We have been studying the growth regulation of nontumorigenic versus tumorigenic cells in vitro and in vivo. In these studies, we have found that a variety of mammalian cell cultures which are nontumorigenic in athymic nude mice or in syngeneic hosts grow exponentially if implanted in vivo within sealed Millipore diffusion chambers. The growth rates of the cells in vivo within the diffusion chambers were comparable to those observed in vitro. Moreover, the cells grown in vivo within the chambers exhibited density-dependent growth inhibition which was similar, at least superficially, to the in vitro phenomenon.

The ability of the nontumorigenic cells to grow when implanted in vivo within diffusion chambers was correlated with and was thought due, in part, to the formation of serous clots within the implanted chambers. Studies of the clots by light, scanning electron, and electron microscopy showed cells throughout the clot and adherent to the clot fibers. We tested the growth of each of the cell lines for dependence on serum growth factors by observing the ability of the cells to grow in medium supplemented with serum versus defibrinogenated, platelet-poor plasma (PPP). BALB/c-3T3 cells and other nontumorigenic, fibroblast-like cell lines failed to proliferate in vitro when defibrinogenated, PPP was substituted for serum, demonstrating a requirement by the cells for growth factors unique to serum. Virally transformed, tumorigenic fibroblasts grew in media supplemented with either PPP or serum. Thus, for these fibroblast-like cell lines, the ability of the cells to grow in PPP is correlated with tumorigenicity in athymic nude mice.

Since the interstitial fluid surrounding cells in vivo resembles PPP more than serum and since many of the nontumorigenic fibroblast-like cell lines demonstrated a requirement for serum growth factor, the growth of these cell lines within the implanted diffusion chambers suggests that serum growth factors might be required for growth of such cells in vivo and that serum growth factors can accumulate within the diffusion chambers. Analysis of the serous clot material provided evidence to coroborate this hypothesis. The fluid from implanted chambers can stimulate growth of BALB/c-3T3 cells. The growth factor activity of the serous chamber fluid is more comparable to that of mouse serum than to that of mouse PPP.

Our studies suggest that two of the possible factors provided by implanted Millipore chambers in vivo and required for growth of nontumorigenic fibroblast-like cells are: (a) a substrate for cell anchorage provided in this case by the fibrous clots and by the nitrocellulose surface of the Millipore filters; and (b) growth factors which accumulate within the microenvironment of the diffusion chamber implants.


This research was supported in part by USPHS Grant 1-R01-CA-16521-01A1MB4 and CA22042 and American Cancer Society Grant BC-224.

This content is only available via PDF.