- Email: [email protected]
The protein tyrosine phosphatase α modifies insulin secretion in INS-1E cells
BBRC Biochemical and Biophysical Research Communications 311 (2003) 361–364 www.elsevier.com/locate/ybbrc
The protein tyrosine phosphatase a modifies insulin secretion in INS-1E cellsq Katja Kapp, Elisabeth Metzinger, Monika Kellerer,1 Hans-Ulrich H€ aring, * and Reiner Lammers Medical Clinic IV, Otfried-M€uller-Str. 10, 72076 T€ ubingen, Germany Received 17 September 2003
Abstract Increasing evidence indicates a role of insulin signalling for insulin secretion from the pancreatic b-cells. Therefore, regulators of insulin signalling, like protein tyrosine phosphatases, could also have an impact on insulin secretion. Here, we investigated a possible role of the negative regulator protein tyrosine phosphatase a (PTPa) for insulin secretion. RT-PCR analysis confirmed that both splice variants of the extracellular domain of PTPa that vary by an insert of 9 amino acids are expressed in human islets and insulinoma cells (INS-1E, RIN1046-38). Overexpression of the wild type PTPa splice variant containing the 9 amino acids reduced insulin secretion, as did a mutant form unable to bind Grb2 (Tyr798Phe). By contrast, overexpression of a phosphatase inactive mutant improved insulin secretion. These data reveal a functional relevance of PTPa for insulin secretion. Ó 2003 Elsevier Inc. All rights reserved. Keywords: Insulin secretion; Protein tyrosine phosphatase; PTPa; Splice variant
Insulin secretion from pancreatic b-cells is achieved by diverse secretagogues with glucose being the major physiological stimulus. Although basic mechanisms of insulin secretion are understood, not all signalling molecules involved are yet known or fully characterized. Several lines of evidence indicate that insulin signalling via the insulin receptor plays a role for insulin secretion. The pancreatic knockout of the insulin receptor (IR) [1] or IRS-1 [2] leads to reduced insulin secretion. Protein tyrosine kinase inhibitors were shown to decrease insulin secretion [3] while other data reveal a stimulation of insulin secretion [4,5]. Vanadate, an inhibitor of protein tyrosine phosphatases (PTPs), can also modulate insulin secretion [6]. Since PTPs are modulators of the insulin signalling cascade, these data point to a role of PTPs in insulin secretion. Up to now, for two different members q
Abbreviations: PTPa, protein tyrosine phosphatase a; IR, insulin receptor; PTP, protein tyrosine phosphatase. * Corresponding author. Fax: +49-7071-29-5974. E-mail address: [email protected] (R. Lammers). 1 Present address: Marienhospital Stuttgart, B€ oheimstr. 37, 70199 Stuttgart, Germany. 0006-291X/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2003.10.011
of the PTP family a role in insulin secretion has been described. The overexpression of PTPr is linked to an impaired glucose induced insulin secretion in Goto– Kakizaki rats [7]. By contrast, the genomic deletion of the enzymatically inactive PTP IA-2, which is localized in the secretory granules, impairs insulin secretion [8]. In this study, we wanted to investigate the effect of PTPa on insulin secretion. PTPa can directly dephosphorylate the insulin receptor (IR) and is, therefore, a negative regulator of the insulin signal [9,10]. Human PTPa consists of a short extracellular, a transmembrane, and a cytoplasmic domain with two phosphatase domains. There are two isoforms due to an alternatively spliced, 27 bp containing exon occurring in the extracellular domain. Previous data on the presence of these isoforms revealed that both forms are tissue-specifically expressed and that they occur in mice and rats [11]. PTPa is involved in the translocation of glucose transporter 4 in rat adipose cells [12] and specifically inhibits insulin-increased prolactin gene expression [13], but a function in insulin secretion has yet not been observed. Here we show that overexpression of PTPa has an effect on insulin secretion.
362
K. Kapp et al. / Biochemical and Biophysical Research Communications 311 (2003) 361–364
Materials and methods Cell culture. INS-1E cells [14] were cultured in RPMI 1640 medium supplemented with 10% fetal calf serum, 10 mM Hepes (pH 7.4), 2 mM glutamine, 1 mM Na-pyruvate, and 50 lM b-mercaptoethanol. For stable expression, cells were infected with retroviruses generated from the retroviral expression vector pLXSN as described [15]. Geneticin selection was started 48 h after infection and resistant cells were pooled. RIN 1046-38 cells were cultured in medium 199 with Earle’s Salts, supplemented with 2 mM glutamine, 10% fetal calf serum and glucose was adjusted to 11.1 mM. RNA extraction and RT-PCR. Total RNA from human islets or cells was isolated using the RNAeasy Kit (Quiagen). For reverse transcription, first strand cDNA synthesis kit was used (Roche). PCR was done as suggested by the manufacturer (Eppendorf) with the primers for endogenous human PTPa (forward 50 -AGCAAGCACC AATTCTATAGGC, reverse 50 GTTGGATAAGCGGAAAGAAT TG), for endogenous rat PTPa (forward 50 -CTGATAACCAGT TCACGGATGC, reverse 50 -TGGCCAGAAGTGGTACACTTTG), and for the transfected human PTPa (forward 50 -GATGGAGG GTCCTCGCTGGGTCAGACACTGGAACAGT ATGAG and reverse 50 -CTTTCCACACCCTAACT GACAC [pLXSN-30 ]), 4–5 ll of the reverse transcribed mRNA or 200–250 ng of total RNA, and 2.5 U Taq-Polymerase. A fifth of each PCR was separated on a 2.5% agarose gel and stained with ethidium bromide. Insulin secretion assay. INS-1E cells were seeded 2 days before use at a density of 1 105 /cm2 . To determine insulin secretion, cells were washed twice with modified Krebs–Ringer bicarbonate buffer (25 mM Hepes, pH 7.4; 125 mM NaCl, 4.74 mM KCl, 1 mM CaCl2 , 1.2 mM KH2 PO4 , 1.2 mM MgSO4 , 5 mM NaHCO3 , 0.1% BSA, and 2.8 mM D (+)glucose) and incubated for 2 h at 37 °C for adaptation, 1 h at 37 °C (basal secretion), and 1 h at 37 °C (stimulated secretion) in Krebs–Ringer bicarbonate buffer. Supernatants were centrifuged at 250g for 5 min and stored at )20 °C until used. Insulin content was measured using a rat insulin RIA kit (Linco Research). Statistical analysis. Data were analyzed by one-way analysis of variance (ANOVA), followed by Student’s t tests for unpaired groups. The statistical software package JMP (SAS Institute, Cary, NC) was used.
Results INS-1E cells are a derivative of the INS-1 cell line and at low passage number have the unique property to secrete insulin in a glucose responsive way [14]. We therefore used this cellular model system to investigate a possible role of PTPa for insulin secretion. PTPa occurs with two splicing variants, which differ in size by 27 bp and are referred to as long or short isoform, respectively. First, we investigated if both forms are present in human islets and in the rat pancreatic b-cell lines, RIN 1046-38 and INS-1E. Total RNA was prepared and RTPCR was performed. The PCR covered the alternatively spliced exon and yielded DNA fragments of the expected size (276 and 303 bp, respectively; Fig. 1A). They indicate that most of the PTPa is present in primary islets and both cell lines as the short form. As control, a PCR with RNA did not result in the identification of any product. Since PTPa is expressed in pancreatic cells, we investigated the effect of PTPa overexpression on insulin
Fig. 1. Expression of PTPa in human islets or insulinoma-derived cell lines. RNA was isolated, reverse transcribed as indicated, subjected to PCR, and analyzed on an agarose gel. (A) Expression of endogenous PTPa (+) and control without reverse transcription ()). (B) Overexpression of PTPa isoforms in transfected INS-1E cell pools (+) and control without reverse transcription ()). DNA marker: 50 base pair ladder; CS, inactive PTPa (Cys 442, 732 to Ser); WT, native PTPa; and YF, Tyr 798 to Phe.
secretion. INS-1E cells were infected with retroviruses encoding the long form of wild type PTPa, a mutant form unable to bind Grb2 (Tyr 798 to Phe; [16]) or a phosphatase inactive mutant (Cys 442 and 732 to Ser; [16]). The Tyr 798 to Phe mutant was included since in previous experiments this mutant showed a different localization than the wild type phosphatase, which could affect its substrate specificity [15]. Following G418 selection, positive cells were pooled and directly used for insulin secretion to maintain a low passage number required for an optimal glucose response of the cells. To prove that PTPa was expressed from the transgene, RNA from pools of infected cells was prepared and used for RT-PCR. To this end, a reverse primer complementary to the 30 UTR of the transcripts ensured the detection of transgene-derived transcripts. Fig. 1B showed the expected PCR products of about 200 bp. As control, a PCR was performed directly with RNA, which resulted only in background signals. The amplification products were confirmed by sequencing. Employing the established cell pools, we investigated the function of the different PTPa forms by measuring insulin secretion upon glucose stimulation (Fig. 2). In cells overexpressing the wild type phosphatase or the tyrosine mutant insulin secretion was significantly reduced upon glucose stimulation when compared with
K. Kapp et al. / Biochemical and Biophysical Research Communications 311 (2003) 361–364
363
the wild type PTPa. Beyond that, it was shown that PTPa dephosphorylates Kv2.1 upon overexpression in 293 cells [19], adding another possible substrate in pancreatic b-cells. A closer investigation of b-cells may now help us to reveal the differential activity of the PTPa mutants and to elucidate the role of this new regulator of secretion.
Acknowledgments Fig. 2. Glucose stimulated secretion of insulin. Basal and glucose stimulated insulin secretion was determined in parental and in PTPa overexpressing INS-1E cells. The cells were counted in parallel to insulin secretion. Abbreviations are as in Fig. 1. Data are shown as means SEM from quadruplicate samples, with p values indicated as *p < 0:01 and **p < 0:05.
We thank P. Maechler and C.B. Wollheim for providing us with INS-1E cells and we are grateful to the group of M. Stumvoll for help with the statistical analysis. This work was supported by a grant from the DFG to R.L.
References parental cells (P values as indicated in the figure). By contrast, overexpression of the phosphatase inactive mutant resulted in higher amounts of secreted insulin in comparison to the control cells (P < 0:01; n ¼ 4). We conclude that PTPa is involved in the regulation of glucose-dependent insulin secretion in b-cells.
Discussion Insulin signalling is important for insulin secretion of the pancreatic b-cells. Therefore, PTPs as mediators of the insulin signalling could also play a role for insulin secretion. For some PTPs, a functional relevance for this process has been demonstrated [7,8]. Several studies have suggested that PTPa negatively regulates the insulin signal, likely by dephosphorylation of the IR [9,12,13]. We therefore investigated a possible role of PTPa for secretion of insulin. In the present study, we showed that both splice variants of PTPa occur in human islets as well as in rat insulinoma cells and that an overexpression of various forms of PTPa changed insulin secretion in response to glucose. Overexpression of a phosphatase inactive PTPa led to higher insulin secretion while the phosphatase active forms reduced secretion. Since PTPa is a PTP that directly dephosphorylates the IR, the reduced secretion upon overexpression may be mediated by an impaired signal from the IR. Alternatively, a PTPa substrate different from the IR could as well be important for secretion. Possible candidate substrates are the voltage-dependent (delayed rectifier) Kþ channels (Kv channels) that are involved in the glucose response of b-cells. In general, dephosphorylation leads to activation of these channels and thereby to cessation of insulin secretion [17]. Kv1.2 has been shown to be a substrate of PTPa and is activated upon dephosphorylation [18], which would fit to the results for
[1] R.N. Kulkarni, J.C. Bruning, J.N. Winnay, C. Postic, M.A. Magnuson, C.R. Kahn, 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 (1999) 329–339. [2] R.N. Kulkarni, J.N. Winnay, M. Daniels, J.C. Bruning, S.N. Flier, D. Hanahan, C.R. Kahn, Altered function of insulin receptor substrate-1-deficient mouse islets and cultured beta-cell lines, J. Clin. Invest. 104 (1999) R69–R75. [3] S.J. Persaud, T.E. Harris, C.J. Burns, P.M. Jones, Tyrosine kinases play a permissive role in glucose-induced insulin secretion from adult rat islets, J. Mol. Endocrinol. 22 (1999) 19–28. [4] R.L. Sorenson, T.C. Brelje, C. Roth, Effect of tyrosine kinase inhibitors on islets of Langerhans: evidence for tyrosine kinases in the regulation of insulin secretion, Endocrinology 134 (1994) 1975–1978. [5] E.J. Verspohl, B. Tollkuhn, H. Kloss, Role of tyrosine kinase in insulin release in an insulin secreting cell line (INS-1), Cell. Signal. 7 (1995) 505–512. [6] S. Gogg, J. Chen, S. Efendic, U. Smith, C. Ostenson, Effects of phosphotyrosine phosphatase inhibition on insulin secretion and intracellular signaling events in rat pancreatic islets, Biochem. Biophys. Res. Commun. 280 (2001) 1161–1168. [7] C.G. Ostenson, A.C. Sandberg-Nordqvist, J. Chen, M. Hallbrink, D. Rotin, U. Langel, S. Efendic, Overexpression of protein– tyrosine phosphatase PTP sigma is linked to impaired glucoseinduced insulin secretion in hereditary diabetic Goto–Kakizaki rats, Biochem. Biophys. Res. Commun. 291 (2002) 945–950. [8] K. Saeki, M. Zhu, A. Kubosaki, J. Xie, M.S. Lan, A.L. Notkins, Targeted disruption of the protein tyrosine phosphatase-like molecule IA-2 results in alterations in glucose tolerance tests and insulin secretion, Diabetes 51 (2002) 1842–1850. [9] N.P.H. Moller, K.B. Moller, R. Lammers, A. Kharitonenkov, E. Hoppe, F.C. Wiberg, I. Sures, A. Ullrich, Selective downregulation of the insulin receptor signal by protein–tyrosine phosphatases alpha and epsilon, J. Biol. Chem. 270 (1995) 23126–23131. [10] R. Lammers, N.P.H. Moller, A. Ullrich, The transmembrane protein tyrosine phosphatase alpha dephosphorylates the insulin receptor in intact cells, FEBS Lett. 404 (1997) 37–40. [11] G. Daum, S. Regenass, J. Sap, J. Schlessinger, E.H. Fischer, Multiple forms of the human tyrosine phosphatase RPTP alpha. Isozymes and differences in glycosylation, J. Biol. Chem. 269 (1994) 10524–10528. [12] L.N. Cong, H. Chen, Y. Li, C.H. Lin, J. Sap, M.J. Quon, Overexpression of protein tyrosine phosphatase-alpha (PTP-alpha)
364
K. Kapp et al. / Biochemical and Biophysical Research Communications 311 (2003) 361–364
but not PTP-kappa inhibits translocation of GLUT4 in rat adipose cells, Biochem. Biophys. Res. Commun. 255 (1999) 200–207. [13] K.K. Jacob, J. Sap, F.M. Stanley, Receptor-like protein–tyrosine phosphatase alpha specifically inhibits insulin-increased prolactin gene expression, J. Biol. Chem. 273 (1998) 4800–4809. [14] D. Janjic, P. Maechler, N. Sekine, C. Bartley, A.S. Annen, C.B. Wollheim, Free radical modulation of insulin release in INS-1 cells exposed to alloxan, Biochem. Pharmacol. 57 (1999) 639–648. [15] R. Lammers, M.M. Lerch, A. Ullrich, The carboxyl-terminal tyrosine residue of protein–tyrosine phosphatase alpha mediates association with focal adhesion plaques, J. Biol. Chem. 275 (2000) 3391–3396.
[16] R. Lammers, N.P.H. Moller, A. Ullrich, Mutant forms of the protein–tyrosine phosphatase alpha show differential activities towards intracellular substrates, Biochem. Biophys. Res. Commun. 242 (1998) 32–38. [17] L.H. Philipson, Beta-cell ion channels: keys to endodermal excitability, Horm. Metab. Res. 31 (1999) 455–461. [18] W. Tsai, A.D. Morielli, T.G. Cachero, E.G. Peralta, Receptor protein tyrosine phosphatase alpha participates in the m1 muscarinic acetylcholine receptor-dependent regulation of Kv1.2 channel activity, EMBO J. 18 (1999) 109–118. [19] H. Gil-Henn, G. Volohonsky, A. Elson, Regulation of protein– tyrosine phosphatases alpha and epsilon by calpain-mediated proteolytic cleavage, J. Biol. Chem. 276 (2001) 31772–31779.
The protein tyrosine phosphatase a modifies insulin secretion in INS-1E cellsq Katja Kapp, Elisabeth Metzinger, Monika Kellerer,1 Hans-Ulrich H€ aring, * and Reiner Lammers Medical Clinic IV, Otfried-M€uller-Str. 10, 72076 T€ ubingen, Germany Received 17 September 2003
Abstract Increasing evidence indicates a role of insulin signalling for insulin secretion from the pancreatic b-cells. Therefore, regulators of insulin signalling, like protein tyrosine phosphatases, could also have an impact on insulin secretion. Here, we investigated a possible role of the negative regulator protein tyrosine phosphatase a (PTPa) for insulin secretion. RT-PCR analysis confirmed that both splice variants of the extracellular domain of PTPa that vary by an insert of 9 amino acids are expressed in human islets and insulinoma cells (INS-1E, RIN1046-38). Overexpression of the wild type PTPa splice variant containing the 9 amino acids reduced insulin secretion, as did a mutant form unable to bind Grb2 (Tyr798Phe). By contrast, overexpression of a phosphatase inactive mutant improved insulin secretion. These data reveal a functional relevance of PTPa for insulin secretion. Ó 2003 Elsevier Inc. All rights reserved. Keywords: Insulin secretion; Protein tyrosine phosphatase; PTPa; Splice variant
Insulin secretion from pancreatic b-cells is achieved by diverse secretagogues with glucose being the major physiological stimulus. Although basic mechanisms of insulin secretion are understood, not all signalling molecules involved are yet known or fully characterized. Several lines of evidence indicate that insulin signalling via the insulin receptor plays a role for insulin secretion. The pancreatic knockout of the insulin receptor (IR) [1] or IRS-1 [2] leads to reduced insulin secretion. Protein tyrosine kinase inhibitors were shown to decrease insulin secretion [3] while other data reveal a stimulation of insulin secretion [4,5]. Vanadate, an inhibitor of protein tyrosine phosphatases (PTPs), can also modulate insulin secretion [6]. Since PTPs are modulators of the insulin signalling cascade, these data point to a role of PTPs in insulin secretion. Up to now, for two different members q
Abbreviations: PTPa, protein tyrosine phosphatase a; IR, insulin receptor; PTP, protein tyrosine phosphatase. * Corresponding author. Fax: +49-7071-29-5974. E-mail address: [email protected] (R. Lammers). 1 Present address: Marienhospital Stuttgart, B€ oheimstr. 37, 70199 Stuttgart, Germany. 0006-291X/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2003.10.011
of the PTP family a role in insulin secretion has been described. The overexpression of PTPr is linked to an impaired glucose induced insulin secretion in Goto– Kakizaki rats [7]. By contrast, the genomic deletion of the enzymatically inactive PTP IA-2, which is localized in the secretory granules, impairs insulin secretion [8]. In this study, we wanted to investigate the effect of PTPa on insulin secretion. PTPa can directly dephosphorylate the insulin receptor (IR) and is, therefore, a negative regulator of the insulin signal [9,10]. Human PTPa consists of a short extracellular, a transmembrane, and a cytoplasmic domain with two phosphatase domains. There are two isoforms due to an alternatively spliced, 27 bp containing exon occurring in the extracellular domain. Previous data on the presence of these isoforms revealed that both forms are tissue-specifically expressed and that they occur in mice and rats [11]. PTPa is involved in the translocation of glucose transporter 4 in rat adipose cells [12] and specifically inhibits insulin-increased prolactin gene expression [13], but a function in insulin secretion has yet not been observed. Here we show that overexpression of PTPa has an effect on insulin secretion.
362
K. Kapp et al. / Biochemical and Biophysical Research Communications 311 (2003) 361–364
Materials and methods Cell culture. INS-1E cells [14] were cultured in RPMI 1640 medium supplemented with 10% fetal calf serum, 10 mM Hepes (pH 7.4), 2 mM glutamine, 1 mM Na-pyruvate, and 50 lM b-mercaptoethanol. For stable expression, cells were infected with retroviruses generated from the retroviral expression vector pLXSN as described [15]. Geneticin selection was started 48 h after infection and resistant cells were pooled. RIN 1046-38 cells were cultured in medium 199 with Earle’s Salts, supplemented with 2 mM glutamine, 10% fetal calf serum and glucose was adjusted to 11.1 mM. RNA extraction and RT-PCR. Total RNA from human islets or cells was isolated using the RNAeasy Kit (Quiagen). For reverse transcription, first strand cDNA synthesis kit was used (Roche). PCR was done as suggested by the manufacturer (Eppendorf) with the primers for endogenous human PTPa (forward 50 -AGCAAGCACC AATTCTATAGGC, reverse 50 GTTGGATAAGCGGAAAGAAT TG), for endogenous rat PTPa (forward 50 -CTGATAACCAGT TCACGGATGC, reverse 50 -TGGCCAGAAGTGGTACACTTTG), and for the transfected human PTPa (forward 50 -GATGGAGG GTCCTCGCTGGGTCAGACACTGGAACAGT ATGAG and reverse 50 -CTTTCCACACCCTAACT GACAC [pLXSN-30 ]), 4–5 ll of the reverse transcribed mRNA or 200–250 ng of total RNA, and 2.5 U Taq-Polymerase. A fifth of each PCR was separated on a 2.5% agarose gel and stained with ethidium bromide. Insulin secretion assay. INS-1E cells were seeded 2 days before use at a density of 1 105 /cm2 . To determine insulin secretion, cells were washed twice with modified Krebs–Ringer bicarbonate buffer (25 mM Hepes, pH 7.4; 125 mM NaCl, 4.74 mM KCl, 1 mM CaCl2 , 1.2 mM KH2 PO4 , 1.2 mM MgSO4 , 5 mM NaHCO3 , 0.1% BSA, and 2.8 mM D (+)glucose) and incubated for 2 h at 37 °C for adaptation, 1 h at 37 °C (basal secretion), and 1 h at 37 °C (stimulated secretion) in Krebs–Ringer bicarbonate buffer. Supernatants were centrifuged at 250g for 5 min and stored at )20 °C until used. Insulin content was measured using a rat insulin RIA kit (Linco Research). Statistical analysis. Data were analyzed by one-way analysis of variance (ANOVA), followed by Student’s t tests for unpaired groups. The statistical software package JMP (SAS Institute, Cary, NC) was used.
Results INS-1E cells are a derivative of the INS-1 cell line and at low passage number have the unique property to secrete insulin in a glucose responsive way [14]. We therefore used this cellular model system to investigate a possible role of PTPa for insulin secretion. PTPa occurs with two splicing variants, which differ in size by 27 bp and are referred to as long or short isoform, respectively. First, we investigated if both forms are present in human islets and in the rat pancreatic b-cell lines, RIN 1046-38 and INS-1E. Total RNA was prepared and RTPCR was performed. The PCR covered the alternatively spliced exon and yielded DNA fragments of the expected size (276 and 303 bp, respectively; Fig. 1A). They indicate that most of the PTPa is present in primary islets and both cell lines as the short form. As control, a PCR with RNA did not result in the identification of any product. Since PTPa is expressed in pancreatic cells, we investigated the effect of PTPa overexpression on insulin
Fig. 1. Expression of PTPa in human islets or insulinoma-derived cell lines. RNA was isolated, reverse transcribed as indicated, subjected to PCR, and analyzed on an agarose gel. (A) Expression of endogenous PTPa (+) and control without reverse transcription ()). (B) Overexpression of PTPa isoforms in transfected INS-1E cell pools (+) and control without reverse transcription ()). DNA marker: 50 base pair ladder; CS, inactive PTPa (Cys 442, 732 to Ser); WT, native PTPa; and YF, Tyr 798 to Phe.
secretion. INS-1E cells were infected with retroviruses encoding the long form of wild type PTPa, a mutant form unable to bind Grb2 (Tyr 798 to Phe; [16]) or a phosphatase inactive mutant (Cys 442 and 732 to Ser; [16]). The Tyr 798 to Phe mutant was included since in previous experiments this mutant showed a different localization than the wild type phosphatase, which could affect its substrate specificity [15]. Following G418 selection, positive cells were pooled and directly used for insulin secretion to maintain a low passage number required for an optimal glucose response of the cells. To prove that PTPa was expressed from the transgene, RNA from pools of infected cells was prepared and used for RT-PCR. To this end, a reverse primer complementary to the 30 UTR of the transcripts ensured the detection of transgene-derived transcripts. Fig. 1B showed the expected PCR products of about 200 bp. As control, a PCR was performed directly with RNA, which resulted only in background signals. The amplification products were confirmed by sequencing. Employing the established cell pools, we investigated the function of the different PTPa forms by measuring insulin secretion upon glucose stimulation (Fig. 2). In cells overexpressing the wild type phosphatase or the tyrosine mutant insulin secretion was significantly reduced upon glucose stimulation when compared with
K. Kapp et al. / Biochemical and Biophysical Research Communications 311 (2003) 361–364
363
the wild type PTPa. Beyond that, it was shown that PTPa dephosphorylates Kv2.1 upon overexpression in 293 cells [19], adding another possible substrate in pancreatic b-cells. A closer investigation of b-cells may now help us to reveal the differential activity of the PTPa mutants and to elucidate the role of this new regulator of secretion.
Acknowledgments Fig. 2. Glucose stimulated secretion of insulin. Basal and glucose stimulated insulin secretion was determined in parental and in PTPa overexpressing INS-1E cells. The cells were counted in parallel to insulin secretion. Abbreviations are as in Fig. 1. Data are shown as means SEM from quadruplicate samples, with p values indicated as *p < 0:01 and **p < 0:05.
We thank P. Maechler and C.B. Wollheim for providing us with INS-1E cells and we are grateful to the group of M. Stumvoll for help with the statistical analysis. This work was supported by a grant from the DFG to R.L.
References parental cells (P values as indicated in the figure). By contrast, overexpression of the phosphatase inactive mutant resulted in higher amounts of secreted insulin in comparison to the control cells (P < 0:01; n ¼ 4). We conclude that PTPa is involved in the regulation of glucose-dependent insulin secretion in b-cells.
Discussion Insulin signalling is important for insulin secretion of the pancreatic b-cells. Therefore, PTPs as mediators of the insulin signalling could also play a role for insulin secretion. For some PTPs, a functional relevance for this process has been demonstrated [7,8]. Several studies have suggested that PTPa negatively regulates the insulin signal, likely by dephosphorylation of the IR [9,12,13]. We therefore investigated a possible role of PTPa for secretion of insulin. In the present study, we showed that both splice variants of PTPa occur in human islets as well as in rat insulinoma cells and that an overexpression of various forms of PTPa changed insulin secretion in response to glucose. Overexpression of a phosphatase inactive PTPa led to higher insulin secretion while the phosphatase active forms reduced secretion. Since PTPa is a PTP that directly dephosphorylates the IR, the reduced secretion upon overexpression may be mediated by an impaired signal from the IR. Alternatively, a PTPa substrate different from the IR could as well be important for secretion. Possible candidate substrates are the voltage-dependent (delayed rectifier) Kþ channels (Kv channels) that are involved in the glucose response of b-cells. In general, dephosphorylation leads to activation of these channels and thereby to cessation of insulin secretion [17]. Kv1.2 has been shown to be a substrate of PTPa and is activated upon dephosphorylation [18], which would fit to the results for
[1] R.N. Kulkarni, J.C. Bruning, J.N. Winnay, C. Postic, M.A. Magnuson, C.R. Kahn, 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 (1999) 329–339. [2] R.N. Kulkarni, J.N. Winnay, M. Daniels, J.C. Bruning, S.N. Flier, D. Hanahan, C.R. Kahn, Altered function of insulin receptor substrate-1-deficient mouse islets and cultured beta-cell lines, J. Clin. Invest. 104 (1999) R69–R75. [3] S.J. Persaud, T.E. Harris, C.J. Burns, P.M. Jones, Tyrosine kinases play a permissive role in glucose-induced insulin secretion from adult rat islets, J. Mol. Endocrinol. 22 (1999) 19–28. [4] R.L. Sorenson, T.C. Brelje, C. Roth, Effect of tyrosine kinase inhibitors on islets of Langerhans: evidence for tyrosine kinases in the regulation of insulin secretion, Endocrinology 134 (1994) 1975–1978. [5] E.J. Verspohl, B. Tollkuhn, H. Kloss, Role of tyrosine kinase in insulin release in an insulin secreting cell line (INS-1), Cell. Signal. 7 (1995) 505–512. [6] S. Gogg, J. Chen, S. Efendic, U. Smith, C. Ostenson, Effects of phosphotyrosine phosphatase inhibition on insulin secretion and intracellular signaling events in rat pancreatic islets, Biochem. Biophys. Res. Commun. 280 (2001) 1161–1168. [7] C.G. Ostenson, A.C. Sandberg-Nordqvist, J. Chen, M. Hallbrink, D. Rotin, U. Langel, S. Efendic, Overexpression of protein– tyrosine phosphatase PTP sigma is linked to impaired glucoseinduced insulin secretion in hereditary diabetic Goto–Kakizaki rats, Biochem. Biophys. Res. Commun. 291 (2002) 945–950. [8] K. Saeki, M. Zhu, A. Kubosaki, J. Xie, M.S. Lan, A.L. Notkins, Targeted disruption of the protein tyrosine phosphatase-like molecule IA-2 results in alterations in glucose tolerance tests and insulin secretion, Diabetes 51 (2002) 1842–1850. [9] N.P.H. Moller, K.B. Moller, R. Lammers, A. Kharitonenkov, E. Hoppe, F.C. Wiberg, I. Sures, A. Ullrich, Selective downregulation of the insulin receptor signal by protein–tyrosine phosphatases alpha and epsilon, J. Biol. Chem. 270 (1995) 23126–23131. [10] R. Lammers, N.P.H. Moller, A. Ullrich, The transmembrane protein tyrosine phosphatase alpha dephosphorylates the insulin receptor in intact cells, FEBS Lett. 404 (1997) 37–40. [11] G. Daum, S. Regenass, J. Sap, J. Schlessinger, E.H. Fischer, Multiple forms of the human tyrosine phosphatase RPTP alpha. Isozymes and differences in glycosylation, J. Biol. Chem. 269 (1994) 10524–10528. [12] L.N. Cong, H. Chen, Y. Li, C.H. Lin, J. Sap, M.J. Quon, Overexpression of protein tyrosine phosphatase-alpha (PTP-alpha)
364
K. Kapp et al. / Biochemical and Biophysical Research Communications 311 (2003) 361–364
but not PTP-kappa inhibits translocation of GLUT4 in rat adipose cells, Biochem. Biophys. Res. Commun. 255 (1999) 200–207. [13] K.K. Jacob, J. Sap, F.M. Stanley, Receptor-like protein–tyrosine phosphatase alpha specifically inhibits insulin-increased prolactin gene expression, J. Biol. Chem. 273 (1998) 4800–4809. [14] D. Janjic, P. Maechler, N. Sekine, C. Bartley, A.S. Annen, C.B. Wollheim, Free radical modulation of insulin release in INS-1 cells exposed to alloxan, Biochem. Pharmacol. 57 (1999) 639–648. [15] R. Lammers, M.M. Lerch, A. Ullrich, The carboxyl-terminal tyrosine residue of protein–tyrosine phosphatase alpha mediates association with focal adhesion plaques, J. Biol. Chem. 275 (2000) 3391–3396.
[16] R. Lammers, N.P.H. Moller, A. Ullrich, Mutant forms of the protein–tyrosine phosphatase alpha show differential activities towards intracellular substrates, Biochem. Biophys. Res. Commun. 242 (1998) 32–38. [17] L.H. Philipson, Beta-cell ion channels: keys to endodermal excitability, Horm. Metab. Res. 31 (1999) 455–461. [18] W. Tsai, A.D. Morielli, T.G. Cachero, E.G. Peralta, Receptor protein tyrosine phosphatase alpha participates in the m1 muscarinic acetylcholine receptor-dependent regulation of Kv1.2 channel activity, EMBO J. 18 (1999) 109–118. [19] H. Gil-Henn, G. Volohonsky, A. Elson, Regulation of protein– tyrosine phosphatases alpha and epsilon by calpain-mediated proteolytic cleavage, J. Biol. Chem. 276 (2001) 31772–31779.