Properties of a sarcoma-growth-factor-like peptide from cells transformed by a temperature-sensitive sarcoma virus

Serum-free conditioned media was collected from three sarcoma virus-transformed cell lines and an untransformed cell line. All three virally transformed lines produced and released growth factors into their serum-free media. The major activity in all cases, whether the cells were transformed by Moloney sarcoma virus (MSV) or Kirsten sarcoma virus (KiSV), or whether they were mouse or rat, was a sarcoma-growth-factor (SGF)-like activity with an apparent molecular weight of 10,000. The SGF-like pools from a Moloney sarcoma virus-transformed mouse 3T3 cell and a Kirsten sarcoma virus-transformed NRK cell were further purified by carboxymethyl cellulose chromatography. The elution profiles of these peptides were very similar. The serum-free conditioned media from the untransformed cells showed no detectable growth stimulating activity. The temperature sensitivity of an SGF-like growth factor from the supernate of a NRK cell transformed by a wild-type Kirsten sarcoma virus (KiSV) was compared with that of the SGF-like activity from the supernates of a NRK cell transformed by a ts-mutant of KiSV that is temperature sensitive with respect to transformation (ts-371 Cl 5). Neither the cells transformed by the wild-type sarcoma virus nor those transformed by the temperature sensitive virus released a SGF-like activity that was temperature sensitive under the conditions of the assays.     

Rous sarcoma virus-transformed fibroblasts adhere primarily at discrete protrusions of the ventral membrane called podosomes

Rous sarcoma virus-transformed BHK cells (RSV/B4-BHK) adhere to a fibronectin-coated substratum primarily at specific dot-shaped sites. Such sites contain actin and vinculin and represent close contacts with the substratum as revealed by interference reflection microscopy. Only a few adhesion plaques and actin filament bundles can be detected in these cells as compared to untransformed parental fibroblasts. In thin sections examined with transmission electron microscopy (TEM) these adhesion sites correspond to short protrusions of the ventral cell surface that contact the substratum at their apical portion. These structures, which may represent cellular feet, are therefore called podosomes. By screening a number of different transformed fibroblasts plated on a fibronectin-coated substratum we find that podosomes are common to mammalian and avian cell lines transformed either by Rous sarcoma virus (RSV) or by Fujinami avian sarcoma virus (FSV), whose oncogenes encode specific tyrosine kinases. Using antibodies reacting with phosphotyrosine in immunofluorescence experiments, we show that phosphotyrosine-containing molecules are concentrated in podosomes. Podosomes are not detected in fibroblasts transformed by other retroviruses (Snyder-Theilen sarcoma virus, Abelson leukemia virus and Kirsten sarcoma virus) or by DNA tumor viruses (polyoma, SV40), indicating that podosome-mediated adhesion in transformed fibroblasts is related to the peculiar properties of some oncoproteins and possibly to their tropism for adhesion systems. Podosomes and adhesion plaques, although similar in cytoskeletal protein composition, have different mechanisms and kinetics of formation. Assembly of podosomes, in fact (i) does not require fetal calf serum (FCS) in the adhesion medium, that is necessary for the organization of adhesion plaques; (ii) does not require protein synthesis; and (iii) is insensitive to the ionophore monensin, that prevents adhesion plaque formation. Moreover, during attachment to fibronectin-coated dishes, podosomes appear in the initial phase (60 min) of attachment, while adhesion plaques require a minimum of 180 min. In conclusion podosomes of RSV- and FSV-transformed fibroblasts represent a phenotypic variant of adhesion structures.     

Fibronectin from chicken embryo fibroblasts contains covalently bound phosphate

Fibronectin isolated from cultures of chicken embryo fibroblasts (CEF) contains phosphorus linked to serine and threonine by monoester bonds. Normal and Rous sarcoma virus (RSV)-transformed cells were incubated with [32P]orthophosphate, and fibronectin was isolated from the cell surfaces and conditioned media. 32P was stably associated with fibronectin during immunoprecipitation, SDS-polyacrylamide gel electrophoresis, phospholipid solvent extraction, and hot acid but not alkaline treatment. After a limited acid hydrolysis of fibronectin, both phosphoserine and phosphothreonine were found. The specific radioactivity of the 32P-labeled fibronectin from the conditioned medium of normal CEF was higher than that from the cultures of transformed CEF.     

Loss of epidermal growth factor receptors and release of transforming growth factors do not correlate with sarcoma virus-transformation in clonally-related NIH/3T3-derived cell lines.

Transformation of NIH/3T3 cells by Kirsten murine sarcoma virus (MSV) caused a dramatic reduction in the number of cell-surface receptors for epidermal growth factor (EGF). However, the number of EGF receptors remained at a very low level in a non-tumourigenic revertant cell line isolated from the virus-transformed cells, indicating that an increase in EGF receptors is not a requirement for the phenotypic reversion of Kirsten MSV-transformed 3T3 cells. Serum-free conditioned medium from normal and virus-transformed cell lines contained similar amounts of cell growth-promoting activity as assayed by the ability to stimulate DNA synthesis in quiescent Swiss 3T3 cell cultures. However, the concentrated conditioned medium from these cell lines showed no evidence of beta-transforming growth factor (TGF) activity as assayed by promotion of anchorage-independent growth of untransformed normal rat kidney (NRK) fibroblasts in agarose. The cellular release of alpha-TGF activity was assayed by measuring the ability of concentrated conditioned medium to inhibit the binding of 125I-EGF to Swiss 3T3 cells. Conditioned medium protein from the virus-transformed cell line inhibited 125I-EGF binding but only to the same extent as conditioned medium protein prepared from the untransformed cell line. The alpha-TGF secretion by these cell lines was estimated to be 30-45-fold lower than the level of alpha-TGF released by a well-characterized alpha-TGF-producing cell line (3B11). These results suggest that the induction of TGF release is not a necessary event in the transformation of NIH/3T3 cells by Kirsten MSV.     

Density-dependent changes in hexose transport, glycolytic enzyme levels, and glycolytic rates, in uninfected and murine sarcoma virus-transformed rat kidney cells

In normal rat kidney (NRK) cell cultures, increased cell density results in a decrease in the rates of hexose transport, glucose utilization, and lactate production and an increase in the level of hexokinase activity. A murine sarcoma virus (Kirsten)-transformed cell line (KNRK) showed little or no density-dependent variation in sugar uptake, glucose consumption, or lactate production. On the other hand, hexokinase, phosphofructokinase, pyruvate kinase, glucose-6-phosphate dehydrogenase, and 6-phosphogluconate dehydrogenase activities were elevated in dense transformed cultures as compared to sparse or uninfected cultures. In another virus-transformed cell line (ts339/NRK) exhibiting temperature-dependent morphology, growth pattern, and transport of 2-deoxy- -glucose, the levels of glycolytic enzyme activity were related to cell density but not to the culture temperature. The lack of correlation between glycolytic enzyme activity and lactate production by either uninfected or murine sarcoma virus-transformed cultures supports the suggestion that enhanced growth and/or hexose transport capacity rather than elevated glycolytic enzyme activity are responsible for the increased rate of lactate production by virus-transformed NRK cells.     

Characterization of intra- and extracellular transforming growth factors from normal and avian sarcoma virus-transformed rat cells

Intra- and extracellular transforming growth factors (TGFs) have been characterized in an avian sarcoma virus-transformed rat cell line, 77N1, and the nontransformed parental cell line, NRK. Serum-free conditioned medium from 77N1 and cell extracts from NRK and 77N1 were subjected to ion exchange column chromatography. Intracellular TGFs in cell extracts of NRK and 77N1 cells showed a major peak of DNA synthesis-stimulating and colony-forming activities which eluted in the 0.10 to 0.15 M salt concentration region. Extracellular TGF in conditioned medium of 77N1 cells showed a major peak of activity which eluted at 0.05 to 0.06 M salt concentration. Furthermore partially purified extracellular TGF had a molecular weight of about 40,000 daltons, whereas that for intracellular TGF was about 12,000 daltons. These intra- and extracellular TGFs, as well as TGF gamma 2 which was purified from 77N1 cell extract, were heat- and acid-labile polypeptides sensitive to treatment with dithiothreitol. Radioimmunoprecipitation analyses with antiserum against TGF gamma 2 demonstrated that intra- and extracellular TGFs in 77N1 and NRK cells were immunologically identical or very closely related to each other and suggested the possibility that extracellular TGF from 77N1 cells was released into serum-free conditioned medium after formation of complex with other cellular components.     

Serum-free growth of normal and transformed fibroblasts in milk: differential requirements for fibronectin

Bovine milk may be used as a supplement for the serum-free growth of certain fibroblastic cells in culture. The growth properties of three representative cell types in milk-supplemented medium were examined; fibroblastic cell strains, fibroblastic cell lines, and transformed fibroblasts. Transformed fibroblasts, which included RNA and DNA tumor virus-transformed cells and carcinogen-transformed cells, grew in milk. Instead of growing attached to the culture dishes, as they normally do in serum, transformed fibroblasts grew in milk as large clusters in suspension. In contrast, nontransformed fibroblastic cell strains and cell lines did not grow in milk-supplemented medium. Fibroblasts transformed by a temperature-sensitive transformation mutant of Rous sarcoma virus were temperature-sensitive for growth in milk. The failure of cells to adhere to the substratum in milk-supplemented medium suggested that milk might be deficient in attachment factors for fibroblasts. When the attachment of fibroblastic cells in milk- supplemented medium was facilitated by pretreating culture dishes with fibronectin, (a) transformed cells grew attached rather than in suspension, (b) normal cell lines attached and grew to confluence, and (c) normal cell strains adhered and survived but did not exhibit appreciable cell proliferation.     

Reduced synthesis of pp60src and expression of the transformation-related phenotype in interferon-treated Rous sarcoma virus-transformed rat cells.

Treatment of Rous sarcoma virus-transformed rat cells with rat interferon-alpha (specific activity, 10(6) U/mg of protein) for 24 h caused a 50% reduction in intracellular pp60src-associated protein kinase activity. Staphylococcus aureus V8 protease digestion of pp60src, derived from 32P-labeled monolayer cultures incubated with or without interferon, revealed no differences either in the phosphopeptide pattern or in the phosphoserine-phosphotyrosine ratio. However, [3H]leucine pulse-labeling experiments showed that the synthesis of pp60src was reduced by 42 to 48%, relative to the level of bulk protein synthesis, in the interferon-treated cultures. Rat interferon-alpha also reduced the growth rate of Rous sarcoma virus-transformed rat cells in a dose-dependent manner over a 72-h period. The decrease in growth rate was accompanied by increases in the thickness and number of actin fibers per cell and by a decline in intracellular tyrosine phosphorylation by pp60src. The results suggest that interferon can inhibit the expression of the transformation-related phenotype by selectively reducing the synthesis of the Rous sarcoma virus transforming gene product. However, the interferon effects on the cytoskeletal organization and proliferation of Rous sarcoma virus-transformed cells may be due at least in part to the predominance of interferon-induced phenotypic changes over those caused by pp60src.     

Adhesion-promoting factor from embryonic fibroblasts for normal liver epithelial cells

In the present study we show that adhesion of normal rat liver epithelial cells (RL34) to substratum coated with type I collagen (collagen substratum) is promoted by a factor involved in 80% ammonium sulfate precipitated proteins from serum-free conditioned medium (PCM) of rat embryo fibroblasts. Adhesion of RL34 cells to collagen substratum was promoted dose dependently by whole PCM and the maximum effects on adhesion could be achieved by 200 micrograms/ml whole PCM. Kinetics studies with 100 micrograms/ml whole PCM showed that adhesion proceeded very slowly, taking 16 h to reach a plateau. Adhesion-promoting activity in whole PCM was sensitive to treatments with trypsin, acid, and heat but stable to dithiothreitol treatment. Further purification of whole PCM was performed using a combination of chromatography on blue Sepharose column, gel filtration column and heparin Sepharose column. The partially purified proteins, referred to as heparin PCM, are not bound or only weakly bound to heparin under physiological ion strength and pH, and the apparent molecular weight (Mr) range is estimated to be 40,000 to 60,000 from gel filtration chromatography and SDS-polyacrylamide gel electrophoresis. When whole PCM or heparin PCM was used for coating on plastic or collagen substratum, they no longer exerted the promoting activity.     

Activation of ornithine decarboxylase in monolayer cells treated with trypsin

When monolayer Chinese hamster cells are treated with trypsin for short periods of time, ornithine decarboxylase (ODCase) activity increases two- to fourfold. This increase can be blocked by aprotinin, a protease inhibitor, and is not observed when cultures are dislodged from substrate mechanically prior to contact with exogenous trypsin. The trypsin-induced increase in ornithine decarboxylase activity is not due to degradation of enzyme or inhibitor molecules or to new enzyme synthesis. Immunoprecipitable protein, radiolabeled with [3H]alpha-difluoromethylornithine in vitro, is the same molecular weight in cells harvested with or without trypsin. Protein-bound levels of this specific enzyme-activated irreversible inhibitor of ornithine decarboxylase are unchanged by trypsin treatments that increase enzyme activity. Trypsin treatment of rat embryonic fibroblasts, transformed by a temperature-sensitive mutant of Rous sarcoma virus, increases ODCase activity in cells growing at the nonpermissive, but not at the permissive, temperature for the transformed phenotype. These results suggest that ornithine decarboxylase can be activated by exogenous trypsin treatment in a manner that is dependent on cell adhesion properties, which are modified in transformed cells.     

Evidence that a target-derived soluble factor is necessary for the selective lysis of SV40-transformed fibroblasts by activated mouse macrophages

Our laboratory is investigating the basis for the selective recognition of transformed cells by activated mouse macrophages. As targets we are using a panel of SV40-transformed, C3H.OL fibroblast cell lines (SV-COL) that display widely different levels of sensitivity to lysis, from highly sensitive to completely resistant. Our results show that adding conditioned medium from the macrophage-sensitive target SV-COL-E8 (CM(E8] to a cytolysis assay with the macrophage-resistant target SV-COL-F5f causes the macrophages to kill the resistant targets in a contact dependent fashion. We have termed this activity "macrophage cell lysis factor" (MCLF). MCLF activity was not detected in conditioned media from cells not killed by activated macrophages (i.e., 3T3-like cell lines or embryo fibroblasts) but was present in conditioned media from six other SV-COL cell lines at levels that were directly proportional to the sensitivity of those targets (r = 0.98). These data suggest that MCLF plays a key role in determining the lytic sensitivity of SV40-transformed fibroblasts. Finally, to ask whether the production of MCLF is sufficient to explain the selective recognition of SV40-transformed fibroblasts by activated macrophages we have tested whether CM(E8) will cause macrophages to kill normal cells. Our results show that in the presence of CM(E8) macrophages will kill immortalized, 3T3-like fibroblasts but will not kill normal embryo fibroblasts. These results suggest that the production of MCLF, or a similar activity, is necessary but not sufficient for macrophage cytolysis to occur and that a change in target cell phenotype that occurs during the process of immortalization is also required.     

Comparison of adhesive bond strength of different cell types in vitro

Summary An established in vitro assay for quantitating cell-substratum adhesion has been utilized to measure the adhesiveness of 10 cell lines to a colloidal overlay. The procedure, a derivation of the William’s blister test for adhesives, involves growing cells to confluency on a polystyrene surface and then overlaying the monolayers with a Bacto-agar substratum. The cell-agar substratum systems are debonded and the^ra,adhesive bond strength, of the separation of the two interfaces calculated. The^ra’s were determined for the following cell types: SGL (gingival epithelial-like), L (transformed mouse fibroblasts), HeLa (human carcinoma), MDCK (canine kidney epithelial), WI-38 (human embryonic lung), Flow 1000 (human embryonic skin—muscle), Flow 4000 (human embryonic kidney), Flow 5000 (whole human embryo), BALB/c 3T3 (mouse fibroblasts) and SV₄₀-transformed BALB/c 3T3 (simian virus 40-transformed mouse fibroblasts). Transformed cells (L, HeLa and SV₄₀-transformed BALB/c 3T3) proved to be quantitatively less adhesive (^ra/cell) than either fibroblast or epithelial-like cell lines. Of the “normal” cells tested the kidney cells, human embryonic fibroblast and canine epithelial cells, and the gingival epithelial-like cells demonstrated a weaker binding to the colloidal overlay than the mouse fibroblasts (BALB/c 3T3), the human embryonic lung, the human embryonic skin-muscle, and the whole human embryo fibroblast cell lines. Concanavalin A increased the bonding strength of Flow 5000 cells after 24 hr incubation; however, the adhesiveness of the treated BALB/c 3T3 cells remained similar to the unterested samples while the^ra of the treated SV₄₀-transformed BALB/c 3T3 cells decreased. This research was supported by National Institute of Dental Research Grant DE03983.     

Quail embryo fibroblasts transformed by four v-myc-containing virus isolates show enhanced proliferation but are non tumorigenic.

Quail embryo fibroblasts infected with any of the four natural avian myc gene-containing virus strains (MC29, CMII, OK10 and MH2) or with the myb, ets-containing E26 acute leukemia virus, were examined for their expression of several transformation-associated parameters. All myc-containing viruses, but not E26 or Rous sarcoma virus (used as a control) induced a dramatic stimulation of cell proliferation. In addition, the myc virus-transformed cells exhibited prominent nucleoli, possibly as a consequence of their increased proliferation. Cells transformed by MC29, OK10, MH2 and E26 were capable of growing in semi-solid medium and showed a loss of actin cables and, in most cases, of an ordered fibronectin distribution. All of the myc virus-transformed fibroblasts, as well as the E26-transformed cells, were unable to form tumors in nude mice, indicating that the myc gene (and the myb/ets genes) are not sufficient for the induction of a fully malignant phenotype in avian fibroblasts.     

Demonstration of T antigens on the surface of cells transformed with primate polyoma viruses

Primate polyoma virus-transformed hamster, mouse, and rat cell lines were examined by indirect immunofluorescence staining for cell surface-associated T antigens, by using a rabbit antiserum prepared against sodium dodecyl sulfate-denatured large T antigen of simian virus 40 (anti-SV40-SDS-T serum). Positive surface staining was shown not only on SV40-transformed cells, but also on BK and JC virus-transformed cells. In contrast, normal cells and cells transformed with mouse polyoma-, human adeno-, and murine sarcoma viruses were negative. The data on SV40-transformed cells confirmed the reports of others demonstrating the cell surface location of SV40 large T antigen, and the data on BK and JC virus-transformed cells proved that these cells have cell-surface T antigens that cross-react with anti-SV40-SDS-T serum.     

Fetal rat lung fibroblasts produce a TGF beta homolog that blocks alveolar type II cell maturation

Normal growth and differentiation of the lung depends upon mesenchymal-epithelial interactions during development. Recombination experiments using immature (Day 17) and mature (Day 21) fetal rat lung fibroblasts (FRLF) revealed that the stimulatory effect of mature fibroblasts on fetal type II epithelial cells is blocked by immature fibroblasts. Similarly, conditioned medium from Day 17 FRLFs blocks the stimulatory effect (fibroblast-pneumonocyte factor) of Day 21 conditioned medium on type II epithelial cells. This blocking activity is nondialyzable, trypsin sensitive, and heat stable. Its activity is neutralized by an antibody to TGF beta, in both conditioned media and recombined cell studies, and its activity is mimicked by TGF beta. Developmentally, TGF beta-like activity is present in conditioned medium from 15- to 19-day FRLF, decreasing precipitously between 19 and 21 days gestation. Northern blot analysis of mRNAs from fetal rat lung fibroblasts on Days 17, 19, and 21 revealed expression of TGF beta at all three stages of development.     

Coordinate regulation of the levels of type III and type I collagen mRNA in most but not all mouse fibroblasts

A mouse genomic clone was isolated by cross-hybridization with a DNA fragment which codes for the NH2-propeptide of chick alpha1(III) collagen. The region of cross-hybridization within the mouse clone was localized, its sequence determined, and an exon coding for the NH2-propeptide of mouse alpha1(III) collagen was identified. This DNA fragment hybridizes to an RNA species of approximately 5300 nucleotides, slightly larger than the major alpha2(I) collagen RNA species. The mouse type III collagen probe was used to examine the effect of transformation on alpha1(III) collagen RNA levels in mouse fibroblasts. The levels of type III and type I collagen mRNA levels were compared in control and sarcoma virus-transformed murine cell lines, as well as in NIH 3T3 cells transformed by members of the human ras oncogenes. The levels of type III RNA decreased about 10-15-fold in Moloney sarcoma virus-transformed cells and in a cell line transformed with a v-mos-containing plasmid, but showed only a 50% decrease in a Kirsten murine sarcoma virus-transformed BALB 3T3 cell line, and increased 4-fold in a Rous sarcoma virus (RSV)-transformed BALB 3T3 cell line. In contrast, the levels of alpha2(I) collagen mRNA are 8- to 10-fold lower in all these cell lines when compared to untransformed cells. NIH 3T3 cells transformed with two human ras oncogenes showed decreased levels of alpha2(I) and alpha1(III) mRNAs. In contrast to the RSV-transformed mouse cell line, RSV-transformed chick embryo fibroblasts contained much smaller amounts of type III RNA than control chick embryo fibroblasts. We conclude that the levels of alpha1(III) and alpha2(I) collagen mRNA are often but not necessarily coordinately regulated by transformation in mouse cells.     

Conditions leading to the establishment of the N (a gene dependent) and A (a gene independent) transformed states after polyoma virus infection of rat fibroblasts.

Infection of normal rat fibroblasts (FR 3T3) with the early tsa mutant of polyoma virus may lead to either the A or the N phenotype, tsa-A transformants, originally derived by agar selection, are not temperature dependent for maintenance of the transformed phenotype, whereas tsa-N transormants revert at high temperature to normal growth control. A transformants did not result from an independent cellular mutation selected in agar medium, but rather from a transformation process distinct from that leading to the N state. It occurred in both liquid and agar media when the infected cells were maintained under growth-restricting conditions, such as absence of anchorage and contact inhibition at confluency. N transformation occurred in cells maintained in active growth after virus infection (sparse cultures on a solid substratum). Physiological conditions during a critical period after virus infection thus appear to be a crucial parameter of the transformation process.     

Neurite outgrowth induced by the substrate associated material from nonneuronal cells

Cultured embryonic heart cells release a powerful inducer of neurite outgrowth into the surrounding medium. The present report demonstrates that these cells also deposit material which induces neurite outgrowth directly onto their culture substratum. Thus, embryonic heart cells condition both the culture medium and the culture substratum with respect to neurite outgrowth. Conditioned substrata were prepared by incubating heart cell monolayers in EDTA until the cells released from the substratum and were discarded. When dissociated neurons from ciliary or sympathetic chain ganglia were plated in fresh medium onto a conditioned substratum, neurite outgrowth was initiated in 80–95% of the neurons within 60 min. The neurite-inducing activity is trypsin sensitive, but is not inactivated by antibodies to the cell attachment protein fibronectin, by the membrane-solubilizing detergent Triton X-100, or by the enzymes collagenase, RNase, or DNase. The factor in conditioned medium which also induces neurite outgrowth depends for its activity on attachment to an artificial polyornithine substratum, under which condition it appears to promote adhesion of neuronal filopodia to the substratum. Thus, neurite outgrowth in these two culture systems occurs only if the substratum is conditioned by the appropriate extracellular materials: conditioned either directly by the deposition of heart cell products or indirectly by the binding of a conditioned medium factor to the polyornithine substratum. These substratum-conditioning factors may be related to those components of the extracellular matrix which support neurite outgrowth in vivo.     

Effects of acidified fetal bovine serum on the fibrinolytic activity and growth of cells in culture.

The fibrinolytic activity of cells in culture varied with the type of serum employed in the growth medium. Degradation of iodinated fibrin occurred slowly when Rous sarcoma virus-transformed chick embryo fibroblasts were grown in medium containing fetal bovine serum (FBS), and rapidly when chicken serum was employed. This difference reflected the low plasminogen and high inhibitor content of FBS. The inhibitors were found to be serum macromolecules that were precipitated with ammonium sulfate or polyethylene glycol, and were inactivated by boiling or upon exposure to acidic conditions. No inhibitor activity was detected in fetuin, one of the major proteins present in FBS. Acidified FBS was similar to chicken serum in that both supported high rates of cell-mediated fibrinolytic activity. Although virally transformed hamster, mouse and chicken cells grew well in acid-treated FBS, their normal counterparts did not. Apparently, acifification resulted in the formation of materials that were toxic to normal cells. These agents rapidly blocked cellular DNA synthesis.     

Proteins secreted by Caco-2 cells support growth of other tumor cells in Chee's Essential Medium without supplements.

The growth promoting effect of several hormones and growth factors on two human colon tumor cell lines (Caco-2 and SW 48) was studied using six different chemically defined serum-free media (SFM). Caco-2 grew in a simple SFM [GF3: Chee's Essential Medium (CEM) plus insulin, transferrin and selenium], whereas, SW 48 cells did not grow in GF3 medium. This suggested that Caco-2 cells probably secrete proteins in SFM which influence attachment and growth of Caco-2 and other tumor cells. Lyophilized Caco-2 conditioned medium and substratum, when added to plain CEM, supported growth of SW 48 and SW 948 cells. The substratum material was more effective than conditioned medium in promoting growth of the cell lines. The substratum material helps attachment and spreading of the cells and, thus, improves growth of the cells over conditioned medium. Caco-2 conditioned medium and substratum were analyzed for their components using SDS-PAGE system and gel filtration chromatography. The substratum was analyzed for the presence of fibronectin and laminin by the ELISA technique. The conditioned medium does not contain TGF alpha and TGF beta. The growth stimulating activity of the conditioned medium is due to a protein component, approximately 58Kd in size.