Glucagon-Like Peptide 2 Dose-Dependently Activates Intestinal Cell Survival and Proliferation in Neonatal Piglets
http://www.100md.com
内分泌学杂志 2005年第1期
United States Department of Agriculture, Agricultural Research Service, Children’s Nutrition Research Center (D.G.B., B.S., X.G., L.C., X.C.), Department of Pediatrics, Baylor College of Medicine, Houston, Texas 77030; and Department of Medical Physiology (J.J.H.), University of Copenhagen, DK-2200 Copenhagen, Denmark
Address all correspondence and requests for reprints to: Douglas G. Burrin, Ph.D., Children’s Nutrition Research Center, 1100 Bates Street, Houston, Texas 77030. E-mail: dburrin@bcm.tmc.edu.
Abstract
Glucagon-like peptide 2 (GLP-2) is a gut hormone that stimulates mucosal growth in total parenteral nutrition (TPN)-fed piglets; however, the dose-dependent effects on apoptosis, cell proliferation, and protein synthesis are unknown. We studied 38 TPN-fed neonatal piglets infused iv with either saline or GLP-2 at three rates (2.5, 5.0, and 10.0 nmol·kg–1·d–1) for 7 d. Plasma GLP-2 concentrations ranged from 177 ± 27 to 692 ± 85 pM in the low- and high-infusion groups, respectively. GLP-2 infusion dose-dependently increased small intestinal weight, DNA and protein content, and villus height; however, stomach protein synthesis was decreased by GLP-2. Intestinal crypt and villus apoptosis decreased and crypt cell number increased linearly with GLP-2 infusion rates, whereas cell proliferation and protein synthesis were stimulated only at the high GLP-2 dose. The intestinal activities of caspase-3 and -6 and active caspase-3 abundance decreased, yet procaspase-3 abundance increased markedly with increasing infusion rate and plasma concentration of GLP-2. The GLP-2-dose-dependent suppression of intestinal apoptosis and caspase-3 activity was associated with increased protein kinase B and glycogen-synthase kinase-3 phosphorylation, yet the expression phosphatidylinositol 3-kinase was unaffected by GLP-2. Intestinal endothelial nitric oxide synthase mRNA and protein expression was increased, but only at the high GLP-2 dose. We conclude that the stimulation of intestinal epithelial survival is concentration dependent at physiological GLP-2 concentrations; however, induction of cell proliferation and protein synthesis is a pharmacological response. Moreover, we show that GLP-2 stimulates intestinal cell survival and proliferation in association with induction of protein kinase B and glycogen-synthase kinase-3 phosphorylation and Bcl-2 expression.
Introduction
GLUCAGON-LIKE PEPTIDE 2 (GLP-2) is a 33-amino-acid peptide derived from differential posttranslational processing of proglucagon in intestinal endocrine L-cells (1, 2). GLP-2 secretion is stimulated in response to food intake, and the circulating concentration is significantly reduced by total parenteral nutrition (TPN) (3, 4, 5). In parenterally fed neonatal piglets given increasing proportions of enteral nutrition, we observed a strong positive correlation between the circulating GLP-2 concentration and intestinal mucosal growth (3). Similarly, recent studies in rats showed that the intestinal adaptive trophic response to small bowel resection was positively correlated with circulating GLP-2 and colonic proglucagon gene expression (6). Results from these two different models of intestinal adaptation suggest that the circulating GLP-2 concentration is an important physiological stimulus of mucosal growth. Consistent with this idea, inducing supraphysiological circulating GLP-2 concentrations by infusion of exogenous GLP-2 can prevent TPN-induced mucosal atrophy (7, 8), yet it is unknown whether this response occurs within the physiological range of circulating GLP-2.
The biological effects of GLP-2 are mediated via activation of a G protein-linked, membrane receptor (GLP-2R) expressed mainly in the gastrointestinal tract and brain (9, 10, 11). The dominant biological effect of GLP-2 is stimulation of small intestinal mucosal growth, which is associated with reduced apoptosis and proteolysis and increased protein synthesis in TPN-fed piglets (7) and with increased cell proliferation and reduced apoptosis in chow-fed mice (12, 13, 14). In addition, trophic effects have been observed in the stomach and colon (15, 16); however, there is limited information as to whether GLP-2 affects growth and metabolism in non-gastrointestinal tissues in neonates. The cellular mechanisms of GLP-2 action subsequent to binding and activation of the GLP-2R are poorly defined and complicated by uncertainty of the specific cellular localization. Conflicting reports have demonstrated the presence of the GLP-2R in human enteroendocrine cells (10) and in murine enteric neurons (17). Studies in GLP-2R-transfected fibroblasts have demonstrated that GLP-2 induces cAMP production, immediate early gene expression, and cell proliferation (18). Subsequent studies with this model showed that GLP-2 prevents apoptosis via protein kinase A-dependent phosphorylation of Bad and glycogen-synthase kinase-3 (GSK-3) and downstream inhibition of caspase-3 activity but does not involve phosphatidylinositol 3-kinase (PI3-kinase) or MAPK pathways (19, 20). In contrast, studies with a human colon carcinoma cell line (Caco-2) have indicated that GLP-2 increases cell proliferation in association with a transient increase in ERK phosphorylation, and the response is suppressed by inhibitors of PI3-kinase and MAPK (21). Despite the caveats of these GLP-2 studies in transformed cell lines, a recent report with isolated primary intestinal cells has shown that the GLP-2R is expressed and that GLP-2 induced cAMP production and [3H]thymidine incorporation (22). Whether any of these aforementioned signaling pathways are activated by GLP-2 in vivo is unknown. In addition, our recent study in TPN-fed piglets suggests that the GLP-2 stimulation of intestinal blood flow is NO dependent and associated with increased endothelial nitric oxide synthase (eNOS) expression, possibly implicating nitric oxide in other downstream GLP-2R signaling events (14), such as cell apoptosis and protein metabolism (23, 24).
The aim of the current study was to determine whether the intestinal trophic actions of GLP-2 are dose dependent in TPN-fed piglets given GLP-2 infusion rates that produce circulating GLP-2 concentrations within the physiological and pharmacological range. In investigating the GLP-2 dose response, we quantified several endpoints of intestinal growth and also sought to establish whether any of the cellular signaling proteins that are activated by GLP-2 in cultured cells are also increased in the intestine in response to GLP-2 treatment in vivo.
Materials and Methods
Animals, surgery, and study design
The study protocol was approved by the Animal Care and Use Committee of Baylor College of Medicine and was conducted in accordance with the Guide for the Care and Use of Laboratory Animals [Department of Health and Human Services publication no. (NIH) 85-23, revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD]. Pregnant, crossbred sows were obtained from the Texas Department of Criminal Justice (Huntsville, TX) approximately 1 wk before estimated date of parturition. Sows were housed, fed, and kept under surveillance in the animal facility at the Children’s Nutrition Research Center (Houston, TX) until delivery. At the time of parturition, vaginally born newborn piglets were removed from the sow to prevent colostrum ingestion and immediately placed in heated cages (30 C) and given water until surgery.
The piglets were surgically catheterized within 12 h of birth under isoflurane general anesthesia. Silastic catheters were inserted into the jugular vein and carotid artery via a procedure as previously described (3). Catheters were secured in a jacket that was attached to a tether, which allowed free movement and secure administration of TPN to the piglets in their cages. Pre- and postoperatively on each day, piglets received enrofloxacin (2.5 mg·kg–1; Bayer, Shawnee Mission, KS). Postoperatively, each piglet received one dose of analgesic (0.1 mg·kg–1 butorphenol tartrate, Fort Dodge Labs, Fort Dodge, IA).
Piglets were administered TPN daily that provided 240 ml fluid, 25 g glucose, 13 g amino acids, and 5 g lipid per kilogram body weight for 7 d; piglets were weighed daily to adjust the TPN infusion rate. The parenteral nutrient solution consisted of dextrose (104 g/liter), a complete amino acid mixture (55 g/liter;), lipid (21 g/liter; Intralipid 20%; Fresenius Kabi, Bad Homburg, Germany), electrolytes, trace minerals, and vitamins as described previously (3). The nutrient solution was administered continuously at 50% the full rate during the initial 24-h period after surgery and thereafter increased to 100%. Within 6 h of surgery, the piglets were randomly assigned to one of four treatment groups; a control group received TPN plus a continuous iv infusion at 0.5 ml·kg–1·h–1 of 0.45 g/liter NaCl containing 0.1% human serum albumin (Bayer Corp., Elkhart, IN) or the same infusion rate of 0.45 g/liter NaCl containing 0.5% human serum albumin containing human GLP-2 (1–33) (California Peptide Research, Inc., Napa, CA) at one of three peptide concentrations (0.8, 1.6, or 3.6 μg/ml). The approximate daily molar infusion rates were 2.5, 5.0, and 10.0 nmol·kg–1 for the low, medium, and high GLP-2 treatment groups, respectively. As with the TPN solution, infusion rates of control, and GLP-2 solutions were adjusted daily based on body weight. A total of 38 piglets were studied in the four groups: TPN (n = 10), low GLP-2 (n = 7), medium GLP-2 (n = 8), and high GLP-2 (n = 13).
After 6 d of full TPN or TPN plus GLP-2 treatment, piglets received an iv bolus of bromodeoxyuridine (BrdU) at 50 mg/kg body weight (Sigma Aldrich, St. Louis, MO) 8 h before the pigs were killed to estimate crypt cell proliferation (see below). In addition, a bolus dose of [13C]phenylalanine (1.5 mmol/kg phenylalanine containing 0.15 mmol/kg [13C6]phenylalanine) (Cambridge Isotope Laboratories, Andover, MA) was given 30 min before pigs were killed to measure the rate of tissue protein synthesis. Pigs were killed with a venous injection of pentobarbital sodium (50 mg/kg body weight) and sodium phenytoin (5 mg/kg body weight) (Beuthanasia-D, Schering-Plough Animal Health, Kenilworth, NJ). The abdomen was opened, and the whole small intestine was excised and quickly flushed with ice-cold saline, weighed, and divided into parts of equal length, designated as jejunum and ileum. Samples of jejunum were placed in formalin for morphological and BrdU analysis. The liver, spleen, and stomach were removed and weighed. Samples of liver, spleen, stomach, and hind limb skeletal muscle were also collected and with intestinal tissues were frozen in liquid nitrogen and used for subsequent measurements.
Morphometry, cell proliferation, and apoptosis
Morphometry analysis of intestinal mucosal tissue was performed on formalin-fixed, hematoxylin- and eosin-stained sections as described previously (3, 25). In vivo crypt cell proliferation was measured as described previously (3). BrdU-labeled cells were detected by immunohistochemistry in formalin-fixed, paraffin-embedded sections and expressed as a percentage of total nuclei per crypt observed in approximately 15–20 well-oriented crypt sections from two to three tissue sections from each animal. Measurements of apoptosis were made based on cell morphology observed in x400 images by a single, trained observer that was blinded of the treatments. The apoptotic cells were characterized by evidence of condensed irregular chromatin, nuclear fragmentation, and intensely eosinophilic cytoplasm as shown previously (25). Apoptotic cells were expressed as a percentage of the total epithelial cell numbers in the villus and crypt compartment of the same section; approximately 1000–1500 total epithelial cells were counted from two to three tissue sections per animal. Tissue samples were assayed for DNA content as previously described (3).
In vivo protein synthesis and mass spectrometry
Samples of jejunum, liver, spleen, and stomach tissue were homogenized and deproteinized with 2 M perchloric acid, and the perchloric acid-soluble (tissue-free pool) and acid-insoluble (protein-bound pool) fractions were subjected to mass spectrometric analysis similar to that described previously (7, 26). The acid-insoluble fraction was hydrolyzed with 6 N HCl for 24 h before gas chromatography mass spectrometry analysis. The isotopic enrichment of [U-13C]phenylalanine (M+6 isotopomer) in the two tissue pools was determined by gas chromatography mass spectrometry analysis of the n-propyl ester heptafluorobutyramide derivative using methane-negative chemical ionization. The analyses were performed with a 5890 series II gas chromatograph linked to a model 5989B (Hewlett-Packard, Palo Alto, CA) quadrupole mass spectrometer. The isotopic enrichment of phenylalanine was determined by monitoring ions at a mass-to-charge ratio of 383 to 389. Protein synthesis was calculated as described previously as the fractional protein synthesis rate (FSR, %/d): FSR = (IEbound/IEfree) x (1440/t) x 100, where IEbound and IEfree are the isotopic enrichments (mol% excess) of [13C6]phenylalanine of the perchloric acid-insoluble (protein-bound) and perchloric acid-soluble (tissue-free) pool, t is the time of labeling (in minutes), and 1440 is the number of minutes in a day. Tissue samples were assayed for protein using the BCA method (Pierce, Rockford, IL).
Caspase-3 and -6 activity
Intestinal tissue caspase activities were measured with EnzChek caspase-3 assay kit (Molecular Probes, Inc., Eugene, OR), based on the incubation of caspase-3 [Z-DEVD-(Asp-Glu-Val-Asp)] and caspase-6 [Ac-VEID-(Ac-Val-Glu-Ile-Asp)] (Calbiochem, San Diego, CA) specific substrates conjugated to the 7-amino-4-methylcoumarin compound, which yields a fluorescent product upon cleavage. Frozen tissue (100 mg) was homogenized in 1.5 ml buffer containing 50 mM Tris/HCl (pH 8.0), 25 mM MgCl2, and 0.1 mM phenylmethylsulfonyl fluoride. The homogenate was centrifuged at 5000 x g for 15 min at 4 C. Extracts (50 μl) were mixed with 50 μl of working solution and applied to micro-well plates and incubated at room temperature (covered from light) for 30 min (caspase-3 assay) or at 37 C for 45 min (caspase-6 assay). Fluorescence readings were obtained at 5-min intervals for 45 min at 441 nm in a SPECTRAmax Gemini XS Microplate Spectrophotometer (Molecular Devices). The linearity of the reaction was confirmed, and the time period between 5 and 20 min was used for caspase-6 and the time period between 5 and 30 min for caspase-3. Activity rates were corrected for extract protein concentration determined using the BCA method with albumin as a standard.
Western blotting analysis
Caspase-3 and Bcl-2.
For all Western blots, frozen intestinal tissue samples (100 mg) were homogenized in 50 mM HEPES buffer (pH 7.4) containing 1 mM EDTA, 1 mM dithiothreitol, 5 mg/liter phenylmethylsulfonyl fluoride, 5 mg/liter aprotinin, 5 mg/liter chymostatin, and 5 mg/liter pepstatin. The homogenate was then sonicated and centrifuged at 12,000 x g for 15 min at 4 C. Equal amounts (30–120 μg) of supernatant protein extracts were separated on a 15% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk in the Tris-buffered saline (TBS). Membranes were incubated with a primary antibody [rabbit polyclonal antibody against human caspase-3, H-277, 1:1000; and rabbit polyclonal antibody against human Bcl-2, C21, 1:200 (Santa Cruz Biotechnology, Santa Cruz, CA)] diluted in the 5% nonfat milk in TBS with added Tween 20 solution (0.1%). Membranes were incubated with a secondary antibody [goat antirabbit IgG-horseradish peroxidase (HRP), 1:5000 (Santa Cruz Biotechnology)]. The molecular mass of caspase-3 precursor is 34 kDa, and the active form is 19 kDa, whereas that of Bcl-2 is 29 kDa.
PI3-kinase, protein kinase B (PKB), GSK-3.
Equal amounts (30–120 μg) of supernatant protein extracts were separated on a 9% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk in the TBS. Membranes were incubated with a primary antibody (mouse monoclonal antibody against human PI3-kinase, 1:1000 (MBL International Corp., Watertown, MA); mouse monoclonal antibody against human PKB1, 1:500; and mouse monoclonal antibody against Xenopus GSK-3?, 1:1000 (Santa Cruz Biotechnology)], diluted in the 5% nonfat milk in TBS with added Tween 20 solution (0.1%). Membranes were incubated with a secondary antibody [goat antimouse IgG1-HRP, 1:5000 (Santa Cruz Biotechnology)]. The molecular mass of PI3-kinase is 85 kDa, PKB/Akt is 60 kDa, GSK-3 is approximately 51/46 kDa, and phospho-GSK-3 is 46 kDa.
Phosphorylation of GSK-3 and PKB in intestinal tissue was measured as follows. Tissue was extracted as described above, except with added phosphatase inhibitor, sodium orthovanadate, to a final concentration of 2 mM. For phosphor-GSK-3, equal amounts (30–120 μg) of supernatant protein extracts were separated on a 9% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk in the TBS. Membranes were incubated with a primary antibody (rabbit polyclonal antibody recognizes both phospho-GSK-3/?, Ser21/Ser9) (Santa Cruz Biotechnology) and then secondary antibody as described above for GSK-3 protein. PKB-Ser473 and PKB-Thr308 were measured by Western blotting after immunoprecipitation with anti-PKB agarose beads. Extracts were combined with 4 μg of anti-PKB, PH Domain, and agarose (mouse monoclonal IgG; Upstate, Charlottesville, VA) and incubated overnight at 4 C. Agarose beads were centrifuged at 12,000 rpm, washed three times with ice-cold 1x PBS, resuspended in 60 μl 2x Laemmli sample buffer, and boiled for 5 min. Ten microliters of immunoprecipitation product (roughly 160 μg of total protein) were separated on a 9% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk in the TBS. Membranes were incubated with a primary antibody [rabbit anti-phospho-PKB1-S473, 1:500 (R&D Systems, Minneapolis, MN), and rabbit anti-phospho-PKB1-T308, 1:500 (Cell Signaling Technology, Inc., Beverly, MA)] diluted in TBS with added Tween 20 solution (0.1%). Membranes were incubated with a secondary antibody [goat antirabbit IgG1-HRP, 1:5000 (Santa Cruz Biotechnology)]. The membranes were then stripped (Restore Western blot stripping buffer, Pierce) at 37 C for 15 min, washed with TBS with added Tween 20 solution (0.1%) three times, and reprobed with mouse anti-PKB1 antibody (1:1000; Santa Cruz Biotechnology) diluted in 5% milk plus TBS with added Tween 20 solution (0.1%). The molecular mass of PKB is 60 kDa. The quantity of the two PKB-Ser473 and PKB-Thr308 forms was expressed relative to the amount of PKB protein on each membrane.
eNOS.
Equal amounts (100 μg) of supernatant protein extracts were separated on a 7.5% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 10 ml Pierce Superblock solution and then incubated with a primary antibody [rabbit polyclonal antibody against human eNOS (H-159) (Santa Cruz Biotechnology)] diluted 1:500 in the TBS with added Tween 20 solution (0.1%). The apparent molecular mass of the eNOS band was 124 kDa. Membranes were incubated with a secondary antibody [antirabbit IgG conjugated with biotin, 1:25,000 (Santa-Cruz Biotechnology)] enhanced with Neutravidin-HRP (1:25,000; Pierce).
All Western blots were allowed to react with HRP substrate (ECL-plus, Amersham Biosciences, Piscataway, NJ) and then exposed to x-ray film for 30–120 sec, and the image was scanned and quantified by ImageQuant 5.0 software (Molecular Dynamics, Amersham Biosciences, Sunnyvale, CA). All Western blots were run with six pigs from each treatment group and used for statistical analysis, and treatment means and SEs are shown as bar graphs. Western blots of pooled samples from six pigs in each treatment also are shown in each figure for the respective proteins.
Real-time RT-PCR of eNOS mRNA.
Total RNA was extracted from the frozen porcine jejunal tissue samples using RNeasy Mini kit (QIAGEN Inc., Valencia, CA). After the RNA concentration was quantified, approximately 25–50 ng of total RNA for each reaction was used for real-time qRT-PCR. Primers and probe of porcine eNOS for real-time RT-PCR was based on the sequence of porcine eNOS mRNA (GenBank accession no. AY266137), i.e. using eNOS TaqMan TAMRA probe (100 nM), 6FAM-CTT CAC CGC GTT GGC CAC TTC CT-TAMRA; eNOS forward primer (10 mM), GGC ATC GCC AGA AAG AC; and eNOS reverse primer (10 mM), CAT CAC GGT GCC CAT GAG T. Primers and probe of ribosomal RNA (18S rRNA, Applied Biosystems, Foster City, CA) were used as an internal control. Assays were performed in triplicate with an ABI Prism 7700 sequence detector (Applied Biosystems). Data were normalized to 18S rRNA.
GLP-2 RIA
After 6 d of full TPN or TPN plus GLP-2 treatment and before bolus injections of BrdU and [13C]phenylalanine, blood samples were collected in EDTA tubes and centrifuged at 3000 x g at 4 C, and plasma was frozen immediately in liquid nitrogen. Plasma GLP-2 (1–33) concentrations were quantified by RIA as described previously (3, 14). This assay recognizes both the human and porcine GLP-2 peptides. The clearance rates for plasma GLP-2 were calculated using the following equation, CR = IR/Cp, where CR is the clearance rate (liter·kg–1·h–1), IR is the GLP-2 infusion rate (pmol·kg–1·h–1), and Cp is the steady-state plasma GLP-2 concentration (pmol/liter).
Statistical analysis
Data for the four treatment groups were analyzed using Minitab statistical software (Minitab Inc., State College, PA). Data were first analyzed by one-way ANOVA with GLP-2 infusion dose as a main effect, followed by a Tukey’s means comparison test. Means comparisons were done specifically to test for statistical differences between the control group and the three GLP-2 infusion rate groups. In some cases, data were analyzed using multiple regression analysis, and best-fit lines were generated using Origin software program (Microcal Software Inc., Northampton, MA). Because of the general lack of response to GLP-2, measurements of tissue mass and protein synthesis in the stomach, liver, spleen, and muscle tissue were obtained only for control and high GLP-2 groups; these data were analyzed by ANOVA using Minitab statistical software. Results are expressed as means and their respective SEs, and a P value less than 0.05 was considered statistically significant.
Results
There were no significant differences among the four treatment groups in either final body weight or daily body weight gain. The mean final body weights (kg) were 2.49 ± 0.16 and 2.64 ± 0.14, and the daily body weight gains (g/d) were 60.0 ± 4.5 and 69.7 ± 3.9 in the control and high GLP-2 groups, respectively. An important objective in this study was to target the GLP-2 infusion rates to achieve circulating plasma GLP-2 concentrations within the physiological range and the higher pharmacological range based on previously reported studies. The plasma GLP-2 (1–33) concentrations (pmol/liter) in all three GLP-2 treatment groups were significantly higher than in the TPN control group (31.5 ± 4.0) (Fig. 1). The mean plasma GLP-2 concentration in the low GLP-2 infusion group (166 ± 25) was higher than we observed previously in enterally fed pigs (4, 7). The mean plasma GLP-2 concentration was 75.4 ± 7.0 pmol/liter in a group of formula-fed piglets studied in parallel to this study; the piglets were the same age and body weight and the plasma GLP-2 was determined in the same assay conditions. Interestingly, the estimated plasma GLP-2 (1–33) clearance rates (liter·kg–1·h–1) were not different, being 0.697 ± 0.122, 0.620 ± 0.122, and 0.841 ± 0.089 for the low (2.5 nmol·kg–1·d–1), medium (5.0 nmol nmol·kg–1·d–1), and high (10.0 nmol·kg–1·d–1) GLP-2 groups, respectively.
FIG. 1. Plasma GLP-2 (1–33) concentrations in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. Results are means ± SEM for n = 7–10 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
Based on Tukey’s means comparison, the medium and high, but not the low, GLP-2 treatment groups had significantly higher intestinal mass, protein and DNA content, villus height and area, and crypt cell number compared with the control group (Table 1). Similar increases associated with GLP-2 infusion rate were observed in these latter endpoints in the distal small intestine segment (results not shown). There were no significant treatment differences in intestinal length or muscularis thickness. In addition, there were no significant treatment differences in the weight (Table 2) or the protein and DNA content (data not shown) of the stomach, liver, or spleen.
TABLE 1. GLP-2 infusion increases intestinal mass, protein, and DNA content in TPN-fed neonatal piglets
TABLE 2. Growth and protein synthesis rates in stomach, liver, spleen, and skeletal muscle in response to GLP-2 infusion
To determine whether the changes in intestinal protein mass were mediated by synthesis or degradation, we measured the FSR. The FSR in the intestine was significantly (P < 0.01) increased by GLP-2 treatment, but only the high GLP-2 treatment was different (P < 0.05) from the control group (Fig. 2). In contrast, stomach protein synthesis was significantly lower (P < 0.05) in the high GLP-2 than the control group, whereas protein synthesis in the liver, spleen, and muscle were not affected by GLP-2 (Table 2). Representative images of hematoxylin and eosin and BrdU histochemistry in intestinal sections from control and high GLP-2 pigs are shown (Fig. 3). The rates of crypt and villus epithelial cell apoptosis and crypt cell proliferation were measured to explain the GLP-2-induced increase in intestinal DNA content (Fig. 4). GLP-2 dose-dependently suppressed (P < 0.05) apoptosis in both crypt and villus cell compartments, even at the lowest GLP-2 infusion rate. However, crypt cell proliferation, expressed as a percentage (Fig. 4) and total number of BrdU-positive cells per crypt (data not shown), was increased only at the highest GLP-2 infusion rate, consistent with the dose response in protein synthesis.
FIG. 2. GLP-2 infusion increases intestinal protein synthesis. A, Multiple regression analysis indicated a highly significant positive relationship between protein synthesis and plasma GLP-2 concentration (y = 61.1 + 0.0042x + 3.061E –5x2; R2 = 0.781; P < 0.001; n = 36). B, Mean intestinal fractional protein synthesis rates in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. Results are means ± SEM for n = 7–8 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 3. Histological images of BrdU-stained (A and B; x200 magnification) and hematoxylin- and eosin-stained (C and D; x60 magnification) jejunal tissue sections from TPN-fed piglets infused with 0.0 (A and C) or 10.0 (B and D) nmol·kg–1·d–1 of human GLP-2 for 7 d. In A and B, BrdU-positive crypt cells show dark brown staining (black arrows) with gray hematoxylin-counterstained nuclei.
FIG. 4. GLP-2 infusion decreases apoptosis and increases cell proliferation. A–C, Intestinal apoptosis rates in the villus (A) and crypt (B) compartments and crypt cell proliferation rates (C) in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. Results are means ± SEM for n = 7–9 pigs per group. Differences between control and specific GLP-2-treated groupsare based on Tukey’s means comparison; *, P < 0.05.
The intestinal activities of caspase-3 and -6 (Fig. 5) and protein expression of caspase-3 (Fig. 6) and Bcl-2 (Fig. 7) were measured given their integral role in intestinal epithelial cell apoptosis. Similar to apoptosis rates, the activities of caspase-3 and -6 were dose-dependently (P < 0.05) suppressed with medium and high GLP-2 infusion rates, yet we found no difference between the control and low GLP-2 groups. GLP-2 also dose-dependently increased the abundance of the procaspase-3 form and decreased the abundance of the active caspase-3 form, resulting in a marked suppression of the active to inactive ratio of the caspase-3 enzyme protein. Consistent with increased cell survival, the intestinal expression of Bcl-2 protein was significantly (P < 0.05) increased with increasing GLP-2 infusion rates. Furthermore, regression analysis indicated that the ratio of active to inactive caspase-3 was effectively suppressed (P < 0.001) and Bcl-2 protein expression increased with increasing circulating plasma GLP-2 concentrations, especially within the physiological range (<200 pM; Fig. 7). Regression analysis indicated that Bcl-2 protein abundance was significantly and negatively correlated (R2 = 0. 74; P < 0.001) to the amount of active caspase-3. Thus, based on either the suppression of active caspase-3 or stimulation of Bcl-2 expression, the maximal cell survival response occurred in the pharmacological circulating GLP-2 concentration range of 600–800 pM.
FIG. 5. GLP-2 infusion decreases intestinal caspase-3 (A) and caspase-6 (B) activities in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. Results are means ± SEM for n = 7–8 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 6. A, GLP-2 decreases the ratio of active caspase-3 (lower band, 19 kDa) to procaspase-3 (upper band, 34 kDa) in the intestine (Western blot) of TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. B, Quantitative analysis of procaspase-3 (solid bars) and active caspase-3 (hatched bars) protein abundance expressed in arbitrary units. Percentage values show the calculated ratio of active caspase-3 to procaspase-3. Results are means ± SEM for n = 6 pigs per group. Differences in procaspase-3, active caspase-3, and percentage of active/procaspase-3 abundance between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 7. GLP-2 increases the intestinal expression of Bcl-2 protein of TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. A, Western blot result of Bcl-2 expression in pooled intestinal tissue extracts from piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. B, Quantitative results of Bcl-2 expression expressed in arbitrary units, where results are means ± SEM for n = 6 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05. C, Relationship between activated caspase-3 ratio (, solid line) and Bcl-2 (, dotted line) expression with plasma GLP-2 concentrations (C). Multiple regression analysis indicated the following equation: y = 37.1 –0.069x + 0.0004x2; R2 = 0.73; P < 0.001; n = 24 for activated caspase-3 ratio; and y = 68.7 + 1.331x –0.00078x2; R2 = 0.89; P < 0.001; n = 24 for Bcl-2 expression.
We next measured various upstream signaling molecules reported to be involved in the inhibition of apoptosis and particularly GLP-2 receptor signaling, including PI3-kinase, PKB, and GSK-3. GLP-2 infusion did not affect the protein expression of either PI3-kinase or PKB (data not shown). However, the medium and high GLP-2 doses significantly (P < 0.05) increased the phosphorylation of both PKB (Fig. 8) and GSK-3 (Fig. 9). Moreover, we found that the relative abundance of both Ser473 and Thr308 phospho-PKB was increased by all three GLP-2 infusion rates. We also measured the intestinal expression of eNOS, which is responsible for vascular nitric oxide release and may be involved in the acute (4-h) GLP-2-induction of intestinal blood flow reported previously (14). Consistent with previous work, we found that both eNOS mRNA and protein (Fig. 10) expression were increased (P < 0.05) and sustained even after a week of continuous GLP-2 infusion, but only at the highest GLP-2 infusion rate.
FIG. 8. GLP-2 increases the phosphorylation but not the abundance of intestinal PKB protein in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. A, Results of PKB-Ser473, PKB-Thr308, and PKB protein from Western blot of pooled samples. B, Quantitative results of PKB-Ser473, PKB-Thr308 abundance expressed relative to PKB protein. Results are means ± SEM for n = 6 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 9. GLP-2 increases the phosphorylation but not the abundance of intestinal GSK-3 protein in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. A, Results of GSK-3 protein and phospho-GSK abundance in pooled intestinal samples. B, Quantitative results of phospho-GSK abundance. Results are means ± SEM for n = 6 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 10. GLP-2 increases the intestinal expression of eNOS mRNA of TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. A, Western blot result of eNOS expression in pooled intestinal tissue extracts from piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. B, Quantitative results of eNOS expression expressed in arbitrary units, where results are means ± SEM for n = 6 pigs per group. C, Results of RT-PCR analysis of eNOS mRNA abundance expressed relative to 18S mRNA abundance in intestinal tissue. Results are means ± SEM for n = 6 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
Discussion
The primary aim of this study was to determine intestinal responsiveness to various GLP-2 infusion rates that were targeted to produce circulating GLP-2 concentrations within the physiological and pharmacological range in TPN-fed piglets. Plasma GLP-2 concentrations considered to be in the physiological range are based on reports in enterally fed piglets (50–100 pM) (4, 7). A key finding was that the activation of intestinal epithelial cell survival, based on apoptosis and caspase-3 expression, occurred at relatively low circulating GLP-2 concentrations, whereas only high GLP-2 concentrations induced the stimulation of cell proliferation and protein synthesis. The results suggest that the activation of intestinal epithelial survival is more sensitive than cell proliferation to circulating GLP-2 concentrations in neonatal piglets. The combined effect of these cell kinetic responses resulted in dose-dependent increases in intestinal DNA mass and villus growth. This study also provides the first in vivo evidence that GLP-2 activates intestinal cell survival in association with induction of PKB and GSK-3 phosphorylation and Bcl-2 expression.
A key unresolved issue with respect to GLP-2 function is its physiological relevance to the well-documented intestinal trophic response to enteral nutrition. In the current study, we sought to examine the effect of reproducing a physiological circulating GLP-2 concentration, typical for enterally fed pigs (50–100 pM) (4, 7), but in a model of the TPN-fed piglet where the plasma GLP-2 concentration and intestinal mucosal growth are relatively suppressed because of the absence of luminal nutrition. Based on regression analysis, we observed that key endpoints of intestinal growth, namely mass and mucosal morphology, increased significantly with increasing infusion rate and circulating concentrations of GLP-2. However, means comparison analysis indicated that intestinal trophic effect of GLP-2 occurred mainly at the medium and high GLP-2 treatment groups, when compared with the control group. It is notable that the plasma GLP-2 concentrations in the medium and high GLP-2 groups are within the range (400–1200 pM) of those reported in GLP-2-treated short-bowel patients (27). Thus, in the absence of any enteral nutrition, artificially increasing the circulating GLP-2 concentration above the physiological range (i.e. 75 pM enteral fed) appears to be insufficient alone to induce intestinal growth in the neonatal piglet. Therefore, it appears that the intestinal trophic response to physiological GLP-2 concentrations requires the presence of additional contributing factors, such as luminal nutrients or other gut hormones. It is possible that the TPN decreases the intestinal GLP-2 responsiveness via disruption of GLP-2R signaling pathways. In addition, the increased circulating GLP-2 concentrations that we achieved in TPN-fed piglets may not reproduce the local tissue concentration, which occurs with enteral feeding perhaps because of high activity of dipeptidyl peptidase IV enzyme. Thus, under the conditions of TPN, producing a trophic effect may require a higher plasma level of GLP-2 than what otherwise might occur during enteral feeding.
An especially important finding in this study was the differential responsiveness of cell survival and proliferation to the circulating GLP-2 concentration. We previously showed that a pharmacological GLP-2 infusion rate prevented intestinal mucosal atrophy in TPN-fed piglets by decreasing apoptosis, but without a stimulation of cell proliferation (7). This was somewhat in contrast to several reports demonstrating that GLP-2 treatment both decreased intestinal apoptosis and increased crypt cell proliferation in rodents (12, 13) and cultured cells (18, 19, 21). In the current study, we observed significant decreases in apoptosis and active caspase-3 expression at all GLP-2 infusion rates; this was especially evident in the inverse, nonlinear response between active caspase-3 expression and plasma GLP-2. Activated caspase-3 is a key executioner protease responsible for most of the cellular destruction during apoptosis. Moreover, we found that both caspase-3 and caspase-6 activities were suppressed in parallel with epithelial cell apoptosis rate, consistent with reports showing their rapid activation in intestinal epithelial cells during detachment-induced apoptosis (28). However, contrary to our previous study, we found that GLP-2 increased crypt cell proliferation and protein synthesis, but only at the highest infusion rate and circulating concentrations of GLP-2. This differential responsiveness of cell death vs. cell growth mechanisms may be model dependent, because TPN is marked by significant mucosal atrophy compared with the fed condition. These results suggest that intestinal cell survival mechanisms are more sensitive than cell proliferation to the circulating GLP-2 concentration in neonatal piglets.
The differential responsiveness of intestinal cell survival and proliferation to GLP-2 has been observed in cultured cells and may be a consequence of different intracellular signaling pathways. The nature of intestinal GLP-2R signaling in vivo is largely unknown because of uncertainty about its precise cellular localization; reports indicate the presence of the receptor in human endocrine cells (10) and murine enteric neurons (17). However, studies in fibroblasts transfected with the GLP-2R have shown that much lower GLP-2 concentrations are necessary to inhibit apoptosis (20 nM) (19, 20) than to stimulate cell proliferation (100 nM) (18). These studies also have shown that the GLP-2-mediated inhibition of apoptosis in GLP-2R-transfected BHK cells occurs via a cAMP-dependent, protein kinase A pathway that is associated with increased GSK-3 phosphorylation (19, 20) but does not involve activation of PI3-kinase or PKB. This observation is consistent with evidence that phosphorylation-dependent inhibition of GSK-3 by G protein-coupled receptors seems to be mediated by PKA, whereas PI3-kinase/PKB-mediated GSK-3 inhibition is associated with stimulation by growth factors, such as insulin (29). Contrary to this general idea are studies with cultured colon carcinoma cells demonstrating that GLP-2 increased cell proliferation at supraphysiological (10 nM) GLP-2 concentrations and that this response was associated with increased MAPK kinase phosphorylation and blocked by inhibitors of MAPK and PI3-kinase. In the current study, we found that none of the GLP-2 treatments increased PI3-kinase, PKB, or GSK-3 protein expression. However, both PKB and GSK-3 phosphorylation were significantly increased in the GLP-2-treated groups. Therefore, our results provide the first in vivo evidence suggesting that the GLP-2-induced suppression of intestinal apoptosis and stimulation of cell proliferation is associated with PKB activation and GSK-3 inhibition.
Recent studies have linked the cardioprotective actions of GSK-3 inhibition to the delayed activation of the mitochondrial permeability transition, a key regulator of apoptosis (30). GSK-3 inhibition also is a prerequisite for caspase-3 activation associated with thapsigargin-induced endoplasmic reticulum stress (31). Mitochondrial dysfunction is generally associated with the intrinsic apoptotic pathway and is largely controlled by the activity of Bcl-2 family proteins. The Bcl-2-related member, Bad, is a proapoptotic downstream target of PI3-kinase/PKB phosphorylation and has been associated with GSK-3 inhibition of apoptosis (32). GLP-2 stimulates Bad phosphorylation in association with the inhibition of GSK-3 and apoptosis in GLP-2R-transfected fibroblasts (20). We did not measure the protein expression or phosphorylation of Bad in intestinal extracts; however, we found that the expression of the anti-apoptotic protein, Bcl-2, was dose-dependently increased with GLP-2 infusion. Bcl-2 deficiency leads to intestinal dysfunction and is positively correlated with the attenuation of apoptosis (33, 34). Overexpression of Bcl-2 inhibits ischemia-reperfusion-induced apoptosis in the intestinal epithelium in transgenic mice (35). The anti-apoptotic action of Bcl-2 occurs via stabilization of mitochondrial membrane integrity and prevents cytochrome c release, which is a critical early event in apoptosis. Bcl-2 is also a target of caspase-3, whereby caspase-3 activation promotes Bcl-2 cleavage in a positive feedback loop, which accelerates the release of mitochondrial cytochrome c and eventually cell death (36). Thus, our results are consistent with the idea that GLP-2-induced suppression of caspase-3 activity limits the degradation of Bcl-2, leading to increased cellular availability for Bcl-2 stabilization of mitochondrial function and increased cell survival. Whether the inhibition of GSK-3 is mechanistically linked to the GLP-2-associated increase in Bcl-2 expression warrants further study.
Nitric oxide (NO) is a ubiquitous, cell-permeable anti-apoptotic signaling molecule involved in a variety of cell functions, including vasodilatation, apoptosis, and inflammation (37, 38). We previously demonstrated that the GLP-2-mediated up-regulation of intestinal blood flow and glucose uptake is NO dependent and associated with increased expression of eNOS. However, the role of NO in the GLP-2-mediated intestinal survival and proliferation responses is unknown. Nitric oxide protects the structure and function of enterocytes under stress (e.g. oxidative stress, injury, and colitis) (39, 40, 41) and inhibits mitochondrial dysfunction-induced apoptosis (42). Studies also indicate that NO can inhibit apoptosis by nitrosylation-mediated suppression of caspase-3 activation (37). It is also of interest that eNOS is not only a PKB substrate but also is activated in a protein kinase A (PKA)-dependent manner (43, 44). Consistent with our previous study showing that GLP-2 acutely up-regulates intestinal eNOS expression, we found that eNOS expression was significantly increased, but only at the high GLP-2 infusion rate. Additional studies are warranted to determine whether GLP-2 induces eNOS expression via a GLP-2R- mediated pathway involving PKA and PKB and whether the GLP-2-mediated increase in intestinal epithelial cell survival, proliferation, or protein synthesis is NO dependent.
Another aim of this study was to quantify the growth and protein synthesis response to GLP-2 in other tissues, besides the small intestine. The expression of the GLP-2R mRNA is largely confined to the gastrointestinal tissues, including stomach and intestine, but not the liver, spleen, or skeletal muscle (10, 11); however, the reports of GLP-2-mediated inhibition of bone resorption may be linked to GLP-2R expression in peripheral sites. Consistent with the expression pattern, we found no change in either growth (protein and DNA content) or protein synthesis in the liver, spleen, or skeletal muscle in response to GLP-2 treatment. However, the high GLP-2 infusion rate significantly decreased the stomach protein synthesis rate, despite the lack of effect on protein and DNA mass. This result is consistent with reports that GLP-2 also suppresses gastric secretion and motor function (45).
In summary, the current study extends our previous finding that GLP-2 prevents mucosal atrophy in TPN-fed piglets by showing the dose dependence of this response and the cellular signals potentially involved. We found that the GLP-2 dose-dependent stimulation of intestinal growth is mediated largely by stimulation of epithelial cell survival and to a lesser extent by cell proliferation, and this leads to preserved villus length. Although our results suggest that the stimulation of intestinal cell survival occurs within the physiological range of circulating GLP-2, it did not translate into significant increases in intestinal mucosal structure and mass. This indicates that reproducing a physiological circulating concentration of GLP-2 does not produce the intestinal trophic effect observed with enteral nutrition. Thus, if GLP-2 has a physiological role in enteral nutrient-mediated intestinal growth, it appears to require other factors such as luminal nutrients, other gut hormones, or neural stimulation. In addition, we showed that GLP-2 stimulates intestinal epithelial cell survival and proliferation in association with suppression of caspase-3 and induction of PKB and GSK-3 phosphorylation and Bcl-2 expression. This study did not attempt to localize the activation of these signaling molecules to specific cells within the mucosa, namely the epithelial cells where we observed changes in apoptosis and cell proliferation. However, given the current evidence that the GLP-2R expression is confined to a limited cell population (enteroendocrine cells or enteric neurons), we postulate that the changes observed in these signaling pathways occur in epithelial cells downstream of GLP-2R activation. Therefore, it will be important to establish whether the GLP-2R and activation of these signaling molecules occurs within the same cell type or whether GLP-2R signaling occurs via a heterotypic, intercellular signaling pathway.
Acknowledgments
We acknowledge the technical assistance of Xiaofeng Lu.
References
Drucker DJ 2002 Gut adaptation and the glucagon-like peptides. Gut 50:428–435
Burrin DG, Stoll B, Guan X 2003 Glucagon-like peptide 2 function in domestic animals. Domest Anim Endocrinol 24:103–122
Burrin DG, Stoll B, Jiang RH, Chang XY, Hartmann B, Holst JJ, Greeley GH, Reeds PJ 2000 Minimal enteral nutrient requirements for intestinal growth in neonatal piglets: how much is enough? Am J Clin Nutr 71:1603–1610
van Goudoever JB, Stoll B, Hartmann B, Holst JJ, Reeds PJ, Burrin DG 2001 Secretion of trophic gut peptides is not different in bolus- and continuously fed piglets. J Nutr 131:729–732
Xiao QA, Boushey RP, Drucker DJ, Brubaker PL 1999 Secretion of the intestinotropic hormone glucagon-like peptide 2 is differentially regulated by nutrients in humans. Gastroenterology 117:99–105
Dahly EM, Gillingham MB, Guo Z, Murali SG, Nelson DW, Holst JJ, Ney DM 2003 Role of luminal nutrients and endogenous GLP-2 in intestinal adaptation to mid-small bowel resection. Am J Physiol Gastrointest Liver Physiol 284:G670–G682
Burrin DG, Stoll B, Jiang R, Petersen Y, Elnif J, Buddington RK, Schmidt M, Holst JJ, Hartmann B, Sangild PT 2000 GLP-2 stimulates intestinal growth in premature TPN-fed pigs by suppressing proteolysis and apoptosis. Am J Physiol Gastrointest Liver Physiol 279:G1249–G1256
Chance WT, Foley-Nelson T, Thomas I, Balasubramaniam A 1997 Prevention of parenteral nutrition-induced gut hypoplasia by coinfusion of glucagon-like peptide-2. Am J Physiol 273:G559–G563
Munroe DG, Gupta AK, Kooshesh F, Vyas TB, Rizkalla G, Wang H, Demchyshyn L, Yang ZJ, Kamboj RK, Chen HY, McCallum K, Sumner-Smith M, Drucker DJ, Crivici A 1999 Prototypic G protein-coupled receptor for the intestinotrophic factor glucagon-like peptide 2. Proc Natl Acad Sci USA 96:1569–1573
Yusta B, Huang L, Munroe D, Wolff G, Fantaske R, Sharma S, Demchyshyn L, Asa SL, Drucker DJ 2000 Enteroendocrine localization of GLP-2 receptor expression in humans and rodents. Gastroenterology 119:744–755
Petersen YM, Burrin DG, Sangild PT 2001 GLP-2 has differential effects on small intestine growth and function in fetal and neonatal pigs. Am J Physiol Regul Integr Comp Physiol 281:R1986–R1993
Drucker DJ, Ehrlich P, Asa SL, Brubaker PL 1996 Induction of intestinal epithelial proliferation by glucagon-like peptide 2. Proc Natl Acad Sci USA 93:7911–7916
Tsai CH, Hill M, Asa SL, Brubaker PL, Drucker DJ 1997 Intestinal growth-promoting properties of glucagon-like peptide-2 in mice. Am J Physiol Endocrinol Metab 36:E77–E84
Guan XF, Stoll B, Lu XF, Tappenden KA, Holst JJ, Hartmann B, Burrin DG 2003 GLP-2-mediated up-regulation of intestinal blood flow and glucose uptake is nitric oxide-dependent in TPN-fed piglets. Gastroenterology 125:136–147
Lovshin J, Yusta B, Iliopoulos I, Migirdicyan A, Dableh L, Brubaker PL, Drucker DJ 2000 Ontogeny of the glucagon-like peptide-2 receptor axis in the developing rat intestine. Endocrinology 141:4194–4201
Litvak DA, Hellmich MR, Evers BM, Banker NA, Townsend Jr CM 1998 Glucagon-like peptide 2 is a potent growth factor for small intestine and colon. J Gastrointest Surg 2:146–150
Bjerknes M, Cheng H 2001 Modulation of specific intestinal epithelial progenitors by enteric neurons. Proc Natl Acad Sci USA 98:12497–12502
Yusta B, Somwar R, Wang F, Munroe D, Grinstein S, Klip A, Drucker DJ 1999 Identification of glucagon-like peptide-2 (GLP-2)-activated signaling pathways in baby hamster kidney fibroblasts expressing the rat GLP-2 receptor. J Biol Chem 274:30459–30467
Yusta B, Boushey RP, Drucker DJ 2000 The glucagon-like peptide-2 receptor mediates direct inhibition of cellular apoptosis via a cAMP-dependent protein kinase-independent pathway. J Biol Chem 275:35345–35352
Yusta B, Estall J, Drucker DJ 2002 Glucagon-like peptide-2 receptor activation engages bad and glycogen synthase kinase-3 in a protein kinase A-dependent manner and prevents apoptosis following inhibition of phosphatidylinositol 3-kinase. J Biol Chem 277:24896–24906
Jasleen J, Ashley SW, Shimoda N, Zinner MJ, Whang EE 2002 Glucagon-like peptide 2 stimulates intestinal epithelial proliferation in vitro. Dig Dis Sci 47:1135–1140
Walsh NA, Yusta B, Dacambra MP, Anini Y, Drucker DJ, Brubaker PL 2003 Glucagon-like peptide-2 receptor activation in the rat intestinal mucosa. Endocrinology 144:4385–4392
Li J, Bombeck CA, Yang S, Kim YM, Billiar TR 1999 Nitric oxide suppresses apoptosis via interrupting caspase activation and mitochondrial dysfunction in cultured hepatocytes. J Biol Chem 274:17325–17333
Rajendram R, Marway JS, Mantle D, Peters TJ, Preedy VR 2003 Skeletal muscle and jejunal protein synthesis in normal and ethanol-treated rats: the effect of the nitric oxide synthase inhibitors, L-omega-nitro-L-arginine methyl ester and N(G)-nitro-L-arginine in vivo. Metabolism 52:397–401
Niinikoski H, Stoll B, Guan X, Kansagra K, Lambert BD, Stephens J, Hartmann B, Holst JJ, Burrin DG 2004 Onset of small intestinal atrophy is associated with reduced intestinal blood flow in TPN-fed neonatal piglets. J Nutr 134:1467–1474
Stoll B, Chang X, Fan MZ, Reeds PJ, Burrin DG 2000 Enteral nutrient intake level determines intestinal protein synthesis and accretion rates in neonatal pigs. Am J Physiol Gastrointest Liver Physiol 279:G288–G294
Jeppesen PB, Hartmann B, Thulesen J, Graff J, Lohmann J, Hansen BS, Tofteng F, Poulsen SS, Madsen JL, Holst JJ, Mortensen PB 2001 Glucagon-like peptide 2 improves nutrient absorption and nutritional status in short-bowel patients with no colon. Gastroenterology 120:806–815
Grossmann J, Mohr S, Lapentina EG, Fiocchi C, Levine AD 1998 Sequential and rapid activation of select caspases during apoptosis of normal intestinal epithelial cells. Am J Physiol 274:G1117–G1124
Fang X, Yu SX, Lu Y, Bast Jr RC, Woodgett JR, Mills GB 2000 Phosphorylation and inactivation of glycogen synthase kinase 3 by protein kinase A. Proc Natl Acad Sci USA 97:11960–11965
Juhaszova M, Zorov DB, Kim SH, Pepe S, Fu Q, Fishbein KW, Ziman BD, Wang S, Ytrehus K, Antos CL, Olson EN, Sollott SJ 2004 Glycogen synthase kinase-3? mediates convergence of protection signaling to inhibit the mitochondrial permeability transition pore. J Clin Invest 113:1535–1549
Song L, De Sarno P, Jope RS 2002 Central role of glycogen synthase kinase-3? in endoplasmic reticulum stress-induced caspase-3 activation. J Biol Chem 277:44701–44708
Stoica BA, Movsesyan VA, Lea PM, Faden AI 2003 Ceramide-induced neuronal apoptosis is associated with dephosphorylation of Akt, BAD, FKHR, GSK-3?, and induction of the mitochondrial-dependent intrinsic caspase pathway. Mol Cell Neurosci 22:365–382
Kamada S, Shimono A, Shinto Y, Tsujimura T, Takahashi T, Noda T, Kitamura Y, Kondoh H, Tsujimoto Y 1995 Bcl-2 deficiency in mice leads to pleiotropic abnormalities: accelerated lymphoid cell death in thymus and spleen, polycystic kidney, hair hypopigmentation, and distorted small intestine. Cancer Res 55:354–359[Abstract]
Merritt AJ, Potten CS, Watson AJ, Loh DY, Nakayama K, Nakayama K, Hickman JA 1995 Differential expression of bcl-2 in intestinal epithelia: correlation with attenuation of apoptosis in colonic crypts and the incidence of colonic neoplasia. J Cell Sci 108:2261–2271[Abstract]
Coopersmith CM, O’Donnell D, Gordon JI 1999 Bcl-2 inhibits ischemia-reperfusion-induced apoptosis in the intestinal epithelium of transgenic mice. Am J Physiol 276:G677–G686
Kirsch DG, Doseff A, Chau BN, Lim DS, Souza-Pinto NC, Hansford R, Kastan MB, Lazebnik YA, Hardwick JM 1999 Caspase-3-dependent cleavage of Bcl-2 promotes release of cytochrome c. J Biol Chem 274:21155–21161
Kim PK, Kwon YG, Chung HT, Kim YM 2002 Regulation of caspases by nitric oxide. Ann NY Acad Sci 962:42–52
Shah V, Lyford G, Gores G, Farrugia G 2004 Nitric oxide in gastrointestinal health and disease. Gastroenterology 126:903–913
Thomas S, Ramachandran A, Patra S, Vidyasagar S, Balasubramanian KA 2001 Nitric oxide protects the intestine from the damage induced by laparotomy and gut manipulation. J Surg Res 99:25–32
Sasaki M, Bharwani S, Jordan P, Elrod JW, Grisham MB, Jackson TH, Lefer DJ, Alexander JS 2003 Increased disease activity in eNOS-deficient mice in experimental colitis. Free Radic Biol Med 35:1679–1687
Bersimbaev RI, Yugai YE, Hanson PJ, Tzoy IG 2001 Effect of nitric oxide on apoptotic activity in the rat gastrointestinal tract. Eur J Pharmacol 423:9–16
Kim YM, Kim TH, Seol DW, Talanian RV, Billiar TR 1998 Nitric oxide suppression of apoptosis occurs in association with an inhibition of Bcl-2 cleavage and cytochrome c release. J Biol Chem 273:31437–31441
Fulton D, Gratton JP, McCabe TJ, Fontana J, Fujio Y, Walsh K, Franke TF, Papapetropoulos A, Sessa WC 1999 Regulation of endothelium-derived nitric oxide production by the protein kinase Akt. Nature 399:597–601
Boo YC, Sorescu G, Boyd N, Shiojima I, Walsh K, Du J, Jo H 2002 Shear stress stimulates phosphorylation of endothelial nitric-oxide synthase at Ser1179 by Akt-independent mechanisms: role of protein kinase A. J Biol Chem 277:3388–3396
Wojdemann M, Wettergren A, Hartmann B, Hilsted L, Holst JJ 1999 Inhibition of sham feeding-stimulated human gastric acid secretion by glucagon-like peptide-2. J Clin Endocrinol Metab 84:2513–2517(Douglas G. Burrin, Barbar)
Address all correspondence and requests for reprints to: Douglas G. Burrin, Ph.D., Children’s Nutrition Research Center, 1100 Bates Street, Houston, Texas 77030. E-mail: dburrin@bcm.tmc.edu.
Abstract
Glucagon-like peptide 2 (GLP-2) is a gut hormone that stimulates mucosal growth in total parenteral nutrition (TPN)-fed piglets; however, the dose-dependent effects on apoptosis, cell proliferation, and protein synthesis are unknown. We studied 38 TPN-fed neonatal piglets infused iv with either saline or GLP-2 at three rates (2.5, 5.0, and 10.0 nmol·kg–1·d–1) for 7 d. Plasma GLP-2 concentrations ranged from 177 ± 27 to 692 ± 85 pM in the low- and high-infusion groups, respectively. GLP-2 infusion dose-dependently increased small intestinal weight, DNA and protein content, and villus height; however, stomach protein synthesis was decreased by GLP-2. Intestinal crypt and villus apoptosis decreased and crypt cell number increased linearly with GLP-2 infusion rates, whereas cell proliferation and protein synthesis were stimulated only at the high GLP-2 dose. The intestinal activities of caspase-3 and -6 and active caspase-3 abundance decreased, yet procaspase-3 abundance increased markedly with increasing infusion rate and plasma concentration of GLP-2. The GLP-2-dose-dependent suppression of intestinal apoptosis and caspase-3 activity was associated with increased protein kinase B and glycogen-synthase kinase-3 phosphorylation, yet the expression phosphatidylinositol 3-kinase was unaffected by GLP-2. Intestinal endothelial nitric oxide synthase mRNA and protein expression was increased, but only at the high GLP-2 dose. We conclude that the stimulation of intestinal epithelial survival is concentration dependent at physiological GLP-2 concentrations; however, induction of cell proliferation and protein synthesis is a pharmacological response. Moreover, we show that GLP-2 stimulates intestinal cell survival and proliferation in association with induction of protein kinase B and glycogen-synthase kinase-3 phosphorylation and Bcl-2 expression.
Introduction
GLUCAGON-LIKE PEPTIDE 2 (GLP-2) is a 33-amino-acid peptide derived from differential posttranslational processing of proglucagon in intestinal endocrine L-cells (1, 2). GLP-2 secretion is stimulated in response to food intake, and the circulating concentration is significantly reduced by total parenteral nutrition (TPN) (3, 4, 5). In parenterally fed neonatal piglets given increasing proportions of enteral nutrition, we observed a strong positive correlation between the circulating GLP-2 concentration and intestinal mucosal growth (3). Similarly, recent studies in rats showed that the intestinal adaptive trophic response to small bowel resection was positively correlated with circulating GLP-2 and colonic proglucagon gene expression (6). Results from these two different models of intestinal adaptation suggest that the circulating GLP-2 concentration is an important physiological stimulus of mucosal growth. Consistent with this idea, inducing supraphysiological circulating GLP-2 concentrations by infusion of exogenous GLP-2 can prevent TPN-induced mucosal atrophy (7, 8), yet it is unknown whether this response occurs within the physiological range of circulating GLP-2.
The biological effects of GLP-2 are mediated via activation of a G protein-linked, membrane receptor (GLP-2R) expressed mainly in the gastrointestinal tract and brain (9, 10, 11). The dominant biological effect of GLP-2 is stimulation of small intestinal mucosal growth, which is associated with reduced apoptosis and proteolysis and increased protein synthesis in TPN-fed piglets (7) and with increased cell proliferation and reduced apoptosis in chow-fed mice (12, 13, 14). In addition, trophic effects have been observed in the stomach and colon (15, 16); however, there is limited information as to whether GLP-2 affects growth and metabolism in non-gastrointestinal tissues in neonates. The cellular mechanisms of GLP-2 action subsequent to binding and activation of the GLP-2R are poorly defined and complicated by uncertainty of the specific cellular localization. Conflicting reports have demonstrated the presence of the GLP-2R in human enteroendocrine cells (10) and in murine enteric neurons (17). Studies in GLP-2R-transfected fibroblasts have demonstrated that GLP-2 induces cAMP production, immediate early gene expression, and cell proliferation (18). Subsequent studies with this model showed that GLP-2 prevents apoptosis via protein kinase A-dependent phosphorylation of Bad and glycogen-synthase kinase-3 (GSK-3) and downstream inhibition of caspase-3 activity but does not involve phosphatidylinositol 3-kinase (PI3-kinase) or MAPK pathways (19, 20). In contrast, studies with a human colon carcinoma cell line (Caco-2) have indicated that GLP-2 increases cell proliferation in association with a transient increase in ERK phosphorylation, and the response is suppressed by inhibitors of PI3-kinase and MAPK (21). Despite the caveats of these GLP-2 studies in transformed cell lines, a recent report with isolated primary intestinal cells has shown that the GLP-2R is expressed and that GLP-2 induced cAMP production and [3H]thymidine incorporation (22). Whether any of these aforementioned signaling pathways are activated by GLP-2 in vivo is unknown. In addition, our recent study in TPN-fed piglets suggests that the GLP-2 stimulation of intestinal blood flow is NO dependent and associated with increased endothelial nitric oxide synthase (eNOS) expression, possibly implicating nitric oxide in other downstream GLP-2R signaling events (14), such as cell apoptosis and protein metabolism (23, 24).
The aim of the current study was to determine whether the intestinal trophic actions of GLP-2 are dose dependent in TPN-fed piglets given GLP-2 infusion rates that produce circulating GLP-2 concentrations within the physiological and pharmacological range. In investigating the GLP-2 dose response, we quantified several endpoints of intestinal growth and also sought to establish whether any of the cellular signaling proteins that are activated by GLP-2 in cultured cells are also increased in the intestine in response to GLP-2 treatment in vivo.
Materials and Methods
Animals, surgery, and study design
The study protocol was approved by the Animal Care and Use Committee of Baylor College of Medicine and was conducted in accordance with the Guide for the Care and Use of Laboratory Animals [Department of Health and Human Services publication no. (NIH) 85-23, revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD]. Pregnant, crossbred sows were obtained from the Texas Department of Criminal Justice (Huntsville, TX) approximately 1 wk before estimated date of parturition. Sows were housed, fed, and kept under surveillance in the animal facility at the Children’s Nutrition Research Center (Houston, TX) until delivery. At the time of parturition, vaginally born newborn piglets were removed from the sow to prevent colostrum ingestion and immediately placed in heated cages (30 C) and given water until surgery.
The piglets were surgically catheterized within 12 h of birth under isoflurane general anesthesia. Silastic catheters were inserted into the jugular vein and carotid artery via a procedure as previously described (3). Catheters were secured in a jacket that was attached to a tether, which allowed free movement and secure administration of TPN to the piglets in their cages. Pre- and postoperatively on each day, piglets received enrofloxacin (2.5 mg·kg–1; Bayer, Shawnee Mission, KS). Postoperatively, each piglet received one dose of analgesic (0.1 mg·kg–1 butorphenol tartrate, Fort Dodge Labs, Fort Dodge, IA).
Piglets were administered TPN daily that provided 240 ml fluid, 25 g glucose, 13 g amino acids, and 5 g lipid per kilogram body weight for 7 d; piglets were weighed daily to adjust the TPN infusion rate. The parenteral nutrient solution consisted of dextrose (104 g/liter), a complete amino acid mixture (55 g/liter;), lipid (21 g/liter; Intralipid 20%; Fresenius Kabi, Bad Homburg, Germany), electrolytes, trace minerals, and vitamins as described previously (3). The nutrient solution was administered continuously at 50% the full rate during the initial 24-h period after surgery and thereafter increased to 100%. Within 6 h of surgery, the piglets were randomly assigned to one of four treatment groups; a control group received TPN plus a continuous iv infusion at 0.5 ml·kg–1·h–1 of 0.45 g/liter NaCl containing 0.1% human serum albumin (Bayer Corp., Elkhart, IN) or the same infusion rate of 0.45 g/liter NaCl containing 0.5% human serum albumin containing human GLP-2 (1–33) (California Peptide Research, Inc., Napa, CA) at one of three peptide concentrations (0.8, 1.6, or 3.6 μg/ml). The approximate daily molar infusion rates were 2.5, 5.0, and 10.0 nmol·kg–1 for the low, medium, and high GLP-2 treatment groups, respectively. As with the TPN solution, infusion rates of control, and GLP-2 solutions were adjusted daily based on body weight. A total of 38 piglets were studied in the four groups: TPN (n = 10), low GLP-2 (n = 7), medium GLP-2 (n = 8), and high GLP-2 (n = 13).
After 6 d of full TPN or TPN plus GLP-2 treatment, piglets received an iv bolus of bromodeoxyuridine (BrdU) at 50 mg/kg body weight (Sigma Aldrich, St. Louis, MO) 8 h before the pigs were killed to estimate crypt cell proliferation (see below). In addition, a bolus dose of [13C]phenylalanine (1.5 mmol/kg phenylalanine containing 0.15 mmol/kg [13C6]phenylalanine) (Cambridge Isotope Laboratories, Andover, MA) was given 30 min before pigs were killed to measure the rate of tissue protein synthesis. Pigs were killed with a venous injection of pentobarbital sodium (50 mg/kg body weight) and sodium phenytoin (5 mg/kg body weight) (Beuthanasia-D, Schering-Plough Animal Health, Kenilworth, NJ). The abdomen was opened, and the whole small intestine was excised and quickly flushed with ice-cold saline, weighed, and divided into parts of equal length, designated as jejunum and ileum. Samples of jejunum were placed in formalin for morphological and BrdU analysis. The liver, spleen, and stomach were removed and weighed. Samples of liver, spleen, stomach, and hind limb skeletal muscle were also collected and with intestinal tissues were frozen in liquid nitrogen and used for subsequent measurements.
Morphometry, cell proliferation, and apoptosis
Morphometry analysis of intestinal mucosal tissue was performed on formalin-fixed, hematoxylin- and eosin-stained sections as described previously (3, 25). In vivo crypt cell proliferation was measured as described previously (3). BrdU-labeled cells were detected by immunohistochemistry in formalin-fixed, paraffin-embedded sections and expressed as a percentage of total nuclei per crypt observed in approximately 15–20 well-oriented crypt sections from two to three tissue sections from each animal. Measurements of apoptosis were made based on cell morphology observed in x400 images by a single, trained observer that was blinded of the treatments. The apoptotic cells were characterized by evidence of condensed irregular chromatin, nuclear fragmentation, and intensely eosinophilic cytoplasm as shown previously (25). Apoptotic cells were expressed as a percentage of the total epithelial cell numbers in the villus and crypt compartment of the same section; approximately 1000–1500 total epithelial cells were counted from two to three tissue sections per animal. Tissue samples were assayed for DNA content as previously described (3).
In vivo protein synthesis and mass spectrometry
Samples of jejunum, liver, spleen, and stomach tissue were homogenized and deproteinized with 2 M perchloric acid, and the perchloric acid-soluble (tissue-free pool) and acid-insoluble (protein-bound pool) fractions were subjected to mass spectrometric analysis similar to that described previously (7, 26). The acid-insoluble fraction was hydrolyzed with 6 N HCl for 24 h before gas chromatography mass spectrometry analysis. The isotopic enrichment of [U-13C]phenylalanine (M+6 isotopomer) in the two tissue pools was determined by gas chromatography mass spectrometry analysis of the n-propyl ester heptafluorobutyramide derivative using methane-negative chemical ionization. The analyses were performed with a 5890 series II gas chromatograph linked to a model 5989B (Hewlett-Packard, Palo Alto, CA) quadrupole mass spectrometer. The isotopic enrichment of phenylalanine was determined by monitoring ions at a mass-to-charge ratio of 383 to 389. Protein synthesis was calculated as described previously as the fractional protein synthesis rate (FSR, %/d): FSR = (IEbound/IEfree) x (1440/t) x 100, where IEbound and IEfree are the isotopic enrichments (mol% excess) of [13C6]phenylalanine of the perchloric acid-insoluble (protein-bound) and perchloric acid-soluble (tissue-free) pool, t is the time of labeling (in minutes), and 1440 is the number of minutes in a day. Tissue samples were assayed for protein using the BCA method (Pierce, Rockford, IL).
Caspase-3 and -6 activity
Intestinal tissue caspase activities were measured with EnzChek caspase-3 assay kit (Molecular Probes, Inc., Eugene, OR), based on the incubation of caspase-3 [Z-DEVD-(Asp-Glu-Val-Asp)] and caspase-6 [Ac-VEID-(Ac-Val-Glu-Ile-Asp)] (Calbiochem, San Diego, CA) specific substrates conjugated to the 7-amino-4-methylcoumarin compound, which yields a fluorescent product upon cleavage. Frozen tissue (100 mg) was homogenized in 1.5 ml buffer containing 50 mM Tris/HCl (pH 8.0), 25 mM MgCl2, and 0.1 mM phenylmethylsulfonyl fluoride. The homogenate was centrifuged at 5000 x g for 15 min at 4 C. Extracts (50 μl) were mixed with 50 μl of working solution and applied to micro-well plates and incubated at room temperature (covered from light) for 30 min (caspase-3 assay) or at 37 C for 45 min (caspase-6 assay). Fluorescence readings were obtained at 5-min intervals for 45 min at 441 nm in a SPECTRAmax Gemini XS Microplate Spectrophotometer (Molecular Devices). The linearity of the reaction was confirmed, and the time period between 5 and 20 min was used for caspase-6 and the time period between 5 and 30 min for caspase-3. Activity rates were corrected for extract protein concentration determined using the BCA method with albumin as a standard.
Western blotting analysis
Caspase-3 and Bcl-2.
For all Western blots, frozen intestinal tissue samples (100 mg) were homogenized in 50 mM HEPES buffer (pH 7.4) containing 1 mM EDTA, 1 mM dithiothreitol, 5 mg/liter phenylmethylsulfonyl fluoride, 5 mg/liter aprotinin, 5 mg/liter chymostatin, and 5 mg/liter pepstatin. The homogenate was then sonicated and centrifuged at 12,000 x g for 15 min at 4 C. Equal amounts (30–120 μg) of supernatant protein extracts were separated on a 15% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk in the Tris-buffered saline (TBS). Membranes were incubated with a primary antibody [rabbit polyclonal antibody against human caspase-3, H-277, 1:1000; and rabbit polyclonal antibody against human Bcl-2, C21, 1:200 (Santa Cruz Biotechnology, Santa Cruz, CA)] diluted in the 5% nonfat milk in TBS with added Tween 20 solution (0.1%). Membranes were incubated with a secondary antibody [goat antirabbit IgG-horseradish peroxidase (HRP), 1:5000 (Santa Cruz Biotechnology)]. The molecular mass of caspase-3 precursor is 34 kDa, and the active form is 19 kDa, whereas that of Bcl-2 is 29 kDa.
PI3-kinase, protein kinase B (PKB), GSK-3.
Equal amounts (30–120 μg) of supernatant protein extracts were separated on a 9% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk in the TBS. Membranes were incubated with a primary antibody (mouse monoclonal antibody against human PI3-kinase, 1:1000 (MBL International Corp., Watertown, MA); mouse monoclonal antibody against human PKB1, 1:500; and mouse monoclonal antibody against Xenopus GSK-3?, 1:1000 (Santa Cruz Biotechnology)], diluted in the 5% nonfat milk in TBS with added Tween 20 solution (0.1%). Membranes were incubated with a secondary antibody [goat antimouse IgG1-HRP, 1:5000 (Santa Cruz Biotechnology)]. The molecular mass of PI3-kinase is 85 kDa, PKB/Akt is 60 kDa, GSK-3 is approximately 51/46 kDa, and phospho-GSK-3 is 46 kDa.
Phosphorylation of GSK-3 and PKB in intestinal tissue was measured as follows. Tissue was extracted as described above, except with added phosphatase inhibitor, sodium orthovanadate, to a final concentration of 2 mM. For phosphor-GSK-3, equal amounts (30–120 μg) of supernatant protein extracts were separated on a 9% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk in the TBS. Membranes were incubated with a primary antibody (rabbit polyclonal antibody recognizes both phospho-GSK-3/?, Ser21/Ser9) (Santa Cruz Biotechnology) and then secondary antibody as described above for GSK-3 protein. PKB-Ser473 and PKB-Thr308 were measured by Western blotting after immunoprecipitation with anti-PKB agarose beads. Extracts were combined with 4 μg of anti-PKB, PH Domain, and agarose (mouse monoclonal IgG; Upstate, Charlottesville, VA) and incubated overnight at 4 C. Agarose beads were centrifuged at 12,000 rpm, washed three times with ice-cold 1x PBS, resuspended in 60 μl 2x Laemmli sample buffer, and boiled for 5 min. Ten microliters of immunoprecipitation product (roughly 160 μg of total protein) were separated on a 9% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk in the TBS. Membranes were incubated with a primary antibody [rabbit anti-phospho-PKB1-S473, 1:500 (R&D Systems, Minneapolis, MN), and rabbit anti-phospho-PKB1-T308, 1:500 (Cell Signaling Technology, Inc., Beverly, MA)] diluted in TBS with added Tween 20 solution (0.1%). Membranes were incubated with a secondary antibody [goat antirabbit IgG1-HRP, 1:5000 (Santa Cruz Biotechnology)]. The membranes were then stripped (Restore Western blot stripping buffer, Pierce) at 37 C for 15 min, washed with TBS with added Tween 20 solution (0.1%) three times, and reprobed with mouse anti-PKB1 antibody (1:1000; Santa Cruz Biotechnology) diluted in 5% milk plus TBS with added Tween 20 solution (0.1%). The molecular mass of PKB is 60 kDa. The quantity of the two PKB-Ser473 and PKB-Thr308 forms was expressed relative to the amount of PKB protein on each membrane.
eNOS.
Equal amounts (100 μg) of supernatant protein extracts were separated on a 7.5% denatured SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blocked with 10 ml Pierce Superblock solution and then incubated with a primary antibody [rabbit polyclonal antibody against human eNOS (H-159) (Santa Cruz Biotechnology)] diluted 1:500 in the TBS with added Tween 20 solution (0.1%). The apparent molecular mass of the eNOS band was 124 kDa. Membranes were incubated with a secondary antibody [antirabbit IgG conjugated with biotin, 1:25,000 (Santa-Cruz Biotechnology)] enhanced with Neutravidin-HRP (1:25,000; Pierce).
All Western blots were allowed to react with HRP substrate (ECL-plus, Amersham Biosciences, Piscataway, NJ) and then exposed to x-ray film for 30–120 sec, and the image was scanned and quantified by ImageQuant 5.0 software (Molecular Dynamics, Amersham Biosciences, Sunnyvale, CA). All Western blots were run with six pigs from each treatment group and used for statistical analysis, and treatment means and SEs are shown as bar graphs. Western blots of pooled samples from six pigs in each treatment also are shown in each figure for the respective proteins.
Real-time RT-PCR of eNOS mRNA.
Total RNA was extracted from the frozen porcine jejunal tissue samples using RNeasy Mini kit (QIAGEN Inc., Valencia, CA). After the RNA concentration was quantified, approximately 25–50 ng of total RNA for each reaction was used for real-time qRT-PCR. Primers and probe of porcine eNOS for real-time RT-PCR was based on the sequence of porcine eNOS mRNA (GenBank accession no. AY266137), i.e. using eNOS TaqMan TAMRA probe (100 nM), 6FAM-CTT CAC CGC GTT GGC CAC TTC CT-TAMRA; eNOS forward primer (10 mM), GGC ATC GCC AGA AAG AC; and eNOS reverse primer (10 mM), CAT CAC GGT GCC CAT GAG T. Primers and probe of ribosomal RNA (18S rRNA, Applied Biosystems, Foster City, CA) were used as an internal control. Assays were performed in triplicate with an ABI Prism 7700 sequence detector (Applied Biosystems). Data were normalized to 18S rRNA.
GLP-2 RIA
After 6 d of full TPN or TPN plus GLP-2 treatment and before bolus injections of BrdU and [13C]phenylalanine, blood samples were collected in EDTA tubes and centrifuged at 3000 x g at 4 C, and plasma was frozen immediately in liquid nitrogen. Plasma GLP-2 (1–33) concentrations were quantified by RIA as described previously (3, 14). This assay recognizes both the human and porcine GLP-2 peptides. The clearance rates for plasma GLP-2 were calculated using the following equation, CR = IR/Cp, where CR is the clearance rate (liter·kg–1·h–1), IR is the GLP-2 infusion rate (pmol·kg–1·h–1), and Cp is the steady-state plasma GLP-2 concentration (pmol/liter).
Statistical analysis
Data for the four treatment groups were analyzed using Minitab statistical software (Minitab Inc., State College, PA). Data were first analyzed by one-way ANOVA with GLP-2 infusion dose as a main effect, followed by a Tukey’s means comparison test. Means comparisons were done specifically to test for statistical differences between the control group and the three GLP-2 infusion rate groups. In some cases, data were analyzed using multiple regression analysis, and best-fit lines were generated using Origin software program (Microcal Software Inc., Northampton, MA). Because of the general lack of response to GLP-2, measurements of tissue mass and protein synthesis in the stomach, liver, spleen, and muscle tissue were obtained only for control and high GLP-2 groups; these data were analyzed by ANOVA using Minitab statistical software. Results are expressed as means and their respective SEs, and a P value less than 0.05 was considered statistically significant.
Results
There were no significant differences among the four treatment groups in either final body weight or daily body weight gain. The mean final body weights (kg) were 2.49 ± 0.16 and 2.64 ± 0.14, and the daily body weight gains (g/d) were 60.0 ± 4.5 and 69.7 ± 3.9 in the control and high GLP-2 groups, respectively. An important objective in this study was to target the GLP-2 infusion rates to achieve circulating plasma GLP-2 concentrations within the physiological range and the higher pharmacological range based on previously reported studies. The plasma GLP-2 (1–33) concentrations (pmol/liter) in all three GLP-2 treatment groups were significantly higher than in the TPN control group (31.5 ± 4.0) (Fig. 1). The mean plasma GLP-2 concentration in the low GLP-2 infusion group (166 ± 25) was higher than we observed previously in enterally fed pigs (4, 7). The mean plasma GLP-2 concentration was 75.4 ± 7.0 pmol/liter in a group of formula-fed piglets studied in parallel to this study; the piglets were the same age and body weight and the plasma GLP-2 was determined in the same assay conditions. Interestingly, the estimated plasma GLP-2 (1–33) clearance rates (liter·kg–1·h–1) were not different, being 0.697 ± 0.122, 0.620 ± 0.122, and 0.841 ± 0.089 for the low (2.5 nmol·kg–1·d–1), medium (5.0 nmol nmol·kg–1·d–1), and high (10.0 nmol·kg–1·d–1) GLP-2 groups, respectively.
FIG. 1. Plasma GLP-2 (1–33) concentrations in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. Results are means ± SEM for n = 7–10 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
Based on Tukey’s means comparison, the medium and high, but not the low, GLP-2 treatment groups had significantly higher intestinal mass, protein and DNA content, villus height and area, and crypt cell number compared with the control group (Table 1). Similar increases associated with GLP-2 infusion rate were observed in these latter endpoints in the distal small intestine segment (results not shown). There were no significant treatment differences in intestinal length or muscularis thickness. In addition, there were no significant treatment differences in the weight (Table 2) or the protein and DNA content (data not shown) of the stomach, liver, or spleen.
TABLE 1. GLP-2 infusion increases intestinal mass, protein, and DNA content in TPN-fed neonatal piglets
TABLE 2. Growth and protein synthesis rates in stomach, liver, spleen, and skeletal muscle in response to GLP-2 infusion
To determine whether the changes in intestinal protein mass were mediated by synthesis or degradation, we measured the FSR. The FSR in the intestine was significantly (P < 0.01) increased by GLP-2 treatment, but only the high GLP-2 treatment was different (P < 0.05) from the control group (Fig. 2). In contrast, stomach protein synthesis was significantly lower (P < 0.05) in the high GLP-2 than the control group, whereas protein synthesis in the liver, spleen, and muscle were not affected by GLP-2 (Table 2). Representative images of hematoxylin and eosin and BrdU histochemistry in intestinal sections from control and high GLP-2 pigs are shown (Fig. 3). The rates of crypt and villus epithelial cell apoptosis and crypt cell proliferation were measured to explain the GLP-2-induced increase in intestinal DNA content (Fig. 4). GLP-2 dose-dependently suppressed (P < 0.05) apoptosis in both crypt and villus cell compartments, even at the lowest GLP-2 infusion rate. However, crypt cell proliferation, expressed as a percentage (Fig. 4) and total number of BrdU-positive cells per crypt (data not shown), was increased only at the highest GLP-2 infusion rate, consistent with the dose response in protein synthesis.
FIG. 2. GLP-2 infusion increases intestinal protein synthesis. A, Multiple regression analysis indicated a highly significant positive relationship between protein synthesis and plasma GLP-2 concentration (y = 61.1 + 0.0042x + 3.061E –5x2; R2 = 0.781; P < 0.001; n = 36). B, Mean intestinal fractional protein synthesis rates in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. Results are means ± SEM for n = 7–8 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 3. Histological images of BrdU-stained (A and B; x200 magnification) and hematoxylin- and eosin-stained (C and D; x60 magnification) jejunal tissue sections from TPN-fed piglets infused with 0.0 (A and C) or 10.0 (B and D) nmol·kg–1·d–1 of human GLP-2 for 7 d. In A and B, BrdU-positive crypt cells show dark brown staining (black arrows) with gray hematoxylin-counterstained nuclei.
FIG. 4. GLP-2 infusion decreases apoptosis and increases cell proliferation. A–C, Intestinal apoptosis rates in the villus (A) and crypt (B) compartments and crypt cell proliferation rates (C) in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. Results are means ± SEM for n = 7–9 pigs per group. Differences between control and specific GLP-2-treated groupsare based on Tukey’s means comparison; *, P < 0.05.
The intestinal activities of caspase-3 and -6 (Fig. 5) and protein expression of caspase-3 (Fig. 6) and Bcl-2 (Fig. 7) were measured given their integral role in intestinal epithelial cell apoptosis. Similar to apoptosis rates, the activities of caspase-3 and -6 were dose-dependently (P < 0.05) suppressed with medium and high GLP-2 infusion rates, yet we found no difference between the control and low GLP-2 groups. GLP-2 also dose-dependently increased the abundance of the procaspase-3 form and decreased the abundance of the active caspase-3 form, resulting in a marked suppression of the active to inactive ratio of the caspase-3 enzyme protein. Consistent with increased cell survival, the intestinal expression of Bcl-2 protein was significantly (P < 0.05) increased with increasing GLP-2 infusion rates. Furthermore, regression analysis indicated that the ratio of active to inactive caspase-3 was effectively suppressed (P < 0.001) and Bcl-2 protein expression increased with increasing circulating plasma GLP-2 concentrations, especially within the physiological range (<200 pM; Fig. 7). Regression analysis indicated that Bcl-2 protein abundance was significantly and negatively correlated (R2 = 0. 74; P < 0.001) to the amount of active caspase-3. Thus, based on either the suppression of active caspase-3 or stimulation of Bcl-2 expression, the maximal cell survival response occurred in the pharmacological circulating GLP-2 concentration range of 600–800 pM.
FIG. 5. GLP-2 infusion decreases intestinal caspase-3 (A) and caspase-6 (B) activities in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. Results are means ± SEM for n = 7–8 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 6. A, GLP-2 decreases the ratio of active caspase-3 (lower band, 19 kDa) to procaspase-3 (upper band, 34 kDa) in the intestine (Western blot) of TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. B, Quantitative analysis of procaspase-3 (solid bars) and active caspase-3 (hatched bars) protein abundance expressed in arbitrary units. Percentage values show the calculated ratio of active caspase-3 to procaspase-3. Results are means ± SEM for n = 6 pigs per group. Differences in procaspase-3, active caspase-3, and percentage of active/procaspase-3 abundance between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 7. GLP-2 increases the intestinal expression of Bcl-2 protein of TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. A, Western blot result of Bcl-2 expression in pooled intestinal tissue extracts from piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. B, Quantitative results of Bcl-2 expression expressed in arbitrary units, where results are means ± SEM for n = 6 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05. C, Relationship between activated caspase-3 ratio (, solid line) and Bcl-2 (, dotted line) expression with plasma GLP-2 concentrations (C). Multiple regression analysis indicated the following equation: y = 37.1 –0.069x + 0.0004x2; R2 = 0.73; P < 0.001; n = 24 for activated caspase-3 ratio; and y = 68.7 + 1.331x –0.00078x2; R2 = 0.89; P < 0.001; n = 24 for Bcl-2 expression.
We next measured various upstream signaling molecules reported to be involved in the inhibition of apoptosis and particularly GLP-2 receptor signaling, including PI3-kinase, PKB, and GSK-3. GLP-2 infusion did not affect the protein expression of either PI3-kinase or PKB (data not shown). However, the medium and high GLP-2 doses significantly (P < 0.05) increased the phosphorylation of both PKB (Fig. 8) and GSK-3 (Fig. 9). Moreover, we found that the relative abundance of both Ser473 and Thr308 phospho-PKB was increased by all three GLP-2 infusion rates. We also measured the intestinal expression of eNOS, which is responsible for vascular nitric oxide release and may be involved in the acute (4-h) GLP-2-induction of intestinal blood flow reported previously (14). Consistent with previous work, we found that both eNOS mRNA and protein (Fig. 10) expression were increased (P < 0.05) and sustained even after a week of continuous GLP-2 infusion, but only at the highest GLP-2 infusion rate.
FIG. 8. GLP-2 increases the phosphorylation but not the abundance of intestinal PKB protein in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. A, Results of PKB-Ser473, PKB-Thr308, and PKB protein from Western blot of pooled samples. B, Quantitative results of PKB-Ser473, PKB-Thr308 abundance expressed relative to PKB protein. Results are means ± SEM for n = 6 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 9. GLP-2 increases the phosphorylation but not the abundance of intestinal GSK-3 protein in TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. A, Results of GSK-3 protein and phospho-GSK abundance in pooled intestinal samples. B, Quantitative results of phospho-GSK abundance. Results are means ± SEM for n = 6 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
FIG. 10. GLP-2 increases the intestinal expression of eNOS mRNA of TPN-fed piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. A, Western blot result of eNOS expression in pooled intestinal tissue extracts from piglets infused with 0.0, 2.5, 5.0, and 10.0 nmol·kg–1·d–1 of human GLP-2 for 7 d. B, Quantitative results of eNOS expression expressed in arbitrary units, where results are means ± SEM for n = 6 pigs per group. C, Results of RT-PCR analysis of eNOS mRNA abundance expressed relative to 18S mRNA abundance in intestinal tissue. Results are means ± SEM for n = 6 pigs per group. Differences between control and specific GLP-2-treated groups are based on Tukey’s means comparison; *, P < 0.05.
Discussion
The primary aim of this study was to determine intestinal responsiveness to various GLP-2 infusion rates that were targeted to produce circulating GLP-2 concentrations within the physiological and pharmacological range in TPN-fed piglets. Plasma GLP-2 concentrations considered to be in the physiological range are based on reports in enterally fed piglets (50–100 pM) (4, 7). A key finding was that the activation of intestinal epithelial cell survival, based on apoptosis and caspase-3 expression, occurred at relatively low circulating GLP-2 concentrations, whereas only high GLP-2 concentrations induced the stimulation of cell proliferation and protein synthesis. The results suggest that the activation of intestinal epithelial survival is more sensitive than cell proliferation to circulating GLP-2 concentrations in neonatal piglets. The combined effect of these cell kinetic responses resulted in dose-dependent increases in intestinal DNA mass and villus growth. This study also provides the first in vivo evidence that GLP-2 activates intestinal cell survival in association with induction of PKB and GSK-3 phosphorylation and Bcl-2 expression.
A key unresolved issue with respect to GLP-2 function is its physiological relevance to the well-documented intestinal trophic response to enteral nutrition. In the current study, we sought to examine the effect of reproducing a physiological circulating GLP-2 concentration, typical for enterally fed pigs (50–100 pM) (4, 7), but in a model of the TPN-fed piglet where the plasma GLP-2 concentration and intestinal mucosal growth are relatively suppressed because of the absence of luminal nutrition. Based on regression analysis, we observed that key endpoints of intestinal growth, namely mass and mucosal morphology, increased significantly with increasing infusion rate and circulating concentrations of GLP-2. However, means comparison analysis indicated that intestinal trophic effect of GLP-2 occurred mainly at the medium and high GLP-2 treatment groups, when compared with the control group. It is notable that the plasma GLP-2 concentrations in the medium and high GLP-2 groups are within the range (400–1200 pM) of those reported in GLP-2-treated short-bowel patients (27). Thus, in the absence of any enteral nutrition, artificially increasing the circulating GLP-2 concentration above the physiological range (i.e. 75 pM enteral fed) appears to be insufficient alone to induce intestinal growth in the neonatal piglet. Therefore, it appears that the intestinal trophic response to physiological GLP-2 concentrations requires the presence of additional contributing factors, such as luminal nutrients or other gut hormones. It is possible that the TPN decreases the intestinal GLP-2 responsiveness via disruption of GLP-2R signaling pathways. In addition, the increased circulating GLP-2 concentrations that we achieved in TPN-fed piglets may not reproduce the local tissue concentration, which occurs with enteral feeding perhaps because of high activity of dipeptidyl peptidase IV enzyme. Thus, under the conditions of TPN, producing a trophic effect may require a higher plasma level of GLP-2 than what otherwise might occur during enteral feeding.
An especially important finding in this study was the differential responsiveness of cell survival and proliferation to the circulating GLP-2 concentration. We previously showed that a pharmacological GLP-2 infusion rate prevented intestinal mucosal atrophy in TPN-fed piglets by decreasing apoptosis, but without a stimulation of cell proliferation (7). This was somewhat in contrast to several reports demonstrating that GLP-2 treatment both decreased intestinal apoptosis and increased crypt cell proliferation in rodents (12, 13) and cultured cells (18, 19, 21). In the current study, we observed significant decreases in apoptosis and active caspase-3 expression at all GLP-2 infusion rates; this was especially evident in the inverse, nonlinear response between active caspase-3 expression and plasma GLP-2. Activated caspase-3 is a key executioner protease responsible for most of the cellular destruction during apoptosis. Moreover, we found that both caspase-3 and caspase-6 activities were suppressed in parallel with epithelial cell apoptosis rate, consistent with reports showing their rapid activation in intestinal epithelial cells during detachment-induced apoptosis (28). However, contrary to our previous study, we found that GLP-2 increased crypt cell proliferation and protein synthesis, but only at the highest infusion rate and circulating concentrations of GLP-2. This differential responsiveness of cell death vs. cell growth mechanisms may be model dependent, because TPN is marked by significant mucosal atrophy compared with the fed condition. These results suggest that intestinal cell survival mechanisms are more sensitive than cell proliferation to the circulating GLP-2 concentration in neonatal piglets.
The differential responsiveness of intestinal cell survival and proliferation to GLP-2 has been observed in cultured cells and may be a consequence of different intracellular signaling pathways. The nature of intestinal GLP-2R signaling in vivo is largely unknown because of uncertainty about its precise cellular localization; reports indicate the presence of the receptor in human endocrine cells (10) and murine enteric neurons (17). However, studies in fibroblasts transfected with the GLP-2R have shown that much lower GLP-2 concentrations are necessary to inhibit apoptosis (20 nM) (19, 20) than to stimulate cell proliferation (100 nM) (18). These studies also have shown that the GLP-2-mediated inhibition of apoptosis in GLP-2R-transfected BHK cells occurs via a cAMP-dependent, protein kinase A pathway that is associated with increased GSK-3 phosphorylation (19, 20) but does not involve activation of PI3-kinase or PKB. This observation is consistent with evidence that phosphorylation-dependent inhibition of GSK-3 by G protein-coupled receptors seems to be mediated by PKA, whereas PI3-kinase/PKB-mediated GSK-3 inhibition is associated with stimulation by growth factors, such as insulin (29). Contrary to this general idea are studies with cultured colon carcinoma cells demonstrating that GLP-2 increased cell proliferation at supraphysiological (10 nM) GLP-2 concentrations and that this response was associated with increased MAPK kinase phosphorylation and blocked by inhibitors of MAPK and PI3-kinase. In the current study, we found that none of the GLP-2 treatments increased PI3-kinase, PKB, or GSK-3 protein expression. However, both PKB and GSK-3 phosphorylation were significantly increased in the GLP-2-treated groups. Therefore, our results provide the first in vivo evidence suggesting that the GLP-2-induced suppression of intestinal apoptosis and stimulation of cell proliferation is associated with PKB activation and GSK-3 inhibition.
Recent studies have linked the cardioprotective actions of GSK-3 inhibition to the delayed activation of the mitochondrial permeability transition, a key regulator of apoptosis (30). GSK-3 inhibition also is a prerequisite for caspase-3 activation associated with thapsigargin-induced endoplasmic reticulum stress (31). Mitochondrial dysfunction is generally associated with the intrinsic apoptotic pathway and is largely controlled by the activity of Bcl-2 family proteins. The Bcl-2-related member, Bad, is a proapoptotic downstream target of PI3-kinase/PKB phosphorylation and has been associated with GSK-3 inhibition of apoptosis (32). GLP-2 stimulates Bad phosphorylation in association with the inhibition of GSK-3 and apoptosis in GLP-2R-transfected fibroblasts (20). We did not measure the protein expression or phosphorylation of Bad in intestinal extracts; however, we found that the expression of the anti-apoptotic protein, Bcl-2, was dose-dependently increased with GLP-2 infusion. Bcl-2 deficiency leads to intestinal dysfunction and is positively correlated with the attenuation of apoptosis (33, 34). Overexpression of Bcl-2 inhibits ischemia-reperfusion-induced apoptosis in the intestinal epithelium in transgenic mice (35). The anti-apoptotic action of Bcl-2 occurs via stabilization of mitochondrial membrane integrity and prevents cytochrome c release, which is a critical early event in apoptosis. Bcl-2 is also a target of caspase-3, whereby caspase-3 activation promotes Bcl-2 cleavage in a positive feedback loop, which accelerates the release of mitochondrial cytochrome c and eventually cell death (36). Thus, our results are consistent with the idea that GLP-2-induced suppression of caspase-3 activity limits the degradation of Bcl-2, leading to increased cellular availability for Bcl-2 stabilization of mitochondrial function and increased cell survival. Whether the inhibition of GSK-3 is mechanistically linked to the GLP-2-associated increase in Bcl-2 expression warrants further study.
Nitric oxide (NO) is a ubiquitous, cell-permeable anti-apoptotic signaling molecule involved in a variety of cell functions, including vasodilatation, apoptosis, and inflammation (37, 38). We previously demonstrated that the GLP-2-mediated up-regulation of intestinal blood flow and glucose uptake is NO dependent and associated with increased expression of eNOS. However, the role of NO in the GLP-2-mediated intestinal survival and proliferation responses is unknown. Nitric oxide protects the structure and function of enterocytes under stress (e.g. oxidative stress, injury, and colitis) (39, 40, 41) and inhibits mitochondrial dysfunction-induced apoptosis (42). Studies also indicate that NO can inhibit apoptosis by nitrosylation-mediated suppression of caspase-3 activation (37). It is also of interest that eNOS is not only a PKB substrate but also is activated in a protein kinase A (PKA)-dependent manner (43, 44). Consistent with our previous study showing that GLP-2 acutely up-regulates intestinal eNOS expression, we found that eNOS expression was significantly increased, but only at the high GLP-2 infusion rate. Additional studies are warranted to determine whether GLP-2 induces eNOS expression via a GLP-2R- mediated pathway involving PKA and PKB and whether the GLP-2-mediated increase in intestinal epithelial cell survival, proliferation, or protein synthesis is NO dependent.
Another aim of this study was to quantify the growth and protein synthesis response to GLP-2 in other tissues, besides the small intestine. The expression of the GLP-2R mRNA is largely confined to the gastrointestinal tissues, including stomach and intestine, but not the liver, spleen, or skeletal muscle (10, 11); however, the reports of GLP-2-mediated inhibition of bone resorption may be linked to GLP-2R expression in peripheral sites. Consistent with the expression pattern, we found no change in either growth (protein and DNA content) or protein synthesis in the liver, spleen, or skeletal muscle in response to GLP-2 treatment. However, the high GLP-2 infusion rate significantly decreased the stomach protein synthesis rate, despite the lack of effect on protein and DNA mass. This result is consistent with reports that GLP-2 also suppresses gastric secretion and motor function (45).
In summary, the current study extends our previous finding that GLP-2 prevents mucosal atrophy in TPN-fed piglets by showing the dose dependence of this response and the cellular signals potentially involved. We found that the GLP-2 dose-dependent stimulation of intestinal growth is mediated largely by stimulation of epithelial cell survival and to a lesser extent by cell proliferation, and this leads to preserved villus length. Although our results suggest that the stimulation of intestinal cell survival occurs within the physiological range of circulating GLP-2, it did not translate into significant increases in intestinal mucosal structure and mass. This indicates that reproducing a physiological circulating concentration of GLP-2 does not produce the intestinal trophic effect observed with enteral nutrition. Thus, if GLP-2 has a physiological role in enteral nutrient-mediated intestinal growth, it appears to require other factors such as luminal nutrients, other gut hormones, or neural stimulation. In addition, we showed that GLP-2 stimulates intestinal epithelial cell survival and proliferation in association with suppression of caspase-3 and induction of PKB and GSK-3 phosphorylation and Bcl-2 expression. This study did not attempt to localize the activation of these signaling molecules to specific cells within the mucosa, namely the epithelial cells where we observed changes in apoptosis and cell proliferation. However, given the current evidence that the GLP-2R expression is confined to a limited cell population (enteroendocrine cells or enteric neurons), we postulate that the changes observed in these signaling pathways occur in epithelial cells downstream of GLP-2R activation. Therefore, it will be important to establish whether the GLP-2R and activation of these signaling molecules occurs within the same cell type or whether GLP-2R signaling occurs via a heterotypic, intercellular signaling pathway.
Acknowledgments
We acknowledge the technical assistance of Xiaofeng Lu.
References
Drucker DJ 2002 Gut adaptation and the glucagon-like peptides. Gut 50:428–435
Burrin DG, Stoll B, Guan X 2003 Glucagon-like peptide 2 function in domestic animals. Domest Anim Endocrinol 24:103–122
Burrin DG, Stoll B, Jiang RH, Chang XY, Hartmann B, Holst JJ, Greeley GH, Reeds PJ 2000 Minimal enteral nutrient requirements for intestinal growth in neonatal piglets: how much is enough? Am J Clin Nutr 71:1603–1610
van Goudoever JB, Stoll B, Hartmann B, Holst JJ, Reeds PJ, Burrin DG 2001 Secretion of trophic gut peptides is not different in bolus- and continuously fed piglets. J Nutr 131:729–732
Xiao QA, Boushey RP, Drucker DJ, Brubaker PL 1999 Secretion of the intestinotropic hormone glucagon-like peptide 2 is differentially regulated by nutrients in humans. Gastroenterology 117:99–105
Dahly EM, Gillingham MB, Guo Z, Murali SG, Nelson DW, Holst JJ, Ney DM 2003 Role of luminal nutrients and endogenous GLP-2 in intestinal adaptation to mid-small bowel resection. Am J Physiol Gastrointest Liver Physiol 284:G670–G682
Burrin DG, Stoll B, Jiang R, Petersen Y, Elnif J, Buddington RK, Schmidt M, Holst JJ, Hartmann B, Sangild PT 2000 GLP-2 stimulates intestinal growth in premature TPN-fed pigs by suppressing proteolysis and apoptosis. Am J Physiol Gastrointest Liver Physiol 279:G1249–G1256
Chance WT, Foley-Nelson T, Thomas I, Balasubramaniam A 1997 Prevention of parenteral nutrition-induced gut hypoplasia by coinfusion of glucagon-like peptide-2. Am J Physiol 273:G559–G563
Munroe DG, Gupta AK, Kooshesh F, Vyas TB, Rizkalla G, Wang H, Demchyshyn L, Yang ZJ, Kamboj RK, Chen HY, McCallum K, Sumner-Smith M, Drucker DJ, Crivici A 1999 Prototypic G protein-coupled receptor for the intestinotrophic factor glucagon-like peptide 2. Proc Natl Acad Sci USA 96:1569–1573
Yusta B, Huang L, Munroe D, Wolff G, Fantaske R, Sharma S, Demchyshyn L, Asa SL, Drucker DJ 2000 Enteroendocrine localization of GLP-2 receptor expression in humans and rodents. Gastroenterology 119:744–755
Petersen YM, Burrin DG, Sangild PT 2001 GLP-2 has differential effects on small intestine growth and function in fetal and neonatal pigs. Am J Physiol Regul Integr Comp Physiol 281:R1986–R1993
Drucker DJ, Ehrlich P, Asa SL, Brubaker PL 1996 Induction of intestinal epithelial proliferation by glucagon-like peptide 2. Proc Natl Acad Sci USA 93:7911–7916
Tsai CH, Hill M, Asa SL, Brubaker PL, Drucker DJ 1997 Intestinal growth-promoting properties of glucagon-like peptide-2 in mice. Am J Physiol Endocrinol Metab 36:E77–E84
Guan XF, Stoll B, Lu XF, Tappenden KA, Holst JJ, Hartmann B, Burrin DG 2003 GLP-2-mediated up-regulation of intestinal blood flow and glucose uptake is nitric oxide-dependent in TPN-fed piglets. Gastroenterology 125:136–147
Lovshin J, Yusta B, Iliopoulos I, Migirdicyan A, Dableh L, Brubaker PL, Drucker DJ 2000 Ontogeny of the glucagon-like peptide-2 receptor axis in the developing rat intestine. Endocrinology 141:4194–4201
Litvak DA, Hellmich MR, Evers BM, Banker NA, Townsend Jr CM 1998 Glucagon-like peptide 2 is a potent growth factor for small intestine and colon. J Gastrointest Surg 2:146–150
Bjerknes M, Cheng H 2001 Modulation of specific intestinal epithelial progenitors by enteric neurons. Proc Natl Acad Sci USA 98:12497–12502
Yusta B, Somwar R, Wang F, Munroe D, Grinstein S, Klip A, Drucker DJ 1999 Identification of glucagon-like peptide-2 (GLP-2)-activated signaling pathways in baby hamster kidney fibroblasts expressing the rat GLP-2 receptor. J Biol Chem 274:30459–30467
Yusta B, Boushey RP, Drucker DJ 2000 The glucagon-like peptide-2 receptor mediates direct inhibition of cellular apoptosis via a cAMP-dependent protein kinase-independent pathway. J Biol Chem 275:35345–35352
Yusta B, Estall J, Drucker DJ 2002 Glucagon-like peptide-2 receptor activation engages bad and glycogen synthase kinase-3 in a protein kinase A-dependent manner and prevents apoptosis following inhibition of phosphatidylinositol 3-kinase. J Biol Chem 277:24896–24906
Jasleen J, Ashley SW, Shimoda N, Zinner MJ, Whang EE 2002 Glucagon-like peptide 2 stimulates intestinal epithelial proliferation in vitro. Dig Dis Sci 47:1135–1140
Walsh NA, Yusta B, Dacambra MP, Anini Y, Drucker DJ, Brubaker PL 2003 Glucagon-like peptide-2 receptor activation in the rat intestinal mucosa. Endocrinology 144:4385–4392
Li J, Bombeck CA, Yang S, Kim YM, Billiar TR 1999 Nitric oxide suppresses apoptosis via interrupting caspase activation and mitochondrial dysfunction in cultured hepatocytes. J Biol Chem 274:17325–17333
Rajendram R, Marway JS, Mantle D, Peters TJ, Preedy VR 2003 Skeletal muscle and jejunal protein synthesis in normal and ethanol-treated rats: the effect of the nitric oxide synthase inhibitors, L-omega-nitro-L-arginine methyl ester and N(G)-nitro-L-arginine in vivo. Metabolism 52:397–401
Niinikoski H, Stoll B, Guan X, Kansagra K, Lambert BD, Stephens J, Hartmann B, Holst JJ, Burrin DG 2004 Onset of small intestinal atrophy is associated with reduced intestinal blood flow in TPN-fed neonatal piglets. J Nutr 134:1467–1474
Stoll B, Chang X, Fan MZ, Reeds PJ, Burrin DG 2000 Enteral nutrient intake level determines intestinal protein synthesis and accretion rates in neonatal pigs. Am J Physiol Gastrointest Liver Physiol 279:G288–G294
Jeppesen PB, Hartmann B, Thulesen J, Graff J, Lohmann J, Hansen BS, Tofteng F, Poulsen SS, Madsen JL, Holst JJ, Mortensen PB 2001 Glucagon-like peptide 2 improves nutrient absorption and nutritional status in short-bowel patients with no colon. Gastroenterology 120:806–815
Grossmann J, Mohr S, Lapentina EG, Fiocchi C, Levine AD 1998 Sequential and rapid activation of select caspases during apoptosis of normal intestinal epithelial cells. Am J Physiol 274:G1117–G1124
Fang X, Yu SX, Lu Y, Bast Jr RC, Woodgett JR, Mills GB 2000 Phosphorylation and inactivation of glycogen synthase kinase 3 by protein kinase A. Proc Natl Acad Sci USA 97:11960–11965
Juhaszova M, Zorov DB, Kim SH, Pepe S, Fu Q, Fishbein KW, Ziman BD, Wang S, Ytrehus K, Antos CL, Olson EN, Sollott SJ 2004 Glycogen synthase kinase-3? mediates convergence of protection signaling to inhibit the mitochondrial permeability transition pore. J Clin Invest 113:1535–1549
Song L, De Sarno P, Jope RS 2002 Central role of glycogen synthase kinase-3? in endoplasmic reticulum stress-induced caspase-3 activation. J Biol Chem 277:44701–44708
Stoica BA, Movsesyan VA, Lea PM, Faden AI 2003 Ceramide-induced neuronal apoptosis is associated with dephosphorylation of Akt, BAD, FKHR, GSK-3?, and induction of the mitochondrial-dependent intrinsic caspase pathway. Mol Cell Neurosci 22:365–382
Kamada S, Shimono A, Shinto Y, Tsujimura T, Takahashi T, Noda T, Kitamura Y, Kondoh H, Tsujimoto Y 1995 Bcl-2 deficiency in mice leads to pleiotropic abnormalities: accelerated lymphoid cell death in thymus and spleen, polycystic kidney, hair hypopigmentation, and distorted small intestine. Cancer Res 55:354–359[Abstract]
Merritt AJ, Potten CS, Watson AJ, Loh DY, Nakayama K, Nakayama K, Hickman JA 1995 Differential expression of bcl-2 in intestinal epithelia: correlation with attenuation of apoptosis in colonic crypts and the incidence of colonic neoplasia. J Cell Sci 108:2261–2271[Abstract]
Coopersmith CM, O’Donnell D, Gordon JI 1999 Bcl-2 inhibits ischemia-reperfusion-induced apoptosis in the intestinal epithelium of transgenic mice. Am J Physiol 276:G677–G686
Kirsch DG, Doseff A, Chau BN, Lim DS, Souza-Pinto NC, Hansford R, Kastan MB, Lazebnik YA, Hardwick JM 1999 Caspase-3-dependent cleavage of Bcl-2 promotes release of cytochrome c. J Biol Chem 274:21155–21161
Kim PK, Kwon YG, Chung HT, Kim YM 2002 Regulation of caspases by nitric oxide. Ann NY Acad Sci 962:42–52
Shah V, Lyford G, Gores G, Farrugia G 2004 Nitric oxide in gastrointestinal health and disease. Gastroenterology 126:903–913
Thomas S, Ramachandran A, Patra S, Vidyasagar S, Balasubramanian KA 2001 Nitric oxide protects the intestine from the damage induced by laparotomy and gut manipulation. J Surg Res 99:25–32
Sasaki M, Bharwani S, Jordan P, Elrod JW, Grisham MB, Jackson TH, Lefer DJ, Alexander JS 2003 Increased disease activity in eNOS-deficient mice in experimental colitis. Free Radic Biol Med 35:1679–1687
Bersimbaev RI, Yugai YE, Hanson PJ, Tzoy IG 2001 Effect of nitric oxide on apoptotic activity in the rat gastrointestinal tract. Eur J Pharmacol 423:9–16
Kim YM, Kim TH, Seol DW, Talanian RV, Billiar TR 1998 Nitric oxide suppression of apoptosis occurs in association with an inhibition of Bcl-2 cleavage and cytochrome c release. J Biol Chem 273:31437–31441
Fulton D, Gratton JP, McCabe TJ, Fontana J, Fujio Y, Walsh K, Franke TF, Papapetropoulos A, Sessa WC 1999 Regulation of endothelium-derived nitric oxide production by the protein kinase Akt. Nature 399:597–601
Boo YC, Sorescu G, Boyd N, Shiojima I, Walsh K, Du J, Jo H 2002 Shear stress stimulates phosphorylation of endothelial nitric-oxide synthase at Ser1179 by Akt-independent mechanisms: role of protein kinase A. J Biol Chem 277:3388–3396
Wojdemann M, Wettergren A, Hartmann B, Hilsted L, Holst JJ 1999 Inhibition of sham feeding-stimulated human gastric acid secretion by glucagon-like peptide-2. J Clin Endocrinol Metab 84:2513–2517(Douglas G. Burrin, Barbar)