BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL INSULIN RAPIDLY STIMULATES ERK2 IN THE MEMBRANE OF OSTEOBLAST-LIKE UMR-106 CELL

Transcription

1 Vol. 43, No. 5, December 1997 Pages INSULIN RAPIDLY STIMULATES ERK2 IN THE MEMBRANE OF OSTEOBLAST-LIKE UMR-16 CELL Sung-Jin Kim*, Kyung-Ha Kim Department of Pharmacology, School of Dentistry, Kyung-Hee University Seoul, KOREA Received September 18, 1997 Received after revision September 22, 1997 Summary We investigated subcellular distribution of ERK2 in osteoblast-like UMR-16 cell and explored to determine if its activities are regulated by insulin. 23%, 34% and 43% of total ERK2 were distributed in membrane, cytosol and nucleus, respectively. Insulin caused 4% increase of ERK2 content in membrane in 1 min whereas it induced approximately 5 /; decrease of ERK2 in cytosol in 1 min. In terms ofkinase activity, insulin stimulated phosphorylation of the membrane-associated ERK2 by 2-fold and 1.8- fold in 1 min and 1 min and cytosolic ERK2 by 2.7-fold and 2.3-fold in 1 min and 1 min, respectively. In contrast, the phosphorylation of nuclear ERK2 was stimulated by insulin in time-dependent manner with maximal (3-fold) activity observed at 3 min. Insulin also increased the content of MEK2 in membrane by 2.2- to 2.6-fold in 1 rain. MEK2 translocated into mambrane in response to insulin may play a role in the activation of the membrane-associated ERK2 via phosphorylation. Key words : insulin signaling, osteoblast, ERK, MEK, membrane Introduction ERK2 (extracellular signal regulated protein kinase 2 or pp42 MAP kinase) plays essential roles in insulin signal transduction following activation oftyrosine kinase intrinsic to insulin receptor in response to insulin (1). Regulation of ERK2 involves activation of a series of signaling molecules such as Grb2/SOS, ras, raf and MEK (MAP kinase kinase) (2-7). The MEKs (MEK1, MEK2, MEK3) are dual specificity protein kinases that phosphorylate ERKs on tyrosine and threonine residues to make it active (8-1). MEK2 is the most active ERK activator (11). ERKs exist as several isozymes including ERK1 (pp44 MAP kinase), ERK2 (pp42 MAP kinase) and ERK3 (pp53 MAP kinase) (12). ERK2 is present mainly in the cytoplasm; however, upon stimulation with growth factors, it undergoes translocation into the nucleus and phosphorylates 4, To whom correspondence should be addressed /97/ / Capyright 1997 by Academic Press Australia. All rights of reproduction in any form reserved.

2 transcription factors such as AP-1 in many cells (13,14). UMR-16 cell contains insulin receptor and glucose transporters (15) but signaling mechanisms regulated by insulin are not fully understood in the cell. In the present study, we explored to determine iferk2 could be regulated by insulin in subcellular compartments of UMR-16 cells. We present evidence that ERK2 is widely distributed in the cell and its phosphorylation is stimulated by insulin in membrane as well as in cytosol and nucleus. The potential role of MEK2 in the activation of membrane-associated ERK2 via phosphorylation is proposed. Methods Materials Pork insulin was obtained from Elanco Products Co. (Indianapolis, IN). The reagents used for polyacrylamide gel electrophoresis were from Bio-Rad. Nitrocelluose filters (.45~tm) were purchased from Schleicher & Schuell. Antibodies to ERK2, and MEK2 were purchased from Transduction laboratories. Antibody to phospho-erk2 was purchased from New England Biolab. ECL kit was purchased from Amersham. UMR- 16 cells were puchased from ATCC. Lactate dehydrogenase assay kits and all other chemicals were purchased from Sigma Chemical Co. Cell culture UMR-16 cells were grown in DMEM media with 1% fetal calf serum in 5% humidified CO2 atmosphere at 37 ~ Isolation of nuclei, membrane, cytosol Nuclei were isolated as previously described (16). Confluent cells were serum starved for 18 hrs and stimulated with insulin (1 nm) for the indicated times. Media was removed and cultures were washed with ice-cold phosphate buffered saline. The cells were scraped into a buffer containing 2 mm Hepes, 5 mm MgC12, 25 ram KC1, 2 mm PMSF,.1 mg/ml aprotinin, 1 mm sodium vanadate, 1 mm sodium molybdate, 1 mm 13-glycerophosphate, 5 mm sodium pyrophosphate, 1 mm EDTA, 1 mm EGTA and.25 M sucrose. Cells were dounce homogenized (3 strokes) with a tight-fitting glass- Teflon homogenizer. The homogenate was loaded onto 1 ml of 1M sucrose in lysis buffer and centrifuged at 1,6 x g for 1 min at 4 ~ to pellet nuclei. The nuclei were washed with 1M sucrose in lysis buffer and centrifuged at 1,6 x g for 1 rain at 4 ~ The resulting pellets were resuspended with lysis buffer and subjected to brief sonication. 1,6 x g supernatants were centrifuged at 33, x g for 3 min to separate plasma membrane enriched fractions and cytosol (19). The 33, x g pellets was washed with lysis buffer and centrifuged at 33, x g for 3 rain. The resulting pellets were incubated with 1% Triton X-1 for 3 min at 4~ The suspension was centrifuged at 33~ x g for 1 min and the supernatant was regarded as solubilized membranes. Lacate dehydrogenase assay Measurements of lactate dehydrogenase, a cytosolic marker enzyme, were performed using a kit according to the instruction provided by the manufacturer. Western blot analysis Electrotransfer of proteins from the gels to nitrocelluose paper (Schleicher & SchueU) was carried out for 1 hr at 1 V (constant) as described by Towbin et al. (17). 124

3 The filter papers were preincubated for 1 hr at 23 ~ with PBS containing.1% Tween 2 and 3% bovine serum albumin and washed with PBS containing.1% Tween 2 three times for 1 min each. The blots were probed with primary antibodies for 1 hr at 23 ~ The blots were then incubated with HRP-conjugated anti-rabbit IgG for 3 min and washed with PBS containing Tween 2 five times for 1 min each. The detection of immobilized specific antigens was carried out by ECL (NEN). Quantitation of MAP kinase bands were carried out by scanning densitometry. Results Subcellular distribution of ERK2 and MEK2 To address the possibility of cross contamination among subcellular fractions, lactate dehydrogenase, a cytoplasmic marker, activities were measured with nuclear, cytosolic and membrane fractions (Fig. 1A). More than 95% was recovered in cytosolic fractions whereas only 2% and 3% were identified in nuclear and membrane fractions, respectively. Identification of ERK2 and MEK2 was performed by SDS-polyacrylamide gel electrophoresis and Western blot analysis using corresponding antibodies. Scanning densitometry revealed that 23%, 34% and 43% of total ERK2 were found in membrane, cytosol, and nucleus (Fig. 1B) while 2%, 41% and 39% of total MEK2 in membrane, cytosol and nucleus, respectively (Fig. 1C). Effect of insulin on the phosphorylation of ERK2 in membrane, cytosoi and nucleus antibody. Identification of ERK2 was performed by Western blot analysis using anti-erk2 Phosphorylated EKK2 was detected using anti-phospho ERK2 antibody and regarded as activated form of the enzyme. Insulin treatment caused approximately 5% reduction of ERK2 content in cytosol in 3 min whereas ERK2 content of membrane were increased to 25 to 4% over basal in 3 rain (Fig. 2), suggesting ERK2 appeared in the membrane in response to insulin might be transported from cytoplasm. Tyrosine phosphorylated ERK2 was measured with anti-phosphospecific ERK2 antibody to address the activity of ERK2. Insulin stimulated phosphorylation of ERK2 in membrane by 2-fold and 1.8-fold at 1 rain and 1 min, respectively. Insulin also stimulated phosphorylation ofcytosolic ERK2 by 2.7-fold and 2.3-fold at 1 rain and 1 min, respectively. These insu. fin, stimulated phosphorylation of membrane and cytosolic ERK2 decreased to basal levels in 3 rain. On the contrary, the phosphorylation of nuclear ERK2 was stimulated by insulin in time-dependent manner reaching maximal activity (3- fold) at 3 rain. (Fig. 3). 125

4 A. O I- B. O I- C. O I, :~ M C N M C N M C N Figure 1: Subcellular distribution of lactate dehydrogenase, ERK H and MEK H in UMR-16 cell UMP,-16 cells were cultured to confluence, scaped into isolation buffer, and followed by homogenization. The cell homogenates were subjected to differential centrifugation to isolate membrane, cytosol, and nucleus as described in Materials and Methods. Each fraction was subjected to LDH assay (A). ERK II (B) and MEK II (C) contents were analyzed by SDS-PAGE, Western blot analysis and scanning densitometry. Data were presented as mean 4-- S.E.M. (n=3). Abbreviation : M, membrane ; C, cytosol ; N, nucleus. Effect of insulin on the level of MEK2 in the membrane Since MEK2 is known as an upstream activator of ERK2, we explored the possibility that MEK2 could translocate into membrane in response to insulin to phosphorylate membrane-associated ERK2. Insulin increase the content of MEK2 in the membrane by 2.6-fold and 2.2-fold in l min and 1 min, respectively (Fig. 4). The time course ofmek2 translocation into membrane and fold-stimulation by insulin were similar to those oferk2 phosphorylation in membrane by insulin (see Fig. 3). 126

5 BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL Discussion Considering bone cells contain a number of growth factors such as IGF-1, PDGF, TGF-13, IL, TNF (18), it is obvious that growth factor-mediated signal transduction plays significant roles in the functions of bone cells. The recent identification of insulin receptors in osteoblasts prompted us to study insulin action in UMR-16 cell (15). Since ERK2 is an important target of insulin actions (1,2), we tested to determine if ERK2 is regulated by insulin in subcellular compartments ofumr-16 cells. A. Membrane Cytosol Nucleus i i J, f 1 Blot : Time (rain) B o 2.~ a membrane ~ 1 r.cj "-G o.5 leus... cytosol [] Time (min) Figure 2: Distribution of ERK II in response to insulin in UMR-16 cell After treatment with insulin (1 nm) for the indicated durations, UMR-16 cells were subjected to subcellular fractionation as described in the Materials and Methods section. Equal amount (2 ~g) of membrane, cytosolic and nuclear fractions were applied to SDS/polyacrylamide gel, followed by electrophoresis and transfer to nitrocellulose membranes. The membranes were probed with anti-erk II antibody and detected with an enhanced chemiluminescence system in accordance with the protocol provided by the manufacturer (NEN) (A). The content oferk II in membrane, cytosol and nucleus was quantified by scanning densitometry (B). The results are expressed as average of two separate experiments. 127

6 BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL A. Membrane u r Cytosol I Nucleus Blot: i anti-perk2 Tree(rain) B ~ o +.~ 2.5 nucleus....-~ 2 r 1.5 o 1.5 I Time (min) Figure 3: Phosphorylation of ERK II in response to insulin in UMR-16 cell After treatment with insulin (1 nm) for the indicated durations, UMR-16 cells were subjected to subcellular fractionation as described in the Materials and Methods section. Equal amount (2 I~g) of membrane, cytosolic and nuclear fractions were applied to SDS/polyacrylamide gel, followed by eleetrophoresis and transfer to nitrocellulose membranes. The membranes were probed with anti-phospho ERK II antibody and detected with an enhanced chemiluminescence system in accordance with the protocol provided by the manufacturer (NEN) (A). The content of phosphorylated ERK I] in membrane, cytosol and nucleus was quantified by scanning densitometry (B). The results are expressed as average &two separate experiments. Cytosolic marker enzyme assay indicates that membrane and nuclear fractions prepared in the present study are free of significant contamination from cytosol. Thus, contamination of cytosolic ERK2 and MEK2 in the membrane and nuclear preparation could not account for the presence of these enzymes. Relatively high levels of ERK2 and MEK2 in membrane and nucleus suggest that ERK2 could participate in membrane and nuclear functions in osteoblasts. The high levels of ERK2 and MEK2 in the nucleus 128

7 A Membrane Cytosol Nucleus [ lr 1 i 1 ~'~ ~ ~=~ ), = ~i~!~ Blot " _anti-mek2 Time (min) B 9 membfane o o,..~ 4-~ m ,> Time (min) Figure 4: Translocation of MEK II into memebrane in response to insulin in UMR-16 cell After treatment with insulin (1 nm) for the indicated durations, UMR-i6 cells were subjected to subcellular fractionation as described in the Materials and Methods section. Equal amount (2 ~g) of membrane, cytosolic and nuclear fractions were applied to SDS/polyacrylamide gel, followed by electrophoresis and transfer to nitrocellulose membranes. The membranes were probed with anti-mek II antibody and detected with an enhanced chemiluminescence system in accordance with the protocol provided by the manufacturer (NEN) (A). The content of MEK I1 in membrane, cytososl and nucleus was quantified by scanning densitometry (B). The results are expressed as average of two separate experiments. 129

8 BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL were also identified in other cell lines such as Chinese Hamster Ovary (CHO) cell overexpressing human insulin receptor and Fao Fepatoma cell (12). The membraneassociated ERK2 and MEK2 were recovered from 33, x g pellets solubilized with Triton X-1, suggesting that ERK2 and MEK2 may associate with membrane lipids by hydrophobic interaction or they may be tightly associated with integral membrane proteins. The 33, x g pellets are plasma membrane-enriched fraction which is used to isolate membrane receptors such as insulin receptors (19). Thus, the majority of the ERK2 and MEK2 identified in membrane fractions in the present study is probably associated with plasma membrane of the cell. Insulin had some effect on the subcellular distribution of ERK2. The increase in ERK2 in the membrane in response to insulin may be caused by translocation of the enzyme from cytoplasm to the membrane, considering the corresponding decrease in ERK2 in cytoplasm. Insulin stimulated the phosphorylation of ERK2 in the membrane as well as in cytosol and the nucleus. The time course oferk2 phosphorylation in the membrane and eytosol is more rapid than that in the nucleus. Since ERK2 activation requires tyrosine and threonine phosphorylation of the enzyme by MEK (8-11), the insulin-induced ERK2 phosphorylation represent ERK2 activation in the cell. In terms of understanding an upstream kinase responsible for ERK2 phosphorylation induced by insulin, we tested to determine if insulin causes MEK2 translocation into the membrane and nucleus, since MEK2 is known to activate ERK2 via phosphorylation (8). Interestingly, insulin caused significant increase in the amount of MEK2 in membranes with similar extent and kinetics to those Of phosphorylation of membrane ERK2 by insulin. This raises an interesting possibility that MEK2 translocated into the membrane could phosphorylate the membrane-associated ERK2. Recently, it has been found that ERK1 and ERK2 associate with p75 nerve growth factor receptors present in the plasma membrane of PC 12 cell (21). Likewise, the insulin-stimulated ERK2 in the membrane of UMR-16 cell could interact with membrane receptors, regulating their functions. In summary, insulin causes translocation of ERK2 from cytoplasm to membrane and it simultaneously activates ERK2 in the membrane as well as in cytosol and nucleus. Insulin also rapidly increase the content of MEK2 in the membrane. We propose that MEK2 translocated into the membrane in response to insulin may phosphorylate the 13

9 membrane-associated ERK2, activating it. The activated ERK2 in membrane could regulate membrane-bound proteins such as membrane receptors by phosphorylation. The insulin-stimulated nuclear ERK2 could phosphorylate DNA binding proteins, resulting in the regulation of transcription of osteoblast genes. Acknowledgments This work was supported in part by grants from Korea Science and Engineering Foundation (project # ), Ministry of Education (96' Basic Medical Science Grant) and Ministry of Health and Welfare, KOREA. The authors acknowledge support from Kyung-Hee University. References 1. Lawrence, J. C., Roach, P. J. (1997) Diabetes 46, Kyriakis, J. M., App, H., Zhang, X. F., Banerjee, P., Brautigan, D. L., Rapp, U. R., and Avuruch, J. (1992) Nature (London) 358, Lange-earter, C. A., Pleiman, C. M., Gardner, A. M., Blumer, K J., and Johnson, G L. (1993) Science 26, Marshall, C. J. (1994) Curr. Opin. Genet. Dev. 4, Kolch, W., Heidecker, G., Kochs, G., Hummel, R., Vahidi, H., Mischak, H., Finkenzeller, G., Marme, D., Rapp, U. R. (1993)Nature 364, Pulverer, B. J., Kyriakis, J. M., Avruch, J., Nikolakaki, E., and Woodgett, J. R. (1991) Nature 353, Boulton, T. G., Nye, S. H., Robbins, D. J., Ip, N. Y., Radziejewska, E., Morgenbesser, S. D., Depinho, R. A., Panayotatos N, Cobb, M. H., Yancopoulos, G. D. (1991) Cell 65, Camarillo, I. G., Linebaugh, B. E, Rillema, J. A. (1997) Proc. Soc. Exp. Biol. Med. 215, Crews, C., Alessandrini, A., Erikson, R. L. (1992) Science 258, Nishida, E. and Kotoh, Y. (1993)Trends. Biochem Sci. 18, Zheng, C. F., Guan, K. L. (1993) J. Biol. Chem. 268, Kim, S. -J. and Kahn, C. R. (1997) Biochem. J. 323, Mizukami, Y., Yoshida, K. (1997) Biochem. J. 323, Kim, S. -J. and Kahn, C. R. (1994)J. Biol. Chem. 269, Thomas, D. M., Hards, D. K., Rogers, S. D., NG, K W., Best, J. D. (1996) J. Bone. Miner. Res. 11, Chen, R. H., Sarnecki, C., Blenis, J. (1992) Mol. Cell. Biol. 12, Towbin, H., Staehelin, J. and Gordon, J. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, Manolagas, S.C., Jilka, R. (1995)N. Engl. J. Med. 232, Kim, S.-L, Kim, H., Pillion, D. J. (1991). Biochem. Biophys. Res. Commun. 179: Kim, S.-J., Kahn, C. R. (1995) J. Biomed. Res. 5, Volente, C., Angelastro, J. M., Geene, L. A. (1993) J. Biol. Chem. 268,