In humans and pigs, hydrolysis of dietary polyglutamyl folates is carried out by intestinal brush border folate hydrolase [glutamate carboxypeptidase II (GCPII)], whereas the transport of the monoglutamyl folate derivatives occurs via the intestinal brush border reduced folate carrier (RFC). The study objective was to measure the expression of intestinal GCPII and RFC during postnatal development of pigs and their effects on plasma and liver folate concentrations. Duodenum, jejunum, ileum, liver, and plasma samples were collected from female Yorkshire pigs at birth, 24 h, 1 wk, 3 wk, and 6 mo (n = 6 at each time point). GCPII mRNA transcripts and protein (normalized using β-actin), and enzyme activity (normalized per mg mucosal protein) were highest in all segments of small intestine at birth and were undetectable in ileum after 1 wk, whereas jejunal protein and activity predominated at 6 mo. RFC mRNA transcripts were present in all segments of small intestine at birth and declined significantly throughout development to 6 mo. Conversely, RFC protein increased twofold during the first 24 h and remained constant throughout development in all segments of small intestine. Liver RFC mRNA transcripts were detected at birth but were reduced by 6 mo. Liver folate concentration increased throughout postnatal development, whereas plasma folate levels increased during the first 24 h but decreased over time, reflecting the pattern of RFC expression in small intestine. These findings show that intestinal GCPII and intestinal and hepatic RFC all exhibit ontogenic changes in the pig that are reflected in postnatal folate status.
- glutamate carboxypeptidase II
- reduced folate carrier
folate is an essential water-soluble vitamin required for normal growth and development. Folate serves as a coenzyme in many important reactions including DNA synthesis, amino acid metabolism, and the synthesis of S-adenosylmethionine. Therefore, the requirement for folate is greatest during periods of rapid growth and cell division, such as gestation and postnatal development.
Mammals cannot synthesize folate and must acquire the vitamin from dietary sources. Dietary folates are a mixture of monoglutamyl and polyglutamyl forms that are hydrolyzed to the monoglutamyl form prior to transport across the jejunal brush border membrane. In humans and pigs, hydrolysis of dietary polyglutamyl folates is carried out by intestinal glutamate carboxypeptidase II (GCPII), whereas the transport of the monoglutamyl folate derivatives occurs in part via the intestinal reduced folate carrier (RFC) (5, 6, 10, 32). Each protein is anchored in the jejunal brush border membrane, GCPII by a single amino terminal transmembrane domain and RFC by 12 transmembrane domains (16, 22).
Folate in milk is largely bound to species-specific folate binding protein (FBP) that strongly enhances the uptake of folate in the newborn intestine via an endocytotic process (17). This protein-bound folate is resistant to gastric digestion (28) and is less available to microorganisms residing in the intestinal lumen (8). In contrast to adult human, rat, and pig, milk folate bound to FBP is most avidly absorbed in the ileum compared with the jejunum of suckling rat small intestine (17). This finding parallels the well-documented ileal endocytotic absorption of other macromolecules in the neonatal rat, a process that is markedly reduced at the time of weaning (11, 12, 20).
In mammals, the newborn small intestine undergoes dramatic changes from the time of birth to adulthood, including changes in mechanisms and capacity for nutrient digestion and absorption. Although these changes occur throughout the entire developmental period, there are two abrupt transitions that are of prime importance (29). At birth, mammals transition from obtaining all nutrients from the maternal blood supply to deriving all nutrients orally from mother's milk. The second major shift in intestinal development is at the time of weaning, when the animal begins to consume solid food with dramatically different nutrient composition compared with milk. Many intestinal brush border enzymes and transporters exhibit large shifts in expression and activity during postnatal development. For example, lactase expression and activity in the rodent are high at birth and decrease as the animal approaches weaning, whereas sucrase expression and activity are relatively low at birth and increase dramatically at the time of weaning (1, 11, 12). Concomitantly, intestinal glucose transporters are expressed highly at birth whereas the expression of fructose transporters increases during the time of weaning, a pattern that parallels the presence of lactose and sucrose degradation products, respectively (1, 3, 11, 12, 29).
Currently, little is known about the postnatal ontogeny of GCPII or RFC in any species. Balamurugan and Said (2) reported higher RFC activity, protein, and transcript expression in suckling rats compared with weanling or adult animals, whereas no data are available on the developmental patterns of expression of GCPII in any species. The objective of this study was to identify the ontogenic patterns of expression and activity of GCPII in pig intestine and the expression of RFC in pig intestine and liver during postnatal development and to determine the effects of these changes on plasma and liver folate concentrations. We chose the pig as a model to study the ontogeny of folate hydrolysis and transport based on previous studies in our laboratory that showed jejunal brush border membrane expression of GCPII in human and pig but not in rat or monkey (34) and confirmed that RFC is expressed in pig jejunal brush border membranes and liver (32).
RNALater solution was purchased from Ambion. Complete Protease Inhibitor tablets were purchased from Roche Applied Science. [14C]PteGlu3 was a gift from Dr. Carlos Krumdieck (University of Alabama, Birmingham). Bio-Rad Protein Assay was obtained from Bio-Rad Laboratories. Trizol Reagent and the First Strand Synthesis Kit were purchased from Invitrogen. All real-time PCR reagents, including Amplitaq Gold, PCR buffer, SYBRgreen, and the Primer Express Software for primer design, were obtained from Applied Biosystems (Foster City, CA). Primers were synthesized by the Molecular Structure Facility at the University of California, Davis. All other chemicals were obtained from various commercial sources.
Animals and tissue collection.
All procedures in this study were reviewed and approved by the Animal Welfare Committee of the University of California, Davis. Nonfasted female Yorkshire market pigs, raised for commercial purposes, were obtained from the University of California, Davis Swine Center at various ages during postnatal development: 0 h (time of birth, prior to colostrum consumption), 24 h, 1 wk, 3 wk (after first chow meal on day of weaning), and 6 mo (n = 6 at each time point). Piglets were exclusively suckled from birth up to 3 wk of age and those that were 3 wk or older were fed standard pig chow containing adequate amounts of all nutrients. Animals ages 0 h to 3 wk were euthanized by cardiac injection with Euthanol 6. Six-month-old animals were euthanized by electric shock and exsanguination. Immediately following euthanasia, liver and small intestine were removed from each animal and whole blood samples were collected in EDTA tubes. Segments of the small intestine were identified, excised, and immediately placed in ice-cold saline. Samples of 12–15 cm each were collected from duodenum (between the pyloric sphincter and the ligament of Trietz), jejunum (immediately distal to the ligament of Trietz), and ileum (immediately proximal to the ileocecal junction). Segments were sectioned longitudinally, placed on a glass plate over ice, and mucosa was collected by scraping with glass slides. Similarly, interior portions of liver (2–3 g) were excised and immediately placed in ice-cold saline. All mucosal and liver samples were then wrapped in foil and immediately frozen in dry ice. In addition, 20 mg of liver and mucosa from each intestinal segment were placed in RNALater solution to stabilize RNA for future isolation. Blood samples were centrifuged for 20 min to separate plasma and red blood cells. All tissue samples were maintained at −80°C until analysis.
Frozen intestinal mucosal samples were thawed on ice in Tris-mannitol buffer containing protease inhibitor cocktail and homogenized with a Polytron homogenizer (Brinkman Instruments, Westbury, NY). Brush border membranes were isolated from mucosal scrapings by the calcium precipitation method (26), and purity was determined by measuring the enrichment of alkaline phosphatase activity. In all samples, brush border membrane fractions showed 6- to 10-fold higher alkaline phosphatase activity compared with total homogenate (data not shown). The final membrane pellets were resuspended in 150 μl of homogenization buffer containing protease inhibitor cocktail.
Liver and plasma folate concentrations were analyzed by Dr. Tsunenobu Tamura (University of Alabama, Birmingham) using the Lactobacillus casei microbiological assay. Frozen liver tissue was homogenized in HEPES-CHES buffer by using a previously published protocol (36).
GCPII enzyme activity.
Intestinal folate hydrolase activity was measured in duodenal, jejunal, and ileal brush border membranes by the method of Krumdieck and Baugh (14) using 90 μM [14C]PteGlu3 as substrate. The 14C radiolabel is located on the terminal glutamate residue of the polyglutamyl folate. This configuration allows the brush border exopeptidase activity of intestinal GCPII to liberate the labeled glutamate residue. The 3,3-dimethylglutarate buffer was adjusted to pH 6.5, and para-hydroxymercuribenzoate (pHMB, 0.17 mM) was added to the reaction mixture to inhibit intracellular folate hydrolase activity. pHMB was shown in prior studies to completely inhibit intracellular folate hydrolase activity whereas brush border membrane folate hydrolase activity is unaffected (25, 34, 35). Samples were incubated with radiolabeled substrate for 40 min at 37°C before the reaction was terminated with 10% trichloroacetic acid, and uncleaved substrate was precipitated by adding 2% charcoal in 0.1 M acetic acid. After centrifugation, supernatants containing radiolabeled product were collected and measured with a liquid scintillation counter. Protein concentration was measured by spectrophotometry using Bio-Rad reagent at 595 nm. Folate hydrolase specific activity was expressed as picomoles cleaved substrate per milligram mucosal protein per minute.
Western blot analysis.
Brush border membrane proteins (15 μg for GCPII and 50 μg for RFC) were resolved on 10% Tris-glycine SDS-polyacrylamide gels (Bio-Rad) and were transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA). Immunodetection was performed by using J591 antibody against brush border folate hydrolase (GCPII) (a gift from Dr. N. Bander, Weill Medical College of Cornell University) and reduced folate transporter-1 (RT1) antibody against RFC (gift of Dr. Sylvia Smith, Medical College of Georgia).
J591 is a murine monoclonal antibody designed against the human extracellular domain of GCPII (15). Membranes were incubated with J591 antibody (1:2,000) for 1 h followed by a 1 h incubation with goat anti-mouse horseradish peroxidase-conjugated secondary antibody (1:30,000) (Pierce, Rockford, IL). h-RFT1 is a polyclonal antibody raised in rabbits against human peptide sequence RPKRSLFFNRDDRGRC (4). Membranes were incubated with h-RFT1 antibody (1:1,000) for 2 h followed by a 1-h incubation with goat anti-rabbit horseradish peroxidase conjugated secondary antibody (1:3,000) (Pierce, Rockford, IL). Data were normalized to β-actin, detected by use of a multispecies β-actin primary antibody (1:10,000) and anti-mouse horseradish peroxidase-conjugated secondary antibody (1:10,000) (Sigma). Bands were detected via the SuperSignal West Dura Extended Duration Substrate (Pierce) and a charge-coupled device camera (NucleoTech, Hayward, CA) by using chemiluminescent detection.
Total RNA from all tissue samples was isolated using the Trizol reagent method (7). cDNA was prepared by reverse transcription using Invitrogen First Strand Synthesis kit. Pig-specific primers for real-time PCR for GCPII and RFC were designed by using Primer Express software. Each PCR reaction was performed in triplicate by use of the ABI Prism 7900 sequence detection system (Applied Biosystems). The GCPII primers included forward (GTGAAGTCCTATCCAGATGGTTGG) and reverse (GCTGCTCCACTCTGAGGGTC) sequences, and the RFC primers included forward (TCGCGTCCTCTCTTTCTCAAG) and reverse (GTGTTGACTCCGAAGACCAGG) sequences. Separate standard-curves for each gene were generated using purified transcript fragments (method previously described; Ref. 33). Both genes were normalized to β-actin by using the forward (TCGATCATGAAGTGCGACGT) and reverse (CGTGTTGGCTAGAGGTCCT) primers.
All measurements were performed in triplicate. One-way ANOVA was used to analyze within tissue differences in mRNA expressions, protein expressions, and folate hydrolase activities in intestinal mucosal and liver samples and folate concentrations in plasma and liver at each time point compared with 0 h. Significant differences were confirmed by a two-tailed paired t-test. Values are presented as means ± SE, and differences were considered significant at P < 0.05. GCPII mRNA transcript data were log transformed, as shown in Fig. 1.
GCPII and RFC mRNA expression in pig small intestine.
At 0 h, GCPII transcripts were detected in all three segments of the small intestine with ∼10-fold higher transcript expression in duodenal and jejunal mucosa than in ileal mucosa (Fig. 1). Ileal GCPII expression declined following birth and was barely detectable after 1 wk of age. Duodenal and jejunal GCPII expressions were variable throughout postnatal development and were not significantly lower at 6 mo compared with levels at birth.
RFC mRNA transcripts were also detected in all three segments of pig small intestine at birth with expression levels that were 1.5- to 2.2-fold higher in duodenal and jejunal mucosa compared with ileal mucosa (Fig. 2). RFC mRNA expression levels declined throughout postnatal development in all three tissues. RFC transcript levels were significantly lower (2.5-, 3.4-, and 4.2-fold) in duodenal, jejunal, and ileal mucosa at 6 mo compared with levels at birth. Ileal RFC expression remained detectable throughout all time points during postnatal development.
GCPII and RFC protein expression in pig small intestine.
Protein expressions of GCPII (Fig. 3) and RFC were analyzed in brush border membranes of all segments of pig small intestinal mucosa. Equal amounts of GCPII protein were detected in all segments of the small intestine at birth. Duodenal mucosa GCPII protein expression remained constant until 3 wk and was significantly reduced (2.5-fold) at 6 mo compared with all other time points during postnatal development (Fig. 3A). Jejunal mucosa GCPII protein expression levels showed no significant differences at any point in postnatal development (Fig. 3B). Ileal GCPII protein expression was significantly reduced by 1 wk compared with birth and continued to decline through 6 mo (Fig. 3C). Ileal GCPII protein expression was significantly lower compared with duodenum by 24 h and to jejunum by 1 wk. By 6 mo, ileal GCPII protein expression declined to levels approaching zero, whereas jejunal GCPII protein expression was significantly higher compared with all other segments of the small intestine (tissue comparison data not shown).
Duodenal, jejunal, and ileal mucosal RFC protein expression levels were detected at relatively low levels at birth and increased more than twofold by 24 h (Fig. 4) . Levels declined slightly after 24 h and remained constant for the remainder of postnatal development in all tissues. There were no significant differences in RFC expression among duodenum, jejunum, and ileum at any time point during postnatal development.
GCPII enzyme activity in pig small intestine.
GCPII enzyme activity was measured in duodenal, jejunal, and ileal brush border membranes using the previously described method of Krumdieck and Baugh (14) (Fig. 5). All reactions included 0.17 mM pHMB to inhibit intracellular folate hydrolase activity (35). At birth, activity was detected in all segments of pig small intestine with significantly higher activity in duodenum and jejunum compared with ileum at all time points in postnatal development. Duodenal and jejunal activities remained constant throughout the first 3 wk, whereas only jejunal activity was sustained at 6 mo. Ileal activity decreased rapidly by 1 wk. This pattern of change in GCPII activity is similar to changes in both GCPII transcript and protein expression over time (Figs. 1 and 3).
RFC mRNA expression in pig liver.
Since RFC activity controls folate transport across both intestinal brush border and liver plasma membrane in pig (32) we measured the mRNA expression of RFC in pig liver at all ages (Fig. 6). RFC transcripts were detected at birth in pig liver and were sustained for 3 wk. However, liver RFC transcript expression was significantly reduced by 6 mo compared with birth. These results of liver RFC mRNA expression show a pattern similar to intestinal RFC mRNA expression (Fig. 2).
Liver and plasma folate concentrations.
Pig plasma and liver folate concentrations were measured to examine the physiological implications of the observed developmental changes in expression of GCPII and RFC in pig small intestine and liver. Fig. 7 shows that pig liver folate concentrations gradually increased after birth and were significantly higher at 6 mo compared with birth. By contrast, pig plasma folate concentrations were significantly increased by 24 h, followed by a gradual decrease throughout the remainder of postnatal development. Plasma folate levels were significantly lower at 6 mo compared with birth. This initial increase and subsequent decrease in plasma folate mirrored the pattern of RFC protein expression shown in pig duodenum, jejunum, and ileum (Fig. 4).
The present study was designed to identify the ontogenic pattern of intestinal GCPII expression and activity and intestinal and hepatic RFC expression in the pig. The pig serves as a useful model of human intestinal folate hydrolysis and transport because both GCPII and RFC are expressed in jejunal brush border membranes (6, 10, 32), similar to humans (5, 18). Since these proteins regulate the intestinal assimilation of folates, whereas RFC also regulates folate uptake by the liver, we sought to identify the physiological impact of these regulations on plasma and liver folate status. Here, our findings show that intestinal GCPII and intestinal and hepatic RFC all exhibit ontogenic changes in expression and activity in the pig and that folate status is influenced by these changes.
Intestinal GCPII is expressed in the duodenal and jejunal mucosa of adult human and pig, but not rat, mouse, or monkey (5, 6, 9, 10, 34). GCPII is not reported to be expressed in adult ileum in any species. However, we identified the presence of GCPII transcripts, protein expression, and brush border membrane folate hydrolase activity in all segments of newborn pig intestine at birth. Whereas jejunal activity was sustained through out 6 mo, ileal GCPII expression and activity decreased dramatically after 1 wk and minimal detection was observed by 6 mo (Figs. 1, 3, and 5). This is consistent with reports that the adult pig and human exhibit GCPII expression and activity in duodenum and jejunum, but not ileum (6, 23). The sustained expression of GCPII in the jejunal mucosa along with the decline in ileal mucosa during postnatal development is similar to that of other brush border membrane hydrolases, such as amino-oligopeptidase, which increases in jejunal mucosa and decreases in ileal mucosa at weaning in rats (24). Since milk folates are primarily found as 5-methyl-tetrahydrofolates in the polyglutamyl form (19, 27), newborn pig intestine may express GCPII along the entire small intestine to maximize mucosal hydrolysis of the large influx of polyglutamyl folates from milk. At weaning (3 wk), piglets transition from milk polyglutamyl folates to monoglutamyl folic acid, which is found in pig chow. Folic acid does not require GCPII activity prior to absorption but its continual requirement for RFC transport could explain the sustained levels of RFC expression seen at all stages of development. The reason for a decline in GCPII expression in ileal mucosa is less clear, although the transition from milk to pig chow would decrease its requirement for folate absorption. More studies are needed to determine the regulation of ileal GCPII expression by polyglutamyl folates.
Previous studies indicate that the onset of suckling stimulates several brush border membrane enzymes and transporters in rodents and pigs, and this stimulation is due to species-specific factors present in colostrum (13, 37, 38). Low levels of RFC protein were detected at birth in all segments of pig small intestine but increased more than twofold after 24 h of consuming colostrum. RFC transcripts showed a similar but less dramatic increase in expression after 24 h in duodenum and jejunum. This suggests a stimulatory effect of suckling or age on the expression of RFC in pig intestine.
The newborn intestine of many species has the ability to absorb intact macromolecules including milk-derived proteins and immunoglobulins, a feature unique to the newborn intestine that is typically lost as the animal matures (20). Mason and Selhub (17) showed that milk folate, almost entirely bound to folate binding protein, is absorbed intact in the newborn ileum by an endocytotic mechanism, rather than by dissociation and uptake of free monoglutamyl folate by RFC. Interestingly, FBP is expressed in adult pig liver and kidney, but not in the intestine, suggesting intestinal FBP expression is a unique characteristic of the newborn (30, 31). Given the low expression of RFC in newborn pig intestine, the newborn intestine may be less reliant on RFC for dietary folate transport since folate bound to milk can be absorbed intact in the ileum. Additionally, the onset of suckling induces an increase in intestinal RFC expression perhaps as a response to the decline in endocytotic ability in the newborn intestine during development.
The ontogenic pattern of hepatic RFC mRNA expression in the pig mirrors that of intestinal RFC, increasing after 24 h of suckling followed by a gradual decrease throughout postnatal development (Fig. 2 and 6). These patterns of RFC expression seems to regulate plasma folate levels in pig, which also increase after 24 h of suckling followed by a decline over the first 6 mo (Fig. 7). On the other hand, liver folate concentrations do not seem to follow the same pattern but show a constant increase over time. This finding may represent the accumulation of folate pools in the liver over time and perhaps a decrease in folate utilization as the period of rapid growth and development slows toward maturity.
Both external (dietary) and intrinsic (hormonal) factors are known to regulate the expression of many genes during postnatal development (12, 29). The results of this study suggest that mechanisms of dietary folate hydrolysis and transport in the developing piglet undergo dynamic changes during the first 6 mo of life. Specifically, our findings help elucidate the ontogenic changes in GCPII and RFC expression during postnatal development in the pig. A separate intestinal transporter, proton-coupled folate transporter (PCFT), has recently been characterized and found to play an important role in human intestinal folate absorption (21, 39). Additional studies are needed to determine the interaction of PCFT with GCPII and RFC in dietary folate absorption, as well as to identify the molecular mechanisms that regulate the expression and activities of these gene products during development.
This work was partially funded by the Jastro-Shields Award through the Department of Nutritional Biology at University of California, Davis.
The authors thank Dr. Neil Bander, Dr. Sylvia Smith, and Dr. Tsunenobu Tamura for generous contribution to this study.
Present address for T. Shafizadeh: Tethys Bioscience, 3410 Industrial Blvd., Suite 103, West Sacramento, CA 95691.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2009 the American Physiological Society