Mutated FOXC1s Are Involved In Congenital Heart Disease By Abnormally Activating The Expression Of NKX2-5 Directly

As an important transcription factor,FOXC1 plays a key role in cardiovascular development of vertebrate.Although the phenotype of Axenfeld-Rieger syndrome patients with congenital heart disease(CHD)are co-segregated with FOXC1 mutation,there is no direct evidence to support that FOXC1 mutation is the cause of CHD.To clarify the role and mechanism of FOXC1 mutation in CHD,we used rat cardiomyocyte cell line(H9c2)as a model to explore the regulation relationship between FOXC1 and NKX2-5.The results showed that FOXC1 regulates NKX2-5 expression in a dose-dependent manner via direct binding to three classical binding sites in the proximal NKX2-5 promoter.These results demonstrate that the mechanism of Foxc1a in zebrafish directly regulating the expression of nkx2.5 is conserved in mammalian cardiomyocytes cell lines.Based on the above findings,we used the rat cardiomyocyte cell line(H9c2)as a model to validate the function of 8 kinds of FOXC1 mutations that are identified in the patients suffering from Axenfeld-Rieger Syndrome(ARS)with CHD.The results showed that,compared to wild-type FOXC1,FOXC1(p.R170W)and FOXC1(p.G149D)could abnormally inhibit the transcription of NKX2-5,FOXC1(p.Q70Hfs*8),and FOXC1(p.S82T)lose the transcriptional activation ability of NKX2-5,and FOXC1(p.L86F)and FOXC1(p.F112S)abnormally increase the transcriptional activation of NKX2-5,while FOXC1(p.A85P)and FOXC1(p.R127L)normally regulate the transcription of NKX2-5.Interestingly,after overexpression of these mutant FOXC1s in zebrafish embryos,the ability of seven mutants FOXC1s except FOXC1(p.G149D)regulating the endogenous nkx2.5 expression was consistent with the results observed in the rat cardiomyocyte cell line H9c2.FOXC1(p.G149D)has a stronger ability to activate expression of nkx2.5 than wild-type FOXC1 in zebrafish embryo,which was completely opposite to its transcriptional inhibition results in cell lines H9c2.To confirm the clinicopathologic significance of these mutant FOXC1s,we overexpressed two representative mutatant FOXC1s in foxc1a homozygous mutant zebrafish embryos,and the results showed that overexpression of FOXC1(p.R170W)further reduced the ventricle minor axis shortening fraction of foxc1a homozygous mutant embryos.While overexpression either FOXC1(p.G149D)or wild-type FOXC1 could rescue the decreased ventricle minor axis shortening fraction of foxc1a homozygous mutant embryos.The results are consistent with the inhibition of NKX2-5 expression by FOXC1(p.R170W)and the activation of nkx2.5 expression by FOXC1(p.G149D)in zebrafish embryos.To understand the frequency of FOXC1 mutations in sporadic CHD patients,we collected and analyzed the genotypes of FOXC1 in 48 CHD abortion fetuses and 13 non-CHD abortion fetuses.The results showed that two"mutations"of suspected gene polymorphism FOXC1(p.G380?A381ins G)and FOXC1(p.G456?Q457ins G)were found in both CHD abortion fetuses and non-CHD abortion fetuses.Compared with overexpression of wild-type FOXC1,overexpression of these two"mutations"FOXC1 in rat cardiomyocyte cell line(H9c2)resulted in abnormal up-regulation of the transcriptional activation of NKX2-5promoter.While overexpression of these two"mutations"FOXC1 in zebrafish embryos had the same ability to up-regulate the expression of nkx2.5 in embryos as overexpression of wild-type FOXC1.Consistent with the results obtained from zebrafish embryos,there was no significant difference between the expression of NKX2-5 in heart tissue of various FOXC1genotypes between non-CHD abortion fetuses and CHD abortion fetuses.In order to further confirm the inconsistency of cell line model in evaluating the function of human mutant gene,we repeated the evaluation experiment of FOXC1 function in human embryonic renal epithelial cell line HEK293T.The results showed that FOXC1 inhibited the activity of NKX2-5 promoter in a dose-dependent manner in HEK293T,contrary to its role in H9c2 cell line.To reveal the mechanism of these opposite effects,we compared the transcriptome data of H9c2 and HEK293T cell lines and found that 30 transcription factors including Hoxc11 were specifically expressed in H9c2.Further studies have shown that Hoxc11 can regulate NKX2-5 expression in cell line H9c2 via its homeobox domain directly binding to at least five binding sites in the proximal NKX2-5 promoter.In order to verify whether the mechanism that Hoxc11 directly regulate the activity of NKX2-5 promoter is conserved in vivo,zebrafish embryos were selected as the research object.We genearted hoxc11a knockout zebrafish and hoxc11b knockout zebrafish by CRISPR/Cas9.Consistent with no expression of genes hoxc11a and hoxc11b in the embryonic developing heart,the embryonic development,growth and reproduction of knockout zebrafish were normal.Because RT-PCR showed that hoxc11a and hoxc11b were expressed in adult zebrafish hearts,transcriptome data of adult zebrafish hearts carrying hoxc11a knockout or hoxc11b knockout alleles were analyzed.The results showed that nkx2.5 was not in differentially expressed genes'lists,which was supported by RT-q PCR results.On the other hand,the expression of hoxc11b was not up-regulated in hoxc11a knockout adult zebrafish hearts,and vice versa,which were suggesting that there was no gene functional compensation effect between hoxc11a and hoxc11b in adult zebrafish hearts.These results suggest that hoxc11a and hoxc11b are not necessary for the formation and function of zebrafish heart.The molecular biological process that Hoxc11 directly regulates the expression of NKX2-5 in rat cardiomyocyte cell line H9c2 does not exist in modal animal individuals.In conclusion,our results demonstrated that the data from zebrafish model are more consistent with the molecular biology data from patients than the ones from the cell line model in evaluating the function of FOXC1 mutations.Evaluating the function of mutant FOXC1s showed that the FOXC1(p.R170W)may be involved in CHD through abnormal inhibition of NKX2-5 transcription;FOXC1(p.Q70Hfs*8)and FOXC1(p.S82T)may be involved in CHD through loss of transcriptional activation of NKX2-5;FOXC1(p.G149D),FOXC1(p.L86F)and FOXC1(p.F112S)may be involved in CHD through abnormal increase of transcriptional activation of NKX2-5;while FOXC1(p.A85P)and FOXC1(p.R127L),together with two suspected polymorphisms found in sporadic CHD patients,FOXC1(p.G380?A381ins G)and FOXC1(p.G456?Q457ins G),are not involved in CHD(not through abnormal regulation of NKX2-5 expression at least).