We recently described novel regulatory roles for protein histidine phosphorylation of key islet proteins (e.g., nucleoside diphosphate kinase and succinyl thiokinase) in insulin secretion from the islet β-cell (Kowluru A. Diabetologia 44: 89-94, 2001; Kowluru A, Tannous M, and Chen HQ. Arch Biochem Biophys 398: 160-169, 2002). In this context, we also characterized a novel, ATP- and GTP-sensitive protein histidine kinase in isolated β-cells that catalyzed the histidine phosphorylation of islet (endogenous) proteins as well as exogenously added histone 4, and we implicated this kinase in the activation of islet endogenous G proteins (Kowluru A. Biochem Pharmacol 63: 2091-2100, 2002). In the present study, we describe abnormalities in ATP- or GTP-mediated histidine phosphorylation of nucleoside diphosphate kinase in islets derived from the Goto-Kakizaki (GK) rat, a model for non-insulin-dependent diabetes. Furthermore, we provide evidence for a marked reduction in the activities of ATP- or GTP-sensitive histidine kinases in GK rat islets. On the basis of these observations, we propose that alterations in protein histidine phosphorylation could contribute toward insulin-secretory abnormalities demonstrable in the diabetic islet.
- pancreatic islet
- nucleoside diphosphate kinase
- GTP-binding proteins
- insulin secretion
in most cells, transduction of extracellular signals involves ligand binding to a receptor, often followed by the activation of one (or more) GTP-binding proteins (G proteins) and their effector systems (3, 4, 6). The pancreatic β-cell is unusual in that glucose, the major physiological stimulus for insulin secretion, lacks an extracellular receptor. Instead, events consequent to glucose metabolism promote insulin secretion via the generation of and/or altered distribution of diffusible second messengers, such as ions, cyclic nucleotides, and biologically active lipids (17, 26, 27, 34, 36). Changes in calcium concentration not only initiate insulin secretion but also regulate various enzymes, such as protein kinases, phosphodiesterases, adenylate cyclases, and phospholipases, thereby facilitating insulin secretion (17, 26, 27, 34, 36). In addition to calcium-dependent protein kinases, several other kinases, including calmodulin-, cyclic nucleotide-, and phospholipid-dependent protein kinases, tyrosine kinases, and mitogen-activated protein kinases, have been described in β-cells (10). The majority of these kinases mediate phosphorylation of endogenous β-cell proteins by use of ATP as phosphoryl donor.
Several studies have appeared in recent years on the localization, characterization, and regulation of protein histidine kinases in multiple cell types (2, 11, 28, 30, 39, 42). Wei and Matthews (42) first reported a filter paper-based protein kinase assay that selectively quantitated acid-labile and alkali-stable phosphorylation reactions. Utilizing this assay, they were able to purify and characterize a protein histidine kinase from Saccharomyces cerevisiae by use of histone 4 as the substrate and demonstrated that the yeast histidine kinase activity selectively phosphorylated the histidine residue at the 75th position, but not the 18th residues on histone 4 (42). Along these lines, we (18, 21) recently identified several proteins that underwent histidine phosphorylation in islet β-cells. Some of these include the nucleoside diphosphate (NDP) kinase, the β-subunit of trimeric G proteins (22), and the succinyl thiokinase (STK; Ref. 13). We reported (18, 21) that the NDP kinase undergoes autophosphorylation at a histidine residue, which in turn is transferred to the NDPs to yield nucleotide triphosphates (NTPs). It has also been proposed that NDP kinase mediates the transphosphorylation of GDP bound to the G proteins (inactive conformation) to yield their respective GTP-bound (active) conformation (12, 18, 37). We also reported that, unlike the NDP kinase, the β-subunit of trimeric G proteins does not undergo autophosphorylation but requires a membrane-associated factor or kinase for its phosphorylation at a histidine residue (14, 22). Furthermore, recent studies from our laboratory (13, 23) have localized in the mitochondrial fraction from isolated β-cells a novel isoform of NDP kinase, which appears to form a complex with the mitochondrial STK, possibly mediating its functional regulation. Using histone 4 as the substrate, we (14) recently characterized a novel protein histidine kinase activity in normal rat islets, human islets, and clonal β-cell preparations. This novel histidine kinase, with an apparent molecular mass of 60-70 kDa, utilized either ATP or GTP as phosphoryl donors. We also observed that mastoparan, a global stimulator of G proteins and insulin secretion from isolated rat islets and clonal β-cells, but not mastoparan-17, its inactive analog, stimulated this histidine kinase activity in isolated β-cells (14). Together, these data suggested a critical role for protein histidine phosphorylation in the β-cell stimulus-secretion coupling.
Because we showed earlier that islet endogenous G proteins relevant to insulin secretion are activated by histidine phosphorylation, we undertook the present study to quantitate the degree of protein histidine phosphorylation in islets derived from the Goto-Kakizaki (GK) rat, a model for non-isulin-dependent diabetes mellitus (NIDDM) (29) to determine whether insulin-secretory abnormalities demonstrable in these cells relate to alterations in protein histidine phosphorylation. We provide evidence for a significant defect in the histidine phosphorylation of NDP kinase as well the histone 4-phosphorylating histidine kinase in islets derived from the GK rat. On the basis of these observations, we propose that alterations in protein histidine phosphorylation could contribute toward insulin-secretory abnormalities demonstrable in the diabetic islet.
MATERIALS AND METHODS
Materials. [γ-32P]GTP (30 Ci/mmol) and [γ-32P]ATP (600 Ci/mmol) were purchased from NEN-DuPont. NDPs, NTPs, and histone 4 were purchased from Boehringer Mannheim. Nytran filter papers were purchased from Schleicher and Schuell. All other reagents used in the present study were of the highest purity available.
Isolation of islets from control and diabetic rats. GK rats were generously provided by Dr. Robert Farese (Tampa, FL). They were provided to our laboratory (at the Veterans Affairs Medical Center, Madison, WI) at the age of 8 wk and were then housed at our animal care facility. Offspring were allowed to reach the ages of 8-14 wk, when they were studied in parallel with age- and sex-matched Wistar rats (Harlan, Indianapolis, IN). As indicated in Ref. 29, islets were isolated by collagenase digestion from control (Wistar) and diabetic (GK) rats. In brief, pancreases were inflated with collagenase solution (1 and 2 U/ml for Wistar and GK rats, respectively) in Hanks' balanced salt solution supplemented with 1% fetal bovine serum. After digestion with collagenase, islets were washed twice, passed through mesh (92-μm pore size), and then purified on a Ficoll gradient. Islets were then hand picked twice under stereomicroscopic control to exclude extraislet debris. Also, as indicated in our earlier studies (29), we made certain to isolate islets with grossly normal size and appearance to minimize any extrinsic artifacts of the isolation procedure and abnormal pancreatic morphology. Islets from control Wistar and diabetic GK rats were homogenized in a buffer consisting of 230 mM mannitol, 70 mM sucrose, and 5 mM HEPES buffer, pH 7.4, containing 1 mM EGTA, 1 mM DTT, and 2.5 μg/ml each of leupeptin and pepstatin.
Quantitation of histidine phosphorylation of NDP kinase in lysates from control and diabetic rat islets. The phosphorylation reaction was carried out (100 μl total volume), in a buffer consisting of 50 mM Tris · HCl, pH 7.4, 2 mM DTT, islet protein (30 μg), and [γ-32P]ATP or [γ-32P]GTP (1 μCi/tube) at 37°C for 2 min. It was terminated by the addition of Laemmli stop solution. Because the phosphohistidine of NDP kinase is heat sensitive (18, 21), the samples were incubated with the sample buffer at room temperature for 30 min before SDS-PAGE. Moreover, it has been shown (18, 22) that typical fixation conditions used for SDS gels (methanol, acetic acid, and water medium) ablates phosphate labeling from [32P]phosphohistidine. Therefore, to verify that our gel fixation conditions did not underestimate NDP kinase phosphoenzyme formation, gels were fixed in some experiments in 50 mM sodium phosphate buffer, pH 8.0, containing 18.5% formaldehyde for 1.5 h, as described by Wei and Matthews (42); these results were compared with those obtained using gels fixed in methanol (40%) and acetic acid (7%) medium. We observed no significant differences in the residual labeling on NDP kinase between these two conditions (additional data not shown). Therefore, after separation of proteins by SDS-PAGE, gels were fixed in methanol-acetic acid medium for 1.5 h and dried at room temperature, as described by us earlier (14, 18, 22). Labeled proteins were identified by autoradiography. Molecular weights of labeled proteins were determined using prestained molecular weight standards. Labeling intensity of the proteins was quantitated by scanning individual lanes with a Zeineh Video Laser Densitometer (Biomed Instruments, Fullerton, CA) that was interfaced to an IBM computer equipped with software to calculate the individual peak areas.
Quantitation of histidine kinase activity in lysates from control and diabetic rat islets. This quantitation was done according to the method we recently described (14). In brief, the assay mixture (100 μl total volume) consisted of 0.6 mg/ml histone 4, 0.2 mM [γ-32P]ATP or [γ-32P]GTP, 15 mM magnesium chloride, 50 mM Tris · HCl, pH 7.5, and islet lysates, as indicated (see legend to Fig. 1). The reaction was carried out at 37°C for 5 min and was terminated by incubating the mixture in 0.5 N NaOH at 60°C for 30 min. The base-treated reaction mixture was transferred directly to the Nytran filter papers previously soaked overnight in 1 mM ATP, pH 9.0, at room temperature and air dried. The filter papers were then transferred to a beaker with 200 ml of 10 mM sodium pyrophosphate at pH 9.0 and gently stirred at room temperature for 30 min to remove unreacted ATP or GTP. The filters were air-dried under an infrared lamp, and the radioactivity associated with the filters was quantitated by scintillation spectrometry. The degree of histidine phosphorylation of endogenous proteins (no added histone 4) or exogenously added histone 4 was expressed as picomoles of 32P incorporated per minute per milligram of islet protein (14).
Other methods. The protein concentration in the samples was assayed according to the method of Bradford, using serum albumin as standard as we described (13, 15, 18, 20-22). The statistical significance of the differences between control and diabetic animal groups was determined by Student's t-test. P values <0.05 were considered significant.
Using [γ-32P]ATP or [γ-32P]GTP as the phosphoryl donor, we previously demonstrated (18) that a protein in the molecular mass region of 20-22 kDa was phosphorylated in normal rat islets, human islets, and insulin-secreting β-cells. On the basis of its sensitivity to pH, temperature, and histidine-modifying agents, we identified the phosphorylated amino acid residue as histidine (18). Subsequent immunoprecipitation studies conclusively identified this protein as the subunit of NDP kinase (18). Mechanistically, several earlier studies in multiple cell types, including our own, in the isolated β-cell (12, 18, 21) have shown that NDP kinase undergoes autophosphorylation transiently at a histidine residue and that phosphate, in turn, is transferred to NDPs to yield their corresponding NTPs. In the present series of studies, using the stability criteria described above (and in materials and methods), we determined the degree of histidine phosphorylation of NDP kinase in lysates derived from control Wistar and diabetic GK rat islets.
Figure 1 represents an autoradiogram demonstrating the autophosphorylation of NDP kinase in control (Wistar) and diabetic (GK) rat islets carried out using either [γ-32P]ATP (A) or [γ-32P]GTP (B) as phosphoryl donors. These data indicate a substantial reduction in NDP kinase phosphorylation in islets derived from the GK rat in the presence of either of the substrates. Densitometric quantitation of the labeled bands (Fig. 1C) indicated a >50% reduction in the autophosphorylation of NDP kinase. These data are compatible with our recent findings of a significant attenuation (-40%) in the catalytic activity of NDP kinase in islets derived from the GK rats compared with their control counterparts (29).
In addition to NDP kinase, we reported (14) localization of a novel histidine kinase activity in isolated rat islets and clonal β-cells and implicated this enzyme in the activation of islet G proteins and insulin secretion. Herein, we measured such a histidine kinase activity, using a Nitran filter paper assay, in homogenates derived from the control and GK rat islets and in the absence or presence of histone 4 as the substrate. Data in Fig. 2A indicate a marked reduction (>90%) in the degree of endogenous protein histidine phosphorylation (4.62 pmol 32P incorporated·min-1·mg protein-1 in control rat islets vs. 0.40 pmol 32P incorporated·min-1· mg protein-1 in GK rat islets) when [γ-32P]ATP was used as the phosphoryl donor (see materials and methods for additional details). Furthermore, under similar experimental conditions, we observed a substantial degree of inhibition (-64%) of histidine phosphorylation of exogenously added histone 4 in the homogenates of diabetic GK rat islets compared with those from the control Wistar rats (7 pmol 32P incorporated·min-1·mg protein-1 in Wistar rat islets vs. 2.5 pmol 32P incorporated·min-1·mg protein-1 in GK rat islets). Although the [γ-32P]GTP-dependent phosphorylation of endogenous proteins was reduced modestly, but significantly, the histone 4 phosphorylation was markedly attenuated (-64%) in GK rat islets compared with the control Wistar rat islets. The corresponding values for degree of endogenous protein histidine phosphorylation were 1.12 pmol 32P incorporated·min-1·mg protein-1 in control rat islets vs. 0.8 pmol 32P incorporated·min-1·mg protein-1 in GK rat islets. However, a nearly identical degree of reduction was demonstrable in GTP-dependent phosphorylation of histone 4 in GK rat islets (i.e., 1.84 pmol 32P incorporated·min-1·mg protein-1 in control rat islets vs. 0.66 32P incorporated·min-1·mg protein-1) in GK rat islets. Together, these data (Figs. 1 and 2) suggest significant alterations in the protein histidine phosphorylation in the diabetic rat islets.
A growing body of evidence suggests novel regulatory roles for protein histidine phosphorylation in cellular regulation in multiple cell types (2, 14, 28, 42). One of the main objectives of the present study was to assess the functional status of protein histidine phosphorylation in islets derived from the GK rat, an animal model for NIDDM in humans. Our data indicated significant defects in the ATP- as well as GTP-sensitive phosphorylation of NDP kinase in islets derived from the GK rat. Furthermore, we observed a significant reduction in the ATP- and GTP-sensitive histidine kinase activity in GK rat islets, suggesting that the overall protein histidine phosphorylation is markedly reduced in the GK islet.
In the present studies, pancreatic islets were isolated from overtly diabetic, but nonobese, GK rats, a genetic model for NIDDM in humans and studied ex vivo compared with islets derived from age- and sex-matched control Wistar rats. However, we cannot exclude some contribution to the islet dysfunction by the antecedent, ambient diabetic milieu in vivo, since we have not attempted to restore normoglycemia in vivo or in cultured islets in vitro for prolonged periods at normal glucose concentration. Our previous data and the existing body of evidence (Ref. 29 and references therein) argue against the possibility that some degree of glucotoxicity or lipotoxicity might contribute to the attenuated histidine phosphorylation. Thus it seems likely that the defect in protein histidine phosphorylation that we report herein is intrinsic to the GK islet.
The possibility that a decreased β-cell mass or decreased β-cell protein could contribute to the observed defects in histidine phosphorylation needs to be addressed as well in the context of our current findings. Earlier studies have reported significant reductions in the DNA content or β-cell mass in GK islets (31, 32), although others have failed to see any differences in β-cell mass between control and GK rat islets (8, 35). Recent data from our laboratory have indicated no major differences in insulin content, protein content, and the stereomicroscopic morphology between the Wistar and GK rat islets (29). Furthermore, the degree of reduction in ATP- or GTP-sensitive NDP kinase phosphorylation and histidine kinase activity were significantly lower in GK rat islets even after normalization of these values to equal protein (present studies). Together, these points argue against a selective reduction in β-cell mass as the (sole) islet lesion in the GK rat.
Emerging evidence indicates that several proteins of carbohydrate metabolism undergo phosphorylation at histidine residues; these include glucose-6-phosphatase (7), ATP-citrate lyase (41), and aldolase (40). Recent studies from our laboratory (18, 21) have identified at least three proteins that underwent histidine phosphorylation. These include NDP kinase, the β-subunit of trimeric G proteins (22), and the mitochondrial STK (13). In our present studies, we observed that the ATP- and GTP-sensitive autophosphorylation of the NDP kinase activity was significantly reduced in GK rat islets, data compatible with marked reductions in NDP kinase catalytic activity from these cells (29). In this context, we have reported localization of at least three forms of NDP kinase in isolated β-cells (23). The nm-23 H1 (NDP kinase A) was found to be localized predominantly in the soluble fraction, in contrast to nm23-H2 (NDP kinase B), which was found to be localized in the soluble as well as the membranous fractions. Furthermore, a third form of NDP kinase (i.e., nm23-H4) was found to be localized exclusively in the mitochondrial fraction. Because in the present studies we utilized total cell lysates instead of individual subcellular fractions, it is not possible to identify conclusively which of these isoforms is defective in the diabetic islet.
In the present study, we also observed significant reduction in ATP- and GTP-sensitive histidine kinase activity in islets derived from the GK rat. We have implicated this enzyme in the functional activation of G proteins in the islet β-cell (14). We observed that, in addition to histone 4-phosphorylating activity, histidine phosphorylation of endogenous proteins was also decreased. This could potentially represent histidine phosphorylation of several islet endogenous proteins, including NDP kinase, the mitochondrial STK, and the β-subunit of trimeric G proteins that we have identified as proteins undergoing phosphorylation at histidine residues (see above). On the basis of the data presented in Fig. 1, one of the potential proteins might represent the NDP kinase. Our unpublished evidence along these lines indicates no major differences in the histidine phosphorylation of the β-subunit. Additional studies are needed to determine the identity of those proteins whose histidine phosphorylation is significantly impaired in the GK islet.
In addition to the functional regulation of several enzymes, such as STK, ATP-citrate lyase, glucose-6-phosphatase, aldolase, etc. (7, 40, 41), recent evidence implicates protein histidine phosphorylation in the activation of G proteins. For example, several earlier studies, including our own (Ref. 14 and references therein) demonstrated direct stimulatory effects of mastoparan, a tetradecapeptide from wasp venom, on G protein function and insulin secretion in normal rat islets, human islets, and clonal β-cells. Recently, we (14) also demonstrated that mastoparan, but not its inactive analog mastoparan-17, markedly stimulated the histone 4-phosphorylating histidine kinase activity in lysates derived from normal rat islets and clonal β-cells. These data indicate a possible regulatory role for this novel histidine kinase in the activation of G proteins, presumably at the level of their conversion from a GDP-bound, inactive form of the G proteins to its GTP-bound, active form. In this context, independent studies from our laboratory in isolated β-cells (22) and by Wieland et al. (43) in HL-60 cells have demonstrated that the β-subunit of trimeric G proteins undergoes phosphorylation at a histidine residue and that phosphate, in turn, is transferred to the GDP-bound α-subunit (inactive conformation) to yield its GTP-bound, active conformation. We have also provided evidence (14, 22) to suggest that, unlike NDP kinase, which undergoes autophosphorylation at a histidine residue (Fig. 1), the β-subunit requires the intermediacy of a histidine kinase to catalyze its phosphorylation at a histidine residue. These observations of novel activation mechanisms of trimeric G proteins via subunit phosphorylation of β-subunits were confirmed recently by several studies (5, 9, 33).
Together, it appears that protein histidine phosphorylation plays major regulatory roles in metabolic regulation in multiple cell types, including the islet β-cell. Therefore, it is likely that a marked attenuation in the histidine kinase activity demonstrable in diabetic rat islets (present study) could result in the reduction in the activation of specific G proteins, which we (1, 13, 15, 16, 19, 20) and others (24, 25, 28) have shown to be essential for insulin secretion. Our current data also provide further evidence to suggest that abnormalities in insulin secretion demonstrable in GK rat islets may be due, in part, to decreased histidine phosphorylation of specific islet proteins; compatible with this formulation are our data in GK rats that indicate a possible defect in the activation by NDP kinase of a mastoparan-sensitive G protein step in the exocytotic secretion of insulin in GK rats (29). Additional studies are needed to precisely identify the phosphoprotein substrates for the histidine kinase, whose phosphorylation may be relevant to insulin secretion. Studies are also needed to identify the candidate G proteins (trimeric as well as monomeric) whose activation is under the fine control of NDP kinase and histidine kinase activity that we proposed to be essential for the exocytotic secretion of insulin.
This research work was supported by grants from the Department of Veterans Affairs (Merit Review and Research Enhancement Awards), National Institute of Diabetes and Digestive and Kidney Diseases (DK-56005), and the American Diabetes Association. A. Kowluru is the recipient of a Research Career Scientist award from the Department of Veterans Affairs.
I thank Mary Rabaglia for expert help in islet isolation and Lisa Modrick in histidine kinase assays.
A significant portion of this work was carried out during the author's stay at the Veterans Affairs Medical Center in Madison, WI, and the University of Wisconsin School of Medicine-Madison.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2003 by American Physiological Society