar to self-assembly of the heterotrimer in that the process is entropy driven. The process is, however, less favorable in that the delta G value was 10 kJ/mol less negative. The results suggest that the presence of the alpha 2(I) chain in type I collagen helps drive the self-assembly process, probably because the alpha 2(I) chain is more hydrophobic than the alpha 1(I) chain and, therefore, smaller amounts of structured water may be lost during self-assembly of the homotrimer than during self-assembly of the heterotrimer.(ABSTRACT TRUNCATED AT 250 WORDS)

Temperature-induced post-translational over-modification of type I procollagen. Effects of over-modification of the protein on the rate of cleavage by procollagen N-proteinase and on self-assembly of collagen into fibrils.

Previous observations suggested that incubating fibroblasts at elevated temperature caused over-modification of type I procollagen by post-translational enzymes because of a delay in folding of the collagen triple helix. Here, human skin fibroblasts were incubated at 40.5 instead of 37 degrees C, and the type I procollagen secreted into the medium was isolated. Analysis of the protein indicated that there was an increase of about 5 residues of hydroxylysine/alpha chain and about 1 residue of glycosylated hydroxylysine/alpha chain. Assays with procollagen N-proteinase indicated that the N-propeptide of the over-modified collagen was cleaved at a decreased rate, apparently because the over-modification altered the conformation-dependent cleavage site for the enzyme. Assays in a system for assembly of collagen into fibrils demonstrated that the over-modified protein had a higher critical concentration for self-assembly. Also, the fibrils formed from the over-modified collagen at 31 and 29 degrees C had smaller diameters than fibrils formed from normal type I collagen. The results provide direct evidence for earlier suggestions that post-translational over-modification of a fibrillar collagen can alter the morphology of the fibrils formed. The results also indicate that some of the biological consequences of the mutations in type I procollagen causing heritable disorders must be ascribed to the effects of post-translational over-modifications that frequently occur as secondary consequences of changes in the primary structure of the protein.

Copolymerization of pNcollagen III and collagen I. pNcollagen III decreases the rate of incorporation of collagen I into fibrils, the amount of collagen I incorporated, and the diameter of the fibrils formed.

Previous observations suggested that pNcollagen III, the partially processed form of type III procollagen, coats fibrils of collagen I and thereby helps regulate the diameter of fibrils formed by collagen I. The previous observations, however, did not exclude the possibility that pNcollagen III was deposited on preformed collagen I fibrils after the fibrils were assembled. Here, mixtures of pNcollagen III and collagen I were generated simultaneously by enzymatic cleavage of precursor forms of the proteins. The results demonstrated that pNcollagen III forms true copolymers with collagen I. The presence of pNcollagen III both inhibited the rate at which collagen I assembled into fibrils and decreased the amount of collagen I incorporated into fibrils at steady-state equilibrium. In addition, the results demonstrated that copolymerization of pNcollagen III with collagen I generated fibrils that were thinner than fibrils generated under the same conditions from collagen I alone. Increasing the initial molar ratio of pNcollagen III to collagen I in the solution-phase increased the amount of pNcollagen III copolymerizing with collagen I and progressively decreased the diameter of the fibrils. Therefore, the copolymers were heterogeneous in that the stoichiometry of the two monomers in the fibrils varied. The results are consistent with a model in which pNcollagen III can regulate the diameter of collagen I fibrils by coating the surface of the fibrils and thereby allow tip growth but not lateral growth of the fibrils.

A type I collagen with substitution of a cysteine for glycine-748 in the alpha 1(I) chain copolymerizes with normal type I collagen and can generate fractallike structures.

Type I procollagen was purified from cultured fibroblasts of a proband with a lethal variant of osteogenesis imperfecta. The protein was a mixture of normal procollagen and mutated procollagens containing a substitution of cysteine for glycine in either one pro alpha 1(I) chain or both pro alpha 1(I) chains, some or all of which were disulfide-linked through the cysteine at position alpha 1-748. The procollagen was then examined in a system for generating collagen fibrils de novo by cleavage of the pCcollagen to collagen with procollagen C-proteinase [Kadler et al. (1987) J. Biol. Chem. 262, 15696-15701]. The mutated collagens and normal collagens were found to form copolymers under a variety of experimental conditions. With two preparations of the protein that had a high content of alpha 1(I) chains disulfide-linked through the cysteine alpha 1-748, all the large structures formed had a distinctive, highly branched morphology that met one of the formal criteria for a fractal. Preparations with a lower content of disulfide-linked alpha 1(I) chains formed fibrils that were 4 times the diameter of control fibrils. The formation of copolymers was also demonstrated by the observation that the presence of mutated collagens decreased the rate of incorporation of normal collagen into fibrils. In addition, the solution-phase concentration at equilibrium of mixtures of mutated and normal collagens was 5-10-fold greater than that of normal collagen.(ABSTRACT TRUNCATED AT 250 WORDS)

Collagen fibrils in vitro grow from pointed tips in the C- to N-terminal direction.

Growth of collagen fibrils was examined in a system in which collagen monomers are generated by specific enzymic cleavage of type IpCcollagen with procollagen C-proteinase. Fibrils formed at 37 degrees C had highly tapered and symmetrical pointed tips. The pattern of cross-striations in the pointed tips indicated that all the molecules were oriented so that the N-termini were directed towards the tip. At 29 degrees C and 32 degrees C, the fibrils formed were thicker. One end of fibrils formed at 29 degrees C was blunt, and the other was pointed. Growth of the fibrils was exclusively from pointed tips. Occasionally a spear-like projection appeared at a blunted end. The spear-like projection then became a new pointed tip for growth in the opposite direction. The results suggested a model for fibril growth with at least three distinct binding sites for monomers. In the model, the pointed tip is the site with the highest affinity for the binding of monomers and most probably defines the critical concentration for fibril assembly. The main shaft of the fibril is a site with very low affinity for binding. The blunted end defines a low-affinity binding site where monomers can bind in opposite orientation to produce growth from a new pointed end.

Pleomorphism in type I collagen fibrils produced by persistence of the procollagen N-propeptide.

The assembly of type I collagen and type I pN-collagen was studied in vitro using a system for generating these molecules enzymatically from their immediate biosynthetic precursors. Collagen generated by C-proteinase digestion of pC-collagen formed D-periodically banded fibrils that were essentially cylindrical (i.e. circular in cross-section). In contrast, pN-collagen generated by C-proteinase digestion of procollagen formed thin, sheet-like structures that were axially D-periodic in longitudinal section, of varying lateral widths (up to several microns) and uniform in thickness (approximately 8 nm). Mixtures of collagen and pN-collagen assembled to form a variety of pleomorphic fibrils. With increasing pN-collagen content, fibril cross-sections were progressively distorted from circular to lobulated to thin and branched structures. Some of these structures were similar to fibrils observed in certain heritable disorders of connective tissue where N-terminal procollagen processing is defective. The observations are considered in terms of the hypothesis that the N-propeptides are preferentially located on the surface of a growing assembly. The implications for normal diameter control of collagen fibrils in vivo are discussed.

Type I procollagen: the gene-protein system that harbors most of the mutations causing osteogenesis imperfecta and probably more common heritable disorders of connective tissue.

Recent data from several laboratories have established that most variants of osteogenesis imperfecta (OI) are caused by mutations in the 2 structural genes for type I procollagen. There are 2 general reasons for the large number of mutations in type I procollagen in OI. One reason is that most of the structure of the procollagen monomer is essential for normal biological function of the protein. The second reason is that most of the mutations cause synthesis of structurally altered pro alpha chains of type I procollagen. The deleterious effects of the structurally altered pro alpha chains are then amplified by at least 3 mechanisms. One mechanism is a phenomenon referred to as "procollagen suicide" whereby altered pro alpha chains cause degradation of normal pro alpha chains synthesized by the same cell. Another mechanism involves the fact that many of the structurally altered pro alpha chains prevent normal processing of the N-propeptides of procollagen and persistence of the N-propeptide interferes with normal fibril assembly. A third mechanism is a recently discovered phenomenon in which a substitution of a bulkier amino acid for glycine can cause a kink in the triple helix of the molecule. The kinked collagen, in turn, causes formation of abnormally branched fibrils. Because the deleterious effects of abnormal pro alpha chains are amplified by these 3 mechanisms, most of the mutations are dominant and many are dominant lethal. The conclusion that most variants of OI are caused by mutations in the structural genes for type I procollagen has broad implications for other diseases that affect connective tissue, diseases such as chondrodystrophies, osteoarthritis, and osteoporosis.

Type I procollagen N-proteinase from chick embryo tendons. Purification of a new 500-kDa form of the enzyme and identification of the catalytically active polypeptides.

Procollagen N-proteinase (EC 3.4.24.14), the enzyme that cleaves the NH2-terminal propeptides from type I procollagen, was purified over 20,000-fold with a yield of 12% from extracts of 17-day-old chick embryo tendons. The procedure involved precipitation with ammonium sulfate, adsorption on concanavalin A-Sepharose, and five additional column chromatographic steps. The purified enzyme was a neutral, Ca2+-dependent proteinase (5-10 mM) that was inhibited by metal chelators. It had a molecular mass of 500 kDa as determined by gel filtration. The enzyme contained unreduced polypeptides of 61, 120, 135, and 161 kDa that were separated by polyacrylamide gel electrophoresis in sodium dodecyl sulfate. The 135- and 161-kDa polypeptides were catalytically active after elution from the polyacrylamide gel. Other properties of 500-kDa enzyme are: 1) the Km for type I procollagen is 54 nM at pH 7.5 and 35 degrees C, and the kappa cat is 350 h-1; 2) the activation energy for reaction with type I procollagen is 7,100 cal mol-1; 3) the isoelectric point is 3.6; and 4) the enzyme specifically cleaves the NH2-terminal propeptides of type I and II procollagen, but not of type III procollagen. A minor form of N-proteinase with a 300-kDa mass was also purified and was found to contain a 90-kDa polypeptide as the major active polypeptide. The enzyme appeared to be a degraded form of the 500-kDa N-proteinase. The properties of the 300-kDa enzyme were similar to those observed for the 500-kDa enzyme.

A substitution of cysteine for glycine 748 of the alpha 1 chain produces a kink at this site in the procollagen I molecule and an altered N-proteinase cleavage site over 225 nm away.

In previous work (Vogel, B. E., Minor, R. R., Freund, M., and Prockop, D. J. (1987) J. Biol. Chem. 262, 14737-14744), we identified a single-base mutation that converted the glycine at position 748 of the alpha 1 chain of type I procollagen to a cysteine in a proband with a lethal variant of osteogenesis imperfecta. In addition to posttranslational overmodification, the abnormal molecules displayed decreased thermal stability and a decreased rate of secretion. An unexplained finding was that procollagen was poorly processed to pCcollagen in postconfluent cultures of skin fibroblasts. Here, we show that the procollagen synthesized by the proband's cells is resistant to cleavage by procollagen N-proteinase, a conformation-sensitive enzyme. Since the only detectable defect in the molecule was the cysteine for glycine substitution, we assembled several space-filling models to try to explain how the structure of the N-proteinase cleavage site can be affected by an amino acid substitution over 700 amino acid residues or 225 nm away. The models incorporated a phase shift of a tripeptide unit in one or both of the alpha 1 chains. The most satisfactory models produced a flexible kink of 30 degrees or 60 degrees at the site of the cysteine substitution. Therefore, we examined the procollagen by electron microscopy. About 25% of the molecules had a kink not seen in control samples, and the kink was at the site of the cysteine substitution.

Assembly of type I collagen fibrils de novo. Between 37 and 41 degrees C the process is limited by micro-unfolding of monomers.

The effects of temperature on the assembly of collagen fibrils were examined in a system in which collagen monomers are generated de novo and in a physiological buffer by specific enzymic cleavage of type I pC-collagen, an intermediate in the normal processing of type I procollagen to type I collagen. Increasing the temperature of the reaction in the range of 29-35 degrees C decreased the turbidity lag and increased the rate of propagation as assayed by turbidity. The effect of temperature on the turbidity propagation rate gave a linear Arrhenius plot with a negative slope. The predicted value of the activation energy of propagation was 113 kJ/mol. However, the effects of temperature on the rate of assembly above 37 degrees C were opposite to the effects seen at temperatures below 37 degrees C. In the range of 37-41 degrees C, the turbidity propagation rate decreased markedly with temperature. Also, the turbidity lag increased. Therefore, much longer times were required for monomers to reach equilibrium with fibrils. A large fraction of the collagen monomers remaining in solution at temperatures above 37 degrees C was sensitive to rapid digestion by trypsin and alpha-chymotrypsin. Cooling the solutions to 25 degrees C made the monomers resistant to protease digestion. The results are consistent with the conclusion that, although formation of collagen fibrils is a classical example of an entropy-driven process of self-assembly, the rate of assembly between 37 and 41 degrees C is limited by reversible micro-unfolding of the monomer.

Expression of type I procollagen genes.

All of the type I collagen in connective tissue is the product of one structural gene for the pro alpha 1(I) chain and another for the pro alpha 2(I) chain of type I procollagen. An intriguing question therefore is how the expression of the two genes differs in mineralizing and non-mineralizing tissues. One approach that our laboratory has pursued to answer this and related questions is to develop a new system whereby one can examine the self-assembly of collagen fibrils de novo by controlled enzymic cleavage of procollagen to collagen under physiological conditions. The system has made it possible for the first time to define thermodynamic parameters for the self-assembly process. We are now using the system to define the normal kinetics for fibril formation. The results should make it possible to study the effects of other components of extracellular matrix on fibril assembly, including the effects of bone-specific components that initiate mineralization. A second approach has been to define mutations in type I procollagen genes that cause increased brittleness of bone. Over a dozen mutations in type I procollagen genes have been found in probands with osteogenesis imperfecta. One of the surprises has been that at least 25% of the probands with lethal variants of osteogenesis imperfecta have mutations in type I procollagen genes. Another surprise has been the observation that a number of the mutations are tissue specific in terms of their phenotypic manifestations even though the same abnormal pro alpha chains are being synthesized in a variety of tissues.