The Role of Calcium/Calmodulin-Dependent Protein Kinase Cascade in Glucose Upregulation of Insulin Gene Expression Xiao Yu,1 Koji Murao,1 Yoshitaka Sayo,1 Hitomi Imachi,1 Wen M. Cao,1 Shouji Ohtsuka,1 Michio Niimi,1 Hiroshi Tokumitsu,2 Hiroyuki Inuzuka,2 Norman C.W. Wong,3,4 Ryoji Kobayashi,2 and Toshihiko Ishida1
A number of factors have been reported to affect insulin synthesis in -cells. Although glucose is the most important regulator of insulin gene expression in pancreatic -cells, the mechanisms whereby glucose stimulates insulin gene transcription in response to changes in glucose concentration have not been clarified yet. In this study, we examined the role of the Ca2ⴙ/calmodulin (CaM)-dependent protein kinase (CaM-K) cascade in transcriptional activation of insulin. RT-PCR, Western blotting, and immunohistochemical staining analysis revealed that CaM-K kinase-␣ (CaM-KK␣) and CaM-KIV were localized in rat pancreatic -cells and their cell line, INS-1. Exposure of INS-1 cells to 11.2 mmol/l glucose elicited an increase of insulin promoter activity as well as upregulation of CaM-KIV activity within 2 min after stimulation. We investigated the influence on insulin promoter activity of the constitutively active form (CaM-KIVc) or dominant-negative mutant (CaMKIVdn) of CaM-KIV in transfected INS-1 cells. CaMKIVc alone was sufficient, and the upstream kinase, CaM-KK, was enhanced to upregulate the insulin promoter activity in INS-1 cells. Furthermore, cotransfection of CaM-KIVdn suppressed to a significant degree the glucose-upregulated activity of the insulin promoter. Taken together, these results indicated that the CaM-KK/CaM-KIV cascade might play an important role in glucose-upregulated transcriptional activation of the insulin gene. Diabetes 53:1475–1481, 2004
From the 1First Department of Internal Medicine, Kagawa Medical University, Kagawa, Japan; the 2Department of Signal Transduction Sciences, Kagawa Medical University, Kagawa, Japan; the 3Department of Medicine, Faculty of Medicine, University of Calgary, Health Sciences Center, Calgary, Alberta, Canada; and the 4Department of Biochemistry & Molecular Biology, Faculty of Medicine, University of Calgary, Health Sciences Center, Calgary, Alberta, Canada. Address correspondence and reprint requests to Koji Murao, MD, PhD, First Department of Internal Medicine, Kagawa Medical University, 1750-1, Mikicho, Kita-gun, Kagawa, 761-0793, Japan. E-mail:
[email protected]. Received for publication 31 May 2003 and accepted in revised form 15 March 2004. ATF, activating transcription factor; CaM, calmodulin; CaM-K, CaM-dependent protein kinase; CaM-KIVc, constitutively active CaM-KIV; CaM-KIVdn, dominant-negative mutant of CaM-KIV; CaM-KK, CaM-K kinase; CaM-KKc, constitutively active CaM-KK; CREB, cAMP-responsive element binding protein; ELISA, enzyme-linked immunosorbent assay; GST, glutathione S-transferase. © 2004 by the American Diabetes Association. DIABETES, VOL. 53, JUNE 2004
T
he maintenance of blood glucose concentrations within narrow limits is of critical physiologic importance in mammals. Elevated glucose concentrations stimulate the transcription of the insulin and several other genes that regulate glucose homeostasis (1,2). However, the molecular mechanisms by which glucose regulates the transcription of these genes, especially that of insulin, are only partially understood (3). Although glucose stimulation of insulin biosynthesis at the posttranscriptional and translational level has been studied in detail, the glucose stimulation of insulin gene transcription has not. The exclusive pancreatic islet -cell expression of the insulin gene is controlled by cis-acting elements that reside within the ⫺340 to ⫺91 bp of the gene, which interact with selected nuclear factors (4). Studies (5,6) show that sites A3 (⫺201 to ⫺196), C1 (⫺115 to ⫺107), and E1 (⫺100 to ⫺91) are essential for efficient expression of the insulin gene in the -cells. Under physiologic conditions, insulin gene transcription is controlled by a glucose concentration through a mechanism that regulates the activity of both C1 and A3 (CT2, Flat-E) cis-acting elements (4). Intracellular Ca2⫹ plays an important role in -cell function. This ion is the key intracellular mediator of glucose-stimulated insulin secretion (7,8) and directly affects the biosynthesis of insulin (9). In addition, Ca2⫹ also enhances transcription of the human insulin gene. This effect of calcium is completely blocked by exposure to a Ca2⫹/calmodulin (CaM)-dependent protein kinase (CaM-K) inhibitor (10) and thus suggests that the actions of calcium on the insulin gene are mediated in part by Ca2⫹/CaM-Ks. These kinases belong to a diverse group of enzymes that participate in many cellular responses and are activated by increasing concentrations of intracellular Ca2⫹. There are two multifunctional CaM-Ks, named CaM-KI and -IV, which are activated by an upstream CaM-K kinase (CaMKK) through phosphorylation of a threonine residue within the active loop of the protein. This phosphorylation strongly upregulates the catalytic activity of both enzymes (11). Since the CaM-K inhibitor blocked calcium stimulation of insulin gene transcription (10), this finding suggests the participation of CaM-Ks in glucose control of the gene. In addition, previous studies showed that activity of the transcription factor ATF (activating transcription factor)-2, 1475
CaM-K CASCADE ON INSULIN GENE EXPRESSION
which binds to the human insulin gene for cAMP-responsive elements is enhanced by Ca2⫹/CaM-KIV (12). These findings pointed to the potential role of CaM-Ks and prompted us to examine whether the CaM-KK/CaM-KIV cascade existed in pancreatic -cells. The results of our studies suggest a mechanism requiring the CaM-KK/CaM-KIV cascade in glucose control of insulin gene transcription. RESEARCH DESIGN AND METHODS Recombinant rat CaM was expressed in E. coli BL21 (DE3) using the vector porcine endothelin–CaM (a gift from Dr. Nobuhiro Hayashi, Fujita Health University) and purified by elution from a phenyl-Sepharose column (13). Recombinant rat CaM-KK␣ was expressed in E. coli BL21 (DE3) by using porcine endothelin–16 vector and then purified by CaM-Sepharose column chromatography. Recombinant glutathione S-transferase (GST)-CaM-KIs (wild type and Thr177Ala mutant) were expressed in E. coli JM-109 and purified by glutathione-Sepharose column chromatography as previously described (14). Recombinant mouse CaM-KIV was expressed using the bacculovirus system, kindly provided by Dr. Thomas R. Soderling (Vollum Institute, Oregon Health Sciences University, Portland, OR). Cell culture. The INS-1 cells originated from a rat insulinoma cell line that was developed and propagated at the Division of Biochimie Cliniqe (courtesy of Dr. C.B. Wollheim, Geneva, Switzerland). The present experiments were performed using cell passages 61–72. These cells were cultured in RPMI 1640 media (Gibco, Tokyo, Japan) containing 11.2 mmol/l glucose (unless otherwise stated) and supplemented with 10% heat-inactivated fetal bovine serum (Dainippon Pharmaceutical, Tokyo, Japan), 50 mol/l 2-mercaptoethanol, 100 units/ml penicillin, and 0.1 mg/ml streptomycin. Islets of Langerhans were isolated from pancreata of 250-g male Sprague-Dawley rats as previously described (15). PCR. Total RNA was extracted from the INS-1 and the pancreatic islet cells from Sprague-Dawley rats according to methods described previously (16). The expression of CaM-KIV and CaM-KK was examined using RT-PCR as described previously (16). A primer pair matching the published sequence (17,18) of CaM-KIV and CaM-KK (sense: 5⬘-GCCCTATGCTCTCAAAGTGT-3⬘ and 5⬘-GGAGGTGAAGAACTCAGTC-3⬘; antisense: 5⬘-CACACCGTCTTCATG AGCAC-3⬘and 5⬘-GGATGCAGCCTCATCTTCCT-3⬘, respectively) was used in separate RT-PCRs. Thirty cycles of PCR for CaM-KIV and CaM-KK were carried out using a thermal cycler (Sanko Junyaku, Tokyo, Japan) according to a step program of 60 s at 94°C, 60 s at 51°C, and 60 s at 72°C, followed by a 15-min extension at 72°C. The identity of the PCR-amplified DNA products was confirmed by sequence analysis. Immunohistochemical localization. Tissue specimens of interest were fixed in formalin and embedded in paraffin. Sections were incubated overnight with a goat polyclonal antibody directed against CaM-KIV (sc-1546; Santa Cruz Biotechnology, Santa Cruz, CA) or a goat polyclonal antibody directed against CaM-KK (sc-1548; Santa Cruz Biotechnology) in 5% normal rabbit serum in PBS. Sections were rinsed in PBS, incubated for 60 min with a biotinylated antibody (Vectastain Elite kit; Vector Laboratories, Burlingame, CA) in 1% normal rabbit serum in PBS, rinsed in PBS, and incubated with an avidinbiotinylated peroxidase complex (Vectastain Elite kit) in PBS, as recommended by the manufacturer. Antibody binding was visualized with the diaminobenzidine reaction, and sections were counterstained with Mayer’s hemotoxylin. Double immunofluorescence for insulin and CaM-KK or CaMKIV was visualized using an indirect immunofluorescence technique. To perform double staining, slides were incubated with a rabbit polyclonal insulin antibody (1:100) mixed with one of the following antisera: goat polyclonal antibody to CaM-KIV (1:100) and goat polyclonal antibody to CaM-KK (1:100). After rinsing in PBS, the sections were incubated with a mixture of rhodamine-conjugated donkey anti-goat immunoglobulins (1:100) and fluorescein-conjugated donkey anti-rabbit antibody (1:100). The processed slides were examined using fluorescence microscopy. Activation and phosphorylation of CaM-KI by INS-1 cell extract. GST-CaM-KI (1.0 g) was incubated with either INS-1 cell extract (2.4 g) or buffer at 30°C for 20 min in 10 l of 50 mmol/l HEPES (pH 7.5), 10 mmol/l Mg(Ac)2, 1 mmol/l dithiothreitol, and 400 mol/l ATP containing either 2 mmol/l CaCl2 or 8 mol/l CaM. The reaction was terminated by diluting 20-fold at 4°C with 50 mmol/l HEPES (pH 7.5), 2 mg/ml BSA, 10% ethylene glycol, and 2 mmol/l EDTA. CaM-KI activity was measured at 30°C for 10 min in 25 l of 50 mmol/l HEPES (pH 7.5), 10 mmol/l Mg(Ac)2, 1 mmol/l dithiothreitol, 400 mol/l [␥-32P]ATP (1,000 –2,000 cpm/pmol), and 40 mol/l syntide-2 containing 2 mmol/l CaCl2/8 mol/l CaM. The reaction was initiated by the addition of 5 l CaM-KI and terminated by spotting 15-l aliquots onto phosphocellulose paper (Whatman P81; Whatman, Kent, U.K.) followed by washing in 75 mmol/l 1476
phosphoric acid. Phosphorylation of GST-CaM-KI by INS-1 cell extract was essentially the same as that of the activation assay except for the use of 400 mol/l [␥-32P]ATP (1,000 –2,000 cpm/pmol). After a 20-min incubation at 30°C, the reaction was terminated by adding 5 l of SDS-PAGE sample buffer. Then the samples were subjected to SDS-10%PAGE, followed by autoradiography. CaM-KIV activity in INS-1 cells. The day before testing, the INS-1 cells were incubated overnight with regular medium containing 2.8 mmol/l glucose. INS-1 cells were treated with 11.2 mmol/l glucose for 1 min before harvest at indicated time periods, and then the cells were lysed and endogenous CaM-KIV was immunoprecipitated with anti–CaM-KIV antibody (an affinitypurified goat polyclonal antibody, sc-1546; Santa Cruz Biotechnology). CaMKIV activity in the immunocomplex was measured at 30°C for 10 min in 30 l of 50 mmol/l HEPES (pH 7.5), 10 mmol/l Mg(Ac)2, 1 mmol/l dithiothreitol, 400 mol/l [␥-32P]ATP (1,000 –2,000 cpm/pmol), and 40 mol/l syntide-2 containing 2 mmol/l CaCl2/8 mol/l CaM. Transfection of INS-1 cells and luciferase reporter gene assay. The reporter gene used in our studies was kindly provided by Dr. Roland Stein (Vanderbilt University Medical Center, Nashville, TN). The wild type (⫺238 WT-LUC) contains the rat insulin-2 gene sequences spanning the region from ⫺238 to 2 bp that are linked to the luciferase reporter gene. Purified reporter plasmid was transfected into INS-1 (at 60% confluence) using a conventional cationic liposome transfection method (lipofectamine; Life Technologies, Gaithersburg, MD). All assays were corrected for -galactosidase activity, and total amounts of protein per reaction were identical (16). Both cDNA of Ca2⫹/CaM-independent mutant of CaM-KIV (CaM-KIVc, 305 HMDT to DEDD), CaM-KIV kinase-negative mutant (CaM-KIVdn, 305 HMDT to DEDD, K71E), and CaM-KK mutant (CaM-KKc, residues 1– 434) were constructed as described previously (19,20). Transfected cells were harvested, and -galactosidase activity was measured in an aliquot of the cytoplasmic fraction (16). Twenty-microliter aliquots were taken for the luciferase assay and performed according to the manufacturer’s instructions (ToyoInk, Tokyo, Japan). Expression of CaM-KIV mutants in INS-1 cells using recombinant adenovirus infection. Recombinant adenoviruses carrying cDNAs of CaMKIV were produced using the Adeno-X Expression System (Clontech, Palo Alto, CA) as previously described (21,22). For virus infection, confluent INS-1 cells in six-well culture plates were infected with viruses at a multiplicity of infection of 20 plaque-forming units/cell at 37°C for 1 h. After infection, viruses were aspirated and cells were further cultured in RPMI medium containing 10% fetal bovine serum. The insulin gene expression and insulin secretion were determined by Northern blot analysis and enzyme immunoassay. Northern blot analysis. A single-step acid guanidinium thiocyanate-phenolchloroform extraction technique was used to isolate total RNA from INS-1 cells transfected with adenovirus carrying cDNA of CaM-KIVc or CaM-KKc ⫹ CaM-KIVc as described previously (22). Separation of the RNA samples, transfer to membrane, and hybridization with the insulin probe was described previously (23). To detect rat insulin mRNA, we used a 355-bp probe derived from INS-1 cDNA obtained using PCR amplification using insulin-specific oligonucleotide primers. The rat insulin sense and antisense primers were 5⬘-ATAGACCATCAGCAAGCAGG-3⬘ and 5⬘-CTCCAGTTGTGGCACTTGCG-3⬘, respectively (24). Probes used in the hybridization were labeled with digoxigenin (Boehringer Mannheim, Indianapolis, IN) using nick translation and then purified as previously described (24). Blots were also probed with rat -actin to assess equal loading of samples (24). After posthybridization washes, the membranes were incubated with anti-Dig antibody (Boehringer Mannheim), followed by treatment with AMPPD, and then exposed to X-ray film for 15– 45 min. Insulin enzyme-linked immunosorbent assay. The levels of immunoreactive insulin were quantified using a commercially available sandwich-type enzyme-linked immunosorbent assay (ELISA) (Shibayagi, Tokyo, Japan). This ELISA is sensitive to 1 ng/ml insulin, and it has an intra-assay coefficient of variation of ⬍0.5% and an interassay coefficient of variation of ⬍10%. Others. Western blotting was carried out using either anti–CaM-KK antibody (Transduction Laboratories, Lexington, KY) or anti–CaM-KIV antibody (Transduction Laboratories). Detection was performed by using chemiluminescence reagent (NEN Life Science Products, Boston, MA). Statistical analysis. Statistical comparisons were made by one-way ANOVA and Student’s t test, with P ⬍ 0.05 considered significant.
RESULTS
Expression of CaM-KK/CaM-KIV in INS-1 cells and rat pancreatic islets. Whether the CaM-K cascade is present in rat pancreatic islets and rat insulinoma cell line, INS-1, is not known. Therefore, we examined the mRNA expression of CaM-KK␣ and CaM-KIV using RT-PCR. ReDIABETES, VOL. 53, JUNE 2004
X. YU AND ASSOCIATES
FIG. 2. Identification of the CaM-KK/CaM-KIV cascade in INS-1 cells. A: Total cell protein extracted from the INS-1 cells and recombinant rat CaM-KK or CaM-KIV were blotted with anti–CaM-KK or anti–CaM-KIV antibody, respectively. B: Activation of CaM-KI by CaM-KK. Either wild type or mutants (positions 1–295, T179A, and T199D) of GSTCaM-KI (0.5 g), which was expressed and purified from E. coli, was incubated with either buffer (ⴚ) or 9 ng of recombinant CaM-KK (ⴙ) of 2 mmol/l CaCl2 and 8 mol/l CaM. The protein kinase activity of each CaM-KI toward syntide-2 was measured as described under RESEARCH DESIGN AND METHODS. The results are presented as the mean ⴞ SE of three experiments. The 32P incorporated into each recombinant CaM-KI during the activation reaction under the same condition as described above was analyzed by SDS-PAGE followed by autoradiography (inset).
FIG. 1. Expression of CaM-KK/CaM-KIV in the pancreas. A: RT-PCR of CaM-KK/CaM-KIV. -Actin was amplified as an internal control. Lane M, markers; lane 1, INS-1 cells; lane 2, rat pancreatic islets. B: Immunohistochemical localization of CaM-KK/CaM-KIV in pancreatic islets was shown using anti–CaM-KK or anti–CaM-KIV antibodies or nonimmune IgG as a negative control. a: Control IgG for anti–CaM-KK. b: Anti–CaM-KK. c: Control IgG for anti–CaM-KIV. d: Anti–CaM-KIV. C: Double immunostaining. CaM-KK staining (a) in rat pancreas (the field shows a pancreatic islet), and insulin staining (b) of the same islet. CaM-KIV staining (c) in rat pancreas (the field shows a pancreatic islet), and insulin staining (d) of the same islet.
sults (Fig. 1) showed that CaM-KIV and upstream protein kinase, CaM-KK␣, were detected not only in rat pancreatic islets but also in INS-1 cells. In addition, both mRNAs appeared to be more abundantly expressed in the INS-1 cells. This observation was consistent with previous data DIABETES, VOL. 53, JUNE 2004
showing the presence of CaM-KIV mRNA in other insulinoma cell lines and in RINm5F and HIT-T15 cells (12). To examine subcellular localization of CaM-KK␣ and CaM-KIV in rat pancreas, we performed immunohistochemical analyses of pancreatic islets from 8-week-old rats using anti–CaM-KIV or anti–CaM-KK antibodies. Results (Fig. 1B) showed immunostaining of both CaM-KIV and CaM-KK␣ in the pancreatic islets. Immunostaining for the two species was most prominent in the cytoplasm of the islet cells. Identity of the pancreatic -cells (Fig. 1C) was determined by staining the cytoplasm with the antiinsulin antibody (Fig. 1C, panels a and c). In addition, the cytoplasm of ⬎90% of pancreatic -cells showed staining by anti–CaM-KK (Fig. 1C, panel b) or anti–CaM-KIV (Fig. 1C, panel d) antibodies. Together, these findings show that both CaM-KK␣ and CaM-KIV were present in rat pancreatic islets and INS-1 cells. Furthermore, both genes were expressed at the protein level (Fig. 2A) and found to be present in the cytoplasm of islet cells. Identification of Ca2ⴙ/CaM-K cascade in INS-1 cells. The presence of the CaM-K cascade in the pancreatic -cells prompted us to ask whether it participates in the regulation of gene expression and specifically in connec1477
CaM-K CASCADE ON INSULIN GENE EXPRESSION
FIG. 3. Effect of the CaM-KK/CaM-KIV cascade on insulin promoter activity. CaM-KK/CaM-KIV cascade activated insulin promoter activity in INS-1 cells. The cells were transfected with ⴚ238WT-LUC and empty vector, CaM-KKc, and CaM-KIVc/CaM-KIVw expression vectors. The results are expressed as relative luciferase activity compared with control cells arbitrarily set at 100. Each data point shows the mean ⴞ SE (n ⴝ 3) of separate transfections. *Significant difference (P < 0.05).
tion with the insulin gene. Therefore, we examined components of the cascade in the insulinoma cell line, INS-1, using both immunological and biochemical methods. Results (Fig. 2A) from Western blot analysis showed that both CaM-KK␣ and CaM-KIV were detected in the cells probed with their respective specific antibodies. Next, we wondered whether the CaM-KK in the cells was functionally active. To address this question, we added recombinant CaM-KI to extract from INS-1 cells for 20 min in the presence of Mg-ATP and Ca2⫹/CaM and measured both protein kinase activity and [32P] incorporation of GST-CaM-KI. Results (Fig. 2B) showed a dramatic upregulation of the catalytic activity of GST-CaM-KI that was ⬃10-fold higher than control as well as upregulation of [32P] incorporation into GST-CaM-KI. Previous studies demonstrated that CaM-KK activated both CaM-KI and CaM-KIV through phosphorylation of a Thr residue (Thr196 in CaM-KIV and Thr177 in CaM-KI). To show that CaM-KK activity in INS-1 extract activates GST-CaM-KIV via the same mechanism, we added to the reaction a mutant GST-CaM-KI that contains a Thr177Ala substitution. As predicted, phosphorylation and activation of the mutant GST-CaM-KI by INS-1 cell extract was not observed. These results clearly demonstrate that INS-1 cell extract contains both CaM-KK and CaM-KIV. Effect of CaM-K cascade on insulin gene transcription. Previous studies (10,12) showed that secreted insulin may upregulate insulin gene transcription via signaling through activation of CaM-KIV. However, whether the CaM-K cascade is a part of this process is not known. Published evidence (25) supports a role for CaM-KIV in transcriptional regulation, and because this kinase is activated by CaM-KK, it is likely that this upstream kinase is also involved. Therefore, we examined the potential role of the CaM-K cascade by cotransfecting INS-1 cells with ⫺238 WT-LUC plus CaM-KIV and/or CaM-KK to determine whether these latter proteins affected insulin gene transcription. Results (Fig. 3) showed that constitutively active CaM-KIV (CaM-KIVc) alone stimulated a sixfold increase in insulin promoter activity in INS-1 cells. Furthermore, the introduction of nonactivated wild-type CaM-KIV along 1478
with constitutively active CaM-KK (CaM-KKc) caused a seven- to eightfold increase in transcriptional activation of insulin gene. In contrast, the cotransfection of nonactivated wild-type CaM-KIV and wild-type CaM-KK did not activate the reporter pINS-LUC (Fig. 3, last column). CaM-KIV upregulation of insulin gene expression and secretion. To further solidify results from the preceding studies, we generated recombinant adenoviruses carrying the Ca2⫹/CaM-independent form of CaM-KIV (CaM-KIVc) (22). In these studies, INS-1 cells were infected with recombinant adenoviruses that carried the CaM-KIVc, and extract from these cells was tested for their ability to phosphorylate syntide-2. This is a peptide substrate that can be phosphorylated by CaM-KIV (14). To confirm the upregulation of insulin by CaM-KIVc, we performed Northern blot analysis using total cellular RNA prepared from adenovirus-infected INS-1 cells. Results (Fig. 4A) showed that insulin mRNA was significantly upregulated by the expression in the presence of CaM-KIVc, while -actin was not. In support of this finding, the ELISA of insulin peptide secretion from the adenovirus-infected INS-1 cells was fourfold higher in the presence of CaM-KIVc. The addition of CaM-KKc to the preceding cells caused a further increase of insulin secretion to fivefold above baseline (Fig. 4B). These findings show that expression of CaM-KIVc strongly stimulates both transcription and secretion of insulin from INS-1 cells. Glucose upregulation of CaM-KIV activity. To keep blood glucose concentrations within narrow limits, the body must impose strict regulation and have fast-acting mechanisms to guarantee efficient insulin secretion and biosynthesis (1,2). In these processes, glucose itself has been shown to act as the major nutrient regulator that triggers a cascade, referred to as the stimulus-response coupling. To determine whether the CaM-K cascade is part of this mechanism, we examined the ability of glucose to stimulate activity of CaM-KIV in INS-1 cells. Therefore, INS-1 cells were treated with 11.2 mmol/l glucose for 1 min before harvest at indicated time periods. The cell extracts were subjected to immunoprecipitation using an anti– CaM-KIV antibody, and kinase activities were also assayed DIABETES, VOL. 53, JUNE 2004
X. YU AND ASSOCIATES
FIG. 5. Glucose stimulates the activity of CaM-KIV. INS-1 cells were exposed to 11.2 mmol/l glucose for 1 min before harvest at indicated time periods, and CaM-KIV activity was determined as described in RESEARCH DESIGN AND METHODS. Each data point shows the mean ⴞ SE (n ⴝ 4) of separate determinations. *Significant difference (P < 0.05).
posed to 2.8 mmol/l glucose was the same (Fig. 6, columns 1 and 2). In contrast, when the same cells were exposed to 11.2 mmol/l glucose there was a marked eightfold increase in ⫺238WT-LUC activity, but this increase was abrogated by overexpression of CaM-KIVdn. Together, these findings support the idea that CaM-KIV is required for glucose upregulation of insulin expression in INS-1 cells. DISCUSSION
FIG. 4. The CaM-KK/CaM-KIV cascade stimulates insulin gene expression and secretion. A: CaM-KIVc or CaM-KKc/CaM-KIVc upregulation of insulin gene expression in INS-1 cells was detected using Northern blot analysis with [32P]-labeled rat insulin cDNA (top), and [32P]labeled rat -actin cDNA was used as a control (bottom). Lane 1, INS-1 without treatment; lane 2, INS-1 cell expressing lacZ; lane 3, INS-1 cell expressing CaM-KIVc; lane 4, INS-1 cell expressing CaM-KIVc/CaMKKc. B: CaM-KIVc or CaM-KKc/CaM-KIVc upregulation of basal insulin secretion. LacZ; INS-1 cell expressing lacZ, CaM-KIVc; INS-1 cell expressing CaM-KIVc, CaM-KKc ⴙ CaM-KIVc; INS-1 cell expressing CaM-KKc and CaM-KIVc. Each data point shows the mean ⴞ SE (n ⴝ 4) of separate transfection experiments. *Significant difference (P < 0.05) from the control.
using syntide-2. In comparison with nonstimulated cells, glucose treatment caused a roughly twofold increase in CaM-KIV activity. The upregulation kinetics of CaM-KIV activity reached a peak after 5 min of exposure to the glucose and returned to a basal level after 30 min (Fig. 5). These findings show that glucose stimulates the activation of CaM-KIV and is consistent with the idea that glucose stimulation of insulin activity may involve a role for the CaM-K cascade. Participation of CaM-KIV in glucose upregulation of insulin gene expression. To examine whether CaM-KIV plays a role in glucose stimulation of signaling to enhance insulin promoter activity, we cotransfected INS-1 cells with ⫺238WT-LUC and with either an expression plasmid encoding a dominant-negative mutant of CaM-KIV (CaMKIVdn) or control, an empty expression plasmid. Results showed that activity of the reporter, ⫺238WT-LUC, plus either empty vector or CaM-KIVdn in transfected cells exDIABETES, VOL. 53, JUNE 2004
In the present study, we have examined the role of the CaM-KK/CaM-KIV cascade on insulin gene expression in response to glucose. CaM-Ks are implicated in the regulation of a wide variety of biological events requiring the participation of intracellular Ca2⫹, such as muscle contraction, neurotransmitter release, and gene expression
FIG. 6. CaM-KIVdn inhibits the upregulation of insulin promoter activity by glucose stimulation. INS-1 cells were transfected with ⴚ238WT-LUC and CaM-KIVdn and cultured in media containing either 2.8 or 11.2 mmol/l glucose for 24 h prior to measuring luciferase activity. Control, ⴚ238WT-LUC and pcDNA3; dominant-negative CaMKIV, ⴚ238WT-LUC and CaM-KIVdn expression vector. The results are expressed as relative luciferase activities compared with control cells arbitrarily set at 100. Each data point shows the mean ⴞ SE (n ⴝ 4) of separate transfections. 1479
CaM-K CASCADE ON INSULIN GENE EXPRESSION
(11,26). Recent studies (19,26) have indicated that socalled multifunctional CaM-Ks, including CaM-KI, -II, and -IV, are regulated by phosphorylation either by itself or an upstream protein kinase, CaM-KK. Previous reports (12) showed that CaM-KIV was expressed in pancreatic islet and other insulin-secreting cells, including MIN6 and HIT cells. In the first set of studies, we established that the respective mRNAs encoding CaM-KIV and its upstream protein kinase, CaM-KK, were present in rat pancreatic islet and the insulin-secreting cell line, INS-1. Furthermore, the mRNA was expressed because both Western blot and immunohistochemistry of the cells showed the presence of the protein products in the cells. The expression of these genes does not reveal any information regarding its function in the cell. The possible existence of a short-term effect of glucose on insulin gene transcription was suggested indirectly, but as yet has never been clearly shown by previous reports (27,28). The results of Efrat et al. (29) clearly demonstrated that increased levels of nuclear insulin mRNA can be detected in TC-3 cells as early as 10 min after glucose stimulation. Similar dynamics obtained in the previous study (30), when analyzing transcription initiation in HIT cells after a 15-min glucose stimulus on top of a 0.1-mmol/l glucose background, support these earlier findings. These data clearly demonstrate a short-term regulation of insulin gene transcription by glucose. We showed here that the enzyme activity of CaM-KIV was elevated within 2 min by glucose stimulation and that activated CaM-KK/CaM-KIV cascade stimulated the insulin transcription. Additionally, CaM-KIVdn prevented the activation of insulin transcription by glucose stimulation; therefore, not only CaM-KIV but also CaM-KK/CaM-KIV cascade might participate in glucose-stimulated insulin gene transcription. Recently, Leibiger et al. (10) reported that secreted insulin in response to glucose stimulation is a key factor in glucosestimulated insulin gene transcription. On the other hand, several studies have indicated that glucose can also regulate the insulin gene independently of insulin secretion (30,31); furthermore, insulin gene expression appears to be largely unaltered after targeted disruption of the insulin receptor gene in the islet -cell, at least in younger animals (32). Although the possibility that secreted insulin in response to glucose stimulation may activate the insulin gene transcription cannot be excluded at this point, we found that in our in vitro study with INS-1 cells, glucokinase, a key enzyme of glucose metabolism in the pancreatic -cell, was regulated by CaM-KK/CaM-KIV cascade (Y.S., K.M., unpublished data). The role of the CaM-KK/CaM-KIV cascade in insulin synthesis as well as in glucose metabolism in pancreatic -cells deserves further investigation. One of the most common mechanisms by which elevated intracellular Ca2⫹ regulates cellular events is through its association with CaM. The Ca2⫹/CaM complex binds to and modulates the functions of multiple key regulatory proteins, including a family of CaM-Ks (33). A role for transcriptional regulation by CaM-K is suggested by the observation that the Ca2⫹-dependent transcription of three immediate early genes (c-fos, NGFI-A, and NGFI-B) was 80% blocked by the CaM-K inhibitor, KN-62 (25). CaM-KK has been identified and cloned as an activator for two multifunctional CaM-KKs, CaM-KI and -IV. 1480
Phosphorylation by CaM-KK of Thr, located in the activation loop of the catalytic domain of CaM-KIV, results in a large increase in the catalytic efficiency of CaM-K (20,34). Numerous studies have demonstrated that the CaM-KK/ CaM-KIV cascade is present and functional in various cell types, such as the Jurkat cell (26), cultured hippocampal neurons (35), and transfected COS-7 cells (19). We showed that both pancreatic -cells and the insulin-secreting cell line, INS-1, have this signal cascade. CaM-KIV, which has significant nuclear localization (36), phosphorylates transcription factors such as cAMP-responsive element binding protein (CREB) and serum response factor (37). Several studies have shown that CaM-KIV can mediate transcriptional stimulation through CREB phosphorylation (38,39). More recently, we have demonstrated (34) that cotransfection of nonactivated wild-type CaM-KIV with CaM-KKc stimulates CREB-mediated transcription 14-fold relative to either kinase alone. This is consistent with the in vitro observation that the Vmax-to-Km ratio of CREB phosphorylation was increased 30-fold by activation of CaM-KIV by CaM-KK (40). The promoter region of the insulin gene has been cloned, sequenced, and shown to contain putative consensus binding sites for a variety of transcription factors. The insulin promoter contains several cis-acting regulatory elements located within the 5⬘ flanking region. These elements interact with transacting factors. Some of these factors are found ubiquitously, whereas others are more restricted to the -cell. Important transcriptional regulatory elements have been identified in the promoter regions of the insulin gene; two of the more important elements are the A3 and C1 (4 – 6). It is well known that glucose increases the concentration of both cAMP and Ca2⫹ in pancreatic -cells (7) and triggers insulin secretion from pancreatic -cells. In rat primary cultured islets, ATF-2 (also called CREBP1) increased the glucose-upregulated activities of the human insulin promoter (12). ATF-2 was originally isolated as a protein binding to cAMP response elements (41) and belongs to the CREB/ATF family characterized by kination and basic leucine zipper DNA-binding domains, which is consistent with the previous report (42) that ATF-2 is a target of JNK and p38 MAP kinase. ATF-2 is capable of forming homodimers or heterodimers with c-Jun and of binding to cAMP response elements, including the promoter region in the human insulin gene (43). The transcription of the human insulin gene, enhanced by an elevated intracellular concentration of calcium ions, was completely blocked by Ca2⫹/CaM-K inhibitor (10). The activity of the transcription factor ATF-2, which binds to the cAMP responsive elements of the human insulin gene, was enhanced by CaM-KIV, but CaM-KII or protein kinase A has no effect on ATF-2 activity (12). Further investigation will be needed to clarify the mechanism of the CaM-KK/CaM-KIV cascade on insulin gene transcription. In summary, we examined the role of the CaM-KK/CaMKIV cascade on insulin gene expression in response to glucose stimulation in the insulinoma cell line. The results indicate that the CaM-KK/CaM-KIV cascade stimulated insulin gene transcription, suggesting that the activation of the CaM-KK/CaM-KIV cascade may play an important role in insulin biosynthesis in pancreatic -cells. DIABETES, VOL. 53, JUNE 2004
X. YU AND ASSOCIATES
ACKNOWLEDGMENTS
K.M. is supported in part by a Grant-in-Aid for Scientific Research (no. 14770601). We thank Kazuko Yamaji and Kiyo Ueeda for their technical assistance. REFERENCES 1. Welsh M, Nielsen DA, MacKrell AJ, Steiner DF: Control of insulin gene expression in pancreatic beta-cells and in an insulin-producing cell line, RIN-5F cells. II. Regulation of insulin mRNA stability. J Biol Chem 260:13590 –13594, 1985 2. Brun T, Roche E, Kim KH, Prentki M: Glucose regulates acetyl-CoA carboxylase gene expression in a pancreatic beta-cell line (INS-1). J Biol Chem 268:18905–18911, 1993 3. Rutter GA, Tavare JM, Palmer DG: Regulation of mammalian gene expression by glucose. News Physiol Sci 15:149 –154, 2000 4. Melloul D, Marshak S, Cerasi E: Regulation of insulin gene transcription. Diabetologia 45:309 –326, 2002 5. Shieh SY, Tsai MJ: Cell-specific and ubiquitous factors are responsible for the enhancer activity of the rat insulin II gene. J Biol Chem 266:16708 – 16714, 1991 6. Peshavaria M, Gamer L, Henderson E, Teitelman G, Wright CV, Stein R: XIHbox 8, an endoderm-specific Xenopus homeodomain protein, is closely related to a mammalian insulin gene transcription factor. Mol Endocrinol 8:806 – 816, 1994 7. Wollheim CB, Sharp GW: Regulation of insulin release by calcium. Physiol Rev 61:914 –973, 1981 8. Worley JF 3rd, McIntyre MS, Spencer B, Mertz RJ, Roe MW, Dukes ID: Endoplasmic reticulum calcium store regulates membrane potential in mouse islet beta-cells. J Biol Chem 269:14359 –14362, 1994 9. Meldolesi J, Pozzan T: The endoplasmic reticulum Ca2⫹ store: a view from the lumen. Trends Biochem Sci 23:10 –14, 1998 10. Leibiger IB, Leibiger B, Moede T, Berggren PO: Exocytosis of insulin promotes insulin gene transcription via the insulin receptor/PI-3 kinase/ p70 s6 kinase and CaM kinase pathways. Mol Cell 1:933–938, 1998 11. Soderling TR: The Ca-calmodulin-dependent protein kinase cascade. Trends Biochem Sci 24:232–236, 1999 12. Ban N, Yamada Y, Someya Y, Ihara Y, Adachi T, Kubota A, Watanabe R, Kuroe A, Inada A, Miyawaki K, Sunaga Y, Shen ZP, Iwakura T, Tsukiyama K, Toyokuni S, Tsuda K, Seino Y: Activating transcription factor-2 is a positive regulator in CaM kinase IV–induced human insulin gene expression. Diabetes 49:1142–1148, 2000 13. Hayashi N, Matsubara M, Takasaki A, Titani K, Taniguchi H: An expression system of rat calmodulin using T7 phage promoter in Escherichia coli. Protein Expr Purif 12:25–28, 1998 14. Eto K, Takahashi N, Kimura Y, Masuho Y, Arai K, Muramatsu MA, Tokumitsu H: Ca(2⫹)/calmodulin-dependent protein kinase cascade in Caenorhabditis elegans: implication in transcriptional activation. J Biol Chem 274:22556 –22562, 1999 15. Sutton R, Peters M, McShane P, Gray DW, Morris PJ: Isolation of rat pancreatic islets by ductal injection of collagenase. Transplantation 42:689 – 691, 1986 16. Murao K, Wada Y, Nakamura T, Taylor AH, Mooradian AD, Wong NCW: Effects of glucose and insulin on rat apolipoprotein A-I gene expression. J Biol Chem 273:18959 –18965, 1998 17. Anderson KA, Means RL, Huang QH, Kemp BE, Goldstein EG, Selbert MA, Edelman AM, Fremeau RT, Means AR: Components of a calmodulindependent protein kinase cascade: molecular cloning, functional characterization and cellular localization of Ca2⫹/calmodulin-dependent protein kinase kinase beta. J Biol Chem 273:31880 –1889, 1998 18. Hsu LS, Tsou AP, Chi CW, Lee CH, Chen JY: Cloning, expression and chromosomal localization of human Ca2⫹/calmodulin-dependent protein kinase kinase. J Biomed Sci 5:141–149, 1998 19. Tokumitsu H, Soderling TR: Requirements for calcium and calmodulin in the calmodulin kinase activation cascade. J Biol Chem 271:5617–5622, 1996 20. Tokumitsu H, Brickey DA, Glod J, Hidaka H, Sikela J, Soderling TR: Activation mechanisms for Ca2⫹/calmodulin-dependent protein kinase IV: identification of a brain CaM-kinase IV kinase. J Biol Chem 269:28640 – 28647, 1994 21. Tokumitsu H, Inuzuka H, Ishikawa Y, Ikeda M, Saji I, Kobayashi R:
DIABETES, VOL. 53, JUNE 2004
STO-609, a specific inhibitor of the Ca(2⫹)/calmodulin-dependent protein kinase kinase. J Biol Chem 277:15813–15818, 2002 22. Inuzuka H, Tokumitsu H, Ohkura N, Kobayashi R: Transcriptional regulation of nuclear orphan receptor, NOR-1, by Ca(2⫹)/calmodulin-dependent protein kinase cascade. FEBS Lett 522:88 –92, 2002 23. Sayo Y, Murao K, Imachi H, Cao WM, Sato M, Dobashi H, Wong NC, Ishida T: The multiple endocrine neoplasia type 1 gene product, menin, inhibits insulin production in rat insulinoma cells. Endocrinology 143:2437–2440, 2002 24. Sayo Y, Hosokawa H, Imachi H, Murao K, Sato M, Wong NC, Ishida T, Takahara J: Transforming growth factor beta induction of insulin gene expression is mediated by pancreatic and duodenal homeobox gene-1 in rat insulinoma cells. Eur J Biochem 267:971–978, 2000 25. Enslen H, Soderling TR: Roles of calmodulin-dependent protein kinases and phosphatase in calcium-dependent transcription of immediate early genes. J Biol Chem 269:20872–20877, 1994 26. Park IK, Soderling TR: Activation of Ca2⫹/calmodulin-dependent protein kinase (CaM-kinase) IV by CaM-kinase kinase in Jurkat T lymphocytes. J Biol Chem 270:30464 –30469, 1995 27. Permutt MA, Kipnis DM: Insulin biosynthesis. II. Effect of glucose on ribonucleic acid synthesis in isolated rat islets. J Biol Chem 247:1200 – 1207, 1972 28. Koh G, Seino Y, Takeda J, Fukumoto H, Kurose T, Tsuji K, Tsuda K, Taminato K, Imura H: Short term effects of glucose and arginine on the preproinsulin messenger ribonucleic acid level in the perfused rat pancreas: comparison with insulin secretion. Endocrinology 124:707–711, 1989 29. Efrat S, Surana M, Fleischer N: Glucose induces insulin gene transcription in a murine pancreatic beta-cell line. J Biol Chem 266:11141–11143, 1991 30. de Vargas LM, Sobolewski J, Siegel R, Moss LG: Individual beta cells within the intact islet differentially respond to glucose. J Biol Chem 272:26573– 26577, 1997 31. Goodison S, Kenna S, Ashcroft SJH: Control of insulin gene expression by glucose. Biochem J 285:563–568, 1992 32. Kulkarni RN, Bruning JC, Winnay JN, Postic C, Magnuson MA, Kahn CR: Tissue-specific knockout of the insulin receptor in pancreatic beta cells creates an insulin secretory defect similar to that in type 2 diabetes. Cell 96:329 –339, 1999 33. Nairn AC, Picciotto MR: Calcium/calmodulin-dependent protein kinases. Semin Cancer Biol 5:295–303, 1994 34. Tokumitsu H, Enslen H, Soderling TR: Characterization of a Ca2⫹/ calmodulin-dependent protein kinase cascade: molecular cloning and expression of calcium/calmodulin-dependent protein kinase kinase. J Biol Chem 270:19320 –19324, 1995 35. Bading H, Ginty DD, Greenberg ME: Regulation of gene expression in hippocampal neurons by distinct calcium signaling pathways. Science 260:181–186, 1993 36. Jensen KF, Ohmstede CA, Fisher RS, Sahyoun N: Nuclear and axonal localization of Ca2⫹/calmodulin-dependent protein kinase type Gr in rat cerebellar cortex. Proc Natl Acad Sci U S A 88:2850 –2853, 1991 37. Cruzalegui FH, Means AR: Biochemical characterization of the multifunctional Ca2⫹/calmodulin-dependent protein kinase type IV expressed in insect cells. J Biol Chem 268:26171–2618, 1993 38. Enslen H, Sun P, Brickey D, Soderling SH, Klamo E, Soderling TR: Characterization of Ca2⫹/calmodulin-dependent protein kinase IV: role in transcriptional regulation. J Biol Chem 269:15520 –15527, 1994 39. Matthews RP, Guthrie CR, Wailes LM, Zhao X, Means AR, McKnight GS: Calcium/calmodulin-dependent protein kinase types II and IV differentially regulate CREB-dependent gene expression. Mol Cell Biol 14:6107– 6116, 1994 40. Enslen H, Tokumitsu H, Soderling TR: Phosphorylation of CREB by CaM-kinase IV activated by CaM-kinase IV kinase. Biochem Biophys Res Commun 207:1038 –1043, 1995 41. Maekawa T, Sakura H, Kanei-Ishii C, Sudo T, Yoshimura T, Fujisawa J, Yoshida M, Ishii S: Leucine zipper structure of the protein CRE-BP1 binding to the cyclic AMP response element in brain. EMBO J 8:2023–2028, 1989 42. Gupta S, Campbell D, Derijard B, Davis RJ: Transcription factor ATF2 regulation by the JNK signal transduction pathway. Science 267:389 –393, 1995 43. Inagaki N, Maekawa T, Sudo T, Ishii S, Seino Y, Imura H: c-Jun represses the human insulin promoter activity that depends on multiple cAMP response elements. Proc Natl Acad Sci U S A 89:1045–1049, 1992
1481
