Changes of phospholipase A2 inhibitory activity in the K+-sensitive actin gelation factor during the differentiation of myeloid leukemia cells

320

Biochimica et Biopt~vsica Acta 930 (1987) 320-325

Elsevier BBA 12116

Changes of phospholipase A 2 inhibitory activity in the K +-sensitive actin gelation factor during the differentiation of myeloid leukemia cells Kuniaki Takagi, Hiroshi Hotta and Yasunobu Suketa Department of Environmental Biochemistry, Shizuoka College of Pharmao', Shizuoka (Japan)

(Received 29 December 1986) (Revised manuscript received 29 June 1987)

Key words: Lipocortin:Dephosphorylation: Differentiation; Actin gelation protein; (Myeloid leukemia)

The phospholipase A 2 inhibitory activity of a 38 k D a K +-sensitive actin gelation factor in a murine leukemia cell line (M1) was examined. A specific antibody against 38 kDa protein was found to cross-react with 37 k D a protein (lipocortin) in rat peritoneal exudates. Although the native 38 k D a protein from M I cells did not block phospholipase A 2 activity, pretreatment with alkaline phosphatase produced a form that did inhibit this enzyme. However, a purified 38 k D a protein from differentiated M I cells blocked phospholipase A 2 activity without pretreatment with alkaline phosphatase. Phospholipase A 2 inhibitory activity of the 38 kDa protein was not altered by addition of actin. These findings suggest that the phospholipase A 2 inhibitory of our 38 kDa protein was induced during differentiation. We also proposed that our 38 kDa protein has the same epitope as lipocortin.

Introduction Several oncogene products and growth factor receptors exhibit tyrosine kinase activity [1,2]. The characterization of tyrosine kinases and their substrates is important for the understanding of the mechanism of cell proliferation. One of the substrates for these tyrosine kinases is a K+-sensitive 38 kDa actin gelation protein [3] in mouse myeloid leukemic cells (M1 cells). This factor is referred to as 'abp38' in Refs. 3-8. Serine and tyrosine in the 38 kDa protein were phosphorylated in vivo; the rate of phosphorylation in 38 kDa protein changes

The abbreviations used are: SDS, sodium dodecyl sulfate: V8 protease, proteinase from Staphylococcus aureus strain V8. Correspondence : K. Takagi, Department of Environmental Biochemistry, Shizuoka College of Pharmacy, Shizuoka 422. Japan.

during differentiation of M1 cells [4]. The 38 kDa protein localizes in the cytosol before differentiation, but half of the 38 kDa protein is in the cytoskeleton after cell differentiation [5,6]. The 38 kDa protein is widely distributed in cultured cells and tissues [5]. Recently, some independent studies have demonstrated that lipocortin (a 37 to 40-kDa protein which has phospholipase A 2 inhibitory activity) [7--9] and calpactin I (a 36-kDa protein which has calcium-dependent phospholipid- and actin-binding ability) are substrate for tyrosine kinases [10-12]. These two proteins are homologous proteins according to their amino acid sequences [12]. They are ubiquitous proteins associated with the cytoskeleton and are localized on the cytoplasmic face of the plasma membrane [13-16]. In this report, the phospholipase A 2 inhibitory activity of the 38 kDa protein was found to change with cell differentiation of M1 cells. Phosphorylation and phospholipase A 2 inhibitory activity of

0167-4889/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)

321 the 38-kDa protein were shown to play an important role in the regulation of cell differentiation. Materials and Methods

Materials. Reagents and their sources were as follows. EGTA, Tris and hydroxylapatite, Nakarai Chemical Co., Ltd., Japan; phenylmethylsulfonyl fluoride, pepstatin, antipain'and N-tosyl-L-phenylalanine chloromethyl ketone and alkaline phosphatase, Sigma Chemical Co., St. Louis, U.S.A.' phospholipase A 2, Boehringer Mannheim, F.R.G.; nitrocellulose membrane and horseradish peroxidase conjugated goat anti rabbit IgG antibody, Bio-Rad Laboratories, CA, U.S.A.); CNBr-activated Sepharose 4B and DEAE-Sepharose CL-6B, Pharmacia Fine Chemicals, Uppsala, Sweden. Actin was purified from the acetone powder of rabbit skeletal muscle by the method of Spudich and Watt [17]. Cell line and cell culture. The M1 cell line was isolated from myeloid leukemic cells of an SL strain mouse [18]. The cells were cultured in Eagle's minimum essential medium (Nissue Seiyaku Co., Japan) containing a 2-fold concentration of amino acids and vitamins. To induce differentiation, the cells were incubated for 3 days in the above medium containing 20% conditioned medium harvested from rat embryo fibroblasts. Differentiation was checked by a phagocytosis test as reported previously [19]. Preparation and characterization of 38-kDa protein purified from M1 cells. The 38 kDa protein was purified from M1 cells as described previously [3]. For actin binding tests, the purified 38 kDa protein was mixed with purified rabbit skeletal muscle actin (0.575 m g / m l ) in the presence of 2 mM MgC12, with or without KC1 at 4°C. After incubation for 1 h at room temperature, the mixtures were centrifuged at 100 000 × g for 1 h. The pellet, which contained actin-bound 38 kDa protein, was washed with buffer C (1 mM EGTA, 1 mM ATP, 1 mM dithiothreitol, various proteinase inhibitors and 10 mM Tris-HC1, p H 7.0) and resuspended in the same buffer using a sonicator. To the supernatant and the pellet solution (150 ~1) 50 /LI of 4-fold concentrated SDS-polyacrylamide gel electrophoresis sample buffer was added. Then,

50 /tl of the mixture was electrophoresed in an SDS-containing 10% polyacrylamide gel slab [20]. The gels were scanned after electrophoresis in a densitometer. Peptide maps were examined by the method of Cleveland et al. [21] and the protein concentration was determined by the method of Bradford [22]. Preparation of antiserum. Antiserum against the 38 kDa protein was prepared as described previously [5]. Alkaline phosphatase (200 #g) was emulsified in complete Freund's adjuvant and injected subcutaneously at multiple sites into New Zealand male rabbits. The rabbits were given booster injections of 100 /~g alkaline phosphatase at 4 weeks after the first injection, and blood was removed 10 days later. The 7-globulin fraction of the antiserum against alkaline phosphatase was conjugated to CNBr-activated Sepharose 4B. Immunoblotting. Protein were separated by 10% SDS-polyacrylamide gel electrophoresis and transferred onto nitrocellulose sheets and 60 V for 4 h in the Tris/glycine/methanol buffer of Towbin et al. [23]. One nitrocellulose sheets was stained with amido black 10B. Another sheet was incubated for 1 h at room temperature in buffer A (0.15 M NaC1, 0.01 M Tris-HC1, 1% bovine serum albumin, pH 7.4) and was subsequently incubated for 2h in an appropriate dilution of polyclonal antibodies. This sheet was rinsed three times in buffer B (0.15 M NaC1, 0.01 M Tris-HC1, 0.05% Tween 20, 0.02% NaN 3, pH 7.4) followed by incubation for 1 h with horseradish peroxidase-conjugated goat anti-rabbit IgG antibody. After incubation, the sheet was rinsed three times in buffer B and reacted with 4-chloro-l-naphthol and H202 in methanol solution. Phospholipase A 2 assay. Samples were tested for phospholipase A 2 inhibitory activity by an in vitro assay as described previously [24]. Briefly, the substrate for phospholipase A 2 was prepared as autoclaved [9,10-3H]oleic acid-labeled Escherichia coli. For each experiment, a 200-/~1 sample was combined with 50 ~1 7-fold buffer (0.7 M Tris-HC1 (pH 8.0) and 60 mM CaC12) and with 50 ~1 of a dilute preparation of procine pancrease phospholipase A 2 which contained 25 ttg enzyme and 125 #g bovine serum albumin. Samples were mixed and kept on ice for 1 h, and then the reaction was performed by addition of 50 ~tl of E.

322

coli suspension at room temperature for 3 rain. The reaction was stopped by adding 100/~l HC1. The mixture was centrifuged for 5 min at 10 000 × g. Each supernatant (200/~1) was mixed with 5 ml scintillation fluid and the residual phospholipase A 2 activity was quantitated by liquid scintillation counting. In all analyses, samples were assayed in triplicate and adjusted for nonspecific release by subtracting a control value (without phospholipase A2). 1 unit of activity inhibits 15 ng phospholipase A 2.

~ m . . _

Results

Immunological comparison of 38 kDa protein with lipocortin To determine whether 38 kDa protein was similar to lipocortin, we have partially purified a 37 kDa protein that inhibited phospholipase A 2 activity from rat peritoneal exudates as described earlier [26] and analyzed it by a Western blotting method, using a specific polyclonal antibody raised against 38 kDa protein [5]. As shown in Fig. 1, immunoreactive 37-kDa protein was contained in the DEAE flow-through fraction. The phospholipase A 2 inhibitory activity of the 37-kDa protein was reduced by incubation with the antibody (1 : 50 dilution) against 38-kDa protein. Controls, 37-kDa protein incubated with preimmune serum (1:50 dilution), had no effect on the phospholipase A 2 inhibitory activity of 37-kDa protein (data not shown). This 37-kDa protein from rat peritoneal exudates blocked phospholipase A 2 activity (data not shown). In addition, the 38-kDa protein and the 37-kDa protein and the 37-kDa protein from peritoneal exudates were cleaved by V8 protease by Clevelands' method [21] and were compared with each other by Western blot analysis (Fig. 2. The major cleavage fragments of 38 kDa protein were the 33.7-kDa, 30.9-kDa, 29.2-kDa, 26.5-kDa and 25.2-kDa peptides. However, the major fragment from the 37-kDa protein from rat peritoneal exudates was a 27.8-kDa peptide. Phospholipase A e inhibitory activity of 38-kDa protein We have purified the 38-kDa protein from undifferentiated M1 cells as described previously [3] and we examined the actin-binding ability of

J

f

Fig. 1. Immunoblot analysis of lipocortin with anti-abp38 antibody. (A) DEAE flow-through fraction of rat peritoneal exudates was fractionated by SDS-polyacrylamide (13% gel) and transferred to nitrocellulose membranes. Lane 1 was blotted with rabbit anti-abp38 antibody (serum dilution, 1 : 1000), followed by horseradish peroxidase-goat anti-rabbit IgG antibody. Lanes 2 and M were stained with amido black 10B. Lanes 1 and 2, DEAE flow-through fraction; lane M, molecular weight markers (phosphorylase b (92.5 kDa), serum albumin (66.2 kDa), ovalbumin (45 kDa), carbonic anhydrase (31 kDa), soybean trypsin inhibitor (21.5 kDa) and lysozyme (14.4 kDa); all from Bio-Rad Co.).

the 38-kDa protein. As shown in Fig. 3, in the absence of KC1, more than 99% of the 38-kDa protein present in the starting mixture was recovered in the sediment. In the presence of KC1, only 15% of 38-kDa protein was sedimented with F-actin. We then assayed the phospholipase A 2 inhibitory activity of this protein. The 38-kDa protein did not block phospholipase A 2 activity

323

2

1

Fig. 2. Immunoblot analysis of peptide maps of lipocortin and abp38. The abp38 purified from undifferentiated and differentiated M1 cells at a concentration of 0.300 m g / m l was digested at 3 7 ° C with a final concentration of 25 /~g/ml of V8 protease. The 37-kDa protein from rat peritoneal exudates at a concentration of 0.030 m g / m l was digested at 3 7 ° C with a final concentration of 2.5 # g / m l of V8 protease. Each sample was fractionated by SDS-polyacrylamide gel electrophoresis (13% gel) and transferred to nitrocellulose membrane. This sheet was immunoblotted with rabbit anti-abp38 antibody (serum dilution, 1 : 1000). Lanes 1, 2, 3 and 4, digestion of the abp38 purified from M1 cells incubated for 2, 10, 30 and 60 min, respectively. Lanes 5, 6, 7 and 8, digestion of the abp38 purified from differentiated M1 cells incubated 2, 10, 30 and 60 min, respectively. Lanes 9 and 10, digestion of the 37-kDa protein of rat peritoneal exudates incubated for 2 and 30 rnin, respectively.

4

Fig. 3. Binding of abp38 to F-actin. A mixture of abp38 (0.204 m g / m l ) and actin (0.575 m g / m l ) was incubated in the presence of 2 mM MgC12 with or without 50 mM KC1 at 25 ° C for 60 min. After centrifugation at 100000× g for 60 rain, the pellet and the supernatant were electrophoresed in an SDScontaining 10% gel slab. Lane 1, pellet without KC1; lane 2, supernatant without KCI; lane 3, pellet with KCI; lane 4, supernatant with KC1.

,

over the range 0-64.5 /~g/ml of 38-kDa protein concentrations (except at 6.45 ffg/ml; Fig. 4). Since the 38 kDa protein of undifferentiated M1 cells was highly phosphorylated as reported earlier [4], we examined the effect of dephosphorylation of this protein on the inhibitory activity of this protein against phospholipase A 2. The 38-kDa protein from undifferentiated M1 cells was pretreated with alkaline phosphatase; the enzyme was removed using anti-alkaline phosphatase antibody linked to Sepharose 4B. The dephosphorylated abp38 exhibited phospholipase A 2 inhibitory activity in a dose-dependent manner (Fig. 4). On the other hand, the 38-kDa protein from differentiated M1 cells blocked phospholipase A 2

3

x10-3

-- 3 E

j

J

2

o

o

o~:

//•

,(%.

,,

... .~o0~i

o Io "

I 10

o 100 I

Mg of a~p38

Fig. 4. Phospholipase A 2 inhibitory activity of abp38 purified from undifferentiated M1 cells. Aliquots of the purified abp38 were incubated with 50 ~tg of porcine pancreatic phospholipase A2, and the activities were assayed for residual phospholipase activity. The data presented show an average in triplicate. O, native abp38; e, abp38 pretreated with alkaline phosphatase,

324 the 38-kDa protein is converted into a cytoskeletal component. We examined the effects of actin on the phospholipase A 2 inhibitory activity of 38-kDa protein. As indicated in Table I, G- and F-actin did not block the phospholipase A 2 inhibitory activity by itself. In fact, the phospholipase A 2 inhibitory activity of 38-kDa protein slightly increased after the addition of actin.

u

~8 ~5

2~

2

/ Discussion

o

"

~

1;

1;0

pg of abp38

Fig. 5. Phospholipase A 2 inhibitory activity of abp38 purified from differentiated M1 cells. Aliquots of the purified abp38 were incubated with the phospholipase A 2 and the activities were assayed. activity without pretreatment with alkaline phosphatase (Fig. 5). The purified 38-kDa protein from differentiated M1 cells had a spec. act. of 10 0 0 0 - 1 4 000 u n i t s / m g protein.

The effect of actin on the phospholipase A: inhibitory activity of 38-kDa protein During cell differentiation of M1 cells, some of

TABLE I THE EFFECT OF ACTIN ON THE PHOSPHOLIPASE A 2 INHIBITORY ACTIVITY OF abp38 Abp38 was purified from differentiated M1 cells. The abp38 was mixed with purified rabbit skeletal muscle actin in the presence of 2 mM MgCI2 at 4 ° C. After incubation for 30 min at room temperature, the samples were tested for phospholipase A 2 inhibitory activity by an in vitro assay. abp38 (fig) Exp. 1 0 102 102 102 Exp. 2 0 0 0 102 102 102

Actin (fig)

Mg 2+ (mM)

Substrate hydrolyzed (cpm)

0 0 11.5 115

2 0 2 2

25909+_ 12974+_ 14653+_ 10845+_

0 11.5 115 0 11.5 115

0 2 2 0 2 2

31997+_2113(0%) 30762+_2243(3.9%) 30714+_ 723 (4.0%) 15866+_ 467 (50.4%) 16796+_1461 (47.5%) 15042_+1511 (53.0%)

698 (0%) 689 (49.9%) 332 (43.4%) 317(58.1%)

In this study, we have demonstrated phospholipase A 2 inhibitory activity of a 38-kDa protein from M1 cells. The 37-kDa protein (lipocortin) in rat peritoneal exudates cross-reacted with the a n t i b o d y prepared against 38-kDa protein (Figs. 1 and 2). This suggest that lipocortin has the same antigenic determinant as the 38-kDa protein. However, cleavage fragments of 38-kDa protein from undifferentiated and differentiated M1 cells were different from those of lipocortin (Fig. 2). These differences m a y be due to the different species studied or the different types of protein (38-kDa protein was endogenous and lipocortin was exogenous). Phospholipase A 2 inhibitory activity of the 38k D a protein was correlated with p h o s p h o r y l a t i o n of the 38-kDa protein (Fig. 4) and the 38-kDa protein in differentiated M1 cells exhibited phospholipase A 2 inhibitory activity without treatment with alkaline phosphatase (Fig. 5). Tyrosine and serine residues of the 38-kDa protein in undifferentiated M1 cells are phosphorylated and each a m i n o acid was dephosphorylated during cell differentiation [4]. We have not yet clarified whether b o t h or either p h o s p h o a m i n o acid residues of tyrosine and serine are related to the phospholipase A 2 inhibitory activity in the 38-kDa protein. We have reported that the 38-kDa protein in M1 cells increased about 8-fold (3.89% of the total protein) and was converted into a cytoskeletal form during differentiation [5]. F r o m our findings that actin did not block the phospholipase A 2 inhibitory activity of the 38-kDa protein (Table I) and that the 38-kDa protein differentiated M1 cells had a spec. act. of 1 0 0 0 0 - 1 4 0 0 0 u n i t s / m g , we propose that the 38-kDa protein in differentiated M1 cells is a physiologically functional molecule. Further studies on the interaction of the

325 38-kDa protein with endogenous phospholipase A 2 a n d p r o d u c t s o f p h o s p h o l i p a s e A 2 in M 1 cells a r e in p r o g r e s s .

Acknowledgments W e w i s h to t h a n k Dr. Y. I w a m o t o f o r his gift o f E. coli; Mr. M . T a k a d a a n d M r . H , N a k a g a m i f o r a s s i s t a n c e o f this w o r k .

References 1 Bishop, J.M. (1985) Cell 42, 23-38 2 Heldin, C. and Westermark, B. (1984) Cell 37, 9-20 3 Takagi, K., Ichikawa, Y. and Nagata, K. (1985) J. Biochem. 97, 605-616 4 Takagi, K. and Ichikawa, Y. (1987) J. Biochem. 101, 73-79 5 Takagi, K., Zenita, K. and Ichikawa, Y. (1987) J. Biochem. 101, 63-71 6 Takagi, K., Okabe, Y., Yoshimura, K. and Ichikawa, Y. (1986) Cell Struct. Funct. 11,235-243 7 Wallner, B.P., Mattaliano, R.J., Hession, C., Care, R.L., Tizard, R., Sinclair, L.K., Foeller, C., Pingchang, E., Browning, J.L., Ramachandran, K.L. and Pepinsky, R.B. (1986) Nature 320, 77-81 8 Pepinsky, R.B. and Sinclair, L.K. (1986) Nature 321, 81-84 9 Touqui, L., Rothhut, B., Shaw, A., M., Fradin, A., Vargaftig, B. and Russo-Marie, F. (1986) Nature 321 177-180

10 Gerke, V. and Weber, K. (1984) EMBO J. 3, 227-233 11 Glenney, J. (1986) J. Biol. Chem. 261, 7247-7252 12 Kristensen, T., Saris, C.J.M., Hunter, T., Hicks, L.J., Noonan, D.J., Greneney, J.R. and Tack, B.F. (1986) Biochemistry 25, 4497-4503 13 Greenberg, M.E., Brackenbury, R. and Edelman, G.M. (1984) J. Cell Biol. 98, 473-486 14 Gould, K.L., Cooper, J.A. and Hunter, T. (1984) J. Cell Biol. 98, 487-497 15 Cheng, Y.E. and Chen, L.B. (1981) Proc. Natl. Acad. Sci. USA 78, 2388-2392 16 Greenberg, M.E. and Edelman, G.M. (1983) Cell 33, 767-779 17 Spuidch, J.A. and Watt, S. (1971) J. Biol. Chem. 246, 4866-4871 18 Ichikawa, Y. (1969) J. Cell. Physiol. 74, 223-234 19 Ichikawa, Y. (1970) J. Cell. Physiol. 76, 175-184 20 Laemmli, U.K. (1970) Nature 227, 680-685 21 Cleveland, D.W., Fischer, S.G., Kirchner, M.W. and Laemmli, U.K. (1977) J. Biol. Chem. 252, 1102-1106 22 Bradford, M. (1976) Anal. Biochem. 72, 248-254 23 Towbin, H., Staehelin, T. and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350-4354 24 Rothhut, B., Russo-Marie, F., Wood, J., DiRosa, M. and Flower, R.J. (1983) Biochem. Biophys. Res. Commun. 117, 878-884 25 Hirata, F. (1981) J. Biol. Chem. 256, 7730-7733 26 Pepinsky, R.B., Sinclair, L.K., Browning, J.L., Mattaliano, R.J., Smart, J.E., Chow, E.P. Falbel, T., Ribolini, A. Garwin, J.L. and Wallner, B.P. (1986) J. Biol. Chem. 261, 4239-4246