Orally Administered Lactoferrin Increases Hepatic Protein Synthesis in Formula-Fed Newborn Pigs 1

Lactoferrin is a polypeptide which is abundant in colostrum; however, its biologic effect in the neonate is unknown. The objective was to determine the potentially anabolic effect of orally administered lactoferrin on visceral organ growth and protein synthesis in newborn pigs. We studied a total of 18 unsuckled newborn pigs from six litters. Three pigs from each litter were randomly assigned to one of three dietary treatment groups (n = 6) and bottle-fed (10 mL/h) formula, formula containing physiologic levels (1 mg/mL) of added bovine lactoferrin (bLF), or colostrum. After 24 h of feeding, we measured visceral organ protein synthesis in vivo using a flooding dose of [³H]phenylalanine. We also measured visceral organ protein and DNA mass, as well as intestinal hydrolase activities and villus morphology. Hepatic protein synthesis in pigs fed either formula containing bLF or colostrum was similar and in both groups was significantly higher than in pigs fed formula. Splenic protein synthesis was not significantly different in pigs fed either formula or formula containing bLF, but was significantly higher in colostrum-fed animals. There were no significant differences in small intestinal growth, protein synthesis, or hydrolase activities between newborn pigs fed formula, formula containing bLF, or colostrum. Our results demonstrate that feeding formula containing physiologic concentrations of added bLF increased hepatic protein synthesis in newborn pigs, suggesting that colostrumborne lactoferrin serves an anabolic function in neonates.

Abbreviations: bLF, bovine lactoferrin; BW, body weight; PCA, perchloric acid

Lactoferrin is a 78-kD, single-chain polypeptide, containing two high affinity iron-binding sites, and belongs to the transferrin gene family^{(1, 2)}. Lactoferrin was initially isolated from bovine ⁽³⁾ and human ⁽⁴⁾ milk. Although lactoferrin is found in various biologic fluids, including pancreatic and bronchial secretions and circulating neutrophil granules⁽⁵⁾, it is particularly abundant in human, porcine, and bovine colostrum with concentrations ranging from 1 to 5 mg/mL^{(6, 7)}.

Despite its abundance in colostrum and milk, the in vivo biologic significance of colostral-borne lactoferrin in the neonate is unknown. Lactoferrin has been shown to have pronounced biologic effects in cell culture systems. For example lactoferrin inhibits bacteria growth⁽⁸⁾, facilitates intestinal iron absorption^{(9, 10)}, and reduces myelopoiesis via inhibition of granulocyte/macrophage-colony stimulating factor production^{(11, 12)}. Studies in vitro also have shown that lactoferrin stimulates proliferation of intestinal crypt cells⁽¹³⁾, lymphoctyes ⁽¹⁴⁾, and hepatocytes ⁽¹⁵⁾. Although in vivo mitogenic effects of lactoferrin have not been demonstrated, in vivo studies have shown that lactoferrin is rapidly cleared from the circulation by hepatic parenchymal cells via a specific receptor-dependent mechanism^{(16, 17)}, and specific receptor-binding for lactoferrin has been characterized in both intestinal enterocytes⁽¹⁸⁾ and peripheral lymphocytes⁽¹⁹⁾. Therefore, it is conceivable that lactoferrin has biologic effects on the gastrointestinal tract, liver, and lymphoid tissues, including the spleen.

We hypothesized that the mitogenic effect of relatively high lactoferrin concentrations in porcine colostrum contributes to the marked stimulation of gastrointestinal protein synthesis that we observed previously in colostrum-fed newborn pigs ⁽²⁰⁾. Furthermore, evidence suggests that maternal lactoferrin is absorbed in human milk-fed preterm infants ⁽²¹⁾. Therefore, it is conceivable that ingested colostrum-borne lactoferrin may be absorbed into the blood and stimulate growth of other responsive tissues, particularly the liver and perhaps the spleen. To test the hypothesis that colostrum-borne lactoferrin stimulates intestinal, hepatic, and splenic growth in vivo, we measured growth and protein synthesis in newborn pigs fed an artificial formula containing bLF for 24 h after birth. We compared the organ growth response in newborn pigs fed formula containing bLF with that of pigs fed either formula without added lactoferrin or colostrum.

METHODS

Animals and design. Six litters of conventional cross-bred pigs(Texas A&M University, College Station, TX) were obtained immediately after birth (before suckling), weighed, and randomly assigned by weight to receive their respective dietary treatment. In each of the six litters, three pigs were bottle-fed 20 mL/kg of BW every 2 h either porcine colostrum, formula, or formula containing bLF for 24 h. Before initiating the feeding protocol, the umbilical artery of each pig was catheterized with polyvinyl chloride catheters (Sherwood Medical, St. Louis, MO) under general isoflurane anesthesia (Aerrane, Anaquest, Madison, WI). The pigs were allowed to recover approximately 1-2 h before beginning the feeding protocol. Animals were housed in separate cages with a dry towel for bedding; ambient temperature was maintained at approximately 28-29 °C. 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.

Feeding protocol. The colostrum was a pooled sample collected from conventional sows within the first 24 h postpartum. The formula was designed to be comparable to porcine milk and contained 60 g/L protein, 58 g/L fat, 52 g/L lactose, and 1000 kcal/L gross energy (Ross Laboratories, Columbus, OH). At the beginning of the 24-h feeding period, purified bLF (New Zealand Milk Products, Wellington, NZ) was added to the formula to yield a final concentration of 1 g/L and mixed continuously at 4 °C. Immediately before each 2-h feeding, approximately 20 mL/kg of each of the three diets were warmed to 37 °C and bottle-fed to the respective animal. The actual dietary intake was determined by weighing the bottles before and after each feeding.

Infusion and tissue collection. After 24 h on the experimental diets, the animals were infused via the umbilical arterial catheter slowly over 2 min with 10 mL/kg BW of a flooding dose of L-[4-³H]phenylalanine(Amersham Corp., Arlington Heights, IL) which provided 1.5 mmol and 37 MBq/kg of BW. At 5, 15, and 30 min after the midpoint of the infusion, arterial blood samples were collected in heparinized tubes and placed on ice for measurement of blood phenylalanine-specific radioactivity. Immediately after withdrawing the 30-min blood sample, animals were anesthetized with an i.v. dose of pentobarbital (50 mg/kg of BW) and exsanguinated by withdrawing approximately 30 mL of blood. The abdomen was opened and flushed with ice-cold saline, and the small intestine, from the pylorus to the ileocecal junction, was removed free of mesenteric tissue and placed in ice-cold saline. The small intestine was then divided into three segments. The duodenum, defined as the segment from the pylorus to the peritoneal reflection (analogous to the ligament of Trietz), was removed. The remaining small intestine, from the peritoneal reflection to the ileocecal junction, was then divided in half, and the proximal and distal halves were designated as jejunum and ileum, respectively. Each segment was flushed with cold saline and weighed. Samples for histologic analysis were obtained from the midportion of each segment, and the remaining tissue was scraped and frozen in liquid nitrogen. After the small intestine had been excised, the liver and spleen were quickly removed and weighed, and a sample of each was frozen in liquid nitrogen.

Enzyme analyses. Disaccharidase activities were determined in homogenates of scraped jejunal mucosa suspended in 40 mM phosphate buffer, pH 8.0, using 0.3 M sucrose, 0.112 M maltose or 0.6 M lactose, and 0.122 mMp- hydroxymercuribenzoate ⁽²²⁾. Alkaline phosphatase activity was assayed in 0.3 M 2-amino-2-methyl-1-propanol buffer, pH 10.3, using 5 mM p-nitrophenol phosphate as the substrate and 1 mM MgCl₂ in the assay medium (a modification of kit no. 104, Sigma Chemical Co., St. Louis, MO) ⁽²²⁾. Leucine aminopeptidase was measured using L-leucine-β-naphthylamide as the substrate (kit no. 251, Sigma Chemical Co.). DNA was measured using the method of Labarca and Paigen ⁽²³⁾. Protein was determined with a bicinchoninic acid protein assay kit (Pierce, Rockford, IL). Enzyme activities are expressed as micromoles/min/g protein.

Histologic analysis. The small intestine samples were processed and stained with hematoxylin and eosin as previously described⁽²⁴⁾. Fresh proximal jejunum and ileum (2 cm) were opened, pinned villus side up on corks, and fixed in 4% paraformaldehyde in PBS for 24-48 h. The sections were examined in a blined manner with a Zeiss Axiophot (Carl Zeiss Inc., Germany) using a calibrated grid to quantify villus height from at least 30 well oriented crypt-villus columns from each section of tissue.

[³H]Phenylalanine specific activity measurement. The specific radioactivity of [³H]phenylalanine was determined in whole blood samples obtained 5, 15, and 30 min after infusion of the [³H]phenylalanine and in tissue samples. The tissue samples were homogenized in 0.2 mol/L PCA, and the specific radioactivity of[³H]phenylalanine in the PCA-soluble and -insoluble fractions were determined by anion-exchange HPLC as described previously⁽²⁵⁾.

Calculations. The rate of jejunal and ileal protein synthesis was estimated from the absolute rate of phenylalanine incorporation into PCA-insoluble tissue fractions as described previously, to avoid the confounding effect of endocytosis of colostral Ig ⁽²⁰⁾. The rate of phenylalanine incorporation into tissue protein was determined using the following equation: where C is the radioactivity of phenylalanine in the PCA-insoluble or protein-bound pool(Bq), t is time of labeling in hours, V_s is the rate of phenylalanine incorporation (μmol/h),S_a is the specific activity of the PCA-soluble or tissue free phenylalanine pool (Bq/μmol), V_d is the rate of phenylalanine appearance resulting from protein degradation(μmol/h), and S_b is the specific activity of the PCA-insoluble or protein-bound phenylalanine pool (Bq/μmol). This equation can be further simplified to the following: Equation assuming that the termV_d·S_b is negligible. This assumption was made because the observed value forS_a was approximately 100-fold greater than the value for S_b. In addition, the value forS_b is likely to be inaccurate because of dilution by phenylalanine in ingested proteins.

Equation 2B

Equation 2C

The value used for S_a was corrected to represent the average tissue phenylalanine specific activity at the midpoint(t½) of the 30-min labeling period. The correctedS_a for each pig was calculated by adding individual tissue S_a (Bq/μmol) after time (t), and the rate of change in blood S_a(Bq·μmol^-1·min^-1) was estimated from the regression of 5-, 15-, and 30-min blood samples of all pigs within a treatment group as follows: Equation

Equation 2D

The absolute rate of phenylalanine incorporation calculated for both the jejunal and ileal segment was expressed per unit of BW(μmol·h^-1·kg BW^-1).

For all other tissues besides the jejunum and ileum, protein synthesis was calculated as a fractional rate (K_s,%/d) from the equation described by Garlick ⁽²⁶⁾:Equation where all parameters are as described above.

Equation 2E

Statistics. Treatment means were analyzed by one-way analysis of variance; dietary treatment was the main effect. Differences between treatments were determined by Fisher's least significant difference test. Results are presented as means with the pooled SD from the one-way analysis of variance. A probability value of less than 0.05 was considered significant.

RESULTS

The mean initial BW of the colostrum, formula, and formula plus bLF groups were 1.27, 1.27 and 1.26 kg, respectively; the BW gain during the 24-h feeding period did not differ among the groups (data not shown). There were no differences among treatment groups in the weight, protein, or DNA content in either the liver or the spleen (Table 1). Among the three treatment groups, there were no significant differences in weight, length, villus height, or mucosal DNA content in either jejunum or ileum(Table 2). However, the mucosal protein content of the jejunum and ileum in the colostrum-fed pigs was greater than in either the formula or formula plus bLF group (Table 2). Among the three treatment groups, there also were no differences in jejunal or ileal(data not shown) enzyme activities, either specific or total, of lactase, sucrase, maltase, aminopeptidase, or alkaline phosphatase(Table 3).

Table 1

Table 2

Table 3

Hepatic protein synthesis in pigs fed either colostrum or formula plus bLF was similar (118 ± 8 and 124 ± 6%/d, respectively), and in both groups was significantly (p < 0.05) higher than in pigs fed formula (97 ± 6%/d) (Fig. 1). Splenic protein synthesis in pigs fed either formula or formula plus bLF was not different (59± 6 and 48 ± 3%/d, respectively), but was significantly(p < 0.05) lower than in the colostrum-fed animals (81 ± 7%/d) (Fig. 1). The rates of small intestinal protein synthesis were not significantly different among the three treatment groups(Fig. 1). No differences between jejunal and ileal protein synthesis were observed among the three treatment groups.

Figure 1

DISCUSSION

Lactoferrin is an iron-binding protein that is particularly abundant in mammalian milk and colostrum. Although a number of in vitro studies have shown that lactoferrin has mitogenic effects on various cell types, there is virtually no information regarding the biologic effects of lactoferrinin vivo, especially in neonates. Our results are the first to demonstrate that orally administered bLF stimulates hepatic protein synthesis in formula-fed newborn pigs. Moreover, we showed that the stimulation of hepatic protein synthesis in pigs fed formula plus bLF was comparable to that of colostrum-fed pigs and in both groups was significantly higher than in pigs fed formula only.

We speculate that the effects of oral lactoferrin on hepatic protein synthesis are direct for the following reasons. First, several studies indicate that lactoferrin is relatively resistant to proteolytic digestive enzymes, suggesting that milk-borne lactoferrin may remain biologically active in the intestinal lumen ^(27-29). Second, recent reports using isolated porcine enterocytes and brush-border membranes^{(18, 30)} have demonstrated the presence of functional lactoferrin receptors, suggesting that binding and receptor-mediated endocytosis of luminal lactoferrin by mucosal enterocytes is possible. It should be noted that the enterocyte receptor-binding studies using competitive inhibition assays have shown that bLF is capable of binding porcine lactoferrin receptors. Furthermore, the urinary excretion, and therefore intestinal absorption, of maternally derived lactoferrin has been demonstrated in human milk-fed premature infants ⁽²¹⁾. Third, studies have demonstrated the presence of hepatic lactoferrin receptors and the key role of the liver in clearing lactoferrin from the circulation^{(16, 17, 31)}. Fourth, in vitro studies have shown that iron-saturated lactoferrin is a comitogenic substance for neonatal hepatocytes ⁽¹⁵⁾. Therefore, we suggest that orally administered bLF survives luminal digestion and is absorbed from the intestine into the portal blood where it binds to hepatocytes perhaps via specific receptors and stimulates protein synthesis.

Given that lactoferrin potentially may have a stimulatory role in erythropoiesis, an alternative hypothesis may be that the increased hepatic protein synthesis was due to the stimulation of hematopoietic cells within the liver by lactoferrin absorbed from the intestinal lumen. This possibility seems unlikely given that most of the hematopoiesis in the newborn pig is like that of the human and occurs in the bone marrow ⁽³²⁾. Furthermore, the possibility that orally administered bLF enhanced intestinal nutrient absorption which in turn stimulated hepatic protein synthesis is unlikely, because there were no effects of bLF on either intestinal mass or digestive enzyme activity.

We originally hypothesized that orally administered bLF has potential anabolic effects on the small intestine and spleen. In vitro lactoferrin had been shown to stimulate intestinal crypt cell proliferation⁽¹³⁾. This observation, combined with the recent evidence of intestinal lactoferrin receptors in neonatal pigs, suggested that oral lactoferrin could have anabolic or mitogenic potential. However, the data from this study do not demonstrate any effect of orally administered bLF on either small intestine growth or enzyme activity in newborn pigs. It may be that for bLF to affect small intestine growth or protein synthesis, prolonged exposure is required, and we may not have seen an effect because of the short duration of the present study.

In vitro studies have demonstrated that lymphocytes contain specific lactoferrin receptors and are responsive to lactoferrin^{(14, 19)}. We hypothesized that orally administered bLF might affect the spleen, because it is one of the largest lymphatic organs in the body and is a major site of lymphocyte production. Although orally administered bLF did not affect splenic growth or protein synthesis in this study, splenic protein synthesis was significantly higher in animals fed colostrum than in those fed formula. This finding suggests that either the increased concentration of nutrients, or perhaps other growth factors, stimulated splenic protein synthesis in those animals.

Acknowledgments. The authors thank Judy Rosenberger, Scott Perkinson, Kathy McKee, and Susan Amick for the technical assistance. We also thank Leslie Loddeke for editorial assistance and Julia Redmond for secretarial assistance in the preparation of this manuscript.

Footnotes

This work is a publication of the USDA/ARS Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, TX. This project has been funded in part with federal funds from the U.S. Department of Agriculture, Agricultural Research Service under Cooperative Agreement number 58-6250-1-003. The contents of this publication do not necessarily reflect the views or policies of the U.S. Department of Agriculture, nor does mention of trade names, commercial products, or organizations imply endorsement from the U.S. Government. Cited Here...