Abstract
Treatment of type 1 diabetes by islet transplantation is currently limited by loss of functional β-cell mass after transplantation. We investigated here whether adenovirus-mediated changes in AMP-activated protein kinase (AMPK) activity, previously shown to affect insulin secretion in vitro, might affect islet graft function in vivo. In isolated mouse and rat islets, insulin secretion stimulated by 17 (vs 3) mmol/l glucose was inhibited by 36.5% (P<0.01) and 43% (P<0.02) respectively after over-expression of constitutively-active AMPK- (AMPK CA) versus null (eGFP-expressing) viruses, and glucose oxidation was decreased by 38% (P<0.05) and 26.6% (P<0.05) respectively. Increases in apoptotic index (terminal deoxynucleotide transferase-mediated deoxyuridine trisphosphate biotin nick end-labelling) (TUNEL)) were also observed in AMPK CA- (22.8 ± 3.6% TUNEL-positive cells, P<0.001), but not AMPK DN- (2.72 ± 3.9%, positive cells, P=0.05) infected islets, versus null adenovirus-treated islets (0.68 ± 0.36% positive cells). Correspondingly, transplantation of islets expressing AMPK CA into streptozotocin-diabetic C57 BL/6 mice improved glycaemic control less effectively than transplantation with either null (P<0.02) or AMPK-DN-infected (P<0.01) islets. We conclude that activation of AMPK inhibits β-cell function in vivo and may represent a target for therapeutic intervention during islet transplantation.
Introduction
In the wake of improvements achieved by Shapiro et al.(2000), human islet transplantation is now considered a potentially useful treatment modality for type 1 diabetes (Korsgren et al. 2005). At present, however, the shortage of transplantable human islet material and the requirement for multiple donors severely limit the usefulness of this approach. This problem is further aggravated by the substantial (60–80%) and time-dependent loss of islet material that appears to occur after transplantation (Korsgren et al. 2005, Rickels et al. 2005). A similar phenomenon is observed in other species (Davalli et al. 1995) and probably occurs by multiple, non-immune and immune-mediated mechanisms (Korsgren et al. 2005). Strategies which inhibit these losses, and thus enhance β-cell survival, are therefore likely to reduce the number of islet equivalents required initially, and to extend the usefulness of islet transplantation. To this end, adenovirus-mediated introduction into islets of potentially cyto-protective genes, such as bcl-2 (Contreras et al. 2001), insulin-like growth factor-1 (Giannoukakis et al. 2000a), Iκ-B kinase inhibitor (Rehman et al. 2003), Iκ-B repressor (Giannoukakis et al. 2000b), erythropoietin (Fenjves et al. 2004) or a tumour necrosis factor (TNF) receptor fusion decoy (Machen et al. 2004), is considered an exciting approach.
Adenosine monophosphate-activated protein kinase (AMPK), an evolutionarily conserved serine/threonine kinase (Carling 2004), is ubiquitously expressed in mammalian tissues and is involved in the regulation of substrate (notably fatty acid) oxidation. AMPK is seen as a potentially interesting therapeutic target for type 2 diabetes (Ruderman & Prentki 2004), since activation of the enzyme in muscle and liver enhances glucose uptake and suppresses glucose release respectively (Zhou et al. 2001). Changes in AMPK activity in these tissues have thus been proposed to be involved in the anti-hyperglycaemic effects of metformin and the glitazone class of antidiabetic drugs (Fryer et al. 2002). On the other hand, activation of AMPK in isolated rodent and human islets, as well as clonal β cells, suppresses glucose metabolism and glucose-stimulated insulin secretion (GSIS) (da Silva Xavier et al. 2000, Eto et al. 2002, Rutter et al. 2003, Leclerc et al. 2004), and enhances β-cell death through apoptosis (Kefas et al. 2003b) (F Diraison and G A R, unpublished data). Importantly, the extent to which AMPK activation in the β cell limits the otherwise beneficial effects of the above antihyperglycaemic agents is presently unexplored (Rutter et al. 2003).
The role of AMPK in the regulation of insulin secretion is largely untested in vivo. Thus, while whole-body inactivation of the α2-isoform of the catalytic subunit by homologous recombination in mice (Viollet et al. 2003) leads to abnormal glucose tolerance, this appears largely to be the result of increased sympathetic tone. Importantly, islets isolated from α2-knockout mice display no evident abnormalities in glucose-stimulated insulin secretion ex vivo (Viollet et al. 2003), consistent with the predominant expression of α1 in islets and β cells (da Silva Xavier et al. 2000). Moreover, compensatory increases in AMPKα1 activity in β cells cannot be ruled out in the α2 knockout mouse model.
The present study was undertaken with two aims in mind:
-
to clarify the role of AMPK in the control of insulin secretion and β-cell mass in vivo
-
to assess the therapeutic potential of islet-specific regulation of the enzyme in a rodent model.
We demonstrate that adenovirus-mediated over-expression of an activated form of AMPK (Woods et al. 2000) markedly inhibits the improvement of glycaemic control achieved by transplantation of islets under the kidney capsule of streptozotocin-diabetic syngeneic mice, demonstrating the potential importance of AMPK in controlling functional β-cell mass in vivo.
Materials and Methods
Adenoviral preparation
Adenoviruses expressing the constitutively active NH2-terminal fragment common to mammalian AMPKα1 and α2 (AMPK1–312,T172D; Ad-AMPK CA), and the full-length AMPKα2 mutated in the active site (D157A; Ad-AMPK DN) or enhanced green fluorescent protein alone (Ad-eGFP) were generated as described (da Silva Xavier et al. 2003) by subcloning the corresponding cDNAs into pAdTrackCMV (He et al. 1998). Viruses were amplified in HEK293 cells and purified as described on CsCl gradients (Ainscow & Rutter 2001).
Animals
Inbred male C57 BL/6 mice aged 8–14 weeks (20–25 g) were used as donors and transplant recipients. All experimental mice were bred and housed in the specific-pathogen-free facility, University of Bristol. They were housed at 22 °C with 12-h light/dark cycle and ad libitum access to chow and water. Procedures were performed in accordance with UK Home Office regulations, and the NIH ‘Principles of Laboratory Animal Care’ were followed throughout. Male Wistar rats (220–250 g) were supplied by the University of Bristol Medical School Animal Facility.
Islet isolation
Donor mice were killed by cervical dislocation and underwent laparotomy to expose the common bile duct. The duct was then clamped at its insertion into the duodenum and cannulated proximally with a 27-gauge needle and 3 ml collagenase P solution (1 mg/ml, Sigma) in Hanks’ buffered salt solution (HBSS; Gibco) supplemented with 0.01 M HEPES, pH 7.4) injected. Isolated pancreata were digested at 37 °C for 8–9 min. After incubation, the digested tissue was washed, hand-shaken and centrifuged at 170 g for 3 min with HEPES-buffered HBSS (see above) augmented with 5% fetal calf serum (FCS). After passing through a 1 mm tissue-collecting sieve, islets were purified on a discontinuous Ficoll density gradient (Sigma) (Gotoh et al. 1985). Islets were hand-picked and transferred into Dulbecco’s modified Eagle’s medium (DMEM) (Sigma) supplemented with 10% FCS, 11 mM glucose, 100 i.u./ml penicillin and 100 μg/ml streptomycin for further culture.
Ex vivo gene transfer into mouse islets
Islets were infected with Ad-eGFP, Ad-AMPK CA or Ad-AMPK DN at a multiplicity of infection (MOI) of 50–100 plaque-forming units per islet cell (assuming islets contained an average of 2000 cells) (Flotte et al. 2001) for 16 h. Infection of islets with the null virus at this MOI did not provoke any changes in apoptotic index or glucose-stimulated insulin secretion in comparison to uninfected islets (results herein) (Diraison et al. 2004, Parton et al. 2004). Islets were then incubated for 48 h at 37 °C in 5% CO2 prior to ex vivo studies, or were transplanted immediately.
Extraction and assay of AMPK activity
For uniformity, mass islet isolations were performed and the resulting islets divided into aliquots for viral infection and comparison. Equal numbers and sized islets were infected with one of each of the three constructs (AdeGFP, AdAMPK-CA or AdAMPK-DN). Quantification of AMPK activity was performed as described previously (da Silva Xavier et al. 2003). In brief, 400 transduced islets were scraped into ice-cold lysis buffer (50 mM Tris–HCl, pH 7.4, at 4 °C/250 mM sucrose/50 mM NaF/1 mM sodium pyrophosphate/1 mM EDTA/1 mM EGTA/1 mM DTT/1% (v/v) Triton X-100/complete protease inhibitor mixture; Roche Diagnostics). Extracts were centrifuged (13 000 g for 5 min at 4 °C), and protein concentration was determined with DC protein assay reagent (Bio-Rad). This protocol releases over 90% of total cellular content of each AMPK isoform (Salt et al. 1998). AMPK was immunoprecipitated from 100 μl cell extract with either sheep anti-α1 or anti-α2 antibodies. AMPK activity was measured in 10 μl crude extract by using ‘SAMS’ peptide (the synthetic peptide HMRSAMSGLHLVKRR) and γ-32P-ATP (specific activity, 1000 c.p.m./pmol).
Insulin secretion
At 48 h after adenoviral infection, islets were incubated for 60 min at 37 °C in 2 ml Krebs-Ringer bicarbonate HEPES buffer (KRBH) (130 mM NaCl, 3.6 mM KCl, 1.5 mM CaCl2, 0.5 mM MgSO4, 0.5 mM KH2PO4, 2.0 mM NaHCO3 and 10 mM HEPES) supplemented with 3 mM glucose and 0.5% (w/v) BSA pre-equilibrated with 95% O2:5% CO2, pH 7.4. Islets were separated into three groups of five islets per condition and incubated for 30 min in 1 ml KRBH as above containing either 3 or 17 mM glucose. Total insulin was extracted in acidified ethanol (75% EtOH, 23.5% H2O and 1.5% HCl). Insulin was measured by radioimmunoassay by competition with 125I-labelled rat insulin (Linco Research, St Charles, MO, USA) according to the manufacturer’s instructions. Released insulin is expressed as percentage of the total cellular (extractable) insulin content.
[U-14C]Glucose oxidation
Islets were preincubated for 30 min at 37 °C in KRBH supplemented with 3 mM glucose and 0.5% BSA (w/v). Triplicate groups of 100 islets were then placed in 250 μl KRHB containing 3 or 17 mM unlabelled glucose, 1.7 μCi [U-14C]glucose and 0.5% BSA (w/v) in a 24-well plate. A rubber gasket, the size of the 24-well plate and containing 0.5 cm holes, was aligned over the plate. A CO2 capture chamber was created as described previously (Collins et al. 1998). A UniFilter-24 GF/B plate (Packard Instrument, Research Parkway Meridien, Connecticut, USA) was sealed with an adhesive sheet, and 100 μl of 40% (w/v) KOH were pipetted onto each filter. The filter plate was inverted and aligned over the rubber gasket to form a small CO2 capture chamber. Finally, the chamber was sealed with a 6 mm glass plate, a 6 mm metal plate and a lead weight to ensure an airtight seal. The apparatus was incubated for 2 h at 37 °C. Filters were removed, and captured 14CO2 was measured by scintillation counting. Control incubations lacking islets were included in each incubation series.
Preparation of islet cryostat sections and immunocytochemistry
Islets were infected with adenoviruses, cultured for 48 h and then washed with PBS and fixed with 3% (w/v) formaldehyde for 16 h at 4 °C. For sectioning, islets were fixed with Zamboni’s fixative (Stefanini et al. 1967) overnight at 4 °C, immersed in a solution of 30% (w/v) sucrose in PBS overnight, and subsequently frozen in OCT compound (Tissue-Tek; Sakura Finetecnical, Tokyo, Japan). Islet sections (10 μm) were obtained with a cryostat (Bright OTF5000, Jencons, Leighton Buzzard, UK). Islet slices were permeabilized with 0.3% (w/v) Triton X-100 overnight, and then blocked in 3% (w/v) BSA in PBS for 15 min. Slices were incubated with guinea pig anti-insulin antiserum at 1:500 dilution in a humidified chamber at 4 °C overnight. After washing with PBS, islets were incubated again overnight at 4 °C with tetramethyl rhodamine isothiocyanate (TRITC)-conjugated anti-guinea pig antibody (1:500 dilution). After a final wash in PBS, islet slices were mounted in a mixture of Mowiol (Merck) and glycerol. Images were captured on a Leica SP2 laser scanning confocal microscope with a 63 oil immersion objective with excitation at 350, 488 (Ar) and 543 nm (He-Ne). Emitted light was detected at >515 nm for eGFP (green) or >560 nm for insulin (red).
Terminal deoxynucleotide transferase-mediated deoxyuridine trisphosphate nick end-labelling (TUNEL) assay
Islets were cultured for 72 h with null, AMPK CA or DN adenoviruses at 100 MOI as indicated. Islets were then washed twice with ice-cold PBS and 10 μm cryostat sections prepared as above for immunocytochemistry. Sections were permeabilized in 0.2% (v/v) triton X-100 in PBS for 1 h at 4 °C before incubation with TUNEL-labelling solution containing terminal deoxynucleotidyl transfersase (Roche) for 1 h at 37 °C. Slides were then washed twice with PBS and incubated in DAPI (Sigma; 10 μg/ml) in PBS for 10 min. After a final wash in PBS, sections were mounted in movial solution. Positive controls to test the specificity of the TUNEL assay were prepared by incubating in DNase 1 (Roche; 100 U/ml) in 50 mM Tris–HCl, pH 7.5, 1 mg/ml BSA for 10 min at room temperature. Islets were imaged on an SP2 Leica confocal microscope with excitation at 560 nm for TMR-red-labelled nucleotides (red), 488 nm for eGFP (green) and 350 nm for DAPI nuclear staining, and the number of apoptotic cells was counted. The number of TUNEL-positive (apoptotic) cells was counted and expressed as a percentage of total number of cells per islet. At least 12 islets were counted in each case.
Preparation of islets for transplantation
After three gentle washes, transduced islets were hand-picked and carefully counted. Islets were placed into a 1.5 ml tube on ice and pelleted by centrifugation. Supernatant was removed and the islet pellet transferred into a sterile P200 pipette tip plugged with wax. The pipette tip was centrifuged at 170 g for 3 min, and the supernatant was aspirated to yield a dry islet pellet at its tip.
Islet transplantation
To induce diabetes, animals were fasted overnight and then administered 160 mg/kg of streptozotocin (STZ) (Sigma) in citrate buffer (0.1 M sodium citrate and 0.1 M citric acid, pH 4.5) by an intraperitoneal route. Transplantation was carried out 3 days after administration of STZ. Blood glucose levels were determined by an Accucheck II blood glucose monitor (Roche) using whole blood collected from the tail vein. Mice with a blood glucose value over 20 mmol/l were used as transplant recipients. Mice were anaesthetized by an intraperitoneal injection of fentanyl (10 mg/ml; Hypnorm, Janssen, Wantage, UK) and midazolam (5 mg/ml; Hypnovel) mixed with sterile distilled water in a ratio of 1:1:2, and a volume of 50–100 μl was administered. The left kidney was accessed through a small subcostal incision. A small puncture was made in the renal subcapsule over the inferior pole, and the pipette tip containing the islets was inserted under the capsule and advanced to the superior pole. The islet pellet was then deposited and the pipette tip carefully removed without spillage. The kidney was replaced in the peritoneal cavity and the skin stapled with clips (SLS, Nottingham, UK).
Blood glucose analysis was performed on postoperative day (POD) 1 and thrice weekly to 30 days and thereafter weekly to day 100. In accordance with UK Home Office regulations, animals losing over 15% of preoperative body weight or in distress were killed.
Intraperitoneal glucose tolerance tests
An intraperitoneal glucose tolerance test (IPGTT) was performed on surviving recipients at POD 30. The animals were fasted overnight for not more than 16 h. Each animal was then weighed, and a starved blood glucose value recorded. Animals were fasted overnight, and after administration of a 10% intraperitoneal glucose solution (2 g/kg body mass), samples of blood were taken from the tail vein at 15, 30, 60 and 120 min after the glucose challenge.
Graft nephrectomy
Animals were anaesthetized and the old incision was opened. The left kidney containing the graft was carefully dissected away from peritoneum and adhesions. The renal pedicle was identified and suture ligated with 4/0 vicryl (Ethicon, Edinburgh, UK). The kidney was then excised and the skin closed once more with clips. Blood glucose analysis was performed at 1 and 24 h after graft nephrectomy.
Statistics
Data are given as means ± s.e.m. Comparisons between means were performed with Student’s t-test for unpaired data, or one-way ANOVA with post hoc Bonferroni correction for measurements of glycaemia, using Graphpad Prism (GraphPad Software Inc., San Diego, CA, USA).
Results
Effects of changes in AMPK activity on glucose metabolism, insulin secretion and apoptosis in isolated mouse islets
Infection of islets with Ad-eGFP, Ad-AMPK CA or Ad-AMPK DN (see Materials and Methods) was confirmed by confocal microscopy to detect eGFP fluorescence. Similar levels of eGFP expression were apparent after infection with each virus (Fig. 1A). The efficiency of infection ranged from ~25% in the majority of islets (>150 μm) to ~70% in smaller islets (diameter of <100 μm), with a predominant localization in peripheral, insulin-positive cells as observed previously in rat islets (Diraison et al. 2004) (Fig. 1A). Transduction with AMPK-expressing viruses was associated with the expected changes in extractable AMPK activity (Fig. 1B).
Over-expression of AMPK CA in isolated mouse islets inhibited glucose-stimulated insulin secretion (17 versus 3 mM) by approximately 40% (P<0.01), but there was no marked effect after expression of dominant-negative AMPK (AMPK DN). Similar findings were observed in rat pancreatic islets where AMPK CA inhibited the secretion of insulin in response to glucose by ~50% (P<0.01) (results not shown) as consistent with previous results (Diraison et al. 2003, Leclerc et al. 2004). No effect of either virus (CA or DN) on total islet insulin content was observed after infection of islets from either species (data not shown).
We next determined whether the effects of over-expression of AMPK CA may be explained in part by a reduction in glucose oxidation, given the previous finding of a decrease in glucose-induced changes in β-cell ATP content and reduced pyridine nucleotide fluorescence (da Silva Xavier et al. 2003). In control mouse islets expressing eGFP alone (Null virus-infected), a five-fold induction in the oxidation of [U-14C]glucose was observed at elevated (17 versus 3.0 mM) glucose concentrations (Fig. 2). This increase was significantly (30%, P<0.05) inhibited in mouse islets that over-expressed AMPK CA. In contrast, islets transduced with AMPK DN showed no significant difference in the oxidation of glucose at either glucose concentration. Essentially similar data were obtained with rat islets (data not shown).
Previous reports have suggested that pharmacological activation of AMPK, or expression of the constitutively activated enzyme in β-cell lines (Kefas et al. 2003a,b) and purified rat β cells (Kefas et al. 2003b, 2004), can lead to apoptosis. To determine whether over-expression of activated AMPK may induce apoptosis in β cells in the context of the intact, isolated mouse islet, we monitored DNA cleavage in islet slices by incorporating tetramethyl rhodamine-labelled nucleotides into the free 3′-OH end of DNA strand breaks, using terminal deoxynucleotidyl transferase (TUNEL assay). The number of apoptotic cells was then counted in islets infected with null (eGFP), Ad-AMPK CA or Ad-AMPK DN viruses. Whereas expression of AMPK DN had no significant effect on the number of apoptotic cells compared with null virus, AMPK CA caused a ~30-fold increase after 72 h of infection (Table 1).
Impact of changes in islet graft AMPK activity on glycaemic control
We next explored the cytoprotective and metabolic effects of modulating β-cell AMPK activity in vivo by monitoring the survival and function of a suboptimal islet mass transplanted into syngeneic diabetic mice. To this end, a suboptimal islet transplant mass was first determined. Islets were treated with adenoviruses in culture for 16 h before transplantation. This period was chosen to ensure adequate uptake of viral particles by the islet, while avoiding the deleterious effects (e.g. central islet necrosis or changes in islet vasculature) of extended islet culture in vitro. Indeed, we observed no recovery of glycaemia after transplantation with islets previously cultured for 48 h, even at high islet numbers (600 islets/animal; n=8). We defined a suboptimal mass as the number of adenovirally infected transplanted islets that would have less than a 25% success rate in restoring euglycaemia (defined as two consecutive post-transplant blood glucose readings below 15 mmol/l). Preliminary data showed that transplantation of 600 islets resulted in a cure rate of 100% (n=5), while transplantation of 300 islets resulted in a cure rate of 25% (n=4). To confirm that the transplantation of 300 islets constituted a suboptimal islet mass, diabetic mice were transplanted with 200 islets, resulting in no improvement in glycaemia compared with untransplanted animals (n=6). These experiments (see Figure 7) established that transplantation of 300 islets constituted a suboptimal, marginal islet mass, likely to be most sensitive to any effect of increasing or decreasing AMPK activity.
To test whether AMPK manipulation could affect β-cell function during the early post-transplantation period, 300 Ad-AMPK CA- or Ad-AMPK DN-infected islets were transplanted into diabetic mice, and blood glucose levels were monitored over a 20-day period. Mice transplanted with 300 Ad-eGFP-infected islets were used as controls. Comparing the mean blood glucose value for mice receiving 300 Ad-AMPK CA islets (n=8) to those mice receiving Ad-eGFP(control) virus-infected islets (n=8) over the 20-day period showed significant impairment in glycaemic control in the Ad-AMPK CA group (P=0.0016; Fig. 3A). By contrast, the mean blood glucose levels for mice receiving Ad-AMPK DN (n=7) islets were not significantly different from those observed in mice receiving Ad-eGFP (control) virus-infected islets (n=8) (P=0.64) (Fig. 3B). However, when mice transplanted with Ad-AMPK CA-infected islets (n=8) and Ad-AMPK DN-infected islets (n=7) (Fig. 3C) were compared, a significant impairment in glycaemic control was seen in the former cohort (P=0.009).
Mice were weighed and monitored regularly over a 30-day postoperative period. All mice transplanted with Ad-eGFP-transduced islets survived this period (n=8). Of the mice receiving Ad-AMPK-CA- or Ad-AMPK-DN-infected islets, 6/8 (75%) and 6/7 (86%) survived respectively. The cohorts’ 30-day weight change is shown in Fig. 4. The mean weight change in surviving mice receiving Ad-AMPK CA-infected islets was significantly less than in those mice receiving Ad-AMPK DN-infected islets (−9.0 ± 4.4% vs 3.25 ± 3.25%; P=0.03). There was no significant difference when comparing surviving mice transplanted with Ad-eGFP (control) islets (1.31 ± 4.1%) and either Ad-AMPK CA islets (P=0.06) or Ad-AMPK DN-infected islets (P=0.36).
For a finer assessment of the function of the transplanted islets in vivo, an intraperitoneal glucose tolerance test (IPGTT) was performed. The model was established with control non-diabetic mice and STZ-induced diabetic mice without islet transplantation (Fig. 5A). Each contemporaneous cohort of mice underwent IPGTT on POD 30 (Fig. 5A). The mean area under the curve (AUC) in surviving mice receiving Ad-AMPK CA-infected islets (2947 ± 178.6, n=6) was significantly greater than in animals receiving Ad-eGFP-infected islets (2341 ± 183.3, n=8, P<0.04) or Ad-AMPK DN-infected islets (1971 ± 243.8, n=6, P=0.01), demonstrating poorer glucose tolerance (Fig. 5B).
To verify that the changes reported above reflected differences in graft function rather than recovery of endogenous β-cell mass, nephrectomy was performed on a single cohort of three previously transplanted animals, 100 days after the original transplantation. Each animal showed an elevation in blood glucose concentration within 1 h, and at 24 h all were hyperglycaemic (>20 mmol/l) (Fig. 6).
Discussion
Effects of AMP-activated protein kinase on glucose-stimulated insulin secretion in vitro
Over-expression of activated AMPK in intact mouse islets is shown here markedly to suppress glucose oxidation, consistent with previous measurements of ATP content and reduced pyridine nucleotide fluorescence in rat islets and clonal β-cells (da Silva Xavier et al. 2000, Tsuboi et al. 2003). Similarly, glucose-stimulated insulin secretion was inhibited to a similar extent by over-expression of AMPK CA, suggesting that changes in glucose metabolism may be the principal mechanism through which AMPK activation affects glucose-stimulated insulin secretion. Interestingly, these changes were observed despite the relatively low infection efficiency (~20–30%) with the virus, an observation which probably reflects the fact that the outermost layers of β cells are the most active metabolically and in terms of glucose-stimulated insulin release after a period of culture, where degradation of the intraislet vasculature is likely. These levels of infection were not significantly increased in medium-large islets (over ~150 μm diameter) even up to an MOI of ~1000 in our hands, a point at which apoptosis becomes apparent after expression of null (Ad-eGFP) virus (L Parton, F Diraison and G A R, unpublished results; see also Diraison et al. 2004).
Expression of AMPK CA also led to a dramatic increase in islet cell apoptosis in vitro, a phenomenon which seems likely to contribute substantially to the increased incidence of islet failure after islet transplantation in vivo. These observations are consistent with very recent findings showing that activation of AMPK leads to cell-cycle arrest in a number of cell types (Nagata et al. 2004, Xiang et al. 2004, Jones et al. 2005), and with the identification of the tumour suppressor LKB1 as an important upstream AMPK kinase (Hawley et al. 2003, Woods et al. 2003). Whether apoptosis in β cells is the cause or the consequence of altered glucose metabolism remains to be firmly established, but the fact that treatment of islet or β cells with 5-amino-imidazole carboxamide riboside (AICAR) (da Silva Xavier et al. 2003, Leclerc et al. 2004) or the AMPK-activator metformin (Leclerc et al. 2004) rapidly inhibits insulin release suggests that apoptosis may be a downstream consequence of the metabolic changes (Kefas et al. 2004).
By contrast, introduction of AMPK DN was largely without effect on glucose oxidation or insulin release. The latter finding in mouse islets differs from our previous results obtained in rat (da Silva Xavier et al. 2000) or human (Leclerc et al. 2004) islets, where introduction of AMPK DN caused an increase in insulin release at sub-maximal glucose concentrations. It should be emphasized, however, that no effort was made in the present study to investigate the effects of AMPK selectively on smaller islets, where higher efficiencies of transfection (>70%) were achieved in our earlier studies (Leclerc et al. 2004).
Effects of AMPK activation on islet function in vivo
The findings described here extend to the in vivo setting the results of the present and previous (da Silva Xavier et al. 2000, 2003, Leclerc et al. 2004) studies using isolated islets. The decrease in apparent β-cell function observed after the transplantation of islets expressing AMPK CA presumably reflects both the poorer acute response to glucose, and a decrease in β-cell mass due to enhanced apoptosis (see above) (Kefas et al. 2003b).
A loss of functional β-cell mass of up to 70% occurs after transplantation in man (2). In the present work, activation of AMPK is shown to be detrimental to β-cell function and survival in vivo. However, despite a tendency to improved glucose tolerance compared with null virus treatment (Fig. 4A), we were unable to affect islet function significantly after transplantation by expression in islets of a dominant-negative form of AMPK. There may be two possible explanations for this observation. Firstly, endogenous AMPK may remain essentially unstimulated in transplanted islets, at least at early time points. Secondly, the relatively small (~15%; Fig. 1B) decrease in AMPK activity observed in vitro seems likely to be due to changes in cells situated at the islet periphery which were efficiently transduced with the virus, leaving cells deeper within the islet core unaffected.
In conclusion, we demonstrate that expression of activated AMPK selectively in pancreatic islets affects glucose metabolism, insulin secretion and β-cell survival, and is associated with decreased β-cell function in vivo. We propose that suppression of islet AMPK activity may thus represent a potential therapeutic target for intervention after islet transplantation in type 1 diabetics.
Apoptosis assessed by DNA nick end-labelling (TUNEL) of islet sections. Islets were isolated and cultured for 72 h in adenoviruses as indicated. TUNEL labelling was performed as described in Materials and Methods. The numbers of TUNEL-positive cells are expressed as a percentage of total number of cells per islet. At least 12 islets were counted for each experimental condition
| TUNEL-positive cells (%) | |
|---|---|
| ***P<0.001 versus control adenovirus. | |
| Null (eGFP) | 0.68+0.36 |
| AMPK CA | 22.8+3.60*** |
| AMPK DN | 2.72+1.13 |
| Positive control | 76.8+6.37*** |
Expression of Ad-eGFP, Ad-AMPK CA and Ad-AMPK DN in isolated islets and effect on extractable AMPK activity. (A) Islets from a single isolation were infected with adenoviruses as described (Materials and Methods). Images were captured by confocal scanning microscopy, and overlay projections were constructed. Representative islets are shown. (B) 400 islets per condition then underwent protein extraction and AMPK assay with SAMS peptide (see Materials and Methods). Data are the means ± s.e.m. of three separate experiments.
Citation: Journal of Endocrinology 187, 2; 10.1677/joe.1.06413
Effect of infection with Ad-eGFP, Ad-AMPK CA or Ad-AMPK DN on glucose-stimulated insulin secretion and glucose oxidation in isolated islets. Islets were infected with the indicated adenoviruses as described (Materials and Methods). (A) After 48 h, insulin secretion was measured during static incubation of islets for 1 h with KRB medium supplemented with indicated glucose concentrations. Total and released insulin were quantified by radioimmunoassay. (B) Glucose oxidation in the three groups of transduced islets was also determined as described (Materials and Methods).
Citation: Journal of Endocrinology 187, 2; 10.1677/joe.1.06413
Increased expression of activated AMPK is detrimental to glycaemic control. STZ-diabetic C56BL/6 mice were transplanted with 300 islets infected with Ad-AMPK CA (n=8), Ad-AMPK DN (n=7) or Ad-eGFP/control (n=8) at an MOI of 50–100. Blood glucose analysis was performed preoperatively at day 1 and three times weekly thereafter. (A–C) Mean blood glucose concentration for mice transplanted with Ad-eGFP-, Ad-AMPK CA- or Ad-AMPK DN-infected islets. Data from PODs 4–19 are omitted for clarity. Data are the means ± s.e.m. from the animals shown per condition. ** P<0.01 for the effect of Ad-AMPK CA versus Ad-eGFP at POD 3 (A) and versus Ad-AMPK DN at POD 20 (C). *P<0.05 for the effect of Ad-AMPK CA versus Ad-AMPK DN at POD 3 (Unpaired, two-tailed Student’s t-test). ANOVA analysis with Bonferroni’s multiple comparison test of blood glucose concentration values at all time points in the three animal cohorts confirmed that Ad-AMPK CA-infected islets displayed impaired glycaemic control when compared with either Ad-eGFP- (P<0.01) or Ad-AMPK DN-(P<0.01) transduced islets.
Citation: Journal of Endocrinology 187, 2; 10.1677/joe.1.06413
Impact of transplantation with islets infected with Ad-eGFP, Ad-AMPK CA or Ad-AMPK DN on recipient body weight. On postoperative day (POD) 30, surviving mice were weighed, and percentage weight change over the post-transplantation period was compared with the weight immediately before transplantation (POD 0). Data are means ± s.e.m. for the number of animals per condition as follows: Ad-eGFP, n=8; Ad-AMPK CA, n=6; and Ad-AMPK DN, n=6. *P<0.05 for the difference in weight change in Ad-AMPK CA- and Ad-AMPK DN-transplanted animal cohorts (unpaired Student’s t-test).
Citation: Journal of Endocrinology 187, 2; 10.1677/joe.1.06413
Transplantation of islets transduced with Ad-AMPK CA results in further impairment of glucose tolerance. (A) Intraperitoneal (IP) glucose tolerance tests (IPGTT) were performed in control non-diabetic C57BL/6 mice (n=8) and other animals 4 days after STZ administration (n=7). After standardized glucose challenge (2 g/kg IP), blood glucose analysis was undertaken. Data were collected for all animals surviving to 30 days. Error bars show mean blood glucose values for each cohort ± s.e.m. (Ad-eGFP, n=8 animals; Ad-AMPK CA, n=6; Ad-AMPK DN, n=6). *P<0.05 at 60 and 120 min when Ad-AMPK CA- and Ad-AMPK DN-infected islets were compared. (B) Data are mean AUC per condition ± s.e.m. **P<0.01 for the difference in glucose responsiveness between animals transplanted with Ad-AMPK CA- and Ad-AMPK DN-transduced islets. *P<0.05 for the difference in glucose responsiveness between animals transplanted with Ad-eGFP- and Ad-AMPK-CA-transduced islets.
Citation: Journal of Endocrinology 187, 2; 10.1677/joe.1.06413
Glycaemic control is dependent upon syngeneic islet graft. A single cohort of three animals (i.e. a single animal per condition) underwent graft nephrectomy, in accordance with UK Home Office regulations. Blood glucose analysis was performed prior to the procedure, and at the indicated times after nephrectomy.
Citation: Journal of Endocrinology 187, 2; 10.1677/joe.1.06413
Transplantation of 300 islets represents a suboptimal islet mass. Diabetic mice were transplanted with 600 (n=5), 400 (n=7), 300 (n=4) or 200 (n=6) isolated islets that had been incubated for 16 h. All mice transplanted with 600 islets were euglycaemic by POD 3 (blood glucose of<15 mmol/l). Forty-three per cent and 25% of animals transplanted with 400 and 300 islets respectively were euglycaemic by POD 3. None of the animals transplanted with 200 islets achieved euglycaemia over the same time period.
Citation: Journal of Endocrinology 187, 2; 10.1677/joe.1.06413
This study was supported by grants to G A R from the Wellcome Trust (Research Leave Fellowship and Programme 067081/Z/02/Z) and the Juvenile Diabetes Research Foundation International (JDRFI; 1–2003–235). IL is a Wellcome Trust Advanced Fellow. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
References
Ainscow EK & Rutter GA 2001 Mitochondrial priming modifies Ca2+ oscillations and insulin secretion in pancreatic islets. Biochemical Journal 353 175–180.
Carling D 2004 The AMP-activated protein kinase cascade – a unifying system for energy control. Trends in Biochemical Science 29 18–24.
Collins CL, Bode BP, Souba WW & Abcouwer SF 1998 Multiwell 14CO2-capture assay for evaluation of substrate oxidation rates of cells in culture. BioTechniques 24 803–808.
Contreras JL, Bilbao G, Smyth C, Eckhoff DE, Xiang XL, Jenkins S, Cartner S, Curiel DT, Thomas FT & Thomas JM 2001 Gene transfer of the Bcl-2 gene confers cytoprotection to isolated adult porcine pancreatic islets exposed to xenoreactive antibodies and complement. Surgery 130 166–174.
da Silva Xavier G, Leclerc I, Salt IP, Doiron B, Hardie DG, Kahn A & Rutter GA 2000 Role of AMP-activated protein kinase in the regulation by glucose of islet beta-cell gene expression. PNAS 97 4023–4028.
da Silva Xavier G, Leclerc I, Varadi A, Tsuboi T, Moule SK & Rutter GA 2003 Role for AMP-activated protein kinase in glucose-stimulated insulin secretion and preproinsulin gene expression. Biochemical Journal 371 761–774.
Davalli AM, Ogawa Y, Ricordi C, Scharp DW, Bonner-Weir S & Weir GC 1995 A selective decrease in the beta cell mass of human islets transplanted into diabetic nude mice. Transplantation 59 817–820.
Diraison F, Parton L, Ferre P, Foufelle F, Briscoe CP, Leclerc I & Rutter GA 2003 Over-expression of sterol response element binding protein-1c or culture with 5-aminoimidazole-4-carboxamide ribonucleoside induces lipogenesis and decreases glucose-stimulated insulin secretion from rat pancreatic islets. Diabetic Medicine 20 A77.
Diraison F, Parton L, Ferre P, Foufelle F, Briscoe CP, Leclerc I & Rutter GA 2004 Over-expression of sterol-regulatory-element-binding protein-1c (SREBP1c) in rat pancreatic islets induces lipogenesis and decreases glucose-stimulated insulin release: modulation by 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR). Biochemical Journal 378 769–778.
Eto K, Yamashita T, Matsui J, Terauchi Y, Noda M & Kadowaki T 2002 Genetic manipulations of fatty acid metabolism in beta-cells are associated with dysregulated insulin secretion. Diabetes 51 S414–S420.
Fenjves ES, Ochoa MS, Gay-Rabinstein C, Molano RD, Pileggi A, Mendez AJ, Inverardi L & Ricordi C 2004 Adenoviral gene transfer of erythropoietin confers cytoprotection to isolated pancreatic islets. Transplantation 77 13–18.
Flotte T, Agarwal A, Wang J, Song S, Fenjves ES, Inverardi L, Chesnut K, Afione S, Loiler S, Wasserfall C et al.2001 Efficient ex vivo transduction of pancreatic islet cells with recombinant adeno-associated virus vectors. Diabetes 50 515–520.
Fryer LG, Parbu-Patel A & Carling D 2002 The anti-diabetic drugs rosiglitazone and metformin stimulate AMP-activated protein kinase through distinct signaling pathways. Journal of Biological Chemistry 277 25226–25232.
Giannoukakis N, Mi Z, Rudert WA, Gambotto A, Trucco M & Robbins P 2000a Prevention of beta cell dysfunction and apoptosis activation in human islets by adenoviral gene transfer of the insulin-like growth factor I. Gene Therapy 7 2015–2022.
Giannoukakis N, Rudert WA, Trucco M & Robbins PD 2000b Protection of human islets from the effects of interleukin-1 beta by adenoviral gene transfer of an I kappa B repressor. Journal of Biological Chemistry 275 36509–36513.
Gotoh M, Maki T, Kiyoizumi T, Satomi S & Monaco AP 1985 An improved method for isolation of mouse pancreatic islets. Transplantation 40 437–438.
Hawley SA, Boudeau J, Reid JL, Mustard KJ, Udd L, Makela TP, Alessi DR & Hardie DG 2003 Complexes between the LKB1 tumor suppressor, STRADalpha/beta and MO25 alpha/beta are upstream kinases in the AMP-activated protein kinase cascade. Journal of Biology 2 28.
He TC, Zhou S, da Costa LT, Yu J, Kinzler KW & Vogelstein B 1998 A simplified system for generating recombinant adenoviruses. PNAS 95 2509–2514.
Jones RG, Plas DR, Kubek S, Buzzai M, Mu J, Xu Y, Birnbaum MJ & Thompson CB 2005 AMP-activated protein kinase induces a p53-dependent metabolic checkpoint. Molecular Cell 18 283–293.
Kefas BA, Cai Y, Ling Z, Heimberg H, Hue L, Pipeleers D & Van de Casteele M 2003a AMP-activated protein kinase can induce apoptosis of insulin-producing MIN6 cells through stimulation of c-Jun-N-terminal kinase. Journal of Molecular Endocrinology 30 151–161.
Kefas BA, Heimberg H, Vaulont S, Meisse D, Hue L, Pipeleers D & Van de Casteele M 2003b AICA-riboside induces apoptosis of pancreatic beta cells through stimulation of AMP-activated protein kinase. Diabetologia 46 250–254.
Kefas BA, Ca Y, Kerckhofs K, Ling Z, Martens G, Heimberg H, Pipeleers D & Van de Casteele M 2004 Metformin-induced stimulation of AMP-activated protein kinase in β-cells impairs their glucose-responsiveness and can lead to apoptosis. Biochemical Pharmacology 68 409–416.
Korsgren O, Nilsson B, Berne C, Felldin M, Foss A, Kallen R, Lundgren T, Salmela K, Tibell A & Tufveson G 2005 Current status of clinical islet transplantation. Transplantation 79 1289–1293.
Leclerc I, Woltersdorf WW, da Silva Xavier G, Rowe RL, Cross SE, Korbutt GS, Rajotte RV, Smith R & Rutter GA 2004 Metformin, but not leptin, regulates AMP-activated protein kinase in pancreatic islets: impact on glucose-stimulated insulin secretion. American Journal of Physiology. Endocrinology and Metabolism 286 E1023–E1031.
Machen J, Bertera S, Chang Y, Bottino R, Balamurugan AN, Robbins PD, Trucco M & Giannoukakis N 2004 Prolongation of islet allograft survival following ex vivo transduction with adenovirus encoding a soluble type 1 TNF receptor-Ig fusion decoy. Gene Therapy 11 1506–1514.
Nagata D, Takeda R, Sata M, Satonaka H, Suzuki E, Nagano T & Hirata Y 2004 AMP-activated protein kinase inhibits angiotensin II-stimulated vascular smooth muscle cell proliferation. Circulation 110 444–451.
Parton LE, Diraison F, Neill SE, Ghosh SK, Rubino MA, Bisi JE, Briscoe CP & Rutter GA 2004 Impact of PPAR over-expression and activation on pancreatic islet gene expression profile analysed with oligonucleotide microarrays. American Journal of Physiology. Endocrinology and Metabolism 287 E390–E404.
Rehman KK, Bertera S, Bottino R, Balamurugan AN, Mai JC, Mi Z, Trucco M & Robbins PD 2003 Protection of islets by in situ peptide-mediated transduction of the Ikappa B kinase inhibitor Nemo-binding domain peptide. Journal of Biological Chemistry 278 9862–9868.
Rickels MR, Schutta MH, Markmann JF, Barker CF, Naji A & Teff KL 2005 β-Cell function following human islet transplantation for type 1 diabetes. Diabetes 54 100–106.
Ruderman N & Prentki M 2004 AMP kinase and malonyl-CoA: targets for therapy of the metabolic syndrome. Nature Reviews. Drug Discovery 3 340–351.
Rutter GA, da Silva Xavier G & Leclerc I 2003 Roles of 5′-AMP-activated protein kinase (AMPK) in mammalian glucose homoeostasis. Biochemical Journal 375 1–16.
Salt IP, Johnson G, Ashcroft SJ & Hardie DG 1998 AMP-activated protein kinase is activated by low glucose in cell lines derived from pancreatic beta cells, and may regulate insulin release. Biochemical Journal 335 533–539.
Shapiro AMJ, Lakey JRT, Ryan EA, Korbutt GS, Toth E, Warnock GL, Kneteman NM & Rajotte RV 2000 Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. New England Journal of Medicine 343 230–238.
Stefanini M, De Martino C & Zamboni L 1967 Fixation of ejaculated spermatozoa for electron microscopy. Nature 216 173–174.
Tsuboi T, da Silva Xavier G, Leclerc I & Rutter GA 2003 5′ AMP-activated protein kinase controls insulin-containing secretory vesicle dynamics. Journal of Biological Chemistry 278 52042–52051.
Viollet B, Andreelli F, Jorgensen SB, Perrin C, Geloen A, Flamez D, Mu J, Lenzner C, Baud O, Bennoun M et al.2003 The AMP-activated protein kinase alpha2 catalytic subunit controls whole-body insulin sensitivity. Journal of Clinical Investigation 111 91–98.
Woods A, Azzout-Marniche D, Foretz M, Stein SC, Lemarchand P, Ferre P, Foufelle F & Carling D 2000 Characterization of the role of AMP-activated protein kinase in the regulation of glucose-activated gene expression using constitutively active and dominant negative forms of the kinase. Molecular and Cellular Biology 20 6704–6711.
Woods A, Johnstone SR, Dickerson K, Leiper FC, Fryer LG, Neumann D, Schlattner U, Wallimann T, Carlson M & Carling D 2003 LKB1 is the upstream kinase in the AMP-activated protein kinase cascade. Current Biology 13 2004–2008.
Xiang X, Saha AK, Wen R, Ruderman NB & Luo Z 2004 AMP-activated protein kinase activators can inhibit the growth of prostate cancer cells by multiple mechanisms. Biochemical and Biophysical Research Communications 321 161–167.
Zhou G, Myers R, Li Y, Chen Y, Shen X, Fenyk-Melody J, Wu M, Ventre J, Doebber T, Fujii N et al.2001 Role of AMP-activated protein kinase in mechanism of metformin action. Journal of Clinical Investigation 108 1167–1174.
