| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL NUTRITION |
Department of Animal and Poultry Sciences, Virginia Polytechnic Institute and State University, Blacksburg 24061-0306
 |
Abstract |
|---|
Key Words: amino acid absorption • dietary regulation • intestine • PepT1 • peptide • transporter
|
INTRODUCTION |
|---|
eloquent way of stating that the contribution of peptides to total AA absorption and utilization had been ignored for far too long. Long before the cloning and characterization of the intestinal peptide transporter, PepT1 (Fei et al., 1994), a carrier-mediated mechanism was shown to exist for the uptake of small peptides across the brushborder membrane of the enterocyte. Although the first evidence for peptide transport was provided in the 1950s (Newey and Smyth, 1959, 1960), acceptance of AA absorption across the gut wall in the form of peptides was slow to emerge, even as recently as the late 1980s (Webb, 1990). In addition to transport through PepT1, peptides may also be absorbed through alternate routes including paracellular movement and by cell-penetrating peptides (CPP) that are capable of moving cargo across the plasma membrane (Figure 1). The purpose of this review is to describe, both from a historical and current perspective, intestinal peptide transport, which has been characterized primarily in laboratory species such as the rat and relate these findings to livestock and poultry. We hope that, through this review, we can impart to the reader the nutritional importance of peptide transport in agriculturally important animals and with this knowledge, the potential to advance the field of nutrition and diet formulation. Unless it is otherwise stated, the use of the term peptide will imply collectively di- and tripeptides. Although excellent reviews on peptide transport and the intestinal peptide transporter, PepT1, have been previously published (Webb, 1990; Webb et al., 1993; Leibach and Ganapathy, 1996; Daniel, 2004; Daniel and Kottra, 2004), this is the first to combine information from classical nutrition studies with recent molecular and gene regulatory studies to define a novel approach to how we view protein nutrition and gut physiology across a wide range of species.
|
|
FATE OF LUMINAL PROTEIN DIGESTION PRODUCTS |
|---|
; Kanai and Hediger, 1992; Segawa et al., 1997; Chairoungdua et al., 1999; Rajan et al., 2000; Broer et al., 2004). |
PEPTIDE TRANSPORT VS. FREE AA TRANSPORT |
|---|
). In terms of energetic efficiency, 2 or 3 AA can be transported into the cell by PepT1 for the same expenditure of energy required to transport a single free AA (Daniel, 2004). In addition, individuals suffering from deficiencies in free AA transport were still able to assimilate essential AA, pointing to the possibility that PepT1 transports enough dietary AA to compensate for a deficiency in free AA transport (Adibi, 1997).
Transport of AA in the form of peptides was demonstrated to be a faster route of uptake per unit of time than their constituent AA in the free form (Adibi and Phillips, 1968; Craft et al., 1968; Adibi, 1971, 1986; Cheng et al., 1971; Burston et al., 1972). In rats, AA in soy protein hydrolysate (Kodera et al., 2006) or egg white protein hydrolysate (Hara et al., 1984) was absorbed faster into the portal blood after duodenal infusion than those representing an AA mixture or intact protein with the same respective AA composition. Similar results were observed using milk protein hydrolysate vs. a free AA mixture in pigs (Rerat et al., 1988) and humans (Silk et al., 1980). Intestinal perfusion of 0, 50, 100, or 200 g/L solutions of intact soy protein or 0, 100, 200, 300, or 400 g/L solutions of hydrolyzed soy protein caused a load-dependent slowing of intestinal transit in dogs (Zhao et al., 1997). Interestingly, the intact protein slowed transit more effectively than the hydrolysate, supporting the notion that digestion is a rate-limiting step in nutrient assimilation. Furthermore, a greater amount of AA was absorbed in the proximal small intestine when protein was infused as a hydrolysate instead of in the intact form, suggesting that AA as peptides were more readily available for absorption.
|
PEPTIDES AND PORTAL-DRAINED VISCERA AA FLUX |
|---|
; Stoll and Burrin, 2006). For example, 90% of dietary glutamate is sequestered by the portal-drained viscera (PDV) in swine. Intestinal mucosal cells use AA for energy, protein synthesis, nucleosides, polyamides, and maintenance of the intestinal immune system (i.e, glutathione and mucin synthesis; Reeds et al., 2000).
Thus, although it is certain that AA derived from dietary protein are transported into the enterocyte in the form of di- and tripeptides, there is still much to be learned about the quantitative significance and origin of luminally derived peptides and the relative postabsorptive use and portal appearance of free AA vs. peptides. Determination of amounts of free and peptide-bound AA in plasma involves protein removal followed by AA analysis. A variety of plasma deproteinization methods are used, including chemical precipitation with sulfosalicylic acid (Koeln et al., 1993) or methanol (Tagari et al., 2004, 2008), physical removal using ultrafiltration (Seal and Parker, 1996), or a combination of chemical precipitation followed by the ultrafiltration step (Han et al., 2001b; Tagari et al., 2004, 2008). The traditional method of AA analysis is separation by ion-exchange chromatography (Koeln et al., 1993), in which samples are applied to a column and eluted by buffers with increasing pH or ionic strength, or both (White et al., 1986). Postcolumn derivatization with ninhydrin generates a colorimetric product that can be measured. The Pico-Tag Waters method (Seal and Parker, 1996; Delgado-Elorduy et al., 2002; Tagari et al., 2004, 2008) is a newer approach in which phenylisothiocyanate is used for precolumn derivatization of AA. After derivatization, reversed-phase gradient elution high-performance liquid chromatography separates the phenylisothiocyanate derivatives that are quantified based on UV absorbance. In the simplest terms, peptide-bound AA are calculated as the difference between total AA observed after hydrolysis of protein-free preparations and free AA present before hydrolysis. Based on the different combinations of methods for AA analysis, there are likely to be variations in quantities of plasma peptides reported in the literature.
Little or no detection of absorbed intact peptides was reported in dogs (Levenson et al., 1959) and rats (Wiggans and Johnston, 1959). Others have reported that up to 85% of AA entering the portal blood was in the form of small peptides in rats (Gardner, 1975), Holstein calves (Koeln et al., 1993), lactating Holstein cows (Tagari et al., 2004, 2008), Friesian steers (Seal and Parker, 1996), sheep (Remond et al., 2000), and yak cows (Han et al., 2001a,b). In some of these studies, total appearance of free and peptide AA in the portal vein was greater than the intake of dietary protein, suggesting contributions from degradation products resulting from tissue protein turnover in the gastrointestinal tract, spleen, and pancreas (Han et al., 2001a). In studies in which lower contributions of peptide-bound AA to total portal AA flux were observed (32%, Remond et al., 2000 and Han et al., 2001b; up to 20%, Tagari et al., 2004, 2008), the authors attributed the conflicting reports to differences in methodology. Refinements in analytical technique, including chemical deproteinization of plasma followed by physical filtration, result in a more accurate determination of peptide AA concentrations. The more recent reports indicating a lower PDV flux of peptides may represent a more reasonable estimate of this value. Although estimates of the proportion and absolute concentration of peptide AA entering the portal circulation vary across and within species, the fact still remains that peptides are detected in the blood and their concentration is influenced by diet, suggesting their importance as a nutritional resource to the liver and extrahepatic tissues.
DiRienzo (1990) quantified the in vivo flux of both free and peptide AA across the mesenteric and non-mesenteric portions of the PDV in both calves and sheep and made 2 important findings: 1) the mesenteric flux (drainage into mesenteric vein from jejunum, ileum, cecum, colon, and pancreas) of free and peptide AA was similar and 2) the flux of peptide AA across the nonmesenteric-drained viscera (from rumen, reticulum, omasum, abomasum, duodenum, and spleen) contributed the largest proportion of total PDV AA flux in both calves and sheep. This study confirmed the contribution of peptides to the portal appearance of AA and also demonstrated significant uptake of peptides from the rumen and omasum. This was later confirmed by Matthews and Webb (1995), McCollum and Webb (1998), and by Tagari et al. (2004, 2008). This observation has important nutritional implications in ruminants in which dietary supply of AA nitrogen and microbial metabolism of dietary protein influences the nutrients that are made available to the animal.
Matthews and Webb (1995) found uptake of 2 dipeptides and free methionine across sheep ruminal and omasal epithelia to occur by a non carrier-mediated route of uptake as demonstrated by a linear response in uptake with greater absorption in omasal tissues as compared with ruminal tissues. Because serosal appearance of peptides was a nonsaturatable process, the data suggested that absorption occurred by a non carrier-mediated process. However, in a later report characterizing a peptide transporter in omasal epthelia (Matthews et al., 1996), the authors mentioned that attempts to identify saturatable uptake of several dipeptides may have been confounded by the use of substrate concentrations that were greater than mediated transport capacity.
|
FATE OF PLASMA PEPTIDES |
|---|
; Odoom et al., 1990). Peptides are likely transported into cells and hydrolyzed into free AA, which are then used for protein synthesis (Krzysik and Adibi, 1979) with the extent of peptide utilization in cells influenced by dietary status (Webb, 1990; Webb et al., 1993). In calves fed a diet of hay, corn, and soybean meal, circulating peptide AA were present at a 3-fold greater concentration than free AA with a preferential removal of peptide AA from plasma by tissues of the hind limbs (McCormick and Webb, 1982). When fed purified diets containing either urea or soybean meal as the sole supplier of nitrogen, peptide removal varied depending on the protein source with the greatest removal by the hind limbs in calves fed the soybean meal and negligible removal by the hind limbs in calves fed the urea (Danilson et al., 1987). These results suggest that diet influences concentrations of circulating peptides and availability to extrahepatic tissues. That there is tissue selectivity for peptide removal suggests that there may be tissue-specific ability to utilize circulating plasma peptides.
The hydrophobicity of peptides, which may also be related to dietary protein source, may affect their absorption and subsequent utilization by tissues. Hydrophobic peptides and peptides resistant to mucosal hydrolysis were utilized or absorbed at a faster rate than peptides that are hydrophilic and susceptible to hydrolysis in C2C12 myogenic and MAC-T mammary cells (Pan et al., 1996). Burston and Matthews (1990), however, did not find a correlation between hydrophobicity and transport rate of peptides in hamster jejunum. Pan and Webb (1998) demonstrated that molecular structure influenced availability of methionine-containing peptides as a methionine source in ovine skeletal muscle cells. There were distinct differences in utilization of methionine-containing dipeptides in 3 cell types, suggesting that there were cell-specific differences in transport and hydrolytic events.
|
PEPTIDES AS A SOURCE OF AA FOR THE MAMMARY GLAND |
|---|
; Metcalf et al., 1996; Bequette et al., 1999). Wang et al. (1996) evaluated peptide-bound methionine as a source for protein synthesis in mammary tissue explants from lactating mice and found no difference in the ability of 5 of 7 pairs of dipeptides to stimulate [3H]-leucine incorporation into proteins, but the di- and tripeptides promoted a greater incorporation than did free methionine. Although PepT1 mRNA was not detected in mammary cells (Chen et al., 1999), aminopeptidase N mRNA was detected in caprine mammary cells and was found to be regulated by the postabsorptive form of circulating AA (Mabjeesh et al., 2005), thus suggesting a mechanism for utilization of AA circulating in the form of small peptides.
In goats at early and late lactation, methionine, histidine, threonine, proline, and phenylalanine in milk protein output were not accounted for in estimates of uptake of free AA (Mabjeesh et al., 2002). An in vivo isotope kinetic technique was used to determine the sources of AA for milk protein synthesis in which a long-term (greater than 20 h) i.v. infusion of a labeled AA (e.g., methionine) was performed until an isotopic steady state was reached in the plasma-free AA pool and in the secreted milk casein. A dilution of isotope was observed in milk casein indicating an alternate source of AA being used for protein synthesis. The contribution of AA from intracellular protein turnover was removed as a possibility due to the high fractional rate of protein synthesis in mammary tissue and observation of complete turnover of proteins at the end of a 30-h infusion of isotope. Mabjeesh et al. (2005) concluded that circulating unlabeled peptides contributed to AA use for milk protein synthesis by mammary tissue with an estimated 7 to 18% of methionine in casein being derived from peptides. When the estimate of methionine derived from peptides was summed with an estimate of the contribution of free AA, total uptake of methionine by the udder was in close balance with the estimate for milk output.
Tagari et al. (2004, 2008) demonstrated that peptide-bound AA represented an important portion of total AA flux across the PDV and also represented a key contribution to the mammary gland for milk protein synthesis. They reported, in the 2004 paper, that the PDV flux of total free essential AA was greater in cows fed steam-flaked corn versus steam-rolled corn (571.2 vs. 366.4 g/12 h, respectively). The PDV flux of essential peptide-bound AA was 69.3 ± 10.8 and 51.5 ± 13.2 g/12 h for cows fed steam-flaked or steam-rolled corn, respectively. Mammary uptake of essential peptide-bound AA was greater for cows fed steam-flaked corn than steam-rolled corn (25.0 vs. 15.1 g/12 h, respectively). In the 2008 study in which processing effects of sorghum rather than corn were evaluated, PDV flux of free AA was greater in cows fed dry-rolled as compared with steam-flaked sorghum, whereas PDV flux of peptide AA was greater in cows fed steam-flaked sorghum. In these cows, milk production output did not differ between the 2 dietary groups, and the authors attributed this to a possible compensatory effect of peptide-bound AA in the cows containing lower plasma concentrations of free AA in which peptide-bound AA could be used in response to the free AA pool shortage. Tagari and colleagues suggested that peptide-bound AA may play a role in contributing to AA for milk protein synthesis. With all of the accumulated data demonstrating the presence of significant levels of circulating peptides in the blood, especially in ruminants, and with the demonstration of significant stomach and intestinal absorption of AA in the form of small peptides, peptide-bound AA in the blood may be important from a nutritional perspective.
|
EVIDENCE FOR PEPTIDE TRANSPORT OUT OF ENTEROCYTES |
|---|
detected peptidase activity in the brushborder and soluble fractions of rat intestinal mucosa for 13 dipeptides and 5 tripeptides with the soluble fractions constituting 80 to 90% and brushborder constituting 10 to 15% of total activity. The authors acknowledged that significant cross-contamination of each fraction occurred and lability of the peptide hydrolases prevented further purification and functional characterization of specific enzymes. Although these data demonstrate the presence of intracellular peptidases, they in no way demonstrate complete hydrolysis of all dietary protein-derived peptides. Inside the cell, peptides may either be hydrolyzed into free AA by peptidases and transported across the basolateral membrane by free AA transporters or intact peptides may be transported out of the cell by a peptide transporter.
The discovery of a peptide transporter in the basolateral membrane of enterocytes (Terada et al., 1999) and the knowledge that some peptides are relatively resistant to hydrolysis provides at least presumptive evidence that a carrier-mediated mechanism exists for transport of intact peptides to the bloodstream. In recent years, peptide transport from inside the cell to the basolateral compartment has been studied to evaluate transport characteristics of the basolateral peptide transporter. Irie et al. (2004) characterized the efflux properties of the basolateral peptide transporter in Caco-2 cells by preloading cells with [14C]-Gly-Sar and sampling the medium on the basolateral side after incubation. Efflux to the basolateral side was not affected by basolateral pH and was saturatable with a transporter affinity (Kt) of 24.8 ± 6.4 mM and maximum transport velocity (Vmax) of 1.61 ± 0.35 nmol/mg of protein per minute, indicating a low affinity. The Kt in the efflux direction was 5 to 10 times greater than in the influx direction, indicative of asymmetry in substrate affinity.
The identity of the basolateral peptide transporter remains elusive. Shepherd et al. (2002) identified a candidate protein in the basolateral membrane of rat jejunum through the use of photoaffinity labeling with [4-azido-D-Phe]-L-Ala. They observed that perfusion of the label at the serosal side had no effect on transepithelial movement of a nonhydrolyzable dipeptide D-Phe-L-Gln from the lumen, thus suggesting that [4-azido-D-Phe]-L-Ala was unable to enter the mucosal side from the basolateral side. This is supportive of a strongly asymmetric basolateral peptide transporter. However, when the label was perfused luminally, there was a 40% reduction in the rate of transepithelial transport of D-Phe-L-Gln and a 60% increase in the accumulation of this dipeptide in the mucosa, indicating that the label enters the cell from the lumen and binds to the endofacial side of the basolateral transporter and competes with and inhibits transepithelial movement of D-Phe-L-Gln. The labeled candidate protein was extracted from SDS-PAGE gels and analyzed by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Interestingly, database searches for the 112-kDa candidate protein with an isoelectric point of 6.5 revealed no similarities to PepT1 or other known proteins, thus establishing that this is a novel protein. Identification and molecular characterization of the basolateral peptide transporter will facilitate studies aimed at maximizing transport of peptides and peptide-like drugs across the small intestine.
|
CLONING AND CHARACTERIZATION OF PEPT1 |
|---|
). Table 1 summarizes the currently identified members of the peptide transporter family. In 1994, the first PepT1 mRNA was cloned from the rabbit (Fei et al., 1994). The remainder of this review will focus on structural and functional characteristics of PepT1 and the factors that regulate its expression including diet, developmental stage, hormones, diurnal rhythm, and disease.
|
; Fei et al., 1994). The first characterization of the peptide transporter in a livestock animal was reported by Matthews et al. (1996), in which sheep omasal RNA encoding for the peptide transporter was expressed in Xenopus oocytes and dipeptide transport was demonstrated. They observed that Gly-Sar uptake was saturatable (Kt = 0.4 mM) and inhibited 44% by carnosine, 94% by methionylglycine, and 91% by glycylleucine, but not by free glycine. The presence of mRNA that encodes for a peptide transporter in sheep omasal epithelia was confirmed by Pan et al. (1997) when they demonstrated transport for di- and tripeptides in oocytes injected with poly(A)+ RNA.
The peptide transporter, PepT1, has been cloned in multiple vertebrate species including the rabbit, rat, mouse, sheep, chicken, turkey, dog, human, pig, cattle, monkey, Atlantic cod, and zebrafish (Table 2). The size of PepT1 ranges from 707 to 729 AA. Expression or activity, or both, of PepT1 has been detected in other species including the guinea pig (Himukai et al., 1983), hamster (Burston and Matthews, 1990), and the black bear (Gilbert et al., 2007b). Peptide transporters have also been found in bacteria, yeast, plants, and invertebrates (Daniel, 2004). Recently, a prokaryotic H+-dependent peptide transporter, YdgR, was characterized and found to have features very similar to mammalian PepT1 (Weitz et al., 2007).
|
|
SUBSTRATE SPECIFICITY OF PEPT1 |
|---|
). Metal ions (i.e., Zn2+, Mn2+, and Cu2+) also can interact with transporter proteins to enhance peptide absorption (Leibach and Ganapathy, 1996).
It has been shown that PepT1 accepts pharmaceutical compounds with structural similarities to peptides called peptidomimetics and participates in their absorption, which is of major therapeutic value (Leibach and Ganapathy, 1996; Brandsch et al., 2004). This allows for the design of pharmacological drugs that possess acceptable oral availability because of transport by PepT1. In addition, PepT1 and PepT2 have been found to transport cephalosporins, penicillins, an aminopeptidase inhibitor bestatin, valine ester prodrugs of acyclovir and ganciclovir, and inhibitors of angiotensin-converting enzyme, to name a few (Steffansen et al., 2004). For an excellent review on the topic, see Brodin et al. (2002).
Peptide transport was determined to be proton-dependent (Figure 1A
). Using the 2-microelectrode voltage-clamp technique in PepT1 cRNA-injected oocytes and measuring currents for Gly-Sar transport, an inward H+ current was detected, demonstrating that peptide transport via PepT1 was proton-dependent among all species tested (Adibi, 1997; Pan et al., 2001; Chen et al., 2002a; Klang et al., 2005; Van et al., 2005). The proton-to-substrate ratio for neutral and cationic AA transported by PepT1 is 1, whereas the ratio for charged anionic AA is 2 (Steel et al., 1997). The unstirred water layer at the brushborder membrane is an isolated microenvironment free from the influence of the luminal contents and maintains a high extracellular concentration of protons (Adibi, 1997). After uptake of peptides and H+ by PepT1, protons are then transported out of the cell by the Na+/H+ exchanger on the brushborder membrane in exchange for Na+. The Na+ in turn is taken out of the cell by the Na+/K+ ATPase pump on the basolateral membrane where 3 Na+ are transported out of the cell in exchange for 2 K+, restoring the electrochemical gradient.
Binding affinity for charged peptides and peptidomimetic drugs changes when pH is altered (Daniel and Kottra, 2004). At brushborder membrane pH (5.5 to 6.0), neutral and acidic peptides were preferred substrates for ovine PepT1 (Pan et al., 2001) and rabbit and human PepT1 (Amasheh et al., 1997; Steel et al., 1997) when expressed in Xenopus oocytes. Chen et al. (2002a) tested uptake of [3H]-Gly-Sar in CHO cells transfected with chicken PepT1 and observed that uptake was greater at pH 6.0 and 6.5 than at 5.0, 5.5, 7.0, or 7.5, with similar results observed in oocytes expressing chicken PepT1. It is important to point out though that pH dependence has been studied in vitro. In the intact intestine, an acidic microclimate is maintained at the brushborder membrane independent of the intestinal luminal contents (Daniel, 2004). Changes within the lumen do not directly affect the unstirred water layer, and thus, the results of in vitro studies are difficult to relate to the intact, functioning intestine.
The affinity constants of peptide substrates for PepT1 vary widely from micromolar to millimolar. Two methods have been employed to measure transport kinetics or binding affinity of peptide substrates. The transport (Kt) of peptides is measured in Xenopus oocytes expressing PepT1. The transport of the proton, which is cotransported with the peptide, is measured by voltage-clamping studies. Alternatively, the substrate affinity of PepT1 is determined by an uptake inhibition assay. In this assay, the concentration of unlabeled peptide that inhibits 50% of labeled Gly-Sar uptake (IC50) is determined. In Table 3, the Kt and IC50 values for sheep, chicken, and pig PepT1 are shown compared with human PepT1. The Kt/IC50 values for Gly-Sar were similar for sheep (0.61 mM), chicken (0.47 mM), pig (0.94 mM), turkey (0.69 mM; Van et al., 2005), and human (1.2 mM) PepT1. In contrast, the Kt values for the di- and tripeptides tested ranged from 0.027 to 6.9 mM, and the IC50 values ranged from 0.005 to 7.9 mM. The reported IC50 values for human PepT1 were greater than for sheep, pig, and chicken PepT1 and may reflect the different cells used for the assays (MDCK cells for human PepT1 vs CHO cells for sheep, chicken, and pig PepT1). In general, tetrapeptides were poor substrates for PepT1 as indicated by the IC50 values ranging from 0.95 to 27 mM or no response in the voltage clamp as-says. In most cases, the IC50 values are comparable to the Kt values for sheep and chicken PepT1. Peptides with 2 basic AA (Arg-Lys, Lys-Lys, Lys-Trp-Lys, Lys-Tyr-Lys) were poorly transported, which suggests PepT1 can accommodate 2 positive charges, 1 from the peptide and 1 from the proton, but not 3 positive charges. An interesting species difference was seen with the dibasic Lys-Lys, which was a good substrate for the ruminant sheep PepT1 but a poor substrate for the mongastric chicken, pig, and human PepT1. The peptides that were used in these studies represented a variety of molecular weights, net charges, and hydrophobicities, but there was no correlation between any of these 3 variables and affinity constants. It should also be pointed out that although PepT1 is generally classified as a low-affinity transporter, affinity constants for substrates can vary depending on the model system and transport conditions used. In a review, Brodin et al. (2002) emphasized a point made by Meredith and Boyd (2000) that affinity is the result of a sum of a number of discrete interactions and cannot be determined by a single variable.
|
; Taylor et al., 1980; Pan et al., 2001; Klang et al., 2005) and stimulatory effects of peptides on transport activity (Addison et al., 1974), thus further complicating the relationship between peptide composition and transporter affinity. Peptide composition may also influence the rate and extent of intracellular and extracellular hydrolysis, as demonstrated by Adibi et al. (1986) when they observed that rate of appearance of constituent AA comprising different dipeptides differed after intravenous injection.
Vig et al. (2006) conducted a comprehensive study aimed at determining PepT1 substrate specificity and structural characteristics. They tested 73 peptides (mostly dipeptides) for their binding affinity and ability to enhance transport activity of PepT1 in MDCK cells using Gly-Sar as a reference substrate. Twenty-one substrates, many of which showed affinity for PepT1, did not activate PepT1, suggesting that they are not substrates for transport. This finding contradicts the current mode of thought that all di- and tripeptides serve as substrates for PepT1. Of these 21 nontransported substrates, Trp-Trp was further evaluated for transport activity. Uptake of [3H]-Trp-Trp in MDCK-PepT1 transfected and MDCK-mock transfected cells showed similar kinetics, confirming that Trp-Trp was not a substrate for transport. For peptides that did activate PepT1, Vig et al. (2006) concluded that, in general, N-terminal, large, hydrophobic AA enhanced PepT1 activity. In terms of charge, neutral AA were better substrates for PepT1. Aromaticity enhanced activity, and the authors suggested that the inability of Trp-Trp to be transported by PepT1 was the result of the large total volume of this dipeptide exceeding total capacity of the binding pocket of the PepT1 protein.
There has been much effort to characterize the regions of the PepT1 protein important in substrate recognition and affinity as well as proton binding. The N-terminal half of PepT1 consisting of transmembrane domains (TMD) 1 to 6 is thought to contain the AA residues associated with pH changes during peptide uptake and substrate binding (Terada et al., 2000a), whereas the C-terminal AA (TMD 7 to 9) are thought to play an important role in determining substrate affinity (Fei et al., 1998). Histidine residues are important in the substrate-binding domain of peptide transporters. Pretreatment of renal brushborder membrane vesicles with diethyl pyrocarbonate, a histidine-modifying reagent, abolished peptide transport (Meredith and Boyd, 2000). Mutations of human or rat PepT1 His-57 and rat His-121 (Terada et al., 1996; Fei et al., 1997) either markedly reduced or abolished transport activity and were therefore considered to be extremely important for PepT1 activity. Mutations of rabbit PepT1 His-57 also produced a nonfunctional transporter, and mutations of the tyrosine residues that flank rabbit His-57 (Tyr-56 or Tyr-64) also produced a transporter with no activity (Chen et al., 2000). Immunostaining revealed that transporter number on the plasma membrane was not impaired by His-57 or His-121 mutation (Terada et al., 1996), indicating that protein function rather than production was impaired. Mutations of rabbit His-121 resulted in reduced uptake and a much lower affinity, suggesting a role in substrate recognition. The largest decline in affinity occurred for anionic substrates, indicating that His-121 plays a role in protonating negatively charged substrates (Chen et al., 2000). It was then proposed that His-57 is critical for H+ coupling and that the adjacent aromatic residues stabilize the positive charge on the proton through cation-
interactions. Doring et al. (2002) suggested that rabbit PepT1 AA residues 1 to 59 are an important part of the substrate-binding domain and interact with side chains of substrates, and AA residues 60 to 91 are important for the pH profiles of peptide uptake, consistent with previous studies. Collectively, these studies are in agreement with a molecular modeling approach to studying structure and function of PepT1, in which it was predicted that TMD 7 to 10 form part of a peptide transport channel and TMD 1, 3, and 5 form the other half, with specific AA residues in each TMD playing an important role in substrate binding (Bolger et al., 1998).
There are other various structural characteristics that have a marked effect on peptide transport activity. Peptide transporters exhibit a greater affinity for peptides containing the L-isomers of AA compared with the D-isomers (Leibach and Ganapathy, 1996; Brandsch et al., 2004). Terada et al. (2000b) demonstrated that dipeptides with a modified 
-amino group, such as N-methyl-Gly-Gly and N-formyl-Met-Ala, showed reduced affinity for PepT1 than their original counterparts (Gly-Gly and Met-Ala, respectively). A peptide bond is not required for substrate recognition. Doring et al. (1998) found that PepT1 and PepT2 recognized and transported substrates as simple as omega-amino fatty acids with a positive amino and negatively charged carboxyl head group, which were separated by a minimum of 4 methylene groups.
|
TISSUE AND CELLULAR DISTRIBUTION OF PEPT1 |
|---|
). The renal peptide transporter, PepT2, in contrast to PepT1, is expressed at the greatest levels in the kidney tubule and to a lesser extent in the lung, mammary gland, choroid plexus, and glia cells in the nervous system, but not in the intestine (Daniel, 2004). Sites with lower levels of PepT1 expression or activity, or both, include the colon, human bile duct epithelium, and rabbit brain and liver (Miyamoto et al., 1996; Knutter et al., 2002; Ford et al., 2003). Although almost no dietary peptides or AA reach the colon, there is a large supply of endogenous proteins in the colon, possibly serving as substrates for proteolysis by microflora. Intestinal-specific expression of PepT1 is regulated by CDX2, a transcription factor that plays an important role in the proliferation, differentiation, and maturation of intestinal epithelial cells (Shimakura et al., 2006b). This regulation most likely occurs through interactions with the ubiquitous Sp1 transcription factor for which there is a recognition site in the PepT1 promoter that controls basal activity (Shimakura et al., 2005).
Within the small intestine there are interesting species differences in PepT1 expression. Expression of PepT1 mRNA was detected in the greatest quantities in the duodenum, jejunum, and ileum of the chicken, pig, and ruminant, respectively (Chen et al., 1999). These differences may correspond to differences in the site of maximal digestion and absorption. In the chicken, there was considerable expression of PepT1 in the ceca in addition to the small intestine and kidney (Chen et al., 2002a). In rats, protein expression was greatest in the distal small intestine (ileum) as compared with the jejunum (Tanaka et al., 1998). Howard et al. (2004) found that rat PepT1 mRNA was equally expressed throughout the length of the small intestine. In the black bear, mRNA abundance of PepT1 was greatest in the mid-region of the intestinal tract (Gilbert et al., 2007b). In contrast to most species, ruminants, such as sheep and cattle, express PepT1 mRNA in the stomach, specifically in the omasum and rumen (Chen et al., 1999; Pan et al., 2001).
Expression of intestinal PepT1 increases with enterocyte maturation. In situ hybridization has been used to examine cellular distribution of PepT1 mRNA. Freeman et al. (1995) found that rabbit PepT1 mRNA staining was restricted to enterocytes between the cryptvillus junction and villus tip with increased expression toward the tip. Within the colon, expression was restricted to the surface columnar epithelial cells. Immunohistochemical analyses showed that in rats, PepT1 protein was localized to the brushborder membrane of cells lining the villi with increasing intensity from the cryptvillus junction to the villus tip; no protein was detected in the crypts (Ogihara et al., 1996; Tanaka et al., 1998). Interestingly, restriction of PepT1 to the brushborder membrane may be developmentally regulated. Hussain et al. (2002) observed PepT1 protein at fetal d 18 and day of birth in the duodenum, and right after birth, PepT1 spread to the subapical cytoplasm, basal cytoplasm, and basolateral membrane in addition to the apical membrane. At weaning and adulthood, PepT1 expression was restricted to the brushborder membrane. Because its expression is mainly detected in the apical membrane of small intestinal enterocytes lining the villi, PepT1, as a broad specificity, low-affinity, high-capacity transporter, is strategically located to maximize assimilation of AA from the diet.
|
DIETARY REGULATION OF PEPT1 |
|---|
rationalized the differences observed in transporter regulation among different nutrients. They suggested that transporters for metabolizable nontoxic nutrients, such as sugars, nonessential AA, and short-chain fatty acids, should be upregulated with increasing dietary level and that transporters for essential nutrients, which are toxic in large quantities, such as water-soluble vitamins and minerals, should be downregulated by increasing concentrations of substrates. For AA, the matter is complicated by the fact that AA can be also used as sources of energy. Some AA are more essential to a cell than others, or more toxic. Furthermore, enterocytes express transporters with overlapping substrate specificity and transporters that mediate the movement of both essential and nonessential AA (Karasov et al., 1987). Thus, it becomes difficult to predict whether a transporter should be upregulated in response to certain AA deficiencies or imbalances.
Substrates can regulate transporters specifically and nonspecifically. Nonspecific regulation involves general changes in mucosal surface area, transcellular electrochemical gradient, paracellular permeability, and plasma membrane lipid composition and fluidity (Ferraris, 1994, 2000). Specific regulation includes changes in the site density of specific transporters in enterocytes as a result of changes in protein synthesis or degradation rate or an increased insertion of preformed cytoplasmic transporters into the brushborder membrane (Ferraris and Diamond, 1989).
The influence of dietary protein on PepT1 expression and activity has been a popular area of research in recent years. From a commercial livestock industry perspective, dietary protein is a costly nutrient, and even fractional improvements in its utilization have the potential to save the industry millions of dollars as well as reduce nitrogen excretion into the environment. Although dietary protein regulation of PepT1 expression and activity has not been as well studied in agriculturally important species, much of the work conducted in biomedical species is broadly applicable and provides a starting point for other studies aimed at determining the nutritional relevance of peptide transport.
Our laboratory was the first to evaluate the effect of dietary protein quality on expression of PepT1. Broiler chicks fed a diet containing a low-quality protein (corn gluten meal) showed decreased expression of PepT1 from d 3 to d 7 followed by an increase in expression to d 14. In chicks that consumed an equal amount of the same diet substituted with a greater quality protein (soybean meal), expression of PepT1 mRNA rose continuously with age from d 3 to d 14, and birds were heavier (Gilbert et al., 2008). It was thought that in birds consuming the soybean meal, continuous up-regulation of the transporter served as a mechanism to maximize assimilation of AA from a well-balanced mixture, whereas in birds consuming the corn gluten meal, the effect was more complicated due to the severely imbalanced diet.
In general, peptide transporter expression and activity increases with dietary protein and peptide level. Shiraga et al. (1999) maintained rats on a 20% casein diet for 1 wk. A group of rats was switched to a protein-free diet, and others were switched to diets consisting of 50, 20, or 5% casein, 20% of a dipeptide, or 10% of a single AA, fed for 3 d. Greater levels of dietary protein were associated with greater expression of PepT1 mRNA and protein. Compared with rats fed the protein-free diet, there was an increase in PepT1 mRNA and protein for rats fed the Gly-Phe diet and significantly greater PepT1 mRNA for the phenylalanine-fed but not the glycine-fed rats. Results from this study demonstrated that PepT1 may be regulated by specific AA and is very responsive to changes in dietary protein, in particular quantity and composition.
A high-protein diet (72 vs. 18%) resulted in increased absorption of the dipeptide carnosine in mice (Ferraris et al., 1988), and feeding a protein-free diet for 40 to 84 d resulted in an increased ability of rat jejunum to transport a methionine dipeptide but not free methionine (Lis et al., 1972). It is perhaps counterintuitive that both abundance and lack of substrate will stimulate PepT1. Upregulation by a high-protein diet appears to be a mechanism to take advantage of the abundant resource, whereas upregulation in response to a lack of substrate may be a compensatory mechanism to scavenge AA in the lumen. The magnitude of the response in changes of PepT1 expression and activity will probably be dependent on length of time for a particular dietary manipulation, availability of transportable substrate, AA composition (concentrations of free and peptide-bound), and presence of other components in the lumen that can change digestive and absorptive dynamics (i.e., sugars, vitamins, minerals, fats, etc.).
In chickens, an increase in PepT1 mRNA was observed in the intestine of chickens fed 18 and 24% CP diets with restricted food intake, and a decrease in PepT1 mRNA was observed in chickens fed a 12% CP diet (Chen et al., 2005). In chicks fed the 24% CP diet ad libitum, there was lower expression of PepT1 as compared with chicks consuming restricted amounts of the diet. Thus, the increase in PepT1 was probably not due to an increase in CP but instead to the feed restriction. Similarly, Gilbert et al. (2008) found that feed restriction increased expression of PepT1 mRNA from d 3 to 14 posthatch in broilers.
Regulation of PepT1 by dietary substrate appears to occur by 2 mechanisms: 1) by increasing mRNA stability and 2) by increasing gene transcription rate (Adibi, 2003). Upregulation of PepT1 in response to low-protein or protein-free diets may be a mechanism to compensate for reduced surface area, because protein malnutrition is known to cause significant reductions in intestinal surface area and mucosal mass (Ferraris and Carey, 2000). In protein-malnourished rats (5% CP diet), intestinal villi were shorter in comparison with control rats (20% CP diet). There were older enterocytes on villi in malnourished rats compared with those at the same villus height in well-fed rats. Hence, the mechanism leading to a fasting or malnutrition-related increase in nutrient transport may be a combination of increased gene expression and ratio of transporting to nontransporting cells (Ferraris and Carey, 2000).
The mechanism of dietary protein regulation of nutrient transporters may be through a pathway of direct substrate regulation. Shiraga et al. (1999) found elements in the 5' upstream region of rat PepT1 responsive to peptides and free AA. Fei et al. (2000) identified a similar AA responsive element in the promoter region of mouse PepT1. To study responses of the PepT1 gene to specific AA or peptide substrates, in vitro experiments using cultured cells that express PepT1 were conducted. This method establishes a controlled model for observing the effect of a single substrate on regulation of the PepT1 gene. Addition of Gly-Sar to the medium of Caco-2 cells caused a 3-fold and 2-fold increase in the expression of human PepT1 mRNA and protein, respectively (Thamotharan et al., 1998). Treatment with brefeldin, an inhibitor of protein transport from the endoplasmic reticulum to the Golgi, abolished transport, showing that increased synthesis and processing by the trans Golgi network accounted for increased expression at the apical membrane. Similar uptake experiments were performed with a naturally occurring dipeptide, Gly-Gln, confirming the physiological relevance of these findings (Walker et al., 1998). These results demonstrate direct regulation of PepT1 by its own substrate, which has important implications for nutritional supplements aimed at improving AA uptake. Additionally, it is worth noting that free AA uptake may be indirectly regulated by PepT1 activity. Because many AA transporters serve as exchangers, filling of cells with a variety of AA as peptides by PepT1 may be important for net movement of AA. For example, Wenzel et al. (2001) demonstrated that uptake of dipeptides caused stimulation of AA uptake by the bo,+ system.
Short-term feed deprivation is common in domestic animals. In cattle and pigs, it can occur under conditions such as weaning and relocation, and in chickens, it is common for a delayed access to feed for 48 h or more after hatch to occur due to bird processing and transport (Noy and Sklan, 1998; Vieira and Moran, 1999; Bigot et al., 2003). Short-term starvation in rats has been shown to increase mRNA and protein expression of PepT1 (Ihara et al., 2000). Rats starved for 4 d exhibited a 179% increase in mRNA and protein expression. Rats that were fed 50% of the intake of controls for 10 d and rats given total parenteral nutrition (TPN) for 10 d exhibited a 161 and 164% increase in PepT1 mRNA, respectively. This dramatic upregulation occurred in spite of the fact that the mucosal weight decreased in the starved and TPN group by 41 and 50%, respectively. In a different study, PepT1 mRNA and protein increased 3-fold, and the rate of peptide transport in rats increased dramatically after only 1 d of fasting (Thamotharan et al., 1999a). Ferraris and Diamond (1989) described the regulation of nutrient transporters as a way to match uptake capacity to requirements without wasting energy on unnecessary transporters.
Howard et al. (2004) examined the effects of TPN and administration of glucagon-like peptide (GLP)-2 on the mRNA expression of PepT1 in rat small intestine. Total parenteral nutrition for 7 d upregulated PepT1 mRNA in the distal intestine, whereas proximal (duodenal) mRNA was unchanged. Administration of GLP-2 inhibited the effect of TPN on mRNA expression of PepT1. It is known that GLP-2 is a trophic factor that maintains cellular protein synthesis during luminal starvation, and perhaps infusion of GLP-2 reduced the need for upregulation of PepT1. During luminal starvation, the need to absorb endogenous protein products increases, and thus, apical membrane transporters, which in this study included NBAT, EAAC1, ASCT2, and PepT1, were upregulated in the distal intestine to maximize assimilation of AA (Howard et al., 2004).
Peptide transporter expression and activity may also be altered by diurnal rhythm (Pan et al., 2002, 2003, 2004). To date, studies involving this phenomenon have been conducted in rodents, which are nocturnal mammals. In rats, PepT1 expression increased at night, paralleling the normal feeding behavior of these rodents. Interestingly, fasting or an imposed daytime feeding abolished this pattern, and expression patterns quickly changed to accommodate the nutritional needs of the gut (Pan et al., 2004). This hints to the plasticity of peptide transporter expression and how quickly the intestine is able to adapt in response to environmental changes. This has important implications across mammalian and avian species that can vary dramatically in feeding behavior.
Increased expression and activity of PepT1 during feed deprivation or restriction may serve as a mechanism to compensate for reduced mucosal surface area. Delayed access to feed for 36 h posthatch resulted in depressed villus height and crypt depth and reduced growth in all intestinal segments (Uni et al., 1998). Silva et al. (2007) subjected male broiler chicks to a feed restriction at 30% of ad libitum intake from 7 to 14 d and found that feed restriction decreased the surface area of the tip of the enterocytes in the small intestine at 14 d. Ihara et al. (2000) found that starvation or TPN for 4 d in rats was accompanied by reductions in mucosal weight by 41 and 50%, respectively. Feed restriction causes reduced absorptive surface area due to intestinal mucosal atrophy, which may explain why peptide transporter gene expression and protein activity is increased. As we begin to understand the factors that control expression of PepT1 and discover how changes in the diet can manipulate these factors to our advantage, we can begin to effect changes in the phenotype (BW gain, feed conversion, milk yield, etc.) through dietary manipulation of gene expression.
Shimakura et al. (2006a) proposed a mechanism for induction of intestinal PepT1 during situations of fasting or starvation. They demonstrated that PepT1 induction is mediated by peroxisome-proliferator-activated receptor (PPAR), a member of a family of nuclear receptors activated by fatty acid ligands, thus playing an important role in the adaptive response to starvation. In PPAR-null mice, the fasting-induced expression of PepT1 was abolished, whereas in wild-type mice, there were significant increases in PPAR and PepT1 after 48 h of fasting. When the PPAR ligand WY-14643 was orally gavaged to fed rats for 5 d, expression of PepT1 mRNA increased, and when Caco-2 cells were treated with the same ligand, expression of PepT1 increased and uptake of [3H]-Gly-Sar increased. There was no increase in PepT1 expression or peptide uptake when ligands specific for PPARY or PPARβ/
were administered to cells. Although the functional response element or other regulatory region was not determined, the authors suggested that PepT1 is either directly regulated by PPAR through binding to a regulatory region or that PPAR induces expression of transcription factors, such as SP1 or CDX2, which were also upregulated in this study in response to fasting.
|
DEVELOPMENTAL AND HORMONAL REGULATION OF PEPT1 |
|---|
), leptin (Buyse et al., 2001), epidermal growth factor (Nielsen et al., 2001, 2003), and thyroid hormone, T3 (Ashida et al., 2002). To describe each of these hormones in detail would be beyond the scope of this review; hence, we will only discuss insulin and leptin, because these may be most relevant from a nutritional perspective. Insulin is not a hormone located in the gut lumen, but during normal physiological conditions, circulating insulin can bind to receptors located on the basolateral membrane of enterocytes (Adibi, 2003). Addition of 5 nM insulin to Caco-2 cells increased the number of PepT1 transporters and stimulated Gly-Gln uptake (Thamotharan et al., 1999b). Treatment of the basolateral membrane with 50 ng/mL of insulin for 1 h increased Gly-Sar uptake, whereas treatment of the apical membrane had no effect (Nielsen et al., 2003). Leptin, a hormone that suppresses appetite and increases metabolism, is secreted by both adipocytes and by the stomach and has been found to be released into the stomach and reach the intestine in a nondegraded form where there are apical leptin receptors on enterocytes (Buyse et al., 2001). Addition of leptin to only the apical membrane of Caco-2 cells or mouse jejunum increased cephalexin and Gly-Sar transport and increased membrane PepT1 while reducing intracellular quantities. Addition to only the basolateral membrane had no effect on transport or PepT1 expression. Hindlet et al. (2007) demonstrated that in leptin deficient ob/ob mice, there was impaired peptide transport activity (50% less Gly-Sar uptake as compared with wild-type mice) and gene expression (40% and 2-fold less PepT1 protein and mRNA, respectively, as compared with wild-type mice) that was restored upon 7 d of continuous subcutaneous administration of leptin. Stimulation of peptide transport by insulin or leptin is thought to involve increased trafficking of cytoplasmic PepT1 proteins to the apical membrane.
Peptide transport activity varies with age (Leibach and Ganapathy, 1996), and dietary changes during the prenatal, suckling, or weaning period may have irreversible effects on nutrient transport that carry over into adulthood (Karasov et al., 1985; Pacha, 2000). Early ontogenetic development of the gut is characterized by morphogenesis and cytodifferentiation during fetal development, at which time the intestine becomes prepared for postnatal life when it will assume complete responsibility for nutrient absorption (Puchal and Buddington, 1992; Pacha, 2000). Prenatal expression of glucose, peptide, and AA transporters was reported to be present in humans, guinea pigs, sheep, rabbits, and rats (Guandalini and Rubino, 1982; Pacha, 2000; Shen et al., 2001).
At birth, the intestine becomes the site of nutrient assimilation, and the animal begins to consume a high-protein milk diet. At weaning, the animal shifts to the adult diet consisting of predominantly carbohydrates. Xiao (2006) conducted one of few studies involving developmental regulation of peptide and AA transporters in a domestic animal. This study investigated the development of nutrient transporter expression during the 2 most rapid stages of intestinal development: 1) at birth when the pig shifts diet from the amniotic fluid to milk of the mother and 2) at weaning when the pig shifts diet from the milk of the mother to a solid diet (Pacha, 2000). During suckling (d 0 to 21) and postweaning (d 21 to 35), pig intestinal PepT1 mRNA remained constant except for a decline 1 d after birth and a peak 1 d after weaning. Protein expression of PepT1 generally decreased with age in the duodenum and increased with age in the jejunum and ileum (Xiao, 2006). Protein and mRNA expression of PepT1 decreased in all segments during the first few days of suckling in contrast to several brushborder AA transporters and peptide hydrolases that increased, suggesting a more complete hydrolysis of peptides in the lumen and less availability of peptides during this time period.
Other mammals show similar changes in expression of PepT1 that correlate with changes in the diet and intestinal development. Rat intestinal PepT1 mRNA and protein were present at fetal d 20, increased at birth, and reached maximal expression levels during d 3 to 5 (Shen et al., 2001). The PepT1 mRNA levels then dropped to about 12% of maximal levels by d 14 and rose to 23 to 58% of maximal expression levels at weaning, after which expression plateaued. Similarly, PepT1 protein dropped sharply after d 5, rose to 59 to 88% of maximal expression at weaning, and then plateaued at adulthood (d 75). Colonic PepT1 mRNA and protein were detected at d 1 to d 5, dropped to almost undetectable levels at d 7, and were undetectable afterwards at all days. The ability of the colon to transport peptides was suggested to serve as a compensatory mechanism for the temporary low capacity for nutrient absorption in the small intestine.
The developmental regulation of ovine PepT1 in the dorsal rumen, ventral rumen, omasum, duodenum, jejunum, and ileum of lambs was studied at 2, 4, 6, and 8 wk of age (Poole et al., 2003). Ovine PepT1 was present in all tissues examined at 2 wk of age and was not influenced by age. Because expression was evaluated beginning at 2 wk, there may have been earlier changes occurring, for example, induction by suckling as reported in other species. Similarly, the last sampling date was 8 wk, and lambs were not weaned in this study, a dietary shift that induces changes in the transporter in other mammals. Poole et al. (2003) observed that lambs allowed to nurse but not allowed access to a creep diet at birth expressed greater quantities of PepT1 mRNA in the rumen. Because of the lack of luminal substrate (i.e., the presence of the reticular groove should prevent milk from entering the rumen) to act as inducers of expression, it was suggested that blood-derived factors may be involved in the developmental regulation of ovine PepT1.
The turkey and chicken, although exhibiting a different mode of embryological development than mammalian species, exhibit a similar developmental regulation of PepT1 to prepare the intestine for immediate uptake of nutrients at hatch. In the bird, however, the diet shifts from the lipid-rich yolk as the main source of nutrients during embryological development to a carbohydrate- and protein-rich diet posthatch. Van et al. (2005) observed a 3.3-fold increase in mRNA expression levels of PepT1 in turkey intestinal tissue from 5 d before hatch to day of hatch, which is similar to rats in which PepT1 mRNA spiked at birth (Shen et al., 2001). Chen et al. (2005) observed a 14- to 50-fold increase in intestinal chicken PepT1 mRNA from embryo d 18 to day of hatch, with expression peaking right before hatch. More recently, Gilbert et al. (2007a) conducted a comprehensive study of the developmental regulation of peptide, AA, and sugar transporter mRNA in the small intestine of 2 genetically selected lines of broiler chicks from embryo d 18 to 14 posthatch. For all brushborder membrane transporters evaluated, mRNA abundance increased with age with dramatic induction from embryo d 18 to day of hatch. It will be of interest to determine if embryonic nutritional modulations (e.g., in ovo nutrient administration) have effects on the developmental timing of PepT1 gene expression. Expression of PepT1 mRNA increased with age with the greatest expression at d 14 posthatch, suggesting that the capacity for peptide absorption in broilers continues to increase with maturity. Interestingly, in that same study, it was observed that out of 14 transporter genes evaluated for mRNA expression, only PepT1 was influenced by genetic line. There was an approximately 2-fold difference in PepT1 mRNA, suggesting a greater capacity for absorption of peptide-bound AA. Although transporter expression during development may be controlled by genetic determinants, expression is also controlled at the transcriptional levels by many factors, including the diet, suggesting that developmental programming is not absolutely predetermined (Pacha, 2000).
|
REGULATION OF PEPT1 DURING DISEASE |
|---|
It has been found that PepT1, which is expressed in very small quantities in the colon compared with the proximal intestine (Ford et al., 2003), is upregulated in the colon of patients suffering from short-bowel syndrome (Ziegler et al., 2002), perhaps serving as a mechanism to maximize absorption of dietary AA in patients with a restricted capacity (Ziegler et al., 2002). This same pattern of expression is also exhibited in patients suffering from chronic ulcerative colitis and Crohn’s disease (Merlin et al., 2001). Also, N-formyl-methionyl-leucyl-phenylalanine, a major proinflammatory peptide of the human colonic lumen, is found at greater concentrations in the colon where bacterial loads are greater compared with the small intestine and where PepT1 is normally minimally expressed. However, during a diseased state, PepT1 expression increases in the colon, transports N-formyl-methionyl-leucyl-phenylalanine, and stimulates expression of MHC-1 genes (Merlin et al., 2001). Interestingly, leptin is increased in inflamed colonic mucosa similar to PepT1. Nduati et al. (2007) demonstrated that leptin induces colonic expression of hPepT1 via the transcription factors CREB and CDX2.
Bacterial infections caused by endotoxin administration in rats were shown to regulate expression and activity of PepT1 in the intestinal tract (Shu et al., 2002). However, in contrast to the other types of diseases discussed, endotoxin treatment served to downregulate PepT1 mRNA and protein (32 to 62% of controls) in rats, most likely through the action of proinflammatory cytokines (IL-1β and tumor necrosis factor-) that are increased by lipopolysaccharide treatment. Sekikawa et al. (2003) found that infection with the nematode Nippostrongylus brasiliensis in rats downregulated mRNA or protein expression levels, or both, of GLUT5, PepT1, LAT2, and SGLT1 7 to 14 d after infection. This downregulation of nutrient transporters could contribute to the malnutrition that ensues in patients with a nematode infection (Sekikawa et al., 2003).
Barbot et al. (2003) examined the effects of Cryptosporidium parvum, a cause of diarrhea, on PepT1 in rat intestine from d 4 to 50. Rats were gastrically infused with C. parvum on d 4, and the parasite disappeared on d 21. On d 10, the parasitic infection was greater in the ileum than in the proximal small intestine. Villus atrophy occurred throughout the small intestine but was most pronounced in the ileum with villus morphology returning to normal on d 21. PepT1 mRNA levels increased on d 10, during the peak of the infection, and returned to normal levels after removal of the parasite (d 21). In the control animals, PepT1 mRNA was evenly distributed throughout the small intestine from d 4 to 50. Immunohistochemical staining revealed expression of PepT1 protein from d 4 to 50 on the brush-border membrane throughout the small intestine, but during the peak of infection (d 10 through 12), PepT1 was also detected intracellularly, after which it disappeared. This intracellular staining was not detected in the control animals. It is not clear how the parasite may regulate PepT1 expression. One theory is that it may interfere with trafficking of the protein from the cytoplasm to the membrane (Barbot et al., 2003).
|
ALTERNATIVE MECHANISMS OF PEPTIDE UPTAKE |
|---|
). In earlier studies of peptide transport in the ruminant, Matthews and Webb (1995) and McCollum and Webb (1998) alluded to paracellular transport as a potential route of peptide uptake in the ruminant. Madara and Pappenheimer (1987) and Pappenheimer and Reiss (1987) showed that mediated uptake of a substrate (e.g., peptides via PepT1) was a prerequisite for paracellular uptake. In normal rats that ingested a 5% glucose solution containing [123I]-labeled octapeptides of D-AA, assumed to be resistant to hydrolysis and to not be a substrate for the peptide transporter, there was as much as 50% of the peptide excreted into the urine intact (Pappenheimer et al., 1994). This suggests that transcellular Na+-dependent glucose transport triggers dilation of the tight junctions permitting absorption of other solutes through solvent drag (Pappenheimer and Reiss, 1987). Chediack et al. (2006) reported paracellular uptake of 2 D-dipeptides, serine-aspartic acid and serine-lysine, in the house sparrow gut and negligible activity of the peptide transporter.
In species in which the permeability of tight junctions prevents significant passive epithelial transport, there may be strategies for overcoming this barrier. Motlekar et al. (2006) reported efforts to enhance uptake of the low molecular weight heparin, ardeparin, through the use of zona occludens toxin, which is an enterotoxin obtained from Vibrio cholerae. This toxin was shown to reversibly and safely open tight junctions. They demonstrated that in the presence of 100 µg/kg of AT1002, a novel synthetic hexapeptide derived from zona occludens toxin, oral bioavailability of ardeparin was improved in the rat with no detectable cytotoxicity or morphological damage to intestinal cells. Paracellular uptake of peptides or peptidomimetics may also be influenced by bioactive components in the diet. Capsaicin, the chemical responsible for the pungent properties of hot peppers, was shown to reduce uptake of cephalexin in the jejunum and ileum of rats, primarily due to a decrease in transcellular transport mediated by PepT1 (Komori et al., 2007). Interestingly, capsaicin increased the paracellular permeabililty of cephalexin. These results indicated that the transcellular and paracellular transport of peptides can be independently altered.
An additional alternate route of peptide uptake could be through the non carrier-mediated transcellular route. In recent years, there has been much research focused on cell-penetrating peptides (CPP) as routes of delivery for pharmacological substances. Cell-penetrating peptides are peptides that are capable of translocating across the plasma membrane and may at the same time deliver cargo to the interior of the cell (Palm et al., 2007) including oligonucleotides (Morris et al., 1997), peptide nucleic acids (Eriksson et al., 2002), plasmids (Morris et al., 1999), proteins (Saalik et al., 2004), and liposomes (Torchilin et al., 2001). Hallbrink et al. (2001) demonstrated that 4 different CPP were capable of translocating a labeled pentapeptide into Bowes human melanoma cells. The mechanism of how CPP are able to translocate the plasma membrane is still unclear. This may occur through direct penetration of the plasma membrane or by various types of endocytosis followed by endosomal release. Translocation may be initiated by binding of cationic moieties on the CPP to negative charges on the plasma membrane. Sai et al. (1998) demonstrated that intestinal absorption of a novel fluorescence-derivatized cationic peptide 001-C8 [H-MeTyr-Arg-MeArg-D-Leu-NH(CH2)8NH2] occurred by adsorptive-mediated endocytosis, which occurs through binding of the cationic moieties of the peptide to negative charges on the plasma membrane. The cationic peptide Tat, derived from the protein transduction domain of HIV-1, was reported to enter cells by macropinocytosis (Kaplan et al., 2005) or by binding to heparin sulfate receptors and entering by either caveolar endocytosis (Fittipaldi and Giacca, 2005) or clathrin-mediated endocytosis (Richard et al., 2005). Cell-penetrating peptides could serve as a novel nutrient delivery agent to overcome the impenetrability of the cell barrier. It would also be of interest to further explore the possibility of delivering larger peptides that are not transported by PepT1 to the interior of the cell.
|
DIETARY PEPTIDES |
|---|
). Peptides offer an advantage over free AA that are unstable or insoluble and may represent a more cost-effective way of supplementing AA (Zimmerman et al., 1997; Lindemann et al., 2000; Dabrowski et al., 2003). Because of the labile nature of glutamine and insoluble nature of tyrosine or tryptophan, absorption in a small peptide form increases the availability of these AA to the body (Adibi, 1997). Also, the use of dipeptides reduces the hypertonicity that results from a free AA feeding solution.
In all animal species evaluated thus far, there appears to be considerable capacity in the small intestine for the absorption of AA in the form of small peptides. Thus, it is reasonable to predict that incorporation of small peptides or hydrolyzed proteins into the diet would exploit this ability and potentially enhance animal growth and development. Additionally, di- and tripeptides are absorbed quickly and efficiently by the intestine without initial pancreatic digestion (Zambonino Infante et al., 1997).
There are some reports of feeding hydrolyzed proteins to fish, which express high levels of PepT1 in the small intestine (Verri et al., 2003; Ronnestad et al., 2007). Weight gain and survival rate were improved for sea bass larva fed 20 or 40% of total nitrogen as peptides, respectively, as compared with those fed only fish meal (Zambonino Infante et al., 1997). Similarly, survival, final weight, total biomass, and malformation rates of sea bass or carp larvae were optimal when 52% CP diets were fed that contained equal amounts of yeast and fish protein hydrolysate as contributors of CP as compared with equal parts of yeast and soy protein or fish meal (Cahu et al., 1998). In sea bass larvae fed fish meal or fish meal plus hydrolysate as 25% of total nitrogen, weight gains were similar in both groups (Cahu et al., 1999). It can be inferred from these studies that substitution of fish protein hydrolysate for fish meal can improve performance of sea bass or carp larvae. In these studies, the fish meal hydrolysate was not produced from the same fish meal to which it was compared; thus, these results must be considered preliminary.
In a more recent study, sea bass larvae were fed diets containing either 10% of a sardine by-product hydrolysate (SH; 42% CP; main fraction of peptides between 200 to 500 Da), a commercial enzymatic hydrolysate (CPSP; 82% CP; main fraction of peptides between 500 to 2500 Da), or 19% of either of the 2 hydrolysates (Kotzamanis et al., 2007). The remainder of the protein in the diet was supplied by fish meal. The SH hydrolysates were less soluble and contained a larger proportion of di- and tripeptides than CPSP. Kotzamanis et al. (2007) observed that the 10% incorporation of CPSP yielded the best growth and survival rates and intestinal development of the larvae as indicated by early induction of brushborder membrane enzyme activites. This group also appeared to have improved immunological status as indicated by lower levels of Vibrio spp. in the larvae. The poorest performance data were obtained with the diet containing 10% of SH. Both the molecular weight distribution and concentration of dietary peptides may influence growth performance and immunological development. The use of different protein sources, hydrolysis conditions, and dietary concentrations makes it very difficult to compare across studies and make claims regarding efficacy. The differences in substrate affinity of PepT1 for the different peptides, the effect of peptides on gene expression, as well as the 8,400 possible di- and tripeptides creates a challenge for the nutritionist in providing a peptide profile that accommodates the digestive enzyme and transporter physiology in the gut.
Although a variety of peptide-based products are currently on the market, there has been some debate over their efficacy due to the reasons already described. Technological advances in AA analysis of peptide profiles will improve the design of future feeding studies involving dietary protein replacement or supplementation with peptides. Promising results have been obtained in studies involving pigs. Dried porcine solubles is a peptide product consisting of partially hydrolyzed residues from porcine intestine (Nutra-Flo Company, Sioux City, IA). Growth performance was improved in weaned pigs that consumed diets containing dried porcine solubles (Zimmerman et al., 1997; Lindemann et al., 2000; DeRouchey et al., 2003). Additionally, an enzymatically digested protein product produced from a proprietary blend of swine blood, selected poultry tissues, and hydrolyzed feathers (Griffin Industries Inc., Cold Spring, KY) was observed to be similar in feeding value to menhaden fish meal, spray-dried animal blood cells, and spray-dried plasma protein in early-weaned pigs.
|
CONCLUSIONS AND IMPLICATIONS FOR FUTURE RESEARCH |
|---|
Although practical applications of peptides to livestock nutrition are lacking, we hope that the readers will exploit the information presented herein to stimulate further investigations, the results of which may enable improved nutritional management decisions. Providing a diet that best accommodates the profile of digestive enzymes and nutrient transporters present in the gut will result in improved utilization of dietary protein, reduced nitrogen excretion into the environment, improved health, and improved growth. The future of dietary peptides in livestock diets remains unclear, but we hope that this review has served the purpose of enlightening some readers to embrace the concept of peptide absorption and better understand the factors that contribute to overall animal performance and health. Future studies should continue to address the influence of dietary protein composition on peptide and AA transporter expression in the gut. Many of the studies reported to this point include mRNA abundance data, and it is unclear if protein expression and activity parallel changes in the level of the transcript. As we gain a better understanding of how AA assimilation may be improved through dietary peptides, we should strive for improved technology that allows for the accurate determination of peptide profiles to better fine-tune these protein sources to address the nutritional requirements of the animal.
1 Corresponding author: webbk{at}vt.edu
Received for publication December 21, 2007. Accepted for publication April 11, 2008.
|
LITERATURE CITED |
|---|
. Am. J. Physiol. Gastrointest. Liver Physiol. 291:G851–G856.This article has been cited by other articles:
|
|
 |
|
  R. Martineau, I. Ortigues-Marty, J. Vernet, and H. Lapierre Technical note: Correction of net portal absorption of nitrogen compounds for differences in methods: First step of a meta-analysis J Anim Sci, October 1, 2009; 87(10): 3300 - 3303. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Top
Abstract
INTRODUCTION