The heterogeneity of collagen solutions and its effect on fibril formation.

Abstract

The reconstitution of fibrils from the collagen in solutions extracted from developing connective tissue has attracted a great deal of attention because of the possibility that it is related to collagen fibrillogenesis in vivo. Study of the kinetics of this reconstitution (Gross, 1956; Gross & Kirk, 1958; Bensusan & Hoyt, 1958; Wood, 1958; Wood & Keech, 1960; Wood, 1960a, b; Bensusan, 1960; Bensusan & Scanu, 1960) has helped to elucidate the mechanism of the process. Wood & Keech (1960) found that fibrils form in collagen solutions that have been extracted from purified calf dermis by dilute acetic acid in two distinct, effectively consecutive steps. These occupy an initial lag period and a growth phase. Kinetic and electron-microscope data were satisfactorily explained by postulating that the first step consists of the aggregation of asymmetric collagen particles in solution to form nuclei and that during the second these nuclei grow into fibrils by the accretion of further collagen from solution. The rate of the growth during the second step and the diameter of the fibrils ultimately produced depend to a large extent on the nucleation process, and are related to the number and shape of the nuclei (Wood, 1960a). Nucleation and growth were regarded as closely similar aggregation reactions, the main difference between them being that nucleation is a homogeneous reaction whereas in growth both solid and liquid phases are involved. Certain observations, particularly some aspects of the effect of chondroitin sulphate on the rate of fibril formation (Wood, 1960b; Keech, 1961), could only be reconciled with some difficulty with this simple model; the results suggested that nucleation and growth might be fundamentally different processes. One possibility is that nuclei are the product of aggregation of only a small fraction of the total collagen in solution, the remaining collagen being unable to form nuclei but able to aggregate on nuclei to form fibrils. A similar type of heterogeneity of acidsoluble collagen had been suggested by Buzagh (1942). Boedtker & Doty (1956) found evidence from light-scattering data for the presence of a small amount of aggregated material in solutions of citrate-soluble collagen, which could only be removed by prolonged centrifuging. Fessler (1957, 1960a, b) observed that, under certain experimental conditions, collagen which was precipitated on warming solutions of neutral-salt-soluble collagen to 370 could be divided into two fractions. One of these ('fraction A') redissolved when the precipitate was kept at 0° for 48 hr.; the other ('fraction C') did not. In addition, a portion of the original collagen ('fraction B') was not precipitated. Except for these studies there appears to be little evidence from their physical properties that the collagen molecules in a solution of collagen are not identical. However, Jackson & Bentley (1960) have recently concluded that extracts of collagenous tissue contain collagen molecules of different ages. The work reported here was designed to test the hypothesis that collagen molecules in solution are heterogeneous with respect to their ability to aggregate into fibrils. A preliminary account of this work has been published (Wood, 1962).