Hepatocyte growth factor preserves beta cell mass and mitigates hyperglycemia in streptozotocin-induced diabetic mice.

Type I diabetes is an autoimmune disease that results in destructive depletion of the insulin-producing beta cells in the islets of Langerhans in pancreas. With the knowledge that hepatocyte growth factor (HGF) is a potent survival factor for a wide variety of cells, we hypothesized that supplementation of HGF may provide a novel strategy for protecting pancreatic beta cells from destructive death and for preserving insulin production. In this study, we demonstrate that expression of the exogenous HGF gene preserved insulin excretion and mitigated hyperglycemia of diabetic mice induced by streptozotocin. Blood glucose levels were significantly reduced in mice receiving a single intravenous injection of naked HGF gene at various time points after streptozotocin administration. Consistently, HGF concomitantly increased serum insulin levels in diabetic mice. Immunohistochemical staining revealed a marked preservation of insulin-producing beta cells by HGF in the pancreatic islets of the diabetic mice. This beneficial effect of HGF was apparently mediated by both protection of beta cells from death and promotion of their proliferation. Delivery of HGF gene in vivo induced pro-survival Akt kinase activation and Bcl-xL expression in the pancreatic islets of diabetic mice. These findings suggest that supplementation of HGF to prevent beta cells from destructive depletion and to promote their proliferation might be an effective strategy for ameliorating type I diabetes.

Type I diabetes is an autoimmune disease that results in destructive depletion of the insulin-producing beta cells in the islets of Langerhans in pancreas. With the knowledge that hepatocyte growth factor (HGF) is a potent survival factor for a wide variety of cells, we hypothesized that supplementation of HGF may provide a novel strategy for protecting pancreatic beta cells from destructive death and for preserving insulin production. In this study, we demonstrate that expression of the exogenous HGF gene preserved insulin excretion and mitigated hyperglycemia of diabetic mice induced by streptozotocin. Blood glucose levels were significantly reduced in mice receiving a single intravenous injection of naked HGF gene at various time points after streptozotocin administration. Consistently, HGF concomitantly increased serum insulin levels in diabetic mice. Immunohistochemical staining revealed a marked preservation of insulin-producing beta cells by HGF in the pancreatic islets of the diabetic mice. This beneficial effect of HGF was apparently mediated by both protection of beta cells from death and promotion of their proliferation. Delivery of HGF gene in vivo induced prosurvival Akt kinase activation and Bcl-xL expression in the pancreatic islets of diabetic mice. These findings suggest that supplementation of HGF to prevent beta cells from destructive depletion and to promote their proliferation might be an effective strategy for ameliorating type I diabetes.
Diabetes is a devastating illness with significant morbidity and mortality, and its incidence has increased steadily worldwide (1). Type I diabetes is an autoimmune disease that results in extensive destruction of the insulin-producing beta cells in the islets of Langerhans in pancreas. Current clinical therapy for type I diabetes is to replace insulin through multiple injections, or replace the destroyed beta cell mass via islet or whole pancreas transplantation (2)(3)(4). Strategies in development include the establishment of human cell lines that can produce, process, store, and release insulin (5); direct intra-pancreatic injection of adenoviral vector containing human insulin gene (6,7); and the induction of endocrine differentiation of stem cells or/and beta cell precursors (8 -11). However, these approaches have their inherent limitations. For instance, injections at best only produce fluctuated, dysregulated levels of circulating insulin. Likewise, the shortage of human islets precludes the clinical use of islet transplantation on a large patient population. In addition, it remains a question whether the transplanted beta cells can survive for a long duration, in light of the continuous presence of the autoimmune triggers responsible for initial destruction of beta cells. In this context, we hypothesized that protecting pancreatic beta cells from destructive death may represent a new, novel strategy for preserving insulin production and for prevention and treatment of type I diabetes.
Hepatocyte growth factor (HGF) 1 is a mesenchyme-derived, multifunctional protein that plays a critical role in cell survival, proliferation, migration, and differentiation (12)(13)(14). Earlier studies demonstrated that HGF and its specific receptor, a receptor tyrosine kinase encoded by c-met protooncogene, are highly expressed during pancreas development. HGF is also shown to function as an insulinotropic factor and promotes beta cell proliferation and regeneration (15)(16)(17). Transgenic mice overexpressing HGF in the islet beta cell resist the diabetogenic effects of the beta cell toxin streptozotocin (STZ) (18,19). Furthermore, intraperitoneal injection of HGF protein exhibits a favorable effect for amelioration of hyperglycemia in diabetic mice receiving a marginal mass of islet grafts (20). Collectively, these observations led us to speculate that HGF might play a critical role in promoting insulin-producing beta cell survival after injurious stimuli; if so, delivery of HGF may provide an effective means for preserving insulin excretion and for mitigating hyperglycemia in diabetes in vivo.
Earlier studies show that exogenous HGF protein has a very short half-life in blood circulation, presumably owing to its rapid clearance by the liver in vivo (21,22). Thus, it is tremendously difficult to sustain circulating levels of exogenous HGF protein through direct injection, even at very short intervals. This problem may be potentially solved, at least partially, by developing an HGF gene transfer strategy that allows persistent expression in vivo. To this end, we recently described an efficient approach to introduce exogenous HGF gene in vivo by systemic administration of a naked plasmid vector. This hydrodynamics-based in vivo gene transfer strategy results in a high level of exogenous HGF protein in the circulation. Hence, this approach could provide an efficient way to examine the poten-tial of exogenous HGF on preservation of beta cells in the pancreatic islets of diabetic animals.
In the present study, we report the effects of exogenous HGF on the development of diabetes in a mouse model induced by STZ. Our results suggest that exogenous HGF promotes beta cell survival and proliferation and, thereby, mitigates hyperglycemia in STZ-induced diabetic mice.

EXPERIMENTAL PROCEDURES
Animals-Male CD-1 mice weighing 18 -22 g were purchased from Harlan Sprague-Dawley (Indianapolis, IN). They were housed in the animal facilities of the University of Pittsburgh Medical Center with free access to food and water. Animals were treated humanely in accordance with National Institutes of Health guidelines and by use of approved procedures of the Institutional Animal Use and Care Committee at the University of Pittsburgh.
Plasmid Injection-The recombinant human HGF expression plasmid (pCMV-HGF) that contains full-length human HGF cDNA driven under a human cytomegalovirus (CMV) promoter was cloned as described previously (23). The empty expression plasmid vector pcDNA3 was purchased from Invitrogen (Carlsbad, CA). Plasmid DNA was administrated into mice by a hydrodynamic-based gene transfer technique via rapid injection of a large volume of DNA solution through the tail vein, as previously described (24 -26). Briefly, 20 g of plasmid DNA was diluted in 1.6 ml of saline and injected via the tail vein into the circulation within 5-10 s. Mice were injected with pCMV-HGF plasmid at 16 h prior to administration of STZ (Sigma, St. Louis, MO). Control mice were injected with 20 g of empty vector pcDNA3 plasmid in an identical manner. STZ was given by a single intravenous injection at the dose of 100 mg/kg body weight (BW), unless indicated otherwise. Groups of mice (n ϭ 6) were sacrificed at 7 days after administration of STZ, respectively, and the pancreas was removed. For some experiments, additional groups of mice were sacrificed at 6, 12, 24, and 48 h after administration of STZ, respectively. One part of the pancreas was fixed in 10% phosphate-buffered formalin for histological studies after paraffin embedding. Another part was immediately frozen in optimal cutting temperature compound for cryosectioning.
Blood Glucose Measurement-Blood was collected at 1, 3, and 7 days after administration of STZ by retro-orbital bleeding. Blood glucose levels were determined by using a colorimetric glucose assay kit, according the procedures specified by the manufacturer (Sigma).
Determination of Insulin Level-Mouse serum insulin levels were determined by using an ultrasensitive insulin enzyme-linked immunosorbent assay (ELISA) kit (Crystal Chem Inc., Chicago, IL). Briefly, serum samples were collected at 1, 3, and 7 days after administration of STZ. Five-microliter samples or different concentrations of mouse insulin standard solutions and 95 l of sample diluents were added to the 96-well microtiter plates pre-coated with the guinea pig anti-rat insulin antibody. After incubating for 2 h at 4°C and followed by extensive washing, a 100-l aliquot of horseradish peroxidase-conjugated anti-rat insulin antibody was added, and the plates were incubated for 30 min at room temperature. Following washing, they were then incubated with 100 l of enzyme substrate solution. The plates were allowed to stand for 40 min at room temperature, and the reaction was stopped by addition of enzyme reaction stopping solution. The absorbance was read at 405 nm by an EL800 Universal Microplate Reader (Bio-Tek Instruments, Inc., Winooski, VT). The concentration of serum insulin was expressed as nanograms per milliliter. For determination of the insulin content in pancreatic islets, isolated islets were resuspended in acid/ ethanol (0.18 M HCl in 70% ethanol), sonicated, and extracted at Ϫ20°C, as described previously (27). After neutralization, the insulin content in the islet extracts was determined by using the same ELISA kit. The data were expressed as nanograms per microgram of total protein.
Immunofluorescence Staining-Indirect immunofluorescence staining was performed using an established procedure (14,28). Briefly, pancreatic cryosections were prepared and mounted onto poly L-lysinecoated slides and fixed in 3% paraformaldehyde in PBS for 30 min, followed by incubation with 0.2% Triton X-100 for 15 min. The slides were stained with the specific primary antibodies using the Vector M.O.M. immunodetection kit according to the protocols specified by the manufacturer (Vector Laboratories, Burlingame, CA). The primary antibodies used were monoclonal mouse insulin antibody (Clone 2D11-H5, Ventana Medical Systems Inc., Tucson, AZ) and a mixture of polyclonal rabbit anti-glucagon (A0565, alpha cell marker), anti-somatostatin (A0566, delta cell marker), and anti-pancreatic polypeptide (A0619, PP cell marker) antibodies (Dako, Carpenteria, CA). To visualize the pri-mary antibodies, tissues were stained with either Cy-3 TM or fluorescein isothiocyanate-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories). The slides were then stained with 4Ј,6-diamidino-2-phenylindole, HCl (DAPI) to visualize the nuclei. Stained slides were mounted with anti-fade mounting medium (Vector Laboratories) and viewed with a Nikon Eclipse E600 Epi-fluorescence microscope equipped with a Nikon DXM 200 digital camera (Melville, NY). For quantitation, the percentages of a particular cell type in the pancreatic islets were calculated, which were determined on at least five islets per animal, six animals per group.
Immunohistochemical Staining-Pancreatic sections from the mice were prepared at 4 m thick by a routine procedure. Sections were deparaffinized and rehydrated through a graded series of alcohol to distilled H 2 O. Endogenous peroxidase was blocked in a solution of 3% H 2 O 2 in methanol for 10 min at room temperature. Double immunohistochemical staining was performed using the M.O.M. immunodetection kit according to the instructions specified by the manufacturer (Vector Laboratories). The slides were first stained with monoclonal mouse insulin antibody (Clone 2D11-H5, Ventana Medical Systems Inc.) or a mixture of polyclonal rabbit anti-glucagon, anti-somatostatin, and antipancreatic polypeptide antibodies, followed by incubating with their respective secondary antibodies. For visualizing proliferating cell nuclear antigen (PCNA), the slides were reincubated with anti-PCNA antibody (Santa Cruz Biotechnologies, Santa Cruz, CA). As a negative control, the primary antibody was replaced with nonimmune mouse IgG, and no staining occurred. Slides were viewed with a Nikon Eclipse E600 microscope equipped with a Nikon DXM200 digital camera (Melville, NY). The percentages of PCNA-positive cells in the pancreatic islets were calculated, which were determined on at least five islets per animal, six animals per group.
TUNEL Staining-In situ detection of DNA fragmentation was performed using terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick-end labeling (TUNEL) staining, according to the procedure described previously (26,29). A fluorescein-based TUNEL staining protocol was employed by using an Apoptosis Detection System (Promega, Madison, WI). Cryosections of the pancreas were washed with PBS, fixed with 3% paraformaldehyde, and then treated with 0.2% Triton X-100 for 10 min. After pre-equilibration in 100 l of buffer containing 200 mM potassium cacodylate, 25 mM Tris-HCl, pH 6.6, 0.2 mM dithiothreitol, 0.25 mg/ml bovine serum albumin, and 2.5 mM cobalt chloride, strands of DNA were end-labeled by incubation at 37°C for 1 h in 50 M fluorescein-12-dUTP, 100 M dATP, 10 mM Tris-HCl, pH 7.6, 1 mM EDTA, and TdT. The reaction was stopped by adding 2ϫ sodium chloride/sodium citrate hybridization buffer for 15 min. To localize the pancreatic islets, immunofluorescence staining was performed on cryosections with anti-insulin antibody, as described elsewhere (14). For visualizing cell nuclei, the slides were also stained with DAPI. After washing, the slides were mounted with Vectashield anti-fade mounting media (Vector Laboratories) and observed on Nikon Eclipse E600 Epifluorescence microscope. The apoptotic index in the pancreatic islets was defined as a percentage of apoptotic cells in total islet cell population, which were determined on at least five islets per animal, six animals per group.
Isolation of Pancreatic Islets-Mouse pancreatic islets were isolated by collagenase digestion followed by separation on Ficoll-Conray gradients, according to the procedure described previously (19). Briefly, the pancreas was injected through the pancreatic duct with 3 ml of 1.7 mg/ml collagenase (Sigma) in Hanks' buffered saline solution (HBSS), removed, incubated at 37°C for 20 min, and then passed through a 500-m wire nylon mesh. The digested pancreas was rinsed with HBSS, and the islets were separated by density gradient in the Ficoll-Conray device (Sigma). After several washes with HBSS, islets were picked manually using a Pasteur pipette with the aid of a dissecting microscope and pooled for protein extraction.
Western Blot Analysis-The isolated pancreatic islets as described above were homogenized in radioimmune precipitation assay lysis buffer (1% Nonidet P-40, 0.1% SDS, 100 g/ml phenylmethylsulfonyl fluoride, 0.5% sodium deoxycholate, 1 mM sodium orthovanadate, 2 g/ml aprotinin, 2 g/ml antipain, and 2 g/ml leupeptin in PBS) on ice, and the supernatants were collected after centrifugation at 13,000 ϫ g at 4°C for 20 min. Protein concentration was determined using a bicinchoninic acid protein assay kit (Sigma), and islet lysates were mixed with an equal amount 2ϫ SDS loading buffer (100 mM Tris-HCl, 4% SDS, 20% glycerol, and 0.2% bromphenol blue). Samples were heated at 100°C for 5-10 min before loading and separated on precasted 10% SDS-polyacrylamide gels (Bio-Rad, Hercules, CA). The proteins were electro-transferred to a nitrocellulose membrane (Amersham Biosciences, Piscataway, NJ) in transfer buffer containing 48 mM Tris-HCl, 39 mM glycine, 0.037% SDS, and 20% methanol at 4°C for 1 h. Nonspecific binding to the membrane was blocked for 1 h at room temperature with 5% Carnation nonfat milk in TBS buffer (20 mM Tris-HCl, 150 mM NaCl, and 0.1% Tween 20). The membranes were then incubated for 16 h at 4°C with various primary antibodies in blocking buffer containing 5% milk. The antibodies against phosphospecific Akt kinase and total Akt were obtained from Cell Signaling (Beverly, MA). The antibodies against Bcl-xL and actin were purchased from Santa Cruz Biotechnology, Inc. Following extensive washing for three times, the membranes were then incubated with horseradish peroxidase-conjugated secondary antibody (Bio-Rad) for 1 h at room temperature in 5% nonfat milk. The signals were visualized by the enhanced chemiluminescence system (ECL, Amersham Biosciences).
Immunoprecipitation-Immunoprecipitation was carried out essentially according to the procedures described previously (30). Briefly, pancreatic islets isolated from various groups were homogenized in radioimmune precipitation assay lysis buffer (1% Nonidet P-40, 0.1% SDS, 100 g/ml phenylmethylsulfonyl fluoride, 0.5% sodium deoxycholate, 1 mM sodium orthovanadate, 2 g/ml aprotinin, 2 g/ml antipain, and 2 g/ml leupeptin in PBS) on ice, and the supernatants were collected after centrifugation at 13,000 ϫ g at 4°C for 20 min. After protein concentration was determined, 200 g of the lysates was precleared by adding 0.25 g of normal rabbit IgG and 20 l of protein A/G Plus-agarose (Santa Cruz Biotechnologies) into 1 ml of whole islet lysates. After incubation for 1 h at 4°C, supernatants were collected by centrifugation at 1,000 ϫ g for 5 min at 4°C. Lysates were immunoprecipitated overnight at 4°C with 1 g of anti-mouse c-Met (Santa Cruz Biotechnologies) and anti-phosphotyrosine-PY20 (BD PharMingen/Transduction Laboratories, San Diego, CA), respectively, followed by precipitation with 20 l of protein A/G Plus-agarose for 3 h at 4°C. After four washes with radioimmune precipitation assay buffer, the immunoprecipitates were boiled for 5 min in SDS sample buffer. The resulting precipitated complexes were separated on SDS-polyacrylamide gels and blotted with various antibodies as described above.
Statistical Analysis-Animals were randomly assigned to control and treatment groups. Data were expressed as mean Ϯ S.E. Data were analyzed using SigmaStat software (Jandel Scientific, San Rafael, CA).
Comparison between groups was made using one-way analysis of variance followed by the Student-Newman-Keuls test. A p value of less than 0.05 was considered significant.

HGF Mitigates Hyperglycemia and Preserves Insulin Production in Streptozotocin-induced Diabetic
Mice-Mice were administrated intravenously with HGF plasmid at 16 h prior to injection of a specific diabetogenic agent, STZ. As previously reported (25), following a single, intravenous injection of naked plasmid encoding human HGF cDNA driven under a cytomegalovirus promoter (pCMV-HGF), serum HGF levels increased as early as 4 h, reaching a peak at 12 h. Quantitative determination by a specific ELISA for human HGF protein revealed ϳ8 ng/ml exogenous HGF in the circulation at the peak level (data not shown). Although serum HGF levels tended to decline, significant HGF protein (more than 1 ng/ml) was detectable in the circulation at 7 days after a single injection.
Blood glucose levels at different time points in mice after STZ injection are presented in Fig. 1A. We found that HGF significantly mitigated the STZ-induced hyperglycemia in diabetic mice. In mice injected with empty pcDNA3 vector, blood glucose levels started to elevate as early as 1 day and continued to increase to more than 300 mg/dl at 7 days after intravenous administration of STZ at 100 mg/kg BW. The dynamics of hyperglycemia observed in this study manifested the onset and development of diabetes in mice, presumably due to the destruction and depletion of beta cells in the islet of Langerhans in pancreas, as typically seen in this model. However, in mice given pCMV-HGF plasmid, blood glucose levels at various time points were significantly reduced, compared with the pcDNA3 control group (Fig. 1A). This reduction of blood glucose levels by HGF was reproducible in the diabetic mice induced by different doses of STZ (Fig. 1C), indicating that delivery of HGF gene substantially mitigates the hyperglycemia in diabetic animals.
We next examined the serum insulin levels in the STZinduced diabetic mice. As shown in Fig. 1B, during the course of the development of diabetes induced by STZ, serum insulin levels progressively declined and reached to ϳ20% of normal level at 7 days after injection of STZ at 100 mg/kg BW. The reduction of serum insulin levels was closely correlated with an increase in blood glucose levels (Fig. 1A). Single injection of HGF gene significantly improved serum insulin levels at various time points in the diabetic mice induced by STZ. At 7 days, serum insulin levels in mice receiving HGF plasmid injection were more than 3-fold of that receiving empty vector (Fig. 1B). Consistently, HGF also increased serum insulin levels in the diabetic mice induced by various doses of STZ (Fig. 1D).
HGF Preserves Insulin-producing Beta Cells in Mouse Pancreas-The beneficial effects of exogenous HGF described above prompted us to further investigate the impacts of exogenous HGF on the survival and preservation of insulin-producing beta cells in the islets of pancreas in diabetic mice. To distinguish the beta cells from non-beta cells in the islets, we employed a double immunofluorescence staining technique. Beta cells were stained with anti-insulin antibody (red), whereas non-beta islet cells were recognized by a mixture of antibodies against glucagon (alpha cell marker), somatostatin (delta cell marker), and pancreatic polypeptide (PP cell marker) (green). Fig. 2 shows representative micrographs of the pancreatic islets of diabetic mice at different time points after STZ administration. In mice injected with STZ at 100 mg/kg BW, the numbers of insulin-positive beta cells in the pancreatic islets dropped very rapidly. At 48 h after STZ, very few, if any, islet cells with positive staining with insulin were observed in mice receiving empty pcDNA3 vector injection (Fig. 2C). Of note, there was a significant number of cells that were not stained with either insulin (red) or non-beta islet cell markers (green) within the islets after STZ. Such cells are presumably the degranulated beta cells caused by severe hyperglycemia. It is of interest to note that, owing to marked depletion of beta cells after STZ, non-beta islet cells (green) apparently aggregated in the center of the islets where beta cells normally were positioned (Fig. 2, C and D).
Delivery of HGF gene significantly mitigated the depletion of the beta cells in diabetic mice. Substantial numbers of insulinpositive beta cells was still present at 7 days after STZ in mice receiving pCMV-HGF plasmid injection (Fig. 2H). Fig. 2I shows the percentage of beta cells in the pancreatic islets at different time points after STZ between pcDNA3 and pCMV-HGF groups. At 7 days after injection of STZ at 100 mg/kg BW, the percentage of beta cells in the pancreatic islets dropped below 20%; most of them, if not all, were insulin-negative, degranulated beta cells. However, in mice receiving pCMV-HGF injection, about 60% of the total cell population within the islets was beta cells, and most of them displayed insulin positivity (Fig. 2,  H and I). Consistent with these observations, we found a marked difference in the insulin content in the isolated islets at 7 days after STZ between pcDNA3 and pCMV-HGF groups (Fig. 2J), as determined by a quantitative measurement using specific ELISA.
HGF Protects Beta Cells from Death in Diabetic Mice Induced by STZ-We further investigated the cellular events that led to an increase in insulin-positive beta cells in STZ-induced diabetic mice by HGF. Because STZ is a diabetogenic agent that specifically induces beta cell depletion, we first examined the effects of HGF on beta cell death in the pancreatic islets of diabetic mice in vivo. Fig. 3 shows the effects of HGF gene transfer on apoptotic cell death in the pancreatic islets of diabetic mice induced by STZ, as demonstrated by TUNEL stain- ing. Apoptosis in the pancreatic islets was clearly evident during the onset and development of diabetes induced by STZ (Fig.  3C). Time course studies revealed that apoptotic cell death in the islets peaked at 12 h after STZ injection in vivo (Fig. 3G). Intravenous injection of the HGF gene markedly prevented islet cells from apoptosis (Fig. 3F). The apoptotic index was significantly lower in diabetic mice receiving HGF gene than those given the empty vector (Fig. 3H). Of note, HGF was capable of inhibiting cell apoptosis in the pancreatic islets of diabetic mice induced by different doses of STZ (Fig. 3H). Thus it appears that preservation of insulin-producing beta cells by HGF in diabetic mice is mediated, at least in part, through promoting beta cell survival.
HGF Promotes Beta Cell Proliferation in Vivo-Because alterations in either cell death or cell growth could potentially play a role in the preservation of beta cells, we next investigated the effects of HGF on beta cell proliferation in the pancreatic islets of diabetic mice in vivo. As shown in Fig. 4, HGF significantly promoted cell proliferation in the pancreatic islets in the early time points after injection of STZ. Immunohistochemical staining for proliferating cell nuclear antigen (PCNA), an indicative marker for G 1 to S phase transition, revealed a significant increase in the percentages of positive cells in the pancreatic islets of diabetic mice pre-administrated with HGF plasmid (Fig. 4). This induction of PCNA expression in the pancreatic islets was prominent at 6 and 12 h after STZ (Fig. 4E); much less increase in PCNA was observed beyond 48 h after the onset of diabetes. Double-immunohistochemical staining showed that most PCNA-positive cells in the islets did not express non-beta cell markers (Fig. 4, A and B). Instead, co-localization of PCNA and insulin was clearly evident in the islets of pancreas in mice receiving HGF plasmid (Fig. 4D), indicating that HGF may preferentially promote beta cell proliferation. These results suggest that HGF preservation of the beta cell mass in the pancreatic islets of diabetic mice may also be mediated by enhancing beta cell proliferation.

HGF Promotes Beta Cell Survival by Activating Cell-survival
Signaling-To elucidate the potential mechanism underlying HGF promotion of beta cell survival and amelioration of hyperglycemia in diabetic mice, we investigated the signal transduction pathways leading to cell survival by using isolated pancreatic islets from diabetic animals. As shown in Fig. 5, HGF receptor activation was confirmed by detecting the tyrosine phosphorylation of c-Met protein in the isolated islets. Immunoprecipitation studies revealed that more than 2-fold induction of tyrosine-phosphorylated c-Met was observed in the pancreatic islets from the mice receiving pCMV-HGF injection, compared with that in mice receiving pcDNA3 or in normal control mice (Fig. 5B), suggesting that the c-Met receptor in the pancreatic islets is activated after intravenous injection of naked HGF plasmid vector.
We examined the activation of protein kinase B/Akt kinase, a signaling protein that plays a central role in dictating cell survival or death (31), in the isolated pancreatic islets from the mice receiving either pcDNA3 or pCMV-HGF plasmid. As shown in Fig. 6, STZ at 100 mg/kg BW reduced the levels of phosphorylated Akt at 6 h, but not at 12 h, in the pancreatic islets. However, administration of HGF plasmid markedly induced phosphorylated Akt levels in the islets in a time-dependent manner. More than 4-fold induction of phosphorylated Akt was found in the isolated islets at 6 h after STZ administration. This induction of phosphorylated Akt in the pancreatic islets was transient; the phosphorylated Akt level only slightly increased and tended to return toward baseline at 12 h after the initial injury. Fig. 7 shows the expression levels of pro-survival protein Bcl-xL in the pancreatic islets of the diabetic mice. Administration of the HGF gene markedly induced Bcl-xL overexpression in the pancreatic islets. More than 6-fold induction of Bcl-xL protein was observed at 6 h after administration of STZ at 100 mg/kg BW in the islets from the mice receiving pCMV-HGF plasmid (Fig. 7). Of note, severe diabetes induced by a higher dose of STZ (such as 150 mg/kg BW) was associated with a repressed endogenous Bcl-xL protein expression, and HGF not only reversed this repression of Bcl-xL but also induced its expression in such conditions (data not shown). Thus, exogenous HGF dramatically activates and induces pro-survival signaling that results in Akt phosphorylation and Bcl-xL overexpression in the pancreatic islets of diabetic mice. DISCUSSION Although the pathogenesis underlying the onset and development of type I diabetes remains elusive, it is generally considered as an autoimmune disease that results in extensive destruction and depletion of insulin-producing beta cells in the islets of Langerhans in pancreas by cytotoxic T cells (32,33). In this regard, protection of beta cells and/or promotion of their proliferation by supplementing growth factor could be a novel strategy to enhance beta cell resistance to apoptotic and necrotic damage inflicted by the immune system and various injuries (10,34). In this study, we have demonstrated that systemic delivery of naked HGF plasmid protected beta cells from death and promoted their proliferation in a mouse model of diabetes induced by beta cell toxin STZ. The preservation of beta cells was accompanied by amelioration of hyperglycemia and increased circulating insulin. These findings suggest that protection of beta cells from destructive depletion and promotion of their proliferation by HGF might be an effective approach for ameliorating type I diabetes.
HGF preservation of beta cell mass in the pancreatic islets after injury is likely mediated by a dual mechanism: a decrease in beta cell loss and an increase in beta cell proliferation. However, it remains elusive as to a quantitative contribution of these mechanisms in mediating the beneficial effects of HGF on preservation of beta cell mass. In light of the fact that the primary pathogenesis of diabetes in STZ model is beta cell death, it is plausible to speculate that HGF preservation of the insulin production may be largely mediated by protection of the beta cells against apoptotic cell death after injurious stimulus. This cytoprotective role of HGF for pancreatic beta cells in vivo is consistent with a previous report in which transgenic mice overexpressing the HGF gene under rat insulin promoter resist the diabetogenic effects of STZ (18). It has also been shown that repeated injections of HGF protein ameliorate hyperglycemia in STZ-induced diabetic mice receiving a marginal mass of intrahepatic islet grafts (20). Although the mechanism underlying the beneficial effects of HGF was not determined in that study, it is possible that HGF elicits its actions through a protective effect on grafted islet cells from apoptosis in diabetic mice after transplantation (20), because earlier studies show that islet cell injury in the islet grafts after transplantation is mediated through the mechanisms of apoptosis (35,36). However, cell proliferation may also play a significant role in mediating the beneficial effects of HGF, because there was an induction of the expression of PCNA (Fig. 4), a marker for G 1 to S phase transition, in the insulin-positive beta cells after delivery of HGF gene. Regardless of the mechanisms involved, the fact that HGF inhibits beta cell apoptosis and promotes  5. HGF receptor phosphorylation and activation in the isolated islets after injection of HGF plasmid. Mice were injected with either pcDNA3 or pCMV-HGF plasmid at 16 h prior to STZ administration. Pancreatic islets were isolated from various groups as indicated at 6 h after of STZ. A, the lysates from the islets were immunoprecipitated with antibodies against either c-Met or phosphotyrosine (PY20), respectively, followed by immunoblotting with different antibodies. The picture shows the results of one islet preparation from a pool of three animals. B, graphic presentation shows c-Met tyrosine phosphorylation in various groups. Relative levels of phosphorylated c-Met (-fold induction) were presented after normalization with precipitated total c-Met abundance. Isolated islet samples were prepared from the pancreas pooled from three mice. Data are presented as mean Ϯ S.E. of three preparations (n ϭ 3). *, p Ͻ 0.05. their proliferation in vivo underscores the remarkable ability of HGF to preserve beta cell mass of the pancreatic islets after destructive challenges.
In harmony with the ability to promote beta cell survival, delivery of the HGF gene resulted in marked activation of cell survival signaling in the pancreatic islets in vivo. Activation of PKB/Akt kinase and induction of Bcl-xL in the isolated pancreatic islets as presented in this study (Figs. 6 and 7) provide mechanistic explanations for the impressive beta cell survivalpromoting capacity of HGF. Consistent with this observation, a recent study (27) also demonstrates an Akt activation in the isolated islets after treatment with HGF adenoviral vector. Of note, both Akt activation and Bcl-xL induction occur maximally at 6 h after STZ administration, a time point that significantly precedes the peak of apoptosis in the pancreatic islets (12 h). Akt kinase has recently emerged as a key component of cell survival machinery that plays a central role in regulating cell apoptosis and survival (31,37). Extensive studies demonstrate that Akt activation will lead to multiple signal transduction pathways that result in cell survival. We previously showed that, in renal tubular epithelial cells, HGF induces Akt phosphorylation as early as 5 min through a phosphoinositide 3kinase-dependent pathway (38), which is followed by phosphorylation and resultant inactivation of Bad, a pro-death member of Bcl-2 family (31). Furthermore, HGF can override Bad-induced renal epithelial cell death by stimulating Bad phosphorylation at both Ser 112 and Ser 136 sites (38). Hence, the dramatic activation of Akt kinase by HGF in isolated pancreatic islets implies that the pathway involving sequential phosphorylation of phosphoinositide 3-kinase, Akt, and Bad could also be a major signal mechanism leading to beta cell survival. It should be noted that, in addition to inducing Bad phosphorylation and inactivation, Akt has been shown to phosphorylate the Forkhead family of transcription factors such as FKHRL1, leading to their inactivation and inability to transcriptionally activate the genes essential for cell death (39,40). Therefore, multiple possibilities exist as to the pathways elicited by activated Akt in mediating beta cell survival. It remains a question as to whether one single pathway plays a predominant role, or multiple pathways are necessary for mediating HGF/Akt-induced beta cell survival in the pancreatic islets. Further studies are needed to clarify these issues and to delineate the signaling mechanism leading to beta cell survival.
Besides Akt activation, the protein abundance of Bcl-xL, a pro-survival member of Bcl-2 family proteins, is markedly induced in the isolated pancreatic islets in mice receiving HGF plasmid injection. Of interest, the endogenous Bcl-xL protein level falls in the islets after STZ in a dose-dependent fashion, suggesting that a decrease in pro-survival Bcl-xL protein may account for, at least in part, the increasing apoptosis induced by the beta cell toxin. It is intriguing to note that exogenous FIG. 6. HGF activates pro-survival Akt kinase in the pancreatic islets. A, representative Western blot shows the Akt phosphorylation in the isolated islets at different time points as indicated after administration of STZ at 100 mg/kg BW. Isolated islet samples were prepared from the pancreas pooled from three mice and immunoblotted with antibodies against either phospho-specific Akt or total Akt kinase, respectively. Numbers (1 and 2) indicate two separate sample preparations, each of them representing a pool of isolated islets from three mice. B, graphic presentation shows the relative phospho-Akt abundance after normalizing with total Akt in various treatment groups. Relative levels of phosphorylated Akt (-fold induction over normal control) were presented after normalization with total Akt abundance. Each sample of the isolated islet was prepared from a pool of three mice. Data are presented as mean Ϯ S.E. of three preparations (n ϭ 3). *, p Ͻ 0.05 versus normal control; **, p Ͻ 0.01 versus pcDNA3.

FIG. 7. HGF induces pro-survival Bcl-xL protein expression in the pancreatic islets.
A, representative Western blot shows the Bcl-xL induction in the isolated islets at different time points after administration of STZ at 100 mg/kg BW. Each sample of the isolated islet was prepared from a pool of three mice and immunoblotted with antibodies against either Bcl-xL or actin, respectively. Numbers (1 and 2) indicate two separate sample preparations, each made from a pool of three mice. B, graphic presentation demonstrates the relative Bcl-xL abundance after normalizing with actin in various treatment groups. Relative levels of Bcl-xL (-fold induction over normal control) were presented after normalization with actin. Each sample of the isolated islet was prepared from a pool of three mice. Data are presented as mean Ϯ S.E. of three preparations (n ϭ 3). **, p Ͻ 0.01 versus other groups.