- Email: [email protected]
Dietary protein antigenemia in humoral immunodeficiency
Dietary ProteinAntigenemiain Humorallmmunodeficiency Correlation with Splenomegaly
CHARLOTTE M.D.,
CUNNINGHAM-RUNDLES,
Ph.D.
New York, New York
RONALD I. CARR, M.D., Ph.D. Denver, Colorado
ROBERT
A. GOOD,
M.D., Ph.D.
Oklahoma City, Oklahoma
From the Memorial Sloan-Kettering Cancer Center, New York, New York, the National Jewish Hosoital. Denver, Colorado, and the Oklahoma Medical Research Foundation, Oklahoma City, Oklahoma. This work was supported by United States Public Health Service Grants Al-18589, CA-19267, and CA-29502, American Cancer Society Grant ACS-M-245, and a grant from the Zelda R. Weintraub Cancer Fund. Requests for reprints should be addressed to Dr. Charlotte CunninghamRundles, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021. Manuscript accepted July 7, 1983.
Enhanced gastrointestinal absorption of dietary substances is an important feature of normal neonatal life that also exists in particular disease states such as selective IgA deficiency and atopic allergy. In these studies, it is shown that patients with hypogammaglobulinemia have increased absorption of dietary bovine antigens and that most patients have large amounts of these proteins present in the serum even after an overnight fast. The amounts of such proteins were found to be correlated with spleen size and/or peripheral lymphoid hypertrophy. Interestingly, three patients with X-linked agammaglobulinemia did not have detectable amounts of these proteins in the serum nor did they have splenomegaly or lymphadenopathy. It is speculated that hypogammaglobulinemic patients have a specific gastrointestinal mucosal lesion that permits the chronic excessive absorption of dietary antigens and may result in lymphoid hypertrophy. The gastrointestinal absorption of large amounts of antigenically intact dietary protein is generally regarded as a phenomenon characteristic of neonatal life. The epithelial cells of the immature gastrointestinal tract possess an extensive capacity of pinocytosis that results in a substantially greater absorption of macromolecules than can be normally demonstrated for older infants and adults [ 1,2]. However, other factors besides immaturity can result in excessive gastrointestinal absorption of luminal antigens-examples of these factors are a variety of mucosal lesions [3], IgE-mediated food allergy [4], eczema [s], and, as we have previously shown, selective IgA deficiency [6]. Because IgG and IgM antibody responses are usually normal in these conditions, this excess absorption can result in the periodic circulation of antigen-antibody complexes [3,4,6]. In the present studies, we have quantitated the amounts of two bovine food antigens that are present in the serum of patients with panhypogammaglobulinemia and agammaglobulinemia in order to determine whether a dual lack of secretory IgA and systemic antibody production could result
PATIENTS
in chronic
ANb
antigenemia.
METHODS
Patient and Control Serum Samples. The serum samples of 17 patients with panhypogammaglobulinemia (common varied immunodeficiency), three patients with X-linked (Bruton-type) agammaglobulinemia, and 20 normal subjects (age range 20 to 40) were studied after a 16-hour overnight fast. All patients were completely evaluated by radiographic, medical, and immunologic testing, and each had a complete physical examination. Spleen size was judged by percussion, palpation, and, when clinically indicated, technetium 99 spleen scanning. The hypo- and agammaglobulinemic patients
February 1984
The American Journal of Medicine
Volume 78
181
ANTIGENEMIA
IN IMMUNODEFICIENCY
DISEASE-CUNNINGHAM-RUNDLES
ET AL
centrifuged again, and counted. The results obtained with the normal rabbit serum containing known amounts of added antigen were used to plot standard inhibition curves, and the amounts of antigen present in the test serum samples were determined by comparison with these curves. Antigens for this assay were cow’s milk casein and bovine gammaglobulin (purchased from Pentex, Kankakee, Illinois, and used without further purification). These proteins were labeled with iodine 125 using the lactoperoxidase procedure [8] to a specific activity of about 0.4 pCi/pg. The labeled proteins were stored in small aliquots at -7O’C and were stable with a trichloroacetic acid precipitability of greater than 90 percent for at least three months. This assay detects as little casein or bovine gammaglobulin as 40 ng/ml of serum. The serum
samples of the 20 normal donors were similarly analyzed for comparison to the serum samples of these patients. Figure 1. Patient 3 (with hypogammaglobulinemia) was given 100 ml of milk to drink; serum collected at 30-minute intervals was tested for the presence of bovine K-case/n using a specific antiserum (ceder well). were treated with intramuscular gammaglobulin, 50 mg/kg every two weeks, or intravenous gammaglobulin (pH 4.0 treated immunoglobulin; Swiss Red Cross, Bern, Switzerland), 150 to 300 mgfkg every two to four weeks. After fasting overnight, eight patients with hypogammaglobulinemia, three with agammaglobuiinemia, and four normal subjects were given 100 ml of cow’s milk to drink to test for the appearance of milk antigens in the serum at 30-minute intervals for two to four hours [ 61. Detection of Dietary Proteins In Serum. Radioimmunoassay: Serum samples were tested for the presence of bovine gammaglobulin and bovine casein by radioimmunoassay [7] in which the amount of dietary protein in each serum sample was calculated from the inhibition produced by the patient’s serum on the binding of rabbit anti-casein or anti-bovine gammaglobulin to iodine 125-labeled casein or iodine 125labeled bovine gammaglobulin [7]. For this assay, rabbit anti-casein and anti-bovine gammaglobulin were titered in 10 percent normal rabbit serum in 0.25 M sodium borate-boric acid, 0.075 M saline buffer, pH 8.3, to determine the dilution that specifically bound 50 percent of the appropriate antigen. The dilution of antiserum giving 50 percent binding was mixed with 10 percent normal rabbit serum containing known amounts of casein or bovine gammaglobulin of 1:lO test serum. These tubes were mixed and incubated at 4% for 24 hours. Then the same amount of iodine 125~labeled casein or bovine gammaglobulin was added followed by two volumes of 20 percent polyethylene glycol 6000 (Sigma Chemical, St. Louis, Missouri) in borate buffer (final polyethylene glycol concentration 10 percent). For bovine gammaglobulin, 15 percent polyethylene glycol was used (final concentration 7.5 percent). The tubes were vortexed, incubated at 4% overnight, and then spun at 3,000 rpm X 30 minutes in a Sorvail centrifuge (Ivan Sorvali, Newton, Connecticut). The precipitates were washed once with 1 ml of the appropriate polyethylene glycol solution (10 percent or 7.5 percent),
182
February 1984
Double diffusion in agar: Blood samples obtained from eight of the patients with hypogammaglobulinemia, three with agammaglobulinemia, and four normal subjects just before and at 30-minute intervals for two to three hours after ingestion of 100 ml of cow’s milk were allowed to clot at room temperature for 30 minutes, and aliquots of serum were then immediately tested in micro-double diffusion in agar for the presence of casein [6]. RESULTS
This study was initiated because we observed that not only did four of eight patients with hypogammaglobulinemia have bovine casein detectable in their serum samples by agar diffusion after milk ingestion, but also each of the four had easily detectable antigen even before milk ingestion. An example of this is shown in Flgure 1, in which casein is clearly present in the serum before and at all intervals after milk ingestion. In order to assess the range of levels of food antigens that might be present in the serum of these patients, we quantitated the level of casein and bovine gammaglobulin in the serum samples of 17 patients with hypogammaglobulinemia, and three with X-linked agammaglobuiinemia (Table I). All patients with hypogammaglobulinemia were found to have detectable casein (range 80 to 1,600 ng/ml of serum) and 7 (41 percent) had detectable bovine gammaglobulin (range 250 to 800 ng/ml). In contrast, of the three patients with agammaglobulinemia, only one had trace amounts of casein, and none had detectable bovine gammaglobulin. The serum samples of 20 normal subjects did not contain detectable casein or bovine gammaglobulin in this assay. The serum samples of eight patients with hypogammaglobulinemia and three patients with Xlinked agammaglobulinemia and the normal serum samples were then tested to determine the levels of bovine antigens present after milk ingestion. The serum samples of all eight patients with hypogammagiobulinemia were found to have substantial amounts of casein and bovine gammaglobulin before the milk test;
The American Journal of Medlclne Volume 78
ANTIGENEMIA
TABLE I
IN IMMUNODEFICIENCY
DISEASE-CUNNINGHAM-RUNDLES
ET AL
Casein and Bovine GammaglobulinLevels Casein
Patient Number
(ngiml)
Bovine Gammaglobulin@g/ml)
Hypogammaglobulinemia 1
1,600
650
+tt
1,200 1,280 1,080 1,000 980 980 970 960 960 930 900 900 400 400 300
500 300 250 <40 700 <40 <40 <40 <40 260 <40 800 <40 <40 <40
0 ttt t+ tt tt 0 0 0 t t 0 ++ 0 0 0
80 <40 trace <40
<40 <40 <40 <40
2 3 4 5 6 7 6 9 10 11 12 13 14 15 16 Agammaglobulinemia 17 18 19 20
Spleen: ttt = enlarged by 6 cm or more (to umbilicus or below); tt t On biopsy, benign lymphoid hyperplasia. l
agammaglobulinemia patients had very little antigen detectable at this time. After milk ingestion, the amounts of casein and bovine gammaglobulin in the serum increased for the hypogammaglobulinemia subjects, with peak levels occurring 30 to 120 minutes after ingestion. Figure 2 shows representative data for the casein concentrations found in the serum samples of two hypogammaglobulinemic patients (curves A and B) after milk ingestion. Note that the initial amount of this antigen was already 400 ng/ml in one subject, and 800 ng/ml in the other. For the X-linked agammaglobulinemic subjects, no such increases were observed (example: casein concentration, Figure 2, curve C.) The normal subjects were not found to have any detectable antigenemia under these circumstances. To ascertain what correlation the level of these foreign proteins might have to clinical status, the medical and dietary history, results of physical examination and extensive medical and immunologic analyses of each patient were scrutinized. Most particularly, we sought relationships between the presence of food proteins in the serum and dietary history, evidence of current or previous gastrointestinal disease (including frequent diarrhea, malabsorption, nodular lymphoid hyperplasia, giardiasis, liver disease, or gallbladder dysfunction), level of serum immunoglobulins, and evidence of T cell
February
Other Lymphoid Hyperlrophy
Spleen’
Cervical lymphadenopathy,+ nodular lymphoid hyperplasia Lymphoid pulmonary infiltration
Nodular lymphoid hyperplasia Cervical lymphadenopathyt
Nodular lymphoid hyperplasia
= enlarged by 2 to 6 cm; t
r
= just palpable.
1200
l. . . . ..
1000
E 1
1
0
60
30
Minutes
after
milk
90
120
ingestion
Figure 2. After drinking 100 ml of milk, pafients with hypogammaglobulinemia (curves A and B) had an increase in the amount of bovine casein in their serum, whereas a patient with agammaglobulinemia (curve C) did not.
1984
The American
Journal of Medlclne
Volume 78
183
ANTIGENEMIA
IN IMMUNODEFICIENCY
DISEASE-CUNNINGHAM-RUNDLES
defects (lymphocyte proliferation, lymphocyte surface markers analyses, thymic hormones, and leukocyte migration inhibition). Although correlations between these factors and dietary protein antigenemia were not found, both greater amounts of bovine casein and the presence of bovine gammaglobulin in the serum were strongly correlated with the presence of splenomegaly as repeatedly noted by examining physicians (p
184
February 1984
The American Journal of Medicine
ET AL
agammaglobulinemia do have less frequent gastrointestinal disease than patients with hypogammaglobulinemia [ 121, but we did not find that gastrointestinal disease could be correlated with antigenemia in the hypogammaglobulinemic patients we studied. We have previously sho.wn that 50 to 60 percent of patients with hypogammaglobulinemia have substantial T cell defects [ 13,141 that are not demonstrable in patients with X-linked agammaglobulinemia [ 151. Possibly intact cellular immune function in agammaglobulinemia is responsible for the clinical disparities and the great difference in antigen absorption that we have found between these two populations of patients. We have not, however, been able to correlate a range of T cell functions with the relative level of foreign food protein in the serum of the hypogammaglobulinemic patients, since some patients with normal T cells had large amounts of casein and bovine gammaglobulin, and others with poorly functioning T cells had very little of these proteins in their serum. Our data establish a strong statistical relationship between food protein antigenemia and lymphoid hypertrophy-particularly splenomegaly. The reasons for this association are not known. Can the antigenemia and splenomegaly be attributed to a common, unknown factor present in this disease, or is it possible that a “leaky” gastrointestinal mucosa could lead to chronic lymphoid stimulation? It is known that about 20 percent of patients with hypogammaglobulinemia have splenomegaly and/or nodular hyperplasia; cervical lymphadenopathy is observed in a smaller proportion (about 5 percent) [ 12,141. Again, patients with X-linked agammaglobulinemic differ; lymphoid hyperplasia is rarely seen in this group of patients [ 151. Another aspect of hypogammaglobulinemia (but not agammaglobulinemia) is the increased incidence of lymphoid malignancy; in various studies, it was found to affect 3.4 to 4.3 percent of patients previously diagnosed as having this condition [ 12,14,16]. Although there is no known relationship between lymphoid hypertrophy and lymphoma, we speculate that a chronic gastrointestinal leakage of luminal antigens of food, bacterial, and viral origin could be related directly or indirectly to the increased risk of lymphoma in this group of patients. Our data indicate a need for a better understanding of the gastrointestinal mucosal lesions present in immunodeficient patients. Such work may help to clarify the role of the immune system in antigen clearance. ACKNOWLEDGMENT We thank Dr. F. P. Siegal and Dr. S. Gupta for the referral of patients and the staff of the fmmunodeficiency Clinic for their support.
Volume 78
ANTIGENEMIA IN IMMUNODEFICIENCY DISEASE-CUNNINGHAM-RUNDLES ET AL
REFERENCES
4.
5.
6.
8.
Schloss OM, Worthen TW: The permeability of the gastrointestinal tract of infants to undigested protein. Am J Dis Child 1916; 11: 342. Gruskay FL, Cooke RE: The gastrointestinal absorption of unaltered protein in normal infants and in infants recovering from diarrhea. Pediatrics 1955; 16: 763. Walker WA: Antigen absorption from the small intestine and gastrointestinal disease. Symposium on gastrointestinal and liver disease. Pediatr Clin North Am 1975; 22: 731. Paginelli R, Levinsky RJ, Brostoff D, Wraith DG: Immune complexes containing food protein in normal and atopic subjects after oral challenge and effect of sodium chromoglycate on antigen absorption. Lancet 1979; I: 1270. Cunningham-Rundles C, Kirkwood EM, Ferguson A, Parrott D, Good RA: Circulating immune complexes in sera of allergic patients. Clin Res 1980; 28: 343. Cunninoham-Rundles C. Brandeis WE. Good RA. Dav NK: BoviGe antigens and the formation of circulating immune complexes in selective IgA deficiency. J Clin Invest 1979; 64: 270. Carr RJ, Wold RT, Farr RJ: Antibodies to bovine gammaglobulin and the occurrence of a BGG-like substance in systemic erythematous sera. J Allergy Clin lmmunol 1972; 50: 18. Vitetta ES, Baur S, Uhr JW: Cell surface immunoglobulin isolation and characterization of immunoglobulin from
February
9.
10.
11.
12.
13.
14.
15. 16.
1984
mouse splenic lymphocytes. J Exp Med 1971; 134: 242. Paginelli R, Levinsky RJ: Solid phase radioimmunoassay for detection of circulating food protein antigens in human sera. J lmmunol Methods 1980; 37: 333. Block KJ, Block DB, Stearns M, Walker WA: Intestinal uptake of macromolecules. IV. Uptake of protein antigen in vivo in normal rats and rats infected with Nippostrongylus brasiliensis or subjected to mild systemic anaphylaxis. Gastroenterology 1979; 77: 1039. Griswold WR: Studies of the integrity of iodinated bovine albumin after circulation in vivo. lmmunol Commun 1980; 9: 297. Asherson GL, Webster AD: Diagnosis and treatment of immunodeficiency disease. Oxford: Blackwell Scientific Publications, 1980; 22. Cunningham-Rundles S, CunninghamRundles C, Ma DI, Siegal FP, Kosloff CL, Good RA: Impaired proliferative responses to B lymphocyte activators in common varied immunodeficiency. J Clin lmmunol 1981; 1: 65. Good RA: Common variable immunodeficiency. In: Bergsma D, ed. Birth defects compendium. The National Foundation-March of Dimes. New York: AR Liss, 1979; 547. Good RA: Agammaglobulinemia, X-linked infantile. In Ref 14; 54. Spector BD, Perry GS, Good RA, Kersey JH: lmmunodeficiency diseases and malignancy. In: Twomey JJ, Good RA, eds. The immunopathology of lymphoreticular neoplasms. New York: Plenum Publishing, 1978; 203-222.
The American
Journal of Medicine
Volume 78
185
CHARLOTTE M.D.,
CUNNINGHAM-RUNDLES,
Ph.D.
New York, New York
RONALD I. CARR, M.D., Ph.D. Denver, Colorado
ROBERT
A. GOOD,
M.D., Ph.D.
Oklahoma City, Oklahoma
From the Memorial Sloan-Kettering Cancer Center, New York, New York, the National Jewish Hosoital. Denver, Colorado, and the Oklahoma Medical Research Foundation, Oklahoma City, Oklahoma. This work was supported by United States Public Health Service Grants Al-18589, CA-19267, and CA-29502, American Cancer Society Grant ACS-M-245, and a grant from the Zelda R. Weintraub Cancer Fund. Requests for reprints should be addressed to Dr. Charlotte CunninghamRundles, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021. Manuscript accepted July 7, 1983.
Enhanced gastrointestinal absorption of dietary substances is an important feature of normal neonatal life that also exists in particular disease states such as selective IgA deficiency and atopic allergy. In these studies, it is shown that patients with hypogammaglobulinemia have increased absorption of dietary bovine antigens and that most patients have large amounts of these proteins present in the serum even after an overnight fast. The amounts of such proteins were found to be correlated with spleen size and/or peripheral lymphoid hypertrophy. Interestingly, three patients with X-linked agammaglobulinemia did not have detectable amounts of these proteins in the serum nor did they have splenomegaly or lymphadenopathy. It is speculated that hypogammaglobulinemic patients have a specific gastrointestinal mucosal lesion that permits the chronic excessive absorption of dietary antigens and may result in lymphoid hypertrophy. The gastrointestinal absorption of large amounts of antigenically intact dietary protein is generally regarded as a phenomenon characteristic of neonatal life. The epithelial cells of the immature gastrointestinal tract possess an extensive capacity of pinocytosis that results in a substantially greater absorption of macromolecules than can be normally demonstrated for older infants and adults [ 1,2]. However, other factors besides immaturity can result in excessive gastrointestinal absorption of luminal antigens-examples of these factors are a variety of mucosal lesions [3], IgE-mediated food allergy [4], eczema [s], and, as we have previously shown, selective IgA deficiency [6]. Because IgG and IgM antibody responses are usually normal in these conditions, this excess absorption can result in the periodic circulation of antigen-antibody complexes [3,4,6]. In the present studies, we have quantitated the amounts of two bovine food antigens that are present in the serum of patients with panhypogammaglobulinemia and agammaglobulinemia in order to determine whether a dual lack of secretory IgA and systemic antibody production could result
PATIENTS
in chronic
ANb
antigenemia.
METHODS
Patient and Control Serum Samples. The serum samples of 17 patients with panhypogammaglobulinemia (common varied immunodeficiency), three patients with X-linked (Bruton-type) agammaglobulinemia, and 20 normal subjects (age range 20 to 40) were studied after a 16-hour overnight fast. All patients were completely evaluated by radiographic, medical, and immunologic testing, and each had a complete physical examination. Spleen size was judged by percussion, palpation, and, when clinically indicated, technetium 99 spleen scanning. The hypo- and agammaglobulinemic patients
February 1984
The American Journal of Medicine
Volume 78
181
ANTIGENEMIA
IN IMMUNODEFICIENCY
DISEASE-CUNNINGHAM-RUNDLES
ET AL
centrifuged again, and counted. The results obtained with the normal rabbit serum containing known amounts of added antigen were used to plot standard inhibition curves, and the amounts of antigen present in the test serum samples were determined by comparison with these curves. Antigens for this assay were cow’s milk casein and bovine gammaglobulin (purchased from Pentex, Kankakee, Illinois, and used without further purification). These proteins were labeled with iodine 125 using the lactoperoxidase procedure [8] to a specific activity of about 0.4 pCi/pg. The labeled proteins were stored in small aliquots at -7O’C and were stable with a trichloroacetic acid precipitability of greater than 90 percent for at least three months. This assay detects as little casein or bovine gammaglobulin as 40 ng/ml of serum. The serum
samples of the 20 normal donors were similarly analyzed for comparison to the serum samples of these patients. Figure 1. Patient 3 (with hypogammaglobulinemia) was given 100 ml of milk to drink; serum collected at 30-minute intervals was tested for the presence of bovine K-case/n using a specific antiserum (ceder well). were treated with intramuscular gammaglobulin, 50 mg/kg every two weeks, or intravenous gammaglobulin (pH 4.0 treated immunoglobulin; Swiss Red Cross, Bern, Switzerland), 150 to 300 mgfkg every two to four weeks. After fasting overnight, eight patients with hypogammaglobulinemia, three with agammaglobuiinemia, and four normal subjects were given 100 ml of cow’s milk to drink to test for the appearance of milk antigens in the serum at 30-minute intervals for two to four hours [ 61. Detection of Dietary Proteins In Serum. Radioimmunoassay: Serum samples were tested for the presence of bovine gammaglobulin and bovine casein by radioimmunoassay [7] in which the amount of dietary protein in each serum sample was calculated from the inhibition produced by the patient’s serum on the binding of rabbit anti-casein or anti-bovine gammaglobulin to iodine 125-labeled casein or iodine 125labeled bovine gammaglobulin [7]. For this assay, rabbit anti-casein and anti-bovine gammaglobulin were titered in 10 percent normal rabbit serum in 0.25 M sodium borate-boric acid, 0.075 M saline buffer, pH 8.3, to determine the dilution that specifically bound 50 percent of the appropriate antigen. The dilution of antiserum giving 50 percent binding was mixed with 10 percent normal rabbit serum containing known amounts of casein or bovine gammaglobulin of 1:lO test serum. These tubes were mixed and incubated at 4% for 24 hours. Then the same amount of iodine 125~labeled casein or bovine gammaglobulin was added followed by two volumes of 20 percent polyethylene glycol 6000 (Sigma Chemical, St. Louis, Missouri) in borate buffer (final polyethylene glycol concentration 10 percent). For bovine gammaglobulin, 15 percent polyethylene glycol was used (final concentration 7.5 percent). The tubes were vortexed, incubated at 4% overnight, and then spun at 3,000 rpm X 30 minutes in a Sorvail centrifuge (Ivan Sorvali, Newton, Connecticut). The precipitates were washed once with 1 ml of the appropriate polyethylene glycol solution (10 percent or 7.5 percent),
182
February 1984
Double diffusion in agar: Blood samples obtained from eight of the patients with hypogammaglobulinemia, three with agammaglobulinemia, and four normal subjects just before and at 30-minute intervals for two to three hours after ingestion of 100 ml of cow’s milk were allowed to clot at room temperature for 30 minutes, and aliquots of serum were then immediately tested in micro-double diffusion in agar for the presence of casein [6]. RESULTS
This study was initiated because we observed that not only did four of eight patients with hypogammaglobulinemia have bovine casein detectable in their serum samples by agar diffusion after milk ingestion, but also each of the four had easily detectable antigen even before milk ingestion. An example of this is shown in Flgure 1, in which casein is clearly present in the serum before and at all intervals after milk ingestion. In order to assess the range of levels of food antigens that might be present in the serum of these patients, we quantitated the level of casein and bovine gammaglobulin in the serum samples of 17 patients with hypogammaglobulinemia, and three with X-linked agammaglobuiinemia (Table I). All patients with hypogammaglobulinemia were found to have detectable casein (range 80 to 1,600 ng/ml of serum) and 7 (41 percent) had detectable bovine gammaglobulin (range 250 to 800 ng/ml). In contrast, of the three patients with agammaglobulinemia, only one had trace amounts of casein, and none had detectable bovine gammaglobulin. The serum samples of 20 normal subjects did not contain detectable casein or bovine gammaglobulin in this assay. The serum samples of eight patients with hypogammaglobulinemia and three patients with Xlinked agammaglobulinemia and the normal serum samples were then tested to determine the levels of bovine antigens present after milk ingestion. The serum samples of all eight patients with hypogammagiobulinemia were found to have substantial amounts of casein and bovine gammaglobulin before the milk test;
The American Journal of Medlclne Volume 78
ANTIGENEMIA
TABLE I
IN IMMUNODEFICIENCY
DISEASE-CUNNINGHAM-RUNDLES
ET AL
Casein and Bovine GammaglobulinLevels Casein
Patient Number
(ngiml)
Bovine Gammaglobulin@g/ml)
Hypogammaglobulinemia 1
1,600
650
+tt
1,200 1,280 1,080 1,000 980 980 970 960 960 930 900 900 400 400 300
500 300 250 <40 700 <40 <40 <40 <40 260 <40 800 <40 <40 <40
0 ttt t+ tt tt 0 0 0 t t 0 ++ 0 0 0
80 <40 trace <40
<40 <40 <40 <40
2 3 4 5 6 7 6 9 10 11 12 13 14 15 16 Agammaglobulinemia 17 18 19 20
Spleen: ttt = enlarged by 6 cm or more (to umbilicus or below); tt t On biopsy, benign lymphoid hyperplasia. l
agammaglobulinemia patients had very little antigen detectable at this time. After milk ingestion, the amounts of casein and bovine gammaglobulin in the serum increased for the hypogammaglobulinemia subjects, with peak levels occurring 30 to 120 minutes after ingestion. Figure 2 shows representative data for the casein concentrations found in the serum samples of two hypogammaglobulinemic patients (curves A and B) after milk ingestion. Note that the initial amount of this antigen was already 400 ng/ml in one subject, and 800 ng/ml in the other. For the X-linked agammaglobulinemic subjects, no such increases were observed (example: casein concentration, Figure 2, curve C.) The normal subjects were not found to have any detectable antigenemia under these circumstances. To ascertain what correlation the level of these foreign proteins might have to clinical status, the medical and dietary history, results of physical examination and extensive medical and immunologic analyses of each patient were scrutinized. Most particularly, we sought relationships between the presence of food proteins in the serum and dietary history, evidence of current or previous gastrointestinal disease (including frequent diarrhea, malabsorption, nodular lymphoid hyperplasia, giardiasis, liver disease, or gallbladder dysfunction), level of serum immunoglobulins, and evidence of T cell
February
Other Lymphoid Hyperlrophy
Spleen’
Cervical lymphadenopathy,+ nodular lymphoid hyperplasia Lymphoid pulmonary infiltration
Nodular lymphoid hyperplasia Cervical lymphadenopathyt
Nodular lymphoid hyperplasia
= enlarged by 2 to 6 cm; t
r
= just palpable.
1200
l. . . . ..
1000
E 1
1
0
60
30
Minutes
after
milk
90
120
ingestion
Figure 2. After drinking 100 ml of milk, pafients with hypogammaglobulinemia (curves A and B) had an increase in the amount of bovine casein in their serum, whereas a patient with agammaglobulinemia (curve C) did not.
1984
The American
Journal of Medlclne
Volume 78
183
ANTIGENEMIA
IN IMMUNODEFICIENCY
DISEASE-CUNNINGHAM-RUNDLES
defects (lymphocyte proliferation, lymphocyte surface markers analyses, thymic hormones, and leukocyte migration inhibition). Although correlations between these factors and dietary protein antigenemia were not found, both greater amounts of bovine casein and the presence of bovine gammaglobulin in the serum were strongly correlated with the presence of splenomegaly as repeatedly noted by examining physicians (p
184
February 1984
The American Journal of Medicine
ET AL
agammaglobulinemia do have less frequent gastrointestinal disease than patients with hypogammaglobulinemia [ 121, but we did not find that gastrointestinal disease could be correlated with antigenemia in the hypogammaglobulinemic patients we studied. We have previously sho.wn that 50 to 60 percent of patients with hypogammaglobulinemia have substantial T cell defects [ 13,141 that are not demonstrable in patients with X-linked agammaglobulinemia [ 151. Possibly intact cellular immune function in agammaglobulinemia is responsible for the clinical disparities and the great difference in antigen absorption that we have found between these two populations of patients. We have not, however, been able to correlate a range of T cell functions with the relative level of foreign food protein in the serum of the hypogammaglobulinemic patients, since some patients with normal T cells had large amounts of casein and bovine gammaglobulin, and others with poorly functioning T cells had very little of these proteins in their serum. Our data establish a strong statistical relationship between food protein antigenemia and lymphoid hypertrophy-particularly splenomegaly. The reasons for this association are not known. Can the antigenemia and splenomegaly be attributed to a common, unknown factor present in this disease, or is it possible that a “leaky” gastrointestinal mucosa could lead to chronic lymphoid stimulation? It is known that about 20 percent of patients with hypogammaglobulinemia have splenomegaly and/or nodular hyperplasia; cervical lymphadenopathy is observed in a smaller proportion (about 5 percent) [ 12,141. Again, patients with X-linked agammaglobulinemic differ; lymphoid hyperplasia is rarely seen in this group of patients [ 151. Another aspect of hypogammaglobulinemia (but not agammaglobulinemia) is the increased incidence of lymphoid malignancy; in various studies, it was found to affect 3.4 to 4.3 percent of patients previously diagnosed as having this condition [ 12,14,16]. Although there is no known relationship between lymphoid hypertrophy and lymphoma, we speculate that a chronic gastrointestinal leakage of luminal antigens of food, bacterial, and viral origin could be related directly or indirectly to the increased risk of lymphoma in this group of patients. Our data indicate a need for a better understanding of the gastrointestinal mucosal lesions present in immunodeficient patients. Such work may help to clarify the role of the immune system in antigen clearance. ACKNOWLEDGMENT We thank Dr. F. P. Siegal and Dr. S. Gupta for the referral of patients and the staff of the fmmunodeficiency Clinic for their support.
Volume 78
ANTIGENEMIA IN IMMUNODEFICIENCY DISEASE-CUNNINGHAM-RUNDLES ET AL
REFERENCES
4.
5.
6.
8.
Schloss OM, Worthen TW: The permeability of the gastrointestinal tract of infants to undigested protein. Am J Dis Child 1916; 11: 342. Gruskay FL, Cooke RE: The gastrointestinal absorption of unaltered protein in normal infants and in infants recovering from diarrhea. Pediatrics 1955; 16: 763. Walker WA: Antigen absorption from the small intestine and gastrointestinal disease. Symposium on gastrointestinal and liver disease. Pediatr Clin North Am 1975; 22: 731. Paginelli R, Levinsky RJ, Brostoff D, Wraith DG: Immune complexes containing food protein in normal and atopic subjects after oral challenge and effect of sodium chromoglycate on antigen absorption. Lancet 1979; I: 1270. Cunningham-Rundles C, Kirkwood EM, Ferguson A, Parrott D, Good RA: Circulating immune complexes in sera of allergic patients. Clin Res 1980; 28: 343. Cunninoham-Rundles C. Brandeis WE. Good RA. Dav NK: BoviGe antigens and the formation of circulating immune complexes in selective IgA deficiency. J Clin Invest 1979; 64: 270. Carr RJ, Wold RT, Farr RJ: Antibodies to bovine gammaglobulin and the occurrence of a BGG-like substance in systemic erythematous sera. J Allergy Clin lmmunol 1972; 50: 18. Vitetta ES, Baur S, Uhr JW: Cell surface immunoglobulin isolation and characterization of immunoglobulin from
February
9.
10.
11.
12.
13.
14.
15. 16.
1984
mouse splenic lymphocytes. J Exp Med 1971; 134: 242. Paginelli R, Levinsky RJ: Solid phase radioimmunoassay for detection of circulating food protein antigens in human sera. J lmmunol Methods 1980; 37: 333. Block KJ, Block DB, Stearns M, Walker WA: Intestinal uptake of macromolecules. IV. Uptake of protein antigen in vivo in normal rats and rats infected with Nippostrongylus brasiliensis or subjected to mild systemic anaphylaxis. Gastroenterology 1979; 77: 1039. Griswold WR: Studies of the integrity of iodinated bovine albumin after circulation in vivo. lmmunol Commun 1980; 9: 297. Asherson GL, Webster AD: Diagnosis and treatment of immunodeficiency disease. Oxford: Blackwell Scientific Publications, 1980; 22. Cunningham-Rundles S, CunninghamRundles C, Ma DI, Siegal FP, Kosloff CL, Good RA: Impaired proliferative responses to B lymphocyte activators in common varied immunodeficiency. J Clin lmmunol 1981; 1: 65. Good RA: Common variable immunodeficiency. In: Bergsma D, ed. Birth defects compendium. The National Foundation-March of Dimes. New York: AR Liss, 1979; 547. Good RA: Agammaglobulinemia, X-linked infantile. In Ref 14; 54. Spector BD, Perry GS, Good RA, Kersey JH: lmmunodeficiency diseases and malignancy. In: Twomey JJ, Good RA, eds. The immunopathology of lymphoreticular neoplasms. New York: Plenum Publishing, 1978; 203-222.
The American
Journal of Medicine
Volume 78
185