The Function And Mechanism Of CPA1 Variants And Splice Donor Site Mutations In Chronic Pancreatitis

Part One Identification and Functional Analysis of CPA1 Variants in Chinese Han Patients with Idiopathic Chronic PancreatitisBackground and Objective: Chronic pancreatitis?CP?is a chronic inflammatory process of the pancreas that leads to irreversible morphological changes and progressive impairment of both exocrine and endocrine functions.The leading cause of CP is excessive alcohol consumption,followed by so-called idiopathic chronic pancreatitis?ICP;defined as the absence of any identifiable etiology prior to genetic analysis?.In 2013,Nature Genetics firstly reported that functionally impaired variants of the CPA1?encoding carboxypeptidase A1?gene were strongly associated with CP in a German cohort.In vitro functional analysis of disease-causing missense variants shows that the mechanism of increased pancreatitis risk by CPA1 variants may involve misfolding-induced endoplasmic reticulum stress?ERS?considering that activity of most defective variants was due to markedly reduced secretion,rather than elevated intrapancreatic trypsin activity.In addition,CP patients carrying the CPA1 variants were then found in a Japanese cohort and members of Polish and Hungarian families.Because single nucleotide polymorphisms vary widely between races,replication in the populations of different ethnicity is critical in order to confirm a disease association.Besides,not all of the CPA1 variants found by sequencing were associated with the pathogenesis of CP,so every variant should be confirmed to be whether the disease-causing variant by functional analysis.In order to identify the CPA1 variants in Chinese Han population,we performed sequencing of the CPA1 gene in Chinese Han patients with ICP and ethically matched healthy controls.Then we performed functional analysis to elucidate the pathogenicity and pathogenesis of every newly found variant.Methods: 1.We performed targeted next generation sequencing of the entire coding sequence and exon/intron boundaries of the CPA1 gene in 1436 Chinese Han patients with ICP and 1580 ethically matched healthy controls,and used Sanger sequencing to validate all called variants.2.The wild-type CPA1 c DNA sequence was cloned into the pc DNA3.1/V5-His-TOPO vector by means of TA cloning.3.Variants were introduced into the wild-type CPA1 c DNA expression constructs by means of the Site-Directed Mutagenesis Kit.4.Plasmids were transfected into HEK293 T cells.After 48 hours,total RNA was extracted and reverse transcription was performed to obtain c DNA.Conventional PCR analyses and real-time quantitative PCR analyses were conducted.5.The relative expression level of ERS markers?namely Bip?HSPA5?and XBP1?between the wild-type plasmid and the variant plasmids was measured by real-time quantitative PCR.6.The relative expression level of proteins between the wild-type plasmid and the variant plasmids was measured by Western blotting.Results: 1.A total of 26 rare CPA1 missense variants were characterized,8 of which have not been previously described and ICP patients carry 7 newly found variants: c.118A>C,c.344A>G,c.446G>T,c.542G>A,c.661A>G,c.693C>G and c.1213C>G.2.The wild-type CPA1 c DNA plasmid and variant plasmids were successfully obtained.3.The wild-type plasmid expressed normal transcripts and variant plasmids expressed the transcripts containing the variant sites.4.The relative expression level of transcripts between the wild-type plasmid and the variant plasmids was consistent.5.The relative expression level of Bip?HSPA5?and XBP1 from c.446G>T,c.693C>G and c.1213C>G was highly elevated and it illustrates that c.446G>T,c.693C>G and c.1213C>G elicit ERS.6.The relative expression level of proteins from c.446G>T,c.693C>G and c.1213C>G was highly reduced.Combined with above results,it illustrates that c.446G>T,c.693C>G and c.1213C>G increase pancreatitis risk by involving misfolding-induced ERS that leads to the protein destruction and reduction.Conclusions: 1.The genetic characteristics of idiopathic chronic pancreatitis differ between Chinese Han population and European population.CPA1 variants are ethnic specificity.2.Some of the newly found variants in Chinese Han population are associated with the pathogenesis of chronic pancreatitis due to endoplasmic reticulum stress.Part Two Functional Analysis of Loss-of-Function CPA1 Variants in Chronic PancreatitisBackground and Objective: It has been reported that some genes are associated with the pathogenesis of chronic pancreatitis?CP?.Nemeth et al.made a database about the genetic risk factors in CP?http://www.pancreasgenetics.org/?including PRSS1,PRSS2,SPINK1,CTRC,and CPA1,in which CPA1 non-synonymous variants consist of missense,nonsense,frameshift,and splice site variants.Unlike missense variants,loss-of-function?Lo F?CPA1 variants?i.e.,nonsense,frameshift,and splice site variants?often result in abnormal transcripts and proteins totally different from the normal ones due to the significant change in codon sequence.The mechanism of increased pancreatitis risk by CPA1 variants may involve misfolding-induced endoplasmic reticulum stress?ERS?considering that activity of most defective variants was due to markedly reduced secretion.However,Lo F variants generate transcripts that contain premature termination codons and are thus prone to nonsensemediated RNA decay?NMD?.NMD detects and degrades PTC-containing transcripts,thereby preventing the accumulation of truncated proteins.This implies that most Lo F CPA1 variants would not be able to elicit ERS and hence will not predispose to CP.In order to identify the effect of Lo F CPA1 variants on CP,we searched these variants from the database.Then we performed functional analysis and in silico splicing prediction to elucidate the pathogenicity and pathogenesis of them.Methods: 1.We searched Lo F CPA1 variants in the Genetic Risk Factors in Chronic Pancreatitis Database.2.The wild-type CPA1 full-length sequence and c DNA sequence were cloned into the pc DNA3.1/V5-His-TOPO vector by means of TA cloning;the wild-type CPA1 exon 3 sequence and exon 9 sequence were cloned into the p ET01 vector by means of In-Fusion cloning.3.Nonsense variants were introduced into the wild-type CPA1 c DNA expression constructs and splice site variants were introduced into the wild-type CPA1 exon 3 and exon 9 minigene expression constructs.4.Plasmids were transfected into HEK293 T cells.After 48 hours,total RNA was extracted and reverse transcription was performed to obtain c DNA.Conventional PCR analyses and real-time quantitative PCR analyses were conducted.5.The relative expression level of transcripts and ERS markers?namely Bip?HSPA5?and XBP1?between the wild-type plasmid and the nonsense variant plasmids,and the relative expression level of the nonsense variant plasmids with and without cycloheximide?a NMD inhibitor?treatment were measured by real-time quantitative PCR.6.The relative expression level of proteins between the wild-type plasmid and the nonsense variant plasmids was measured by Western blotting.7.In silico prediction was performed by means of Alamut Visual software suite for the splice site variants.Results: 1.Of the CPA1 variants registered in the Genetic Risk Factors in Chronic Pancreatitis Database,c.79C>T,c.357C>A and c.954₉55del CA can be regarded as nonsense variants whilst c.148-1G>A,c.1072+1G>T and c.1073-2A>G are classified as splice site variants.2.The wild-type plasmids and variant plasmids were successfully obtained.3.c.79C>T,c.357C>A and c.954₉55del CA expressed the transcripts containing the variant sites,whilst c.148-1G>A resulted in the loss of the frst 65 bp of exon 3 from the normal transcripts and c.1072+1G>T led to exon 9 skipping.4.The relative expression level from c.79C>T,c.357C>A and c.954₉55del CA was reduced compared to that of the wild-type plasmid and it was elevated after cycloheximide treatment;the relative expression level of Bip?HSPA5?and XBP1 from these three nonsense variants was reduced.It illustrates that these three nonsense variants generate variant transcripts that would be degraded by the NMD pathway and cannot elicit ERS.5.The relative expression level of proteins from c.79C>T,c.357C>A and c.954₉55del CA was highly reduced.Combined with above results,it is due to the reduced expression of transcripts,rather than eliciting ERS and protein destruction.Therefore,CPA1 nonsense variants are not associated with the pathogenesis of CP.6.c.148-1G>A and c.1072+1G>T generated transcripts containing premature termination codons,which suggests that the transcripts would be reduced due to NMD pathway and the low amount of proteins could not elicit ERS.Therefore,CPA1 splice site variants are not associated with the pathogenesis of CP.7.In silico splicing prediction with Alamut Visual software suite was not able to accurately predict the transcripts of splice site variants.Conclusions: 1.The relative expression level of the transcripts generated by loss-of-function CPA1 variants decreased by virtue of nonsense-mediated m RNA decay so that low amount of proteins is not able to elicit endoplasmic reticulum stress and will not predispose to chronic pancreatitis.2.In silico splicing prediction software is not able to accurately predict the transcripts of splice site variants and functional analysis is still the gold standard.Part Three Expression of Canonical Splice Donor Site GT>GC Mutations and its Application to Chronic Pancreatitis-Predisposing GenesBackground and Objective: The process of coding gene expression comprises transcription and translation.The introns of a newly made precursor RNA?pre-m RNA?are removed via splicing in order to form mature m RNA for translation.The 9-nt consensus sequence for the U2-type splice donor site?SDS?is relatively conserved and the GT dinucleotide in the first two intronic positions?in the context of DNA sequence?plays an important role in the process of splicing.The most frequent variant in Chinese Han patients with chronic pancreatitis?CP?is SPINK1 IVS3+2T>C?c.194+2T>C?.It is a splice site variant leading to the substitution of the canonical SDS GT by GC?termed a SDS GT>GC mutation?and allows normal splicing to occur to express the wild-type transcripts,retaining the function of translating the normal proteins.Besides SPINK1,some other disease-causing SDS GT>GC mutations were also reported to generate wild-type transcripts.Notwithstanding,the actual scale of canonical SDS capable of generating wild-type transcripts in the case of GT>GC substitutions remains unknown and the study about SDS GT>GC mutation in CP is limited to SPINK1 IVS3+2T>C.Therefore,finding those sites where a SDS GT>GC mutation may allow normal splicing to occur in the CP-predisposing genes can better understand the clinical significance of this mutation and help the clinical diagnosis and treatment.We aim to investigate the scale of canonical SDS GT>GC mutations generating wildtype transcripts by combining the “in vivo” dataset and the “in vitro” dataset and to predict the splicing and clinical phenotypes of canonical SDS GT>GC mutations in chronic pancreatitis-predisposing genes.Methods: 1.We performed a meta-analysis of disease-causing SDS GT>GC mutations logged in the Professional version of Human Gene Mutation Database?HGMD?,with a view to generating an “in vivo” dataset to estimate the scale of SDS GT>GC mutations generating wild-type transcripts.2.The wild-type full-length gene sequences were cloned into the pc DNA3.1/V5-His-TOPO vector by means of TA cloning or into the pc DNA3.1?+?vector by means of In-Fusion cloning.3.SDS GT>GC mutations were introduced into the wild-type full-length gene expression constructs.4.Plasmids were transfected into HEK293 T cells.After 48 hours,total RNA was extracted and reverse transcription was performed to obtain c DNA.Conventional PCR analyses were conducted to generate an “in vitro” dataset.5.Integrating results from “in vivo” and “in vitro” datasets,we constructed the pictograms of the 9-nt SDS signal sequences associated with SDS GT>GC mutations and predict the splicing and clinical phenotypes of canonical SDS GT>GC mutations in chronic pancreatitis-predisposing genes.Results: 1.Seven?15.6%?of the 45 informative disease-causing SDS GT>GC mutations in the “in vivo” dataset and nineteen?18.4%?of the 103 SDS GT>GC mutations in the “in vitro” dataset were found to have been capable of generating some correctly spliced transcripts respectively.2.The “in vitro” dataset confirms that SPINK1 IVS3+2T>C could generate wild-type transcripts whilst all the four SDS GT>GC mutations of PRSS2 could not.3.Pictograms showed that the canonical 9-nt SDSs whose substitutions of GT by GC generated normal transcripts exhibited stronger complementarity to the 5' end of U1 sn RNA than those sites whose substitutions of GT by GC did not lead to the generation of normal transcripts.The conserved sequence is ag GUAAG.4.According to the results of pictograms,we predict that CPA1 IVS5+2T>C and CTRC IVS2+2T>C can give rise to wild-type transcripts whilst all the four SDS GT>GC mutations of PRSS1 cannot.Conclusions: 1.Certain SDS GT>GC mutations are capable of generating wild-type transcripts.2.Among the SDS GT>GC mutations in chronic pancreatitis-predisposing genes,some are capable of generating wild-type transcripts and retain the function of translating the normal proteins,presenting milder clinical symptoms.