Reg proteins are normally expressed in pancreatic acinar cells, and the level of several of these proteins was significantly induced upon damage to the endocrine or exocrine pancreas. It has been established that Reg1 and pancreatic islet neogenesis-associated protein [INGAP, Reg3δ] promote the growth or regeneration of the endocrine islet cells. Recent reports suggest that Reg2 is an autoantigen normally expressed in islet β-cells. Reg2 overexpression in vitro offered protection to insulinoma cells. Overexpressed Reg3α increased cyclin D1 and CDK4 levels and the rate of proliferation in insulinoma cells. Acinar-specific overexpression of INGAP increased β-cell mass and protected the animals from streptozotocin-induced diabetes. Moreover, Reg2 gene expression was induced during pancreatitis. We hypothesized that Reg2 is a secreted protein that promotes the growth, survival, and/or regeneration of pancreatic endocrine and exocrine cells. To test its effectiveness, we used elastase-1 promoter (Ela-Reg2) to develop an acinar cell-specific overexpression of the Reg2 gene. Western blot analysis, real-time PCR, and immunohistochemistry revealed barely detectable levels of endogenous Reg2 in the pancreas of normal wild-type mice and increased Reg2 levels in the pancreas of Ela-Reg2 mice that were similar to or higher than Reg2 levels induced in experimental diabetes or pancreatitis. Compared with wild-type littermates, growth, blood glucose and insulin levels, and glucose tolerance were normal in Ela-Reg2 mice; pancreatic histology revealed no change in endocrine or exocrine tissues. Acinar-specific overexpression of the Reg2 gene offered no protection against streptozotocin-induced β-cell damage and diabetes, in hyperglycemia and weight loss, and no advantage in restoring glucose homeostasis and islet function within 3 mo. Furthermore, serum amylase level and pancreatic histochemistry showed that Reg2 overexpression did not protect acinar cells against caerulein-induced acute pancreatitis. In contrast to INGAP or Reg3β, exocrine overexpression of Reg2 offered no protection to the endocrine or exocrine pancreas, indicating clear subtype specificities of the Reg family of proteins.
- pancreatic islets
- transgenic mice
reg proteins constitute a conserved family of seven members (Reg1, Reg2, Reg3α, Reg3β, Reg3δ, Reg3γ, and Reg4) in rodents; their production in the pancreas (including the islets of Langerhans) was induced upon endocrine β-cell or exocrine acinar cell damage (2, 11, 14, 15). While some of these proteins [Reg1 and islet neogenesis-associated protein (INGAP or Reg3δ)] have been implicated in promoting β-cell replication and/or neogenesis in in vivo studies in transgenic and knockout mice (24–26, 32), the roles of the other five proteins have not been well characterized. A unique seven-amino acid insertion (QVAEEDE) near its NH2 terminus distinguishes Reg2 (found only in mice and hamsters) from other Reg proteins (4, 29). Results from a recent dual-labeled immunohistochemistry study indicate that Reg2 is predominantly expressed in islet β-cells (10); its expression in the whole pancreas was drastically induced upon streptozotocin-induced β-cell damage and diabetes (15). On the basis of our knowledge of other Reg proteins, we suspected that Reg2 was mobilized to protect the endocrine islets. Indeed, overexpressed Reg2 protected insulinoma cells against streptozotocin-induced apoptosis (L. Liu et al., unpublished observations); early vaccination with intact Reg2 or the COOH-terminal portion of Reg2 delayed the onset of diabetes in nonobese diabetic (NOD) mice (10). Similarly, overexpressed Reg3α increased cyclin D1 and CDK4 levels and the rate of proliferation in insulinoma cells (5), and INGAP, when overexpressed in acinar cells, induced pancreatic islet hyperplasia and protected transgenic mice from streptozotocin-induced diabetes (26). In the acinar cells, the levels of Reg1 and Reg3β were drastically increased in edematous or necrotizing pancreatitis (2, 11). Administration of Reg1 and Reg3β antibodies worsened sodium taurocholate-induced pancreatitis (30); caerulein-induced pancreatitis was less severe, but with more apoptosis and inflammation in Reg3β-knockout mice (8). Reg1 and Reg3β were thus considered antiapoptotic and/or anti-inflammatory. Moreover, the Reg2 level was markedly upregulated in the pancreas of keratin 8 knockout mice and during the injury and recovery phases of two different models of pancreatitis in wild-type mice. Reg2 was thus proposed to be a cytoprotective protein, the expression of which is regulated by keratin filament organization and phosphorylation (33). We propose that, in pancreatic cells, Reg2 plays a protective role against the development of diabetes and acute pancreatitis. To test its effectiveness, we generated transgenic mice with Reg2 overexpression in pancreatic acinar cells and examined parameters of glucose homeostasis; we challenged the mice with streptozotocin and examined whether Reg2 protects against the development of diabetes or accelerates recovery from diabetes; and we induced acute pancreatitis with caerulein to investigate whether Reg2 has antiapoptotic, anti-inflammatory, or other protective roles.
MATERIALS AND METHODS
Ela-Reg2 transgenic mice.
The transgene was created by insertion of a 530-bp rat elastase I promoter (13) into the pcDNA3.1(−) vector at the HindIII/BamHI site followed by insertion of a 1.9-kb rabbit hemoglobin intron and SV40 polyA tail into the BamHI/XhoI site (9, 17). Then a full-length mouse Reg2 cDNA was inserted into the EcoRI site in the rabbit hemoglobin sequence (29). The DNA construct was confirmed by DNA sequencing, linearized, and used to generate elastase I (Ela)-Reg2 founder mice. The mice were genotyped by PCR, with tail DNA used as the template, as previously described (23). The primers used for transgenic detection were as follows: ElaF (5′-act ttc atg tca cct gtg ct-3′) and Reg2R (5′-cag cga tac aca ata acc acg-3′) (Fig. 1A). The PCR conditions were as follows: 94°C for 3 min followed by 31 cycles of 94°C for 20 s, 57°C for 30 s, and 72°C for 1 min. A control PCR to amplify the endogenous Reg1 gene as a 1.8-kb product was performed using the following primers: Reg1F (5′-tct cat gcc tga tcg tcc-3′) and Reg1R (5′-aac aga gac cca ctg ctc c-3′). The PCR conditions were as follows: 94°C for 3 min followed by 33 cycles at 94°C for 20 s, 57°C for 30 s, and 72°C for 1.5 min. The mice had free access to food and water unless specified for treatments. For breeding, male Ela-Reg2 founder mice were crossed with female C57BL/6 mice purchased from Charles River. The offspring were weaned at 3 wk, their ears were tagged, and an 0.5-cm section of the tail was excised for genotyping. All procedures were approved by the McGill University Animal Care Committee.
Tissue collection and glucose and insulin measurement.
At the end of the experiments, mice were anesthetized with a cocktail of ketamine, xylazine, and acepromazine and then killed by cervical dislocation. Blood was collected and centrifuged at 3,000 rpm for 20 min at 4°C, and serum was stored at −20°C. The pancreas was removed, and 30–40 mg of tissue were homogenized in TRIzol (Invitrogen, Carlsbad, CA) for RNA isolation; half of the remaining tissue was homogenized in protein lysis buffer for Western blotting, and the other half was fixed in 10% formalin for paraffin sections and immunohistochemistry. A glucose tolerance test was carried out on 2.5-mo-old Ela-Reg2 mice and their wild-type littermates, which were fasted for 20 h and then injected with glucose (1 g/kg ip). A OneTouch Ultra glucose meter (LifeScan) was used to measure the glucose level in blood obtained from the tail vein at 0, 15, 30, 60, and 120 min after the injection. Serum insulin was measured using an ELISA kit (ALPCO Diagnostics, Salem, NH).
Streptozotocin (180 mg/kg ip) was prepared fresh in 0.1 M sodium citrate (pH 4.5) and injected into 3- to 4-mo-old male Ela-Reg2 mice and their wild-type littermates. Female mice of the same age were injected with streptozotocin (80 mg/kg ip) for 5 consecutive days. Body weight and tail blood glucose were recorded every 3 days from day 0 to day 16 and then every 1 or 2 wk until 100 days after streptozotocin administration.
Caerulein-induced acute pancreatitis.
After they were fasted for 16 h, 6- to 8-wk-old male Ela-Reg2 and wild-type mice were injected hourly for 7 h with caerulein (50 μg/kg ip) dissolved in normal saline (18); control mice received saline only. At 11 h after the initial injection, the mice were euthanized, and blood and pancreas were collected and processed as described above.
Reverse transcription and quantitative real-time PCR.
Total RNA was isolated from fresh pancreas using TRIzol and measured using a spectrophotometer (model ND-1000, NanoDrop, Wilmington, DE). Reverse transcription was performed using the QuantiTect reverse transcription kit (Qiagen, Valencia, CA). Real-time PCR was performed using the QuantiTect SYBR Green PCR kit (Qiagen) with LightCycler systems (Roche Applied Science, Lewes, UK), and certified Reg2 primers were designed and synthesized by Qiagen (QuantiTect Primer Assay). For each reaction, cDNA synthesized from 20 ng of RNA was used as the template. The relative mRNA level of Reg2 was calculated using the threshold cycle (ΔΔCt) method.
Western blot analysis.
Pancreatic protein was homogenized in lysis buffer (10 mM Tris, pH 7.5, 1 mM EDTA, 150 mM NaCl, 0.1% SDS, 1% Triton X-100, 100 mM NaF, 10 mM sodium pyrophosphate, and 10 mM sodium orthovanadate) containing protease inhibitor cocktail (Roche) and separated by SDS-PAGE. The proteins were transferred to a nitrocellulose membrane, and nonspecific bindings were blocked using a buffer [5% nonfat dry milk in TBST (20 mM Tris, pH 7.5, 150 mM NaCl, and 0.1% Tween 20)]. The membrane was then incubated with rat anti-mouse Reg2 (1:2,000 dilution; R & D Systems, Minneapolis, MN) or rabbit anti-β-actin (1:10,000 dilution; Sigma, St. Louis, MO) overnight at 4°C, washed with TBST, and incubated with horseradish peroxidase-conjugated secondary antibodies (Jackson Immunolabs, West Grove, PA) at room temperature for 1 h. The proteins were detected with enhanced chemiluminensce (GE Healthcare Bio-Sciences, Piscataway, NJ), and the luminescent signal was captured using the Alpha Innotech FluorChem 8900 imaging system. Antibodies from Cell Signaling Technology (Danvers, MA) were used to detect total Akt kinase (Akt1, Akt2, and Akt3) and its phosphorylation at Ser473 (catalog nos. 9272 and 9271, respectively). The antibody against cyclin D1 (M-20; catalog no. sc-718) was obtained from Santa Cruz Biotechnology (Santa Cruz, CA), and the antibody to the p85 subunit of phosphatidylinositol 3-kinase (α-p85) was purchased from Upstate Biotechnology (Lake Placid, NY). Protein levels were quantified by densitometry using AlphaEaseFC software.
Pancreatic sections were dewaxed, rehydrated with ethanol, and permeabilized with 0.1% Triton X-100, and endogenous peroxidase was inactivated with 1% H2O2. The sections were blocked with 10% goat serum in PBS at room temperature for 1 h and incubated with primary antibody against Reg2 (1:100 dilution in 1% goat serum in PBS) overnight at 4°C and then with horseradish peroxidase-conjugated goat anti-rat IgG at room temperature for 3 h. The sections were stained with diaminobenzidine substrate (Vector Labs, Burlingame, CA) and counterstained with hematoxylin for detection of cell nuclei. The pancreas from the pancreatitis experiment was stained with hematoxylin-eosin and rabbit anti-amylase antibody (Sigma). Histological images were captured using a Zeiss microscope and a ×400 or ×630 oil objective.
Data presentation and statistical analysis.
Data are presented as means ± SE and were analyzed by Student's t-test, with statistical significance set at P < 0.05.
Pancreatic acinar-specific overexpression of the Reg2 gene.
To test the effectiveness of Reg2, ectopic expression was driven by acinar-specific rat elastase I promoter in Ela-Reg2 transgenic mice (Fig. 1A). To genotype the mice, we designed a pair of PCR primers (ElaF and Reg2R) to amplify a 1.3-kb fragment of the transgene that was not in the mouse genome. As a control for the quality of the genomic DNA, endogenous Reg1 gene was also amplified by PCR from wild-type and Ela-Reg2 mice as a 1.8-kb band (Fig. 1B). Real-time PCR showed that Reg2 mRNA was drastically increased (75-fold) in the pancreas of Ela-Reg2 mice vs. wild-type littermates (see Fig. 4F). Increased protein level was verified by Western blotting of tissue extracts from the whole pancreas: we detected a moderate (2.8-fold) significant increase in the level of Reg2 protein in transgenic mice that was barely detectable in wild-type littermates (see Fig. 5E). To further confirm cell specificity, in paraffin sections of the pancreas, Reg2 immunostaining was localized in a fraction of acinar cells, but not in the islets or the ductal cells, confirming an acinar-specific overexpression (see Figs. 4G and 5F). Reg2 expression was significant but patchy, similar to that of INGAP transgenic mice (26). This report was derived from data obtained using founder 4, which were subsequently confirmed in founder 1 of Ela-Reg2 mice.
Reg2 overexpression caused no significant change in Akt phosphorylation and levels of phosphoinositide 3-kinase-p85 and cyclin D1.
Of the seven Reg proteins, only the receptor for Reg1 has been characterized (12). As a putative secreted protein, if it acts in a manner similar to Reg1 or INGAP, Reg2 overexpression should affect specific intracellular substrates such as phosphoinositide 3-kinase (PI3K), Akt1, Erk1/2, and cyclin D1. The PI3K/Akt pathway plays a critical role in the proliferation and regeneration of the endocrine and exocrine pancreas (6): an increase in cellular phosphorylated Akt was reported during acinar regeneration in the remnant pancreas following partial pancreatectomy (31), and cyclin D1 was shown to be increased by PI3K/Akt activation in the β-cells (7). In Reg1-induced β-cell regeneration, PI3K/activating transcription factor (ATF)-2/cyclin D1 signaling was activated, and INGAP peptide also increased Akt phosphorylation (1, 25). Class I PI3Ks are heterodimeric enzymes consisting of a 110-kDa catalytic subunit in complex with a regulatory subunit (e.g., p85). To evaluate potential Reg2-induced signals, we performed Western blots on pancreatic proteins extracted from normal untreated wild-type and Ela-Reg2 mice (Fig. 2). We found no change in intracellular p85 (PI3K) level and Akt phosphorylation and a slight increase in cyclin D1 level that did not reach statistical significance.
Reg2 overexpression caused no change in normal islet growth and glucose homeostasis.
As the elastase I gene is only expressed in pancreatic acinar cells starting late in embryonic development (embryonic day 16.5), a Reg2 overexpression driven by its promoter was not expected to cause major changes in general growth and development. Indeed, the litter size was normal, indicating normal fertility, and body weights of the transgenic mice measured at different ages were the same as those of the wild-type littermates (Table 1). Blood glucose level in random-fed and 24-h-fasted mice was not changed; Ela-Reg2 mice also maintained normal insulin concentration (data not shown). Immunohistochemistry of insulin- and hematoxylin-eosin-stained tissue showed no change in the percent ratio of pancreatic islet to whole pancreatic tissues and no change in insulin-positive ratio and morphology of endocrine cells (data not shown). Ela-Reg2 and wild-type mice subjected to a glucose tolerance test displayed no obvious change in the speed of glucose disposal, suggesting a normal rate of insulin secretion upon glucose injection and/or normal glucose uptake/production by insulin target tissues (Fig. 3).
Reg2 overexpression offered no protection against streptozotocin-induced diabetes.
Although streptozotocin only destroyed islet β-cells, it caused significant upregulation of Reg family genes, including Reg2, in acinar cells; Reg2 was believed to be involved in islet cell protection or regeneration (15). To determine whether acinar Reg2 overexpression would protect the islet cells from streptozotocin-induced damage and/or promote islet regeneration after cell death, male Ela-Reg2 and wild-type mice were given a single injection of streptozotocin (180 mg/kg ip) and followed for 3 mo for changes in blood glucose level and body weight. As shown in Fig. 4A, all mice developed significant hyperglycemia only 3 days after the injection, and the blood glucose level continued to rise and reached a peak level between 6 and 9 days and remained at the hyperglycemic plateau for the following 3 mo. Compared with wild-type littermates, Ela-Reg2 mice showed a tendency toward slightly more aggressive hyperglycemia (Fig. 4A). Body weight decreased up to 20% following the development of hyperglycemia and remained lower as hyperglycemia persisted. Compared with wild-type mice, weight loss in Ela-Reg2 mice was slightly greater, similar to the slightly more severe hyperglycemia (Fig. 4B). Within 3 mo, without other treatment, there was no sign of recovery (through islet regeneration) from diabetes in wild-type or Ela-Reg2 mice. Thus, unlike INGAP (Reg3δ), acinar-specific Reg2 expression did not protect against streptozotocin-induced diabetes; if anything, excessive Reg2 protein in the acinar tissues may be slightly detrimental to the pancreatic islets in the face of streptozotocin-induced diabetes.
Similar results were obtained from female mice treated with multiple low-dose injections of streptozotocin, which induced a slower onset (9 days) of hyperglycemia and a peak response from 21 days. Weight loss was <10% in Ela-Reg2 mice and their wild-type littermates, displaying no difference caused by Reg2 overexpression (Fig. 4, C and D). As a result of streptozotocin-induced β-cell death, serum insulin levels in Ela-Reg2 and wild-type mice decreased to a similar 38% of that in untreated wild-type mice (Fig. 4E).
Lack of protection against streptozotocin-induced diabetes might be caused by insufficient Reg2 production. To compare the levels of Reg2 expression caused by transgenic overexpression and streptozotocin induction, we performed real-time PCR (Fig. 4F) and immunohistochemistry (Fig. 4G). As reported previously (15, 19), streptozotocin increased Reg2 mRNA by 2.1-fold, which is far less than the transgenic overexpression of 51- to 75-fold in Ela-Reg2 mice (Fig. 4F). The increase in the level of Reg2 protein revealed using immunohistochemistry was not as significant. In untreated wild-type pancreas, Reg2 was hardly visible; both transgenic Ela-Reg2 expression and streptozotocin treatment caused significant, yet patchy, expression of Reg2 in some acinar cells (Fig. 4G). Thus the level of protein staining indicated that Reg2 overexpression was at least comparable to that induced by streptozotocin and the level of mRNA indicated much greater Reg2 overexpression.
Reg2 overexpression offered no protection against caerulein-induced pancreatitis.
The expression of Reg2 in acinar cells has been reported to be greatly induced during acute pancreatitis and the recovery phase (33). However, the precise role of Reg2 in promoting or preventing pancreatitis is not known. We induced acute pancreatitis in Ela-Reg2 mice by seven hourly injections of caerulein (50 μg/kg ip). At 11 h after the initiation of injection, serum amylase level increased more than ninefold compared with saline-injected control mice (Fig. 5A). In Ela-Reg2 mice, the increase in serum amylase level was not different from that in wild-type littermates, indicating that acinar-specific expression of Reg2 did not decrease the severity of caerulein-induced acute pancreatitis.
Histological analysis revealed acinar swelling, necrosis, interstitial neutrophil infiltration, and decreased amylase content in the pancreas of Ela-Reg2 and wild-type mice, confirming the severity of pancreatitis (Fig. 5, B and C). 1) In untreated wild-type pancreas, hematoxylin staining of the nuclei was strong and clear, and the acinar structure was intact, with rich eosin staining (Fig. 5B1); amylase staining was patchy but strong (Fig. 5C1). 2) In the pancreas of caerulein-treated wild-type mice, the eosin-stained acini were swollen due to edema and vacuolized by lysis (white arrow), and massive neutrophil infiltration was observed (Fig. 5B2, black arrows marked neutrophil nuclei divided into 2–5 lobes, indicating inflammation); there was a significant decrease in amylase level, and the nuclei were disappearing, indicating necrosis (Fig. 5C2). 3) In caerulein-treated Ela-Reg2 pancreas, the acinar structure and cell nuclei were partially dissolved (necrosis), with obvious inflammation (Fig. 5B3, black arrows); amylase staining was further decreased, and necrosis was seemingly more severe than in wild-type mice (Fig. 5C3). The seemingly decreased neutrophil infiltration and inflammation in Ela-Reg2 pancreas was perhaps the consequence of more necrotic cell death, which prevented further inflammation. The histological findings confirmed pancreatitis in wild-type and Ela-Reg2 mice; the latter seemed to have more severe necrosis, indicating that Reg2 overexpression was not protective and, perhaps, even detrimental to the pancreas during pancreatic damage.
Although necrosis was the dominant feature of pancreatitis, acinar cell apoptosis can be detected in the pancreatic sections of wild-type and Ela-Reg2 mice by cell shrinkage, nuclear fragmentation, or chromatin condensation (red arrows in Fig. 5B2 and 5B3). Cleavage of poly(ADP-ribose) polymerase (PARP) occurs downstream of caspase activation and is an indicator of apoptosis. Using Western blotting, we confirmed a slight activation (cleavage) of PARP in pancreatic extracts prepared from caerulein-treated vs. untreated mice (Fig. 5D). Immunohistochemistry and Western blotting showed barely detectable activated caspase-3 in the pancreas; thus no difference between transgenic and wild-type mice was observed (data not shown). These data support the notion that Reg2 overexpression in acinar pancreas did not protect against caerulein-induced apoptosis and acute pancreatitis.
Lack of protection against caerulein-induced pancreatitis might also be caused by insufficient Reg2 production. To compare the levels of Reg2 expression caused by transgenic overexpression and caerulein induction, we used Western blotting (Fig. 5E) and immunohistochemistry (Fig. 5F). Caerulein injections increased pancreatic Reg2 protein 2.8-fold in wild-type mice; untreated Ela-Reg2 pancreas also exhibited a 2.8-fold increase in Reg2 protein level vs. wild-type mice; there was no further increase after the onset of pancreatitis (Fig. 5E). Immunohistochemistry showed a significant caerulein-induced increase in Reg2 protein level in wild-type mice; the increase caused by transgenic expression seemed even greater; the combination of caerulein and transgenic overexpression resulted in the greatest increase in Reg2 level (Fig. 5F). Thus the protein level of Reg2 overexpression was not less than that induced by caerulein, and we have very little reason to suspect that the negative result is due to insufficient Reg2 production.
This study established transgenic, acinar-specific Reg2 overexpression. The level of Reg2 expression in acinar cells, confirmed by real-time PCR, immunohistochemistry, and Western blotting, was significantly higher in transgenic than normal wild-type mice and not less than the reported level of INGAP expression induced by the same elastase promoter (26) and comparable with or greater than that induced by streptozotocin and pancreatitis in wild-type mice (15). Using Ela-Reg2 mice, we have made the following observations. 1) Reg2 overexpression in acinar cells did not affect normal development of the islet cells, β-cell mass, insulin and glucose levels, and normal glucose tolerance. 2) Acinar-specific expression of Reg2 offered no protection against streptozotocin-induced diabetes and did not accelerate the recovery from diabetes. 3) Acinar-specific Reg2 gene expression offered no protection against caerulein-induced pancreatitis, including necrosis and inflammation. If anything, Reg2 overexpression was rather (albeit slightly) detrimental to the endocrine and exocrine pancreatic tissues under those experimental conditions.
We used a rat elastase I promoter to drive Reg2 overexpression in pancreatic acinar cells in mice and achieved sustained Reg2 expression. In normal mouse pancreas, Reg2 protein was barely detectable in Western blots (Fig. 5E) or by immunohistochemistry (Figs. 4G and 5F), and Reg2 protein and mRNA levels were upregulated in the pancreas of streptozotocin-treated diabetic mice (Fig. 4, F and G) (15). In NOD mice made transgenic for β-cell-specific IFNβ overexpression, Reg2 mRNA was upregulated in microdissected islets; the level was further increased after the mice developed diabetes; and Reg2 was colocalized with insulin staining, suggesting that Reg2 might act as an autoantigen in the β-cells to elicit immunological attacks (21). In our model, under normal conditions, excessive Reg2 protein in acinar cells had no effects on normal islet development and function, as revealed by histology, or on insulin levels and glucose tolerance, in contrast to INGAP transgenic mice (26).
Next, we tested the effect on streptozotocin-induced diabetes. We used two doses of streptozotocin: a single injection at 180 mg/kg in male mice and injections at 80 mg/kg for 5 days in female mice. Both doses successfully induced a rapid onset of hyperglycemia. Hyperglycemia appeared to be slightly more severe in Ela-Reg2 mice than their wild-type littermates; correspondingly, Ela-Reg2 mice also displayed a tendency toward more body weight loss. Our data suggest that Reg2, unlike other Reg proteins such as Reg1 and INGAP, does not protect against streptozotocin-induced diabetes or promote islet regeneration after β-cell damage. Reg2 has been proposed to serve as a β-cell autoantigen; vaccination of NOD mice with full-length Reg2 or the COOH-terminal portion of Reg2 delayed the development of diabetes, whereas vaccination with the NH2-terminal portion of Reg2 accelerated the disease (10). However, the effect of Reg2 on streptozotocin-induced diabetes has not been examined. Streptozotocin causes DNA strand breaks, with subsequent activation of PARP and depletion of NAD+ leading to β-cell death; deletion of PARP resulted in mice that were resistant to streptozotocin-induced diabetes (3, 16, 20). On the basis of our results, it is unlikely that acinar-derived Reg2 could protect the β-cells or stimulate islet neogenesis to protect against streptozotocin-induced diabetes.
In contrast, studies of Reg1 knockout and transgenic mice (27, 28) have shown that Reg1 promotes β-cell proliferation and protects against diabetes. Islets isolated from Reg1-deleted mice showed decreased incorporation of [3H]thymidine into cellular DNA, whereas β-cell-specific overexpression of Reg1 increased DNA synthesis. Furthermore, transgenic Reg1 overexpression in β-cells in NOD mice delayed the spontaneous onset of diabetes, supporting the idea that Reg1 has a role in islet growth/neogenesis. Further study revealed that Reg1 activated PI3K and ATF-2, a transcription factor that increased the level of cyclin D1 and caused β-cell proliferation (14, 25). A putative cell membrane receptor has been cloned from the rat islets to mediate Reg1 effects (12). Another Reg protein, INGAP, was induced in acinar cells after partial pancreatic duct obstruction in hamsters (22). The full-length INGAP protein or a 15-amino acid peptide increased the proliferation of cultured hamster duct epithelium and rat pancreatic duct cells, but not a β-cell line or human islets, suggesting that INGAP has a role in stimulating islet neogenesis from pancreatic duct cells (22). In a mechanism similar to Reg1, INGAP was known to activate PI3K, Akt1, and Erk1/2; the latter two elements caused activation of PDX-1 and neurogenin-3; Akt1 also increased the level of cyclin D1, collectively causing β-cell replication or neogenesis (14). The INGAP peptide stimulated islet neogenesis and reversed streptozotocin-induced hyperglycemia (24). In acinar-specific INGAP transgenic mice, β-cell mass and pancreatic insulin content were significantly increased; the mice were resistant to streptozotocin-induced diabetes and remained normoglycemic (26). It was suggested that the altered physiological state (altered gene expression and increased β-cell mass) provided increased protection against β-cell damage. Our Reg2 overexpression did not seem to have affected PI3K, Akt, and cyclin D1 levels.
Acute pancreatitis is a life-threatening inflammation of the pancreas that leads to hyperamylasemia, pancreatic edema, and acinar cell vacuolization. Acting as a nonselective cholecystokinin antagonist, excessive caerulein has been known to induce acute necrotic pancreatitis in animals (18). Reg protein pancreatitis-associated protein/hedgehog-interacting protein/Reg3β was shown to play antiapoptotic and anti-inflammatory roles in caerulein-induced pancreatitis. Reg3β knockout mice exhibited increased sensitivity to apoptosis during the course of pancreatitis, as shown by increased levels of activated caspase-3 and cleaved PARP protein in the pancreas compared with wild-type mice; pretreatment with recombinant Reg3β protein prior to pancreatitis reversed the effects and protected the pancreas (8). The knockout pancreas also showed more inflammatory infiltration and higher levels of expression of inflammatory cytokines, including TNFα, IL-6, and IL-1β; both parameters were significantly reduced when the mice were pretreated with recombinant Reg3β, further supporting an anti-inflammatory role (8). During pancreatitis, acinar cells are injured due to necrosis, apoptosis, and inflammation. Reg3β was shown to be antiapoptotic by inhibiting caspase-3 and PARP activations and anti-inflammatory by inhibiting cytokine expression. Similar to Reg3β, Reg2 has also been proposed to be protective against pancreatitis. In two models of acute pancreatitis, caerulein- and choline-deficient diet-induced, the Reg2 gene was markedly upregulated during the injury, as well as the recovery, phase, suggesting that this protein may play a role (33). In this study, we induced acute pancreatitis by 7 hourly injections of caerulein and observed a nearly ninefold increase in serum amylase level; however, there was no difference in the severity of pancreatitis between the Ela-Reg2 transgenic and wild-type mice. Similarly, a marked increase in Reg2 expression was detected in caerulein-treated wild-type compared with saline-treated pancreas, whereas no further increase was found in caerulein-treated Ela-Reg2 mice. Hematoxylin-eosin staining of paraffin-embedded pancreatic sections exhibited a similar degree of inflammatory infiltration. Our recent data imply that Reg2 may play an antiapoptotic role (L. Liu et al., unpublished observations), which prompted us to examine the activated PARP level in the pancreas by Western blotting, as previously reported (8). The result of this in vivo study seemed to have disproved a protective effect; the role of increased Reg2 expression during acute pancreatitis remains unclear. It is to be noted that the level of Reg2 expression achieved in our Ela-Reg2 pancreas was still in the range of levels that can be induced during acute pancreatitis (Fig. 5, E and F). One cannot exclude the possibility that much further increased Reg2 expression (in a more robust overexpression) may become more effective and protect the acinar cells from pancreatitis or accelerate the deterioration.
In summary, we have created transgenic mice overexpressing Reg2 in acinar cells in the pancreas. The mice appeared normal, with glucose homeostasis that did not differ from wild-type mice. When we induced diabetes with streptozotocin, excessive Reg2 protein in acinar cells did not offer protection against diabetes, nor did it accelerate the recovery from diabetes, in contrast to the effects of INGAP in a similar model. Furthermore, in caerulein-induced acute pancreatitis, Reg2 overexpression did not ameliorate, nor did it significantly deteriorate, the course of pancreatic destruction. Our results suggest a subtype-specific function of Reg family proteins in diabetes and pancreatitis, and, in contrast to Reg1, INGAP, and Reg3β, it is unlikely that acinar-expressed Reg2 is an applicable therapeutic target.
This work was supported by Canadian Institutes of Health Research Operating Grants MOP-84389 and CCI-85675. J.-L. Liu was a senior Chercheurs-Boursier of Fonds de la Recherché en Santé Quebec. B. Li received fellowship support from the Research Institute of McGill University Health Centre. X. Wang was supported by National Natural Science Foundation of China Grant 30700382.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Ela-Reg2 transgenic microinjection was performed by Dr. Qinzhang Zhu and Michel Robillard (Institut de Recherches Cliniques de Montréal and Quebec Transgenic Network). Rat elastase 1 promoter, pKS-RIP/globin vector, and mouse Reg2 cDNA were kindly provided by Drs. G. H. Swift (University of Texas Southwestern Medical Center, Dallas, TX), E. Riu (Universitat Autonoma Barcelona, Spain), and S. Takasawa (Nara Medical University Japan), respectively.
- Copyright © 2010 the American Physiological Society