Elsevier

Matrix Biology

Volume 21, Issue 3, April 2002, Pages 217-226
Matrix Biology

The NH2-terminal propeptides of fibrillar collagens: highly conserved domains with poorly understood functions

https://doi.org/10.1016/S0945-053X(02)00008-2Get rights and content

Abstract

The impetus for this review comes from the recent finding that the absence of the majority of the non-triple-helical sequence in the NH2-terminal propeptide (N-propeptide) of the proα1(I) collagen chain fails to generate a significant phenotype in the mouse (Bornstein et al., J. Biol. Chem., 277:2605–2613, 2002). This result is in apparent conflict with those of numerous studies in vitro that have implicated the N-propeptide in a number of processes that are involved in the biogenesis, maturation and function of type 1 collagen. To seek an explanation for this discrepancy, the sequences of the highly conserved, 55–57-amino acid, cysteine-rich repeats (CRR), which constitute the majority of the globular domains in the N-propeptides, were compared among 13 vertebrate species. Surprisingly, the CRR in mice and rats differs substantially from those in other mammalian species. Indeed, the CRR in birds, fish and amphibia are more similar to those of other mammals than are the CRR in rodents. This finding raises the possibility that the mutant mouse, which lacks exon 2 that encodes the CRR in the N-propeptide, might not be an appropriate model in which to study the function of the N-propeptide in other mammals. Alternatively, compensation, possibly by procollagens II or III, could account for the mild phenotype of the exon 2-deleted mouse. Yet another possibility is that the CRR plays a developmental role in the mouse, akin to that recently proposed for the N-propeptide in type IIA procollagen, rather than a function in collagen biogenesis. Some support for the latter possibility is provided by the observation that, on one background, the breeding of heterozygous exon 2-deleted mice generated homozygous mutants at less than the expected frequency. Experiments to examine these possibilities are proposed.

Introduction

Although the existence of non-triple helical regions or domains in collagens is now considered commonplace, this aspect of collagen structure was not appreciated some 30 years ago. At that time, known sequences that were not composed of repeating Gly-X-Y triplets were limited to the telo- or end-peptides of the α1 and α2 chains of type I collagen. Today, it is recognized that all of the 21 or more of the described collagen types contain ‘non-collagenous’ sequences, many of which separate triple helical segments. Sound experimental evidence for a biosynthetic precursor of type I collagen, termed procollagen, containing non-collagenous sequences, was first published by a number of different laboratories in 1971 (Bellamy and Bornstein, 1971, Layman et al., 1971, Jimenez et al., 1971, Lenaers et al., 1971, Stark et al., 1971). During the course of the following decade, the primary structures of types I, II and III procollagens were determined and a general molecular plan for these proteins was established. This plan consists of a central uninterrupted triple helix and NH2- and COOH-terminal propeptides that differ considerably in amino acid composition and sequence (see Bornstein, 1974, Martin et al., 1975 and Bornstein and Traub, 1979 for details and references to the earlier literature). Types V and XI procollagens also follow this model, but since the structure of the NH2-terminal (N)-propeptides in these proteins differs considerably from that in procollagens I, II and III (Takahara et al., 1995, Gregory et al., 2000), these proteins will not be considered in this review.

Of the fibrillar procollagens, types I, II and III are converted by procollagen N- and C-proteases to monomeric collagens, which are the direct precursors of collagen fibrils (Prockop and Kivirikko, 1995, Prockop et al., 1998, Hojima et al., 1994, Colige et al., 1997), and type V is partially processed by other enzymes (Unsöld et al., 2001). The non-triple helical regions of a number of other collagens are also subject to partial proteolysis. In the case of type XVIII collagen, a COOH-terminal sequence (endostatin) has been shown to be released enzymatically, both in cell culture and in vivo (Marneros and Olsen, 2001). However, such enzymatic processes, which can reveal putative cryptic functions in the precursor proteins, cannot be considered functionally analogous to the conversion of procollagens to collagens. Thus far, only fibrillar procollagens have been shown to require limited proteolysis to achieve a mature, functional state.

This review will compare the primary structures of the cysteine-rich repeats (CRR) in types I, II and III procollagens, which constitute almost the entire globular domains of the N-propeptides in these procollagens (Fig. 1), and will attempt to deduce the functions of these domains from new information obtained for both procollagens and other CRR-containing proteins. Since there is much less published information on the type III N-propeptide, attention will be focused on the N-propeptides of types I and II procollagens. A major conclusion of this review is that few, if any, of the previously proposed functions of the type I N-propeptide, some based on highly credible evidence in vitro, are supported by recent experiments in mice. We will therefore be obliged to rethink the possible functions of this domain in type I procollagen or at least consider the possibility that the mouse may not be an appropriate model for the function of the N-propeptide in other mammals.

Section snippets

The structure and function of cysteine-rich repeats (CRR) in proteins

CRR are found in a wide variety of proteins and are characterized by a signature sequence of 10 cysteines in the sequence CX1CX2CX3CX4CX5CX6CX7CCX8C. The sequences X1 to X8 are variable in length, with X1 and X6 the most variable. When the sequences of evolutionarily distant CRR are compared, only a few amino acids, with the exception of the cysteines, are conserved. These include glycine, tryptophan and proline in X1, and proline in X7.

However, in the N-propeptides of types I–III procollagens,

Amino acid sequences of the CRR in the N-propeptides of procollagens I–III

As shown in Fig. 1, the N-propeptide of the mouse proα1(I) chain is composed of a globular domain and a Gly-X-Y-repeating sequence that interacts with the corresponding sequences in a second proα1(I) chain and a proα2(I) chain to form a short triple helix. However, it should be noted that the proα2(I) chain lacks a CRR domain. In the mouse proα1(I) chain, the CRR constitutes the majority of the globular domain (56 of the 76 amino acids), and exon 2 encodes 65 amino acids that encompass the

Possible functions of the α1(I) collagen N-propeptide

Initial studies documenting the existence of a biosynthetic precursor of type I collagen, termed procollagen (Bellamy and Bornstein, 1971), focused on the presence of N-terminal ‘extensions’ now known as N-propeptides. A number of functions were attributed to these N-propeptides, including initiation of chain association in triple helix formation, inhibition of intracellular fibrillogenesis, and facilitation of both intra- and extracellular transport of procollagen (see Bornstein, 1974 and

Possible functions of the proα1(II) N-propeptide

The existence of a CRR in the proα1(II) N-propeptide was not recognized until well after the sequence of the cDNA encoding the major form of the proα1(II) chain had been determined. Initially, it was thought that the col2a1 gene lacked this sequence and resembled the col1a2 gene in this respect (Su et al., 1989). However, it was subsequently discovered that exon 2 does exist in the col2a1 gene, but is alternatively spliced (Ryan et al., 1990, Ryan and Sandell, 1990) in a pattern that may be

Conclusions and future directions

The possible functions of the N-propeptides of types I, II and III procollagens have remained a subject of interest, and to some extent an enigma, in collagen biology for the past three decades. The generation of mice with a targeted deletion of exon 2 in the col1a1 gene has revealed a phenotype that is unexpectedly mild and is characterized by an apparent absence of defects in procollagen chain synthesis, assembly, secretion, or proteolytic processing. The only abnormality determined thus far

Acknowledgments

Studies from the author's laboratory were supported by Grant AR 11248 from the National Institute of Health. I thank Helene Sage, Lucas Armstrong, and Mary Lou Augustine for helpful comments on the manuscript.

References (67)

  • S.A. Jimenez et al.

    Further evidence for a transport form of collagen: its extrusion and extracellular conversion to tropocollagen in embryonic tendon

    FEBS Letts.

    (1971)
  • S. Kanda et al.

    Role of thrombospondin-1-derived peptide, 4N1K, in FGF-2-induced angiogenesis

    Exp. Cell Res.

    (1999)
  • S. Kuroda et al.

    Biochemical characterization and expression analysis of neural thrombospondin-1-like proteins NELL1 and NELL2

    Biochem. Biophys. Res. Commun.

    (1999)
  • L.F. Lau et al.

    The CCN family of angiogenic regulators: the integrin connection

    Exp. Cell Res.

    (1999)
  • S.T. Lee et al.

    Deletion of the pro-alpha1(I) N-propeptide affects secretion of type-I collagen in Chinese hamster lung cells but not in Mov-13 mouse cells

    J. Biol. Chem.

    (1992)
  • A.G. Marneros et al.

    The role of collagen-derived proteolytic fragments in angiogenesis

    Matrix Biol.

    (2001)
  • G. Marqués et al.

    Production of a DPP activity gradient in the early Drosophila embryo through the opposing actions of the SOG and TLD proteins

    Cell

    (1997)
  • S.H. McLaughlin et al.

    Molecular recognition in procollagen chain assembly

    Matrix Biol.

    (1998)
  • T.M. Misenheimer et al.

    Physical characterization of the procollagen module of human thrombospondin 1 expressed in insect cells

    J. Biol. Chem.

    (2000)
  • L.J. Ng et al.

    Preferential expression of alternatively spliced messenger RNAs encoding type-II procollagen with a cysteine-rich amino-propeptide in differentiating cartilage and non-chondrogenic tissues during early mouse development

    Dev. Biol.

    (1993)
  • D.R. Olsen et al.

    Production of human type I collagen in yeast reveals unexpected new insights into the molecular assembly of collagen trimers

    J. Biol. Chem.

    (2001)
  • S. Piccolo et al.

    Cleavage of chordin by Xolloid metalloprotease suggests a role for proteolytic processing in the regulation of Spemann organizer activity

    Cell

    (1997)
  • S. Piccolo et al.

    Dorsoventral patterning in Xenopus: inhibition of ventral signals by direct binding of chordin to BMP-4

    Cell

    (1996)
  • D.L. Prockop et al.

    Procollagen N-proteinase and procollagen C-proteinase. Two unusual metalloproteinases that are essential for procollagen processing probably have important roles in development and cell signaling

    Matrix Biol.

    (1998)
  • M.C. Ryan et al.

    Differential expression of a cysteine-rich domain in the amino-terminal propeptide of type II (cartilage) procollagen by alternative splicing of mRNA

    J. Biol. Chem.

    (1990)
  • M.C. Ryan et al.

    The human type II procollagen gene: identification of an additional protein-coding domain and location of potential regulatory sequences in the promoter and first intron

    Genomics

    (1990)
  • M. Stark et al.

    Electronoptical studies of procollagen from the skin of dermatosparaxic calves

    FEBS Letts.

    (1971)
  • M.-W. Su et al.

    Organization of the exons coding for pro α1(II) collagen N-propeptide confirms a distinct evolutionary history of this domain of the fibrillar collagen genes

    Genomics

    (1989)
  • K. Takahara et al.

    Complete structural organization of the human α1(V) collagen gene (COL5A1): divergence from the conserved organization of other characterized fibrillar collagen genes

    Genomics

    (1995)
  • M. Wiestner et al.

    Inhibiting effect of procollagen peptides on collagen biosynthesis in fibroblast cultures

    J. Biol. Chem.

    (1979)
  • G. Bellamy et al.

    Evidence for procollagen, a biosynthetic precursor of collagen

    Proc. Natl. Acad. Sci. USA

    (1971)
  • D. Bornthron et al.

    Nucleotide sequence of pre-pro-von Willebrand factor cDNA

    Nucleic Acids Res.

    (1986)
  • P. Bomstein

    The biosynthesis of collagen

    Annu. Rev. Biochem.

    (1974)
  • Cited by (30)

    • Cysteine-rich domain of type III collagen N-propeptide inhibits fibroblast activation by attenuating TGFβ signaling

      2022, Matrix Biology
      Citation Excerpt :

      These data indicate that Col3 plays an important role in regulating wound healing and tumor development, both of which are also affected by TGFβ. The N-terminal propeptides of fibrillar collagens have been studied extensively both as markers in physiologic and pathologic processes, as well as for their roles in these events (reviewed in [33–35]). The N-propeptides of Col1 and Col2 are enzymatically cleaved during fibril assembly [8]; however, Col3 retains its N-propeptide to a greater extent than do other fibrillar collagens, and it persists in many tissues, suggesting that this domain may have a novel regulatory role [36].

    • Quantification of type II procollagen splice forms using alternative transcript-qPCR (AT-qPCR)

      2012, Matrix Biology
      Citation Excerpt :

      Recently, expression of the IID splice form was also observed in human articular chondrocytes during expansion in monolayer culture in the presence of BMP-2 (Claus et al., 2010). The alternatively-spliced exon 2-encoded domain of Col2a1 encodes a highly conserved cysteine-rich (CR) von Willebrand factor C-like domain within the amino (NH2) propeptide (Bornstein, 2002). This exon 2-encoded protein shares some homology with CR domains in other proteins that are known to bind to and regulate growth factors of the TGF-β superfamily (Abreu et al., 2002).

    View all citing articles on Scopus
    View full text