Dietary protein is absorbed from the intestinal lumen as its constituent amino acids as well as small peptides containing two or three residues. The protein responsible for dietary di- and tripeptide transport is peptide transporter 1 (PepT1), a member of the proton-coupled oligopeptide transporter super-family of proteins(Reference Daniel1). PepT1 is localised to the apical membrane of intestinal villi, allowing it access to the digesta passing through the lumen. Studies that have investigated PepT1 mRNA at distinct regions along the intestine have reported variable results(Reference Chen, Wong and Webb2–Reference Wang, Shi and Zhang4), probably due to different animal models. In rats and mice, the amount of PepT1 mRNA does not change across the length of the small intestine(Reference Lu and Klassen5–Reference Erickson, Gum and Lindstrom7); however, in the developing chick intestine, the highest quantities were found in the duodenum and jejunum(Reference Li, Gilbert and Zhang3). In adult human subjects and rats, higher concentrations of PepT1 mRNA in the duodenum and jejunum have also been reported(Reference Terada, Shimada and Pan8, Reference Herrera-Ruiz, Wang and Gudmundsson9). In contrast, Chen et al. (Reference Chen, Wong and Webb2) reported similar expression of PepT1 mRNA across the entire length of the small intestine of mature sheep, dairy cows, pigs and chickens.
Few studies have reported the developmental changes in PepT1. Differences in the location and concentration of PepT1 mRNA along the small intestine with increasing age have been reported in rats(Reference Shen, Smith and Brosius10) and chickens(Reference Chen, Pan and Wong11, Reference Gilbert, Li and Emmerson12). PepT1 protein increased linearly with age in chickens, but there was a decrease in PepT1 mRNA in rats between postnatal days 4 and 50. Shen et al. (Reference Shen, Smith and Brosius10) demonstrated that even though PepT1 mRNA expression was at its highest point at 3–5 d after birth in a neonatal rat model, it declined rapidly at all locations in the small intestine during the following days of suckling, suggesting that PepT1 and peptide transport are potentially important in the early life of the suckling neonate. In piglets, D'Inca et al. (Reference D'Inca, Gras-Le Guen and Che13) determined that there was a dramatic decrease in PepT1 mRNA expression within the first 48 h after birth; however, intra-uterine growth restriction delayed this adaptation, further supporting the importance of peptide transport in the newborn.
In adult human subjects, a significant proportion of amino acids are absorbed as di- and tripeptides, and dipeptides are absorbed faster than free amino acids in the jejunum due to the high capacity for uptake by PepT1(Reference Ganapathy, Brandsch, Leibach and Johnson14). A high turnover rate and expression level of PepT1 contributes to the rapid uptake of peptides(Reference Steinhardt and Adibi15). The efficiency in uptake of small peptides combined with the lower osmolarity of peptide solutions support the use of peptides rather than free amino acids in therapeutic enteral formulas(Reference Boza, Moënnoz and Vuichoud16). In neonates, proteolytic capacity is not as well-developed as in adults, due to a significantly lower production of pepsin(Reference Henderson, Hamosh and Armand17) and trypsin(Reference Austic18). Compared with adults, a higher proportion of amino acids would be present in the neonatal gut as small peptides. As such, the capacity for peptide transport may be high to support optimal nutritional status and growth.
Indeed, alanyl-glutamine is currently being studied as a potential dietary supplement, and has been reported to increase the nutritive status and intestinal barrier function in human neonates(Reference Lima, Soares and Mota19) as well as to aid in the maintenance of cellular glutathione concentrations during oxidative stress(Reference Alteheld, Evans and Gu20). Although the absorption of small peptides is important in healthy individuals, the bioavailability of peptides in times of intestinal duress may be of particular importance. In addition, the added benefits of stability, hypotonicity and palatability(Reference Silk, Hegarty and Fairclough21) support the importance and usefulness of dipeptide preparations for modern clinical treatments.
Characterising the ontogeny of PepT1 in the neonate is of tremendous importance for the development of effective feeding strategies for both healthy and sick infants. The piglet is an excellent model for the human neonates, especially preterm infants(Reference Shulman22). Neonatal pigs share similar patterns of intestinal development with human subjects, as well as similar nutritional requirements. Dipeptide transport capacity via PepT1 and the effects of development have not been previously reported in this model. In the present study, we determined PepT1 transport capabilities and relative quantities of PepT1 mRNA at specific locations along the gastrointestinal tract of piglets during suckling and post-weaning.
Experimental methods
In situ perfusion (gut loop model)
A total of sixteen Yucatan miniature piglets (n 4 per group) were obtained from the breeding herd at the Memorial University of Newfoundland (average birth weight 1·01 (sem 0·03) kg; average growth rate from birth to 1 month of age: 45 (sem 3) g/kg body weight per/d). All procedures were approved by the Institutional Animal Care Committee and were in accordance with the guidelines of the Canadian Council on Animal Care. Piglets were studied at 1, 2, 3 or 6 weeks of age. Piglets in this herd are weaned at 4 weeks of age to a standard pelleted grower pig diet based on wheat, barley, oats and rapeseed meal (15·6 % crude protein, 3·75 % crude fat, 12·1 MJ/kg digestible energy) (Eastern Farmers Co-op); pre-weaning piglets were exclusively suckled by sows that were fed a typical lactation diet with 14 % crude protein and 3 % crude fat. To avoid fasting (which may alter PepT1 expression or function), the in situ study was initiated within 2 h of separating the piglets from the sows at 1, 2 or 3 weeks of age and within 2 h of last feed for 6 week-old piglets. Anaesthesia was administered with an intramuscular injection of 22 mg/kg ketamine hydrochloride (Bimeda-MTC) plus 0·5 mg/kg acepromazine (Vétoquinol Canada, Inc.). General anaesthesia was maintained with 1–2 % isoflurane (Abbot Laboratories Limited), delivered with 1·5 litres O2/min. Body temperature was maintained via a homoeothermic blanket system that automatically adjusted heat output in response to a rectal temperature probe (Harvard Apparatus). A mid-line incision was made in the abdomen to open the peritoneum. The complete length of the small intestine was measured and four equidistant regions were marked for loop placement. The most proximal jejunal loop (0 %) was defined as 15 cm from the ligament of Treitz. Closed loops of intestine consisted of 10 cm sections of intestine, with inlet and outlet cannulas (inner diameter, 0·2 cm; outer diameter, 0·3 cm; Watson Marlow) inserted through a small perforation at both ends of the 10 cm section. In total, four loops were placed at 0, 25, 50 and 75 % of the length of the small intestine. Each isolated gut loop was perfused with a dipeptide-containing buffer that was re-circulated through the loop for the duration of the study by a multi-channel peristaltic pump (Watson Marlow). The perfusate (65 ml) contained 5 mmol/l glycylsarcosine with 37 kBq 3H-glycylsarcosine (Moravek) in PBS (144·6 mm-NaCl, 15·9 mm-Na2HPO4, 1·2 mm-NaH2PO4; pH 6·0), and was maintained at 37°C for the duration of the study. Perfusions through each closed loop continued for 120 min. A 1 ml sample of perfusate from each closed loop was removed at every 10 min to determine isotope disappearance. During the procedure, the exposed intestines were kept moist with warmed saline and covered with gauze and plastic wrap. Heart rate and blood oxygenation were monitored via pulse-oximetry (Surgivet) throughout the experiment. At the end of the perfusion period, the loops were excised by cautery, flushed with cold 0·9 % saline, placed on ice, cut longitudinally and scraped with a microscope slide to remove the mucosa. Samples were also removed from the apex of the spiral colon and flushed, and all tissues were immediately flash-frozen in liquid N2 and stored at − 80°C until further analyses.
Disappearance of 3H-glycylsarcosine
Sampled perfusate (100 μl) was added to 4 ml Scintiverse (Fisher Scientific) for liquid scintillation counting. The specific radioactivity was calculated as the mean dpm/μmol of glycylsarcosine present in the perfusate. Specific radioactivity was calculated at baseline (prior to perfusion) and in each of the perfusate samples taken over the course of the 2 h study. The specific radioactivity was then used to determine the total quantity of glycylsarcosine. Rate of disappearance of glycylsarcosine was determined via AUC using GraphPad Prism 4.0 (Graph pad Software Inc.). Perfusates were also derivatised with phenylisothiocyanate and analysed using HPLC(Reference Bidlingmeyer, Cohen and Tarvin23) complete fraction collection to determine whether the isotope was associated with glycylsarcosine, glycine or some other metabolite.
Disaccharidase activity
Mucosal tissue scraped from each of the gut loop segments was analysed for sucrase and maltase activity, as per the method of Dahlqvist(Reference Dahlqvist24).
Real-time RT-PCR
PepT1 mRNA was measured in mucosa sampled from each of the perfused intestinal loops. Samples of colon were also analysed. RNA was extracted using the Qiagen RNEasy Mini kit (Qiagen), according to the manufacturer's protocol. Relative concentration and purity were determined by measuring the absorbance at 260 and 280 nm using a NanoDrop 1000 (NanoDrop Technologies) spectrophotometer. RNA integrity was visualised via agarose gel visualisation of the 28S:18S ratio. Complementary DNA was created according to the protocol outlined in the QuantiTect (Qiagen) RT manual. A measure of 1 μg of total RNA was used in the reverse transcription reaction. Taqman probes were utilised in the PCR. The sequences of the primers and probes were as follows: PepT1 forward primer 5′-d CTGGAGTTCTCCTATTCTCA-3′, reverse primer 5′-d AACAGCCACGGTCAACAG-3′, probe sequence for PepT1 5′-d BHQ-2-TCCTTCCAACATGAAGTCGGTGC-Pulsar 650-3′. β-Actin was used as an internal control with the following sequences: forward primer 5′-d CCCAGCACGATGAAGA-3′, reverse primer 5′-d CGATCCACACGGAGTC-3′, probe 5′-d FAM-TCAAGATCATCGCGCCTCCAGA-BHQ-1-3′. The accession numbers for the template sequences were AY180903.1 for PepT1(Reference Klang, Burnworth and Pan25) and AY55069 for β-actin. The Lightcycler (Roche) was set to the following conditions: 15 min at 95°C, forty cycles of 1 min at 95°C and 1 min at 55°C and a 40°C incubation for 1 min. Reaction efficiency for PepT1 was 0·94 (sd 0·03) and β-actin reaction efficiency was 0·96 (sd 0·06). Each sample was run in triplicate and analysed using the Livak method (2− ddC t)(Reference Livak and Schmittgen26).
Statistical analyses
Data were analysed by one-way ANOVA with Bonferroni's protected means separation test for PepT1 mRNA. For overall uptake of glycylsarcosine, a mixed model two-way ANOVA was used, with intestinal location as the repeated measure variable; age was the second variable and age × location interaction was also reported. Glycylsarcosine uptake data over time were tested for linearity using regression analysis. Correlations between disaccharidase activity and glycylsarcosine uptake were determined by linear regression analysis. Sample size was n 4 piglets per age group and differences were noted as significant if P< 0·05 (GraphPad Prism 4.0).
Results
Throughout the perfusion studies, all piglets remained stable, well-oxygenated and maintained a core body temperature above 37·5°C.
Dipeptide disappearance
At the end of each perfusion study (2 h), a sample of perfusate from all intestinal segments was analysed to confirm the integrity of the dipeptide. The proportion of 3H associated with glycylsarcosine accounted for greater than 90 % of the radioactivity in the perfusates, while glycine contained less than 5 % of the label. Thus, the dipeptide did not degrade intraluminally and was available for transport. Glycylsarcosine transport occurred linearly in all intestinal loops in all the animals studied, except in the most distal loop (75 %) in piglets aged 1 and 3 weeks, and in the 0 % loop in piglets aged 6 weeks (Fig. 1). In order to assess how dipeptide uptake might change along the longitudinal axis of the small intestine as the animal ages, we compared glycylsarcosine uptake between age groups at the four sites of the small intestine (Fig. 2). Overall, there was a significant effect of age (P< 0·0001), intestinal location (P= 0·001) and age × location interaction (P= 0·0002). When compared by intestinal location, dipeptide uptake was remarkably consistent at the most proximal site (0 %) across all age groups. At the mid-jejunum site (25 %), glycylsarcosine uptake in 1-week-old pigs was twice that in pigs aged 2 and 3 weeks (P< 0·01) with 6-week-old pigs being intermediate. During the suckling phase (weeks 1, 2 and 3), sucrase activity negatively correlated with glycylsarcosine uptake only in the mid-jejunum (25 %) (r 2 0·59, P= 0·006); at the same site, maltase activity also tended to negatively correlate with glycylsarcosine uptake (r 2 0·29, P= 0·09) (data not shown). In the distal jejunum (50 %), greater uptake occurred in the post-weaning animals (6 weeks) compared with the 2-week-old animals (P< 0·001), with uptake in piglets at 1 or 3 weeks of age being intermediate. The uptake of glycylsarcosine in the ileum (75 %) for 6-week-old animals was about twice that for piglets aged 2 and 3 weeks (P< 0·001), and 30 % greater than piglets at 1 week of age (P< 0·05). Within age groups, 1-week-old animals had the greatest uptake in mid-jejunum and ileum (P< 0·01). In 6-week-old animals, there was a notable gradient of glycylsarcosine uptake, with the lowest uptake in the proximal jejunum and up to a 3-fold higher uptake in the ileum (Fig. 2).
Total glycylsarcosine disappeared from closed loops of small intestine during the 2 h perfusion study. Position of each intestinal loop is indicated by its relative position from the ligament of Treitz: (A) 0, (B) 25, (C) 50 and (D) 75 %. Values are means with their standard errors (n 4 for each group). 
, Week 1; 
, week 2; 
, week 3; 
, week 6.
Rate of glycylsarcosine uptake from closed loops of small intestine. Uptake was determined by measuring the disappearance of 3H-glycylsarcosine per g of wet mucosa. Values are means with their standard errors (n 4 for each bar) and were analysed by two-way ANOVA. Significant effects of age (P< 0·0001), intestinal location (P< 0·005) and age × location interaction (P< 0·001) were observed. Lines represent significant differences between intestinal sections within an age group (* P< 0·05, ** P< 0·01). a,b,cMean values with unlike superscript letters were significantly different (P< 0·05) within each specific intestinal site, compared between age groups. Position of each intestinal loop is indicated by its relative position from the ligament of Treitz (0 (
), 25 (
), 50 (
) and 75 % (
) of the total small intestinal length).
Quantification of PepT1 mRNA in the intestine
Samples from all the small-intestinal segments were analysed for PepT1 mRNA, but the results were highly variable and no significant differences in PepT1 mRNA were detected between the age groups or across the length of the intestine (data not shown). PepT1 mRNA was detected in the colon of all animals (Fig. 3); however, the quantity was approximately 10 % of that measured in the proximal jejunum of piglets at 1 week of age and approximately 2 % of that found in the proximal jejunum of post-weaning animals (P< 0·05). Although present in similar abundance at ages of 1, 2 and 3 weeks, colonic PepT1 mRNA in post-weaned pigs was up to 90 % lower than that in suckling pigs (P= 0·0017) (Fig. 3).
Real-time RT-PCR analysis of PepT1 mRNA in the colon. Values are means with their standard errors (n 4 for each bar) and were analysed by one-way ANOVA. a,bMean values with unlike superscript letters were significantly different (P< 0·05).
Discussion
Although the porcine peptide transporter has been cloned and characterised(Reference Klang, Burnworth and Pan25), there has been little investigation into the ontogeny of porcine PepT1. The objective of the present study was to evaluate the ontogenic pattern in peptide transport capability and PepT1 mRNA quantity in the small intestine of the Yucatan miniature pig as it develops from suckling to a post-weaned state. Clinically, such information is necessary to appropriately design feeding regimens for infants. This objective was accomplished through the use of a hydrolysis-resistant dipeptide tracer, 3H-glycylsarcosine, and real-time RT-PCR for mRNA analysis. From these techniques, it was determined that there was little variation in the uptake of glycylsarcosine at any location in the small intestine during the suckling period; however, in the post-weaning period, there was a distinct gradient in rate of uptake, which was greatest in the distal small intestine. Small-intestinal PepT1 mRNA was extremely variable at all locations measured in the small intestine at each age, whereas colonic PepT1 abundance declined dramatically after the animals were weaned.
In the present study, small-intestinal loops were perfused with a dipeptide-containing solution and uptake was calculated from the disappearance of 3H-glycylsarcosine. This in situ model is ideal to isolate specific regions of the intestine for study by allowing multiple loops per animal, limiting systemic effects of the perfusate and still maintaining first-pass metabolism(Reference Nichols and Bertolo27–Reference Adegoke, McBurney and Samuels29). For these reasons, this model was selected to examine dipeptide uptake in the neonatal piglet.
Absorption of glycylsarcosine and PepT1 mRNA was detected in the loops at all four locations in the small intestine. There was a significant effect of intestinal location on the uptake of the dipeptide glycylsarcosine, particularly in post-weaning pigs (Fig. 2), whereas in the older suckling animals (weeks 2 and 3), the uptake of glycylsarcosine was relatively consistent in all intestinal locations studied. The difference in diet composition may be responsible as the protein components of sow milk are readily digested and absorbed predominantly in the proximal parts of the small intestine(Reference Mavromichalis, Parr and Gabert30, Reference Buddington, Elnif and Puchal-Gardiner31). In contrast, products of protein hydrolysis from grain-based diets, with lower digestibility consumed by weaned animals, would probably reach the ileum before absorption(Reference Low32), providing a substrate for PepT1 and perhaps a stimulus for PepT1 up-regulation in the distal intestine of the post-weaned piglet. In this respect, the distal ileum may be important for optimal nitrogen absorption in the period immediately following a diet transition. Alternatively, the transition to solid feeding can be considered a period of intestinal injury and inflammation for the milk-fed piglet(Reference Moeser, Klok and Ryan33, Reference Pié, Lallès and Blazy34), and such injury could lead to a stimulation of peptide uptake capacity(Reference Vavricka, Musch and Fujiya35). With respect to age, dipeptide uptake tended to follow a U-shaped curve, with higher uptake in 1-week-old and post-weaned piglets. To determine whether dipeptide uptake is developmentally regulated, we correlated the data from the suckling period with intestinal sucrase and maltase activities. Disaccharidase activity is an established marker of intestinal maturation, and sucrase and maltase activities are known to rise significantly in piglets at about the second week of life(Reference James, Smith and Tivey36). Because dipeptide uptake negatively correlated with sucrase activity in suckled piglets, PepT1 appears to be developmentally regulated during early neonatal life, but may respond to dietary changes post-weaning.
Although PepT1 is present in the small intestine, we have also identified its presence in the colon of the developing piglet. The present results demonstrated that PepT1 mRNA is present in colonic tissues until after weaning occurs (week 6). This could be an adaptive response that occurs post-weaning; as the distal small intestine increases its capacity for oligopeptide absorption, the colon reduces the transcription of PepT1 mRNA. Shen et al. (Reference Shen, Smith and Brosius10) reported that PepT1 presence in the colon was transient, because by day 7 no PepT1 mRNA was found in the colon of developing rats. PepT1 has been found in the colon of rats at later time points(Reference Shi, Song and Xue37), but its presence has been primarily related to a state of intestinal injury such as gut resection. PepT1 has also been detected in the colon of human subjects who have undergone intestinal resection(Reference Ziegler, Fernández-Estívariz and Gu38). PepT1 is present in the colon of human subjects with inflammatory bowel disease(Reference Merlin, Si-Tahar and Sitaraman39), and it has been shown that PepT1 is capable of transporting N-formyl-methionyl-leucyl-phenylalarine (FMLP), which is a bacterial peptide. As such, PepT1 may be responsible for the exacerbation of inflammatory bowel disease. Colonic PepT1 mRNA and protein were also higher in patients with massive bowel resections, when compared with control subjects(Reference Ziegler, Fernández-Estívariz and Gu38). A study in rats demonstrated an increase in PepT1 protein in the colon of animals with intestinal resection that were fed chow(Reference Shi, Song and Xue37), whereas rats fed an elemental diet did not express any PepT1 mRNA in the colonic tissues(Reference Lardy, Thomas and Noordine40). Findings such as these underscore the importance of peptide presence in the lumen of the colon for the induction of PepT1. Although colonic peptide transport is extremely low when compared with the whole small intestine, in the event of intestinal injury or malnutrition, capacity for oligopeptide uptake may be increased through the up-regulation of PepT1, thereby recovering nutritionally valuable substrates. Whether this also leads to a greater uptake of pathogenic bacterial peptides remains to be determined.
To understand the potential impact of peptide transport, it is important to consider the variable affinity of PepT1 substrates. Alanylalanine, for example, has a high affinity for PepT1 (K i of 0·08 (sem 0·01) mm), whereas glycylsarcosine is classified as a medium-affinity substrate with a K i of 1·1 (sem 0·1) mm, determined in Caco-2 cells(Reference Brandsch, Thunecke and Küllertz41, Reference Brandsch, Knütter and Thunecke42). Previous work on glycylsarcosine transport in porcine jejunum, using an Ussing chamber system, demonstrated dipeptide uptake of approximately 600 nmol/cm2 per 2 h (modified from Winckler et al. (Reference Winckler, Breves and Boll43)). In the present study, using the more physiological ligated loop model, we found that the overall average glycylsarcosine uptake was 424 (sem 146) nmol/cm2 per 2 h. Thus, the data from the present in situ model are comparable with the previous in vitro findings using the same non-nutritional dipeptide and probably underestimate the transport rates of dietary dipeptides with higher PepT1 transporter affinity.
There are two possible mechanisms for the intestinal transport of oligopeptides: PepT1-mediated transcellular transport(Reference Daniel1) and paracellular transport(Reference McCollum and Webb44). The role of these pathways in intestinal oligopeptide transport is not yet fully understood, and there is increasing evidence that the paracellular pathway may be of some importance in the absorption of small peptides and peptidomimetic drugs(Reference Lafforgue, Arellano and Vachoux45, Reference Menon and Barr46). It has been well-documented that the small intestine of suckling animals demonstrates greater paracellular transport at younger ages(Reference Udall, Pang and Fritze47, Reference Weaver, Laker and Nelson48). For this reason, paracellular transport could potentially mask low PepT1 activity in the young piglets that we studied (weeks 1, 2 and 3); however, the impact of paracellular transport would decrease as the animal ages. We observed the greatest disappearance of glycylsarcosine in the ileum of the post-weaning animals. Overall, if paracellular transport had significant impact on the transport of glycylsarcosine in our piglets, it would have been most prominent at the younger ages (weeks 1, 2 and 3) rather than at the post-weaning state. Nevertheless, future studies characterising the contribution of paracellular transport to dietary dipeptide uptake are warranted.
In summary, we have examined the ontogenic changes of the peptide transporter PepT1 in a piglet model of the developing intestine. These results have demonstrated that the capacity for dipeptide transport is present in both the proximal and distal small intestine, with the colon potentially being able to transport peptides during the suckling state. By post-weaning, the ileum is the site of highest dipeptide uptake. The importance of dipeptides to the nutritional status of the developing infant is still unknown; however, when therapeutic formulas are necessary, it is likely that the provision of all or part of the nitrogen as small peptides conveys advantages over preparations of free amino acids, and this may be of particular importance when compromised intestinal function is present.
Acknowledgements
The present study was supported in part by a grant from the Natural Sciences and Engineering Research Council of Canada (grant number 261557–2006 RGPIN). The authors are solely responsible for the work described in the present article. J. A. B. and R. F. B. were responsible for project development. M. G. N. conducted the experiments, the analytical work and generated the results. All authors were involved in the statistical analyses, data interpretation and writing of the manuscript. The authors declare that they have no conflict of interest.
