Collagen turnover was studied in mouse fibroblast cultures (3T6) by radioactive labelling and compartmental analyses. The incorporation of [14C]proline into protein during continuous labelling rapidly reached a maximum value which was directly proportional to the medium specific activity. Radioactivity appeared more slowly in hydroxyproline, and gradually accumulated as cultures became enriched in collagen and its breakdown products. In relation to total new protein synthesis, the proportional synthesis of collagen, as measured by the formation of [14C]hydroxyproline, was less in logarithmically growing than in stationary-phase cultures, and little was deposited in the cell layer.
Newly synthesized hydroxyproline was consistently present in all growth media. In stationary-phase cultures, media contained as much as 60% of the total [14C]hydroxyproline in a form soluble in 0.5 M perchloric acid. Gel filtration chromatography confirmed that this was predominantly free hydroxyproline, only 30% appearing in small peptides whose degree of hydroxylation suggested their origin from larger collagen molecules. This acid-soluble compartment was taken as a convenient index of collagenolysis, which proved to be significant in both growth states, but was proportionately more important throughout logarithmic growth.
Reincubation of prelabelled cultures in fresh medium containing an excess of non-radioactive proline (‘chase’ medium) was followed by the degradative loss of labelled cell layer protein. The released radioactivity could be quantitatively recovered in the growth medium for periods up to 6 days; the rate of its appearance was little influenced by the frequency of feeding. Despite extensive dilution of the proline precursor-pool specific activity, synthesis of [14C]hydroxyproline continued in all chase cultures. The increment appeared largely as collagen breakdown products in the growth medium, and probably arose from 2 principal sources: (1) recently deposited collagen, and (2) the hydroxylation of peptidyl-[14C]proline residues in protocollagen. The balance between these contributions seemed to be dependent upon the extent to which ‘ageing’ of the cell layer collagen had occurred prior to initiating the chase. Radioactive hydroxyproline was rapidly lost from briefly prelabelled cell layers, but was well retained in a macromolecular form when the initial labelling period was sufficiently prolonged. It is proposed that the endogenous collagen-degradative apparatus attacks both young collagen and its polypeptide precursor, but that as the lability of the former substrate rapidly declines, enzyme activity continues to operate on protocollagen to yield [14C]hydroxyproline-containing breakdown products which gradually diminish as the latter substrate pool is exhausted.