Glucose-stimulated DNA Synthesis through Mammalian Target of Rapamycin (mTOR) Is Regulated by KATP Channels

The aim of this study was to define metabolic signaling pathways that mediate DNA synthesis and cell cycle progression in adult rodent islets to devise strategies to enhance survival, growth, and proliferation. Since previous studies indicated that glucose-stimulated activation of mammalian target of rapamycin (mTOR) leads to [3H]thymidine incorporation and that mTOR activation is mediated, in part, through the KATP channel and changes in cytosolic Ca2+, we determined whether glyburide, an inhibitor of KATP channels that stimulates Ca2+ influx, modulates [3H]thymidine incorporation. Glyburide (10–100 nm) at basal glucose stimulated [3H]thymidine incorporation to the same magnitude as elevated glucose and further enhanced the ability of elevated glucose to increase [3H]thymidine incorporation. Diazoxide (250 μm), an activator of KATP channels, paradoxically potentiated glucose-stimulated [3H]thymidine incorporation 2–4-fold above elevated glucose alone. Cell cycle analysis demonstrated that chronic exposure of islets to basal glucose resulted in a typical cell cycle progression pattern that is consistent with a low level of proliferation. In contrast, chronic exposure to elevated glucose or glyburide resulted in progression from G0/G1 to an accumulation in S phase and a reduction in G2/M phase. Rapamycin (100 nm) resulted in an ∼62% reduction of S phase accumulation. The enhanced [3H]thymidine incorporation with chronic elevated glucose or glyburide therefore appears to be associated with S phase accumulation. Since diazoxide significantly enhanced [3H]thymidine incorporation without altering S phase accumulation under chronic elevated glucose, this increase in DNA synthesis also appears to be primarily related to an arrest in S phase and not cell proliferation.

At basal glucose (3 mM), glyburide-induced closure of K ATP channels caused a partial phosphorylation of S6K1. Since mTOR is a regulator of DNA synthesis, these results suggested that glyburide might enhance DNA synthesis at basal glucose. In addition, diazoxide-induced activation of the K ATP channel partially inhibited S6K1 phosphorylation by glucose. These results suggested that diazoxide might attenuate glucose-stimulated DNA synthesis. Guiot et al. (11) reported in 1994 that glibenclamide (glyburide) stimulated ␤-cell replication and increased ␤-cell mass in normal young mice. However, the cellular mechanisms responsible for the effects of glyburide were not addressed. More recent findings have indicated that the L-type Ca 2ϩ channel ␣1D subunit is required for proper ␤-cell development in the postnatal pancreas (12). In ␣1D gene knock-out mice, ␤-cell proliferation was decreased, suggesting that a defect in Ca 2ϩ influx may be partly responsible for a reduction in ␤-cell mass. In addition, recent findings indicated that S6K1 requires an initial Ca 2ϩdependent priming event for activation (13,14).
Diazoxide has been used extensively in vitro to characterize the regulation of stimulus-secretion coupling mechanisms of ␤-cells that are independent of plasma membrane K ATP channels. Although diazoxide has provided important mechanistic insights into the role of plasma membrane K ATP channels in ␤-cells, it is also reported to exert significant effects on mitochondrial energetics (15,16). A direct effect of K ATP channel openers including diazoxide on mitochondrial function has been shown to be critical in the preservation of heart function against injury (17). The cardioprotective effects of diazoxide include decreased Ca 2ϩ uptake due to depolarization of the mitochondrial membrane potential and an activated Ca 2ϩ release that maintains and/or preserves mitochondrial function under conditions of increased metabolic stress. Grimmsmann and Rustenbeck (18) reported that diazoxide at concentrations used routinely to open plasma membrane K ATP channels in ␤-cells also exerted direct effects on ␤-cell mitochondrial function, resulting in a decrease in mitochondrial membrane potential, an efflux of Ca 2ϩ from the mitochondria, and a decrease in ATP concentration. These findings demonstrated that diazoxide exerts multiple effects on ␤-cells that are not limited to plasma membrane K ATP channels. Studies by Grill and co-workers (19 -21) have further demonstrated that chronic glucose exposure of pancreatic islets for 48 h under both in vitro and in vivo conditions produced nonresponsiveness to a subsequent glucose challenge. However, if the chronic exposure of islets to glucose was performed in the presence of diazoxide, the ability of ␤-cells to respond to glucose was maintained. A conclusion drawn from these studies was that the protective effect of diazoxide against glucose-induced desensitization of ␤-cells was not due to a lasting effect of diazoxide on plasma membrane K ATP channels and may involve other targets possibly including the ␤-cell mitochondria.
Our data indicate that [ 3 H]thymidine incorporation in adult rodent islets chronically exposed to glucose or glyburide is largely mediated through mTOR via the K ATP channel and changes in intracellular Ca 2ϩ . Based on [ 3 H]thymidine incorporation and cell cycle analysis, we determined that chronic exposure of adult islets to basal glucose (3 mM) results in DNA synthesis and cell cycle progression consistent with a low level of cell proliferation. In contrast, chronic exposure to elevated glucose or glyburide results in cell cycle progression from G 0 /G 1 to an accumulation of newly synthesized DNA in S phase and a reduction in G 2 /M. Rapamycin (100 nM) resulted in an ϳ62% reduction of S phase accumulation. The aim of this study was to define metabolic signaling pathways that mediate DNA synthesis and cell cycle progression in adult rodent islets to devise strategies to enhance survival, growth, and proliferation. Sprague-Dawley rats were purchased from Harlan  Sprague-Dawley (Indianapolis, IN). Collagenase type XI, Ficoll type 400-DL, diazoxide, glyburide, RNase A, and propidium iodide were obtained from Sigma. CMRL-1066 tissue culture media was from Invitrogen. Defined fetal bovine serum (FBS) was from Hyclone (Logan, UT). Penicillin, streptomycin, Hanks' balanced salt solution, and L-glutamine were obtained from the Washington University Tissue Culture Support Center. Dispase and Versene were from Roche Applied Science. Rapamycin was from Biomol (Plymouth Meeting, PA). [methyl-3 H]Thymidine-aqueous, 2.0 Ci/mmol, 1.0 mCi/ml was from PerkinElmer Life Sciences. Nifedipine was from Calbiochem. The primary antibody for S6K1 was from R&D Systems (Minneapolis, MN). The secondary antibody, peroxidase-conjugated donkey anti-rabbit IgG, was obtained from Jackson ImmunoResearch Laboratories (West Grove, PA). All other chemicals were from commercially available sources.

Materials-Male
Islet Isolation-Islets were isolated from male Sprague-Dawley rats (200 -250 g) by collagenase digestion as described previously (22). Briefly, pancreases were inflated with Hanks' balanced salt solution, and the tissue was isolated, minced, and digested with 20 mg of collagenase/ pancreas for 6 min at 39°C. Islets were separated on a Ficoll step density gradient and then placed in CMRL-1066 culture medium supplemented with 10% FBS, 2 mM L-glutamine, 5.6 mM glucose, 100 units/ml penicillin, 100 g/ml streptomycin (cCMRL) and selected with a stereomicroscope to exclude any contaminating tissues.
[ 3 H]Thymidine Incorporation-Islets (100) were counted into Falcon Petri dishes (35 ϫ 10 mm) and were cultured for 4 days in 1 ml of CMRL (3 or 20 mM glucose, 10% FBS, 2 mM L-glutamine, 100 units/ml penicillin, 100 g/ml streptomycin) containing treatment conditions as indicated in the figure legends at 37°C, under 95% air/5% CO 2 . The culture media were changed everyday or after 2 days as indicated. During the final 24 h of the 96-h incubation, 10 Ci of [ 3 H]thymidine was added to each dish. [ 3 H]Thymidine incorporation was determined by trichloroacetic acid extraction and scintillation counting (23).
Western Blotting-Islets were treated as described above for [ 3 H]thymidine incorporation. Following incubation, islets were processed for SDS-PAGE electrophoresis and Western blotting for S6K1.
Flow Cytometry-Following the 4-day treatment as described above, islets were subjected to cell cycle analysis by flow cytometry. Islets were dispersed into single cells by treatment with Versene, followed by Dispase in Ca 2ϩ -and Mg 2ϩ -free Hanks' solution at 31°C. The dispersed cells were washed once with PBS ϩ 1% FBS and pelleted by low speed centrifugation. Cells were resuspended in 500 l of PBS. Five ml of 100% ethanol was added slowly while vortexing to avoid clumping of cells. Fixed cells were stored at Ϫ20°C until use. Cells were collected by centrifugation, washed once with PBS ϩ 1% FBS, and resuspended in PBS ϩ 1% FBS containing RNase A (250 g/ml) and propidium iodide (10 g/ml). PI-stained cells were analyzed for their DNA content using a FACSCALIBUR instrument and data analyzed with CellQuest software (Immunocytometry Systems).
Expression of Data and Statistics-Data are presented as mean Ϯ S.E. Statistically significant differences between groups were analyzed using unpaired one-tailed t tests, where p Ͻ 0.05 was considered significant.

RESULTS
In our experimental design, isolated primary rat islets were chronically exposed to either basal (3 mM) or elevated glucose (20 mM) for 4 days to quantitate the incorporation of [ 3 H]thymidine that was added to the culture media on day 3 as an index of DNA synthesis. Time course studies established that a 4-day incubation period was optimal for [ 3 H]thymidine incorporation under these conditions. Separate studies also demonstrated, based on S6K1 phosphorylation, that mTOR remained fully activated in response to elevated glucose or glyburide and was not down-regulated over this time period. This same 4-day treatment period was used to measure cell cycle progression by flow cytometry. Both methods measure DNA synthesis, although in the case of [ 3 H]thymidine incorporation only the last 24 h of incubation are measured, whereas with the flow cytometry method, total DNA is intercalated with PI.
Glyburide Dose Response-Our initial approach was to determine whether glyburide, an inhibitor of K ATP channels that enhances Ca 2ϩ influx, modulates DNA synthesis through mTOR. As shown in Fig. 1, chronic exposure of rat islets to elevated glucose for 4 days resulted in an increase in DNA synthesis as assessed by [ 3 H]thymidine incorporation compared with islets exposed to basal glucose (3 mM, lane 2 versus lane 1). Glyburide dose-dependently (1-100 nM) increased DNA synthesis at basal glucose (lanes 4 and 5). These concentrations of glyburide are within the physiological range typically used to modulate K ATP channels.
Role for Ca 2ϩ Influx-Studies were next performed to determine whether glyburide-mediated increases in DNA synthesis were due to cytosolic Ca 2ϩ derived from the extracellular media. In Fig. 2, an increase in glucose concentration enhanced DNA synthesis above basal glucose (lane 2 versus lane 1) and glyburide (10 nM) at basal glucose (lane 4) stimulated DNA synthesis to the same level attained by elevated glucose alone (lane 2). Nifedipine (10 M), an inhibitor of the L-type Ca 2ϩ channel, blocked glyburide-stimulated DNA synthesis (lane 5) to a level comparable with basal glucose (lane 1), and all of these effects were inhibited by rapamycin (25 nM) as shown in lane 6. The level of DNA synthesis in the presence of rapamycin (lane 6) was substantially lower than basal DNA synthesis (lane 1), supporting our previous findings that mTOR mediates both basal and nutrient stimulated DNA synthesis (7). These results support the conclusion that the ability of glyburide to mediate [ 3 H]thymidine incorporation is due, in part, to an increase in Ca 2ϩ influx through the nifedipine-sensitive L-type Ca 2ϩ channel.
These results also correlated with S6K1 phosphorylation based on gel shift mobility assays (upper inset) as an indicator of mTOR activation. Elevated glucose robustly enhanced S6K1 phosphorylation as indicated by the intense and more slowly migrating upper band (lane 2) compared with basal glucose (lane 1). An effective concentration of glyburide (10 nM) resulted in a partial activation of S6K1 phosphorylation at basal glucose (lane 4) as indicated by the doublet band in comparison to basal glucose (lane 1) that is sensitive to nifedipine (lane 5), and all of these effects are blocked by rapamycin (lane 6). These studies, however, suggest that the level of S6K1 activation is not linearly related with [ 3 H]thymidine incorporation, since different levels of S6K1 activation (lane 2 versus lane 4) resulted in comparable levels of DNA synthesis. We propose that S6K1 is an excellent indicator of mTOR activation but other mTOR-mediated signals in addition to S6K1 activation are important for the regulation of DNA synthesis.
Diazoxide Potentiates Glucose-stimulated DNA Synthesis-Our next approach was to determine whether diazoxide, an opener of K ATP channels, would decrease glucose-stimulated DNA synthesis due to its ability to attenuate Ca 2ϩ influx derived from the extracellular media. As shown in Fig. 3, diazoxide (250 M) paradoxically enhanced the ability of elevated glucose to stimulate DNA synthesis (lane 3 versus lane 2). This enhanced DNA synthesis also occurred in the absence of diazoxide treatment during the final 24 h of the total 96-h incubation period as shown in the washout protocol (lane 4). Rapamycin (25 nM) blocked the ability of diazoxide to significantly enhance glucose-stimulated DNA synthesis (lane 5). These unexpected results suggested that diazoxide might exert multiple effects to mediate the potentiation of glucosestimulated DNA synthesis in addition to the opening of plasma membrane K ATP channels. The observation that diazoxide did not have to be present for the total 4 day incubation period to potentiate glucosestimulated DNA synthesis was consistent with this conclusion, since the ability of diazoxide to activate plasma membrane K ATP channels is rapidly reversible following its removal.
Multiple Sites of Action for Diazoxide-Studies were next performed to determine whether the ability of diazoxide to attenuate Ca 2ϩ influx through L-type Ca 2ϩ channels is responsible for the significant potentiation of glucose-stimulated DNA synthesis. If this were the case, an inhibitor of the L-type Ca 2ϩ channel such as nifedipine should mimic  4 -6). These results suggest that diazoxide stimulates DNA synthesis by mechanisms other than solely attenuating Ca 2ϩ influx.
Evidence that the potentiating effect of diazoxide on glucose-stimulated DNA synthesis is associated with a nifedipine-sensitive component of Ca 2ϩ influx is shown in These results indicate that the significant enhancement of glucosestimulated DNA synthesis by diazoxide (lane 3) is dependent on a low level of nifedipine-sensitive Ca 2ϩ influx into ␤-cells. In contrast, chronic exposure of islets to elevated glucose alone as shown in Fig. 4 (lane 2) resulted in a smaller increase in DNA synthesis that is independent of Ca 2ϩ influx due to the lack of inhibition by nifedipine (lanes 4 -6).
The conclusion that the stimulation of Ca 2ϩ influx, under conditions in which primary rat islets are chronically exposed to elevated glucose, will enhance DNA synthesis is further supported by the findings shown in Fig. 6. In this experimental protocol, glyburide due to its ability to directly inhibit K ATP channels was used in combination with elevated glucose to enhance Ca 2ϩ influx. Chronic exposure of islets to glucose (20 mM) in combination with glyburide (10 nM) significantly potentiated DNA synthesis compared with an elevated glucose concentration alone (lane 3 versus lane 2) in a nifedipine-sensitive manner (lane 4).
Rapamycin-mediated Inhibition of Glucose-stimulated DNA Synthesis Is Irreversible in Vitro-An important question raised in these in vitro studies is whether the ability of rapamycin to inhibit basal and glucosestimulated DNA synthesis in primary rodent islets is reversible. This concern is also relevant to the current use of rapamycin as an immunosuppressant in islet transplantation protocols in type 1 diabetes (24). As shown in Fig. 7, the inhibitory effects of rapamycin (1 nM) present for 4 days on glucose-stimulated DNA synthesis (lane 3 versus lane 2) is nearly identical to the inhibition observed when rapamycin is present for only 1 day and absent during the remaining 3 days of the incubation period (lane 6). As shown in the inset, S6K1 phosphorylation, based on gel shift analysis, correlates well with the effects of rapamycin on DNA synthesis. These results indicate that inhibition of DNA synthesis by rapamycin (1 nM) is essentially irreversible during this 3-day exposure period under these in vitro conditions with primary adult islets. Furthermore, these findings suggest that the synthesis of new mTOR protein may be required to restore mTOR activity following the removal of the inhibitor, rapamycin.
Cell Cycle Analysis-To determine to what extent increases in [ 3 H]thymidine incorporation translate into changes in cell proliferation, cell cycle progression was analyzed by flow cytometry with PI-stained   islet cells. As shown in Fig. 8 and Table 1, under basal glucose conditions, quantitation of cell cycle progression indicates that 7.3 Ϯ 0.5% of gated cells were in G 2 /M phase, a result that is consistent with a low level of cell proliferation normally observed in adult islet cells. In contrast, chronic exposure to elevated glucose resulted in a marked accumulation of total gated cells in S phase (50.7 Ϯ 5.5%) and a reduction in G 2 /M to 2.5 Ϯ 0.3%. Furthermore, chronic exposure of islets to basal glucose in combination with glyburide (10 nM) also resulted in an accumulation in S phase (39.6 Ϯ 9.9%) and reduction in G 2 /M to 3.4 Ϯ 0.8%. In separate experiments, rapamycin at 100 nM maximally inhibited S phase accu-mulation by 61.9 Ϯ 2.6% (n ϭ 4), whereas rapamycin at 25 nM resulted in 32.7% (n ϭ 2) inhibition in response to chronic elevated glucose. Although diazoxide significantly enhanced [ 3 H]thymidine incorporation above elevated glucose alone, it did not alter S phase accumulation (data not shown).

DISCUSSION
We have attempted to define the pathways and mediators responsible for modulating glucose-stimulated DNA synthesis through mTOR and to determine the extent that this relates to cell cycle progression and proliferation in adult rodent islets. Since glucose, through its metabolism, activates mTOR continuously for at least 6 days, we utilized an in vitro model of chronic exposure of rat islets to glucose (20 mM) under conditions shown previously to enhance DNA synthesis 1.5-2-fold compared with basal glucose (3 mM) (7). Our studies have shown that appropriately regulating cytosolic Ca 2ϩ levels via the modulation of K ATP channels is critical for mTOR-mediated [ 3 H]thymidine incorporation.
In rodents, chronic exposure of islets to elevated glucose has been shown to increase hypertrophy, proliferation, and/or the neogenesis of ␤-cells. Steil et al. (25) have shown that glucose infusions in rats for 1-2 days resulted in an increase in ␤-cell mass that paralleled increased rates of ␤-cell proliferation. Paris et al. (26) also reported that chronic exposure of rodent islets to glucose in vivo for 48 h provoked a rapid compensatory increase in ␤-cell mass that was independent of plasma insulin levels. Lipsett et al. (27) reported that ␤-cell neogenesis contributed  Western blot. Islets were treated as described above. Following incubation, islets were processed for SDS-PAGE electrophoresis and Western blotting for S6K1. Blot is representative of n ϭ 2 experiments using 1 nM rapamycin and n ϭ 1 using 25 nM rapamycin.   FEBRUARY 10, 2006 • VOLUME 281 • NUMBER 6 significantly to glucose-stimulated expansion of ␤-cell mass during prolonged hyperglycemia in rats infused with elevated glucose. In in vitro models, the mitogenic effects produced by elevated glucose were shown to be long lasting in rat islets and more transitory in human islets (28). In addition, prolonged exposure of rodent islets to elevated glucose resulted in impaired insulin secretion and increased ␤-cell apoptosis that is attributed, in part, to glucose toxicity.

Regulation of mTOR-mediated Islet DNA Synthesis
Our current studies showed that glyburide in the nanomolar range at basal glucose stimulated DNA synthesis to the same magnitude as elevated glucose, and these effects were blocked by rapamycin. Nifedipine prevented glyburide-mediated DNA synthesis indicating that the ability of glyburide to stimulate DNA synthesis was due to an increase in Ca 2ϩ influx from the extracellular media. Maedler et al. (29) reported with isolated human islets that 0.1 and 10 nM glyburide in the presence of basal glucose after 4 days of exposure produced small increases (2.2-and 2.5-fold, respectively) in ␤-cell apoptosis. In the ␤-cell line, INS-1, Briaud et al. (30) reported that glyburide (5 M) increased DNA synthesis at basal glucose (3 mM) but was ineffective at an elevated glucose concentration. Glyburide also failed to stimulate phosphorylation of S6K1 in this ␤-cell line. Their results indicated that glyburide and/or glucose exposure stimulated increases in cytosolic Ca 2ϩ that were primarily associated with Erk1/2 phosphorylation, and DNA synthesis likely occurred through a protein kinase A-dependent pathway. The effects of rapamycin on glucose-or glyburide-mediated DNA synthesis were not evaluated in this study. A plausible explanation for the lack of agreement of the effects of glyburide on S6K1 phosphorylation with our findings is that these studies were performed in a rapid proliferating cell line, INS-1, as opposed to primary islets.
In our previous studies, diazoxide treatment resulted in a partial inhibition of S6K1 phosphorylation suggesting that diazoxide might reduce glucose-stimulated DNA synthesis (7). Paradoxically, diazoxide (250 M) in the presence of glucose (20 mM) resulted in a marked 2-4-fold enhancement of DNA synthesis above that of elevated glucose alone and occurred in a rapamycin-sensitive manner. This potentiating effect also occurred under conditions when diazoxide was absent from the incubation media during the final 24 h of the 96 h total incubation period. These effects of diazoxide were also compared with those of pinacidil, another K ATP channel opener (18). Pinacidil mimicked the effects produced by diazoxide on glucose-stimulated DNA synthesis but to a significantly lesser degree possibly due to its lower potency in opening K ATP channels in comparison with diazoxide in pancreatic ␤-cells (data not shown). The ability of diazoxide to attenuate nutrient overstimulation of mTOR (7) might also be anticipated to down-regulate the negative feedback pathway from mTOR/S6K1 to IRS2. It is possible that attenuating this negative feedback pathway with diazoxide may increase ␤-cell mass by reducing apoptosis through increased growth factor activation of IRS2-, PI3K-, and Akt-regulated pathways other than mTOR in vivo. Functional IRS2/PI3K/Akt signaling is considered essential for ␤-cell survival (1).
Studies were next performed with nifedipine, an inhibitor of L-type Ca 2ϩ channels, to determine whether the ability of diazoxide to attenuate Ca 2ϩ influx was responsible for the marked potentiation of glucose-stimulated DNA synthesis. Nifedipine failed to mimic the stimulatory effects produced by diazoxide (Fig. 4) even though nifedipine completely blocked insulin secretion in response to glucose (data not shown). Overall, these results suggested that diazoxide might be exerting effects independent of the activation of plasma membrane K ATP channels that are responsible for the potentiation of glucose-stimulated DNA synthesis. Findings by Kullin et al. (31) also reported that K ATP channel openers including diazoxide decreased the vulnerability of rat islets to free radicals induced by alloxan, nitroprusside, or IL-1␤. It was proposed that K ATP channel openers by their ability to decrease Ca 2ϩ influx due to ␤-cell membrane hyperpolarization in combination with their direct effect to decrease the mitochondrial membrane potential reduced ATP production necessary for the activation of apoptotic pathways. In cardiac mitochondria, Holmuhamedov et al. (16) have also proposed that K ATP channel openers modulate Ca 2ϩ homeostasis through dissipation of the mitochondrial membrane potential that leads to reduced mitochondrial Ca 2ϩ influx and provides protection against ischemic injury.
Based on cell cycle analysis, our studies suggest that the ability of chronic elevated glucose or glyburide to enhance DNA synthesis in adult islets is linked to an accumulation of cells in S phase. Conditions of chronic hyperglycemia or pregnancy have been shown to cause hypertrophy of islet cells; however, the possibility that hypertrophy is responsible for S phase accumulation of DNA is unlikely, since the synergy of cell growth and proliferation is a G 1 phenomenon regulated by retinoblastoma protein and not by S phase (32). Although the mechanism responsible for this defect in cell cycle progression is unknown, it may be related to excess free radical production which predisposes the mitochondria to Ca 2ϩ overload that mediates DNA damage. It has been shown under chronic hyperglycemic conditions that mRNA levels of some stress-related genes are increased in islets cells such as, heme oxygenase-1, glutathione peroxidase, A20, Fas, iNOS, and c-Myc (33,34). Based on flow cytometry, S phase accumulation represents new DNA synthesis. The arrest of islet cells in S phase may be due to the lack of protein expression, i.e. cyclins, necessary to complete cell cycle progression or possibly a checkpoint in cell cycle progression due to the over expression of stress-related proteins that cause a response to DNA damage.
The ability of diazoxide to significantly enhance [ 3 H]thymidine incorporation above that of elevated glucose alone might be related to the protective effect that diazoxide exerts on the ␤-cell mitochondria. Since the ability of diazoxide to enhance DNA synthesis occurs without any significant alterations in S phase accumulation, this increase in DNA synthesis appears to be linked to an arrest in S phase cell cycle progression and not translated to cell proliferation. In contrast, chronic exposure of islets to basal glucose that only partially activates mTOR results in a typical cell cycle progression pattern that is consistent with a low level of proliferation. Santiago-Walker et al. (35) have recently reported that the ability of PKC␦ to stimulate apoptosis of primary rat thyroid cells is linked to its ability to stimulate G 1 phase cell cycle progression and S phase arrest. Evidence that apoptosis was a direct consequence of replication stress was established by using the PI3K inhibitor, LY294002, that prevented S phase entry and apoptosis. Based on cell cycle analysis by flow cytometry, we have no evidence to indicate that a significant degree of apoptosis occurs under our in vitro conditions with primary rat islets (data not shown). Interestingly, Belanger et al. (36) reported that the anti-apoptotic protein Bcl-2 delays cell proliferation and promotes accumulation of cells in S phase without affecting the rate of apoptosis by human ovarian carcinoma cells. Bcl-2 as a regulator of apoptosis is believed to alter mitochondrial permeability transition through its ion channel activity and inhibits the release of both cytochrome c and apoptosis-inducing factor from the mitochondria, thereby promoting cell survival. Importantly, it has been reported that damage to DNA results in activation of damage response pathways in human epidermal keratinocytes (37). The signal derived from DNA damage to the activation of these response pathways is mediated by mTOR. A similar mechanism may be responsible for enhanced [ 3 H]thymidine incorporation that occurs in a rapamycin-sensitive manner as a consequence of chronic exposure of adult rodent islets to elevated glucose. The ability of rapamycin to partially prevent in a dose-dependent manner the accumulation of DNA in S phase may reflect the necessity of inhibiting mTOR more completely at the higher concentration of 100 nM that maximally inhibits cell cycle progression from G 0 /G 1 to S phase. The reason for the differences in sensitivity to rapamycin of [ 3 H]thymidine incorporation versus cell cycle progression are unknown but may be due to different mTOR-dependent pathways involving DNA damage responses and/or cell cycle checkpoints.
Overall, in our model of chronic glucose exposure we observed an accumulation of DNA in S phase that does not translate to an increase in cell proliferation. This is consistent with the well established finding that only a few adult islet cells are progressing through the cell cycle at any one time. Thus, it appears that the enhanced [ 3 H]thymidine incorporation observed under conditions of prolonged elevated glucose alone or in combination with diazoxide or glyburide may be related to a response to DNA damage that is mediated by mTOR. Although an increase in DNA synthesis based on either [ 3 H]thymidine or bromodeoxyuridine incorporation is required for cell division, this essential prerequisite does not always result in completion of the cell cycle. Our studies suggest that the appropriate regulation of mTOR by nutrients and pharmacologic agents that mediate DNA synthesis and cell cycle progression may be a component of an effective strategy to enhance the survival, growth, and proliferative capacity of adult rodent islets.