Objective In a previous study, the authors have shown that rather than variants in trypsinogen gene(s), mutations in pancreatic secretory trypsin inhibitor (encoded by SPINK1) and cathepsin B (CTSB) are associated with tropical calcific pancreatitis (TCP). Recently, chymotrypsin C (CTRC) variants that diminish its activity or secretion were found to predict susceptibility to chronic pancreatitis (CP). The authors analysed CTRC variants in a large, ethnically matched case-control TCP cohort.
Design The authors sequenced all eight exons and flanking regions in CTRC in 584 CP patients (497 TCP, 87 idiopathic CP) and 598 normal subjects and analysed the significance of association using χ2 test. The authors also investigated interaction of CTRC variants with p.N34S SPINK1 and p.L26V CTSB mutations.
Results The authors identified 14 variants in CTRC, of which non-synonymous variants were detected in 71/584 CP patients (12.2%) and 22/598 controls (3.7%; OR 3.62, 95% CI 2.21 to 5.93; p=6.2×10−8). Rather than the commonly reported p.K247_R254del variant in Caucasians, p.V235I was the most common mutation in Indian CP patients (28/575 (4.9%); OR 7.60, 95% CI 2.52 to 25.71; p=1.01×10−5). Another pathogenic variant, p.A73T was identified in 3.1% (18/584) patients compared with 0.3% (2/598) in controls (OR=9.48, 95% CI 2.19 to 41.03, p=2.5×10−4). The authors also observed significant association for the synonymous variant c.180C>T (p.(=)) with CP (OR 2.71, 95% CI 1.79 to 4.12, p=5.3×10−7). Two novel nonsense mutations, p.G242AfsX9 and p.W113X were also identified exclusively in CP patients. No interaction between CTRC variants and p.N34S SPINK1 or p.L26V CTSB mutations was observed.
Conclusion This study on a large cohort of TCP patients provides evidence of allelic heterogeneity and confirms that CTRC variants play a significant role in its pathogenesis.
- Chymotrypsin C
- chronic pancreatitis
- cathepsin B
- gene-gene interaction
- hepatitis B
- hepatocellular carcinoma
- nonalcoholic steatohepatitis
- non-ulcer dyspepsia
- tropical gastroenterology
- molecular genetics
Statistics from Altmetric.com
- Chymotrypsin C
- chronic pancreatitis
- cathepsin B
- gene-gene interaction
- hepatitis B
- hepatocellular carcinoma
- nonalcoholic steatohepatitis
- non-ulcer dyspepsia
- tropical gastroenterology
- molecular genetics
Significance of this study
What is already known on this subject?
Genetic basis of chronic pancreatitis including tropical calcific pancreatitis is different in Indians.
Variants in CTRC gene predict susceptibility to chronic pancreatitis.
CTRC mutants are pathogenic in chronic pancreatitis due to defect in its secretion and/or activity.
What are the new findings?
We have identified several novel variants in CTRC in a large case-control cohort of tropical calcific pancreatitis.
The mutation spectrum of CTRC in Indians is different from Europeans providing evidence of allelic heterogeneity.
There was no evidence of any relationship between p.N34S SPINK1 or p.L26V CTSB and CTRC variants.
The study established a role of CTRC variants, albeit minor in the pathogenesis of chronic pancreatitis in Indians.
How might it impact on clinical practice in the foreseeable future?
Identification of novel CTRC variants may influence susceptibility screening in Indian chronic pancreatitis patients.
Chronic pancreatitis (CP) is characterised by chronic inflammation of the pancreas resulting in permanent destruction of pancreatic parenchyma eventually leading to loss of exocrine and/or endocrine function. Premature intracellular activation of trypsinogen and other digestive proteases is an early event in the course of pancreatitis and is causally related to the onset of disease. Several genetic risk factors for CP have been identified. The principal genetic influences include gain of function mutations in cationic trypsinogen (PRSS1; OMIM 276000) and loss of function mutations in pancreatic secretory trypsin inhibitor (SPINK1; OMIM 167790) and cystic fibrosis transmembrane regulator (CFTR; OMIM 602421) whereas p.G191R mutation in anionic trypsinogen gene (PRSS2; OMIM 601564) is protective.1–5 Copy number mutations have also been identified in patients with hereditary and idiopathic chronic pancreatitis (ICP).6 Tropical calcific pancreatitis (TCP) is a juvenile form of CP prevalent in tropical developing countries. We and others have earlier shown that mutations in trypsinogen gene(s) are not associated with TCP7–9 while c.101A>G (p.N34S) SPINK1 predicts susceptibility to CP including TCP (p=1.54×10−11)7 and ICP (p=3.2×10−17)10 in Indians. Subsequently, we reported significant association of c.76C>G (p.L26V) mutation in cathepsin B (CTSB; OMIM 116810) with TCP (OR=2.09, p=0.013)11 but a similar study in Europeans did not show any association of p.L26V CTSB variant with ICP, suggesting population specific differences in genetic susceptibility to CP.12 However, many of the above-mentioned mutations do not account for all TCP patients and it is not clear what could trigger abnormal trypsinogen activation and metabolism in the absence of mutations in the trypsinogen genes.
Chymotrypsin C (CTRC; OMIM 601405) encodes a 268 amino acid long serine protease that is secreted from the pancreas and has chymotrypsin-like protease activity. The encoded protein is a member of peptidase S1 family and was initially shown to stimulate the autoactivation of human cationic trypsinogen by cleaving after Phe18.13 Subsequently it was also shown to promote the degradation process of all human trypsin and trypsinogen isoforms with high specificity.14 The activation and degradation is regulated by prevailing Ca2+ concentrations and an altered Ca2+ signalling is the hallmark of pancreatitis.15 Since the intra-acinar activation of trypsinogen is thought to be the primary cause for pancreatitis, non-functional variants of CTRC are supposed to play a role in pathogenesis of the disease and along with defective Ca2+ signalling may lead to onset/progression of pancreatitis.
Recently two independent studies in the European population found mutations in CTRC to be associated with CP.16 ,17 Rosendahl et al also investigated the role of CTRC variants in patients of Indian origin and found them to be associated with TCP, which was further substantiated in another study.18 However, both the studies have focused on a small number of subjects.16 ,18 We screened the complete coding region of CTRC in a large population of CP patients from India, using a case-control cohort approach and also analysed its interaction with other susceptibility genes like SPINK1 that are associated with CP in Indians. Earlier studies have shown that cathepsin B can activate trypsinogen19 while CTRC is capable of inactivating trypsinogen,14 hence, we also investigated the relationship between mutations in CTSB and CTRC genes.
Subjects and methods
Patients and controls
A total of 584 CP (497 TCP, 87 ICP) patients were diagnosed at the Asian Institute of Gastroenterology, Hyderabad; Department of Gastroenterology, SCB Medical College, Cuttack and Department of Gastroenterology, Medical College, Calicut, based on the following WHO criteria: (1) recurrent abdominal pain; (2) large intraductal calculi; (3) sonographic and endoscopic retrograde cholangiopancreaticography evidence of pancreatic calcification; (4) absence of any other aetiological factor such as alcoholism; and (5) diabetes mellitus as defined by WHO Study group E (may or may not be present).20 Affected individuals who did not have intraductal calculi and showed small speckled pancreatic calcification in the absence of known aetiological factors were classified as having ICP.10 ,21 A cohort of 598 ethnically matched individuals who had no complaints or evidence of pancreatitis on imaging were recruited as controls. All the patients and controls gave written informed consent and completed a structured questionnaire detailing their medical history.7 ,8 The study was approved by the Institutional Ethics Committees of all institutes following the guidelines recommended by the Indian Council of Medical Research for research on human subjects.
Genomic DNA was isolated from all the individuals using standard protocols and all eight exons, including the splice site junctions of CTRC were amplified using specific primers (online supplementary table S1). PCR products were purified using Millipore MultiScreen PCRμ96 Filter Plates and sequenced individually using the ABI PRISM BigDye Terminator Cycle Sequencing Kit V.1.1 (PE Applied Biosystems, Foster City, California, USA) on a ABI3730 Genetic Analyser (Applied Biosystems). We initially sequenced the coding region and the flanking intron–exon boundaries in all the patients and 230 controls. Additional 368 controls were screened for all the mutations that were identified in CP patients. All identified mutations and their status were confirmed by sequencing on both the strands. From our earlier studies, we had available data for c.86A>T (p.N29I), c.365G>A (p.R122H) mutations in PRSS1, p.N34S, c.163C>T (p.P55S) mutations in SPINK1 and p.L26V, c.157A>G (p.S53G) mutations in CTSB genes in 306 TCP patients, all ICP patients and 330 normal subjects.10 ,11 We sequenced exons 2 and 3 of PRSS1, exon 3 of SPINK1 and exons 3 and 4 of CTSB in the remaining individuals and analysed for the above-mentioned mutations.
Difference in genotype frequency of mutation between cases and controls was assessed by χ2 test using SPSS V. 18.0. A two tailed p value <0.05 was considered significant in all analyses. Power of the study was calculated using G*Power V.3.1.2 (http://www.psycho.uni-duesseldorf.de/abteilungen/aap/gpower3/).
On comprehensive analysis of the CTRC gene, we identified nine non-synonymous variants (seven missense, one nonsense and one deletion variant), of which four are novel mutations. We also detected two synonymous variants reported earlier and three novel intronic variants (table 1A,B). As reported earlier, majority of the variants resided in exon 3 and 7 of CTRC gene. Instead of the commonly reported c.738_761del24 (p.K247_R254del) variant found in Caucasians, we identified c.703G>A (p.V235I) as the major mutation in Indian CP patients (28/575 (4.9%)). Of the 28 patients carrying the p.V235I variant, 20 were heterozygous and eight homozygous as compared with three heterozygotes and one homozygote in controls (4/598 (0.7%); OR 7.60, 95% CI 2.52 to 25.71; p=1.0×10−5). This mutation which is known to reduce activity and secretion of CTRC was also the major variant in TCP patients (26/488 (5.3%); OR 8.36, 95% CI 2.75 to 28.43; p=3.2×10−6). Another pathogenic variant, c.217G>A (p.A73T) was also over-represented in CP patients (18/584 ;3.1%) compared with 0.3% (2/598) in controls (OR 9.48, 95% CI 2.19 to 41.03; p=2.5×10−4). This mutation was also the major variant identified in ICP patients (5/87 (5.8%)). The loss of function variant c.143A>G (p.Q48R) was detected in a single TCP patient and other reported mutations such as c.514G>A (p.K172E) and c.760C>T (p.R254W) were observed in three patients each. Only one TCP patient was compound heterozygous for p.A73T and p.V235I mutations.
All novel mutations in the coding region were private and present in only one patient each. The novel mutations, c.338G>A (p.W113X) and c.725delG (p.G242AfsX9) are likely to produce prematurely truncated and inactive protein and hence can be considered as causal, whereas the functional relevance of another novel variant c.769G>A (p.A257T) is not clear. We observed a single nucleotide substitution in intron 6, c.640-6G>A exclusively in two patients and c.736C>T (p.R246C) variant only in two controls, suggesting a likely susceptibility and protective role respectively for these mutations. We also identified two novel deletion/insertion variants in intron 6 (c.640-34delG and c.640-34insG), which together were present in 14/575 (2.4%) patients and 9/598 (1.5%) controls, and the difference was not statistically significant (p>0.05) (table 1A). Thus, overall, we identified 14 variants in CTRC, of which non-synonymous and intronic variants were detected in 71/584 CP patients (12.2%) and 22/598 controls (3.7%) (OR 3.62, 95% CI 2.21 to 5.93; p=6.2×10−8).
We also assessed the genotype distribution of common polymorphism c.180C>T (p.(=)) (previously known as p.G60G) that was earlier shown to be significantly associated with familial CP,17 and observed that subjects heterozygous for the variant had a higher risk of CP compared with those carrying the wild allele (OR 2.46, 95% CI 1.84 to 3.29; p=5.5×10−10) and the risk magnified several folds for individuals homozygous for the risk allele (OR 9.89, 95% CI 2.95 to 33.18; p=5.9×10−6) (table 2). However, distribution of another synonymous polymorphism c.285C>T (p.(=)) was not significantly different between patients and controls (table 1B). The study was 96% powered to detect association of common variant c.180C>T (p.(=)) with an OR 2.71 (minor allele frequency=18% in patients and 8% in controls) and 99% powered to detect association of rare variants.
We did not detect the commonly reported variants p.R122H or p.N29I in PRSS1 in any of the patients. In total, 179 of 567 of these individuals (31.6%) carried p.N34S SPINK1 variant (31 homozygotes, 148 heterozygotes), while 297 of 516 (49.4%) were positive for the p.L26V CTSB variant (69 homozygotes and 228 heterozygotes).
In silico analysis of novel CTRC variants
SIFT analysis predicted p.A257T to affect protein function while p.R246C was predicted to be tolerated. Analysis of the intronic variants, c.640-34insG and c.640-34delG using NetGene2, showed an increase in confidence value of a hypothetical alternative splice site at the acceptor ends of the variant. This mutation is likely to lead to an alternatively spliced product with a loss of the initial 30 nucleotides from exon 7 and thus encode a defective CTRC protein. We observed only a slight reduction in the confidence value at the acceptor splice site of intron 6 for the c.640-6G>A variant, suggesting lack of any role for this variant.
Interaction between CTRC variants and p.N34S SPINK1 mutation
Close to one-third of CP patients (179/567 (31.6%)) were positive for p.N34S SPINK1 mutation; 148 of 480 TCP patients (30.8%; 125 heterozygotes, 23 homozygotes) and 31 of 87 ICP patients (35.6%; 23 heterozygotes, 8 homozygotes) carried p.N34S SPINK1 mutation (table 3A). The prevalence of CTRC variants in patients with and without p.N34S SPINK1 mutation was comparable in TCP patients (15/148 (10.1%) and 41/332 (12.4%) respectively) and in ICP patients (5/31 (16.1%) and 7/56 (12.5%) respectively) (online supplementary table S2). Thus, in total, 20 out of 179 CP patients (11.2%) with p.N34S SPINK1 mutation also carried CTRC variants as compared with 48 out of 388 patients (12.4%) without p.N34S SPINK1 mutation (OR 0.88, 95% CI 0.49 to 1.59; p=0.66) (table 3A). There was no evidence of any interaction between c.180C>T (p.(=)) CTRC and p.N34S SPINK1 variants in the patients (online supplementary table S3).
Interaction between CTRC variants and p.L26V CTSB mutation
We observed similar results on analysis of interaction between p.L26V CTSB mutation and CTRC variants (table 3B). A total of 297 out of 516 patients (57.6%) were positive for p.L26V CTSB mutation; 255 of 433 TCP patients (58.9%) and 42 of 83 ICP patients (50.6%) carried p.L26V CTSB mutation. One tenth of p.L26V CTSB positive TCP patients (26/255 (10.2%)) also had CTRC variants, while 25 out of 178 (14.0%) p.L26V CTSB negative TCP patients carried CTRC variants. By contrast, 16.7% of p.L26V CTSB positive ICP patients (7/42) were also positive for CTRC variants as compared with only four out of 41 patients (9.8%) who did not carry p.L26V CTSB mutation (online supplementary table S4). Overall, the prevalence of CTRC variants in p.L26V CTSB positive and negative CP patients was similar (33/297 (11.1%) vs 29/219 (13.2%)) and not statistically significant (OR=0.82, 95% CI 0.47 to 1.44; p=0.46) (table 3B).
In this study, we performed comprehensive screening of CTRC gene in a large, ethnically matched case-control CP cohort from India and observed significant over-representation of rare CTRC variants in CP patients compared with normal individuals (71/584 patients vs 22/598 controls; OR 3.62). A synonymous variant c.180C>T (p.(=)) was also found to be significantly associated with CP (OR 2.71). In addition to identification of reported variants, we identified seven novel variants in CTRC whose functional consequences were analysed by in silico analysis. We did not find any evidence of interaction between CTRC variants and earlier reported mutations in SPINK1 or CTSB genes.
Earlier studies have provided genetic and functional evidence for the role of CTRC variants in the pathogenesis of CP.16 ,17 However, with regard to TCP, the studies have been conducted on a small number of patients and controls.16 ,18 Our study on a large cohort of patients and controls reveals that the spectrum of CTRC mutations identified in TCP is quite different from that in Caucasian population supporting earlier observations that provided evidence of a different genetic basis of CP in Indians.7–10 The most significantly associated micro-deletion mutation p.K247_R254del in Caucasians was not identified in TCP patients, while the secretion defective variant p.R254W that was most common in European subjects was very rare in Indian subjects. Instead, p.V235I was the most common variant being present in 5.3% TCP patients as compared with 0.7% controls. This mutation is known to reduce activity and secretion of the CTRC enzyme. Our results are in contrast to earlier studies on TCP patients that reported a lack of association for p.V235I mutation and negligible difference in allele frequencies of several mutations between TCP patients and controls; probably because of small sample size.16 ,18 The second most common variant, p.A73T also showed significant over-representation in patients in contrast to earlier studies on TCP patients. Its pathogenicity has been convincingly proven since it leads to defective secretion of CTRC protein and also elicits endoplasmic reticulum stress in cultured pancreatic acinar cells.24 We also observed a significant association of the synonymous mutation c.180C>T (p.(=)) with pancreatitis, which is consistent with earlier observations 17 but did not find any interaction with the p.N34S SPINK1 variant. Synonymous single nucleotide polymorphisms can affect gene expression by perturbation of mRNA splicing, altering mRNA secondary structure and stability, and altering the rate of translation and co-translational protein folding.25–27 This variant probably takes one of the above-mentioned routes.
As reported earlier, loss-of-function mutations in CTRC are disease-predisposing. In this regard, identification of mutations such as p.W113X and p.G242AfsX9 is of great significance, since they are highly likely to produce a defective protein. Alternatively, the mRNA of these variants may undergo nonsense mediated decay and thus result in no CTRC protein. In either case, these variants are expected to cause a complete loss of CTRC function. The frame-shift mutant may behave in a manner similar to that of the micro-deletion mutant p.K247_R254del as both of them affect the same region. We used several online tools to obtain useful insights into putative functional consequences of the novel missense variations (p.R246C and p.A257T), but found only SIFT to be reliable because only this tool correctly predicted the status of the earlier functionally proven variants p.Q48R and p.A73T. To date, majority of missense mutations identified in CTRC that influence its activity alter highly conserved amino acid residues. ClustalW analysis revealed that amino acid at position 246 is not conserved at all and so a change at this position is unlikely to have a negative effect on the function of protein. On the other hand, the p.A257T variant alters a highly conserved residue and thus could be deleterious for protein function. These conclusions were substantiated by SIFT analysis of these variants. We did not find other rare variants identified in TCP patients in earlier studies. These observations establish heterogeneous distribution of CTRC mutations in different population groups. However, it confirms that mutations in CTRC play an important role in the pathogenesis of CP.
Since p.N34S SPINK1 is the strongest predictor of risk for TCP, we investigated its interaction with CTRC and found that distribution of CTRC variants in p.N34S carriers and non-carriers was similar. Both p.N34S SPINK1 heterozygotes and homozygotes carried CTRC variants. These results are in contrast to the conclusions from studies on Europeans, where a possible relationship between SPINK1 and CTRC variants was proposed.16 Similar results reported in Indian TCP patients may again be due to a small sample size. Since it is known that cathepsin B can activate trypsinogen while CTRC is capable of inactivating trypsinogen, the probable gain of function mutation in CTSB and loss of function mutations in CTRC could increase susceptibility to pancreatitis in Indians. We did not find any significant difference in distribution of CTRC variants in p.L26V CTSB carriers and non-carriers. However, for major variants, a majority of individuals carrying either p.A73T or p.V235I mutation also carried either of the p.N34S SPINK1 or p.L26V CTSB variant. Thus overall, these results suggest a role for CTRC variants in the pathogenesis of TCP independent of SPINK1 and CTSB mutations.
The major strength of our study is the large sample size that has helped us to identify several rare variants in CTRC. It provides evidence of allelic heterogeneity and confirms that CTRC variants play a significant role in its pathogenesis.
The authors express their gratitude to all the patients and normal volunteers for participating in the study. The help of Mr Inder Deo Mali in DNA isolation from blood samples is gratefully acknowledged. The authors would also like to thank Council of Scientific and Industrial Research, Government of India for financial support (NWP0032). SP received an individual fellowship grant from Department of Biotechnology, Government of India, India.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
Funding This research received a specific funding from Council of Scientific and Industrial Research (NWP0032).
Competing interests None.
Patient consent Obtained.
Ethics approval Ethics approval was provided by the Institutional Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.