A wave of deep intronic mutations in X-linked Alport syndrome

Abstract

X-linked Alport syndrome (XLAS) is an inherited kidney disease caused exclusively by pathogenic variants in the COL4A5 gene. In 10-20% of cases, DNA sequencing of COL4A5 exons or flanking regions cannot identify molecular causes. Here, our objective was to use a transcriptomic approach to identify causative events in a group of 19 patients with XLAS without identified mutation by Alport gene panel sequencing. Bulk RNAseq and/or targeted RNAseq using a capture panel of kidney genes was performed. Alternative splicing events were compared to those of 15 controls by a developed bioinformatic score. When using targeted RNAseq, COL4A5 coverage was found to be 23-fold higher than with bulk RNASeq and revealed 30 significant alternative splicing events in 17 of the 19 patients. After computational scoring, a pathogenic transcript was found in all patients. A causative variant affecting COL4A5 splicing and absent in the general population was identified in all cases. Altogether, we developed a simple and robust method for identification of aberrant transcripts due to pathogenic deep-intronic COL4A5 variants. Thus, these variants, potentially targetable by specific antisense oligonucleotide therapies, were found in a high percentage of patients with XLAS in whom pathogenic variants were missed by conventional DNA sequencing. X-linked Alport syndrome (XLAS) is an inherited kidney disease caused exclusively by pathogenic variants in the COL4A5 gene. In 10-20% of cases, DNA sequencing of COL4A5 exons or flanking regions cannot identify molecular causes. Here, our objective was to use a transcriptomic approach to identify causative events in a group of 19 patients with XLAS without identified mutation by Alport gene panel sequencing. Bulk RNAseq and/or targeted RNAseq using a capture panel of kidney genes was performed. Alternative splicing events were compared to those of 15 controls by a developed bioinformatic score. When using targeted RNAseq, COL4A5 coverage was found to be 23-fold higher than with bulk RNASeq and revealed 30 significant alternative splicing events in 17 of the 19 patients. After computational scoring, a pathogenic transcript was found in all patients. A causative variant affecting COL4A5 splicing and absent in the general population was identified in all cases. Altogether, we developed a simple and robust method for identification of aberrant transcripts due to pathogenic deep-intronic COL4A5 variants. Thus, these variants, potentially targetable by specific antisense oligonucleotide therapies, were found in a high percentage of patients with XLAS in whom pathogenic variants were missed by conventional DNA sequencing. Lay SummaryIn ≈10% to 20% of patients with X-linked Alport syndrome, no molecular cause is identified in COL4A5 exons or flanking regions, suggesting the existence of deep-intronic variants. Identification and interpretation of such variants are often challenging, but of utmost importance for genetic counseling. Moreover, the development of oligotherapies is emerging for such intronic variants affecting splicing and represents a therapeutic hope for families. In this study, we developed a diagnosis-sequencing tool based on targeted RNA sequencing that allows us to accurately identify and interpret splicing effect in patients. We successfully identified a high rate of aberrant splicing linked to deep-intronic variants in COL4A5 in almost 90% of patients with no variant in coding regions (17/19). When studying the effect of each variant on splicing, we showed an apparent genotype-to-phenotype correlation in patients. Altogether, these variants, potentially targetable by specific antisense oligonucleotide therapies, are found in a high percentage of patients with X-linked Alport syndrome in whom pathogenic variants are missed by conventional DNA sequencing. In ≈10% to 20% of patients with X-linked Alport syndrome, no molecular cause is identified in COL4A5 exons or flanking regions, suggesting the existence of deep-intronic variants. Identification and interpretation of such variants are often challenging, but of utmost importance for genetic counseling. Moreover, the development of oligotherapies is emerging for such intronic variants affecting splicing and represents a therapeutic hope for families. In this study, we developed a diagnosis-sequencing tool based on targeted RNA sequencing that allows us to accurately identify and interpret splicing effect in patients. We successfully identified a high rate of aberrant splicing linked to deep-intronic variants in COL4A5 in almost 90% of patients with no variant in coding regions (17/19). When studying the effect of each variant on splicing, we showed an apparent genotype-to-phenotype correlation in patients. Altogether, these variants, potentially targetable by specific antisense oligonucleotide therapies, are found in a high percentage of patients with X-linked Alport syndrome in whom pathogenic variants are missed by conventional DNA sequencing. Alport syndrome (AS) is a chronic hematuric glomerulopathy due to mutations in the COL4A3-5 genes encoding for type IV collagen α3 to α5 (α3[IV]–α5[IV]) chains, respectively, which form a network in the glomerular basement membrane, essential for the long-term stability of the glomerular filtration barrier.1Kalluri R. Shield C.F. Todd P. et al.Isoform switching of type IV collagen is developmentally arrested in X-linked Alport syndrome leading to increased susceptibility of renal basement membranes to endoproteolysis.J Clin Invest. 1997; 99: 2470-2478Crossref PubMed Scopus (272) Google Scholar,2Gunwar S. Ballester F. Noelken M.E. et al.Glomerular basement membrane: identification of a novel disulfide-cross-linked network of alpha3, alpha4, and alpha5 chains of type IV collagen and its implications for the pathogenesis of Alport syndrome.J Biol Chem. 1998; 273: 8767-8775Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar Kidney disease is inconstantly accompanied by sensorineural deafness and ocular lesions. Although mutations in COL4A3 and COL4A4, encoding α3(IV) and α4(IV) chains, respectively, are associated with autosomal AS, mutations in COL4A5 lead to X-linked AS (XLAS) and account for ≈50% of AS.3Morinière V. Dahan K. Hilbert P. et al.Improving mutation screening in familial hematuric nephropathies through next generation sequencing.J Am Soc Nephrol. 2014; 25: 2740-2751Crossref PubMed Scopus (127) Google Scholar In accordance with XLAS, male patients are more severely affected than female patients and typically develop kidney failure during their 20s and 30s,4Jais J.P. Knebelmann B. Giatras I. et al.X-linked Alport syndrome: natural history in 195 families and genotype-phenotype correlations in males.J Am Soc Nephrol. 2000; 11: 649-657Crossref PubMed Google Scholar, 5Bekheirnia M.R. Reed B. Gregory M.C. et al.Genotype-phenotype correlation in X-linked Alport syndrome.J Am Soc Nephrol. 2010; 21: 876-883Crossref PubMed Scopus (194) Google Scholar, 6Gross O. Netzer K.O. Lambrecht R. et al.Meta-analysis of genotype-phenotype correlation in X-linked Alport syndrome: impact on clinical counselling.Nephrol Dial Transplant. 2002; 17: 1218-1227Crossref PubMed Scopus (212) Google Scholar with a clear genotype-to-phenotype correlation and with loss-of-function mutations being associated with a more severe phenotype.4Jais J.P. Knebelmann B. Giatras I. et al.X-linked Alport syndrome: natural history in 195 families and genotype-phenotype correlations in males.J Am Soc Nephrol. 2000; 11: 649-657Crossref PubMed Google Scholar, 5Bekheirnia M.R. Reed B. Gregory M.C. et al.Genotype-phenotype correlation in X-linked Alport syndrome.J Am Soc Nephrol. 2010; 21: 876-883Crossref PubMed Scopus (194) Google Scholar, 6Gross O. Netzer K.O. Lambrecht R. et al.Meta-analysis of genotype-phenotype correlation in X-linked Alport syndrome: impact on clinical counselling.Nephrol Dial Transplant. 2002; 17: 1218-1227Crossref PubMed Scopus (212) Google Scholar, 7Said S.M. Fidler M.E. Valeri A.M. et al.Negative staining for COL4A5 correlates with worse prognosis and more severe ultrastructural alterations in males with Alport syndrome.Kidney Int Rep. 2017; 2: 44-52Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar The α5(IV) chain is expressed in the glomerular basement membrane and at the dermoepidermal junction in skin. In male patients with XLAS, especially with severe phenotype, α5(IV) immunostaining on skin biopsy can prove the diagnosis because α5(IV) expression may be completely absent.7Said S.M. Fidler M.E. Valeri A.M. et al.Negative staining for COL4A5 correlates with worse prognosis and more severe ultrastructural alterations in males with Alport syndrome.Kidney Int Rep. 2017; 2: 44-52Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar On the other hand, female patients with XLAS may show a mosaic pattern of α5(IV) expression following X-chromosome inactivation that occurs in female cells.8Nakanishi K. Iijima K. Kuroda N. et al.Comparison of alpha5(IV) collagen chain expression in skin with disease severity in women with X-linked Alport syndrome.J Am Soc Nephrol. 1998; 9: 1433-1440Crossref PubMed Google Scholar At the genetic level, mutation detection rates are about 80%–90% in male patients.3Morinière V. Dahan K. Hilbert P. et al.Improving mutation screening in familial hematuric nephropathies through next generation sequencing.J Am Soc Nephrol. 2014; 25: 2740-2751Crossref PubMed Scopus (127) Google Scholar,9Yamamura T. Nozu K. Minamikawa S. et al.Comparison between conventional and comprehensive sequencing approaches for genetic diagnosis of Alport syndrome.Mol Genet Genomic Med. 2019; 7: e883Crossref PubMed Scopus (23) Google Scholar Considering the strong genetic homogeneity, it seems likely that some variants, localized in deep-intronic regions not reachable by gene panel or whole exome sequencing, are missed in some patients with a clear clinical diagnosis of XLAS, as suggested by other studies.10Wang X. Zhang Y. Ding J. Wang F. mRNA analysis identifies deep intronic variants causing Alport syndrome and overcomes the problem of negative results of exome sequencing.Sci Rep. 2021; 1118097Google Scholar Indeed, noncoding variants affecting gene expression, regulation, or splicing are presumed to cause ≈15% to 30% of human mendelian diseases.11Cartegni L. Chew S.L. Krainer A.R. Listening to silence and understanding nonsense: exonic mutations that affect splicing.Nat Rev Genet. 2002; 3: 285-298Crossref PubMed Scopus (1788) Google Scholar, 12Faustino N.A. Cooper T.A. Pre-mRNA splicing and human disease.Genes Dev. 2003; 17: 419-437Crossref PubMed Scopus (1014) Google Scholar, 13Lewandowska M.A. The missing puzzle piece: splicing mutations.Int J Clin Exp Google Scholar it that of variants can be by affecting The impact and of exonic PubMed Scopus Google Scholar splicing to the of transcripts that can the or in some lead to of the mRNA through in of of a The of whole sequencing allows identification of deep-intronic variants, but it to the variants in the of splicing because of of in not for splicing variant J. et on and for the molecular of Alport syndrome: the J Genet. 2021; PubMed Scopus Google Scholar Thus, interpretation of deep-intronic variants or coding variants that may affect splicing is an of human and is essential to the of variants on Bulk RNA sequencing of sequencing all in a its in the identification of alternative splicing, especially of a gene. the of in is a and potentially for this T. et al.Improving genetic diagnosis in disease with sequencing.Sci Med. 2017; PubMed Scopus Google Scholar The objective of this was to use RNAseq as a analysis to for COL4A5 splicing in patients with XLAS without identified pathogenic variants in COL4A5 exons and flanking The of the is in The patients with XLAS but no pathogenic variant identified in the COL4A5 coding exons or flanking by targeted gene panel sequencing were in the were The is developed in and and the of COL4A5 gene with a in and of the were The transcript of COL4A5 was and were in the splicing events by the of at at the junction and at of the junction were in events were as of alternative splicing events was using a et T. et al.Improving genetic diagnosis in disease with sequencing.Sci Med. 2017; PubMed Scopus Google the between the of the alternative splicing and the of the was not to events as pathogenic or not but to identified events to their importance in the were compared using a and were expressed with correlation was for correlation was as significant and with and clinical are in and 19 patients were XLAS was in all patients because of clinical and history with the with inconstantly associated with chronic kidney disease and/or and/or skin biopsy was not for in patients and patients a α5(IV) collagen chain staining or of staining at the dermoepidermal junction by In in male skin biopsy revealed no staining staining or staining and of α5(IV) collagen and in whom X-linked not be by skin biopsy all with a familial history of XLAS, a hematuric glomerular deafness affecting male patients more severely than female and a likely X-linked with no and a severe in female patients of the patients patients in skin at at kidney type IV collagen chronic kidney not are as of patients in a of at or at COL4A5 chronic kidney not in a type IV collagen chronic kidney not are as of patients chronic kidney not Kidney was in of female patients. female with glomerular filtration rate of at the of male patients worse than with at a of The male kidney at the of and In of the female expression of α5(IV) collagen chain on skin and severe kidney disease with at the of and with of α5(IV) collagen chain staining at the dermoepidermal junction by sequencing analysis in the revealed a pathogenic variant in COL4A5 gene no variant was identified in the We the of deep-intronic variant in the and in not identified by exome panel sequencing, and in with the identified be for the severe in In all all coding exons their flanking intronic of as as other genes in by gene panel sequencing. to identify a pathogenic or pathogenic variant in their of and a variant in of the COL4A5 gene that was absent the and to lead to to We compared the use of bulk RNAseq without with our targeted RNAseq for the detection of COL4A5 splicing events in a of patients and of the of to the of COL4A5 showed In we a coverage of and COL4A5 for bulk and targeted RNAseq, In of patients and bulk RNAseq to in potentially the identification of causative splicing in these patients. is especially for patients and and in On the other hand, by targeted RNAseq, the a coverage of using targeted RNAseq the coverage of COL4A5 by it by for the splicing we by using targeted RNAseq in all because of its high COL4A5 coverage patients and 15 In we identified alternative splicing events in of 19 patients and were because were not controls and The were specific to patients or controls and specific or significant alternative splicing was in male and female In patients with events identified to the of the and In all male patients but was all and In we identified events with and and each for with a COL4A5 coverage and In of female patients we identified significant splicing a in of In we identified with a higher of and but a COL4A5 coverage and In patients in whom no significant splicing was identified and and in patients with COL4A5 coverage in whom identified splicing and we that to a the identification of a pathogenic We this by COL4A5 transcript with targeted RNAseq in the or of After in patients and and in and the COL4A5 coverage increased by a of and in and respectively, but no was identified in and the the of a transcript to in patients and but not in patients and A of alternative splicing events specific to or more than in were identified in patients and to the of in each of In the was the as in but the increased to In we were to between the identified events with because of in increased its to the other was not found in in patients and in whom no significant alternative splicing was identified by targeted RNAseq, not significant and results targeted RNAseq for the pathogenic events are in splicing events were by chain and sequencing. targeted RNAseq with or without to computational scoring, to the identification of pathogenic events in 17 of 19 patients. In 17 of 19 patients in whom a significant COL4A5 splicing was a and likely for the splicing was identified by sequencing in the of on of the identified the When were sequencing the of the variants with the disease in affected patients and absent in in the families on and exonic exonic intronic in a exonic exonic intronic deep-intronic variants and coding variant in were identified in 15 patients 15 and the deep-intronic and patients and the A of the of variants on COL4A5 splicing is in Altogether, in patients and variants were to lead to the of a or the of an that are by the splicing and in with intronic in the In patients and we identified variants affecting a or the use of a In the of an in was likely for a leading to of these variants and of their are in of these variants was in the or by a a expression of the alternative transcript than patients. In we the variant in leading to a of a this variant was in and in with the variant in COL4A5 on the other our of COL4A5 pathogenic variants in the and the severity of disease in At the level, all of these splicing events are to in the of a for the and variants that lead to and We for a correlation between the effect of each splicing and the of of kidney failure in male patients. was not in female patients because of inactivation of the which is in we a significant correlation between the of aberrant transcript to transcripts in patients and the of to male was not in the analysis because at the of at the of the a of transcripts of In female we to of pathogenic transcripts in but we not the inactivation of the In of we an aberrant splicing in In this female the a of of a in of but it was controls We identified a pathogenic variant in in an exonic to splicing of the identified variant was absent in all controls and other patients. was inherited and with the disease in the the with for and of cause in with suggesting a effect of this variant in male patients. these results revealed the importance of genotype-to-phenotype correlation in which be for the of of the disease and for genetic counseling. In this we the of in a molecular cause identification of an aberrant splicing in of patients of with XLAS without identified variants in COL4A5 coding we showed the effect on splicing of coding variant in We that deep-intronic variants the of of XLAS not a pathogenic variant in the coding of Identification of the molecular cause of AS in these families is of utmost importance because it genetic and to kidney for patients with such splicing variants a for targeted therapies, such as antisense type of approach in to in COL4A5 and the of a with hope that this to a severe and the of kidney T. T. T. et of an for X-linked Alport syndrome with variants in 11: PubMed Scopus Google Scholar The of targeted on deep-intronic variants, the of of this type of In this we identified the variant in in families 15 and Moreover, the variant identified in in T. Nozu K. T. et of splicing and genotype-phenotype correlation in X-linked Alport syndrome.J Am Soc Nephrol. PubMed Scopus Google Scholar, K. and revealed by RNA sequencing through Full Text Full Text PDF PubMed Scopus Google Scholar, et and splicing analysis PubMed Scopus Google Scholar, Y. Y. Y. et the detection of splicing events with PubMed Scopus Google Scholar We can that by more we identify variants more to a The identification of these variants seems for genetic counseling. Indeed, these variants may a pathogenic effect on RNA and lead to a of We showed a significant correlation between the of aberrant transcripts and the severity of in patients. this to be in with more and with patients the variant to the impact of a deep-intronic variant is by chain are and the identification of an intronic In some cases, chain cannot in of or in of identified in at a significant level, because it is not a Moreover, the of a genetic variant on splicing may the of a gene because of between or splicing between K. and revealed by RNA sequencing through Full Text Full Text PDF PubMed Scopus Google Scholar the of a variant that splicing are not at its RNAseq allows the identification of alternative splicing, especially of a gene. results a strong of RNAseq, which is the potentially of genes expressed in the can the detection of alternative splicing especially in female patients in whom the a effect of the this in the of a gene of we developed a and method for analysis of deep-intronic or variants based on targeted RNAseq, us to the coverage of the gene of in our and to the of the In this study, we our conventional kidney gene panel of genes of be to coverage of expressed bioinformatic developed to identify alternative splicing events and et and splicing analysis PubMed Scopus Google Scholar, Y. Y. Y. et the detection of splicing events with PubMed Scopus Google Scholar, S. et robust and detection of alternative splicing 2014; PubMed Scopus Google Scholar, N.A. J. et detection and of alternative splicing events with J 2021; PubMed Scopus Google Scholar these not all identified events to which be In our study, the use of a et T. et al.Improving genetic diagnosis in disease with sequencing.Sci Med. 2017; PubMed Scopus Google Scholar based on the importance of compared with identification of the in in whom events were be for the analysis of a group of genes in which mutations are potentially in a with genetic In patients and we not identify a causative splicing can be The is the of alternative splicing in in but not in Indeed, alternative splicing in a of alternative RNA Rev 2003; PubMed Scopus Google The of alternative RNA splicing in human Genet 2017; PubMed Scopus Google Scholar this this analysis to be on kidney that are not in these patients. the of expression of the α5(IV) collagen chain on skin of these patients not this this be the of a variant that not affect splicing but the of the of the COL4A5 gene. a comprehensive method of of expression the identification of such an our splicing to be the events leading to XLAS in not mutation in the coding COL4A5 for detection is the of caused by the of a in with a J. The and of mRNA Rev PubMed Scopus Google Scholar In our study, we identified patients in whom the splicing variant to and the of aberrant RNA for in with seems to be a for detection of these events because in cases, this revealed the aberrant splicing without affecting other alternative is not we can that the coverage by the of genes at this is the of the of is the for The use of of the of as in the of glomerular but this method cannot be in a diagnosis In kidney it that RNA genes expressed in can be in Y. et of renal Am Soc Nephrol. PubMed Scopus Google Scholar it a for analysis of glomerular transcripts without these genes of allows a for sequencing. The interpretation of intronic variants identified by whole sequencing and in splicing the use of such as the developed K. S. et splicing with deep 2019; Full Text Full Text PDF PubMed Scopus Google Scholar to variants. with a a of the variant is but the to with an the variant is in the deep-intronic K. S. et splicing with deep 2019; Full Text Full Text PDF PubMed Scopus Google Scholar In our study, of 15 splicing events are associated with a In the variants as et of variants in J Genet. 2021; Full Text Full Text PDF PubMed Scopus Google Scholar Thus, identification of deep-intronic variants, for which in a of a analysis of their effect on and analysis of whole sequencing and targeted RNAseq or bulk RNAseq the to identify the and its effect on splicing of the transcript In we the importance of for deep-intronic variants in patients with XLAS without pathogenic variant in COL4A5 coding The of a specific and the development of a us to identify a deep-intronic variant causing the in all patients in whom a splicing was identified of for almost 90% of the patients in our We developed an and targeted RNAseq to identify splicing events in patients without a coding pathogenic it seems to targeted RNAseq in our for molecular of patients without pathogenic variants in the coding regions of kidney gene the no We the use of the of the DNA was by the with for targeted RNA sequencing. about the and in the the transcripts at is the of with at between the of the and the of the with a of and of with to in each and the are with a for each junction through all the of the of the the events with events to the of the of to the of of the alternative splicing in each and controls by a for and The is et T. et al.Improving genetic diagnosis in disease with sequencing.Sci Med. 2017; PubMed Scopus Google Scholar COL4A5 coverage between bulk RNA sequencing and targeted The represents the of on the of with the at The represents the COL4A5 We an of and on COL4A5 junction for bulk and targeted RNAseq, In some patients and we COL4A5 coverage junction on RNAseq increased the of to in the COL4A5 targeted RNAseq the of on COL4A5 by a of COL4A5 coverage by targeted RNA sequencing in all the show on COL4A5 RNA in 19 patients and 15 controls and the COL4A5 junction in patients and controls by targeted of on RNA by targeted RNA sequencing and were to as revealed by the in COL4A5 mRNA by and in patients. On the other hand, patients and not to be to and of the effect of all causative variants identified in 17 of 19 patients. all the variants in COL4A5 are with the likely by in for and the for In patients and was using the The is for each as as the of the variant and the of in of The patients and in of splicing to is in patients and The of the is to its In of by we showed the the higher rate of pathogenic in whom we an due to an in we show the the of the and the of coverage of the on the capture panel the the of coverage on this in the other at the may the leading to because no the junction in this was of genes our panel and for targeted RNA sequencing in glomerular are in events identified in COL4A5 by targeted RNA sequencing without by junction was identified or of the The was not and is by to controls but not to controls are in events the in patients are in not no was identified in the was identified in COL4A5 by targeted RNA sequencing with or without junction was identified or of the the was not on and than in this kidney this is or this is to and kidney for of in the kidney of et a and population of were kidney and a kidney biopsy at kidney these showed an in kidney but this was not associated with on PDF

Topics & Concepts