Insulin mediates glucose-stimulated phosphorylation of PHAS-I by pancreatic beta cells. An insulin-receptor mechanism for autoregulation of protein synthesis by translation.

Although glucose regulates the biosynthesis of a variety of beta cell proteins at the level of translation, the mechanism responsible for this effect is unknown. We demonstrate that incubation of pancreatic islets with elevated glucose levels results in rapid and concentration-dependent phosphorylation of PHAS-I, an inhibitor of mRNA cap-binding protein, eukaryotic initiation factor (eIF)-4E. Our initial approach was to determine if this effect is mediated by the metabolism of glucose and activation of islet cell protein kinases, or whether insulin secreted from the beta cell stimulates phosphorylation of PHAS-I via an insulin-receptor mechanism as described for insulin-sensitive cells. In support of the latter mechanism, inhibitors of islet cell protein kinases A and C exert no effect on glucose-stimulated phosphorylation of PHAS-I, whereas the phosphatidylinositol 3-kinase inhibitor, wortmannin, the immunosuppressant, rapamycin, and theophylline, a phosphodiesterase inhibitor, promote marked dephosphorylation of PHAS-I. In addition, exogenous insulin and endogenous insulin secreted by the beta cell line, betaTC6-F7, increase phosphorylation of PHAS-I, suggesting that beta cells of the islet, in part, mediate this effect. Studies with beta cell lines and islets indicate that amino acids are required for glucose or exogenous insulin to stimulate the phosphorylation of PHAS-I, and amino acids alone dose-dependently stimulate the phosphorylation of PHAS-I, which is further enhanced by insulin. Furthermore, rapamycin inhibits by approximately 62% the increase in total protein synthesis stimulated by high glucose concentrations. These results indicate that glucose stimulates PHAS-I phosphorylation via insulin interacting with its own receptor on the beta cell which may serve as an important mechanism for autoregulation of protein synthesis by translation.

The metabolism of glucose regulates a variety of physiological processes by beta cells of the islets of Langerhans. These processes include insulin biosynthesis and secretion, beta cell replication, and the synthesis of a number of beta cell proteins. Glucose exerts a specific stimulatory effect on insulin gene expression over time intervals of hours (1,2), whereas the biosynthesis of insulin is significantly controlled within minutes at the level of protein translation (3)(4)(5). Although the enhancement of insulin synthesis by glucose at the translational level occurs rapidly and does not require synthesis of new mRNA, limited information is available on processes that regulate translation. More recent studies indicate that insulin is only one of a number of beta cell proteins whose synthesis is regulated by glucose at the level of translation (6). Glucose has been shown to increase the biosynthesis of a secretory granule membrane protein, SGM 110 (7), and also chromogranin A (8) to a similar extent as insulin, suggesting that translational control may extend to a variety of other beta cell proteins.
The initiation phase of mRNA translation is generally ratelimiting for protein synthesis. Initiation is mediated in part by the eIF 1 -4F complex, which is composed of three subunits, eIF-4␥, eIF-4A, and eIF-4E (9). eIF-4␥ is a large subunit (M r 220,000) that binds eIF-4A (M r 45,000) and eIF-4E (M r 25,000). eIF-4A is an ATP-dependent helicase, and eIF-4E is the mRNA cap-binding protein. Thus, eIF-4F is involved in both the recognition of the capped mRNA and melting of secondary structure in the 5Ј-untranslated region. eIF-4E is the least abundant of the eIF-4F subunits, and it is generally believed that the amount of eIF-4E is limiting for translation initiation.
The availability of eIF-4E is regulated by PHAS-I, a heatand acid-stable eIF-4E binding protein first identified in rat adipocytes (10,11). The nonphosphorylated form of PHAS-I binds tightly to eIF-4E, and prevents eIF-4E from binding to eIF-4␥. When phosphorylated in the appropriate site(s), PHAS-I dissociates from eIF-4E, allowing the factor to participate in translation initiation.
PHAS-I is phosphorylated both in vitro and in vivo by a variety of protein kinases. Casein kinase II, protein kinase C, and mitogen-activated protein kinase have been reported to phosphorylate recombinant PHAS-I (12). Insulin-like growth factor-1 and platelet-derived growth factor in smooth muscle cells (13) and insulin in 3T3 L1 adipocytes (14) increase PHAS-I and p70 s6k phosphorylation by a rapamycin-sensitive pathway. This mitogen-induced phosphorylation is also sensitive to the phosphatidylinositol 3-kinase inhibitor, wortmannin (15). Recent studies on PHAS-I have implicated a signaling pathway involving the mammalian target of rapamycin (mTOR). Thus, mTOR appears to be a site by which the rapamycin⅐FKBP12 complex mediates the dephosphorylation of PHAS-I (15)(16)(17)(18). The phosphorylation of PHAS-I and p70 s6k appears to be regulated by mTOR in parallel (18). These studies suggest that PHAS-I has an important role in regulating translation in a variety of cells by signal transduction pathways that involve protein phosphorylation.
Stimulus secretion coupling of insulin secretion from pancreatic beta cells is a complex process involving the transport and metabolism of glucose with the generation of metabolic intermediates including ATP. It is proposed that increases in ATP are associated with the closure of ATP-dependent K ϩ channels, cell depolarization, and Ca 2ϩ entry through voltage-dependent Ca 2ϩ channels (19). There is also strong evidence that protein kinases play a key role in modulating the insulin secretory process by beta cells. The major serine/threonine kinases that have been studied in insulin secretion include Ca 2ϩ and calmodulin protein kinase (20), Ca 2ϩ and phospholipid protein kinase C (21,22), cAMP protein kinase A (23), and more recently mitogen-activated protein kinases (24,25).
Recent evidence also indicates that glucose-stimulated insulin secretion activates a beta cell surface insulin receptor kinase and the intracellular effector substrate, insulin receptor substrate-1, from ␤TC3 cells (26). These results and others (27) have suggested that insulin released from the beta cell of the islet may bind to its own cell-surface receptor and serve as an autocrine mechanism for the regulation of beta cell function. We have therefore examined these aspects of stimulus secretion coupling of insulin secretion by beta cells on PHAS-I activation as reflected by its phosphorylation state, signaling pathways, and proposed functional role to regulate protein translation by the beta cell of the pancreatic islet.

EXPERIMENTAL PROCEDURES
Materials-Male Sprague-Dawley rats were purchased from Sasco (O'Fallon, MO) and Harlan Sprague-Dawley (Indianapolis, IN). Collagenase type P was obtained from Boehringer Mannheim. CMRL-1066 tissue culture medium, penicillin, streptomycin, Hanks' balanced salt solution, L-glutamine, MEM amino acids solution, and MEM nonessential amino acids solution were obtained from Life Technologies, Inc. Fetal bovine serum was from Hyclone (Logan, UT). Dulbecco's modified Eagle's medium (DMEM) was supplied by the Washington University Tissue Culture Support Center. Rapamycin and N-6,2Ј-o-dibutyryladenosine-3Ј:5Ј-cyclic monophosphate (dibutyryl cAMP) were from Biomol (Plymouth Meeting, PA). Ficoll, theophylline, and CPT-cAMP were from Sigma. Insulin was obtained from Lilly and ICN Biomedicals (Aurora, OH). Pro-mix L-[ 35 S] in vitro cell labeling mix was from Amersham Corp. The primary antibody was generated in rabbit with recombinant His-tagged rat PHAS-I (11). The secondary antibody was peroxidase-conjugated donkey anti-rabbit IgG (Jackson Immunologicals). All other chemicals were from commercially available sources.
Amino Acid Composition-Krebs-Ringer bicarbonate buffer (KRBB; 25 mM HEPES, 115 mM NaCl, 25 mM NaHCO 3 , 5 mM KCl, 2.5 mM CaCl 2 , and 1 mM MgCl 2 , pH 7.4, containing 3 mM D-glucose, and 0.1% BSA) was supplemented with MEM amino acids solution, MEM nonessential amino acids solution, and L-glutamine. For these experiments, the 1 ϫ concentration of amino acids was defined as the following (in mM): L-arginine 0.73, L-cystine 0. Islet Isolation and Culture-Islets were isolated from male Sprague-Dawley rats (200 -250 g) by collagenase digestion as described previously (28). Briefly, pancreases were inflated with Hanks' balanced salt solution, and the tissue was isolated, minced, and digested with 7 mg of collagenase/pancreas for 7 min at 39°C. Islets were separated on a Ficoll step density gradient and then selected with a stereomicroscope to exclude any contaminating tissues. Islets were cultured overnight in an atmosphere of 95% air, 5% CO 2 in "complete" CMRL-1066 tissue culture medium (cCMRL) containing 5.5 mM glucose, 2 mM L-glutamine, 10% (v/v) heat-inactivated fetal bovine serum, 100 units/ml penicillin, and 100 g/ml streptomycin. Where "CMRL" is stated, this culture medium does not contain fetal bovine serum.
Pancreatic Beta Cell Lines-␤TC3 and ␤TC6-F7 cells were kindly provided by Norman Fleischer (Albert Einstein College of Medicine, Bronx, NY), and RINm5F cells were obtained from the Washington University Tissue Culture Support Center. RINm5F cells were grown in cCMRL medium, and ␤TC3 as well as ␤TC6-F7 cells were grown in DMEM containing 25 mM glucose and supplemented with 15% heatinactivated horse serum and 2.5% fetal bovine serum as described by Knaack et al. (29). Cells were cultured in Petri dishes (60 ϫ 15 mm) at a concentration of 1 ϫ 10 6 cells/3 ml. RINm5F cells were used after 1 day. ␤TC3 and ␤TC6-F7 cells were used after 5 days of culture, and medium was replaced once during that time. ␤TC3 and ␤TC6-F7 cells were pretreated in DMEM containing 5 mM glucose for 24 h before experiments were begun.
Incubation of Islets and Cells and Preparation of Extracts for PHAS-I Determination-Two separate protocols were used to prepare tissue for PHAS-I immunoblotting. In the first experimental protocol, islets (200 -300/1 ml or 1500 cells/10 ml) were incubated in CMRL or DMEM at 37°C in an atmosphere of 95% air, 5% CO 2 as described in the figure legends. The islets were then centrifuged (Beckman Microfuge B) for 10 s, the supernatant was removed for insulin RIA, and the islets were resuspended in 100 l of homogenization buffer (50 mM ␤-glycerophosphate, pH 7.3, 100 mM NaF, 5 mM EDTA, 2 mM EGTA, 0.1 mM Na 3 VO 4 , 1 mM benzamidine, 10 g/ml leupeptin, 0.2 mM phenylmethylsulfonyl fluoride). The islets were homogenized and centrifuged at 10,000 ϫ g for 20 min, and the supernatants were heated at 100°C for 10 min. Heated supernatants were then centrifuged at 10,000 ϫ g for 30 min, Laemmli sample buffer was added to the supernatants, and 50 l were subjected to SDS-PAGE. An immunoblot was performed using rabbit anti-PHAS-I and donkey, anti-rabbit IgG, as primary and secondary antibodies, respectively. Detection was performed using ECL reagents from Amersham Corp.
To reduce islet tissue requirements, a second protocol for PHAS-I immunoblotting was used. Following experimental treatments as described in the figure legends, islets were washed with phosphate-buffered saline and solubilized in 30 l of Laemmli sample buffer, heated at 100°C for 5 min, and centrifuged at 10,000 ϫ g for 15 min to remove insoluble materials. The supernatants were processed for SDS-PAGE and Western blotting as described above.
␤TC3 and ␤TC6 -7F cells were glucose-and amino acid-depleted in KRBB for 2 h. Fresh KRBB was added containing test reagents as indicated in the figure legends for 30 -120 min. Supernatants were saved for insulin RIA. The cells were washed three times in phosphatebuffered saline, lysed in 300 l of Laemmli sample buffer, boiled for 5 min, and centrifuged at 10,000 ϫ g for 10 min. The supernatants were withdrawn, and 30 l total protein from each reaction were used for PHAS-I immunoblotting analysis.
PHAS-I⅐eIF-4E Binding Assays-For experiments to determine the binding of PHAS-I to eIF-4E, islet extracts were incubated with an affinity resin of the cap homolog 7-methyl-guanosine triphosphate, proteins were eluted and subjected to SDS-PAGE and immunoblotting (11).
Measurement of Total Protein Synthesis-Groups of 100 islets were glucose and serum depleted in 1 ml of CMRL (0.1% BSA, 90% methionine-free) for 90 min. Rapamycin (25, 50, or 100 nM) was added to the medium for an additional 30 min. Medium was removed and replaced with 0.1 ml of 90% methionine-free CMRL ϩ 3 or 20 mM glucose Ϯ rapamycin for 1 h with 80 Ci of Pro-mix L-[ 35 S] in vitro labeling mix. Islets were washed twice with ice-cold phosphate-buffered saline (0.1% BSA, 0.02% NaN 3 ), and protein was precipitated with 1 ml of 10% trichloroacetic acid at 4°C for 30 min. The samples were washed once with 10% trichloroacetic acid and solubilized with 50 l of Protosol. The scintillation mixture was added and the radioactivity was quantitated. Identification of Mediators in Glucose-induced Phosphorylation of PHAS-I-To determine if glucose-induced phosphorylation of PHAS-I is mediated by the metabolism of glucose and activation of major islet cell protein kinases, inhibitors of protein kinases A and C were initially evaluated. The selective protein kinase C inhibitor chelerythrine (10 M) and phorbol 12-myristate 13-acetate (100 nM), an agonist of protein kinase C, exerted no effects on glucose-induced phosphorylation of PHAS-I. The cAMP analogues, CPT-cAMP (0.2 mM) and dibutyryl-cAMP (1 mM) as well as staurosporine (100 nM), a potent and nonselective inhibitor of protein kinase A and protein kinase C were also ineffective in modulating glucose-induced phosphorylation of PHAS-I, suggesting that protein kinase A and most likely mitogen-activating protein kinase were not involved (30) (data not shown).

PHAS-I Phosphorylation by
Recent studies have indicated a role for the insulin receptor, phosphatidylinositol 3-kinase and mTOR in the signaling pathway involved in phosphorylation of PHAS-I by insulin-sensitive cells (15)(16)(17)(18). To determine if insulin secreted from the beta cell stimulates phosphorylation of PHAS-I via an insulin-receptor mechanism, we evaluated the phosphatidylinositol 3-kinase inhibitor, wortmannin, the immunosuppressant, rapamycin, and theophylline, a phosphodiesterase inhibitor that also blocks p70 s6k (31,32), in this signaling pathway by pancreatic islets. As shown in Fig. 2, incubation of islets in CMRL in the presence of 20 mM glucose (lane 2) resulted in increased phosphorylation of PHAS-I␥ in comparison to 3 mM glucose (lane 1). Co-incubation of islets with wortmannin, rapamycin, and theophylline resulted in the dephosphorylation of PHAS-I to a level similar to that observed at basal glucose levels. PHAS-I␣, which is not normally visible at basal glucose concentrations, becomes apparent with these inhibitors. These findings are consistent with insulin-mediated phosphorylation of PHAS-I in insulin-sensitive cells.
Experiments were next performed to determine if insulin secretion from beta cells of the islet may be responsible for enhanced phosphorylation of PHAS-I. As illustrated in Fig. 3A  (lane 2), exposure of islets incubated in CMRL to 20 mM glucose for 30 min resulted in enhanced phosphorylation of PHAS-I␥ in comparison to low glucose levels of 1 mM (lane 1). Furthermore, exposure of islets to exogenous insulin at 2, 20, and 200 nM (lanes 3, 4, and 5) under low glucose levels of 1 mM markedly stimulated in a dose-dependent manner phosphorylation of PHAS-I, similar to that observed with 20 mM glucose. These results suggested that glucose-stimulated phosphorylation of PHAS-I is mediated by insulin secreted into the incubation medium.
To further support this possibility, islets were incubated at 24°C, a condition that blocks almost completely insulin exocytosis from beta cells. Under these conditions of reduced temperature, ion fluxes are minimally affected (33,34)  Amino Acid Dependence of PHAS-I Phosphorylation-Our previous studies were performed with islets incubated in CMRL culture medium. The ability of glucose to stimulate phosphorylation of PHAS-I was also evaluated in KRBB. Unexpectedly, it was found that glucose did not stimulate the phosphorylation of PHAS-I in KRBB in the complete absence of amino acids, even though islets secrete insulin normally in KRBB under these conditions. As shown in Fig. 4 experimental design, islets were exposed to 3 or 20 mM glucose in DMEM for 30 min, and extracts were prepared as described under "Experimental Procedures." Islet extracts were then passed over an affinity resin of the cap homolog 7-methylguanosine triphosphate cross-linked to eIF-4E. Highly phosphorylated PHAS-I␥ and to a lesser extent PHAS-I (␤) do not bind to cross-linked eIF-4E and elute into the "unbound" fraction. Conversely, unphosphorylated PHAS-I (␣) (and to some degree PHAS-I␤) bind to cross-linked eIF-4E and are designated as "bound." As shown in Fig. 7, exposure of islets to 20 mM glucose resulted in a shift of PHAS-I␤ to PHAS-I␥ (lane 2) compared with 3 mM glucose (lane 1) in the islet extract. Theophylline (5 mM), regardless of glucose concentration, caused a dramatic shift in phosphorylation of PHAS-I␥ to ␤ and ␣ (lanes 3 and 4).
PHAS-I that is not bound to eIF-4E, i.e. phosphorylated PHAS-I, is indicated in lanes 5-8. In lane 6, 20 mM glucose induced a small shift in the amount of PHAS-I␤ to ␥ compared with 3 mM glucose (lane 5). These results as anticipated mirrored those in the islet extract (lanes 1 and 2) and correspond to more free eIF-4E available for translation initiation. PHAS-I that is bound to eIF-4E is shown in lanes 9 -12. We anticipated that at 3 mM glucose more eIF-4E would be bound to PHAS-I than at 20 mM glucose (lanes 9 and 10), which was not detected. In a repeat experiment, increasing glucose from 3 to 20 mM produced the expected results in the bound fraction and demonstrated a greater shift in the extract blot from PHAS-I␤ to ␥. In the second part of Fig. 7, theophylline (lanes 7, 8, 11, and 12) caused a dramatic increase in PHAS-I binding to eIF-4E, regardless of the glucose concentration. Almost all of PHAS-I present was bound to eIF-4E due to theophylline (lanes 11 and 12), which was mirrored by the lack of unbound PHAS-I ( lanes  7 and 8). These results suggest the following: 1) glucose stimulation of islets increases phosphorylation of PHAS-I from the ␤ to the ␥ form and is associated with more eIF-4E available to initiate protein translation, and 2) theophylline results in dephosphorylation of PHAS-I, which remains bound to eIF-4E and would be predicted to inhibit initiation of protein translation.
Effects of Rapamycin on Total Protein Synthesis-To evaluate the role of the phosphorylation state of PHAS-I on beta cell function, rapamycin was evaluated for its effects on glucose stimulated protein synthesis by islets. Fig. 8 demonstrates that islets incubated in CMRL in the presence of 20 mM glucose for 60 min increases by 50% the amount of [ 35 S]methionine incorporated into total protein compared with basal conditions of 3 mM glucose. Under these same conditions, rapamycin at 25, 50, and 100 nM significantly inhibited by ϳ62% glucose stimulated incorporation of [ 35 S]methionine into total protein. Rapamycin at 20 and 100 nM had no effect on glucose-stimulated insulin secretion determined over a 1-h incubation (data not shown). These results suggest that the biosynthesis of a number of proteins may be regulated at the level of translation by the beta cell in an autocrine loop through glucose-induced insulin secretion. DISCUSSION This study demonstrates that glucose stimulates in a timeand concentration-dependent manner the rapid phosphorylation of PHAS-I, an inhibitor of mRNA cap-binding protein, eIF-4E, by isolated pancreatic islets. Our initial approach was to determine if this effect was mediated by the metabolism of glucose and subsequent activation of islet cell protein kinases, or whether insulin secreted from the beta cell may stimulate phosphorylation of PHAS-I in a manner analogous to that described for insulin-sensitive cells. Evidence in support of the latter mechanism was that conditions that block insulin exocytosis from beta cells prevented glucose-stimulated phosphorylation of PHAS-I, and exogenous insulin mimicked the effects of elevated glucose concentrations.
During this investigation, in vitro incubations were routinely performed in CMRL and DMEM tissue culture media. To more easily manipulate the components of this tissue culture system, parallel studies were also performed in KRBB in the absence of amino acids, co-factors, and vitamins. Insulin secretion studies are routinely performed in KRBB, and islets secrete insulin normally under these conditions. Unexpectedly, neither elevated glucose concentrations that stimulated insulin secretion from beta cells nor exogenous insulin altered the basal patterns of PHAS-I␥, ␤, or ␣ when islets were incubated in KRBB in the absence of amino acids. However, addition of a normal complement of amino acids to KRBB restored the ability of both endogenous and exogenous insulin to stimulate the phosphorylation of PHAS-I by islets (data not shown). Additional studies with the insulinoma-derived cell line, RINm5F, also indicated that this normal complement of amino acids alone induced in a concentration-dependent manner phosphorylation of PHAS-I, which was further enhanced by insulin. These results suggested that amino acids were by themselves sufficient and also required in the presence of insulin to mediate the phosphorylation of PHAS-I. We did not attempt to determine which amino acids are responsible for these effects. Studies are in progress to define the mechanism whereby amino acids exert their effects on PHAS-I phosphorylation. This requirement for amino acids may explain a recent study by Gilligan et al. (36), who reported that glucose did not alter the phosphorylation state of PHAS-I by islets. These negative results are most likely explained by the fact that these studies were performed in a buffered salt solution in the absence of amino acids. Recently, Patti et al. (37) have also reported that amino acids stimulate the phosphorylation of PHAS-I and at submaximal concentrations synergize with insulin by FAO hepatocytes.
Our findings with the beta cell line, RINm5F, also suggested that the beta cell of the islet is responsible for glucose induced phosphorylation of PHAS-I. Thus, amino acids alone stimulated phosphorylation of PHAS-I and this effect was further enhanced with exogenous insulin by this beta cell line. In addition, conditions which stimulate insulin secretion, i.e. glucose ϩ carbachol, from the beta cell line, ␤TC6-F7, also increased phosphorylation of PHAS-I analogous to that observed with isolated islets. Similar results were also observed with ␤TC3 cells in relating endogenous insulin secretion with the phosphorylation of PHAS-I (data not shown).
PHAS-I is phosphorylated both in vitro and in vivo by a variety of protein kinases. Previous studies in smooth muscle cells and 3T3-L1 adipocytes (13, 14) indicated a role for p70 s6k , a protein kinase that phosphorylates ribosomal protein S6 in mammalian cells (9), in the phosphorylation of PHAS-I. Recent evidence indicates that mTOR regulates the phosphorylation of both p70 s6k and PHAS-I in parallel and that rapamycin is acting on mTOR to prevent phosphorylation of both kinases (17,18). Similar dephosphorylation of p70 s6k and PHAS-I has also been reported with the phosphatidylinositol 3-kinase inhibitor, wortmannin (15,38), and the phosphodiesterase inhibitor, SQ20006 (31,32). Wortmannin and rapamycin promoted significant dephosphorylation of PHAS-I by islets exposed to elevated glucose concentrations. These same inhibitors also induced dephosphorylation of PHAS-I by RINm5F cells exposed to amino acids, further indicating that beta cells of the islet mediate this effect in a similar manner as insulin sensitive cells.
The ability of cAMP to regulate PHAS-I phosphorylation was also evaluated in pancreatic islets. Studies in 3T3-L1 adipocytes and aortic smooth muscle cells suggested that increasing cAMP levels may promote dephosphorylation of PHAS-I (13,14,39). Phosphodiesterase inhibitors such as theophylline and isobutylmethylxanthine are commonly used to modulate cAMP levels in islets. These phosphodiesterase inhibitors significantly enhance glucose-stimulated proinsulin biosynthesis and insulin secretion from beta cells by increasing cAMP levels. Our studies indicated that theophylline results in marked dephosphorylation of PHAS-I, but unexpectedly, CPT-cAMP and dibutyryl-cAMP exerted no effects on the phosphorylation state of PHAS-I (data not shown). These paradoxical results were explained by recent studies indicating that methylxanthines including theophylline block activation of p70 s6k , similar to rapamycin, by a mechanism independent of cAMP and cGMP production (31,32). Our studies demonstrated that rapamycin mimics the ability of theophylline to block glucose-stimulated phosphorylation of PHAS-I. Although theophylline increases cAMP levels and greatly enhances glucose-induced insulin secretion by islets, its effect on mTOR presumably causes significant dephosphorylation of PHAS-I in a similar fashion as rapamycin. In addition, cAMP analogues, CPT-cAMP and dibutyryl-cAMP, had no effect on the phosphorylation level of PHAS-I by glucose-stimulated islets. Although insulin secretion increased further in the presence of these cAMP analogues as anticipated, glucose-induced phosphorylation of PHAS-I␥ was already maximal, and this additional insulin secretion caused no further increase in phosphorylation. Overall, our findings suggest that cAMP independent of insulin secretion does not appear to affect glucose-stimulated phosphorylation of PHAS-I by islets.
Previous studies on the regulation of the biosynthesis of insulin secretory granule proteins by glucose have indicated that a subset of proteins exist whose synthesis is stimulated to a similar extent as insulin (7). These proteins include granule matrix and membrane constituents and possibly other proteins not associated with beta cell granules. Although rapamycin did not affect the acute phase of glucose-stimulated insulin secretion, it inhibited both glucose-stimulated phosphorylation of PHAS-I and total protein synthesis. Even though exogenous insulin stimulated phosphorylation of PHAS-I at basal glucose levels, these conditions did not result in enhanced protein synthesis, suggesting that both elevated glucose concentrations necessary for increased metabolism and the release of insulin are required to mediate increased protein translation by beta cells (data not shown).
Our findings suggest that glucose-stimulated phosphorylation of PHAS-I by beta cells of the islet is mediated via insulin interacting with its own insulin receptor, leading to the phosphorylation of PHAS-I. This raises the possibility that insulin receptor activation may target downstream insulin signaling proteins such as insulin receptor substrate-1 and -2 and may FIG. 8. Rapamycin inhibits glucose-induced protein synthesis in rat islets. Groups of 100 islets were serum-, glucose-, and methionine-depleted in 1 ml of CMRL (0.1% BSA, 90% methionine-free) for 2 h at 37°C. During the final 30 min, rapamycin (25, 50, or 100 nM) was added. The medium was replaced with 0.1 ml of CMRL containing 3 or 20 mM glucose Ϯ rapamycin, and 80 Ci of Pro-mix L-35 S in vitro labeling mix for 1 h. The supernatant was removed, the islets were washed twice with 1 ml of phosphate-buffered saline (0.1% BSA, 0.02% NaN 3 ), and protein was precipitated with 1 ml of 10% trichloroacetic acid at 4°C for 30 min. The samples were washed once with 1 ml of 10% trichloroacetic acid and solubilized with Protosol, and scintillation counting was performed. Results are the percent increase in 35 S incorporation from 3 to 20 mM glucose and are means Ϯ S.E. from six separate experiments in triplicate for each condition. Statistically significant inhibition of [ 35 S]methionine incorporation, compared with the untreated control, was determined at 25, 50, and 100 nM rapamycin (p Ͻ 0.01, 0.04, 0.01, respectively) by unpaired t tests.
provide an important mechanism for the autoregulation of beta cell function in a manner analogous to insulin-sensitive cells. Functional parameters that may be regulated by this autocrine loop in beta cells include protein synthesis, gene expression, DNA synthesis, and cell proliferation. In support of this concept, recent studies by Rothenberg et al. (26) have demonstrated that glucose-induced insulin secretion from ␤TC3 cells activates the beta cell surface receptor tyrosine kinase and insulin receptor substrate-1, which then associates with the 85-kDa ␣-subunit of phosphatidylinositol 3-kinase. Additional studies by Herbeck et al. (27) have also demonstrated expression of insulin receptor mRNA and insulin receptor substrate-1 in single beta cells obtained from isolated rat islets. Insulin receptors have also been identified in the beta cell line, RINm5F (40). Further studies are required to identify specific proteins that may be involved in translational regulation of protein synthesis and characterize additional functional aspects of beta cell physiology that may be affected. Overall, our results support and extend this novel concept at a functional level that insulin secreted by beta cells activates a beta cell insulin receptor-signaling pathway, resulting in autoregulation of protein synthesis. The identification of common insulin signaling pathways in insulin target cells and beta cells may provide new insights with regard to possible defects in the development of diabetes mellitus.