点击显示 收起
1 From the US Department of Agriculture, Agricultural Research Service, Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston (DGB, BS, XC, JBvG, and PJR), and the Department of Biochemistry, Wake Forest University School of Medicine, Winston-Salem, NC (HF and SMH).
2 Dedicated to the memory of Peter J Reeds, who died August 13, 2002. Peter J Reeds was a mentor and inspiration to DG Burrin, B Stoll, and JB van Goudoever. 3 The contents of this publication do not necessarily reflect the views or policies of the US Department of Agriculture, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government. 4 Supported by the US Department of Agriculture, Agricultural Research Service, under Cooperative Agreement no. 58-6250-6-001 and grant 98-35200-6067 (to SMH) and in part by National Institutes of Health grants HD33920 (to DGB) and DK34738 (to SMH). JBvG was supported by the Sophia Foundation for Scientific Research, the Nutricia Research Foundation, and the Royal Netherlands Academy of Science and Arts (Ter Meulen Fund). 5 Address reprint requests to DG Burrin, Children’s Nutrition Research Center, 1100 Bates Street, Houston, TX 77030. E-mail: dburrin{at}bcm.tmc.edu.
ABSTRACT
Background: Preterm infants often receive total parenteral nutrition (TPN) before enteral feeding. Although TPN has been linked to mucosal atrophy, its effects on intestinal digestion, absorption, and metabolism are unknown.
Objective: Our aim was to determine the effects of TPN on rates of intestinal nutrient absorption and metabolism in infant pigs after initiation of enteral feeding.
Design: Piglets were surgically implanted with catheters in the carotid artery, jugular vein, portal vein, and duodenum; an ultrasonic blood flow probe was inserted in the portal vein. Piglets were given TPN (TPN group) or enterally fed formula (enteral group) for 6 d. On day 7, both groups were enterally fed a milk-based formula, and the net portal absorption and metabolism of enteral [2H]glucose and [13C]leucine were measured.
Results: After enteral feeding began, portal blood flow increased by 27% and 41% above the basal rate in the enteral and TPN groups, respectively; oxygen consumption remained lower in the TPN group. During enteral feeding, the net portal absorption of glucose was lower in the TPN group and that of galactose was not significantly different between the groups; lactate release was higher in the TPN group. Portal absorption accounted for only 37% of galactose intake in both groups. The TPN group had lower net portal absorption of arginine, lysine, threonine, and glycine. The portal absorption of dietary leucine was not significantly different between the groups; the arterial utilization and oxidation of leucine were significantly lower in the TPN group.
Conclusion: Short-term TPN results in decreased lactose digestion and hexose absorption and increased intestinal utilization of key essential amino acids when enteral feeding is initiated in piglets.
Key Words: Total parenteral nutrition • preterm infants • amino acids • glucose • galactose
INTRODUCTION
Total parenteral nutrition (TPN) is used in the clinical management of preterm infants and other neonates who cannot tolerate full enteral feedings. Numerous studies have shown that the lack of enteral nutrition suppresses intestinal growth and leads to substantial mucosal atrophy in animals and to a lesser degree in humans (1–4). Studies conducted in formula-fed, preterm infants suggest that their lactose digestion is incomplete; as much as 20–25% of their dietary intake was fermented in the colon (5). The incomplete lactose digestion in preterm infants has been ascribed to insufficient lactase expression, and some have linked carbohydrate malabsorption and fermentation to the risk of necrotizing enterocolitis (6). Because most preterm infants receive some degree of TPN before enteral feeding is initiated, it is conceivable that TPN suppresses intestinal function and contributes to incomplete lactose digestion and hexose absorption when enteral feeding is initiated.
An important nutritional consideration in assessing digestive and absorptive function is the extent to which dietary nutrients are metabolized by the mucosal tissues in the process of absorption. The high rates of turnover and inherent metabolic activity of epithelial cells translate into a substantial nutrient requirement necessary to maintain mucosal tissue structure and function. Early studies by Windmueller and Spaeth, followed by our recent studies in piglets, indicate that the gut preferentially utilizes specific nutrients, particularly glutamate, glutamine, and threonine (7–9). Glutamate and glutamine are key oxidative fuels, whereas the high rate of intestinal threonine utilization is thought to be channeled into mucin glycoprotein synthesis. Glucose is also an important metabolic fuel for intestinal enterocytes (9, 10); however, the proportional intestinal utilization of dietary glucose is substantially lower than that of glutamate (15% compared with 95% of intake) in young pigs (7). However, the proportion of dietary glucose oxidized by the intestine is significantly increased when the dietary protein intake is reduced (11). With respect to essential amino acids, we found that the gut utilizes approximately half of the dietary lysine consumed and is a significant source of whole-body lysine oxidation (12). Thus, the quantity of dietary nutrients absorbed into the portal circulation is significantly influenced both by the metabolic rate and mass of intestinal mucosal tissue.
In the current study, we set out to address 2 questions. The first question was whether a period of chronic TPN reduces the intestinal digestive and absorptive functions in infant piglets. The second question was whether the intestinal mucosal atrophy that occurs during TPN would alter the systemic availability of dietary nutrients on initiation of enteral feeding.
MATERIALS AND METHODS
Animals
We used 3-wk-old female, crossbred piglets (Large White x Hampshire x Duroc) obtained from the Texas Department of Criminal Justice, Huntsville, TX. The piglets were received at the Children’s Nutrition Research Center in Houston at 10 d of age and then fed a liquid milk replacer formula (Litter Life; Merrick, Middleton, WI) at a rate of 50 g • kg body wt-1 • d-1 for 7 d. The composition (per kg dry matter) of the formula fed to the piglets was as follows: 527 g lactose, 100 g fat, and 250 g protein. The protocol was approved by the Animal Care and Use Committee of Baylor College of Medicine and was conducted in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals.
Study design
At 17 d of age, after being deprived of food, piglets were surgically implanted with catheters in the carotid artery (Tygon tubing, 1.77 mm outside diameter; Saint-Gobain Performance Plastics Corp, Akron, OH), portal vein (polyethylene tubing, 1.27 mm outside diameter), and jugular vein and duodenum (silicone tubing, 1.65 mm outside diameter) under general anesthesia, as described previously (7). In addition, the piglets were implanted with an ultrasonic flow probe (6S; Transonics, Ithaca, NY) via the portal vein. After surgery, the piglets were administered TPN for 24 h and then randomly assigned to either continue with TPN (TPN group; n = 7) or begin enteral feeding with a liquid milk replacer formula (enteral group; n = 8) for 6 d. The enteral group received the formula in 4 equal feedings between 0800 and 2000; feeding volumes offered were adapted gradually, such that full feedings were attained by the third postoperative day. The TPN group received a continuous intravenous infusion of an elemental diet containing free amino acids, dextrose, lipid, electrolytes, minerals, and vitamins, as described previously (13). The daily fluid and macronutrient intakes per kg body wt of both the enteral and TPN groups were 240 mL, 13 g protein, and 900 kJ • kg-1 • d-1. Piglets were weighed daily to maintain equal nutrient intakes in both groups.
Enteral feeding and tracer protocol
On the 7th postoperative day, all piglets underwent an enteral feeding and tracer protocol to quantify intestinal digestive and absorptive function. Piglets in the enteral group were deprived of food overnight, and intravenous nutrient administration was stopped 2 h before the protocol began in the TPN group. Piglets were given an oral feeding of formula (20 mL/kg) followed by an intraduodenal infusion of formula (10 mL • kg-1 • h-1) for 6 h. The formula contained isotopes of glucose (6,6-[2H]glucose, 99% D2; Cambridge Isotopes, Woburn, MA) and leucine (1-[13C]leucine, 98% 13C; Cambridge Isotopes) such that the enteral tracer infusion rates were 210 µmol • kg-1 • h-1 for [2H]glucose and 50 µmol • kg-1 • h-1 for [13C]leucine. Five piglets in each treatment group also received a primed, continuous, intravenous infusion of [2H]leucine (45 µmol • kg-1 • h-1; 98% 5,5,5-D3; Cambridge Isotopes) for 6 h.
Arterial and portal blood samples (3 mL) were taken before and 4, 5, and 6 h after the enteral feeding and tracer protocol began. Portal blood flow was monitored continuously via transit-time ultrasound for 30 min before the initial formula feeding and throughout the entire 6-h protocol. Blood gases (Chiron Diagnostics, Halstead, Essex, United Kingdom), glucose, galactose, and lactate (YSI analyzer; Yellow Springs Instrument Co, Yellow Springs, OH) were measured immediately in all samples. An aliquot (1.0 mL) of whole blood was immediately added to an evacuated tube and used for analysis of 13CO2 as described previously (12). An aliquot (0.2 mL) of whole blood was mixed with an equal volume of aqueous internal standard (0.5 mol methionine sulfone/L) and stored at -70 °C for later amino acid analysis. A whole-blood aliquot (0.2 mL) was also mixed with deionized water and stored at -70 °C for later isotopic tracer analysis of [13C]leucine, [2H]leucine, and [2H]glucose. The remaining whole blood was centrifuged at 2500 x g and 4 °C for 10 min, and 0.2 mL plasma was stored at -70 °C for later analysis of ammonia.
At the end of the 6-h protocol, piglets were euthanized with an intravenous injection of pentobarbital sodium (50 mg/kg body wt) and sodium phenytoin (5 mg/kg body wt, Beutanasia-D; Schering-Plough Animal Health, Kenilworth, NJ). The abdomen was opened and the entire small intestine distal to the ligament of Treitz was removed and immediately flushed with ice-cold saline. After being flushed, the small intestine was divided into 2 equal portions; the proximal half was designated the jejunum, and the distal half was designated the ileum. These segments also were divided in half. The 4 small intestinal segments were weighed, a section of each was fixed in 10% buffered formalin for subsequent histology, and a segment (20 cm) of the remaining tissue was snap-frozen in liquid nitrogen and stored at -70 °C until analyzed for protein and DNA concentrations and for activities of lactase, branched-chain amino acid aminotransferase (BCAT), and branched-chain -keto acid dehydrogenase (BCKD).
Sample analysis
Plasma ammonia was measured spectrophotometrically with the use of an assay based on the glutamate dehydrogenase reaction according to the manufacturer’s protocol (Raichem/Sigma Diagnostics, St Louis). Whole-blood samples were prepared for amino acid analysis and mass spectrometry as described previously (7). The amino acid concentrations in whole blood were determined by reversed-phase HPLC of their phenylisothiocyanate derivatives (PicoTag; Waters, Woburn, MA). Gas chromatography–mass spectrometry was performed with the pentaacetate derivative of glucose and the heptafluorobutyramide derivative of leucine. The analyses were performed with a 5890 series II gas chromatograph linked to a model 5989B quadrupole mass spectrometer (Hewlett-Packard, Palo Alto, CA). We used methane negative chemical ionization for leucine and electron impact ionization for glucose. The isotopic enrichment of glucose was determined by monitoring ions at a mass-to-charge ratio of 242–244, whereas for leucine, the ions monitored had a mass-to-charge ratio of 349–352. The isotopic enrichment of 13CO2 was then determined on a continuous-flow gas isotope ratio machine (ANCA; Europa Instruments, Crewe, United Kingdom). The between-sample SD was 0.001 atoms percent excess.
Intestinal tissue samples (100–200 mg) were homogenized in water, and aliquots were removed for the analysis of protein, DNA, and lactase activity as described previously (14). Morphometric analysis of formalin-fixed intestinal tissue was done on paraffin sections stained with eosin and hematoxylin. Villus height, crypt depth, and muscularis thickness were measured by using an Axiophot microscope (Carl Zeiss Inc, Göttingen, Germany) and NIH Image software version 1.60 (National Institutes of Health, Bethesda, MD) in 10–15 vertically well-oriented crypt villus units.
Procedures for the quantitative extraction of BCAT from tissue samples were described previously (15). Frozen tissue was pulverized with a mortar and pestle cooled in liquid nitrogen. The powder (200–300 mg) was transferred to cooled, preweighed centrifuge tubes. The tissue was suspended in buffer (1 g tissue/4 mL extraction buffer) containing 25 mmol HEPES/L (pH 7.4), 0.4% CHAPS (by wt) protease inhibitors (16), 20 mmol EDTA/L, 20 mmol EGTA/L, and 5 mmol dithiothreitol/L. The tissue suspension was then subjected to 2 rounds of freeze-thaw sonication before centrifugation at 10 000 x g and 4 °C for 30 min. The BCAT isoenzymes BCATm and BCATc were extracted quantitatively with the use of this procedure, and the extract was kept frozen at -80 °C. The supernatant fluid was assayed for BCAT activity as described previously with -keto[1-14C]isovalerate and isoleucine (17). A unit of activity was defined as 1 µmol valine formed/min at 37 °C. BCKD activity from tissues was determined essentially as described previously (18, 19), with the following modifications. The frozen tissue was ground to a fine powder with a mortar and pestle on dry ice, and 200–300 mg tissue was homogenized with a Tissuemizer (IKA Works, Inc, Wilmington, NC) at a setting of 20 in 8 volumes of the 3% Triton X 100 extraction buffer (20). Connective tissue was removed during the tissue homogenization step. Insoluble material was removed by centrifugation at 20 000 x g for 5 min at 4 °C. To quantitatively precipitate all BCKD activity, polyethylene glycol (9% final concentration) was added to the tissue extract, and the solution was centrifuged at 12 000 x g and 4 °C for 20 min. BCKD activity was then measured as 14CO2 release from -keto[1-14C]isocaproate. Total BCKD complex activity, which is an estimate of the amount of enzyme, was measured after activation of a separate aliquot of the same sample in the presence of a final concentration of 2 mmol MnCl2/L and lambda protein phosphatase (final concentration: 2000 U/mL) purchased from New England Biolabs, Inc (Beverly, MA), as described previously (20). The activity state of BCKD is the ratio of actual activity before activation to total activity obtained after activation by phosphatase treatment. A unit of activity was defined as 1 µmol 14CO2 formed/min at 37 °C.
Calculations
Whole-body glucose and leucine fluxes were calculated from the following equations. Whole-body flux (Q) in µmol • kg-1 • h-1 was calculated as follows:
RESULTS
Intestinal mass, morphology, and enzyme activities
Consistent with previous studies, we found no differences in either weight gain or final body weight between the TPN and enteral groups during the 6-d treatment period; mean (± SEM) weight gains were 216 ± 25 and 247 ± 16 g/d in the TPN and enteral groups, respectively. At the start of the enteral feeding and tracer-infusion protocol, the body weights of the TPN and enteral groups were 6.17 ± 0.52 and 6.22 ± 0.42 kg, respectively. Even after 6 h of enteral feeding, the small intestinal weight and protein and DNA contents were significantly lower in the TPN group than in the enteral group (Table 1). Villus height and villus area were significantly lower in the TPN group than in the enteral group (Table 1). The changes in intestinal protein content translated into significantly lower total intestinal lactase activity in the TPN group. However, it is noteworthy that the decreased total intestinal activity of lactase in the TPN group was also a function of a decreased mean specific activity of lactase (Table 1).
We measured the activities of 2 key leucine catabolic enzymes—BCAT and BCKD—to assess the biochemical basis for changes in leucine oxidation (Table 1). The first step in the catabolism of a branched-chain amino acid is the reversible transamination to the corresponding -keto acid; for leucine, it is -ketoisocaproate, by BCAT, which is present in both the mitochondria and cytosol (15). The second step in branched-chain amino acid catabolism is the irreversible oxidative decarboxylation of the branched-chain keto acid by the mitochondrial BCKD complex (15). We found that the activities of both BCAT (-21%) and actual BCKD complex (-40%) in the proximal jejunum were lower in the TPN group than in the enteral group; activities of these 2 enzymes in the proximal ileum were not different (data not shown). In addition, even after 6 h of enteral feeding, the proportion of the BCKD measured in the active form was relatively low: 29% and 22% in the enteral and TPN groups, respectively.
Intestinal blood flow and respiratory activity
The rate of blood flow is an underlying physiologic response that increases with feeding and significantly influences intestinal nutrient absorption. Therefore, we measured portal blood flow in the period both before (ie, basal) and during the enteral feeding protocol to assess the hemodynamic response to feeding (Table 2). On the basis of two-factor analysis of variance for portal blood flow, there was no treatment effect (ie, enteral compared with parenteral); however, there was a significant (P < 0.05) main effect of feeding state. Thus, portal blood flow was not significantly different between the enteral and parenteral treatments in either the basal or enteral feeding states. However, compared with the basal state, enteral feeding significantly increased portal blood flow by 27% and 41% in the enteral and TPN groups, respectively. This finding suggests that the normal postprandial hyperemic response was not suppressed by TPN. Despite the lack of significant differences in rates of portal blood flow, the rate of oxygen uptake (P = 0.05) and the oxygen extraction ratio (P = 0.032) were significantly lower in the TPN group than in the enteral group; carbon dioxide production was lower in the TPN group than in the enteral group, but not significantly so (P = 0.099). The respiratory quotient was not significantly different between the 2 treatment groups.
View this table:
TABLE 2 . Portal blood flow, oxygen uptake, and carbon dioxide production in infant pigs fed either enterally or with total parenteral nutrition (TPN) for 6 d and then refed enterally for 6 h1
Portal hexose and lactate balance
To assess the extent of lactose digestion in this study, we quantified the net rates of portal glucose and galactose absorption. As shown in Table 3, the enteral intake of lactose (molar intake of glucose and galactose) and the arterial and portal concentrations of glucose and galactose were not significantly different between groups. Consistent with our previous results in formula-fed piglets (7), the net portal glucose balance in the enteral group was 74% of the glucose intake. However, the net portal balance of galactose in the enteral group was only 38% of the intake and was significantly lower (50%) than that of glucose. In the TPN group, the net portal balance of glucose was significantly lower than that in the enteral group (26% compared with 74%), whereas portal galactose balance (36%) was not significantly different between the treatment groups. In contrast, the arterial lactate concentration and net portal lactate balance were significantly greater in the TPN group than in the enteral group (Table 3).
View this table:
TABLE 3 . Plasma concentrations and net portal balances of glucose, galactose, and lactate in infant pigs fed either enterally or with total parenteral nutrition (TPN) for 6 d and then refed enterally for 6 h1
Whole-body and portal glucose kinetics
The whole-body and PDV glucose kinetics are shown in Table 4. No significant differences in whole-body glucose flux, arterial glucose isotopic enrichment, or portal glucose isotopic enrichment were observed. The arterial glucose isotopic enrichment in both groups also was not significantly different between 4 and 6 h of tracer infusion, indicative of steady state (data not shown). The net portal mass balance of glucose, expressed as a proportion of the enteral lactose input, represents the combined rate of lactose hydrolysis to glucose and galactose, the rate of glucose transport across the epithelial cell, and the rate of mucosal glucose metabolism. Thus, to distinguish between these kinetic processes, we measured the portal absorption of [2H]glucose to quantify the extent of intestinal transport and mucosal metabolism of free glucose. We found that the rate of glucose absorption was significantly lower in the TPN group than in the enteral group: 52.5% compared with 98.5%, respectively. This difference in absorption translated into a higher rate of glucose utilization in the TPN group than in the enteral group. However, it is important to note that this higher rate of glucose utilization also includes any defect in transport as well as glycolytic and oxidative metabolism. Moreover, in both the enteral and TPN groups, when expressed as a proportion of the enteral intake, the rates of [2H]glucose absorption were higher than the net portal glucose mass balance, which reflects the simultaneous uptake of arterial glucose from the systemic circulation, even during enteral feeding. The arterial glucose utilization by the PDV was significantly lower in the TPN group than in the enteral group (Table 4). We estimated the arterial glucose utilization rate by using the fractional extraction estimates from previous studies with similar-age piglets that were either enterally (6.5%) (7) or TPN fed (2.5%) (21). This proportional difference (ie, 6.5% compared with 2.5%) in fractional arterial extraction between the TPN and enteral groups was not significantly different from what we found for [2H]leucine in this study (15% compared with 6%, respectively) for enterally and TPN-fed piglets.
View this table:
TABLE 4 . Rates of whole-body glucose flux and portal [2H]glucose absorption and utilization kinetics in infant pigs fed either enterally or with total parenteral nutrition (TPN) for 6 d and then refed enterally for 6 h1
The kinetics of PDV [2H]glucose metabolism and net portal glucose balance were used to estimate the extent of dietary lactose hydrolysis (Figure 1). To estimate the rate of free glucose input to the luminal epithelial surface, we multiplied the net portal glucose balance in mmol • kg-1 • h-1 (Table 3) by the fractional absorption of the free [2H]glucose tracer (Table 4). The efficiency of lactose hydrolysis was then calculated by dividing the rate of free glucose input on the luminal side by the molar rate of enteral lactose intake (Table 3). On the basis of these calculations, we found that only 76% of the dietary lactose intake was hydrolyzed in the enterally fed piglets, which indicated incomplete digestion. However, in the TPN group, the rate of lactose hydrolysis was significantly lower, such that 49% of the dietary lactose intake was hydrolyzed to glucose and galactose. On the basis of the assumption that 1 mol lactose is hydrolyzed to 1 mol each of glucose and galactose, we also estimated the rate of free galactose input to the luminal epithelial surface by assuming that it was equal to that of free glucose. This calculation allowed us to estimate the fractional absorption of free galactose by the mucosa, which was 49% and 76% in the enteral and TPN groups, respectively.
FIGURE 1. . Schematic diagram showing a model of the kinetics of lactose hydrolysis and hexose absorption in the gut mucosa. Numerical values in shaded boxes indicate the percentage of lactose hydrolysis (A) and the percentage absorption of glucose (B) and galactose (C) from the lumen into the blood in piglets fed enterally (enteral group) or with total parenteral nutrition (TPN group; values in parentheses). All other numerical values represent rates of flux (mmol • kg-1 • h-1) of lactose, glucose, galactose, and lactate in the enteral and TPN (values in parentheses) groups. SGLT-1, sodium-dependent glucose transporter 1; GLUT-2, glucose transporter 2.
Net portal amino acid balances
The net portal balances of amino acids and ammonia are shown in Table 5. The net portal balance of ammonia and several amino acids were significantly lower in the TPN group than in the enteral group, particularly lysine, threonine, arginine, glycine, and glutamine. The portal balance of many other amino acids also tended (P < 0.10) to be lower (-20% to -40%) in the TPN group than in the enteral group, but not significantly so. Note that, because glutamine balance was negative, the lower value in the TPN group than in the enteral group indicates a reduced arterial extraction of glutamine by the PDV tissues. In the TPN group, the glutamine balance was not significantly different from zero, which indicated that there was no net uptake or release of glutamine from the PDV tissues. To assess the extent of protein digestion and amino acid absorption and metabolism by gut tissues, we calculated the net portal balance of essential amino acids as a proportion of the dietary intake (Figure 2). With the notable exception of leucine, the net portal balances of all the essential amino acids measured were numerically lower in the TPN group than in the enteral group; however, these differences were statistically significant (P < 0.05) only for lysine, arginine, threonine, and isoleucine.
View this table:
TABLE 5 . Net portal ammonia and amino acid balance in infant pigs fed either enterally or with total parenteral nutrition (TPN) for 6 d and then refed enterally for 6 h1
FIGURE 2. . Mean (± SEM) net portal balance of essential amino acids in piglets fed enterally (enteral group; n = 7) or with total parenteral nutrition (TPN group; n = 7). *Significantly different from the enteral group, P 0.05 (ANOVA).
Whole-body and portal-drained viscera leucine metabolism
The rates of enteral and arterial isotopically labeled leucine kinetics and metabolism in the PDV and whole body are shown in Table 6. With respect to metabolism of the enteral [13C]leucine tracer, we found no difference in either the absorption or utilization by the PDV tissues. Expressed relative to the enteral intake, the rates of leucine absorption were high in both groups: 95% and 84% in the enteral and TPN groups, respectively. Interestingly, despite the observation that first-pass utilization of leucine was numerically greater in the TPN group than in the enteral group, the rate of [13C]leucine oxidation was significantly lower in the TPN group. The most remarkable aspect of PDV leucine metabolism was the observation that the rate of arterial [2H]leucine utilization in the enterally fed piglets was > 2-fold that in the TPN group. The sum total of first-pass and arterial leucine utilization is an estimate of the rate of protein synthesis in PDV tissues. Despite the significant reduction in arterial leucine use in the TPN-fed piglets, the mean estimates of total PDV protein synthesis were not significantly different between the 2 groups. The calculated estimate of leucine recycling represents combined rates of intracellular and extracellular (ie, luminal proteases) proteolysis within the PDV. The rate of leucine recycling was markedly (-60%) lower in the TPN group than in the enteral group, although this difference was not statistically significant (P = 0.08) because of the large error of the estimate.
View this table:
TABLE 6 . Rates of enteral and arterial leucine kinetic metabolism in the portal-drained viscera (PDV) tissues and in the whole body in infant pigs fed either enterally or with total parenteral nutrition (TPN) for 6 d and then refed enterally for 6 h1
DISCUSSION
Our results showed that piglets that received TPN for 1 wk sustained a marked reduction in the net absorption of both glucose and galactose when they were refed enterally with a lactose-containing formula. These findings show that TPN results in functional defects in lactose digestion and hexose transport in infant piglets. Moreover, they also suggest that substantial amounts of galactose were metabolized by the intestinal mucosa of enterally fed piglets. Although TPN significantly suppressed the net absorption of several key essential amino acids—namely lysine, threonine, isoleucine, and arginine—in general, the adverse effects on protein digestion and essential amino acid absorption were less significant than were those related to carbohydrate digestion.
The finding of reduced net absorption of glucose and galactose in the TPN-fed piglets has several possible explanations. The first is that the total intestinal lactase activity was insufficient to handle the dietary lactose intake. This idea is supported by our observation that the total intestinal lactase activity in the TPN group was reduced to only one-fourth of that in the enteral group and was probably secondary to reduced villus height and surface area. We previously showed that these changes in mucosal structure and cellularity in response to TPN are mediated by the suppression of lactase-phlorizin hydrolase synthesis, total protein synthesis, and crypt cell proliferation and stimulation of villus enterocyte apoptosis (13, 22–24). However, the reduced net absorption of glucose and galactose could also be explained by defects in either mucosal transport across the epithelial cell or increased epithelial metabolism. With respect to glucose transport, studies in animals and humans suggest that TPN reduces the intestinal hexose transport capacity, specifically the brush-border glucose transport activity (25–27). The fact that fractional portal absorption of the free [2H]glucose was significantly lower in the TPN group (53%) than in the enteral group (98%) indicates that there was either a defect in glucose transport or an increase in cellular glucose metabolism. The latter possibility is supported by the fact that, although we did not quantify the metabolic fate of [2H]glucose, we did find that the net portal lactate release in the TPN group was > 2-fold that in the enteral group. Moreover, this increase in net portal lactate release was paralleled by an increase in the circulating lactate concentration. The [2H]glucose tracer and hexose mass balance data were used to determine the amount of free glucose liberated on the luminal surface and suggest that only 49% of the dietary lactose intake was hydrolyzed. It is also notable that only 76% of the dietary lactose intake was hydrolyzed in the enterally fed piglets. Therefore, our results provide strong evidence that the mucosal atrophy resulting from TPN leads to functional defects in intestinal lactase digestion and possibly also in glucose transport.
Another significant finding from our study was that the net portal galactose absorption rate was substantially lower than that of glucose, especially in the enterally fed group. The differential absorption of galactose and glucose in the enterally fed group could be ascribed to differential transport of glucose and galactose across the mucosal epithelium, yet this seems unlikely because both hexoses are transported via sodium-dependent glucose transporter 1 and glucose transporter 2 (28, 29). Although it is generally held that the liver is the primary site of dietary galactose clearance, there is considerable evidence that the infant intestine also metabolizes galactose. In vivo studies with infant lambs show that galactose clearance occurs in the PDV tissues (30). Studies in rodents indicate that galactokinase activity is high at birth and declines during the suckling period, whereas galactose-1-phosphate uridyltransferase and uridine diphosphate galactose-4-epimerase remain high during this period (31–33). Moreover, a study involving isolated porcine enterocytes indicates that both galactokinase activity and galactose oxidation to carbon dioxide are highest at birth and decline in the first 2 wk of life, which points to galactokinase as a rate-limiting factor (34). These results suggest that the intestine is a metabolically significant site of dietary galactose utilization.
Our study was designed to test whether, besides its effect on carbohydrate digestion, TPN also affects the digestion of protein and the intestinal absorption and metabolism of amino acids. Our previous studies showed that the gastrointestinal utilization of essential amino acids has a substantial effect on their systemic availability (8, 12, 35). We predicted that if protein digestion was limiting, we should find a uniform reduction in the net portal absorption of all essential amino acids. The net portal appearance of 4 of the 9 essential amino acids measured was reduced by 22–67% in TPN-fed compared with enterally fed piglets. However, only the net absorption of lysine, arginine, isoleucine, and threonine was significantly different (P < 0.05) between the TPN and enteral groups. We suggest that these differences in net absorption among the essential amino acids indicate that the rate of dietary protein digestion was not limiting in the TPN group. More direct evidence that digestion, per se, was not limiting is found in the observation that the net portal absorption of both free [13C]leucine and total leucine, including that in the form of intact protein, was not significantly different between the TPN and enteral groups. A possible explanation for the relatively low net portal absorption of lysine, arginine, and threonine in the TPN group is their high rate of first-pass mucosal metabolism. Arginine is a precursor for the synthesis of nitric oxide and polyamines, and threonine is a precursor for the biosynthesis of mucin glycoproteins (8, 36, 37).
Another metabolic fate of amino acids utilized by the gut is oxidation to carbon dioxide. Studies indicate that intestinal oxidation affects not only nonessential amino acids—mainly glutamate, glutamine, and aspartate—but also essential amino acids, including leucine, lysine, and phenylalanine (11, 12, 38, 39). Many observations from this study suggest that TPN suppressed intestinal amino acid oxidation. First, [13C]leucine oxidation to 13CO2 was 50% lower in the TPN group than in the enteral group. Second, although previous in vivo studies have shown that leucine oxidation occurs in gut tissues, the current study was the first to measure the activities of the leucine catabolic enzymes BCAT and BCKD in pig intestine and to show that TPN reduces their activity. In addition, the net portal release of ammonia was significantly lower (by 32%) in the TPN group than in the enteral group, a finding that is consistent with the lower amino acid oxidation in the TPN group.
Contrary to our initial hypothesis, the results showed that TPN did not result in an increased systemic availability of amino acids on enteral refeeding. This was evident from the net portal amino acid balances and also from the kinetics of portal leucine metabolism. On the basis of leucine kinetics, the rates of absorption and first-pass utilization of enteral leucine were not significantly different between the TPN and the enteral groups. However, the most striking treatment difference in portal leucine metabolism was the > 2-fold lower rate of arterial leucine utilization in the TPN group than in the enteral group. This observation is consistent with our original hypothesis that reduced intestinal protein mass would decrease amino acid use in TPN-fed piglets. Moreover, the finding concurs with the idea that there is preferential utilization of arterial rather than luminal amino acids for intestinal protein synthesis and that TPN suppresses the fractional rate of protein synthesis (22, 35, 40).
In summary, our results show that TPN reduces both lactose digestion and hexose absorption in infant piglets. Furthermore, the results suggest that the mechanisms underlying the reduction in hexose absorption are probably the diminished transport of glucose and the increased metabolism of glucose to lactate. Moreover, the finding of a low rate of galactose absorption in the enterally fed piglets suggests extensive mucosal metabolism of galactose and warrants further in vivo kinetic investigation. Despite the demonstrable effect of TPN on intestinal carbohydrate metabolism, we found that TPN had only a limited negative effect on intestinal protein digestion and amino acid absorption. Our findings may be relevant to the clinical care and nutritional support of neonates. As with our past results in this animal model, we draw clinical inferences with some trepidation because of the limitations associated with comparisons between piglets and humans. Our hesitation is underscored by the fact that our piglets were neither premature nor deprived of maternal feeding before the study began. Nevertheless, we posit that the current findings may be relevant to the observation of lactose malabsorption in preterm infants and the risk of necrotizing enterocolitis. It has become increasingly evident that the capacity to digest lactose and absorb glucose increases with the gestational age of neonates (6, 41, 42). This immaturity in lactose digestion may lead to malabsorption and increased colonic fermentation, the latter of which is a risk factor for the development of necrotizing enterocolitis. Although the extent of lactose malabsorption in preterm infants is presumed to be largely a function of gestational age, our results suggest that this phenomenon may also be a consequence of the previous duration of TPN.
ACKNOWLEDGMENTS
We thank Leslie Loddeke for editorial review and Jane Schoppe for assistance with manuscript preparation.
DGB, BS, XC, JBvG, and PJR participated in all aspects of the study design, data collection and analysis, and manuscript preparation. HF and SMH were involved with the laboratory analysis and manuscript preparation. None of the authors had any financial or personal interest, including advisory board affiliations, in any company or organization sponsoring the research.
REFERENCES