Human unhydroxylated homotrimeric triple-helical collagen I produced in transgenic plants was used as an experimental model to provide insights into the role of hydroxyproline in molecular folding and fibril formation. By using chemically cross-linked molecules, we show here that the absence of hydroxyproline residues does not prevent correct folding of the recombinant collagen although it markedly slows down the propagation rate compared with bovine fully hydroxylated homotrimeric collagen I. Relatively slow cis-trans-isomerization in the absence of hydroxyproline likely represents the rate-limiting factor in the propagation of the unhydroxylated collagen helix. Because of the lack of hydroxylation, recombinant collagen molecules showed increased flexibility as well as a reduced melting temperature compared with native homotrimers and heterotrimers, whereas the distribution of charged amino acids was unchanged. However, unlike with bovine collagen I, the recombinant collagen did not selfassemble into banded fibrils in physiological ionic strength buffer at 20°C. Striated fibrils were only obtained with low ionic strength buffer. We propose that, under physiological ionic strength conditions, the hydroxyl groups in the native molecule retain water more efficiently thus favoring correct fibril formation. The importance of hydroxyproline in collagen self-assembly suggested by others from the crystal structures of collagen model peptides is thus confirmed experimentally on the entire collagen molecule.

Many biological supramolecular structures are formed by polymerization of macromolecular monomers. Light scattering techniques can provide structural information from such systems, if suitable procedures are used to collect the data and then to extract the relevant parameters. We present an experimental set-up in which a commercial multiangle laser light scattering photometer is linked to a stopped-flow mixer, allowing, in principle, the time-resolved extrapolation of the weight-average molecular weight M(w) and of the z-average square radius of gyration (z) of the polymers from Zimm-like plots. However, if elongated structures are formed as the polymerization proceeds, curved plots rapidly arise, from which M(w) and (z) cannot be recovered by linear fitting. To verify the correctness of a polynomial fitting procedure, polydisperse collections of rod-like or worm-like particles of different lengths, generated at various stages during bifunctional polycondensations of rod-like macromolecular monomers, were considered. Then, the angular dependence of their time-averaged scattered intensity was calculated in the Rayleigh-Gans-Debye approximation, with random and systematic noise also added to the data. For relatively narrow size distributions, a third-degree polynomial fitting gave satisfactory results across a broad range of conversion degrees, yielding M(w) and (z) values within 2% and no greater than 10-20%, respectively, of the calculated values. When more broad size distributions were analyzed, the procedure still performed well for semiflexible polymers, but started to seriously underestimate both M(w) and (z) when rigid rod-like particles were analyzed, even at relatively low conversion degrees. The data were also analyzed in the framework of the Casassa approximation, from which the mass per unit length of the polymers can be derived. These procedures were applied to a set of data taken on the early stages of the thrombin-catalyzed polymerization of fibrinogen, a rod-like macromolecule approximately 50 nm long. The polymers, grown in the absence of Ca(2+) by rate-limiting amounts of thrombin, appeared to be characterized by a much broader size distribution than the one expected for a classical Flory bifunctional polycondensation, and they seem to behave as relatively flexible worm-like double-stranded chains. Evidence for the formation of fibrinogen-fibrin monomer complexes is also inferred from the time dependence of the mass/length ratio. However, our data are also compatible with the presence of limited amounts of single-stranded structures in the very early stages, either as a secondary, less populated pathway, or as transient intermediates to the classical double-stranded fibrils.

Procollagen C-proteinase enhancer (PCPE) is an extracellular matrix glycoprotein that binds to the C-propeptide of procollagen I and can enhance the activities of procollagen C-proteinases up to 20-fold. To determine the molecular mechanism of PCPE activity, the interactions of the recombinant protein with the procollagen molecule as well as with its isolated C-propeptide domain were studied using surface plasmon resonance (BIAcore) technology. Binding required the presence of divalent metal cations such as calcium and manganese. By ligand blotting, calcium was found to bind to the C-propeptide domains of procollagens I and III but not to PCPE. By chemical cross-linking, the stoichiometry of the PCPE/C-propeptide interaction was found to be 1:1 in accordance with enzyme kinetic data. The use of a monoclonal antibody directed against the N-terminal region of the C-propeptide suggested that this region is probably not involved in binding to PCPE. Association and dissociation kinetics of the Cpropeptide domains of procollagens I and III on immobilized PCPE were rapid. Extrapolation to saturation equilibrium yielded apparent equilibrium dissociation constants in the range 150 -400 nM. In contrast, the association/dissociation kinetics of intact procollagen molecules on immobilized PCPE were relatively slow, corresponding to a dissociation constant of 1 nM. Finally, pN-collagen (i.e. procollagen devoid of the C-terminal propeptide domain) was also found to bind to immobilized PCPE, suggesting that PCPE binds to sites on either side of the procollagen cleavage site, thereby facilitating the action of procollagen C-proteinases.Bone morphogenetic protein-1 (BMP-1) 1 and other tolloidrelated metalloproteinases (1, 2), also known as procollagen C-proteinases (PCPs), have recently been shown to be involved in the control of a variety of morphogenetic events during development and tissue repair. These include: (i) the deposition of collagen fibrils in the extracellular matrix following the processing of procollagen propeptides (3-6); (ii) dorsoventral patterning (7-10) through the cleavage of the growth factor inhibitors chordin and SOG; (iii) collagen and elastin crosslinking by the processing of the inactive precursors of lysyl oxidases (11, 12); and (iv) adhesion/migration of epithelial cells by the cleavage of laminin 5 chains (13-15). The activities of PCPs on procollagen substrates may be stimulated up to 20-fold by another glycoprotein of the extracellular matrix, procollagen C-proteinase enhancer (PCPE) (16 -19), which lacks intrinsic proteinase activity. Similarly, in the case of chordin and SOG, the protein TSG or its homologues stimulates cleavage by tolloid proteinases (20, 21), thus raising the possibility that PCP processing of different substrates might be specifically regulated by distinct enhancer proteins.Both tolloid proteinases and PCPE are multidomain glycoproteins containing multiple copies of the so-called CUB domain (22), a protein module common to several extracellular and plasma membrane-associated...