* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Characterization of serine/threonine protein phosphatases in
Survey
Document related concepts
Metalloprotein wikipedia , lookup
Western blot wikipedia , lookup
Amino acid synthesis wikipedia , lookup
Biochemical cascade wikipedia , lookup
Biochemistry wikipedia , lookup
Signal transduction wikipedia , lookup
Butyric acid wikipedia , lookup
Protein–protein interaction wikipedia , lookup
Point mutation wikipedia , lookup
Mitogen-activated protein kinase wikipedia , lookup
Paracrine signalling wikipedia , lookup
Lipid signaling wikipedia , lookup
15-Hydroxyeicosatetraenoic acid wikipedia , lookup
Two-hybrid screening wikipedia , lookup
Phosphorylation wikipedia , lookup
Transcript
Bioscience Reports, Vol. 13, No. 6, 1993 Characterization of Serine/Threonine Protein Phosphatases in RINm5F Insulinoma Cells ~ k e Sjiiholm, 1'2'3's Richard E. Honkanen, 1'4 and Per-Oiof Berggren z Received September 29, 1993; accepted October 22, 1993. This study investigates the occurrence and regulation of serine/threonine protein phosphatases (PPases) in insulin-secreting RINm5F insulinoma cells. PPases types 1 and 2A were identified in crude RINm5F cell homogenates by both enzymatic assay and Western blot analysis. We then characterized and compared the inhibitory actions of several compounds isolated from cyanobacteria, marine dinoflagellates and marine sponges, (viz. okadaic acid, microcystin-LR, calyculin-A and nodularin) cation-independent PPase activities in RINm5F cell homogenates. It was found that okadaic acid was the least potent inhibitor (IC50 ~ 10 - ' M, ICto o ~ 10-6 M), while the other compounds exhibited ICs0 values of = 5 - 1 0 - t ~ and IC~co~5 9 10 9M. The findings indicate that the inhibitory substances employed in this study may be used pharmacologically to investigate the role of serine/threonine PPases in RINm5F cell insulin secretion, a process that is likely to be regulated to a major extent by protein phosphorylation. KEY WORDS: insulin secretion; insulinoma; okadaic acid; protein phosphatase. INTRODUCTION Reversible phosphorylation of certain intracellular proteins is believed to be an important mechanism for regulating their biological activity which, in turn, controls a variety of cellular functions. For instance, significant changes in protein kinase activities and in protein phosphorylation patterns occur after the stimulation of insulin release in response to certain secretagogues [1-3]. Therefore, the molecular mechanisms regulating phosphorylation of proteins involved in the University of Hawaii at Manoa, Cancer Research Center of Hawaii, Molecular Oncology Program, 1236 Lauhala Street, Honolulu, HI 96813, U.S.A. ZDepartment of Endocrinology, The Rolf Luft Center for Diabetes Research, Karolinska Institute, Karolinska Hospital, S-171 76 Stockholm, Sweden. 3 Department of Internal Medicine, L6wenstr6mska Hospital, S-194 89 Upplands V~isby, Sweden. 4Department of Biochemistry (MSB 21981, College of Medicine, University of South Alabama, Mobile AL 36688, U.S.A. 5 To whom correspondence should be addressed (at a). 349 0144-8463/93/1200-0349507.00/0 9 1993 Plenum Publishing Corporation Sj6holm, Honkanenand Berggren 350 insulin secretory process have been extensively investigated. However, far less is known about the role and regulation of protein dephosphorylation by various protein phosphatases (PPases). Biochemically, most serine/threonine PPases can be placed in one of two major classes (PPase-1 and PPase-2) based on their sensitivity to two endogenous heat and acid stable inhibitory proteins (I-1 and I-2) and their ability to dephosphorylate either the c~ or fi subunit of phosphorylase kinase [for rev; see 4-7]. The catalytic subunit of PPase-1 preferentially dephosphorylates the fi-subunit of phosphorylase kinase and is inhibited by nanomolar concentrations of I-1 and I-2. PPase-2 preferentially dephosphorylates the o:-subunit of phosphorylase kinase and is not sensitive to nanomolar concentrations of I-1 or I-2. PPase-2 can be further divided into three subtypes based on their requirements for divalent cations; PPase-2A, like PPase-1, does not require divalent cations, while PPase-2B (calcineurin) and PPase-2C have an absolute requirement for CaZ+/calmodulin and Mg 2+, respectively. Other PPases, (e.g., PPase-X, PPase-Y, PPase-Z [8] and rdgC [9]) have been identified by cloning studies, and a minor PPase (PPase-3) has recently been purified from bovine brain [10]; however, the abundance and tissue distribution of these PPases have yet to be determined [5, 8-10]. This paper investigates the occurrence and regulation of serine/threonine PPases in insulin-secreting RINm5F insulinoma cells. Parts of this study has been published in abstract form [11, 12]. MATERIALS AND METHODS Materials Okadaic acid was obtained from Moanoa Bioproducts (Honolulu, HI). Calyculin-A was from LC Services Corp., Woburn, MA, while nodularin was obtained from Calbiochem. Microcystin-LR was isolated from Microcystis aeruginosa (strain 7820) as described [13] and was identified from its fast atom bombardment spectrum (m/z 995 [MH+]) and [13C]NMR spectrum in MeOH-d4. Amounts of microcystin-LR were calculated by UV analysis [14]. Cyclic adenosine-3',5'-monophosphate (cAMP), protein kinase A (cAMP-dependent), crude histone (type 2-AS) were from Sigma. Ammonium molybdate was obtained from Mallinckrodt, while [y-32p]ATP (350Ci/mol) was from New England Nuclear. Immunoblotting Purified PPase-1 and PPase-2A from rabbit muscle and whole RINm5F cells homogenates were utilized for immunoblotting using type-specific rat polyclonal antibodies generously provided by Dr. D. Alexander [15]. Aliquots of RINm5F cell homogenates and PPase-1 or PPase-2A, as controls, were subjected to SDS-PAGE on 10% polyacrylamide gels. Gels were then electrophoretically transferred to Immobilon-P and immunoreactivity of the primary antibody was visualized by fluorography with an enhanced chemiluminescence system (ECL, Amersham) essentially according to the methods of the manufacturer. Protein Phosphatesin RINm5FCells 351 Preparation of Phosphohistone and Determination of Phosphatase Activity Histone type 2-AS was phosphorylated with rabbit muscle type I protein kinase A (cAMP-dependent) as described previously [16]. Incubation mixtures (4 ml) consisted of 20 mg histone, 1 mg protein kinase A , 20 mM Tris-HCl (pH 7.5), 5mM MgC12, 0.5mM dithiothreitol (DTT), 0.7 gM cAMP and 150gM [732p]ATP. The reaction mixture was incubated at 30~ for 3 h and terminated by addition of trichloroacetic acid (20% final concentration). The phosphorylated histone was recovered according to Meisler and Langan [17]. RINm5F cells [18] were cultured to subconfluence for 2 days in 60 mm plastic Falcon culture dishes in medium BME supplemented with 10% fetal bovine serum, 100 U/ml benzylpenicillin and 0.1 mg/ml streptomycin. The medium was quickly removed, cells washed three times in ice-cold PBS, scraped off plates in 1 ml Tris buffer (20 mM Tris-HCl, 1 mM EDTA and 2 mM DTT, pH 7.4) and disrupted in a Polytron homogenizer. After pelleting debris, homogenates were swiftly transferred to Eppendorf tubes which were immediately plunged into liquid nitrogen and stored at -80~ pending analysis. Phosphatase activity against phosphohistone was determined by measuring the liberation of [32p]. Assays (80/~1 total volume) containing 50mM Tris-HCl (pH7.4), 0.5mM DTT, l mM EDTA and [32p]histone (2/~M PO4) plus the appropriate test substance (or equal amounts of its solvent, dimethyl formamide) were conducted at 30~ for 10 min [10, 14, 16]. Reactions were stopped by addition of 100/~1 1 M H2SO4 containing 1 mM potassium phosphate. Substrate dephosphorylation was kept to less than 10% of total phosphorylated substrate, and preliminary experiments were performed to determine the titration end point [14, 16] ensuring that the reactions were linear with respect to time and enzyme concentration. [32p] released was extracted as a phosphomolybdate complex and measured according to Kiililea et al. [19]. RESULTS AND DISCUSSION In the last decade, a great deal of interest has been focused on how reversible phosphorylation of proteins is involved in regulation of cellular functions [4-7, 20]. Reversible changes in levels of phosphoserine and phosphothreonine at specific residues are a means by which the activity of many key proteins is regulated and a way through which cells convey extracellular signals into such diverse biological responses as mitogenesis, ion channel activity, substrate uptake and metabolism, and hormone secretion [4-7, 21-27]. This phosphorylation/dephosphorylation cycle is now known to be a dynamic process with cellular levels of protein phosphorylation being determined by the combined actions of numerous protein kinases and PPases. However, compared with protein kinases, serine/threonine PPases have received far less attention, partly because the lack of specific inhibitors, and little is known about their role in the insulin-producing pancreatic/3-cell. Studies designed to determine the biological functions of PPases have recently been aided by the discovery of several potent and specific PPase 352 Sj6holm, Honkanen and Berggren OH O H3CO CH3/ o ~ o o Ho , "' r~ 6H ~CH~ H , ' ..... " ,"" ,. 0 _ ! \ OH Okadaic acid H CO~H CH3 HN FI ~ S O N O O NH O H CH / 3 H ~ OO ~n 3 CO2H O NH HN~NH 2 Microcystin-LR Fig. 1. C%H \ II - " " a f O ()CH 3 Calyculin A H3C. H OCH3 OH NH HN~ NH2 Nodularin Structure of the PPase inhibitors okadaic acid, microcystin-LR, calyculin-A and nodularin. inhibitors (Fig. 1). Okadaic acid, a complex polyether fatty acid produced by certain strains of marine dinoflagellates, calyculin-A, a novel spiro-ketal isolated from a marine sponge, and two cyclic peptides produced by cyanobacteria, microcystin-LR and nodularin, have all been shown to specifically inhibit different types of PPases, and in some cases, exert tumorigenic activity [6, 7, 14, 16, 28, 29]. The inhibitory potency of the most extensively investigated compound, okadaic acid, has been carefully characterized on type 1, 2A, 2B, 2C and 3 PPases [6, 10, 16, 30]. These studies revealed that, in a dilute whole cell homogenate, 1 nM of okadaic acid inhibits essentially all PPase-2A activity, while having no appreciable effect on PPase-1, -2B or -2C [6, 10, 16, 31]. One nanomolar okadaic acid also partially inhibit the activity of PPase-3; however, compared to PPase-1 and -2A, PPase-3 represents a minor amount (<3%) of the total okadaic acid sensitive and divalent cation independent PPase activity in a whole cell homogenate, and PPase-3 is generally not in an active state prior to solubilization with cholate [10]. Furthermore, no appreciable amount of PPase-3 activity could be detected in RINm5F cell homogenates or upon Western analysis with polyclonal antisera produced against PPase-3 [A.S., R.E.H. & P.-O.B., (unpublished observations). Okadaic acid, at any concentraion, is without apparent effect on PPase-2C, tyrosine phosphatases, inositol trisphosphatase, mitochondrial pyruvate dehydrogenase, or acid/alkaline phosphatases, as well as numerous kinases (e.g. cAMP-dependent, CaZ+/calmodulin-dependent, and Ca2+/phospholipid-dependent protein kinases [5, 6, 14, 16]. At concentrations >5/~M okadaic acid will partially inhibit the activity of PPase-2B [16, 30]. Protein Phosphates in RINm5F Cells 353 Fig. 2. Immunochemicalanalysis of RINm5F cell homogenates identifying the presence of PPase-1 and PPase-2A. Purified PPase-1 (lane 1) RINm5F cell homogenate (lane 2 and 4), and purified PPase-2A (lane 3) were analyzed by immunolabeling using polyclonal antibodies specific for PPase-1 (1-2) or PPase-2A (3-4) as detailed in the Materials and Methods section. In Fig. 2 it is demonstrated by Western analyses that RINm5F cells contain both PPase-1 and PPase-2A protein. The CaZ+/calmodulin-dependent PPase-2B (calcineurin) has also recently been localized by Western blotting in rat pancreatic islets [32]; however, its functional role in the/3-cell has not yet been studied. In addition, since the PPase assays in this paper were performed in the presence of the cation chelator E G T A , the CaZ+/calmodulin-dependent PPase-2B and the MgZ+-requiring PPase-2C do no contribute to the results obtained. As seen in Fig. 3, the divalent cation independent PPase activity in the dilute RINm5F cell homogenate is completely inhibited by okadaic acid at a concentration of 1 ~M, with partial inhibition occurring at a concentration of less than 0.1 nM. This indicates that both PPase-1 and PPase-2A are active in RINm5F cells (i.e. if only PPase-1 was active then in appropriately diluted samples inhibition should not occur at concentrations below 1 nM [16, 31]). Similarly, if only PPase-2A was active then complete inhibition would be observed at a concentration of 1-2 nM [6, 14, 16, 31]). Of course, other as of yet biochemically uncharacterized PPases (i.e. PPase-Y, PPase-Z etc. [8]) could also be sensitive to okadaic acid and, thus, comprise a portion of the activity ascribed here to PPase-1 or PPase-2A. Comparison of the inhibitory activity of okadaic acid with that of other PPase inhibitors demonstrates that, in dilute RINm5F cell homogenates, okadaic acid is the least potent inhibitor, exhibiting IC50 at --~10-9M and ICloo at ~ 1 0 - 6 M . As mentioned above, these values agree well with previously published results employing a combination of pure PPases types 1 and 2A and other cell homogenates containing PPase-1 and -2A [8, 10, 16, 31]. The inflection point in PPase activity noted at 10-9M okadaic acid (Fig. 3) likely reflects the concentra- 354 Sj6holm, Honkanen and Berggren i f i I I i t i i 100 80 4,g S 60 0 0 #% 0 m N 40 20 0 I 1E-131E-12 1E-11 1 E - I O I I 1E-9 1E-8 [Okadaic i i I acid] I (r 1E-7 1E-6 1E-5 1E-4 (M) i i 1E-8 1E-7 100 80 4.J - - 9~ o ~E U 60 0 0 9 4O 20 0 1E-15 1E-121E-11 I 1E-10 1E-9 [Microcystin-LR] 1E-6 I 1E-5 1E-4- (M) Fig. 3. Different inhibitory profiles of okadaic acid,, microcystin-LR, calyculin-A and nodularin on RINm5F PPase activities. Cell homogenates were incubated for 10 min at 30~ with the indicated agent using [32p]histone as a substrate. PPase activity was assayed as detailed in the Materials and Methods section. Values are mean percent of controls (incubated with solvent only) +S.E.M, for 4 separate experiments. Protein Phosphates in RINm5F Cells 355 100 80 4" D 9~ o ~ ~~ E~ 60 O O (D a_N 40 n ..j 20 0 1E-131E-121E-111E-10 I 1E-9 1E-8 [Calyculin-A] i i i i 1E-7 I 1E-6 1E-5 1E-4 (M) i i i i 100 80 "~ O .--- E 60 if) O a_ N 40 20 1E-131E-121E-111E-101E-9 1E-8 [Nodularin] Fig. 3. (Continued) 1E-71E-6 (M) 1E-51E-4 356 Sj6holm, Honkanen and Berggren tion where all PPase-2A activity is inhibited and type 1 activity remains unaffected [14, 16]. Again, this is in agreement with studies with purified enzymes [8, 10, 14, 16], and suggests that PPase-1 and PPase-2A represent the quantitatively most important cation-independent serine/threonine PPases in RINm5F cells. In comparison with okadaic acid, the other compounds tested, i.e., microcystin-LR, calyculin-A and nodularin, were more potent inhibitors of RINm5F cell PPases, exhibiting IC50 values of ~ 5 - 1 0 - 1 ~ and IC1oo of ~ 5 . 1 0 - 9 M . These findings indicate that such inhibitors can be used pharmacologically to evaluate a possible role of different PPases in the regulation of the stimulus-secretion coupling in insulin-producing cells, which is believed to be controlled to a great extent by protein phosphorylation. More specifically, under the conditions utilized in this study, microcystin-LR, calyculin-A and nodularin all completely inhibit PPases 1 and 2A activity at a concentration of ~ 1 - 5 nM, essentially the same concentration at which okadaic acid inhibits all PPase-2A activity without affecting the activity of PPase-1. In insulin-secreting cells, previous studies have established that stimulation of protein phosphorylation by direct activation of Ca2+/phospholipid-dependent protein kinase (protein kinase C) with phorbol esters or pharmacological elevation of cAMP levels, activating protein kinase A, stimulates insulin secretion [1-3, 33-38]. Further, a number of physiological stimuli of insulin secretion increase fl-cell phosphorylation state [1-3, 33, 34]. In a preliminary report [27], acute exposure of RINm5F cells to the cell-permeant okadaic acid was found to enhance Ca 2+ entry and to stimulate insulin secretion. It should be possible to design experiments distinguishing the functional consequences of inhibiting either PPase-1, PPase-2A or both. For example, if similar results are obtained with ~ l . 0 n M okadaic acid and ~ l . 0 n M calyculin-A, then the observed change produced by these inhibitors is likely to be due to the loss of PPase-2A activity. Alternatively, if a change is observed in the presence of 1.0 nM calyculin-A while 1.0 nM okadaic acid is without effect and 1.0/~M okadaic acid produces an effect similar to that observed with calyculin-A, then PPase-1 is likely to be the PPase involved. However, because of differences in cellular uptake kinetics between the drugs, the utility of such experimental approaches may be confined to studies in permeabilized or homogenized cells to which the different inhibitors or purified PPases may be directly introduced. An equally interesting approach may be to microinject the inhibitors or purified PPases or protein kinases into cells or to use them in patch-clamp studies to determine whether these enzymes alter ion channel activity and thereby the insulin secretory response. Thus in conclusion, we have demonstrated that insulin-secreting cells contain, in addition to previously reported PPase-2B/calcineurin [32], PPases types 1 and 2A. Moreover, the presently reported characterization of doseresponse relationships between the four PPase inhibitors, okadaic acid, microcystin-LR, calyculin-A and nodularin, indicate that such compounds may be useful in probing the roles of different PPases in the regulation of the stimulus-secretion coupling in insulin-producing cells. Protein Phosphates in RINm5F Cells 357 ACKNOWLEDGEMENTS We thank Dr. D. Alexander for generously supplying us with type specific antiserum to PPase-1 and to PPase-2A. Financial support was received from the Swedish Medical Research Council (03X-09890, 19X-00034), the Cancer Foundation, the Sandoz Corporation, the R6nnow Fund, the Pharmacia Fund for Biotechnological Research, the Wallenberg Fund, the Anna Cederberg Fund, the Wenner-Gren Foundation, Funds of the Karolinska Institute, the Bank of Sweden Tercentenary Foundation, the Swedish Diabetes Association, the Swedish Society of Medicine, the Nordic Insulin Fund, Fredrik and Inger Thuring's Foundation and Syskonen Svensson's Fund. REFERENCES 1. 2. 3. 4. 5, 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. Hedeskov, C. J. (1980) Physiol. Rev. 60:442-508. Efendic', S., Kindmark, H. and Berggren, P.-O. (1991) J. Int. Med. 229 (Suppl. 2):9-22. Turk, J., Wolf, B, A. and McDaniel, M. L. (1987) Prog. Lipid Res. 26:125-181. Cohen, P. (1989) Annu. Rev. Biochem. 58:453-508. Shenolikar, S. and Nairn, A. C. (1991)Adv. Sec. Mess, Phosphoprot. Res. 23:1-121. Cohen, P., Holmes, C. F. B. and Tsukitani, Y. (1990) Trends Biochem. Sci. 15:98-102. Bollen, M. and Stalmans, W. (1992) Crit. Rev. Biochem. Mol. Biol. 27:227-281. Cohen, P. T. W., Brewis, N. D., Hughes, V. and Mann, D. J. (1990) FEBS Lett. 268:355-358. Steele, F. R., Washburn, T., Rieger, R. and O'Tousa, J. E. (1992) Cell 69:669-676. Honkanen, R. E., Zwiller, J., Daily, S., Khatra, B. S., Dukelow, M. and Boynton, A. L. (1991) J. Biol. Chem. 266:6614-6619. Sj6holm, A., Honkanen, R. E. and Berggren, P.-O. (1992) Diabetologia 35 (Suppl. 1):A100 (abstract).~ Sj6holm, A., Honkanen, R. and Berggren, P.-O. (1993) J. Cell. Biochem. Suppl. 17A:315 (abstract). Krishnamurthy, T., Carmichael, W. W. and Sarver, E. W. (1986) Toxicon 24:865-873. Honkanen, R. E., Zwiller, J., Moore, R. E., Daily, S. L., Khatra, B. S., Dukelow, M. and Boynton, A. L. (1990)J. Biol. Chem. 265:19,401-19,404. Alexander, D. R., Hexham, J. M. and Crumpton, M. J. (1988) Biochem. J. 256:885-892. Honkanen, R. E., Dukelow, M., Moore, R. E., Zwiller, J., Khatra, B. S. and Boynton, A. L. (1991) Mol. Pharmacol. 40:577-583. Meisler, M. H. and Langan, T. A. (1969) J. Biol. Chem. 244:4961-4968. Sj6holm, ,~., Welsh, N., Bankston, P. W., Hoftiezer, V. and Hellerstr6m, C. (1991) Biochem. J. 277: 533-540. Killilea, S. D., Aylward, J. H., Mellgren, R. L. and Lee, E. Y. C. (1978) Arch. Biochem. Biophys. 191: 638-646. Nishizuka, Y. (1988) Nature 334:661-665. Chiavaroli, C., Vacher, P. and Schlegel, W. (1992) Eur. J. Pharmacol. 227: 173-180. Wagner, P. D. and Vu, N.-D. (1990) J. Biol. Chem. 265"10,352-10,357. Reinhart, P. H., Chung, S., Martin, B. L., Brautigan, D. L. and Levitan, I. B. (1991) J. Neurosci. 11 91627-1635. Wu, Y. N. and Wagner, P. D. (1991) Biochim. Biophys. Acta 1092:384-390. White, R. E., Schonbrunn, A. and Armstrong, D. L. (1991) Nature 351:570-573. ZwiUer, J., Honkanen, R. and Boynton, A. L. (1990) Exp. Cell Res. 187:193-202. Larsson, O., Haby, C., Islam, M. S., Zwiller, J. and Berggren, P.-O. (1993) J. Cell. Biochem. Suppl. 17A: 287 (abstract). Meriluoto, J. A., Eriksson, J. E., Havada, K., Dahlem, A. M., Sivonen, K. and Carmichael, W. W. (1990)J. Chromatogr. 509:390-395. Ishihara, H., Martin, B. L., Brautigan, D. L., Karaki, H., Ozaki, H., Kato, Y., Fusetani, N., Watabe, S., Hashimoto, K. and Uemura, D. (1989) Biochem. Biophys. Res. Commun. 159: 871-877. 358 Sj6holm, Honkanen and Berggren 30. Bialojan, C. and Takai, A. (1988) Biochem. J. 250:283-290. 31. Cohen, P. (1991) Meth. Enzymol. 201:389-397. 32. Gagliardino, J. J., Krinks, M. H. and Gagliardino, E. E. (1991) Biochim. Biophys. Acta 1091: 370-373. 33. Wollheim, C. B. and Biden, T. J. (1986) Ann. N Y A c a d . Sci. 488:317-333. 34. Amm/il~i, C., Ashcroft, F. M. and Rorsman, P. (1993) Nature 363:356-358. 35. Sj6holm, A. (1991) FEBS Lett. 294:257-260. 36. Sj6holm, oA. (1992) Am. J. Physiol. 263:Cl14-C120. 37. Sjfholm, A. (1992) FEBS Len. 311:85-90. 38. Sj6holm, fi,. (1993) Am. J. Physiol. 264:C501-C518.