Nippostrongylus brasiliensis or Subjected

Transcription

1 GASTROENTEROLOGY 77: , 1979 Intestinal Uptake of Macromolecules. VI. Uptake of Protein Antigen In Vivo in Normal Rats and in Rats Infected with Nippostrongylus brasiliensis or Subjected to Mild Systemic Anaphylaxis KURT J. BLOCH, DONALD B. BLOCH, MARILYN STEARNS, and W. ALLAN WALKER The Departments of Medicine and Pediatrics, Harvard Medical School, the Clinical Immunology and Allergy Units of the Medical Services and the Pediatric Gastrointestinal Unit, Massachusetts General Hospital, Boston, Massachusetts Adult Sprague-Dawley rats weighing -250 g were fed bovine serum albumin and sodium bicarbonate by gavage. Serum was obtained at intervals after feeding and tested for immunoreactive bovine serum albumin by radioimmunoassay. Nanogram amounts of immunoreactive bovine serum albumin were detected in serum; peak values were obtained after 4 and 6 hr. The influence of intestinal inflammation on protein uptake was examined in two model systems. Infection of rats with Nippostrongylus brasiliensis was accompanied by partial villous atrophy in the intestinal segments harboring adult worms and mild systemic anaphylaxis in the rat was accompanied by increased intestinal vascular and mucosal permeability. Enhanced uptake of BSA was observed before and shortly after self-cure of infection and during mild systemic anaphylaxis. The molecular size of immunoreactive bovine serum albumin approximated that of the administered bovine serum albumin; no small fragments of bovine serum albumin bearing antigenic determinants were detected. Received February 12, Accepted June 18, Address requests for reprints to: Kurt J. Bloch, M.D., Clinical Immunology and Allergy Units, Massachusetts General Hospital, Fruit Street, Boston, Massachusetts This paper was presented, in part, at the Plenary Session of the American Gastroenterological Association, Las Vegas, Nevada, May This work was supported by Grants AM and AM and 5T32-AI and grants from the National and Massachusetts Chapter of the Arthritis Foundation and the 1. H. Bendit Foundation. W. A. Walker is a recipient of the Research Academic Career Development Award 1-K04-AM from the National Institute of Arthritis. Metabolism and Digestive Diseases by the American Gastroenterological Association Considerable indirect'-5 and limited direct 6 - a evidence suggests that ingested antigenically intact proteins may gain access to the systemic circulation. This evidence includes the elicitation of cutaneous anaphylactic reactions in humans and animals prepared by intradermal injection of homocytotropic antibodies followed by challenge with antigens administered into the gastrointestinal tract 1 2 and the detection of antibodies to ingested antigens in the circulation of patients with alterations of the intestinal mucosal barrier.5 3 or the absence of specific components of that barrier such as secretory IgA antibodies." With regard to the detection of antibodies to food antigen in certain patients, it remains unsettled as to whether this antibody response reflects a normal response to increased amounts of antigen absorbed or an exaggerated immune response to conventional amounts of absorbed antigen. The present communication concerns the serum levels of immunoreactive bovine serum albumin (ibsa) attained in mature rats fed BSA by gavage and the influence of pathologic alterations of the gut on the uptake of this antigen. Materials and Methods Animals Sprague-Dawley rats, both male and female, -250 g were obtained from Charles River Breeding Laboratories (Wilmington, Mass.). Rats were maintained on rat chow (Ralston Purina, Inc., St. Louis, Mo.) which. according to the manufacturer, does not contain cow's milk protein, for 6 wk before use in experiments.

2 1040 BLOCH ET AL. GASTROENTEROLOGY Vol. 77, No.5 Absorption Studies Rats fasted overnight were subjected to light ether anesthesia, a plastic cannula was carefully passed into the stomach, and 2 ml of saline solution containing 1000 mg BSA (Fraction V, Sigma Chemical Co., St. Louis, Mo.) and 25 mg of NaHC0 3 were introduced. Blood was obtained from a cut made in the tail under light ether anesthesia before and at intervals up to 24 hr after feeding. These blood samples consist of a mixture of arterial and venous blood. Preparation of Antisera and Radiolabeled Proteins Rat serum albumin (RSA) was prepared by starch block electrophoresis. 9 Solutions of RSA and crystalline BSA (5X crystallized, Sigma Chemical Co., St. Louis, Mo.) were labeled with 125J.10 Rabbits were immunized with BSA emulsified in complete Freund's adjuvant." Quantitation of Circulating Antigens The amount of ibsa present in serum was detected by a method described by Rothberg et al.6 Briefly, rabbit anti-bsa antiserum was diluted with borate buffer made 10% with respect to normal rat serum; 0.5 ml of various dilutions of rabbit anti-bsa antiserum were tested for their ability to bind 50% of -10 ng 125I-BSA. To 0.5 ml of the appropriate dilution of antiserum was added 50,ILl of rat serum containing ibsa or solutions of known BSA concentration, plus 100,ILl of 125I_BSA solution. After adjusting the total volume to 0.7 ml, the reactants were mixed and incubated at 37 C for 30 min followed by overnight incubation at 4 C. Thereafter, 0.7 ml of cold, saturated (NH')2S0.12 was added, the resulting suspension was rapidly mixed and held on ice for 30 min. The suspension was centrifuged at 3000 rpm for 20 min; the supernate was decanted, and the precipitate washed with 1.0 ml of 50% saturated (NH.)zSO. After centrifugation, the second supernate was combined with the first and the radioactivity in the combined supernate and first precipitate was determined in an automatic crystal scintillation spectrometer (Nuclear Chicago, Chicago, Ill.). The amount of radioactivity in the precipitate was corrected for the amount of radioactivity precipitated in control tubes to which no antiserum had been added. The concentration of ibsa in rat serum was determined from standard curves constructed by plotting the ratio of free to antibody-bound radioactivity as a function of the amount of unlabeled BSA added to control tubes. Induction of Gastrointestinal Lesions. Partial villous atrophy was induced in rats by infection with the nematode parasite, Nippostrongylus brasiliensis Rats were injected with 3000 larvae; the influence of infection on the uptake of BSA was tested at intervals specified in Results. After removal of the last blood sample at 24 hr, rats were killed by ether anesthesia, and histologic sections of the jejunum were prepared. Morphologic alterations were scored according to the method of Schenk and Klipstein.15 The method of maintaining the life-cycle of the parasite in the laboratory has been described.16 Systemic Anaphylaxis For the induction of mild systemic anaphylaxis, rats were injected intravenously with 25,ILl of worm extract l6 on day 40 of a primary infection with N. brasiliensis. Changes in intestinal permeability were assessed according to the method of Byars and Ferraresi. 17 This method was modified by substituting 125I_RSA for 125I_BSA and by measuring trichloroacetic acid-precipitable radioactivity in rinse fluid from the duodenum and the first and second halves of the small intestine. The radioactive counts remaining in the corresponding segments of intestine were also measured. Gel Filtration A pool of rat sera containing ibsa was subjected to gel filtration on a column of Sephadex G-loo (Pharmacia, Piscataway, N.J.) measuring 2.5 cm X 40 cm; 5.0-ml fractions were collected. The eluates in two 5-ml samples between fractions 10 and 20 and in 10 5-ml samples between fractions 20 and 80 were combined (Figure 4). The combined fractions were lyophilized and reconstituted in 1 ml of borate buffer. Combined fractions 10 to 20 were dialyzed before lyophilization; fractions 21 to 80 were not dialyzed to avoid the possible loss of low molecular weight fragments of BSA. The protein and ibsa content of reconstituted fractions was determined. The elution profile of freshly dialyzed 125I_BSA and of 0.6 ml of normal rat serum to which 5 mg of 3,5-diiodotyrosine. 2H20 (mol wt 469) had been added, was determined using the same column. Results Detection of Immunoreactive Bovine Serum Albumin in Serum of Normal Adult Rats Fed BSA by Gavage Preliminary experiments indicated that rats had to be maintained on a diet devoid of BSA for 6 wk before their use in these experiments. This interval was apparently required to dissipate the effect of prior oral immunization resulting from exposure to BSA in the chow used by the breeder.18 After 6 wk, rats were fed by gavage with 100 or 1000 mg BSA, with or without 25 mg NaHC0 3, in 2 ml saline solution. None of four rats fed 100 mg BSA without NaHC0 3 had detectable ibsa in serum obtained 1-24 hr after feeding. Three of four rats fed 100 mg BSA with NaHC0 3 had detectable ibsa; amounts ranged from approximately 0.06 JLg/ml of serum at 1-6 hr to 0.03 JLg/ml at 24 hr. Of four rats fed 1000 mg BSA without NaHC0 3, two had only trace amounts of ibsa in serum and two showed amounts of ibsa similar to those recorded in Figure 1. In contrast, all

3 November 1979 INTESTINAL UPTAKE OF PROTEIN ANTIGEN 1041 nine rats fed 1000 mg BSA with NaHCOa had measurable amounts of ibsa in serum (Figure 1); peak values were obtained at 4 and 6 hr after feeding. The wide range of values attests to the differences in uptake of BSA by different animals. Since feeding of 1000 mg BSA with NaHCOa consistently resulted in the appearance in serum of measurable amounts of ibsa, this dose of protein and buffer was used throughout the following experiments. II1fluence of Gastrointestinal Tract Infection on Uptake of BSA Infection of rats with the nematode N. brasiliensis produces partial villous atrophy in segments of the small intestine harboring the mature worms.1a,14 This lesion is characterized by a reduced villous height to crypt depth ratio and by lymphoid cell infiltration. Intestinal morphology returns to normal following self_curei4,19 which occurs in our rat strain by day 17 of a primary infection.20 In preliminary experiments, we compared the uptake of BSA in four uninfected rats and four rats infected 8 days earlier with 3000 N. brasiliensis larvae. At nearly all intervals during the first 6 hr after feeding, the serum levels of ibsa were greater in infected compared to uninfected rats (Figure 2). The influence of infection on uptake of BSA was further examined in a single group of 24 rats. Twenty of these rats were infected; four served as uninfected controls. Four infected rats were tested on day 12, 15, 19, 29, and 39. The four uninfected rats were tested on day 18. The amount of ibsa in all sera obtained 2 hr after feeding 1000 mg BSA was determined in a single radioimmunoassay (Figure 3). Of the 12 infected rats tested on days 12, 15, and 19, values for serum ibsa exceeded those of the uninfected rats in eight animals. After self-cure, the ibsa values (days 29 and 39) approximated those of the uninfected rats. The jejunum of the rats was examined histologically, and the morphologic alterations were scored.15 The correlation between the numerical score of the morphologic alterations and the concentration of serum ibsa for the entire infected group had a coefficient(r) of 0.6 (P < 0.01). Influence of Mild Systemic Anaphylaxis on Uptake of BSA Mild systemic anaphylaxis was induced on day 40 of a primary infection with N. brasiliensis by the i.v. injection of 25 JLl or worm extract. At this time, the rats possess a population of mast cells sensitized by IgE anti-worm antibodies,20 but have undergone healing of the intestinal lesions accompanying primary infection (see above) and would, in the absence of worm antigen challenge, be expected to show a normal uptake of BSA !:tl HOURS Figure 1, Mean amounts of ibsa (± 1 SD) detected in serum obtained from nine adult rats at various intervals after feeding by gavage with 1000 mg BSA and 25 mg NaHCOa To document the changes in vascular21 and mucosaf2 permeability which accompany systemic anaphylaxis, rats were injected intravenously with 125I_RSA before the induction of systemic anaphylaxis. One hour after challenge with worm extract, the retention of radioactive counts in segments of the intestine and the amount of protein-bound radioactivity present in the lumen was determined. In comparison to four saline-injected rats, four antigenchallenged rats showed mean values of 2.2, 1.8, and 2.2 times the radioactive counts present in the walls of the duodenum and first and second halves of the small intestine. Similarly, intestinal secretions obtained from the corresponding gut segments of the infected rats challenged with antigen showed mean values of protein-bound radioactivity which were 1.4 1,2 o>--d Uninfecled rals 1,0 0,8 0,6 0,4 0,2 I I I I I - N.brasiliensis infected rals I HOURS Figure 2. Amounts of ibsa detected in serum obtained from four uninfected and four N. brasiliensis infected rats at various intervals after feeding by gavage with 1000 mg BSA and 25 mg NaHCOa. Feeding was performed on day 8 of a primary infection.

4 1042 BLOCH ET AL. GASTROENTEROLOGY Vol. 77, No I:l::I " I-,,8,' J. 0 I I I I I DAYS Figure 3. Amounts of ibsa detected in serum obtained 2 hr after feeding rats by gavage with 1000 mg BSA and 25 mg NaHC0 3 Four uninfected rats (open circle) were tested on day 18; 20 rats infected with N. brasiliensis were tested in groups of four animals on days 12, 15, 19, 29, and 39. Increased levels of serum ibsa were observed before and for a short time after self-cure which occurs on day 17 in our rat strain. hr after gavage with 1000 mg BSA was assayed before and after dialysis (molecular size cut-off of tubing -12,000) against phosphate-buffered saline. There was no difference in ibsa detected in these paired samples. Pooled serum containing ibsa was subjected to gel filtration on a Sephadex G-l00 colume (Figure 4). The elution profile of ibsa resembled that of 125I_BSA applied to the same column; no additional ibsa was detected in factions eluted after the bulk of serum proteins (tubes 20-80). Thus small fragments bearing BSA antigenic determinants were not detected in the present experiments. Discussion Before this study, a few other attempts had been made to quantitate the uptake of unlabeled, intact proteins or large fragments of proteins into the systemic circulation of adult animals Our finding of a mean of 0.17-JLg immunoreactive BSA per ml serum in normal adult Sprague-Dawley rats of specified weight fed 6 hr earlier with 1000 mg BSA and 25 mg NaHC0 3 (Figure 1), approximates the values reported. In the present experiments, immunoreactive BSA appeared to have a molecular size similar to that of intact BSA. There was no los!! of ibsa after dialysis of serum in tubing with a pore size ap- 7.5, 6.8, and 11.3 times those of intestinal secretions obtained from saline-injected controls. In subsequent experiments, rats were fed by gavage with 1000 mg BSA on day 40 of a primary infection. One hour later, four rats were challenged with worm extract and four with saline administered intravenously. All animals were exsanguinated by cardiac puncture 2 hr later, and the amount of ibsa present in serum was determined. Serum obtained from rats subjected to mild systemic anaphylaxis contained greater amounts of ibsa than did serum from controls (Table 1). CPM (xio 3 2 Molecular Size of Immunoreactive BSA To test the possibility that small fragments bearing BSA antigenic determinants were absorbed, pooled serum obtained from normal rats, 1, 3, and 6 Table 1. Effect of Mild Systemic Anaphylaxis on Uptake of BSA N. brasiliensis infected rats challenged witha Saline Worm extract Amount of ibsa detected in serum at 2 hr (,tg/ml) 0.005, 0.005, 0.02, ,6,12,40 a All rats received 1000 mg BSA and 25 mg NaHC0 3 by gavage 1 hr before challenge. Challenge was carried out on day 40 of a primary infection. ibsa (jjg/ml) 2 OL-L--- I I I I I so FRACTION NUMBER Figure 4. Sephadex G-100 gel filtration profile of a mixture of normal rat serum and 2,3-diiodotyrosine. 2H20. [Peak concentration of the latter was in fraction 75 (panel a)]; freshly dialyzed 125I_BSA (panel b) and ibsa contained in a pool of sera obtained from rats fed BSA fly gavage (panel c). The 125I_BSA marker was applied to the column after the procedure shown in panel c. Immunoreactive BSA appears to be similar in molecular size to the labeled BSA; no small fragments of ibsa were detected.

5 November 1979 INTESTINAL UPTAKE OF PROTEIN ANTIGEN 1043 proximating 12,000 daltons. On Sephadex G-loo gel filtration, ibsa was detected in the same fractions as the 125I_BSA marker, but not in fractions containing proteins lower in molecular size (Figure 4); identical findings were reported by Thomas and Parrott.23 We designated the BSA detected in serum as ibsa because, despite the apparent similarity in molecular size between administered BSA and ibsa, the latter may, due to proteolysis, be deficient in the number of antigenic determinants per molecule. Wright and Rothberg25 found that after enzymatic digestion of BSA in vitro, BSA molecules were obtained which had a molecular size similar to native BSA, but which were markedly reduced ( fold) in ability to inhibit binding of l3li-bsa to anti BSA antibody. Thus, the radioimmunoassay employed in the present experiments may also underestimate the actual amount of BSA-derived molecules present in serum of rats fed native BSA and the assay is, of course, unable to detect BSA-derived molecules lacking all (surface?) antigenic determinants. The high molecular weight proteins obtained by enzymatic digestion of BSA in vitro eluted from Sephadex G-l00 slightly ahead of native BSA,25 suggesting that alteration of the molecule had occurred during digestion. This explanation may also account for the small difference in the gel filtration profile of ibsa and 125I_BSA observed in our experiments (Figure 4). The development of a spruelike lesion during the course of infection of rats with the nematode parasite N. brasiliensis has been well documented.13,14 We showed a significant correlation between the extent of morphologic alteration of the infected jejunum and the concentration of ibsa achieved in infected rats fed 1000 mg BSA. Enhanced uptake of BSA may be related to gross breaks in the integrity of the intestinal wall in the vicinity of collections of worms or to more subtle alterations such as the breakdown in junctional complexes between epithelial cells lining the villi (reviewed in Reference 26). Enhanced uptake may also be facilitated by stasis of intestinal fluids in the jejunum of infected rats and by the dilution of pancreatic enzymes by the increased fluid found in the jejunum.26 In addition, N. brasiliensis may release an inhibitor of proteolytic enzymes into the intestinal fluid. 26 A marked increase in permeability of the intestinal vasculature in the vicinity of the worms,21 may also contribute to the systemic uptake of protein antigens that have entered the mucosa. Thus parasitic infection may lead to enhanced uptake of BSA by interferring with its luminal breakdown and by providing abnormal channels for uptake from the lumen into the interstitial spaces of the lamina propria and from there into the systemic circulation. The gut is prominently involved during systemic anaphylaxis in the rae 7 ; anaphylaxis produces separation by edema fluid of the intestinal epithelium from the underlying vascular core of the villi. 28 In our experiments, mild systemic anaphylaxis was accompanied by the accumulation of radiolabeled rat serum albumin in the wall of the small intestine and by the transfer of labeled albumin or large fragments of albumin into the intestinal lumen. These observations reflect the enhanced vascular and mucosal permeability which accompanies systemic anaphylaxis.21,22 The leakage of macromolecules into the intestinal lumen probably occurs, at least in part, via the interepithelial spaces.22 It seems likely that during systemic anaphylaxis, the enhanced uptake of BSA from the lumen into the interstitial spaces of the mucosa and subsequently into the systemic circulation occurs via the same channels. The significance of the results 0 btained in the present experiments might be questioned on the basis of the non physiologic amounts of antigen employed and on the special conditions required for the test animals, i.e., abstension from exposure to the test antigen for several weeks before challenge. With respect to the former it should be noted that ibsa was detectable in serum of some rats fed 100 mg BSA together with NaHC03. Under physiologic circumstances, the presence of other ingested foods might decrease the amount of anyone protein antigen required for successful demonstration of uptake; this possibility remains to be tested. With respect to the requirement that animals had to be removed from exposure to the test antigen for several weeks before challenge, presumably in order to allow for the waning of a local immune response to the test antigen, it should be noted that this procedure was followed in order to assure consistent uptake of BSA. Some rats which had been maintained on a diet containing cow's milk protein for many weeks showed considerable uptake of BSA (data not shown), suggesting that if a gastrointestinal immune response was directed against BSA in these animals, then the response was ineffective in excluding all of the protein challenge. In our previous studies it was found that while immunization reduced the uptake of protein antigen by everted gut sacs in vitro, inhibition of uptake was never complete Similarly, Andre et al. observed only partial inhibition of uptake in immunized rats challenged in vivo with a large dose of antigen.32 Thus it may be suggested that small amounts of protein antigen may enter the systemic circulation in normal animals, that larger amounts enter the circulation in rats subjected to intestinal inflammation and that some antigen may enter the circulation despite gastrointestinal immunity to the antigen. Possibly, intestinal inflammation in humans is also accompanied by enhanced uptake of antigens. A normal antibody response to the in-

6 1044 BLOCH ET AL. GASTROENTEROLOGY Vol. 77, No, 5 creased amounts of antigen in the circulation may account for the high incidence of antibodies to food antigens in such patients. 3,5 The biologic functions of proteins that are absorbed antigenically intact may include the induction of an immune response in the lamina propria, mesenteric or remote lymphnid tissue33,6; the induction or maintenance of systemic tolerance to ingested antigens 7,34; or the elicitation of tissue injury by anaphylactic mechanisms or by immune complexes 35 formed in the lamina propria or remote sites in systemically immunized animals exposed to antigens via the gastrointestinal tract. Additional studies are required to define the precise conditions which determine these and other consequences of enteric exposure to protein antigens. References 1. Walzer M: Absorption of allergens. J Allergy 13: , Bernstein ID, Ovary Z: Absorption of antigens from the gastrointestinal tract. Int Arch Allergy Appl Immunol 33: , Taylor KB, Truelove SC, Wright R: Serologic reactions to gluten and cow's milk protein in gastrointestinal disease. Gastroenterology 46:99-108,1964 4, Buckley RH, Dees SC: Correlation of milk precipitins with IgA deficiency. N Engl J Med 281: , Falchuk KR, Isselbacher KJ: Circulating antibodies to bovine albumin in ulcerative colitis and Crohn's disease. Characterization of the antibody response, Gastroenterology 70:5-8, Rothberg RM, Kraft SC, Farr RS: Similarities between rabbit antibodies produced following ingestion of bovine serum albumin and following parenteral immunization. J Immunol 98: , Thomas HC, Parrott DMV: The induction of tolerance to a soluble protein antigen by oral administration. Immunology 27: , Cunningham-Rundles C, Brandeis WE, Good RA, et al: Milk precipitins, circulating immune complexes, and IgA deficiency. Proc Nat! Acad Sci USA. 75: , Kunkel HG, Trautman R: Zone electrophoresis in various types of supporting media. In: Electrophoresis: Theory, Methods and Applications. Edited by M Bier, London, Academic Press, Inc., 1959, p Greenwood FC, Hunter WM, Glover JS: The preparation of l31i-labelled growth hormone of high specific radioactivity, Biochem J 89: , , Garvey JS, Cremer NE, Sussdorf DH: In: Methods in Immunology, Third Edition. Reading, Massachusetts, W.A. Benjamin, Inc., 1977, p , Minden P, Farr RS: The ammonium sulfate method to measure antigen-binding capacity. In: Handbook of Experimental Immunology, Second Edition. Edited by DM Weir. Oxford, Blackwell Scientific Publications, 1973, p , Symons LEA: Kinetics of the epithelial cells, and morphology of villi and crypts in the jejunum of the rat infected by the nematode Nippostrongylus brasiliensis, Gastroenterology 49: , Ferguson A, Jarrett EEE: Hypersensitivity reactions in the small intestine. I. Thymus dependence of experimental partial villous atrophy. Gut 16: , , Schenk EA, Klipstein FA: A protocol for the evaluation of small bowel biopsies. Am J Clin Nutr 25: , Wilson RJM, Bloch KJ: Homocytotropic antibody response in the rat infected with the nematode, Nippostrongylus brasiliensis. II. Characteristics of the immune response, J Immunol 100: , Byars NE, Ferraresi RW: Intestinal anaphylaxis in the rat as a model of food allergy. Clin Exp ImmunoI25: , Dolezel J, Strauss H, Bienenstock J: Antigen in feed as a cause of antibody in unimmunized animals. Int Arch Allergy Appl ImmunoI40: , Taliaferro WH, Sarles MP: The cellular reactions in the skin, lungs and intestine of normal and immune rats after infection with Nippostrongylus muris. J Infect Dis 64: , Bloch, KJ, Cygan RW, Waltin J: The IgE system and parasitism: role of mast cells, IgE and other antibodies in host response to primary infection with Nippostrongylus brasiliensis, In: The Biological Role of the Immunoglobulin E System. Edited by K Ishizaka, DH Dayton, Jr, U.S. Dept. of Health, Education and Welfare, Public Health Sevice, National Institutes of Health, 1972, p , Urquhart GM, Mulligan W, Eadie RM, et al: Immunological studies on Nippostrongylus brasiliensis infection in the rat: the role of local anaphylaxis. Exp ParasitoI17: , , Murray M, Jarrett WFH, Jennings FW: Mast cells and macromolecular leak in intestinal immunological reactions, The influence of sex of rats infected with Nippostrongylus brasiliensis. Immunology 21:17-31, , Thomas HC, Parrott DMV: The induction of tolerance to a soluble proptein antigen by oral administration. Immunology 27: , Hanson DG, Vaz NM, Mai LCS, et al: Inhibition of specific immune responses by feeding protein antigens, Int Arch Allergy Appl Immunol 55: , Wright RN, Rothberg RM: The reaction of pepsin and pepsintrypsin digestion products of bovine serum albumin with antisera from rabbits ingesting this protein. J Immunol 107: , , Ogilvie BM, Jones VE: Nippostrongylus brasiliensis: A review of immunity and the host/parasite relationship in the rat. Exp ParasitoI29: , West GB: Some factors involved in anaphylactic shock. Int Arch Allergy Appl ImmunoI15: , Barth EEE, Jarrett WFH, Urquhart GM: Studies on the mechanism of the self-cure reaction in rats infected with Nippostrongylus brasiliensis. Immunology 10: , Walker WA, Isselbacher KJ, Bloch KJ: Intestinal uptake of macromolecules: effect of oral immunization. Science 177: , Walker WA, Isselbacher KJ, Bloch KJ: Intestinal uptake of macromolecules. II. Effect of parenteral immunization, J Immunoll11: , Walker WA, Wu M, Isselbacher KJ, et al: Intestinal uptake of macromolecules. III. Studies on the mechanism by which immunization interferes with antigen uptake. J Immunol 115: , Andre C, Lambert R. Bazin H, et al: Interference of oral immunization with the intestinal absorption of heterologous albumin. Eur J ImmunoI4: , Crabbe PA, Nash DR, Bazin H, et al: Antibodies of the IgA type in intestinal plasma cells of germ-free mice after oral or parenteral immunization with ferritin. J Exp Med 130: , Kagnoff MF: Induction and paralysis: a conceptio!'.al framework from which to examine the intestinal immune system, Gastroenterology 66: , Carr RI, Rabideau D, Durante D, et al: Studies on immune responses to oral ferritin (abstr), J Allergy Clin Immunol 63:138, 1979