Relationship between leukemogenicity and in vivo inducibility of normal differentiation in mouse myeloid leukemia cells.

Leukemogenicity was studied in sensitive mouse myeloid leukemia cells that could be induced to undergo cell differentiation in vitro into mature granulocytes and macrophages by incubation with the inducer (certain proteins or glucocorticoids) and in resistant myeloid leukemia cells that could not be induced to differentiate into mature cells. Three sensitive and five resistant clones were tested. The resistant cells were much more leukemogenic than the sensitive cells, and the survival time of syngeneic mice inoculated with the sensitive cells. In a diffusion chamber in a syngeneic, inbred SL mouse without any additional manipulation or injection of inducer, the resistant cells remained undifferentiated, but the sensitive cells were induced to differentiate into mature granulocytes and macrophages and their proliferation rate decreased. These results suggested that the greater leukemogenicity of resistant cells is associated with some defect in inducibility of cell differentiation.

Correlation between the carcinogenicities of nitrofuran derivatives and their destructive actions on sebaceous glands of mouse skin.

The effects of six nitrofuran derivatives (including a formerly used food preservative) on mouse skin sebaceous glands were investigated. A close correlation was found between the carcinogenicities and destructive activities of nitrofuran derivatives on the sebaceous glands. 5-Nitro-2-furaldehyde semicarbazone and 4-methyl-1-[(5-nitrofurfurylidene)amino]-2-imidazolidinone, which are carcinogenic, caused marked destruction of the glands at a dose of 1-5 mg/mouse. 2-(5-Nitro-2-furfurylidene)-aminoethanol almost completely destroyed the glands at a dose of 5 mg/mouse; its carcinogenicity has not yet been investigated. 1-[(5-Nitrofurfurylidene)amino]-carcinogenic, did not affect the glands, even at a dose of 5 mg/mouse. 2-(2-Furyl)-3-(5-nitro-2-furyl)acrylamide, which is a potent mutagen but not carcinogenic, had no effect on the glands at a dose of 5 mg/mouse. Under similar conditions, the potent carcinogen 7,12-dimethylbenz(alpha)athracene almost completely destroyed the sebaceous glands at a dose of 0.05 mg/mouse, but dimethyl sulfoxide (used as solvent for the test compounds) had no effect.

Effects of retinoids on induction of differentiation of cultured mouse myeloid leukemia cells.

Retinoic acid, retinol, retinyl acetate, and retinal induced activities of lysosomal enzymes, such as lysozyme, acid protease, and acid phosphatase, in mouse myeloid leukemia cells (M1), while the pyridyl analog of retinoic acid had no effect. Retinoic acid was the most potent inducer of lysosomal enzyme activities. The induction of lysozyme activity by retinoic acid was inhibited by treatment with puromycin. The retinoids did not induce phagocytic and locomotive activities or morphological changes in M1 cells, and they inhibited the induction of these differentiation-associated properties by various inducers without inhibiting cell growth. Retinoic acid was the most potent inhibitor of induction of these differentiation-associated properties. The inhibitory effect of retinoic acid was found to be reversible. These results suggest that distinct mechanisms exist for control of induction of lysosomal enzyme activities and of other differentiation-associated properties of M1 cells, such as phagocytosis, morphological changes, and migration.

Stimulation of differentiation of mouse myeloid leukemic cells and induction of interferon in the cells by double-stranded polyribonucleotides.

Mouse myeloid leukemic MI cells can be induced to differentiate into mature macrophages and granulocytes by differentiation-stimulating factor (D-factor) in conditioned medium of mouse peritoneal macrophages. Double-stranded RNA's, such as the copolymers of polyinosinic and polycytidylic acids and polyadenylic and polyuridylic acids, could not alone induce differentiation of the cells, but enhanced induction of differentiation by low concentrations of the D-factor and induced a significant amount of interferon. Rabbit antiserum to purified L-cell interferon neutralized the antiviral activity of interferon of MI cells. Simultaneous treatment of MI cells with the anti-interferon serum and copolymer of polyinosine and polycytidylic acids and D-factor abolished the enhancing effect of copolymer of polyinosine and polycytidylic acids on the action of the D-factor. These results suggest that the effect of double-stranded RNA's on induction of differentiation of MI cells is mediated by interferon produced by the cells.

Production by mouse spleen cells of factors stimulating differentiation of mouse myeloid leukemic cells that differ from the colony-stimulating factor.

Mouse myeloid leukemic M1 cells can be induced to differentiate into macrophages and granulocytes in vitro by a protein inducer, differentiation-stimulating factor (D-factor), and by various other compounds. Mouse spleen cells produced D-factors when treated with various mitogens, such as concanavalin A, phytohemagglutinin, pokeweed mitogen, lipopolysaccharide, and synthetic double-stranded polyribonucleotide copolymer of polyinosinic and polycytidylic acids. Concanavalin A, phytohemagglutinin, and pokeweed mitogen stimulated spleen lymphocytes, but not spleen macrophages, to produce a D-factor with an apparent molecular weight of 40,000 to 50,000. On the other hand, lipopolysaccharide and copolymer of polyinosinic and polycytidylic acids stimulated both spleen lymphocytes and spleen macrophages to produce D-factors. Spleen macrophages produced D-factors with molecular weights of 40,000 to 50,000 and 20,000 to 25,000, whereas spleen lymphocytes produced only the larger molecules. In addition to D-factor, colony-stimulating factor (CSF), which stimulates growth and differentiation of normal bone marrow cells, and interferon, were detected in conditioned medium of spleen cells treated with concanavalin A or lipopolysaccharide. On gel filtration of the conditioned medium with Sephadex G-100, CSF was eluted between the larger D-factor and the smaller one. The fraction with interferon activity overlapped that of the larger D-factor. Incubation of the conditioned medium at pH 2 abolished the activity of interferon but did not affect the activity of either D-factor or CSF. The addition of cytochalasin B suppressed the production of interferon but not of D-factor or CSF by the spleen cells. These results indicate that the D-factor is a different substance from CSF or type II interferon.

Inhibition of differentiation of cultured mouse myeloid leukemia cells by nonsteroidal antiinflammatory agents and counteraction of the inhibition by prostaglandin E1.

Mouse myeloid leukemia cells (M1) were induced to differentiate into mature macrophages and granulocytes by glucocorticoids or a protein inducer in ascitic fluid from tumor-bearing rats. Addition of nonsteroidal antiinflammatory agents to M1 cells in suspension cultures inhibited the induction of differentiation by glucocorticoid (dexamethasone) or the protein inducer. The inhibition was unrelated to cytotoxicity and was reversible. The nonsteroidal antiinflammatory agent indomethacin inhibited dexamethasone-induced differentiation only when added before the time of commitment of the cells to differentiation. The indomethacin-mediated inhibition was counteracted by prostaglandins E1 or E2 but not by prostaglandins F1alpha or F2alpha. Prostaglandin E stimulated phagocytosis induced by a suboptimal concentration of dexamethasone, but prostaglandin F did not. Moreover, lysozyme activity, which is a typical biochemical marker of macrophages, was induced in M1 cells by prostaglandin E alone, as well as by inducers of differentiation. These results suggest that prostaglandin E may be important in the induction of differentiation of myelod leukemia cells.

Induction of differentiation of Rauscher virus-induced mouse myeloid leukemia cells with a factor(s) in ascitic fluid and inhibitors of nucleic acid and protein syntheses.

Cell line R453, established from a Rauscher virus-induced myeloid leukemia in a C57BL/6 mouse, was induced to differentiate in vitro into macrophages and granulocytes with ascitic fluids from animals bearing various ascites tumors or from mice treated with complete Freund's adjuvant, conditioned media from various cell lines, and glucocorticoid hormone. Differentiated R453 cells had a morphology similar to that of macrophages and granulocytes in normal hematopoietic organs, and they phagocytized small paricles such as latex particles, moved in soft agar showing locomotive activity, and had Fc and C3 receptors on the cell surface. This induction of differentiation of R453 cells was markedly enhanced by addition of inhibitors of RNA synthesis (actinomycin D, nogalamycin, or chromomycin A3), protein synthesis (puromycin or cycloheximide), or DNA synthesis (methotrexate, hydroxyurea, 5-fluorodeoxyuridine, or 1-beta-D-arabinofuranosylcytosine) in the presence of ascitic fluid. Of the inhibitors, actinomycin D was the most effective at a low concentration (5 ng/ml) in stimulating induction of differentiation of R453 cells. However, these inhibitors alone did not induce differentiation of R453 cells. The factor(s) in ascitic fluid that stimulates differentiation of R453 cells was heat labile, nondialyzable, and inactivated by trypsin.

Inhibition by interleukin 4 of leukemia inhibitory factor-, interleukin 6-, and dexamethasone-induced differentiation of mouse myeloid leukemia cells: role of c-myc and junB proto-oncogenes.

Interleukin 4 (IL-4) inhibited the differentiation of mouse myeloid leukemia M1 cells induced by leukemia inhibitory factor (LIF), interleukin 6, or dexamethasone and conversely enhanced the induction of M1 cell differentiation by 1 alpha,25-dihydroxyvitamin D3. IL-4 blocked LIF-induced differentiation of M1 cells when it was added to the culture medium within 10 h after LIF, but IL-4 did not block differentiation when it was added 12 h after LIF. These results indicate that IL-4 inhibited a critical intermediate step in myeloid leukemia cell differentiation. LIF markedly stimulated the expression of junB mRNA within 2 h but suppressed the expression of c-myb and c-myc after 2- and 12-h treatment, respectively. IL-4 did not significantly affect LIF-induced junB expression or suppression of c-myb expression. However, it interfered significantly with the LIF-induced suppression of c-myc gene expression. Similar results were obtained when interleukin 6 was used to induce differentiation of M1 cells. Dexamethasone and 1 alpha,25-dihydroxyvitamin D3 did not induce junB gene expression but suppressed the expression of c-myb and c-myc. IL-4 also interfered with dexamethasone-induced suppression of c-myc gene expression. On the other hand, IL-4 enhanced 1 alpha,25-dihydroxyvitamin D3-induced down-regulation of c-myc gene expression, consistent with its enhancement of differentiation. These results indicate that the change in c-myc expression induced by IL-4 in M1 cells is closely associated with the effect of IL-4 on the induction of differentiation of M1 cells.

Production of differentiation-inhibiting factor in cultured mouse myeloid leukemia cells treated with retinoic acid.

Mouse myeloid leukemia cells (M1) could be induced to differentiate into mature macrophages and granulocytes with dexamethasone or proteinaceous inducer. Retinoic acid inhibited functional and morphological differentiation of M1 cells, but the pyridyl analog of retinoic acid had no effect. M1 cells could be induced to produce a factor(s) inhibiting their own differentiation to macrophage- and granulocyte-like cells by retinoic acid but not by its pyridyl analog. This factor(s) inhibited induction by inducers of phagocytic activity, locomotive activity, lysozyme activity, and morphological changes in M1 cells. The production of the inhibitory factor(s) by M1 cells incubated with retinoic acid was inhibited by a low concentrations (5--10 ng/ml) of actinomycin D. The inhibitory factor seemed to be a protein(s), since it was susceptible to heat treatment and proteases. The effect of retinoic acid in inducing production of the inhibitory factor(s) by M1 cells seemed to be reversible, since it was low on washing the cells with fresh medium. Therefore, induction of this inhibitory factor may be involved in the mechanism of inhibition of functional and morphological differentiation of M1 cells by retinoic acid.

Effects of histone fractions on induction of differentiation of cultured mouse myeloid leukemia cells.

Mouse myeloid leukemic cells (M1) could be induced to differentiate into macrophage-like and granulocyte-like cells by a lysine-rich, histone H1 fraction (10 to 100 microgram/ml). The differentiated M1 cells expressed phagocytic and lysozyme activity and were macrophage-like and granulocyte-like cells. The differentiation-inducing activity of histone H1 was found in histone H1 fractions isolated from calf thymus, rat liver, and mouse leukemia M1 cells. Histone H2A and H2B fractions did not induce differentiation of M1 cells at concentrations of 10 to 100 microgram/ml but did induce differentiation at a high concentration (200 microgram/ml). The histone H3 fraction, poly-L-lysine and poly-L-arginine, inhibited induction of differentiation of M1 cells.

Characterization of a differentiation-inhibitory activity from nondifferentiating mouse myeloid leukemia cells.

Mouse myeloid leukemic M1 cells are induced to differentiate by various differentiation inducers. Activity for inhibition of induction of differentiation of M1 cells (I-activity) has been found in conditioned medium of variant M1 cell clones resistant to differentiation inducers, and this I-activity has been shown to be closely associated with resistance of the cells to differentiation inducers. In this work, the I-activity in the conditioned medium of the resistant M1 cells was shown to bind to Carboxymethyl-Sepharose CL-6B and to be eluted with 0.27-0.4 M NaCl. The profile of gel filtration of I-activity from Sephadex G-200 indicated considerable heterogeneity in molecules with I-activity; the apparent molecular range of the main I-activity was 60,000-80,000. On chromatofocusing, the I-activity was eluted with Polybuffer 96-acetic acid at pH 8.8-9.0. The I-activity was inactivated by treatment with trypsin or by heating at 75 degrees C for 30 min. Therefore, the main I-activity seemed to be due to a basic protein(s).

Prolongation of survival time of mice inoculated with myeloid leukemia cells by inducers of normal differentiation.

Studies were made on the effects of inducers on the leukemogenicity of sensitive mouse myeloid leukemia cells (M1) that could be induced to undergo cell differentiation into mature granulocytes and macrophages in vitro by incubation with inducers (certain proteins, bacterial lipopolysaccharides, or glucocorticoids) and of resistant M1 cells that could not be induced to differentiate into mature cells. Inducers of cell differentiation significantly enhanced the survival times of mice inoculated with sensitive cells but scarcely affected the survival times of mice inoculated with resistant cells. Some mice inoculated with the sensitive cells and treated with lipopolysaccharide did not develop leukemia. The sensitive and resistant clone cells contained similar common tumor-related surface antigens. Treatment with lipopolysaccharide was also effective in athymic nude mice inoculated with the sensitive M1 cells. Lipopolysaccharide or glucocorticoid significantly stimulated differentiation of the sensitive cells cultured in a diffusion chamber in vivo but had little effect on differentiation of resistant cells. These results suggest the possibility of treating, with partial success, leukemia in vivo with differentiation inducers.

Expression of a cell surface glycoprotein (p180) related to cell-substratum adhesion during differentiation of mouse myeloid leukemia cells.

Mouse myeloid leukemia M1 cells were induced to differentiate in vitro into macrophages and granulocytes by various inducers including ascitic fluid. Differentiated M1 cells induced with ascitic fluid expressed a differentiation-associated cell surface glycoprotein with a molecular weight of 180,000 (p180), which can be labeled by lactoperoxidase-catalyzed radioiodination or metabolic labeling with L-[14C]fucose. p180 was also induced by treatment with conditioned medium of hamster embryo cells, dexamethasone, dibutyryl cyclic adenosine 3':5'-monophosphate, and prostaglandin E1. Ascitic fluid, conditioned medium of hamster embryo cells, and dexamethasone induced all the differentiation-associated properties tested, whereas dibutyryl cyclic adenosine 3':5'-monophosphate and prostaglandin E1 induced lysozyme activity and adhesiveness to the substratum but not phagocytosis, locomotive activity, Fc receptors, or morphological changes. The adherent cells induced by dibutyryl cyclic adenosine 3':5'-monophosphate produced a large amount of p180, while the floating cells produced very little, but no difference was detected in the lysozyme activities of the two cell types. These results suggest that p180 is associated with cell-substratum adhesion of differentiated M1 cells.

Enhancement by hemin of the sensitivity of K562 human leukemic cells to 1-beta-D-arabinofuranosylcytosine.

The sensitivity of human leukemia K562 cells to cancer chemotherapeutic drugs during induction of erythroid differentiation of the cells by hemin was examined. Treatment with hemin greatly increased the sensitivity of the cells to 1-beta-D-arabinofuranosylcytosine (ara-C) but did not affect their sensitivities to other chemotherapeutic drugs, including Adriamycin, daunomycin, hydroxyurea, methotrexate, and vincristine. Thymidine and deoxyguanosine, which are known to potentiate the antileukemic effects of ara-C in K562 cells, also induced erythroid differentiation of K562 cells, but other inducers, such as sodium butyrate and delta-aminolevulinic acid, did not increase the sensitivity of K562 cells to ara-C. Hemin did not enhance the sensitivity to ara-C of other leukemia cell lines (Friend erythroleukemic cells, myeloid leukemic M1 cells, and promyelocytic leukemia HL-60 cells). These results indicate that some inducers of erythroid differentiation of K562 cells potentiate the antileukemic effect of ara-C on K562 cells.

Induction of erythroid differentiation of K562 human leukemic cells by herbimycin A, an inhibitor of tyrosine kinase activity.

Herbimycin A, a benzoquinonoid ansamycin antibiotic, is found to reduce intracellular phosphorylation by tyrosine protein kinase. The human chronic myelogenous leukemia cell line K562 expresses a structurally altered c-abl protein with tyrosine kinase activity. When K562 cells are induced to undergo erythroid differentiation by hemin, reduction in the intracellular level of tyrosine phosphorylation occurs. In order to understand the relationship between induction of differentiation and reduction of tyrosine phosphorylation by the c-abl gene product, the effect that herbimycin A, a selective inhibitor of intracellular tyrosine kinase activity, exerts on the differentiation of K562 cells was examined. Reduction of tyrosine phosphorylation in K562 cells by herbimycin A was observed within 1 h. Noncytotoxic concentrations of herbimycin A induced erythroid differentiation of K562 cells but not of murine erythroleukemia 745A cells. The other human myeloid leukemia cell lines (HL-60, THP-1, and U937) tested were not induced to undergo cell differentiation by this antibiotic. Herbimycin A and the other well-known inducers such as hemin, butyric acid, Adriamycin, and 1-beta-D-arabinofuranosylcytosine had additive or more than additive effects on induction of erythroid differentiation of K562 cells. With respect to inhibition of cell growth, the sensitivity of K562 cells to herbimycin A was highest in the human leukemia cell lines we tested. Noncytotoxic concentrations of herbimycin enhanced the antiproliferative effect of Adriamycin or 1-beta-D-arabinofuranosylcytosine on K562 cells. Combination therapy with herbimycin A and its derivatives may be considered for use in the treatment of some types of leukemia where tyrosine kinase activities are implicated as determinants of the oncogenic state.

Control of proliferating potential of myeloid leukemia cells during long-term treatment with vitamin D3 analogues and other differentiation inducers in combination with antileukemic drugs: in vitro and in vivo studies.

Growth inhibition of murine and human myeloid leukemia cells by differentiation inducers during long-term culture was examined to improve the strategy for therapy of myeloid leukemia by differentiation inducers. When the effect of 1 alpha,25-dihydroxyvitamin D3, a typical differentiation inducer, on proliferation of mouse myeloid leukemia M1 cells was examined at a constant product of time and concentration (480 nM in 20 days), the continuous treatment with 24 nM 1 alpha,25-dihydroxyvitamin D3 was the most effective for inhibition of cell proliferation. After 20 days, the cumulative cell number was reduced about 3 X 10(5) times by continuous treatment with 24 nM 1 alpha,25-dihydroxyvitamin D3. Similar results were obtained when M1 cells were treated continuously with dexamethasone. M1 cells resistant to 1 alpha,25-dihydroxyvitamin D3 appeared about 25 days after the start of continuous treatment with 24 nM 1 alpha,25-dihydroxyvitamin D3. On the other hand, when M1 cells were treated continuously with 1 alpha,25-dihydroxyvitamin D3 and noncytotoxic doses of antileukemic drugs such as 1-beta-D-arabinofuranosylcytosine and daunomycin, resistant cells did not appear for at least 35 days. A similar effect of 1 alpha,25-dihydroxyvitamin D3 and antileukemic drugs on cell proliferation was observed with the human monoblast-like cell line U937. The survival of syngeneic SL mice inoculated with M1 cells was prolonged more by treatment with both 1 alpha-hydroxyvitamin D3 and daunomycin than by treatment with either drug alone. These results suggest that continuous treatment with both differentiation inducers and certain antileukemic drugs may be more effective therapeutically than treatment with a differentiation inducer alone.

Hemin enhances the sensitivity of erythroleukemia cells to 1-beta-D-arabinofuranosylcytosine by both activation of deoxycytidine kinase and reduction of cytidine deaminase activity.

The sensitivity of human myelogenous leukemia cells to 1-beta-D-arabinofuranosylcytosine (ara-C) during induction of differentiation was examined. Treatment with hemin greatly increased the sensitivity of erythroid leukemia cells to ara-C. The enhancement of ara-C sensitivity by hemin was not as remarkable in nonerythroid leukemia cells. Hemin altered the metabolism of ara-C in human erythroleukemia K562 cells by reducing ara-C deaminase activity, increasing intracellular accumulation of ara-C, and activating the nucleoside kinases. These alterations may be involved in the enhancing effect of hemin on sensitivity of ara-C. These results suggest that some inducers of differentiation potentiate the antileukemic effect of ara-C on human erythroleukemia cells.

Inhibition by erythroid differentiation factor (activin A) of P-glycoprotein expression in multidrug-resistant human K562 erythroleukemia cells.

The K562/VCR cell line, exhibiting acquired multidrug resistance (MDR) with increased expression of a cell surface glycoprotein (P-glycoprotein), was isolated from human erythroleukemia K562 cells. Various compounds that induced erythroid differentiation of K562 cells were tested for their effects on growth and differentiation of these K562/VCR cells. Sodium butyrate, hemin, 1-beta-D-arabinofuranosylcytosine, and erythroid differentiation factor (EDF) induced erythroid differentiation of K562/VCR cells as well as K562 cells. The MDR of K562/VCR cells was partly overcome by treatment with EDF but not with the other inducers. Expression of P-glycoprotein by K562/VCR cells was measured by radioimmunoassay using MRK16 monoclonal antibody. Results showed that EDF caused down-regulation of P-glycoprotein in K562/VCR cells, whereas the other inducers did not cause its down-regulation. Thus, in addition to inducing erythroid differentiation, EDF enhanced the sensitivity of K562/VCR cells to multidrugs and suppressed expression of P-glycoprotein. These results suggest that differentiation inducers may be useful in chemotherapy of leukemic MDR cells.

Increased synthesis of hyaluronic acid by mouse mammary carcinoma cell variants with high metastatic potential.

Variant subpopulations of FM3A mouse mammary carcinoma cells that have increased lung-colonizing potential were obtained previously by sequentially harvesting pulmonary metastases, culturing their cells in vitro, and reestablishing the metastases in vivo. In the present study, glycosaminoglycan production by the parental and variant cells was studied after metabolic labeling of cultures by [14C]glucosamine for 24 hr. Analysis of the products indicated that the rate of incorporation of the labeled precursor into hyaluronic acid in the high-metastatic variant cells was 27 to 54 times the rate in the low-metastatic variant cells and that the increase in hyaluronic acid synthesis was not associated with an increase in the rate of synthesis of other glycosaminoglycans. Both the cell layers and media of high-metastatic variants contained a much higher proportion of radioactivity in hyaluronic acid than did the corresponding fractions of low-metastatic cell lines. The results provide a basis for further investigation of the potential role of hyaluronic acid in control of the behavior of epithelial tumor cells during metastasis.

Effect of tunicamycin on production by mouse fibroblast L929 cells of the factor-stimulating differentiation of mouse myeloid leukemic cells and the colony-stimulating factor.

Mouse myeloid leukemic M1 cells can be induced to differentiate into macrophages and granulocytes in vitro by a factor(s) stimulating differentiation of the cells (D-factor), which is suggested to be a glycoprotein. On the other hand, growth and differentiation of normal precursor cells of macrophages and granulocytes can be stimulated by a glycoprotein termed colony-stimulating factor (CSF). Mouse fibroblast L929 cells were found to produce both the D-factor and CSF. The properties of the D-factor and CSF and the roles of carbohydrates in the molecules of these factors were examined using tunicamycin, a specific inhibitor of asparaginase-linked glycosylation. Although both the D-factor and CSF were produced by L-cells in usual medium containing fetal calf serum, production of D-factor, but not CSF, was reduced by omission of serum from the medium. The activity of the D-factor was slightly decreased by treating the L-cells with tunicamycin (0.5 microgram/ml) in the presence of 2% fetal calf serum, without any decrease in CSF activity. Conditioned medium of L-cells incubated with or without tunicamycin was fractionated by gel filtration on a Sephadex G-200 column. Normal D-factor appeared as a single peak with an apparent molecular weight of 67,000. D-factor produced in the presence of tunicamycin had an apparent molecular weight of 25,000. On the other hand, most of the CSF was eluted in the void volume, even when it was produced in the presence of tunicamycin. The D-factor produced in the presence of tunicamycin was more sensitive than normal D-factor was to trypsin or heat treatment at 70 degrees. The CSF produced in the presence of tunicamycin was resistant to these treatments. These results indicate that the D-factor is distinct from CSF. Furthermore, the results suggest that the D-factor produced by L-cells is also a glycoprotein and that, although carbohydrate is not essential for production or activity of the D-factor, it contributes to stabilizing the protein portion of D-factor.