Circulating factors controlling cell proliferation.

Dipolid human fibroblast-rich tissues contain a macromolecule with a molecular weight between 30,000 and 50,000 daltons which will inhibit the proliferation of fibroblasts in the G1 phase of the cell cycle (i.e., inhibit both 3H-thymidine uptake as well as the normal increase in cell number). The inhibitor is destroyed by trypsin but not by ribonuclease or deoxyribonuclease, and it is thermolabile. It has an acid IEP. It is not cytotoxic, and its inhibitory activity appears to be completely reversible. This fibroblast endogenous inhibitor does not interfere with the proliferation of DNA synthesis by human lymphocytes, bronchial carcinoma cells, or HeLa cells. The activity does not appear to be species specific. Therefore, we suggest that it is quite possible that the control of fibroblast proliferation resides in a fibroblast chalone. Diploid human fibroblasts, in contrast to chicken or mouse fibroblasts or heteroploid fibroblasts in general, stringently require serum for their proliferation. All of this mitogenic activity of calf serum can be concentrated in a molecular weight range around 100,000 daltons by ultrafiltration. All of the mitogenic activity within this molecular weight class can be concentrated at a pH of 5.2 via isoelectric focusing, and all of the activity at this isoelectric point can be concentrated in one peak on preparative polyacrylamide gel electrophoresis. This latter material is homogeneous at three different pH's in analytical gel electrophoresis as well as in SDS electrophoresis. This purified serum mitogen for diploid human fibroblasts in vitro also works in vivo and represents as much as 0.5% of calf serum protein, albeit there is much less of this protein in adult cow or horse. It is composed of two equal subunits weighing about 60,000 daltons each and contains about 2 moles of sialic acid, one S-S bond, and 6 moles of hexose per subunit. There is a reciprocal relationship between the biological activity of fibroblast inhibitor and serum mitogen, but there is no apparent direct interaction between these two proteins. Addition of pure serum mitogen to diploid human fibroblasts in vitro results in the release of commensurable chalone activity into the medium and a reciprocal loss of mitogen from the medium. Therefore, we propose that serum contains a single macromolecule which competes with endogenous chalone on the surface of diploid human fibroblasts and that this functions as an anti-chalone for the fibroblast.