Autoimmune pancreatitis is characterized by diffuse swelling of the pancreatic parenchyma and irregular narrowing of the main pancreatic duct.1,2 Close association with high serum IgG concentrations, antinuclear antibody, favorable response to corticosteroid treatment, and a specific human leukocyte antigen (HLA) haplotype have been reported to indicate the presence of the autoimmune mechanisms in the pathogenesis of this disease.2-10 We reported previously that patients with autoimmune pancreatitis have high serum IgG4 concentrations and that these high values are closely associated with disease activity.8 We also previously found decreased levels of various complements in some patients with autoimmune pancreatitis in its active stage, which was suggestive of activation of certain complement systems, the classic pathway, the alternative pathway, or the mannose-binding lectin (MBL) pathway.
The characteristic pattern of IgG4 expression in patients with this disease implies the existence of a close association between IgG4 and some complement activation system. IgG4 is unique among the IgG subclasses in its inability to bind C1q, resulting in a lack of activation of the classic pathway. In the pathogenesis of membranous nephropathy, IgG4 subclass immune complexes may lead to the formation of the membrane-attack complex in tissues by fixing the complement through an alternative pathway.11 Similar to the case of membranous nephropathy, autoimmune pancreatitis is closely associated with a high serum concentration of IgG4 subclass immune complexes,8 suggesting the presence of the same pathologic mechanisms in both diseases. However, there has been no convincing evidence to support the contribution of a specific complement activation system.
The aim of this study was to clarify whether or not any complement activation system, the classic pathway, the alternative pathway, or the MBL pathway, is functional during the active stage of autoimmune pancreatitis.
MATERIALS AND METHODS
Patients and Healthy Controls
Between September 1994 and September 2004, we treated and followed 44 patients with autoimmune pancreatitis (35 men and 9 women; age: 47-79 years; median age: 67 years). The following diagnostic criteria for autoimmune pancreatitis were used: (1) enlarged pancreas on ultrasonography or computed tomography (CT), and irregular narrowing of the main pancreatic duct on endoscopic retrograde cholangio-pancreatography (ERCP); (2) increased serum gammaglobulin levels, high serum IgG or IgG4 concentrations, or the presence of autoantibodies; and (3) characteristic histologic findings of lymphoplasmacytic infiltration and fibrosis.12 We treated 32 patients with 40 mg prednisolone daily for 4 weeks, after which we reduced the dose by 5 mg/wk for several weeks. All patients responded favorably to corticosteroid therapy, resulting in the improvement of clinical, laboratory, and image findings. The maintenance dose of prednisolone was 5 to 10 mg. We evaluated 50 patients with chronic calcifying pancreatitis who were treated and followed during the same period. These patients consisted of 44 men and 6 women (age: 30-80 years; median age: 64 years). Healthy controls showed no significant diseases and no abnormalities on laboratory tests, including no immunologic or inflammatory findings.
All participants provided written informed consent for the tests with DNA samples, and serum samples were obtained after receiving permission from the patients. Serum samples were also collected from the normal subjects after informed consent was obtained. The institutional ethics committee approved the study.
Serum complement levels were measured with fresh samples. Serum C3 and C4 concentrations were measured by turbidimetric immunoassay methods, and the serum CH50 concentration was measured by the liposome method. To measure the serum C3, C4, and CH50 concentrations, we excluded serum samples with a high C reactive protein concentration of more than 3 mg/dL. Serum levels of IgG subclasses were measured by single radial immunodiffusion assays (Binding Site, Birmingham, United Kingdom). Serum levels of circulating immune complexes (CICs) were measured by 2 methods: the C1q binding assay or the monoclonal rheumatoid factor (mRF) method (Immune complex mRF Nissui; Nissui Pharmaceutical, Tokyo, Japan).
The serum MBL concentration was measured by an enzyme-linked immunosorbent assay according to the manufacturer's instructions (MBL-oligomer ELISA; AntibodyShop, Gentofte, Denmark).
Two different point mutations in exon 1 of the MBL gene reported in Japanese subjects were studied: the first was at codon 54 (GGC→GAC), which results in a change of glycine to aspartic acid, and the second was at codon 57 (GGA→GAA), which results in a change of glycine to glutamic acid. Detection of the mutations in codons 54 and 57 was performed using the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) technique. To detect the codon 54 point mutation, we used an amplified PCR product (328 bp in length) with the forward primer (5′-GTAGGACAGAGGGCATGCTC-3′) and the reverse primer (5′-CAGGCAGTTTCCTCTGGAAGG-3′), and the product was digested with BanI (New England Biolabs). BanI cleaves codon 54 of the wild-type allele into 2 fragments (245 and 84 bp) and leaves the mutated allele undigested. For detection of the codon 57 point mutation, the PCR product was digested with MboII (New England Biolabs). MboII cleaves the mutated allele at codon 57 into 2 fragments (274 and 55 bp), whereas the wild-type allele remains undigested.
Statistical analysis was performed by the Wilcoxon matched-pairs signed rank test, χ2 test, the Mann-Whitney analysis, or Bonferroni method. A level of corrected P < 0.05 was accepted as statistically significant. All reported P values are 2-sided.
Serum C3, C4, and CH50 Concentrations
We measured serum complements levels, the C3, C4, and CH50 concentrations, in 44 patients with autoimmune pancreatitis and 22 patients with chronic pancreatitis. We found significantly lower CH50 levels (<30 U/mL) in patients with autoimmune pancreatitis than in those with chronic pancreatitis. Although we found no significant difference in C3 and C4 values between patients with autoimmune pancreatitis and those with chronic pancreatitis, we found decreased complement values in some patients with autoimmune pancreatitis as follows: C3 (<86 mg/dL) in 16 patients (36%), C4 (<17 mg/dL) in 16 patients (36%), and CH50 (<30 U/mL) in 7 patients (17%; Table 1).
Serum MBL Concentration
We measured the serum MBL concentrations of 44 patients with autoimmune pancreatitis, 40 patients with chronic pancreatitis who were not experiencing an acute attack, and 48 healthy controls. The median serum MBL values were as follows: autoimmune pancreatitis, 2139 μg/L; ordinary chronic pancreatitis, 4234 μg/L; and healthy controls, 1387 μg/L. The MBL values in cases of both autoimmune pancreatitis and ordinary chronic pancreatitis were significantly higher than those of the healthy controls. We could find no significant difference between autoimmune pancreatitis and ordinary chronic pancreatitis in terms of the MBL values (Fig. 1).
We compared the MBL values of 32 patients with autoimmune pancreatitis before and after corticosteroid therapy; however, we didn't find any significant difference between these 2 groups (Table 2).
MBL Gene Polymorphisms
We measured MBL gene polymorphisms (codon 54 and 57 in exon 1) in 43 patients with autoimmune pancreatitis and 40 patients with ordinary chronic pancreatitis. We found no point mutation of codon 57 in any of the patients. We found a point mutation of codon 54 in a few patients with both autoimmune pancreatitis and ordinary chronic pancreatitis. However, we found no significant differences with regard to frequency between the 2 groups (Table 3).
Four patients with autoimmune pancreatitis and 2 patients with ordinary chronic pancreatitis were homozygous for a mutant allele and had an undetectable MBL value of 0 μg/L. The MBL values of heterozygotes were significantly lower than those of the wild-type allele in both the patient group with autoimmune pancreatitis and the patient group with ordinary chronic pancreatitis (Table 4).
Serum CIC Concentrations and Complement Values
We found high serum CIC values, as determined by 2 assay systems, CIC by mRF assay and CIC by C1q assay, in patients with autoimmune pancreatitis; these results revealed high CIC values by mRF assay in 32 patients (73%) and high CIC by C1q assay in 28 patients (64%). The elevated CIC serum values determined by both assays decreased significantly after corticosteroid therapy, suggesting that CIC is associated with disease activity (Table 2).
To determine the association between elevated CIC and the complement activation system, we compared serum C3, C4 CH50, MBL, and the IgM and IgG subclass values of patients with autoimmune pancreatitis exhibiting high CIC and normal CIC values. In the patients with autoimmune pancreatitis, the elevation of CIC, as determined by mRF assay (>4.1 μg/mL), was significantly associated with an elevation of IgG4 but not with decreases in levels of C3 and C4 (Table 5). On the other hand, the elevation in the CIC level, as determined by C1q assay (>2.9 μg/mL), was significantly associated with an elevation in the level of IgG1 and a decrease in the level of C4, which also showed a tendency toward an association with a decrease in the level of C3 (Table 6).
We studied the contribution of CIC and the complement system in the active stage of autoimmune pancreatitis. We found decreased serum C3, C4, and/or CH50 levels in some patients with autoimmune pancreatitis, suggesting that the complement activation system is associated with the activity and/or the pathogenesis of autoimmune pancreatitis.
We previously confirmed that IgG4 and IgG4-type immune complexes are closely associated with the disease activity in patients with autoimmune pancreatitis,8 suggesting that the IgG4-type autoantibody plays a major role in the pathogenesis of this disease. Because IgG4 does not possess the ability to bind the C1q complement, a complement activation system other than the classic pathway may be involved in this context. At this point in the study, we first investigated the contribution of the MBL pathway.13 These results revealed that the MBL values of patients with autoimmune pancreatitis were significantly higher than those of the healthy controls. However, the MBL values of the patients with chronic pancreatitis were also significantly higher than those of the healthy controls. Furthermore, there were no significant differences in MBL values between patients with autoimmune pancreatitis and those with chronic pancreatitis. Because we found a significant increase in the MBL levels of patients with chronic pancreatitis as well as in those of patients with autoimmune pancreatitis, a possible explanation might be that the increase in MBL levels was caused by an acute-phase response.14 Other reports have also shown that the median MBL concentration was nonsignificantly elevated in the plasma from systemic lupus erythematosus (SLE) patients compared with that of healthy individuals.15,16 Another reason for high serum MBL concentrations in both conditions may be inflammatory change of pancreatic tissue. In patients with type I diabetes, increased levels of MBL were associated with diabetic nephropathy.17 It can be speculated that MBL ligands are present in diabetic kidney, leading to a deposition of MBL in kidney with deleterious effects. Accordingly, inflammatory process of both autoimmune pancreatitis and chronic pancreatitis may be associated with high serum MBL concentration.
In addition, corticosteroid treatment, which ameliorated the clinical symptoms of those with autoimmune pancreatitis, had no effect on the MBL value. In accord with previous reports,18 the serum MBL concentrations observed here reflected the MBL genotype of codon 54 substitution, in which homozygotes for variant MBL alleles were observed at a concentration of 0 mg/dL. The frequency of the MBL gene variant in autoimmune pancreatitis is equivalent to that in chronic pancreatitis, which suggests that the MBL gene polymorphism does not contribute to autoimmune pancreatitis; however, this conclusion would contradict the results of previous reports regarding other autoimmune diseases.19,20 These results suggested that the MBL values have little association with disease activity and that the MBL pathway does not contribute to the pathogenesis of autoimmune pancreatitis.
This study shows that CIC levels as determined by both C1q assay and mRF assay decrease significantly after corticosteroid therapy, which suggests that CIC levels were closely associated with the activity and pathogenesis of this disease. High CIC values determined by C1q assay were significantly associated with elevated IgG1 values, which suggested that high CIC levels determined by C1q assay possibly represent an IgG1-type immune complex. In addition, high CIC levels, when determined by C1q assay, were significantly associated with decreases in levels of C4 and also showed a tendency toward association with decreases in levels of C3, suggesting that the classic pathway is possibly functional in the active stage of this disease. On the other hand, high CIC values, when determined by mRF assay, was significantly associated with elevated IgG4 but not with decreased levels of C3 or C4. These results suggested that a high CIC value determined by mRF assay implies the presence of IgG4-type CIC and that these values are only negligibly associated with complement activation. Alternatively, an IgG1-type immune complex possibly induces complement activation through the classic pathway in the pathogenesis of autoimmune pancreatitis.
Although IgG4 elevation is a characteristic feature of this disease and is associated with disease activity, elevation of IgG4 is not directly associated with complement activation. At this moment, we could not determine the exact roles of IgG4 in this disease. Another possible mechanism of tissue damage through IgG4 antibody may be the reaction to the target tissue antigens. In case of pemphigus vulgaris, which are autoimmune skin diseases, IgG4-type autoantibodies react to the cell adhesion molecules desmoglein-3.21 In some ethnic groups, pemphigus vulgaris is strongly associated with HLA-DRB1*0402, and some of the HLA-DRB1*0402 molecules form a complex with peptides produced by the processing of desmoglein-3.22 The bound peptides stimulate specific T cells, which secrete cytokines and trigger the production of desmoglein-3-specific IgG4-type antibodies, resulting in the formation of skin lesions. Autoimmune pancreatitis is closely associated with HLA-DRB1*0405- DQB1*0401 haplotype.10 Similar to pemphigus vulgaris, these HLA molecules possibly present unknown antigen peptides and trigger the production of antigen-specific IgG4-type antibodies, resulting in the tissue damage. To date, some molecules such as carbonic anhydrase II and lactoferin have been reported to be candidate antigens.23 Further studies are needed to confirm exact target antigens.
In conclusion, the results of this study indicate that autoimmune pancreatitis exhibits serum elevation of CIC and abnormalities in the complement activation system. We excluded the involvement of the MBL pathway and the possibility of an alternative pathway and instead suggest that there is an association between a high serum CIC values and the classic pathway in the active state of this disease. These findings may help elucidate the pathogenesis of autoimmune pancreatitis, and they also are suggestive of possible directions for the development of new therapies for the treatment of this disease.
We thank Drs. Teizo Fujita and Kiyotaka Fujita for helpful discussion.
1. Sarles H, Sarles JC, Muratore R, et al. Chronic inflammatory sclerosis of the pancreas-an autonomous pancreatitis disease? Am J Dig Dis
2. Yoshida K, Toki F, Takeuchi T, et al. Chronic pancreatitis caused by an autoimmune abnormality, proposal of the concept of autoimmune pancreatitis
. Dig Dis Sci
3. Horiuchi A, Kaneko T, Yamamura N, et al. Autoimmune pancreatitis
simulating pancreatic lymphoma. Am J Gastroenterol
4. Ito T, Nakano I, Koyanagi S, et al. Autoimmune pancreatitis
as a new clinical entity: three cases of autoimmune pancreatitis
with effective steroid therapy. Dig Dis Sci
5. Horiuchi A, Kawa S, Akamatsu T, et al. Characteristic pancreatic duct appearance in autoimmune chronic pancreatitis: a case report and review of the Japanese literature. Am J Gastroenterol
6. Erkelens GW, Vleggaar FP, Lesterhuis W, et al. Sclerosing pancreatico-cholangitis responsive to steroid therapy. Lancet
7. Horiuchi A, Kawa S, Hamano H, et al. Sclerosing pancreato-cholangitis responsive to corticosteroid therapy: report of 2 case reports and review. Gastrointest Endosc
8. Hamano H, Kawa S, Horiuchi A, et al. High serum IgG4
concentrations in patients with sclerosing pancreatitis. N Engl J Med
9. Horiuchi A, Kawa S, Hamano H, et al. ERCP features in 27 patients with autoimmune pancreatitis
. Gastrointest Endosc
10. Kawa S, Ota M, Yoshizawa K, et al. HLA DRB10405-DQB10401 haplotype is associated with autoimmune pancreatitis
in the Japanese population. Gastroenterology
11. Oliveira DB. Membranous nephropathy: an IgG4
-mediated disease. Lancet
12. Pearson RK, Longnecker DS, Chari ST, et al. Controversies in clinical pancreatology, autoimmune pancreatitis
: does it exist? Pancreas
13. Kilpatrick DC. Mannan-binding lectin: clinical significance and applications. Biochim Biophys Acta
14. Thiel S, Holmskov U, Hviid L, et al. The concentration of the C-type lectin, mannan-binding protein, in human plasma increases during an acute phase response. Clin Exp Immunol
15. Zimmermann-Nielsen E, Baatrup G, Thorlacius-Ussing O, et al. Complement
activation mediated by mannan-binding lectin in plasma from healthy individuals and from patients with SLE, Crohn's disease and colorectal cancer. Suppressed activation by SLE plasma. Scand J Immunol
16. Takahashi R, Tsutsumi A, Ohtani K, et al. Anti-mannose binding lectin antibodies in sera of Japanese patients with systemic lupus erythematosus. Clin Exp Immunol
17. Saraheimo M, Forsblom C, Hansen TK, et al. On behalf of the FinnDiane Study Group. Increased levels of mannan-binding lectin in type 1 diabetic patients with incipient and overt nephropathy. Diabetologia
18. Sasaki K, Tsutsumi A, Wakamiya N, et al. Mannose-binding lectin
polymorphisms in patients with hepatitis C virus infection. Scand J Gastroenterol
19. Lau YL, Lau CS, Chan SY, et al. Mannose-binding protein in Chinese patients with systemic lupus erythematosus. Arthritis Rheum
20. Ip WK, Lau YL, Chan SY, et al. Mannose-binding lectin
and rheumatoid arthritis in southern Chinese. Arthritis Rheum
21. Shirakata Y, Shiraishi S, Sayama K, et al. Subclass characteristics of IgG auto-antibodies in bullous pemphigoid and pemphigus. J Dermatol
22. Wucherpfnnig KW, Yu B, Bhol K, et al. Structual basis for major histocompatibility complex (MHC)-linked susceptibility to autoimmunity: charged residues of a single MHC binding pocket confer selective presentation of self-peptides in pemphigus vulgaris. Proc Natl Acad Sci USA
23. Okazaki K, Uchida K, Ohana M, et al. Autoimmune-related pancreatitis is associated with autoantibodies and Th1/Th2-type cellular immune response. Gastroenterology
Keywords:© 2006 Lippincott Williams & Wilkins, Inc.
autoimmune pancreatitis; circulating immune complex; classical pathway; complement; IgG4; mannose-binding lectin