Oral transforming growth factor-alpha enhances jejunal mucosal recovery and electrical resistance in piglet rotavirus enteritis.

A randomized, investigator-masked trial determined the effects of oral recombinant human transforming growth factor-alpha (TGF alpha) on jejunal mucosal recovery in 75 piglets with rotavirus diarrhea. Rotavirus inoculation of artificially reared piglets induced subtotal (approximately 50%) villus atrophy and watery diarrhea. Dietary TGF alpha was associated with significant restoration of villus surface area by 4 d postinoculation (p.i.) and complete restoration by 8 d p.i., whereas saline-treated animals required 12 d for recovery. Jejunal segments from clinically recovered TGF alpha-treated piglets showed an increase in electrical resistance across the epithelial barrier in vitro which was proportional to villus height. TGF alpha treatment for 12 d also produced a 30-50% increase in jejunal mucosal mass (protein content and wet weight), compared with the corresponding values in saline-treated piglets and in uninfected controls. However, oral TGF alpha did not hasten the resolution of diarrhea, enhance the specific activities of jejunal mucosal digestive enzymes, or increase jejunal glucose-stimulated Na+ absorption in vitro. We conclude that dietary TGF alpha stimulates jejunal mucosal hypertrophy, improves barrier function, and enhances regrowth of villi in rotavirus enteritis; however, it does not facilitate the restoration of functional activity or mucosal digestive enzymes. Oral TGF alpha can facilitate intestinal epithelial recovery in diseases associated with mucosal damage.

Altered jejunal potassium (Rb+) transport in piglet rotavirus enteritis.

To determine the mechanisms of K+ loss in viral diarrhea, K+ fluxes (estimated by tracer Rb+ flows) across piglet jejunum in Ussing chambers were determined. Normal jejunum was characterized by an indomethacin-sensitive short-circuit current and a small K+ secretory flow. Rotavirus-infected gut secreted K+ at high rates, probably resulting from increased prostaglandin generation because secretion was abolished by indomethacin. Tissues pretreated with indomethacin responded to 8-bromoadenosine 3',5'-cyclic monophosphate acid and 16,16-dimethyl-prostaglandin E2 with K+ secretion. The secretory response in rotavirus-infected jejunum was no greater than that in normal tissue. Serosal addition of Ca2+ ionophore A23187 caused K+ secretion in normal but not rotavirus-infected jejunum. To inhibit the basolateral uptake of K+ and reduce the driving force for secretion, ouabain was added to the bath. Ouabain unmasked a K+ absorptive process in normal intestine, which was not seen in rotavirus-infected tissue. K+ absorption was inhibited by 3-(cyanomethyl)-2-methyl-8-(phenyl-methoxy)imidazo (1,2 alpha)pyridine (Sch-28080) and omeprazole. We speculate that the high fecal K+ losses observed in human rotavirus enteritis might be caused by an imbalance between K+ secretion and an impaired apical K+ absorptive mechanism in the crypt-type epithelium.

L-glutamine with D-glucose stimulates oxidative metabolism and NaCl absorption in piglet jejunum.

To explore the relationship between intestinal fluid absorption and oxidative metabolism, we measured the effects of amino acids and glucose on piglet jejunal ion transport and oxygen consumption (QO2) in vitro. Jejunal QO2 was stimulated by L-glutamine and D-glucose but not by the nonmetabolizable organic solutes methyl beta-D-glucoside or L-phenylalanine. QO2 was maximally enhanced by the combination of D-glucose and L-glutamine (5 mM). Even though 5 mM L-glutamine was previously found to be insufficient to stimulate NaCl absorption, 5 mM L-glutamine enhanced jejunal NaCl flux when combined with equimolar mucosal D-glucose. Either D-glucose or methyl beta-D-glucoside caused an increase in short-circuit current (Isc), an increase in Na+ absorption in excess of Isc, and a decrease in Cl- secretion, when L-glutamine was substituted for D-glucose (10 mM) on the serosal side. This relationship suggests that mucosal sugars, if combined with L-glutamine, enhance neutral NaCl absorption as well as electrogenic Na+ flow. (Aminooxy)acetate, an inhibitor of alanine aminotransferase, abolished the stimulation of QO2 and the NaCl-absorptive response to L-glutamine. We conclude that the oxidative metabolism fueled by L-glutamine is linked to a NaCl-absorptive mechanism in the intestine. We propose that the CO2 produced by glutamine metabolism yields carbonic acid, which dissociates to H+ and HCO3-, which may stimulate parallel antiports in the apical membrane.

Effects of nursing on growth and development of small bowel mucosa in newborn piglets.

Trophic factors in natural milk are potential mediators of the rapid growth of intestine in neonates. To determine whether nursing stimulates growth and development of small bowel mucosa, litters of piglets were divided into suckled and artificially reared groups at birth. The latter animals were raised in an automated feeding device (Autosow) with an artificial diet simulating the nutritional composition of sow milk. At 2, 8, and 15 d of age, animals were killed and 10-cm segments of jejunum, mid-bowel, and ileum were removed. Mucosal homogenates were prepared for enzyme assay and measurement of mucosal mass. Mean body weight, total length of bowel, and circumference of bowel segments did not differ between the two feeding groups at any age studied. As anticipated, mean mucosal ornithine decarboxylase activity decreased (p less than 0.001) and measurements of mucosal mass increased (p less than 0.001) with age; however, mean values for each of these measures were never greater in the nursed animals in comparison to the artificially reared group in any segment at any age. In addition, levels of disaccharidase activity did not correlate with the feeding regimen. To investigate the possibility that unanticipated growth factors in the artificial diet might have accounted for the apparent lack of trophic effect of nursing compared to artificial rearing, we evaluated the effects of this diet and of sow colostrum on 3H-thymidine incorporation in Swiss 3T3 cells in vitro. Colostrum, but not artificial diet, stimulated greater incorporation of 3H-thymidine than culture medium alone (p less than 0.005).(ABSTRACT TRUNCATED AT 250 WORDS)

Protection of agammaglobulinemic piglets from porcine rotavirus infection by antibody against simian rotavirus SA-11.

Rotavirus, a double-stranded RNA virus, has been implicated as a diarrhea-provoking agent in a variety of animal species. Several previous reports have shown that immunization with a single serotype may result in increased in vitro neutralization titers against serotypes not represented in the immunogen. This study was undertaken to determine whether antibody from cows immunized against simian rotavirus strain SA-11 (which is alien to pigs) could protect neonatal piglets from infection with a North Carolina isolate of porcine rotavirus. Accordingly, cows were immunized with SA-11 and an immunoglobulin G (IgG)-rich fraction was isolated from their colostrum. An IgG-rich fraction was similarly isolated from colostrum of nonimmunized cows. At equal concentrations, IgG from SA-11-immunized cows had two- to fourfold higher neutralization titers to seven of eight test strains of rotavirus, including SA-11 (serotype 3); human rotavirus serotypes 1, 3, and 4; North Carolina porcine rotavirus (serotype undetermined); Ohio State porcine rotavirus (serotype 5); and bovine rotavirus (serotype 6). The IgG-rich fractions were fed as dietary supplements to agammaglobulinemic piglets infected with the North Carolina porcine rotavirus. IgG from the SA-11-immunized cows was about eightfold more effective in protecting piglets than was IgG from nonimmunized cows.

Copurification of bovine milk xanthine oxidase and immunoglobulin.

Xanthine oxidase, isolated from bovine milk, exhibited an A280:A450 nm ratio of 5.0. This ratio is reported to be indicative of highly purified enzyme preparations. Serum from a rabbit hyperimmunized against this enzyme fraction exhibited two precipitation lines when incubated with the protein in agarose double diffusion plates. Serum albumin, beta-lactoglobulin, alpha-lactalbumin, lactoferrin, casein, chymosin, and immunoglobulin were tested for reactivity. The second antigen was identified as bovine immunoglobulin. Commercial preparations of xanthine oxidase also contained immunoglobulin as a contaminant. IgG and IgA were present in Sigma (Grade III) fractions and IgM was identified in Boehringer Mannheim preparations. Immunofluorescent studies indicated that xanthine oxidase antiserum reacted with the capillary endothelium of bovine heart. Absorption of this antiserum with bovine IgG abrogated this reaction. These findings may explain apparent discrepancies between reported immunohistological association of xanthine oxidase in heart capillary endothelial cells and the absence of detectable enzymatic activity.

L-glutamine stimulates jejunal sodium and chloride absorption in pig rotavirus enteritis.

Rotavirus enteritis is the leading cause of diarrhea in infants worldwide. A research priority of the World Health Organization is to develop oral rehydration solutions containing amino acids or other additives that will stimulate intestinal absorption more efficiently than the current glucose-based oral rehydration solutions. Glutamine is the principal metabolic fuel of the small bowel and a putative stimulator of mucosal repair. This report describes the transport response to mucosal L-glutamine following intestinal injury caused by porcine rotavirus. Peak symptoms and mucosal damage were observed 2-7 days after oral rotavirus inoculation. In vitro transport studies of the maximally injured region, the midjejunum (80% reduction in lactase), surprisingly, showed transport responses to L-glutamine (30 mmol/L) and L-alanine (30 mmol/L) that were similar qualitatively and quantitatively to those observed in control tissue. Subsequent application of mucosal D-glucose (30 mmol/L) caused additional stimulation of electrogenic Na+ transport, but the response to glucose was blunted (P less than 0.05) in the infected tissues. Glutamine and alanine enhanced Na+ absorption to a similar degree (2-2.5 muEq.cm-2.h-1), but glutamine stimulated equal amounts of electrogenic and electroneutral NaCl absorption, whereas alanine had no significant effect on net Cl- flux. Glutamine is a potentially useful substrate for investigation in oral rehydration solutions for infant diarrhea.

Treatment of rotavirus infection in neonate and weanling pigs using natural human interferon alpha.

Colostrum-deprived neonate piglets challenged with rotavirus and 3-week-old newly weaned piglets naturally exposed to rotavirus were treated with low doses of natural human interferon alpha (nHuIFN alpha) administered into the oral cavity or included in the liquid diet. The colostrum-deprived piglets given the highest dosage of nHuIFN alpha (50 IU/kg body weight) had lower viral excretion scores at 3 (p less than 0.11) and 4 days (p less than 0.001) after virus inoculation. Average group weights and weight gains were consistently greater for all nHuIFN alpha-treated neonate groups; however, these differences were not significant. Mortality rates were lower in neonates for the three highest nHuIFN alpha treatment groups (20%, 30%, and 20%) than in the lowest treatment group and controls (60% and 45%, respectively). Three-week-old weanling piglets did not have significant differences in the total average diarrhea or rotavirus excretion scores. After 10 days, the group receiving the highest dosage of nHuIFN alpha had significantly greater average weight gain than the control group (p less than 0.05). There was a significant (p less than 0.01) dose-dependent effect between the logarithm10 of the nHuIFN alpha dosage and weight gain in weanling piglets. There were no death losses in the 3-week-old weanling piglets from natural rotavirus exposure.

Development of L-glutamine-stimulated electroneutral sodium absorption in piglet jejunum.

Glutamine is the primary metabolic fuel of the small intestine. To determine the effects of glutamine on intestinal electrolyte transport, piglet (3 days to 3 wk old) jejunum was bathed in Ussing chambers in a buffer containing 10 mM serosal glucose, and the effects of different concentrations of mucosal L-glutamine and D-glucose on short-circuit current and transmucosal Na+ and Cl- transport were measured. Resting jejunum secreted Na+ and Cl- in an electrogenic manner. In contrast to mucosal D-glucose (30 mM), which promoted electrogenic Na+ absorption (1.8 mueq.cm-2.h-1), mucosal L-glutamine (30 mM) stimulated both Na+ (2.7 mueq.cm-2.h-1) and Cl- (2.2 mueq.cm-2.h-1) absorption. This NaCl-absorptive jejunal response depended on the presence of both Na+ and Cl-, did not appear until animals were greater than 7 days of age, and was not observed with glucose, phenylalanine, or mannitol. Serosal, as well as mucosal, glutamine (30 mM) promoted electroneutral NaCl absorption. A small electrogenic Na(+)-absorptive response to L-glutamine was also observed. The effect of L-glutamine on jejunal NaCl transport resembles that of other metabolic fuels on colonic transport; its mechanism remains to be determined. We conclude that glutamine promotes electroneutral salt absorption in the small intestine.

Villous atrophy, crypt hyperplasia, cellular infiltration, and impaired glucose-Na absorption in enteric cryptosporidiosis of pigs.

Intestinal morphology and fluid and electrolyte transport were examined in a neonatal porcine model of cryptosporidiosis. Sections of jejunum, ileum, and colon were obtained for morphometric analysis on days 3, 6, 9, and 12 postinfection, and in vivo perfusion studies of jejunum and ileum were conducted on days 3 and 4 postinfection. The most severe morphologic lesion was seen in the ileum on day 3, and consisted of villous atrophy, crypt hyperplasia, and cellular infiltration. Villous surface area was reduced from 2.1 +/- 0.4 x 10(5) microns2 in control ileum to 0.8 +/- 0.1 x 10(5) microns2 in infected ileum, a result associated with enterocytes that were fewer in number and reduced in cross-sectional area. Conversely, the number of inflammatory cells in the lamina propria of the villus increased from 456 +/- 116 in control to 1014 +/- 187 in infected villus without a significant change in the volume of the lamina propria. At the height of infection, there was an approximate 1:2 ratio of both organisms and inflammatory cells to villous enterocytes. In contrast, organisms were not observed in the crypts, and the concentration of inflammatory cells in crypt lamina propria was unaltered. Disappearance of organisms and polymorphonuclear cells from the ileum was associated with restoration of normal structure and was complete by day 12. Although organisms were seen in the colon, the general architecture was not severely affected. On days 3 and 4 postinfection, there was a complete impairment of the glucose-stimulated Na and water absorption in both jejunum and ileum of infected pigs; however, absorption of electrolytes and water from a basic Ringer's solution, in the absence of glucose, was not significantly affected. These results are consistent with a malabsorptive diarrheal disease associated with the morphological damage and are very similar to those seen in enteric viral disease in pigs, except that the upper intestine is more severely affected in the latter.

Purification and properties of sulfhydryl oxidase from bovine pancreas.

Immunofluorescent studies showed that antibodies prepared against bovine milk sulfhydryl oxidase reacted with acinar cells of porcine and bovine pancreas. A close inspection of the specific location within bovine pancreatic cells revealed that the zymogen granules, themselves, bound the fluorescent antibody. Bovine pancreatic tissue was homogenized in 0.3 M sucrose, then separated into the zymogen granule fraction by differential centrifugation. The intact zymogen granules were immunofluorescent positive when incubated with antibodies to bovine milk sulfhydryl oxidase, and glutathione-oxidizing activity was detected under standard assay conditions. Pancreatic sulfhydryl oxidase was purified from the zymogen fraction by precipitation with 50% saturated ammonium sulfate, followed by Sepharose CL-6B column chromatography. Active fractions were pooled and subjected to covalent affinity chromatography on cysteinylsuccinamidopropyl-glass using 2 mM glutathione as eluant at 37 degrees C. The specific activity of bovine pancreatic sulfhydryl oxidase thus isolated was 10-20 units/mg protein using 0.8 mM glutathione as substrate. Ouchterlony double-diffusion studies showed that antibody directed against the purified bovine milk enzyme reacted identically with pancreatic sulfhydryl oxidase. The antibody also immunoprecipitated glutathione-oxidizing activity from crude pancreatic homogenates. Western blotting analysis indicated a 90,000 Mr antigen-reactive band in both bovine milk and pancreatic fractions while sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed a single silver-staining protein with an apparent Mr 300,000. Thus, we believe that sulfhydryl oxidase may exist in an aggregated molecular form. Bovine pancreatic sulfhydryl oxidase catalyzes the oxidation of low-molecular-weight thiols such as glutathione, N-acetyl-L-cysteine, and glycylglycyl-L-cysteine, as well as that of a high-molecular-weight protein substrate, reductively denatured pancreatic ribonuclease A.

Diarrhea: the nemesis of the artificially reared, early weaned piglet and a strategy for defense.

Rearing early weaned piglets artificially for the purpose of increasing the efficiency of the sow is an attractive management concept. However, high death losses resulting from diarrhea in artificially reared piglets have dampered enthusiasm for early weaning. Enterotoxigenic Escherichia coli, transmissible gastroenteritis virus and rotavirus are the three main enteropathogens responsible for causing the diarrhea. The enteropathogens infect the small intestine, which produces a secretory or malabsorptive diarrhea. In nature, the nursing piglet is protected from the enteropathogens by antibody bathing his gut. The source of the antibody is the dam's colostrum and milk. It should be possible to protect artificially reared, early weaned piglets from enteropathogens by feeding them diets that contain antibodies to putative enteropathogens.

Liquid diets fed hourly to pigs weaned at 19 days of age and exposed to rotavirus (weanling diarrhea).

Five diets were tested for their capacity to promote weight gains in newly-weaned, 19-d-old, rotavirus-exposed pigs. The diets were tested under conditions designed to minimize the stress of weaning. That is, rotavirus-exposed pigs were moved at weaning to an isolation unit, caged individually and fed hourly liquid diets that were high (approximately 26%) and low in protein (approximately 11%). In all experiments, pigs experienced postweaning rotavirus-associated diarrhea and depression in rate of gain. Pigs grew faster: when fed diets high in protein (approximately 26% protein) vs low in protein (approximately 11% protein) and when fed diets containing cows' milk proteins vs proteins from soybean flour. A diet containing antibodies to rotavirus did not ameliorate the weanling diarrhea.

Development of resistance of enterocytes to rotavirus in neonatal, agammaglobulinemic piglets.

The first part of this report describes the development of a technique for evaluating the growth of rotavirus under controlled conditions that approximate a natural infection. A standard dose of rotavirus (approximately 10(9) viral particles) was injected into ligated segments in the small intestine of newborn, agammaglobulinemic, colostrum-deprived piglets. After various periods postinoculation, the segments were retrieved and the enterocytes were evaluated for the presence of rotaviral antigens by immunofluorescence and rotaviral particles by transmission electron microscopy. Peak immunofluorescence in enterocytes was detected at 8 h postinoculation in the upper and middle jejunum and ileum. Transmission electron microscopy at this time revealed fully formed virions which were not seen in sections examined before this 8-h period. The second part of our study describes the use of ligated segments in determining the susceptibility to rotavirus of enterocytes in piglets ranging in age from newborn to 2 weeks. By the time piglets were 2 days old, enterocytes in the upper half of the small intestines appeared to be resistant to rotavirus, whereas those in the lower half seemed partially resistant. Between 4 and 8 days of age, enterocytes in the lower half also became resistant. Resistance paralleled the loss in capacity of piglets to transport macromolecules through enterocytes and was not correlated with the loss in capacity to internalize macromolecules.

Tissue distribution of mammalian sulfhydryl oxidase.

Sulfhydryl oxidase activity is present in cow, goat, sow, human, and rat milks, and can also be measured in several rat tissues following homogenization in 1% polyoxyethylene-9-lauryl ether. These include lactating mammary tissue, kidney, and pancreas. Bovine kidney homogenates also exhibit sulfhydryl oxidase activity; however, no activity could be detected in rat thymus, brain, heart, liver, spleen, lung, or small intestinal tissue homogenates. Indirect immunofluorescent staining of tissue sections using rabbit antibodies directed against highly purified bovine milk sulfhydryl oxidase preparations revealed that the enzyme is closely associated with the plasma membrane of lactating cow and rat mammary tissues and the basal-lateral membrane of rat kidney cortex. In addition, the oxidase appears to be associated with endothelial cells lining the capillaries of rat kidney, heart, and small intestine, and centroacinar cells in pancreatic tissue slices also stain for sulfhydryl oxidase. In contrast, liver, brain, and thymus tissues do not exhibit fluorescent staining and appear to be devoid of sulfhydryl oxidase activity.

Effect of dietary regimen on rotavirus-Escherichia coli weanling diarrhea of piglets.

Previously, we induced weanling diarrhea in piglets by infecting them with rotavirus followed by hemolytic enteropathogenic Escherichia coli. We postulated that rotavirus, by damaging the epithelium of the small intestines, produced an enteroenvironment which favored the selection and growth of enteropathogenic E. coli. Furthermore, diet might affect the enteroenvironment and influence the course of the disease. To test this, newly weaned 3-week-old piglets were assigned to one of four dietary regimens and infected with rotavirus followed 24 h later with enteropathogenic E. coli. The course of the disease was followed by monitoring the severity of diarrhea and the fecal shedding of rotavirus and enteropathogenic E. coli in these dually infected piglets. The dietary regimen designed to tax the digestive and absorptive capacities of the piglets (high nutrient intake fed three times a day) produced the most prolonged diarrhea, colonization of the gut by hemolytic enteropathogenic E. coli, and persistent shedding of rotavirus (P less than 0.01). The same nutrient intake divided into 24 equal increments and fed hourly produced a less severe response (P less than 0.01). The least severe response was seen in piglets fed one-third the nutrient intake either hourly or three times a day (P less than 0.01). We conclude that dietary regimen plays an important role in rotavirus-E. coli-induced weanling diarrhea.

Dietary regimen, rotavirus, and hemolytic enteropathogenic Escherichia coli in weanling diarrhea of pigs.

Previously, we induced weanling diarrhea in piglets by infecting them first with rotavirus followed by a hemolytic enteropathogenic Escherichia coli. We postulated that rotavirus, by damaging the epithelium of the small intestines, produced an entero-environment which favored the selection and growth of the enteropathogenic E. coli. Furthermore, diet might affect the entero-environment and influence the course of the disease. To test this, newly weaned 3-week-old piglets were assigned to one of four dietary regimens and infected with rotavirus followed 24 h later with an enteropathogenic E. coli. The course of the disease was followed by monitoring the severity of diarrhea, and the fecal shedding of rotavirus and the enteropathogenic. E. coli in these dually infected piglets. The dietary regimen designed to tax the digestive and absorptive capacity of the piglet (high nutrient intake fed 3 times a day) produced the most prolonged diarrhea, colonization of gut by the hemolytic enteropathogenic E. coli and persistent shedding of rotavirus. The same nutrient intake divided into 24 equal increments and fed hourly produced a less severe response. The least response was seen in piglets fed one-third the nutrient intake either hourly or 3 times a day. We conclude that dietary regimen plays an important role in rotavirus-Escherichia coli induced weanling diarrhea.

Rotavirus and hemolytic enteropathogenic Escherichia coli in weanling diarrhea of pigs.

Since the turn of the century, Escherichia coli has been implicated in the etiology of weanling diarrhea (colibacillosis). However, rotavirus--a virus that destroys enterocytes--has been shown recently to be causally associated with weanling diarrhea of pigs. The role of both rotavirus and hemolytic enteropathogenic E. coli in weanling diarrhea was assessed in this study. Pigs from a closed herd were farrowed and weaned by two markedly different systems: an "intensive care sanitary" system and a "conventional unsanitary" system. Pigs weaned at 3 weeks of age in the sanitary system usually experienced a rotaviral diarrhea about 16 days postweaning. No hemolytic E. coli were detected in feces from these pigs. Peers weaned at the same time by the unsanitary system commenced diarrhea 3 days postweaning. Rotavirus and nonhemolytic E. coli were detected in the feces at the onset of diarrhea and for a few days thereafter. Then, the aerobic fecal flora shifted to nearly pure hemolytic enteropathogenic E. coli. About 10 days later, the diarrhea waned, and the fecal flora shifted back to nonhemolytic E. coli. This hemolytic E. coli shedding pattern could not be duplicated in artificially inoculated sanitary pigs unless they were inoculated with the hemolytic E. coli during a rotaviral-associated diarrhea. Otherwise, the shedding of hemolytic E. coli was fleeting, and the diarrhea, if present, was mild. Pigs developed humoral antibodies to the rotavirus but not to the hemolytic E. coli. We conclude that rotavirus damages the epithelium of the small intestines, which changes the luminal environment to one that favors colonization by enteropathogenic E. coli.

Rotaviral antibodies in cow's milk.

In the past, it has been reported that neonatal diets made from unheated cow's milk were superior to those made from heated cow's milk. It was observed that piglets were equally protected from rotaviral diarrhea when they were fed diets made from either unheated milk that came from a cow immunized against porcine rotavirus or from a cow that was not immunized. Because of this observation, we examined four pools of "normal" cows' colostrum and 58 samples of "normal" cow's milk for the presence of antibody to rotavirus. All pools of colostrum, collected in four different years, had immunofluorescent antibody titers of 1:100 to rotavirus. Seventy-two percent of the samples of milk were also positive--titer no higher than 1:10. Antibodies to rotavirus were found in cow's milk at a creamery prior to but not after pasteurization. Rotaviral antibodies were detected in one out of eight brands of milk bought at the market--perhaps indicating inadequate pasteurization for this brand. These results support the proposition that, at least in part, unheated milk is superior to heated milk because unheated milk contains antibody to an ubiquitous enteropathogen like rotavirus.

Intestinal barriers to water-soluble macromolecules.

Neonates of some species of mammals absorb water-soluble macromolecules from the lumen of the gut to the circulation. This is a means for providing the neonate with passive immunological protection. The accepted model for absorption of macromolecules, particularly immunoglobulin G (IgG), has at least three phases: adherence of the macromolecule to the brush border on enterocytes; internalization of the macromolecule within the enterocytes; and egress of the macromolecule into the lamina propria. With regard to the absorption of IgG, adherence is thought to be a specific reaction of ligand (IgG) with a plasmalemma binding site. Pinocytosis is activated and internalization follows. Egress into the lamina propria occurs at the basal-lateral membrane by a process of reverse pinocytosis. Unbound (unprotected) macromolecules that are internalized in the pinocytosic fluid are shunted off to lysosomes and either digested or stored therein. Neonatal rodents fit this model for macromolecular absorption. However, in another group of neonates (e.g., pig, cow, horse), nonselected absorption takes place, in that IgG and other macromolecules are transported from the gut lumen to the blood. In a third group of neonates, (e.g., human, guinea pig) absorption of IgG is either of low order or nonexistent. Since neonatal mammals possess a mechanism for absorbing macromolecules, there is the potential for internalizing toxic macromolecules if the toxin is presented to the neonate's enterocytes in competitive amounts. Adults retain remnants of the neonatal absorptive mechanism.