Pancreas Vol. 6, No. 5 , pp. 558-563 0 1991 Raven Press, Ltd., New York

Serum Immunoglobulin G Directed Against Porcine Trypsin in Pancreatic Insufficient Cystic Fibrosis Patients Receiving Pancreatic Enzyme Supplements Richard Couper, Steven Lichtman, Geoffrey Cleghorn, Virginia del Castillo, and Peter Durie Division of Gastroc’nterology, Department of Pediatrics, and The Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada; and Department of Pediatrics, University of Toronto, Toronto, Ontario, Canada.

Summary: Cystic fibrosis (CF) patients frequently malabsorb nutrients because of pancreatic failure. Standard therapy entails oral administration of porcine pancreatic enzymes, with meals. Porcine enzymes contain in excess of 25 potentially antigenic proteins. To evaluate antigenicity of one of these (porcine trypsin), we developed ELISA techniques capable of measuring total immunoglobulin G (IgG) and IgG directed against porcine trypsin in patient sera. Cross-sectional evaluation of sera from 12 controls and 41 C F patients showed that IgG directed against porcine trypsin was detectable in 12/17 C F patients receiving porcine enzymes (50.6 2 56.0 ng/ml; range 0-154.0 ng/ml), while none was detected in controls or the 26 CF subjects not receiving enzymes. In the 17 C F patients receiving enzymes, porcine trypsin binding IgG contributed 0.85 2 0.83% of the total IgG pool. Levels of porcine trypsin binding did not correlate with total IgG. Longitudinal evaluation was then performed in 26 C F patients, before and after commencement of enzyme therapy. Prior to commencing therapy, porcine trypsin binding IgG was undetectable in sera from 24/26 patients. Within 4.2 years of commencing therapy, 25/26 patients (96%) developed porcine trypsin binding IgG. Thus, serum IgG responses to porcine trypsin appear to be common in CF patients receiving porcine enzymes and contribute considerably to total IgG levels. Other individual enzymes in porcine extracts are likely to elicit similar antigenic response. Key Words: Immunoglobulin G-Porcine trypsin-Cystic fibrosisPancreatic enzyme supplements.

dividual enzymes, in addition to other proteins and peptides, and most of these foreign proteins are potentially antigenic. Although the antigenicity of porcine pancreatic enzymes has received limited attention, there is some evidence to suggest that an antibody response occurs following oral ingestion. Romeo et a1 (2) used solid phase radioimmunoassay to demonstrate porcine trypsin binding by immunoglobulin G (IgG) in 42 of 53 patients with CF, who were receiving porcine enzymes by mouth. Addi-

Porcine pancreatic enzyme supplements are required in approximately 85% of patients with cystic fibrosis (CF), in order to treat nutrient malassimilation due to pancreatic insufficiency ( I ) . Porcine pancreatic extracts contain at least twenty five in-

Manuscript received April 27, 1990; revised manuscript accepted August 15, 1990. Address correspondence and reprints requests to Dr. P. Durie at Division of Gastroenterology, The Hospital for Sick Children 555 University Avenue, Toronto, Ontario Canada M5G 1x8.

558

IgG AND PORCINE TR YPSIN IN CYSTIC FIBROSIS

tionally, in a preliminary study, we identified immunoglobulin A (IgA), capable of binding to porcine trypsin in proximal intestinal secretions of C F patients receiving enzyme supplements (3). Immunologically-mediated mechanisms may contribute to an adverse respiratory prognosis in patients with CF. Those C F patients with severe respiratory disease appear to have elevated levels of circulating immune complexes, which contain antigens derived from various bacterial sources, including pseudomonas and streptococci as well as from Candida albicans (4-1 0). Immune complex deposition has also been demonstrated in C F lung and pancreatic tissue at postmortem examination. Circulating hypergammaglobulinemia, which may reflect an enhanced immune response, has also been associated with an adverse respiratory prognosis in CF patients ( 1 1). To evaluate the antigenicity of oral enzyme supplements, we developed a sensitive enzyme-linked immunoadsorbent assay (ELISA), capable of detecting IgG directed against porcine trypsin (PTryp-IgG) in sera, and have utilized this technique to assess the frequency, rapidity, and magnitude of response in enzyme-treated and untreated CF patients. METHODS Patients In a preliminary study, cross-sectional evaluation was undertaken in 53 subjects. Forty-one subjects were determined to have CF by characteristic clinical manifestations and by at least two elevated sweat chloride determinations (>60 mmol/L). This CF cohort was further subdivided into 24 pancreatic-sufficient patients (nine women), not receiving porcine pancreatic extracts, and 17 patients (eight women) with documented pancreatic insufficiency, who had been receiving pancreatic extracts for an average of 2.9 +- 3.2 years (range 0.01-17 years). The term pancreatic-sufficiency describes CF patients who have sufficient residual pancreatic exocrine function to permit normal digestion without the need for enzyme supplements. Pancreaticsufficient and pancreatic-insufficient patients were characterized by 3-5 day fecal fat balance studies. Twelve subjects (6 females), with no clinical evidence of cystic fibrosis, growth failure, or pancreatic or renal disease, served as controls. The mean age (+-ISD) and range of the three groups were as follows: CF patients, not receiving enzymes, 12 ?

559

8.4 years (0.7-31.2 years); CF patients, receiving enzymes, 3.4 & 3.1 years (0.1-12.9 years); controls, 1.9 ? 2.0 years (0.1-7.5 years). Controls, and C F patients receiving enzymes, were significantly younger than CF patients not receiving enzymes (p < 0.001). This finding reflects the milder symptoms in CF patients with pancreatic sufficiency, who, in general, are diagnosed at a later age. Longitudinal evaluation was then undertaken. In 26 CF patients (0.84 2 1.2 years, range 0-5 years) with documented pancreatic insufficiency, serum samples were obtained prior to commencement of enzyme supplements. After commencement of enzymes, 49 sera were available for analysis: 14 samples within 1.5 years; 16 samples within 1.5 to 3.0 years; and 19 samples between 3.0 and 5.5 years, following commencement of enzyme therapy. Patient and control sera were also evaluated for human, immunoreactive, cationic trypsinogen levels by a previously described double antibody radioimmunoassay (12). ELISA for total IgG and IgG directed against porcine trypsin Random blood samples, collected from patients and controls, were separated by centrifugation and the sera stored at - 70°C until assayed. The ELISA techniques for total IgG (T-IgG) and specific IgG, directed against porcine trypsin, rely on an antigenantibody sandwich technique. In the P-Tryp-IgG assay, purified Porcine Trypsin Fraction IX (Sigma Chemicals, St. Louis, MO, U.S.A.) was inhibited with tosyl-lysyl-chloromethylketone (TLCK) to prevent autodigestion. Excess unbound TLCK was removed by dialysis in distilled water using a spectraphor-3 membrane. The purity of porcine trypsin was confirmed by finding a single homogenous band on polyacrylamide electrophoresis, using a 12.5% sodium dodecyl sulphate polyacrylamide gel. Three hundred micrograms TLCK-Trypsin in 150 pl coating buffer, pH 9.6, was adsorbed onto plastic microtitre plates (Immulon 2, Dynatech) and incubated overnight, while humidified at 4°C. Wells were washed three times with 150 pl of phosphate buffered saline (PBS) in Tween 0.05%. Excess binding sites were saturated with PBS - 1% Bovine Serum Albumin Fraction V (BSA), incubated for 3 h at room temperature, then washed. One hundred fifty of sera, diluted to 1: 1,000 and 1:2,000with PBS - 1% BSA, were added in duplicate and incubated at 4°C overnight. Wells were washed three times and 150 pl goat antihuman IgG, conjugated with Pancreas, Vol. 6 , No. 5 , 1991

R. COUPER ET A L .

560

alkaline phosphatase (Tago Inc.), diluted to 1:10,000 with PBS - 1% BSA was added and incubated for 3 h at room temperature. Wells were washed three times and 150 pl of paranitrophenol phosphate added, (1 tablet in 5 ml, 10% ethanolamine, pH 9.6). The reaction was stopped at 20 minutes with 50 pl of 3 M NaOH and the optical density (410 nrn) was read on a microplate reader. A standard curve was constructed using simple linear regression analysis with known dilutions of human IgG. The sensitivity range was 6.5-77.5 ng/ml. The intra-assay coefficient of variation was between 2-6%. A modification of the ELISA method for P-TrypIgG was used to evaluate specificity of IgG binding by porcine trypsin. Prior to adding test sera, 25 pg of TLCK-porcine trypsin was added in triplicate to the micotitre plate, then incubated for I h at room temperature. The assay was then performed as described above. The ELISA method for assessing total IgG was essentially the same as that described for specific anti-porcine trypsin IgG. In brief, goat antihuman IgG was applied to the wells, followed by paired sera, diluted to 1:250,000 and 1:500,000 with PBS1% BSA and goat antihuman IgG, conjugated with alkaline phosphatase (1 :10,000). The terminal steps are identical to those used for porcine trypsin binding IgG. Statistical analysis Results are presented as the mean +- one standard deviation. The significance of the difference of means was determined by the unpaired Student’s t test. Correlation coefficients were calculated using simple linear regression analysis. The amount of P-Tryp-IgG was calculated as a percentage of total IgG in all patients in whom P-Tryp-IgG was detected. RESULTS

Cross-sectional evaluation Cross-sectional data for total IgG and specific porcine trypsin IgG in the three patient groups are shown in Table 1 . Controls, and CF patients not receiving enzymes, had no detectable P-Tryp-IgG, whereas 12/17 C F patients receiving enzymes developed detectable titres. There was a highly significant difference in P-Tryp-IgG concentration both between C F patients receiving enzymes and controls, and CF patients not receiving enzymes (p < Pancreas. Vol. 6 , N o . 5 . 1991

TABLE 1. T-IgG and P-Tryp-lgG in sera of CF patients and controls

_

_ _ _ ~ Controls

n ~

C F patients not receiving enzymes CF patients receiving enzymes

~

12 24

17

T-IgG P-Tryp-IgG (mg/ml) (wdml) ~_____ 6.4 t 2.9 0** (3.3-1 3.5) t 11.3 2 8.8* 0** (2.6-40.6) 6.4 t 3.0 50.6 t 55.9 (2.3-12.8) (0-154)

vs C F patients receiving enzymes: * p < 0.035; * * p < 0.001. t Range of values in parentheses.

0.001). There was no significant correlation between patients’ age and T-IgG or P-Tryp-IgG concentration. Similarly, assessment of the five CF patients receiving enzymes, who showed no detectable P-Tryp-IgG, revealed no effect of age or duration of enzyme therapy. No significant correlation was detected between serum immunoreactive pancreatic trypsinogen values and P-Tryp-lgG levels. The contribution of P-Tryp-IgG to T-IgG varied from 0-2.61% (mean 0.85 -+ 0.83%). Preincubation with TLCK-porcine trypsin was performed to assay test specificity in four patients, two with high P-tryp IgG titres and 2 with no detectable P-tryp-lgG. In the two patients with high P-tryp IgG titres, binding of P-tryp IgG was completely abolished; no effect was noted in patients with no measureable P-tryp IgG.

Longitudinal evaluation Only 2/26 patients had detectable P-Tryp-IgG prior to commencement of porcine enzyme therapy. Review of these patients revealed no obvious source of exposure to this antigen. No significant alterations in T-lgG concentrations occurred after the patients began enzyme therapy (Table 2), but 24 of the 25 patients (96%) developed P-TryplgG within 4.2 years of initiation of porcine enzyme therapy. Individual data for each time period is represented graphically (Fig. 1). Eighty-four percent of patients assessed before 1.5 years and 95% of those assessed within three years had detectable PTryp-IgG. At all time periods following commencement of porcine pancreatic extracts, P-Tryp-IgG concentrations were significantly greater than those found in pretreatment sera. P-Tryp-IgG concentrations in sera were not significantly different in patients receiving enzymes for less than 1.5 years, when compared with values obtained after longer periods (>1.5 years) of enzyme therapy. Simple linear regression analysis showed no correlation be-

561

IgG A N D PORCINE TR YPSIN IN CYSTIC FIBROSIS TABLE 2. Longitudinal evaluation of T-IgG and P-Tryp-IgG in CF patient sera before and after commencing porcine enzymes

-

Before enzymes 15

Serum immunoglobulin G directed against porcine trypsin in pancreatic insufficient cystic fibrosis patients receiving pancreatic enzyme supplements.

Cystic fibrosis (CF) patients frequently malabsorb nutrients because of pancreatic failure. Standard therapy entails oral administration of porcine pa...
588KB Sizes 0 Downloads 0 Views