We wished to determine whether exogenous glucagon-like peptide (GLP)-2 infusion stimulates intestinal growth in parenterally fed immature pigs. Piglets (106–108 days gestation) were given parenteral nutrient infusion (TPN), TPN + human GLP-2 (25 nmol · kg−1 · day−1), or sow's milk enterally (ENT) for 6 days. Intestinal protein synthesis was then measured in vivo after a bolus dose of [1-13C]phenylalanine, and degradation was calculated from the difference between protein accretion and synthesis. Crypt cell proliferation and apoptosis were measured in situ by 5-bromodeoxyuridine (BrdU) and terminal dUTP nick-end labeling (TUNEL), respectively. Intestinal protein and DNA accretion rates and villus heights were similar in GLP-2 and ENT pigs, and both were higher (P < 0.05) than in TPN pigs. GLP-2 decreased fractional protein degradation rate, whereas ENT increased fractional protein synthesis rate compared with TPN pigs. Percentage of TUNEL-positive cells in GLP-2 and ENT groups was 48 and 64% lower, respectively, than in TPN group (P < 0.05). However, ENT, but not GLP-2, increased percentage of BrdU-positive crypt cells above that in TPN piglets. We conclude that GLP-2 increases intestinal growth in premature, TPN-fed pigs by decreasing proteolysis and apoptosis, whereas enteral nutrition acts via increased protein synthesis and cell proliferation and decreased apoptosis.
- protein synthesis
- protein degradation
- total parenteral nutrition
- preterm infants
glucagon-like peptide(GLP)-2, along with GLP-1, represent the major secretory peptides derived from the posttranslational processing of the proglucagon gene expressed in the enteroendocrine L cells localized in the distal intestine (4, 14). Studies indicate that a primary stimulus for GLP-2 secretion, like that of GLP-1, is enteral nutrient intake, especially lipid and carbohydrate (13, 30). In addition, after massive small bowel resection the circulating concentrations of GLP-2 and other gut-derived peptides are significantly increased (2, 19, 26). Given that marked intestinal hyperplasia and growth also occur following enteral nutrient ingestion and intestinal resection, it has been postulated that increased secretion of GLP-2 provides a trophic signal that stimulates intestinal growth and adaptation. This is supported by a number of recent studies demonstrating that GLP-2 is an intestinal trophic peptide (8) and upregulates intestinal hexose transport (9, 15). Moreover, administration of GLP-2 can augment or restore growth under conditions of compromised intestinal function, including massive small bowel resection (23), total parenteral nutrition (TPN) (8), and enteritis (3).
A potentially important, yet unexplored, aspect of GLP-2 function is whether it plays a role in the regulation of intestinal growth and maturation during early ontogeny. In the immediate prenatal period, the pig intestine undergoes extensive functional maturation and growth (21). This rapid maturation continues just after birth and appears to be associated with enteral nutrient intake, although the age-related intestinal development may differ between immature and full-term animals (22). Likewise, intestinal growth is markedly suppressed in term newborn pigs after 6 days of TPN (11). Given that enteral nutrition is a critical stimulus of early neonatal intestinal growth and development, it is conceivable that the secretion and trophic actions of GLP-2 are involved. Studies in rodents suggest that proglucagon mRNA is expressed in the fetal intestine and that mucosal immunoreactive glucagon-like activity increases during the early neonatal period (16). Consistent with this, our recent studies (6) in neonatal pigs showed that GLP-2 secretion, as measured by the circulating concentration, is both substantially stimulated by and directly correlated with the level of enteral nutrient intake. More importantly, the stimulation of GLP-2 secretion by enteral nutrient intake level was correlated with intestinal mucosal growth. Thus, although the developing intestine appears capable of GLP-2 secretion in response to a nutrient stimulus, it is unknown whether the intestine in newborns is indeed responsive to GLP-2.
In the present study, we investigated whether the premature neonatal intestine is responsive to GLP-2 under conditions of TPN. By measuring intestinal growth, including villus height, cell proliferation, and apoptosis, we were able to determine whether the actions of GLP-2 in premature pigs are similar to those reported in weanling rodents. In addition, the impact of GLP-2 on the rates of intestinal protein synthesis and degradation was determined in vivo, because intestinal protein synthesis has been shown to be suppressed during TPN (11,25). Finally, we compared the intestinal response to GLP-2 infusion in parenterally fed neonatal pigs with that in pigs fed enterally to investigate the degree to which GLP-2 may mediate the trophic effects of enteral nutrients on the developing intestine.
MATERIALS AND METHODS
A total of 29 crossbred (Large White × Danish Landrace) pig from eight sows (Research Farm Sjaelland III) were used in the experiment. Twenty-one pigs from three sows were obtained by caesarean section at 106–108 days of gestation, and the pigs from each litter were assigned to receive TPN (n = 7), TPN plus GLP-2 (n = 8, GLP-2), or enteral sow's milk (n = 6, ENT) for 6 days after delivery. To estimate intestinal tissue growth and protein accretion after birth, another eight pigs from five sows, which served as an initial control group, were obtained by caesarean section as above and immediately killed after delivery for measurements of intestinal weight as described inMeasurement of in vivo protein synthesis. All procedures were approved by the National Committee on Animal Experimentation, Denmark.
The pregnant sows were sedated with azaperone (0.05 ml/kg im; Janssen, Beerse, Belgium). Anesthesia was induced with thiopental sodium (5 mg/kg iv; Abbott Laboratories, North Chicago, IL) and maintained with isoflurane (Abbott Laboratories) inhalation (1–2% in oxygen) after endotracheal intubation. Following uterine incision, each fetus was removed from the uterus after ligating and transecting the umbilical cord. The premature piglets were kept individually in infant incubators (Air-Shields, Hatboro, PA) maintained at 32–34°C, 80–100% moisture and with extra oxygen supply (1–2 l/min). While still anesthetized, the premature piglets that were assigned to the TPN and GLP-2 groups were fitted with two vascular catheters, one inserted in the dorsal aorta via the umbilical artery (infant feeding tube 4F; Portex, Kent, UK) and one in the umbilical vein (infant feeding tube 5F; PharmaPlast, Lynge, Denmark) via the transected cord vessels. Likewise, pigs assigned to the ENT group were fitted with a dorsal aorta catheter as above and also an orogastric tube (infant feeding tube 6F; PharmaPlast) to be used for enteral feeding. For each pig, the catheters were sutured to the cord and to the skin, and finally, a cotton body suit was fitted onto each pig to protect the catheters.
The TPN and GLP-2 groups received an elemental nutrient solution continuously via the umbilical venous catheter. The nutrient solution contained free amino acids (45.5 g amino acids/l; Vamin 18F), glucose (72.5 g/l), and lipid (30.7 g/l; Intralipid), all supplied by KABI Pharmacia (Copenhagen, Denmark). Mineral and vitamin solutions were added (in %: 2 Calcium-Sandoz, 0.6 Addiphos, 0.6 Peditrace, 1.0 Vitalipid, and 0.1 Soluvit; KABI Pharmacia). The nutrient concentrations were similar to those used and validated previously in studies with newborn TPN-fed pigs (29). The TPN solution was administered continuously from 6 h after birth, and, after an initial 2-day period in which the animals received 50% of the total intake, the animals received 550 kJ · kg−1 · day−1, 8 g amino acid · kg−1 · day−1, and a fluid intake of 170 ml · kg−1 · day−1 for the remainder of the experimental period. To provide some passive immunity, the TPN-fed piglets were given a total of 15 ml of maternal serum over the first 24 h via the arterial catheter. The serum was produced from maternal blood collected from a uterine vein at the time of surgery. Together with the TPN solution that was infused into the umbilical vein, the pigs received two daily 2-h infusions of human GLP-2 (12.5 nmol · kg−1 · 2 h−1; a generous gift from L. Thim, Novo Nordisk, Bagsværd, Denmark) or the vehicle (0.1% porcine serum albumin in 0.9% buffered saline) starting at 0800 and 1600. The ENT group received a continuous orogastric infusion of sow's colostrum (for 1 day) or milk (2–6 days) for 5 days at an hourly rate identical to that of the TPN solution providing ∼700 kJ · kg−1 · day−1 and 12 g protein · kg−1 · day−1. Each day during the 6-day period, the pigs were weighed to adjust their nutrient infusion rates and a 2-ml arterial blood sample was collected at 0800, before the GLP-2 and vehicle infusions.
Measurement of in vivo protein synthesis.
After 6 days of treatment, in vivo fractional tissue protein synthesis rates were measured by the flooding-dose technique as described previously (5). Briefly, pigs were given an intravenous injection of a bolus dose of l-phenylalanine (1.5 mmol/kg body wt) containing ∼0.60 mmol/kg body wt ofl-[13C]phenylalanine (98% [1-13C]phenylalanine; Cambridge Isotopes, Woburn, MA). Thirty minutes after isotope injection, pigs were anesthetized with an intravenous dose of pentobarbital (50 mg/kg body wt; Sigma Chemical, St. Louis, MO). The abdomen was opened and the entire small intestine distal to the ligament of Treitz was immediately flushed with ice-cold saline. The segment of small intestine proximal to the ligament of Treitz was removed and flushed with ice-cold saline. After flushing, the small intestine was divided into two equal portions; the proximal half was designated the jejunum, and the distal was designated the ileum. The weight of the small intestinal segments was determined, and then tissue samples were obtained from the proximal half of each segment. The samples were snap-frozen in liquid nitrogen and stored at −70°C until analysis for tracer enrichment, protein, and DNA as described previously (6, 25).
Blood and tissue preparation.
For isotopic analysis, whole blood and tissue samples were prepared and analyzed as described previously (25). Briefly, extracts of whole blood were obtained by first deproteinizing whole blood and then applying the supernatant to a cation exchange column (Dowex 50W×8 H+-form; Bio-Rad, Richmond, CA). Samples (∼1 g) of small intestinal tissue were homogenized, and aliquots were taken for analysis of protein and DNA as previously described (6). The homogenate was deproteinized with perchloric acid, and the acid-soluble (tissue-free pool) and acid-insoluble (protein-bound pool) hydrolyzed fractions were subjected to mass spectrometric analysis.
The isotopic enrichment of [1-13C]phenylalanine in whole blood and the intestinal tissue-free pool was determined by gas chromatography mass spectrometry method conducted with then-propyl ester heptafluorobutyramide derivative using methane-negative chemical ionization as previously described (24). The analyses were performed with a 5890 series II gas chromatograph linked to a model 5989B (Hewlett-Packard, Palo Alto, CA) quadruple mass spectrometer. The isotopic enrichment of phenylalanine was determined by monitoring ions at a mass-to-charge ratio of 383:386. The isotopic enrichment of [1-13C]phenylalanine in the protein-bound pool was determined by gas chromatography combustion isotope ratio mass spectrometry. Dried protein hydrolysate samples, previously purified by cation-exchange as described in Blood and tissue preparation, were incubated with 400 μl of methylation reagent (methanol:hydrogen bromide:2,2-dimethoxypropane in a ratio of 77:20:3) for 30 min at 110°C. The supernatant was evaporated under a nitrogen stream at 60°C. The methyl esters were incubated with 400 μl of fluoroacetylation reagent (methylene chloride:trifluoracetic anhydride in a 4:1 ratio) for 10 min at 130°C and left at room temperature for 2 h. Then the supernatant was evaporated under nitrogen at room temperature and the resulting trifluoroacetyl methyl ester derivative of phenylalanine was dissolved in 50–100 μl of ethylacetate and analyzed with a 5980 series II gas chromatograph (Hewlett-Packard) linked to a continuous flow gas isotope ratio mass spectrometer (model 20–20 stable isotope analyzer; Europa Scientific). All results were calculated using a matrix approach and expressed as mole percent excess.
A 2- to 3-cm segment of fresh tissue from the proximal region (i.e., within first 10% of segment) of jejunum and ileum segments was fixed in Carnoy's solution (60% ethanol, 30% glacial acetic acid, and 10% chloroform) for 24 h and transferred to 70% ethanol. Samples were embedded in paraffin, sectioned (5 μm), and stained with eosin and hematoxylin. The mean villus height, crypt depth, and muscularis thickness were quantified by a blinded observer using an Axiophot microscope (Zeiss) and NIH Image 1.60 (National Institutes of Health, Bethesda, MD) in at least 15 vertically well-oriented villus-crypt columns.
In vivo bromodeoxyuridine labeling.
In vivo crypt cell proliferation was measured as described previously (6). Briefly, pigs were given an intravenous injection of 5-bromodeoxyuridine (BrdU, 50 mg/kg; Sigma Aldrich, St. Louis, MO) 4 h before they were killed and intestinal tissue samples were collected. Carnoy's-fixed, paraffin-embedded sections were incubated with a mouse anti-BrdU/nuclease reagent (Amersham Pharmacia Biotech, Piscataway, NJ) and then with biotinylated universal secondary antibody (anti-mouse IgG2a; The Binding Site, San Diego, CA). BrdU-labeled cells were visualized using 3,3′-diaminobenzidine (DAB) substrate kit for peroxidase (Vector Labs, Burlingame, CA), and then slides were counterstained with 0.1% hematoxylin. The proportion of proliferating crypt cells was quantified by counting the number of BrdU-labeled nuclei in 15 vertically well-oriented crypts and expressing this as a percentage of total nuclei per crypt.
Apoptosis was detected by using the apoTACS TBL kit (Trevigen, Gaithersburg, MD), which is based on the TUNEL method using terminal deoxynucleotidyl-transferase-mediated dUTP nick-end labeling. Briefly, formalin-fixed, paraffin-embedded sections were incubated with proteinase K (5 μm), washed, and then incubated with a solution containing deoxynucleotidyl-transferase enzyme and biotinylated deoxynucleotides. The labeling reaction was stopped, and biotinylated deoxynucleotide-labeled cells were detected by incubating with streptavidin-horseradish peroxidase. Sections were stained with nuclear fast red. A total of at least 1,000 epithelial cells were counted from 10 villus-crypt sections by a blinded observer. The number of positive-staining apoptotic cells was expressed as the percentage of the total number of villus or crypt cells.
Blood samples were drawn into ice-cold tubes containing, in final concentrations, EDTA (3.9 mmol/l) and the protease inhibitor valine-pyrrolidide (0.01 mmol/l; a generous gift from R. D. Carr and L. B. Christiansen, Novo Nordisk, Bagsværd, Denmark), gently mixed and immediately centrifuged at 2,000 g at 4°C for 5 min to obtain plasma, which was stored at −20°C until analysis. Plasma GLP-2 concentrations were quantified as described previously (13, 28). Plasma samples were extracted with 75% ethanol (final concentration) and centrifuged at 3,000 g, 4°C for 30 min. The supernatant was decanted, lyophilized, and reconstituted to the original plasma volume in assay buffer: 80 mmol/l sodium phosphate buffer, pH 7.5, containing in addition 0.01 mmol/l valine-pyrrolidide, 0.1% wt/vol human serum albumin (Behring, Marburg, Germany), 10 mmol/l EDTA, and 0.6 mmol/l Thimerosal (Sigma Chemical). Approximately 300-μl extracted samples and human GLP-2 (1–33) standards were incubated with 100 μl of rabbit GLP-2 antiserum (final dilution 1:25,000) raised against an NH2 terminal fragment of human GLP-2; this antiserum specifically recognizes the NH2 terminal region of both the human and porcine GLP-2. The experimental detection limit of this assay is <5 pmol/l, and the intra-assay coefficient of variation is 5% at a concentration of 40 pmol/l.
Protein synthesis was calculated as described previously (21) as the fractional synthesis rate (FSR, %/day) Equation 1where IEbound and IEfree are the isotopic enrichments (mol% excess) of [1-13C]phenylalanine of the perchloric acid-insoluble (protein-bound) and perchloric acid-soluble (tissue-free) phenylalanine pool; t is time of labeling in minutes, and 1,440 is the number of minutes per day.
The absolute protein accretion rates (AARs), DNA accretion rates, and the absolute and fractional protein synthesis (ASR, FSR) and degradation (ADR, FDR) rates were calculated as described previously (7). The AAR of the two small intestinal segments was calculated from the difference in tissue protein mass measured at the end of the 6-day treatment period (TPm) and that estimated at day 0 (TPp) Equation 2The small intestinal protein mass of eight initial control pigs at 106 days of gestation (i.e., day 0 of treatment period) was measured to obtain the average intestinal protein mass per kilogram body weight. The intestinal protein mass of all pigs at day 0 (TPp) was estimated by multiplying their respective body weights (measured on day 0) by the average tissue protein mass (TPavg) per unit of body weight measured directly in the eight initial control pigs that were killed onday 0 of the treatment period (data not shown). The absolute rates of DNA accretion were calculated similarly. The ASR was calculated as the FSR measured at the end of the 6-day treatment period times the TPavg Equation 3The TPavg in pigs during the 6-day treatment period was calculated from the TPp and TPm Equation 4The ADR was calculated as the difference between the ASR and AAR Equation 5The FDR was calculated as the ADR divided by the average TPavg Equation 6
Data were subjected to one-way ANOVA to detect differences between any of the three treatments. When a treatment showed statistically significant results, differences between treatment groups were determined using the Fisher's protected multiple comparison test. Values of intestinal protein and DNA accretion rates were tested for difference from zero using a one-sample t-test. Statistical significance was assigned at P < 0.05.
Plasma GLP-2 concentrations and small intestinal morphometry.
At the end of the experiment, the basal plasma GLP-2 (pM; means ± SD) concentrations in the GLP-2 (67 ± 45) and ENT (56 ± 21) piglets were higher (P < 0.05) than in the TPN piglets (28 ± 19). In the GLP-2 group, the peak GLP-2 levels at the end of the 2-h infusion protocol were 390 ± 180 pM, and the concentrations declined to 206 ± 81 pM by 2 h postinfusion (values measured on day 4). The initial and final body weights among the three groups were not significantly different, and so they were pooled to give the averages (± SD) of 1,330 ± 181 and 1,487 ± 238 g, respectively. The rates of weight gain were not different among the three groups, and the mean value for the three groups was 26 ± 5 g · kg−1 · day−1. The values for intestinal weight, circumference, surface area, villus height, and crypt depth in the GLP-2 and ENT pigs were greater (P< 0.05) than in the TPN pigs, except for ileal circumference and villus height (Table 1). There were no differences in muscularis thickness in either the jejunum or ileum among the three groups. The mean (± SD) jejunal muscularis thicknesses in the TPN, GLP-2, and ENT groups were 160 ± 35, 163 ± 28, and 142 ± 27 μm, respectively. Intestinal length was only marginally (∼15%) increased in GLP-2 and ENT pigs and not significantly different from that in the TPN group. Intestinal dimensions were similar in GLP-2 and ENT pigs, except that jejunal circumference and surface area and ileal weight were significantly lower in the GLP-2 group than in the ENT group.
Intestinal protein and DNA accretion.
It is important to note that, during the 6-day treatment period, there was a small, yet significant, net accretion of DNA, but not protein, in both intestinal segments of TPN-fed pigs (Table2). However, in both jejunum and ileum, protein and DNA contents and accretion rates in GLP-2-treated and enterally fed pigs were significantly (P < 0.05) greater than in those given only TPN. Rates of jejunal and ileal protein accretion and of jejunal DNA accretion were lower in GLP-2-treated than in enterally fed pigs.
Stomach, pancreas, and large intestine.
There were no significant differences in the weights of either the stomach or large intestine among the three treatment groups. The mean (± SD) stomach weights in the TPN, GLP-2, and ENT groups were 4.86 ± 0.25, 5.02 ± 0.17, and 5.01 ± 0.23 g/kg body wt, respectively. The mean large intestine weights in the TPN, GLP-2, and ENT groups were 6.58 ± 0.83, 6.45 ± 0.64, and 6.77 ± 0.83 g/kg body wt, respectively. The pancreas weight in ENT pigs was higher (P < 0.05) than in either TPN- or GLP-2-treated pigs. The mean pancreas weights in the TPN, GLP-2, and ENT groups were 1.32 ± 0.06, 1.36 ± 0.06, and 1.79 ± 0.04 g/kg body wt, respectively.
Plasma and tissue [13C]phenylalanine enrichments.
The flooding dose technique is designed to rapidly equilibrate the plasma and tissue phenylalanine pools and minimize differences in isotopic enrichment of all possible precursor pools for protein synthesis. The plasma [13C]phenylalanine enrichments at 5 min after the tracer bolus in all three groups were not significantly different from that of the tracer solution (40 mol% excess), indicating rapid equilibration of the plasma pool. However, the plasma [13C]phenylalanine enrichments in all three groups declined between 5 and 30 min after administration of the tracer bolus; the greatest decrease occurred in TPN-fed (−17%) compared with enterally fed (−13%) and GLP-2-treated (−10%) pigs (Table3). Despite the decline in plasma [13C]phenylalanine enrichments in all three groups 30 min after administration of the tracer, the ratios of tissue-to-plasma [13C]phenylalanine enrichment were relatively high, with an overall mean of 0.860 and a range from 0.720 to 0.953. These results indicate that the bolus tracer infusion effectively minimized the differences between the plasma and tissue-free [13C]phenylalanine enrichment. However, because of the dilution of the mucosal tissue-free [13C]phenylalanine enrichment by unlabeled phenylalanine in the sow's milk, in general, the tissue-to-plasma ratios were lower in enterally fed than TPN-fed pigs. This was evident in the finding that the [13C]phenylalanine enrichment in the ileum in enterally fed pigs was lower (P < 0.05) than in TPN-fed pigs. Interestingly, the [13C]phenylalanine enrichment in the jejunum of GLP-2-treated pigs was greater (P < 0.05) than in TPN-fed pigs.
Intestinal rates of protein synthesis and degradation.
The FSRs in both intestinal segments were significantly (P < 0.05) greater in enterally fed than in TPN-fed pigs (Fig. 1). However, the FSRs in GLP-2-treated and TPN-fed pigs were not significantly different (P > 0.05). In contrast, the fractional protein degradation rate (FDR) in the jejunum, but not the ileum, was significantly (P < 0.05) lower in GLP-2 treated than in TPN-fed pigs. The FDRs in enterally fed pigs were not significantly different from those in TPN-fed pigs.
Intestinal rates of cell proliferation and apoptosis.
The rates of cell proliferation in the jejunum and ileum were significantly higher (P < 0.05) in ENT pigs than in either TPN or GLP-2 pigs (Fig. 2). However, the rates of cell proliferation in TPN and GLP-2 pigs were not different (P > 0.05) in either the jejunum or ileum. In the jejunum, the rates of apoptosis in both ENT and GLP-2 pigs were significantly (P < 0.05) lower than in TPN pigs, whereas the rates were similar in all three groups in the ileum.
The present study has shown that exogenous GLP-2 treatment largely prevents the reduction in intestinal growth associated with TPN in neonates (Fig. 3). For some, although not all, measures of intestinal growth the values were normalized to those found in ENT pigs. Although our results are consistent with those in a previous report in TPN-fed weanling rats (8), this is the first report demonstrating an intestinal trophic effect of GLP-2 in neonates. Moreover, the fact that these results were observed in neonates delivered prematurely by caesarean section indicates that GLP-2 has a trophic effect on gut growth and metabolism even at an early stage of development, when many structural and functional properties of the small intestine are relatively immature (21, 22).
A novel finding from this study is that the trophic actions of GLP-2 on the neonatal intestine appear to be mediated largely by a suppression of proteolysis and apoptosis rather than a stimulation of protein synthesis and cell proliferation. The finding that GLP-2 markedly enhanced intestinal protein accretion by suppressing proteolysis rather than protein synthesis was unexpected in light of our previous evidence that protein synthesis is substantially reduced in parenterally versus enterally fed neonatal pigs (11, 25). The suppression of protein degradation in the jejunum by GLP-2 was evident not only by the decrease in the calculated fractional rate, but also by the fact that the tissue free [13C]phenylalanine enrichment was significantly higher in GLP-2-treated than TPN-fed pigs. It is likely that GLP-2 reduced the rate of intracellular proteolysis and hence the release of unlabeled phenylalanine and thereby decreased the dilution of the [13C]phenylalanine.
There is a lack of information regarding the regulation of protein degradation. This represents a significant gap in our understanding of gastrointestinal protein metabolism and growth. However, recent studies have shown the existence of lysosomal, ubiquitin-proteosomal, and calcium-activated proteolytic systems in the intestinal mucosa (20). Moreover, results suggest strongly that fasting activates (20, 12) and enteral amino acid administration suppresses (1) proteolysis and do so via effects on all three proteolytic pathways. Our results suggest that GLP-2 may be an important gut-derived signal that suppresses intestinal proteolysis. Yet the nature of the protein anabolic response to GLP-2 was different mechanistically from that of enteral nutrition, which stimulated protein accretion mainly via increased protein synthesis. Thus, although GLP-2 is a putatively important gut-derived endocrine signal, increasing its circulating concentration alone, in the absence of a luminal nutrient stimulus, appears insufficient to fully reproduce the intestinal protein anabolic effect of enteral nutrition.
The mechanistic differences observed in protein turnover between GLP-2 and ENT pigs were also apparent in mucosal morphology and epithelial cell turnover, especially in the jejunum. Enteral nutrition and GLP-2 both increased crypt depth, villus height, and DNA content in the jejunum. However, enteral nutrition was associated with both suppression of apoptosis and stimulation of cell proliferation, whereas GLP-2 only suppressed apoptosis. In the ileum, enteral nutrition and GLP-2 both significantly increased DNA content and crypt depth, yet villus height was only marginally affected. Furthermore, cell proliferation was increased by enteral nutrition but not by GLP-2, and neither affected apoptosis in the ileum. Our finding that GLP-2 suppressed apoptosis is consistent with a previous report in mice (27), although this study also showed that GLP-2 stimulates cell proliferation. In light of both the protein metabolic and cell turnover responses, it appears that GLP-2 acts by suppressing the catabolic effects of TPN, whereas enteral nutrition not only suppresses catabolism but also provides a stimulus of cell proliferation and protein synthesis. The fact that GLP-2 did not stimulate cell proliferation in the present study is contrary to a previous report in chow-fed adult mice (27) but could be explained by either the relatively low crypt cell proliferation rate of the immature intestine, the lack of enteral nutrition, or simply differences between species. It is conceivable that increased circulating GLP-2 mediates the suppression of apoptosis associated with enteral nutrition but that the stimulatory actions of enteral nutrition on mucosal growth appear to be mediated locally by luminal nutrients and/or some other indirect signal.
A further question regarding the potential mechanism of GLP-2 action in this study is raised by the finding that, to a large degree, mucosal growth was stimulated throughout the small intestine. The only report describing the distribution of the GLP-2 receptor showed that, although the mRNA is predominantly confined to the gastrointestinal tissues, the highest level of expression occurred in the proximal small intestine and decreased along the longitudinal axis toward the colon (18). This finding may seem paradoxical, since proglucagon mRNA expression and hence the secretion of the GLP-2 peptide have been shown to be localized in the distal intestine (1, 14, 4). It should be noted that we did not observe a stimulation of growth in the stomach, pancreas, or large intestine with GLP-2 treatment, although another recent study has shown that GLP-2 stimulates colonic growth (17). With the exception of protein synthesis and cell proliferation, we found the trophic effects of GLP-2 were evident in every other end point of mucosal growth that we measured in the jejunum. However, in the ileum, a few end points were unaffected and the increases in others were lower than those found in the jejunum. These results suggest either that the GLP-2 receptor is more highly expressed in the proximal versus distal small intestine or that the receptor is strictly localized to the proximal intestine and perhaps stimulates mucosal growth in the distal intestine via an indirect mechanism. The extent of GLP-2 receptor expression along the neonatal intestine is the subject of ongoing investigation in our laboratory.
A final consideration involves the physiological relevance of the present findings and whether it is conceivable that increased circulating GLP-2 concentrations mediate the intestinal trophic effects of enteral nutrition. We found that the circulating GLP-2 concentration was twofold higher in enterally fed than TPN-fed pigs, consistent with our previous study in neonatal pigs (6). In the GLP-2 pigs, the basal peptide levels were similar to those in enterally fed pigs, but the GLP-2 levels during and shortly after stopping the GLP-2 infusion were clearly in the pharmacological range. It is interesting that studies of the GLP-2 receptor transfected into fibroblasts in culture suggest that some of the signaling events only occur in a GLP-2 dose range well above that observed in vivo (31). Thus further studies are necessary to establish whether the intestinal trophic action of GLP-2 occurs at lower circulating concentrations typically found under physiological conditions.
In summary, the results indicate that GLP-2 infusion significantly stimulates small intestinal growth in parenterally fed, premature pigs. We found that the trophic actions of GLP-2 were mediated largely by suppression of proteolysis and apoptosis. However, GLP-2 treatment did not entirely reproduce the trophic effect of enteral nutrition, either quantitatively or qualitatively, which was associated with a stimulation of protein synthesis and cell proliferation and also a suppression of apoptosis. Although GLP-2 produced trophic effects throughout the small intestine, the effects were more pronounced in the jejunum than the ileum. This study is the first to demonstrate the responsiveness of the neonatal intestine to GLP-2, albeit at a pharmacological dose, and warrants further studies to establish the sensitivity to physiological circulating concentrations of GLP-2. It will also be of interest to determine whether the trophic actions of GLP-2 in the neonatal pig are paralleled by enhancements of intestinal function, such as the enhanced glucose transport reported previously (9). Provided that further studies can establish the efficacy and safety of GLP-2 in neonates, it is possible that this peptide could be used as a therapeutic adjuvant in populations of infants with compromised intestinal function, such as those suffering from short bowel syndrome, chemotherapy-induced mucosal lesions, or immaturity following premature birth.
We would like to express our sincere gratitude to Xioayan Chang, Vincent Chan, Inger Heintze, Bente Synnetsvedt, and Anny Pedersen for technical assistance. We thank Darryl Hadsell and Sharon Bonnette for helpful assistance with the TUNEL assays. Susanne Gerlac and the Department of Pharmacy at Rigshospitalet, Copenhagen, are gratefully acknowledged for their help in the preparation of the TPN solutions. Finally, we thank Jane Schoppe for assistance in preparation of the manuscript and Leslie Loddeke for editorial assistance.
This work was supported by The Danish Agricultural and Veterinary Research Council (Program 9702803), National Institute of Child Health and Human Development Grant RO1-HD-33920 (D. G. Burrin), and by federal funds from the U.S. Department of Agriculture, Agricultural Research Service under Cooperative Agreement Number 58–6250–6-001. The contents of this publication do not necessarily reflect the views or policies of the U.S. Department of Agriculture, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
Address for reprint requests and other correspondence: P. T. Sangild, Dept. of Animal Science and Health, Division of Nutrition, Royal Veterinary and Agricultural Univ., 13 Bulowsvej, DK-1870 Frederiksberg, Copenhagen, Denmark (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2000 the American Physiological Society