- Email: [email protected]
Protein Kinase C Modulates the Insulin-Stimulated Increase in Akt1 and Akt3 Activity in 3T3-L1 Adipocytes
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
243, 509–513 (1998)
RC988134
Protein Kinase C Modulates the Insulin-Stimulated Increase in Akt1 and Akt3 Activity in 3T3-L1 Adipocytes Andreas Barthel, Kaname Nakatani, Ajai A. Dandekar, and Richard A. Roth Department of Molecular Pharmacology, Stanford University School of Medicine, Stanford, California 94305
Received January 8, 1998
In the present studies, we have compared the properties of two members of the Akt family of ser/thr kinases, Akt1 and Akt3. First, we demonstrate that both 3T3-L1 fibroblasts and adipocytes express Akt3 mRNA by RT-PCR and sequencing of the resultant PCR product. Second, we show that insulin stimulates the enzymatic activity of Akt1 and Akt3 15- and 7-fold, respectively. We then investigated the ability of protein kinase C to regulate Akt1 and 3. Neither enzyme was activated by stimulation of protein kinase C, however, the insulin-stimulated increases in activity of both isozymes were found to be comparably inhibited by prior protein kinase C activation. Since this inhibition could have resulted from an interaction of the pleckstrin homology domain of the Akt with protein kinase C, we also examined the ability of a mutant Akt1 lacking this domain to be regulated by this enzyme. The insulinstimulated increase in enzymatic activity of this mutant Akt was regulated by PKC activation like the wild type enzyme. These results indicate that Akt1 and 3 are similarly stimulated by insulin and this stimulation is inhibited by prior activation of protein kinase C through a mechanism that is independent of the presence of the pleckstrin homology domain. q 1998 Academic Press
Akt1 (also called PKB and Rac) is a 60 kDa pleckstrin homology domain containing ser/thr kinase that has been shown to be functionally located downstream of PI 3-K and to become activated by a variety of growth factors including insulin (1-4). Constitutively active forms of Akt1 have been shown to induce a variety of biological effects including: activation of the 70 kDa S6 kinase (5); inhibition of glycogen synthase kinase-3 (6); inhibition of apoptosis by stimulation of BAD phosphorylation (7); stimulation of differentiation of 3T3-L1 fiAbbreviations: IRS-1, insulin receptor substrate-1; PI 3-K, phosphatidylinositol 3-kinase; PMA, phorbol 12-myristate 13-acetate; PKC, protein kinase C; WT, wild-type; PH, pleckstrin homology.
broblasts into adipocytes (8); stimulation of glucose uptake and GLUT4 translocation in 3T3-L1 adipocytes (8); and induction of leptin production in 3T3-L1 adipocyes (9). Since all of these biological responses can also be induced by insulin treatment of cells, it is possible that Akt1 is involved as a mediator of these biological responses. Activation of Akt1 has been proposed to be via binding of PI 3-phosphates to its PH domain (10) or alternatively via its phosphorylation (11). Two regulatory phosphorylation sites in Akt1 have been identified, Thr308 and Ser473 (12) . Most recently, a PI 3-phosphate activatable kinase has been identified which can phosphorylate Akt1 at Thr308 (11). Although three members of the Akt ser/thr kinase family have been identified (13) , most studies have focused on Akt1. Of the three isoforms, Akt3 appears to differ most. For example, it lacks one of the two key regulatory phosphorylation sites present in Akt1 and 2, Ser473. It also was reported to have a more limited tissue expression distribution then the other two enzymes, being primarily localized in brain and testis (13). In the present studies we demonstrate that Akt3 is also present in 3T3-L1 fibroblasts and adipocytes. We show that this enzyme, like Akt1, can also be stimulated by insulin. Furthermore, we show that prior activation of protein kinase C can inhibit the subsequent ability of insulin to stimulate the enzymatic activities of both Akt1 and 3. Since, as noted above, the activation of Akt may mediate the ability of insulin to stimulate various biological responses, its negative regulation by protein kinase C could contribute to insulin resistant states and may therefore play a role in the pathogenesis of non-insulin-dependent diabetes mellitus (14). MATERIALS AND METHODS Materials. Cell culture materials were from UCSF Cell Culture Facility (San Francisco, CA). Total RNA was isolated using the RNeasy kit from Quiagen (Chatsworth, CA). The RT-PCR kit was from Stratagene (La Jolla, CA). The primers were from Operon Technologies Inc. (Alameda, CA). [g-32P]-ATP (3000Ci/mmol) was from NEN (Boston, MA). Akt1 and 3-specific polyclonal antibodies were 0006-291X/98 $25.00
509
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC
Vol. 243, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
from UBI (Lake Placid, NY), horseradish-peroxidase conjugated antirabbit immunglobulin was from Amersham Life Science (Cleveland, Ohio), Acrylamide was from National Diagnostics (Atlanta, GA), Nitrocellulose membranes were from Schleicher & Schuell (Keene, NH), insulin and anti-HA monoclonal antibody (12CA5) were from Boehringer Mannheim (Mannheim, Germany). Nucleotides and Taq polymerase were from Gibco Life Technologies (Grand Island, NY). PMA was from Calbiochem (San Diego, CA). Protein A-Sepharose was from Repligen (Cambridge, MA), Protein G-Sepharose was from Pharmacia (Uppsala, Sweden). All other chemicals were from Sigma (St. Louis, MO). Automated DNA sequencing was performed in the Beckman facility (Stanford, CA). Cell culture and differentiation of 3T3-L1 adipocytes. Parental 3T3-L1 fibroblasts and cells expressing either the wildtype Akt or DPH(4-129)Akt were grown and induced to differentiate as described previously (8). The differentiated adipocytes were kept for at least 2 additional days before they were used. The state of differentiation was verified microscopically before use and was comparable for the different cell lines. RNA extraction and RT-PCR. Total RNA was extracted from 3T3L1 adipocytes and fibroblasts and its integrity was verified by agarose gel electrophoresis. The RNA concentration was determined spectrophotometrically. Reverse transcription (10 mg total RNA/sample) was performed in a volume of 50 ml using oligo-dT as primer. Four ml of the resulting cDNAs were subjected to PCR (100 ml total volume, initial denaturation step 2 min 947C, denaturation 1 min 947C, annealing and elongation 2 min 727C, 40 cycles) to generate a 602 bp Akt3 specific fragment using primers spanning bases 340-375 (GAGAAGGATCCATTGTAGTCCAACGTCACAGATT) and bases 904-941 (GCCAGGAATTCTGGTGTACCACAGAATGTCTTCATGGT) according to the numbering used by Konishi et al. (13). Akt assays. After the indicated treatments, cells were chilled on ice, washed once with cold PBS, lysed in 800 ml buffer/dish ( 50 mM Hepes, pH 7.6, 150 mM NaCl, 10% (v/v) glycerol, 1% (v/v) Triton X100, 1 mg/ml bacitracin, 1 mM PMSF, 1 mM benzamidine, 1 mM Na3VO4 , 30 mM NaPPi, 10 mM NaF, 1 mM EDTA, 1 mM DTT, 100 nM okadaic acid ) and the lysate spun for 10 min at 14000 rpm in a tabletop centrifuge. The supernatants were immunoprecipitated for 2 h with 4 mg of either anti-HA, anti-Akt1, or anti-Akt3 antibodies or 2.5 ml anti-Akt1 PH-domain antiserum bound to 25 ml Protein A or G-Sepharose beads, washed three times with 1 ml buffer 2 (25 mM Hepes, pH 7.8, 1% BSA, 10% glycerol, 1 mM DTT, 1% Triton X100, 1 M NaCl) and two times with kinase reaction buffer (50 mM Tris-HCl, pH 7.5, 10 mM MgCl2 , 1 mM DTT). The phosphorylation reaction with GSK-3 peptide as substrate (GRPRTSSFAEG) was performed in 30 ml kinase mix ( 50 mM Tris/HCl, pH 7.5, 10 mM MgCl2 , 1 mM DTT, 5 mM ATP, 100 nM GSK-3 peptide, 2 mCi [g-32P]-ATP) for 30 min at 307C. The reaction was stopped by adding the supernatant to 10 ml gel loading buffer (0.25 M Tris-HCl, pH 6.8, 12 M urea). The phosphorylated peptide was separated from free [32P]-ATP on a 40% polyacrylamide gel containing 6 M urea according to West et al. (15). The gels were dried, autoradiographed and the phosphopeptide spots excised and counted. The immunoprecipitates were subjected to 10% SDS-PAGE and blotted onto nitrocellulose to check for equal amounts of Akt protein in the assay. The blots were probed for Akt1 with a polyclonal antiserum directed against the PH domain or a C-terminal peptide of Akt1 and then incubated with a peroxidase conjugated secondary antibody and peroxidase activity was visualized by ECL.
RESULTS Insulin Stimulation and Expression of Akt3 in 3T3-L1 Adipocytes In an attempt to study the activation of different Akt isoforms in an insulin responsive cell (16) , 3T3-L1 adipo-
cytes were treated with insulin and the enzymatic activity of the two isoforms were determined in an in vitro phosphorylation reaction using a peptide substrate (6) after immunoprecipitation with isoform specific antibodies. Both isoforms were found to be rapidly stimulated by insulin, the Akt1 isoform was maximally stimulated 15-fold whereas Akt3 was stimulated 7-fold (Fig. 1A). The specificity of the antibodies used in the immunoprecipitations was tested by Western blotting. When the immunoprecipitates were probed with an antibody raised against the PH domain of Akt1, a reaction was observed only with the precipitates to the Akt1 isoform (Fig 1B), demonstrating that the Akt3 antibodies do not immunoprecipitate appreciable amounts of Akt1. Since expression of Akt3 had not been previously reported in 3T3-L1 cells and the anti-Akt3 antibody used for the immunoprecipitations was not suitable for Western blotting, we performed RT-PCR with a primer corresponding to a unique sequence of the rat Akt3 and another primer derived from a common region. Both the 3T3-L1 fibroblasts and adipocytes showed the expected 602 bp fragment (Fig. 1C). To confirm the identity of this band, the PCR fragment was purified and sequenced. The nucleotide sequence of this mouse cDNA was more than 97% (563/580) identical to the known rat Akt3 sequence and the deduced amino acid sequence was 100% identical to the rat sequence (13). Effect of PKC Activation on the Insulin-Stimulated Increases in Akt1 and Akt3 Activities To further compare the regulation of the enzymatic activities of Akt1 and 3, cells were treated with PMA to activate PKC with or without a subsequent stimulation with insulin. As previously observed with Akt1 (17), PMA treatment did not stimulate Akt3. However, PMA-treatment was found to inhibit the subsequent ability of insulin to stimulate Akt enzymatic activity in both the fibroblasts as well as the adipocytes. Time course studies showed that after only a 10 min pretreatment with 1 mM PMA the insulin-stimulated activation of Akt1 was maximally inhibited (Fig. 2). This inhibition of Akt1 activation by PMA-pretreatment was observed at a range of insulin concentrations tested (Fig. 3). The overall range of inhibition varied from about 30 to 50 % with an average of 40{7% (nÅ7) inhibition. Finally PMA pretreatment was also found to inhibit the insulin-stimulated increase in enzymatic activity of the Akt3 isoform by 35{10% (nÅ3), a value comparable to that observed with Akt1 (Fig. 4). Since there have been reports indicating a direct interaction of several PKC-isoforms with the PH domain of Akt (13), we compared the ability of PKC to regulate either an expressed intact Akt1 (WT-Akt1) or a mutant lacking the PH-domain (DPH-Akt1), both of which were epitope-tagged. Cells were pretreated or not with PMA, insulin-stimulated, lysed, and the lysates were
510
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC
Vol. 243, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. PMA-pretreatment rapidly inhibits the insulin-stimulated increase in Akt activity. 3T3-L1 fibroblasts expressing a HAtagged WT-Akt1 were preincubated for the indicated times with PMA (1 mM) and then stimulated with 1 mM insulin for another 10 min, chilled on ice and lysed. WT-Akt1 was immunoprecipitated using anti-HA antibodies and its activity was measured as described.
DISCUSSION In the present study, we have compared the regulation of two Akt isoforms in an insulin responsive cell, the 3T3-L1 cells. Akt3, the most recently identified isozyme of the Akt family and its most distinct member (13), was shown to be expressed in 3T3-L1 adipocytes and fibroblasts and found to be stimulated about 7-fold by insulin. This is the first time that expression and activation of Akt3 has been demonstrated in an insulin responsive cell. The total amount of Akt3 enzymatic activity was about 50% of that observed for Akt1 as measured in an in vitro phosphorylation reaction after
FIG. 1. Stimulation of Akt isoforms 1 and 3 by insulin. (A) 3T3L1 adipocytes were stimulated for 5 min with 100 nM insulin, lysed and the cell lysates immunoprecipitated with antibodies directed against each isoform. The immunoprecipitates were assayed for enzymatic activity via the use of a peptide substrate. The activity is expressed as the fold stimulation as compared to the activity in nonstimulated cells and are means{SEM for 3 experiments. (B) Immunoblotting of Akt1 and Akt3 immunoprecipitates. The immunoprecipitates from the 3T3-L1 adipocytes were blotted with a polyclonal antiserum directed against the pH domain of Akt1. (C) Expression of Akt3 mRNA in 3T3-L1 fibroblasts and adipocytes. cDNA was prepared from RNA of 3T3-L1 fibroblasts and adipocytes and was utilized in RT-PCR with primers specific for Akt3 and the reaction mixtures were analyzed on a 1% agarose gel. The band derived from adipocytes was excised and its identity was confirmed by sequencing.
immunoprecipitated with an antibody to the epitope. PMA-pretreatment inhibited the insulin-stimulation of the DPH-Akt1 by 45{9% (nÅ4) a value comparable to that observed with the WT-Akt (Fig. 5).
FIG. 3. PMA-pretreatment inhibits the insulin-stimulated increase in Akt activity at various insulin concentrations. 3T3-L1 adipocytes were treated with either 1 mM PMA (white bars) or with only DMSO (black bars) prior to stimulation with insulin at the indicated doses. After cell lysis, endogenous Akt1 was immunoprecipitated and the precipitates assayed for kinase activity.
511
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC
Vol. 243, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
the pathophysiology of noninsulin dependent diabetes mellitus (14). We therefore examined the ability of PKC to regulate Akt1 and 3. Treatment of 3T3-L1 adipocytes with PMA did not stimulate the enzymatic activity of either isoform, However, pretreatment with PMA resulted in an inhibition of the insulin-stimulated increase in both Akt1 and 3 enzymatic activity. The extent of inhibition observed, approximately 40%, is similar to the observed degree of inhibition of the insulin-stimulated increase in PI 3-kinase activity, consistent with this enzyme being upstream of Akt activation (18). These findings are also consistent with the recent observations of a decreased ability of insulin to stimulate Akt enzymatic activity in the muscle of hyperglycemic rats, a condition known to activate PKC (19). The PH domain has been proposed to be involved in protein-protein-interactions such as binding of b/g subunits of heterotrimeric G proteins or PKC (20). Using GST fusion proteins of the three different Akt PH domains as bait, Konishi et al. (13) demonstrated in vitro binding of PKCa and d isoforms to the PH do-
FIG. 4. PMA-pretreatment inhibits the insulin-stimulated increase in Akt1 and Akt3 activity. (A) The expressed Akt1 was immunoprecipitated from transfected 3T3-L1 adipocytes stimulated either with or without prior PMA-pretreatment. (B) Endogenous Akt3 was immunoprecipitated from the 3T3-L1 adipocytes stimulated either with or without prior PMA-treatment.
immunoprecipitation. Differences in enzymatic activity may be due to differences in the levels of these two isoforms as well as differences in affinities of the antibodies used for the immunoprecipitations or to differences in the abilities of the two isoforms to utilize GSK3 peptide in the kinase assay. The finding that Akt3 is similarly regulated as Akt1 is particularly surprising considering that Akt3 is missing a key regulatory phosphorylation site, Ser473 (12, 13). Modulation of insulin stimulated responses by activated PKC has been demonstrated at various levels of the signalling cascade such as the receptor itself and PI-3-kinase, possibly via modulation of the ser/thr phosphorylation of the insulin receptor and/or its substrates like insulin receptor substrate-1 (18 and references therein). Since activation of PKC has been observed in animals and humans in association with various insulin resistant states, it is a potential player in
FIG. 5. PMA-pretreatment inhibits the insulin-stimulated increase in enzymatic activity of an Akt1 mutant lacking its PH domain. 3T3-L1 fibroblasts expressing either an HA-epitope tagged WT-Akt1 or a DPH-Akt1 were pretreated with or without 1 mM PMA for 15 min, stimulated with the indicated concentrations of insulin, lysed and the expressed enzyme was immunoprecipitated with antiHA antibodies and the precipitates assayed for activity.
512
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC
Vol. 243, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
mains of Akt1, 2 and 3 and binding of PKCz to the PH domains of Akt1 and 2. We therefore raised the question whether the observed modulation of insulin stimulated Akt activity by PKC-activation requires the PH domain. Our studies performed on 3T3-L1 fibroblasts stably expressing either WT-Akt1 or a mutant lacking the PH domain (D4-129-Akt1) showed that the PKCmediated inhibition of the insulin-stimulation in Akt activity was independent of the PH domain. These data suggest that modulation of the insulin-stimulated increase in Akt activity by PKC activation occurs at some upstream step in Akt activation, possibly by modulating the insulin-stimulated increase in PI 3-kinase activity. In summary, the present studies demonstrate that both Akt1 and 3 are similarly stimulated by insulin in a model insulin responsive cell, the 3T3-L1 adipocytes. Moreover, the insulin-stimulated activation of both isoforms are negatively regulated by prior PKC activation, possibly explaining the negative regulation of this enzyme in models of insulin resistance. ACKNOWLEDGMENTS This work was supported in part by NIH grant DK 34926 (to RAR) and a Feodor Lynen fellowship of the Alexander von Humboldt Stiftung (to AB).
REFERENCES 1. Jones, P. F., Jakubowicz, T., Pitossi, F. J., Maurer, F., and Hemmings, B. A. (1991) Proc. Natl. Acad. Sci. USA 88, 4171–4175.
2. Coffer, P. J., and Woodgett, J. R. (1991) Eur. J. Biochem. 201, 475–481. 3. Bellacosa, A., Testa, J. R., Staal, S. P., and Tsichlis, P. N. (1991) Science 254, 274–277. 4. Bos, J. L. (1995) TIBS 20, 441–442. 5. Burgering, B. M. T., and Coffer, P. J. (1995) Nature 376, 599– 602. 6. Cross, D. A. E., Alessi, D. R., Cohen, P., Andjelkovich, M., and Hemmings, B. A. (1995) Nature 378, 785–789. 7. Hemmings, B. A. (1997) Science 275, 628–630. 8. Kohn, A. D., Summers, S. A., Birnbaum, M. J., and Roth, R. A. (1996) J. Biol. Chem. 271, 31372–31378. 9. Barthel, A., Kohn, A. D., Luo, Y., and Roth, R. A. (1997) Endocrinol. 138, 3877–3888. 10. Toker, A., and Cantley, L. C. (1997) Nature 387, 673–676. 11. Alessi, D., James, S., Downes, C., Holmes, A., Gaffney, P., Reese, C., and Cohen, P. (1997) Curr. Biol. 7, 261–269. 12. Alessi, D. R., Andjelkovic, M., Caudwell, B., Cron, P., Morrice, N., Cohen, P., and Hemmings, B. A. (1996) EMBO J. 15, 6541– 6551. 13. Konishi, H., Kuroda, S., Tanaka, M., Matsuzaki, H., Ono, Y., Kameyama, K., Haga, T., and Kikkawa, U. (1995) Biochem. Biophys. Res. Comm. 216, 526–534. 14. Considine, R. V., and Caro, J. F. (1993) J. Cell. Biochem. 52, 8– 13. 15. West, M. H. P., Wu, R. S., and Bonner, W. M. (1984) Electrophoresis 5, 133–138. 16. Green, H., and Kehinde, O. (1974) Cell 1, 113–116. 17. Kohn, A. D., Kovacina, K. S., and Roth, R. A. (1995) EMBO J. 14, 4288–4295. 18. DeFea, K., and Roth, R. A. (1997) J. Biol. Chem. 272, 31400– 31407. 19. Krook, A., Kawano, Y., Song, X. M., Efendic, S., Roth, R. A., Wallberg-Henriksson, H., and Zierath, J. R. (1997) Diabetes 46, 2110–2114. 20. Pawson, T., and Scott, J. D. (1997) Science 278, 2075–2080.
513
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC
243, 509–513 (1998)
RC988134
Protein Kinase C Modulates the Insulin-Stimulated Increase in Akt1 and Akt3 Activity in 3T3-L1 Adipocytes Andreas Barthel, Kaname Nakatani, Ajai A. Dandekar, and Richard A. Roth Department of Molecular Pharmacology, Stanford University School of Medicine, Stanford, California 94305
Received January 8, 1998
In the present studies, we have compared the properties of two members of the Akt family of ser/thr kinases, Akt1 and Akt3. First, we demonstrate that both 3T3-L1 fibroblasts and adipocytes express Akt3 mRNA by RT-PCR and sequencing of the resultant PCR product. Second, we show that insulin stimulates the enzymatic activity of Akt1 and Akt3 15- and 7-fold, respectively. We then investigated the ability of protein kinase C to regulate Akt1 and 3. Neither enzyme was activated by stimulation of protein kinase C, however, the insulin-stimulated increases in activity of both isozymes were found to be comparably inhibited by prior protein kinase C activation. Since this inhibition could have resulted from an interaction of the pleckstrin homology domain of the Akt with protein kinase C, we also examined the ability of a mutant Akt1 lacking this domain to be regulated by this enzyme. The insulinstimulated increase in enzymatic activity of this mutant Akt was regulated by PKC activation like the wild type enzyme. These results indicate that Akt1 and 3 are similarly stimulated by insulin and this stimulation is inhibited by prior activation of protein kinase C through a mechanism that is independent of the presence of the pleckstrin homology domain. q 1998 Academic Press
Akt1 (also called PKB and Rac) is a 60 kDa pleckstrin homology domain containing ser/thr kinase that has been shown to be functionally located downstream of PI 3-K and to become activated by a variety of growth factors including insulin (1-4). Constitutively active forms of Akt1 have been shown to induce a variety of biological effects including: activation of the 70 kDa S6 kinase (5); inhibition of glycogen synthase kinase-3 (6); inhibition of apoptosis by stimulation of BAD phosphorylation (7); stimulation of differentiation of 3T3-L1 fiAbbreviations: IRS-1, insulin receptor substrate-1; PI 3-K, phosphatidylinositol 3-kinase; PMA, phorbol 12-myristate 13-acetate; PKC, protein kinase C; WT, wild-type; PH, pleckstrin homology.
broblasts into adipocytes (8); stimulation of glucose uptake and GLUT4 translocation in 3T3-L1 adipocytes (8); and induction of leptin production in 3T3-L1 adipocyes (9). Since all of these biological responses can also be induced by insulin treatment of cells, it is possible that Akt1 is involved as a mediator of these biological responses. Activation of Akt1 has been proposed to be via binding of PI 3-phosphates to its PH domain (10) or alternatively via its phosphorylation (11). Two regulatory phosphorylation sites in Akt1 have been identified, Thr308 and Ser473 (12) . Most recently, a PI 3-phosphate activatable kinase has been identified which can phosphorylate Akt1 at Thr308 (11). Although three members of the Akt ser/thr kinase family have been identified (13) , most studies have focused on Akt1. Of the three isoforms, Akt3 appears to differ most. For example, it lacks one of the two key regulatory phosphorylation sites present in Akt1 and 2, Ser473. It also was reported to have a more limited tissue expression distribution then the other two enzymes, being primarily localized in brain and testis (13). In the present studies we demonstrate that Akt3 is also present in 3T3-L1 fibroblasts and adipocytes. We show that this enzyme, like Akt1, can also be stimulated by insulin. Furthermore, we show that prior activation of protein kinase C can inhibit the subsequent ability of insulin to stimulate the enzymatic activities of both Akt1 and 3. Since, as noted above, the activation of Akt may mediate the ability of insulin to stimulate various biological responses, its negative regulation by protein kinase C could contribute to insulin resistant states and may therefore play a role in the pathogenesis of non-insulin-dependent diabetes mellitus (14). MATERIALS AND METHODS Materials. Cell culture materials were from UCSF Cell Culture Facility (San Francisco, CA). Total RNA was isolated using the RNeasy kit from Quiagen (Chatsworth, CA). The RT-PCR kit was from Stratagene (La Jolla, CA). The primers were from Operon Technologies Inc. (Alameda, CA). [g-32P]-ATP (3000Ci/mmol) was from NEN (Boston, MA). Akt1 and 3-specific polyclonal antibodies were 0006-291X/98 $25.00
509
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC
Vol. 243, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
from UBI (Lake Placid, NY), horseradish-peroxidase conjugated antirabbit immunglobulin was from Amersham Life Science (Cleveland, Ohio), Acrylamide was from National Diagnostics (Atlanta, GA), Nitrocellulose membranes were from Schleicher & Schuell (Keene, NH), insulin and anti-HA monoclonal antibody (12CA5) were from Boehringer Mannheim (Mannheim, Germany). Nucleotides and Taq polymerase were from Gibco Life Technologies (Grand Island, NY). PMA was from Calbiochem (San Diego, CA). Protein A-Sepharose was from Repligen (Cambridge, MA), Protein G-Sepharose was from Pharmacia (Uppsala, Sweden). All other chemicals were from Sigma (St. Louis, MO). Automated DNA sequencing was performed in the Beckman facility (Stanford, CA). Cell culture and differentiation of 3T3-L1 adipocytes. Parental 3T3-L1 fibroblasts and cells expressing either the wildtype Akt or DPH(4-129)Akt were grown and induced to differentiate as described previously (8). The differentiated adipocytes were kept for at least 2 additional days before they were used. The state of differentiation was verified microscopically before use and was comparable for the different cell lines. RNA extraction and RT-PCR. Total RNA was extracted from 3T3L1 adipocytes and fibroblasts and its integrity was verified by agarose gel electrophoresis. The RNA concentration was determined spectrophotometrically. Reverse transcription (10 mg total RNA/sample) was performed in a volume of 50 ml using oligo-dT as primer. Four ml of the resulting cDNAs were subjected to PCR (100 ml total volume, initial denaturation step 2 min 947C, denaturation 1 min 947C, annealing and elongation 2 min 727C, 40 cycles) to generate a 602 bp Akt3 specific fragment using primers spanning bases 340-375 (GAGAAGGATCCATTGTAGTCCAACGTCACAGATT) and bases 904-941 (GCCAGGAATTCTGGTGTACCACAGAATGTCTTCATGGT) according to the numbering used by Konishi et al. (13). Akt assays. After the indicated treatments, cells were chilled on ice, washed once with cold PBS, lysed in 800 ml buffer/dish ( 50 mM Hepes, pH 7.6, 150 mM NaCl, 10% (v/v) glycerol, 1% (v/v) Triton X100, 1 mg/ml bacitracin, 1 mM PMSF, 1 mM benzamidine, 1 mM Na3VO4 , 30 mM NaPPi, 10 mM NaF, 1 mM EDTA, 1 mM DTT, 100 nM okadaic acid ) and the lysate spun for 10 min at 14000 rpm in a tabletop centrifuge. The supernatants were immunoprecipitated for 2 h with 4 mg of either anti-HA, anti-Akt1, or anti-Akt3 antibodies or 2.5 ml anti-Akt1 PH-domain antiserum bound to 25 ml Protein A or G-Sepharose beads, washed three times with 1 ml buffer 2 (25 mM Hepes, pH 7.8, 1% BSA, 10% glycerol, 1 mM DTT, 1% Triton X100, 1 M NaCl) and two times with kinase reaction buffer (50 mM Tris-HCl, pH 7.5, 10 mM MgCl2 , 1 mM DTT). The phosphorylation reaction with GSK-3 peptide as substrate (GRPRTSSFAEG) was performed in 30 ml kinase mix ( 50 mM Tris/HCl, pH 7.5, 10 mM MgCl2 , 1 mM DTT, 5 mM ATP, 100 nM GSK-3 peptide, 2 mCi [g-32P]-ATP) for 30 min at 307C. The reaction was stopped by adding the supernatant to 10 ml gel loading buffer (0.25 M Tris-HCl, pH 6.8, 12 M urea). The phosphorylated peptide was separated from free [32P]-ATP on a 40% polyacrylamide gel containing 6 M urea according to West et al. (15). The gels were dried, autoradiographed and the phosphopeptide spots excised and counted. The immunoprecipitates were subjected to 10% SDS-PAGE and blotted onto nitrocellulose to check for equal amounts of Akt protein in the assay. The blots were probed for Akt1 with a polyclonal antiserum directed against the PH domain or a C-terminal peptide of Akt1 and then incubated with a peroxidase conjugated secondary antibody and peroxidase activity was visualized by ECL.
RESULTS Insulin Stimulation and Expression of Akt3 in 3T3-L1 Adipocytes In an attempt to study the activation of different Akt isoforms in an insulin responsive cell (16) , 3T3-L1 adipo-
cytes were treated with insulin and the enzymatic activity of the two isoforms were determined in an in vitro phosphorylation reaction using a peptide substrate (6) after immunoprecipitation with isoform specific antibodies. Both isoforms were found to be rapidly stimulated by insulin, the Akt1 isoform was maximally stimulated 15-fold whereas Akt3 was stimulated 7-fold (Fig. 1A). The specificity of the antibodies used in the immunoprecipitations was tested by Western blotting. When the immunoprecipitates were probed with an antibody raised against the PH domain of Akt1, a reaction was observed only with the precipitates to the Akt1 isoform (Fig 1B), demonstrating that the Akt3 antibodies do not immunoprecipitate appreciable amounts of Akt1. Since expression of Akt3 had not been previously reported in 3T3-L1 cells and the anti-Akt3 antibody used for the immunoprecipitations was not suitable for Western blotting, we performed RT-PCR with a primer corresponding to a unique sequence of the rat Akt3 and another primer derived from a common region. Both the 3T3-L1 fibroblasts and adipocytes showed the expected 602 bp fragment (Fig. 1C). To confirm the identity of this band, the PCR fragment was purified and sequenced. The nucleotide sequence of this mouse cDNA was more than 97% (563/580) identical to the known rat Akt3 sequence and the deduced amino acid sequence was 100% identical to the rat sequence (13). Effect of PKC Activation on the Insulin-Stimulated Increases in Akt1 and Akt3 Activities To further compare the regulation of the enzymatic activities of Akt1 and 3, cells were treated with PMA to activate PKC with or without a subsequent stimulation with insulin. As previously observed with Akt1 (17), PMA treatment did not stimulate Akt3. However, PMA-treatment was found to inhibit the subsequent ability of insulin to stimulate Akt enzymatic activity in both the fibroblasts as well as the adipocytes. Time course studies showed that after only a 10 min pretreatment with 1 mM PMA the insulin-stimulated activation of Akt1 was maximally inhibited (Fig. 2). This inhibition of Akt1 activation by PMA-pretreatment was observed at a range of insulin concentrations tested (Fig. 3). The overall range of inhibition varied from about 30 to 50 % with an average of 40{7% (nÅ7) inhibition. Finally PMA pretreatment was also found to inhibit the insulin-stimulated increase in enzymatic activity of the Akt3 isoform by 35{10% (nÅ3), a value comparable to that observed with Akt1 (Fig. 4). Since there have been reports indicating a direct interaction of several PKC-isoforms with the PH domain of Akt (13), we compared the ability of PKC to regulate either an expressed intact Akt1 (WT-Akt1) or a mutant lacking the PH-domain (DPH-Akt1), both of which were epitope-tagged. Cells were pretreated or not with PMA, insulin-stimulated, lysed, and the lysates were
510
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC
Vol. 243, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. PMA-pretreatment rapidly inhibits the insulin-stimulated increase in Akt activity. 3T3-L1 fibroblasts expressing a HAtagged WT-Akt1 were preincubated for the indicated times with PMA (1 mM) and then stimulated with 1 mM insulin for another 10 min, chilled on ice and lysed. WT-Akt1 was immunoprecipitated using anti-HA antibodies and its activity was measured as described.
DISCUSSION In the present study, we have compared the regulation of two Akt isoforms in an insulin responsive cell, the 3T3-L1 cells. Akt3, the most recently identified isozyme of the Akt family and its most distinct member (13), was shown to be expressed in 3T3-L1 adipocytes and fibroblasts and found to be stimulated about 7-fold by insulin. This is the first time that expression and activation of Akt3 has been demonstrated in an insulin responsive cell. The total amount of Akt3 enzymatic activity was about 50% of that observed for Akt1 as measured in an in vitro phosphorylation reaction after
FIG. 1. Stimulation of Akt isoforms 1 and 3 by insulin. (A) 3T3L1 adipocytes were stimulated for 5 min with 100 nM insulin, lysed and the cell lysates immunoprecipitated with antibodies directed against each isoform. The immunoprecipitates were assayed for enzymatic activity via the use of a peptide substrate. The activity is expressed as the fold stimulation as compared to the activity in nonstimulated cells and are means{SEM for 3 experiments. (B) Immunoblotting of Akt1 and Akt3 immunoprecipitates. The immunoprecipitates from the 3T3-L1 adipocytes were blotted with a polyclonal antiserum directed against the pH domain of Akt1. (C) Expression of Akt3 mRNA in 3T3-L1 fibroblasts and adipocytes. cDNA was prepared from RNA of 3T3-L1 fibroblasts and adipocytes and was utilized in RT-PCR with primers specific for Akt3 and the reaction mixtures were analyzed on a 1% agarose gel. The band derived from adipocytes was excised and its identity was confirmed by sequencing.
immunoprecipitated with an antibody to the epitope. PMA-pretreatment inhibited the insulin-stimulation of the DPH-Akt1 by 45{9% (nÅ4) a value comparable to that observed with the WT-Akt (Fig. 5).
FIG. 3. PMA-pretreatment inhibits the insulin-stimulated increase in Akt activity at various insulin concentrations. 3T3-L1 adipocytes were treated with either 1 mM PMA (white bars) or with only DMSO (black bars) prior to stimulation with insulin at the indicated doses. After cell lysis, endogenous Akt1 was immunoprecipitated and the precipitates assayed for kinase activity.
511
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC
Vol. 243, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
the pathophysiology of noninsulin dependent diabetes mellitus (14). We therefore examined the ability of PKC to regulate Akt1 and 3. Treatment of 3T3-L1 adipocytes with PMA did not stimulate the enzymatic activity of either isoform, However, pretreatment with PMA resulted in an inhibition of the insulin-stimulated increase in both Akt1 and 3 enzymatic activity. The extent of inhibition observed, approximately 40%, is similar to the observed degree of inhibition of the insulin-stimulated increase in PI 3-kinase activity, consistent with this enzyme being upstream of Akt activation (18). These findings are also consistent with the recent observations of a decreased ability of insulin to stimulate Akt enzymatic activity in the muscle of hyperglycemic rats, a condition known to activate PKC (19). The PH domain has been proposed to be involved in protein-protein-interactions such as binding of b/g subunits of heterotrimeric G proteins or PKC (20). Using GST fusion proteins of the three different Akt PH domains as bait, Konishi et al. (13) demonstrated in vitro binding of PKCa and d isoforms to the PH do-
FIG. 4. PMA-pretreatment inhibits the insulin-stimulated increase in Akt1 and Akt3 activity. (A) The expressed Akt1 was immunoprecipitated from transfected 3T3-L1 adipocytes stimulated either with or without prior PMA-pretreatment. (B) Endogenous Akt3 was immunoprecipitated from the 3T3-L1 adipocytes stimulated either with or without prior PMA-treatment.
immunoprecipitation. Differences in enzymatic activity may be due to differences in the levels of these two isoforms as well as differences in affinities of the antibodies used for the immunoprecipitations or to differences in the abilities of the two isoforms to utilize GSK3 peptide in the kinase assay. The finding that Akt3 is similarly regulated as Akt1 is particularly surprising considering that Akt3 is missing a key regulatory phosphorylation site, Ser473 (12, 13). Modulation of insulin stimulated responses by activated PKC has been demonstrated at various levels of the signalling cascade such as the receptor itself and PI-3-kinase, possibly via modulation of the ser/thr phosphorylation of the insulin receptor and/or its substrates like insulin receptor substrate-1 (18 and references therein). Since activation of PKC has been observed in animals and humans in association with various insulin resistant states, it is a potential player in
FIG. 5. PMA-pretreatment inhibits the insulin-stimulated increase in enzymatic activity of an Akt1 mutant lacking its PH domain. 3T3-L1 fibroblasts expressing either an HA-epitope tagged WT-Akt1 or a DPH-Akt1 were pretreated with or without 1 mM PMA for 15 min, stimulated with the indicated concentrations of insulin, lysed and the expressed enzyme was immunoprecipitated with antiHA antibodies and the precipitates assayed for activity.
512
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC
Vol. 243, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
mains of Akt1, 2 and 3 and binding of PKCz to the PH domains of Akt1 and 2. We therefore raised the question whether the observed modulation of insulin stimulated Akt activity by PKC-activation requires the PH domain. Our studies performed on 3T3-L1 fibroblasts stably expressing either WT-Akt1 or a mutant lacking the PH domain (D4-129-Akt1) showed that the PKCmediated inhibition of the insulin-stimulation in Akt activity was independent of the PH domain. These data suggest that modulation of the insulin-stimulated increase in Akt activity by PKC activation occurs at some upstream step in Akt activation, possibly by modulating the insulin-stimulated increase in PI 3-kinase activity. In summary, the present studies demonstrate that both Akt1 and 3 are similarly stimulated by insulin in a model insulin responsive cell, the 3T3-L1 adipocytes. Moreover, the insulin-stimulated activation of both isoforms are negatively regulated by prior PKC activation, possibly explaining the negative regulation of this enzyme in models of insulin resistance. ACKNOWLEDGMENTS This work was supported in part by NIH grant DK 34926 (to RAR) and a Feodor Lynen fellowship of the Alexander von Humboldt Stiftung (to AB).
REFERENCES 1. Jones, P. F., Jakubowicz, T., Pitossi, F. J., Maurer, F., and Hemmings, B. A. (1991) Proc. Natl. Acad. Sci. USA 88, 4171–4175.
2. Coffer, P. J., and Woodgett, J. R. (1991) Eur. J. Biochem. 201, 475–481. 3. Bellacosa, A., Testa, J. R., Staal, S. P., and Tsichlis, P. N. (1991) Science 254, 274–277. 4. Bos, J. L. (1995) TIBS 20, 441–442. 5. Burgering, B. M. T., and Coffer, P. J. (1995) Nature 376, 599– 602. 6. Cross, D. A. E., Alessi, D. R., Cohen, P., Andjelkovich, M., and Hemmings, B. A. (1995) Nature 378, 785–789. 7. Hemmings, B. A. (1997) Science 275, 628–630. 8. Kohn, A. D., Summers, S. A., Birnbaum, M. J., and Roth, R. A. (1996) J. Biol. Chem. 271, 31372–31378. 9. Barthel, A., Kohn, A. D., Luo, Y., and Roth, R. A. (1997) Endocrinol. 138, 3877–3888. 10. Toker, A., and Cantley, L. C. (1997) Nature 387, 673–676. 11. Alessi, D., James, S., Downes, C., Holmes, A., Gaffney, P., Reese, C., and Cohen, P. (1997) Curr. Biol. 7, 261–269. 12. Alessi, D. R., Andjelkovic, M., Caudwell, B., Cron, P., Morrice, N., Cohen, P., and Hemmings, B. A. (1996) EMBO J. 15, 6541– 6551. 13. Konishi, H., Kuroda, S., Tanaka, M., Matsuzaki, H., Ono, Y., Kameyama, K., Haga, T., and Kikkawa, U. (1995) Biochem. Biophys. Res. Comm. 216, 526–534. 14. Considine, R. V., and Caro, J. F. (1993) J. Cell. Biochem. 52, 8– 13. 15. West, M. H. P., Wu, R. S., and Bonner, W. M. (1984) Electrophoresis 5, 133–138. 16. Green, H., and Kehinde, O. (1974) Cell 1, 113–116. 17. Kohn, A. D., Kovacina, K. S., and Roth, R. A. (1995) EMBO J. 14, 4288–4295. 18. DeFea, K., and Roth, R. A. (1997) J. Biol. Chem. 272, 31400– 31407. 19. Krook, A., Kawano, Y., Song, X. M., Efendic, S., Roth, R. A., Wallberg-Henriksson, H., and Zierath, J. R. (1997) Diabetes 46, 2110–2114. 20. Pawson, T., and Scott, J. D. (1997) Science 278, 2075–2080.
513
AID
BBRC 8134
/
6948$$$701
01-27-98 09:42:49
bbrcg
AP: BBRC