Expression of the neonatal Fc receptor, FcRn, on human intestinal epithelial cells.

Maternal IgG is transferred to the suckling mouse and rat through a major histocompatibility complex (MHC) class I-related Fc receptor (FcRn) on the brush border of the proximal small intestine. We have previously described a site on the epithelial surface of the human fetal intestine with IgG binding characteristics similar to FcRn. We report here the identification by reverse transcriptase polymerase chain reaction amplification and sequencing of the human orthologue of rat and mouse FcRn in tissue obtained from human fetal and adult intestine. FcRn protein was detected in adult human intestine by western blot. Immunohistochemical studies of sections of human intestine show that the FcRn is localized mostly to the epithelial cells, where it is in the apical region. These data suggest that the binding of IgG previously seen in the fetal intestine is due to the presence of FcRn. Potential roles for this MHC class I-like Fc receptor in the human intestine include the transfer of passive immunity, induction of oral tolerance, and immunosurveillance.

Increased clearance of IgG in mice that lack beta 2-microglobulin: possible protective role of FcRn.

The mechanisms that regulate immunoglobulin G (IgG) catabolism are little understood. We have previously found unusually low IgG concentrations in sera of mice homozygous for a targeted disruption of the beta 2-microglobulin gene. We therefore investigated whether this might result, at least in part, from increased clearance of IgG from the systemic circulation in mice lacking beta 2-microglobulin. We compared the half-lives of radiolabelled mouse IgG1 injected intravenously into beta 2-microglobulin-/- mice and wild-type or heterozygous siblings. The clearance of 125I-labelled IgG1 was strikingly more rapid in the mice lacking beta 2-microglobulin. beta 2-microglobulin-/- mice lack functional molecules of the MHC class I-related Fc receptor, FcRn. Some mutations in mouse IgG1 that increase its clearance have recently been shown to prevent binding to FcRn in the gut. To determine whether the slower degradation of immunoglobulin in mice with beta 2-microglobulin correlated with the ability of the antibody to bind FcRn, we measured the clearance of chicken IgY, which does not bind this receptor. The 125I-labelled IgY was catabolized equally rapidly in beta 2-microglobulin-deficient and wild-type mice. We compared the half-lives of the four subclasses of mouse IgG in beta 2-microglobulin-/-, +/-, and +/+ mice to determine whether the difference we had noted for IgG1 was peculiar to this subclass. The 125I-labelled IgG of all subclasses, with the possible exception of IgG2b. was degraded more rapidly in the beta 2-microglobulin-deficient mice than in heterozygous or wild-type siblings. These data suggest that FcRn can protect IgG from degradation, and is therefore important in maintaining IgG levels in the circulation.

Requirement for a beta 2-microglobulin-associated Fc receptor for acquisition of maternal IgG by fetal and neonatal mice.

There is considerable evidence to suggest that an FcR similar in structure to class I MHC Ags, neonatal Fc receptor (FcRn), transports IgG across the intestinal epithelium of suckling mice. However, this has not previously been shown definitively, nor has it been shown whether FcRn is the only, or even the major, IgG transporter in the neonatal mouse gut. We report here that neonatal mice homozygous for a targeted disruption of the beta 2microglobulin (beta 2m) gene, which encodes one subunit of FcRn, had reduced FcRn alpha-chain at the lumenal plasma membrane of intestinal cells. These mice had strikingly lower serum IgG levels during the first month after birth than littermates that possessed functional FcRn. Furthermore, we found by fostering mice on mothers with a different IgG allotype that all of the IgG in sera of beta 2m-/- mice was endogenous, and that none was obtained from milk. We conclude that FcRn is the only transporter of IgG from mother to young in the mouse. The onset of IgG synthesis in mice that received no milk IgG lagged behind that in siblings with normal IgG transport, suggesting that maternal IgG stimulates Ab production in the neonate. We noted no difference between the IgG concentrations in the milk of beta 2m-/- and beta 2m+/- mice, indicating that FcRn is not involved in the secretion of IgG into milk.

Influence of the polyamine, spermidine, on intestinal maturation and dietary antigen uptake in the neonatal rat.

Polyamines appear to have an important role in postnatal growth of the rat intestine. In the present study, we examined the effect of spermidine on the maturation of the intestine and on its ability to exclude macromolecules. Two litters of Sprague-Dawley rat pups were assigned to one of four experimental groups. These groups received, on Days 7, 8, and 9, either (a) saline by gavage; (b) spermidine, 0.9 mg (6 mumol) by gavage; (c) cortisone acetate, 3.5 mg i.p.; or (d) saline i.p. On Day 10, animals were fed by gavage with a mixture of bovine serum albumin (BSA; 2 mg/g) and the gamma-globulin fraction of mouse antiovalbumin (anti-OVA) antiserum (1 mg/g) and were bled 4 h later. Intestinal tissues were processed for histologic examination, sucrase determination, and identification of neonatal intestinal Fc receptor (FcRn) by Western blot. Serum immunoreactive BSA (iBSA) and mouse IgG1 and IgG2a anti-OVA antibodies were estimated by enzyme-linked immunosorbent assay. Sucrase activity was elevated in cortisone- and spermidine-treated compared to control rats. iBSA and anti-OVA were significantly reduced in cortisone-treated compared to control rats but were not diminished significantly in the spermidine-treated animals. A decrease in the neonatal intestinal Fc receptor was apparent in the spermidine-fed group; cortisone produced a large reduction in FcRn. Spermidine-fed animals showed morphologic evidence of maturation, with loss of giant vacuoles in the distal intestine; cortisone did not produce significant changes in morphology.(ABSTRACT TRUNCATED AT 250 WORDS)

Inflammatory bowel disease: diagnosis and treatment.

The primary feature of IBD which distinguishes pediatric from adult patients is the presence of or potential for significant growth failure. Recognizing this complication before permanent stunting occurs is critically important. The onset of these disorders during childhood also means that many of these children will have a lifetime of disease and therapy, during both growing as well as reproductive years. The impact of this lifelong process on the psychosocial development of the child and adolescent also needs to be recognized and addressed. Treatments must be carefully considered for their chronic adverse consequences (side effects) as well as immediate benefits. This includes the risks of early, and likely repeated, bowel resections as well as potentially mutagenic or teratogenic medications. The resurgence of tuberculosis and other chronic intestinal infections, particularly in certain high-risk pediatric populations, requires a careful consideration of the differential diagnosis of IBD in children. Finally, promising therapies must be tested in children as well as adults to develop the most effective approach to controlling IBD in pediatric patients by taking advantage of potential age-related differences in intestinal immune barrier function.

Neonatal necrotizing enterocolitis, a disease of the immature intestinal mucosal barrier.

Necrotizing enterocolitis (NEC) is an enigmatic process in that one single etiologic factor has been sought and not found. Epidemiologic studies suggest that immaturity of the host plays a very important role. This article reviews the intestinal host defense system and its immature nature early in life in animal models and humans and suggests that it is this immaturity, along with other factors, which allows for the proliferation and invasion of antigens and organism, and the subsequent development of NEC. Data are presented which support the efficacy of pharmacologic maturation of the intestinal barrier with growth factors, either prenatally or postnatally, to decrease the incidence of NEC or, potentially, to provide a more benign course for the disease.

Influence of prenatal corticosteroids on bacterial colonization in the newborn rat.

The interactions between bacteria and the host's intestinal barrier appear to be important regulators of bacterial colonization. In this study we investigated the effect of prenatal corticosteroids, known to accelerate the intestinal maturation of newborn rats, on bacterial colonization in the rat pup. Pregnant rats were treated with either cortisone acetate or normal saline on days 18-21 of gestation and were allowed to deliver spontaneously. The pups, after normal delivery, were sacrificed at different times during the first 10 days of life. The entire small intestine was removed, and each lumen was flushed to exclude nonadherent, transient organisms and homogenized. Tenfold dilutions were plated on horse-blood agar (total bacteria) and MacConkey's medium (gram-negatives). Quantitation and bacterial typification was determined after 24 h of incubation at 37 degrees C. Total bacteria and gram-negatives found in association with the mucosa were significantly lower in pups prenatally treated with steroids. These changes were not related to any changes in motility or intraluminal digestion. This suggests that the developmental condition of the host's intestinal barrier may be an important regulator of the bacterial microenvironment of the newborn small intestinal mucosa.

Immunoglobulin G binding sites on the human foetal intestine: a possible mechanism for the passive transfer of immunity from mother to infant.

In humans, the prenatal transfer of IgG from mother to foetus is facilitated by a receptor for IgG on the placenta. However, amniotic fluid contains IgG of maternal origin, and transfer of swallowed IgG into the circulation from the foetal intestine represents another potential pathway of passive immunization. In this study we assayed for a foetal intestinal IgG receptor to support the hypothesis of this alternate pathway of antibody transfer. Microvillous membrane (MVM) from small bowel of aborted foetuses (18 weeks gestation) were probed with [125I]IgG to detect specific IgG binding sites. Binding was pH dependent and was maximal at pH 6. Competitive inhibition of the binding of [125I]IgG was noted with the addition of increasing amounts of unlabelled IgG. Scatchard analysis showed one binding site with a dissociation constant of 1.58 x 10(-7), similar to that of the IgG receptor described on the suckling rat intestine. The binding of labelled IgG to the human MVM receptor was Fc mediated. These observations provide evidence for an Fc receptor on the human foetal intestine.

Diminished Clostridium difficile toxin A sensitivity in newborn rabbit ileum is associated with decreased toxin A receptor.

Human infants are relatively resistant to Clostridium difficile-associated diarrhea and colitis compared to adults. In that toxin A is the major cause of intestinal damage with this organism, we compared toxin A receptor binding and biological effects in newborn vs adult rabbit ileum. Purified toxin A (M(r) 308 kD) was labeled with tritium or biotin with full retention of biologic activity. Appearance of specific toxin A brush border (BB) binding was strongly age dependent with minimal [3H]toxin A specific binding at 2 and 5 d of life, followed by gradual increase in binding to reach adult levels at 90 d. Absence of toxin A binding sites in newborn and presence in adult rabbits was confirmed by immunohistochemical studies using biotinylated toxin A. Toxin A (50 ng to 20 micrograms/ml) inhibited protein synthesis in 90-d-old rabbit ileal loops in a dose-dependent fashion. In contrast, inhibition of protein synthesis in 5-d-old rabbit ileum occurred only at the highest toxin A doses (5 and 20 micrograms/ml) and at all doses tested was significantly less than the adult rabbit ileum. In addition, toxin A (5 micrograms/ml) caused severe mucosal damage in adult rabbit ileal explants but had no discernable morphologic effect on 5-d-old rabbit intestine. Our data indicate that newborn rabbit intestine lacks BB receptors for toxin A. The absence of the high-affinity BB receptor for toxin A in the newborn period may explain lack of biologic responsiveness to purified toxin, and the absence of disease in human infants infected with this pathogen.

Prevention of necrotizing enterocolitis in the rat with prenatal cortisone.

Cortisone acetate is known to accelerate maturation of the immature intestine. The effect of prenatal administration of cortisone acetate on the morbidity and mortality of necrotizing enterocolitis was examined in a rat pup model. Pregnant rats were administered cortisone acetate, 20 mg/100 g of body weight, or normal saline by daily IP injection from day 18-21 of gestation. Rat pups were taken from the mothers before suckling was initiated, fed a simulated rat milk formula, and subjected to daily ischemic insults to produce an animal model of necrotizing enterocolitis. Both morbidity and the mortality rates were significantly improved with prenatal cortisone treatment. Maturation of the intestinal mucosal barrier was accelerated with the cortisone treatment as measured by decreased serum concentrations of a fed antigen, ovalbumin. Aerobic bacterial colonization of the small intestine and translocation of bacteria to the liver were decreased in the pups pretreated with steroids. These changes observed in a rat model of necrotizing enterocolitis may explain the decreased incidence of necrotizing enterocolitis in human infants born to mothers who received corticosteroids late in gestation.

Uptake and transport of epidermal growth factor by the small intestinal epithelium of the fetal rat.

Epidermal growth factor may play an important part in postnatal gastrointestinal development. However, little is known of its role prenatally. The aim of this study was to detect epidermal growth factor in amniotic fluid and to study its uptake and transfer across the epithelium of fetal rat small intestine. Anesthetized 20-day gestation rats underwent caesarean section. Three fetuses were exteriorized, their abdomens were opened, and ligated loops of proximal and distal small intestine were infused with 100 micrograms epidermal growth factor. Infused segments were removed 30 min later and processed for electron microscopy, and tissue was embedded in LR gold resin. Sections were treated with rabbit anti-rat epidermal growth factor antibody, followed by 5 nm gold-labeled goat anti-rabbit immunoglobulin before staining. Epidermal growth factor was measured in amniotic fluid by radioimmunoassay. In the proximal and distal small intestine epidermal growth factor was found membrane-associated along the luminal surface of microvilli, within apical invaginations and endosomal compartments, free from the membrane in multivesicular bodies, within large clear vesicles, basal vesicles and in association with the basolateral membrane and beyond. Epidermal growth factor was found in amniotic fluid at a concentration of 0.38 +/- 0.07 (SD) ng/ml. This study shows that epidermal growth factor is present in amniotic fluid and is transported across the epithelium of fetal rats by an endocytotic process in both the upper and lower small intestine. It is likely that amniotic fluid epidermal growth factor plays a part in intestinal mucosal development, and may be active systemically after transepithelial passage in utero.

Host defense development in gut and related disorders.

The gastrointestinal tract serves as an important interface between ingested elements from the external environment and the internal milieu of the person. Maturation of this intestinal barrier appears to occur along with the normal development of other organ systems. Evidence is presented for this maturational process, and the relation of this immature barrier to certain disease states seen in the neonatal period (e.g., infectious diarrhea, necrotizing enterocolitis, and allergic disease) is discussed.

Structural and functional maturation of rat gastrointestinal barrier with thyroxine.

It has been noted that the closure of the intestinal barrier to immunoglobulins is a normal maturational process in the rat. It has also been noted that the microvillus membrane (MVM) of newborn animals differs from adult MVM. The purpose of this study is to document whether thyroid hormone can induce closure in vivo in the rat and to relate this effect of thyroxine to the structural and functional maturation of the intestinal MVM. To assess closure, 2-wk-old rats were fed a rat immunoglobulin G (IgG), and serum antibody binding activity was measured 4 h later. The antibody binding activity of treated animals (T) was 1.5-2 times less than that of controls (C) (P less than 0.001), indicating that thyroxine stimulates closure. The MVM similarly showed signs of maturation. Structural maturation was demonstrated by the lower fluidity of the thyroid-treated animals' membranes. Under the influence of thyroxine, the number of receptors on the MVM for IgG had decreased [2.8 X 10(-7) M (C) vs. 1.7 X 10(-7) M (T)], while the Ka remained the same, demonstrating the functional maturation of the MVM. In conclusion, thyroid hormone can induce both structural and functional maturation of the intestinal MVM and can enhance the intestinal mucosal barrier by decreasing the penetration of antibodies.

Development of intestinal mucosal barrier function to antigens and bacterial toxins.

In this paper, we have tried to provide evidence for the association between the microvillus membrane immaturity of young infants/animals and the enhanced attachment and uptake of intestinal antigens and toxins. Finally, we have reviewed observations that growth factors that alter the newborn MVM composition towards maturity may also affect the handling of antigens by the intestinal surface.

Effects of cholic acid infusion in fetal lambs.

The effects of prolonged intravenous infusions of cholic acid into fetal lambs are described in this study. The ewes (n = 10, 11 fetuses) were operated on at 114 days of gestation (term = 150 days) by placing plastic catheters in maternal and fetal vessels and in the amniotic cavity. Gestational ages were confirmed after delivery by radiographic examination of the ossification centers of the fetal legs. Infusions of cholic acid (1.6 mumoles/min-1) started 8 to 10 days after surgery in 5 fetuses (including one twin). The remaining 6 fetuses (also including one twin) were infused with 5% dextrose in water. Total plasma bile acids at the beginning of the experiment were similar in both groups (23.8 +/- 6.6 vs. 24.3 +/- 5.7 microM). No significant changes in fetal heart rate, blood pressure, blood gases or pH were detected during the infusion. Meconium-stained amniotic fluid was observed during the third day of infusion in all the fetuses infused with cholic acid and in one control fetus. Fetuses infused with cholic acid were delivered alive 19-26 days before term. The concentration of plasma bile acids in the experimental group at delivery was 829 +/- 305 microM, i.e. significantly higher than that of the control group (24.4 +/- 5.7 microM). Control fetuses (except one twin) were delivered at term. We concluded that cholic acid, even at the high dose infused, is neither lethal nor severely harmful for the fetus. Meconium passage of the fetuses infused with cholic acid, in our experiment, appeared to be related to the stimulatory effect of cholic acid on fetal colonic motility rather than to fetal hypoxia.(ABSTRACT TRUNCATED AT 250 WORDS)

Maximal response to oxytocin of the isolated myometrium from pregnant patients with intrahepatic cholestasis.

Prematurity and fetal death are common complications in patients with cholestasis of pregnancy. Both conditions appear to be associated with abnormal patterns of uterine activity. We studied the oxytocin-induced contractile activity in uterine strips taken from patients with cholestasis of pregnancy (n = 6) and from women with normal pregnancy (n = 6). Contractile activity of the myometrium in response to oxytocin was significantly higher in patients with cholestasis of pregnancy than in normally pregnant patients, at doses of 10(-6), 10(-4), and 10(-2) M. We found that there is a greater maximal response to oxytocin in strips of myometrium from patients with cholestasis of pregnancy than from normally pregnant patients.