Purification of a Novel Growth Inhibitory Factor for Partially Differentiated Myeloid Leukemic Cells*

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY D 1988 by The American Society for Biochemistry and Molecular Biology, lne. Vol.263, No. 11, Issue of April 15, pp , 1988 Printed in U.S.A. Purification of a Novel Growth Inhibitory Factor for Partially Differentiated Myeloid Leukemic Cells* (Received for publication, October 5, 1987) Takashi Kasukabe, Junko Okabe-Kado, Yoshio Honma, and Motoo Hozumi From the Department of Chemotherapy, Saitama Cancer Center Research Institute, Ina-mmhi, Saitama-362, Japan A novel factor termed growth inhibitory (GI) factor, which specifically inhibits the growth of mouse monocytic leukemia cells including monocytic cell lines (Mm-A and ) and other partially differentiated myeloid leukemic cells, has been purified from conditioned medium of some clones of mouse myeloblastic leukemia M1 cells. The procedure for purification of the GI factor included ammonium sulfate precipitation, CM-Sepharose CL-GB and Sephadex G-200 chromatographies, reverse-phase high-performance liquid Chromatography on a CIS hydrophobicsupport,and high-performance liquid chromatography on a gel filtration column. The purified factor gave a single band of protein with a molecular weight 25,000 of on sodium dodecyl sulfate-polyacrylamide gel. A concentration of 8 X 10"O M GI factor was required for 50% inhibition of growth of Mm-A cells. On chromatofocusing, the GI activity was eluted with Polybuffer 96/acetic acid at The mouse monocytic Mm-A cell line (1) is a highly leukemogenic variant cell line of the monocytic and nonleukemic Mm-1 cell line (2), which developed spontaneously from mouse myeloid leukemia M1 cells (3). Mm-A and Mm-1 cells adhere to culture dishes, have high phagocytic activity, and synthesize lysozyme, whereas M1 cells are myeloblastic and have none of these markers of functional differentiation. We have used Mm-A cells as a model in studies on growth and differentiation of leukemia cells in the intermediate stage of differentiation. We reported previously that the sensitivity of Mm-A cells to inducers of differentiation differed from that of the parent M1 cells and suggested that the responses of leukemic cells to inducers of differentiation might depend on the stage of arrest of differentiation of the cells (4). Therefore, we tried to find more specific inducers of differentiation or inhibitors of proliferation of these partially differentiated leukemic cells. * This work was supported in part by a grant for Cancer Research from the Ministry of Education, Science, and Culture of Japan, and a grant-in-aid from the Ministry of Health and Welfare for a Comprehensive 10-year Strategy for Cancer Control, Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertkement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Recently, we found a growth inhibitory factor (GI factor)' for Mm-A cells in conditioned medium (CM) of differentiation inducer-resistant myeloblastic M1 cells (clone R-1) (5). This GI factor also inhibited growth of MI cells that had been pretreated with differentiation inducer and expressed some differentiation-associated properties but still retained ability to proliferate. On the other hand, it scarcely inhibited growth of untreated M1 cells. Furthermore, the GI factor inhibited growth of other mouse myelomonoblastic leukemic WEHI-3B D' cells that had been pretreated with a differentiation in- ducer and of mouse monocytic leukemia cells (5). These results suggest that the GI factor produced by parental myeloblastic and inducer-resistant M1 cells preferentially inhibited growth of mouse monocytic leukemia cells in intermediate stages of differentiation between myeloblastic leukemia cells and mature macrophages. The GI activity was lost on heat treatment or on incubation with a proteolytic enzyme, but not with mixed glycosidases. These results indicate that GI ph factor is a heat-labile proteinaceous substance (5). The purified GI factor markedly inhibited growth of Thus, the GI factor may play an important role in the mouse bone marrow cells stimulated by macrophage control of proliferation of monocytic cells. Therefore, we colony-stimulating factor. The GI factor appeared to purified the GI factor and examined some of its properties. be a unique cytokine unrelated to known cytokines such as thetumornecrosisfactor,interferons,and MATERIALS AND METHODS oncostatin M. Cells and Cell Culture-The M1 cell line was originally isolated 5431 from a spontaneous myeloid leukemia in an SL strain mouse (3). Sensitive M1 cells responded to various inducers of differentiation, which induced their differentiation into macrophages. Resistant M1 clone R-1 cells did not differentiate on treatment with even high concentrations of these inducers (6). Mm-A cells (1) are a highly leukemogenic variant line of the Mm-1 cell line (2), which is a monocytic, nonleukemogenic cell line that developed spontaneously from mouse myeloid leukemic M1 cells (3). Mm-A cells grow as monolayers in liquid culture (1). M1 cells and Mm-A cells were maintained by subculture in 6-cm Falcon plastic dishes in Eagle's minimum essential medium with twice the normal concentrations of amino acids and vitamins, and supplemented with 10% heat-inactivated calf serum and 160 pg of kanamycin sulfate/ml. They were cultured at 37 "C in a humidified atmosphere of 5% C02 in air. Preparation of CM of R-I Cells-Serum-free cultures were prepared with R-l cells harvested from the spinner cultures. R-1 cells were seeded at 5 X lo5 cells/ml in spinner flasks containing 2 liters of serum-free Eagle's minimum essential medium with twice the normal concentrations of amino acids and vitamins. After 2 days, most of the cells were viable and the culture fluid was collected and centrifuged (10,000 X g, 20 min) to remove cells and cell debris. Assay of Growth Znhibitoly Activity GI Activity-Mm-A cells (1 X lo5 cells/ml) were cultured in the culture medium described above with or without the GI factor preparation for 3 days and then the cells were counted in an automatic cell counter (Coulter Counter, Model ZBI, Coulter Electronics, Inc., Hialeah, FL). The cell number The abbreviations used are: GI factor, growth inhibitory factor; GI activity, growth inhibitory activity; PBS, phosphate-buffered saline; HPLC, high-performance liquid chromatography; SDS, sodium dodecyl sulfate; M-CSF, macrophage colony-stimulating factor; CM, conditioned medium; TNF, tumor necrosis factor.

2 5432 Growth Inhibitory of control culture was10-13 X 10S/ml on day 3. GI activity was calculated by the following formula: (C - 1) - (T- 1) x GI activity (%) = c-i where C is the cell number of the control sample, T is that of the treated sample on day 3, and I is the inoculum cell number on day 0 (5). Purification Procedures-GI factor was precipitated from 135 liters of CM with 90% saturation of ammonium sulfate. The precipitate was suspended in distilled water, dialyzed against 50 mm sodium acetate buffer, ph 5.0, containing 0.14 M NaC1, and applied to a CM- Sepharose CL-GB (Pharmacia Fine Chemicals, Uppsala, Sweden) column (1.9 X 32 cm) at 4 "C. The column was washed with 50 mm sodium acetate buffer, ph 5.0, containing 0.14 M NaCl and then eluted with 50 mm sodium acetate buffer containing a linear concentration gradient of M NaCl. Fractions biologically active from the CM-Sepharose column were concentrated by dialysis against sucrose, dialyzed against phosphatebuffered saline (PBS) (138 mm NaCl, 2.7 mm KCI, 8 mm NaHP04, 1.5 mm KH2P04, ph 7.4), and subjected to gel filtration in PBS on a column (1.9 X 86 cm) of Sephadex G-200 (Pharmacia Fine Chemicals) at 4 "C. Active fractions from the Sephadex G-200 columns were pooled, concentrated by dialysis against sucrose, and dialyzed against PBS. Acetonitrile and trifluoroacetic acid were added to the sample solution at final concentrations of 20% (v/v) and 0.1% (v/v), respectively. The sample was clarified by centrifugation and applied to a reverse-phase high-performance liquid chromatography (HPLC) column. The HPLC system (Waters Associates, Milford, MA) consisted of two M A pumps, a "660 solvent programmer, a model440fixed wavelength detector, and an U6K injector. The sample was applied to a pbondapak C,, column (300 X 3.9 mm, Waters Associates) through a solvent line at room temperature at a flow rate of 1 ml/ min. The column was washed 5 min with a linear gradient from 20% to 30% (v/v) acetonitrile containing 0.1% (v/v) trifluoroacetic acid and then the material was eluted with a linear gradient from 30% to 55% (v/v) acetonitrile containing 0.1% (v/v) trifluoroacetic acid in 150 min. Fractions (2 ml) were collected in tubes containing 20 p1 of 2% (v/v) Tween 20. Active fractions from the reverse-phase HPLC column were pooled, lyophilized, and dissolved in 150 pl of 45% (v/v) acetonitrile, 0.1% (v/v) trifluoroacetic acid solution. The sample was injected onto a TSK G3000 SW gel filtration column (120 X 0.75 cm, Toyo Soda, Tokyo, Japan), eluted with 45% (v/v) acetonitrile, 0.1% (v/v) trifluoroacetic acid solution at a flow rate of 0.5 ml/min, and collected in 0.5-ml fractions. Idination of GI Factor-Purified GI factor was iodinated as described by Tomida et al. (7). Aliquots of GI factor from the TSK G3000 SW column were dried under vacuum in a Speed Vac unit. The preparation was then mixed first with 10 pl of0.2 M sodium phosphate buffer, ph 6.5, containing 0.04% (v/v) polyethylene glycol 6000 and 10 ~l of 50% (v/v) dimethyl sulfoxide, then with 10 pl of carrier-free Iz5I (1 mci, Amersham Japan, Tokyo), and finally with 10 pl of a solution of chloramine T (200 pg/ml). The mixture was left standing at 0 "C for 30 min and then 10 pl of potassium metabisulfite (200 pg/ml) and 10 plof 0.1 M KI were added. The iodinated GI factor was separated from unbound '%I by chromatography on a PD- 10 column (Pharmacia Fine Chemicals) in PBS. Polyacrylumide Gel Electrophoresis-Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS) was performed essentially the same as described by Laemmli (8). The '''Ilabeled GI factor was subjected to electrophoresis on a 15% acrylamide slab gel (100 mm long, 140mmwide,1.5mm thick) at a constant current of 30 ma for 3 h. Autoradiography was performed at -70 "C for 1 day with x-ray film (Fuji, Japan) and intensifying screens (Du Pont Cronex Lightning Plus). Chromatofocusing of GI Factor-Part of the pooled active fractions from the reverse-phase HPLC column were dialyzed against ethanolamine acetate buffer (25 mm, ph 9.5) and then applied to a chromatofocusing column (0.9 X 8.5 cm column; Polybuffer exchanger PBE 94; Pharmacia Fine Chemicals) equilibrated with ethanolamine acetate buffer. Material was eluted with Polybuffer %/acetic acid (ph 6.0, Pharmacia Fine Chemicals) and the GI activity and ph of each fraction were determined. Liquid Swpenswn Culture of Bone Marrow Cells-Bone marrow cella were obtained from the femurs of SL mice of 8-12 weeks old. Factor for Monocytic Leukemia Cells The cells were washed twice with PBS, and suspensions of 5 X lo7 cells in 5 ml of PBS were gently layered over 5 ml of Lympholyte M, a separation medium for isolation of mouse lymphocytes, at a density of g/ml (Cedarlane Laboratories, Ltd., Canada). Centrif- ugation was for 20 min at 400 X g and the cells at the interface were collected and washed twice with PBS. Macrophage colony-stimulating factor (M-CSF) was prepared and partially purified from mouse fibroblast L929 cells as reported by Yamamoto-Yamaguchi et al. (9). L929 cells were cultured in serumfree Eagle's minimum essential medium. M-CSF was concentrated from the CM by ammonium sulfate precipitation and fractionated by gel filtration on a Sephadex G-200 column. The final titer of CSF was about 6600 units/ml. Bone marrow cells were cultured in RPMI 1640 medium (GIBCO) supplemented with 20% (v/v) heat-inactivated horse serum and various amounts of CSF. The cells were seeded at a concentration of 2 X 10' cells/ml in 15-mm tissue culture wells (0.5 ml/well) (Nunc, Denmark). Viable cells were counted after incubation for 6 days. RESULTS Purification Procedures-R-1 cells (5 X lo6 cells/ml) were cultured for 2 days in the absence of calf serum. The CM was treated with 90% saturation of ammonium sulfate and the precipitate was dialyzed against 0.14 M NaCl in 0.05 M sodium acetate buffer, ph 5.0. This treatment generally resulted in a 60- to 80-fold concentration of the CM. The resultant preparation was used as the starting material for purification of the GI factor. The preparation was purified by chromatography on a CM- Sepharose CL-GB column equilibrated with 50 mm sodium acetate buffer, ph 5.0, containing 0.14 M NaC1. Under these conditions, almost all the GI factor activity bound to the CM- Sepharose CL-GB. As shown in Fig. 1, most of the GI activity was eluted with buffer containing M NaC1. The recovery in this purification step was about 50-60%. The GI factor preparation was next chromatographed on a Sephadex G-200 column. The GI activity was eluted in frac- tions corresponding to an apparent molecular weight of 20,000-50,000 (Fig. 2). This step resulted in 25- to 50-fold purification. The GI factor was next chromatographed on a C1, hydrophobic support column (Fig. 3). The bulk of the protein was eluted with less than 40% acetonitrile, whereas the GI activity was eluted with 45-50% acetonitrile. The fractions indicated by a bar were pooled for further purification. The final step of purification was high performance gel filtration chromatography on TSK G3000 SW columns; better separation was obtained by connecting the columns. The volatile solvent was 45% acetonitrile in 0.1% trifluoroacetic acid. The GI activity was eluted as a single broad peak just after the void volume (Fig. 4A). This position was earlier than FIG. 1. Chromatography of GI factor on a CM-Sepharose CL-6B column. The ammonium sulfate precipitate from 135 liters of CM of R-1 cells was dialyzed against 50 mm sodium acetate buffer, ph 5.0, containing 0.14 M NaCl and applied to a CM-Sepharose CL- 6B column. 0, GI factor; ----, absorbance at 280 nm; -, fractions pooled for chromatography on Sephadex G-200.

3 Growth Inhibitory Factor for Monocytic Leukemia Cells 5433 FRction nmbr FIG, 2. Gel filtration of the GI factor on Sephadex G-200. Pooled fractions from the CM-Sepharose CL-GB column were applied to a Sephadex G-200 column. 0, GI activity; , absorbance at 280 nm; -, fractions pooled for reverse-phase HPLC. Arrows show the positions of reference substances: blue dextran (BD), aldolase (Mr 158,000), ovalbumin (Mr 45,000), cytochrome c (M. 12,500), and phenol red (PR). FIG. 3. Reverse-phase HPLC of GI factor. Pooled fractions from the Sephadex G-200 column were applied to a pbondapak Cia column. Material was eluted with a 150-min linear gradient from 30 to 55% (v/v) acetonitrile containing 0.1% (v/v) trifluoroacetic acid. 0, GI activity; ----, absorbance at 280 nm; -, fractions pooled for TSK G3000 SW gel filtration HPLC; -, acetonitrile gradient. expected, since the apparent molecular weight of the GI factor was less than 50,000, as determined by Sephadex G-200 column chromatography (Fig. 2). The fractions indicated by a bar were pooled, lyophilized, and rechromatographed (Fig. 4, B and C). On the third chromatography, a single protein peak was coeluted with GI activity (Fig. 4C). The purification of the GI factor is summarized in Table I. The GI factor was purified 1200-fold with an overall recovery of 1.8%. Chromatofocusing of GI Factor-Part of the GI factor preparation purified by reverse-phase HPLC column chromatography was applied to a chromatofocusing column. Results showed that the apparent isoelectric point of the GI factor was (Fig. 5), indicating that the GI factor is a basic protein. SDS-Polyacrylamide Gel Analysis-An aliquot of the fractions from the TSK G3000 SW high performance gel filtration column (Fig. 4C) was radioiodinated and analyzed by SDSpolyacrylamide gel electrophoresis and autoradiography. A single protein band with an apparent molecular weight of 25,000 was obtained (Fig. 6) and this was associated with GI activity (Fig. 7). The inhibitory effect of the purified GI factor on growth of Mm-A cells is shown in Fig. 8. Assuming that the GI factor has a molecular weight of 25,000, its concentrations for 30 and 50% inhibition of the growth of Mm-A cells were approximately 4 X 10"O and 8 X 10"" M, respectively. Effect of GI Factor on the Growth of Normal Bone Marrow Cells Stimulated by M-CSF-Next we examined whether the GI factor could inhibit the growth of normal bone marrow cells stimulated by M-CSF. For this, we cultured bone marrow FIG. 4. High performance gel filtration chromatography of the GI factor on TSK G3000 SW. A, pooled fractions from the reverse-phase HPLC step were lyophilized and dissolved in 150 pl of 45% acetonitrile containing 0.1% trifluoroacetic acid. Material was eluted with the same buffer at 0.5 ml/min and fractions of 0.5 ml were collected. The horizontal bar indicates the fractions pooled for rechromatography. B, rechromatography of pooled fractions of GI factor from A. The horizontal bar indicates the fractions pooled for the third chromatography. C, third chromatography on a TSK G3000 SW column of pooled fractions of the GI factor from B. 0, GI activity; "", absorbance at 280 nm. Step RlCM 90% (NHJZSO, precipitate CM-Sepharose CL-GB Sephadex G-200 pbondapak CIS TSK G3000 SW 1st 3rd TABLE I Purification of GI factor from RlCM Protein" nag 1, GI factorb ng protein Purification fold 12, ,200 Total GI factor (ab) 102,000 54,882 29,198 12,032 4,571 1,800 Yield % Determined by dye fixation (Bio-Rad). Amount of GI factor (ng of protein) required for 30% GI activity. cells in liquid suspension culture. As reported by Yamamoto- Yamaguchi et al. (9), CSF was essential for survival and growth of bone marrow cells in liquid culture as well as in semisolid agar culture. On day 6 the viable cell number in

4 Growth InhibitoryFactor for Monocytic Leukemia Cells 5434 FIG. 5. Chromatofocusing of GI factor from the reversephase HPLC. Part of the pooled active fractions from the reversephase HPLC step were dialyzed against ethanolamine acetate buffer (25 mm, ph 9.5) and subjected to chromatofocusing. GI factor was eluted with Polybuffer 96/acetic acid, ph , GI activity; , ph. 116K 68K - 30K G I - f a c t a (nglrnl) FIG.8. Inhibition of growth of Mm-A cells by purified GI factor. Mm-A cells were incubated for 3 days with various concentrations of GI factor from the third TSK G3000 SW column and then their cell number was determined. TABLE I1 Effect of GI factor on the growth of normal bone marrow cells stimulated by M-CSF Bone marrow cells (2 X lo5 cells/ml) were incubated with M-CSF (1200 units/ml) in the presence or absence of GI factor for 6 days. Then viable cells were counted bv the trwanblue dve exclusion test. M-CSF GI factor Cell number X IO' cells/ml 1 f 0.5" 44 f 4 20 ng/ml 2 f n d m l 1f K Mean & S.D. of triplicate determinations. FIG. 6. SDS-polyacrylamide gel electrophoresis of purified GI factor. The '251-labeledGI factor was subjected to electrophoresis on 15%acrylamide slab gel. The molecular weight markers used M, 116,000), bovine serum albumin ( M, 68,000), carbonic anhydrase (M. 30,000), and soybean trypsin inhibitor ( M, 20,100). O l-"j 1234sc7~smnuw nice numkr FIG. 7. SDS-polyacrylamide gel electrophoresis of GI activity. The purified GI factor was subjected to electrophoresis by the same procedure as described in the legend to Fig. 6. The gel was sliced and GI activity was eluted from sections (5mm) of gel overnight with serum-free culture medium containing 0.02% Tween 20. cultures containing M-CSF (1200 units/ml) had more than doubled although in the absence of M-CSF, it had decreased to one-fifteenth of that initially (Table 11). Purified GI factor (20-50 ng/ml) markedly inhibited M-CSF-induced growth stimulation of normal bone marrow cells (Table 11). These results suggest that thegi factor may be aregulator of normal hematopoietic cells. DISCUSSION A factor inhibiting growth of mouse monocytic leukemia Mm-A cells has been purified to apparenthomogeneity from the CM of mouse myeloblastic leukemia M1 cells. The apparent molecular weight of the GI factor is 25,000, as determined by SDS-polyacrylamide gel electrophoresis. The GI factor is a basic protein and at 8 X 10"' M causes 50% inhibition of the growth of Mm-A cells. We previously reported that the GI factor in the CMof M1 clone R-1 cells inhibited the growth of not only Mm-A cells, but also mouse monocytic cells, and furthermore, the growth of M1 cells and WEHI-3B D' cells that had been treated with a differentiation inducer and expressed some differentiated functions but retained aproliferation potential (5). In contrast, the GI factor had no effect on the growth of undifferentiated M1 cells or WEHI-3B D' cells (5). Theseresults suggest that the GI factor is a unique cytokine that preferentially inhibits proliferation of partially differentiated leukemic cells. Monocyte-macrophages produce various cytokines with activities to inhibit tumor cells, including tumor necrosis factor (TNF) (10-12), interferons (13-15), interleukin 1 (16, 17), and oncostatin M(18).TNF is produced in vitro by peritoneal macrophages and cloned lines of myeloid leukemia or histiocytoma after stimulation with endotoxin, phorbol esters, and/ or lectins. TNF is an acidic protein with a PI of and is stable on heating at 56 "C for 2 h (10-12). On the other hand, the GI factor was found to be produced constitutively by R-1 cells. The GI factor was a basic protein with a PI of (Fig. 5) andmost of the GI activity was lost on heating at 56 "C for 30 min (5). These properties suggest that the GI factor is not related to TNF. Untreated M1 cells did not produce interferons, although poly(1).poly(c)-treated M1 cells produced interferons (mainly interferon p) (19). Furthermore, the growth of MmA cells was not significantly suppressed by treatment with recombinant mouse interferon, even at a high concentration (2,000 units/ml) (data not shown). Interleukin 1 is an acidic protein of PI with a molecular weight of14,000 as determined by SDS-polyacrylamide gel electrophoresis (16, 17). Oncostatin M is a cytokine that inhibits the replication of A375 melanoma cells and is produced by the human mon-

5 ocytic cell line U937 cells only after stimulation of the cells with a phorbol ester (18). Furthermore, oncostatin M is stable on heating at 56 "C for 1 h. These findings suggest the the GI factor is a unique cell growth regulator that is different from all factors described previously. The GI factor inhibited the proliferation of partially differentiated leukemic cells, but did not significantly affect that of myeloblastic leukemic cells (5). These findings suggest that the GI factor may help to retain the immature leukemic cell population and give some growth advantage to immature leukemic cells. Thus, the GI factor might play an important role in leukemogenesis and/or progression of leukemia. The GI factor also inhibited the growth of normal bone marrow cells stimulated by M-CSF (Table 11). As the cell number and viability of macrophages were scarcely affected by partially purified GI factor (5), the GI factor may preferentially affect M-CSF-responsive macrophage progenitor cells. Thus the GI factor may be a regulator of normal hematopoiesis, although further detailed effects of GI factor on normal hematopoiesis remains to be examined. Studies on the primary structure of the GI factor and its therapeutic effect on monocytic leukemia are also of interest. Growth Inhibitory Factor for Monocytic Leukemia Cells Kasukabe, T., Honma, Y., and Hozumi, M. (1985) Jpn. J. Cancer Res. (Gann) 78, Kasukabe, T., Okabe-Kado, J., Honma, Y., and Hozumi, M. (1987) Jpn. J. Cancer Res. (Gann) 78, Hozumi, M. (1983) Adv. Cancer Res. 38, Tomida, M., Yamamoto-Yamaguchi, Y., and Hozumi, M. (1984) FEBS Lett. 178, Laemmli, U. K. (1970) Nature 227, Yamamoto-Yamaguchi, Y., Tomida, M., and Hozumi, M. (1983) Blood 62, Carswell, E. A., Old, L. J., Kassel, R. L., Green, S., Fiore, N., and Williamson, B. (1985) Proc. Natl. Acad. Sci. U. S. A. 72, Haranaka, K., Carswell, E. A., Williamson, B. D., Prendergast, J. S., Satomi, N., and Old, L. J. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, Beutler, B., and Cerami, A. (1986) Nature 320, Allen, G., and Fantes, K. H. (1980) Nature 287, Gray, W. P., Lenng, D. W., Sherwood, P. J., Wallace, D. M., Berger, S. L., Levinson, A. D., and Goeddel, D. V. (1982) Nature 296, Broxmeyer, H., Lu, L., Platzer, E., Feit, C., Juliano, L., and Rubin, B. (1983) J. Zmmunol. 131, Lachman, L. B., Dinarello, C. A., Llansa, N. D., and Fidler, I. J. (1986) J. Zmmuml. 136, Knudsen, P. J., Dinarello, C. A., and Strom, T. B. (1986) J. Zmmunol. 136, REFERENCES 18. Zarling, J. M., Shoyab, M., Marquardt, H., Hanson, M. B., 1. Kasukabe, T., Honma, Y., and Hozumi, M. (1984) J. Cell. Physiol. Lioubin, M. L., and Todaro, G. (1986) Proc. Natl. Acad. Sci. U. 118, S. A. 83, Maeda, M., and Ichikawa, Y. (1973) Gann 64, Yamamoto, Y., Tomida, M., and Hozumi, M. (1979) Cancer Res. 3. Ichikawa, Y. (1969) J. Cell. Physwl. 74, ,