Discussion

The present study is the first to investigate the impact of genetic variants in miRNA-machinery genes on CHD susceptibility. Our findings demonstrated that the XPO5 rs11077 CC genotype was significantly associated with an increased CHD risk. A clear tendency was also found for DROSHA rs10719 AA genotype, although the association did not reach statistical significance. Interestingly, the analysis of the joint effect of single-nucleotide polymorphisms in miRNA-machinery genes showed that higher scores were associated with increased risk of CHD, supporting the utility of a genetic risk score to provide a stronger predictive value than any single variant.

Accumulating evidence has revealed the highly complicated regulatory networks governing heart development, underscoring the importance of genetic and epigenetic factors in CHD development. Although various forms of CHD have been associated with altered expression of miRNAs, their causal role in the pathogenesis of CHD remains to be elucidated. ^{Reference Pulignani and Andreassi4}

Our analysis focused on potentially functional single-nucleotide polymorphisms located in genes involved in the biogenesis and processing of miRNAs. In general, miRNAs are formed in a multi-step biological process involving proteins and protein complexes, including DICER, DROSHA, and XPO5. ^{Reference Bartel12} The presence of polymorphic variants in the core components of miRNA-machinery genes may impair or enhance miRNA processing efficiency or function, resulting in altered levels of mature miRNAs with potential deleterious effects. ^{Reference Borghini and Andreassi5,Reference Cammaerts, Strazisar, De Rijk and Del Favero6}

XPO5 is a protein responsible for the transport of pre-miRNAs between the nucleus and cytoplasm. The overexpression of XPO5 results in enhanced miRNA activity, ^{Reference Yi, Doehle, Qin, Macara and Cullen13} whereas the loss of XPO5 leads to reduced expression and function of pre-miRNAs. ^{Reference Zeng and Cullen14}

Among the genetic variants in XPO5, rs11077 has received the most attention. Located in the 3’-UTR, the A > C substitution may modify the expression of XPO5 and, consequently, alter the expression of miRNAs. ^{Reference Guo, Shu, Zhou and Zhang7,Reference Shao, Shen, Zhao, Guo, Niu and Liu15} The rs11077 C allele has been associated with an increased risk of developing cancer ^{Reference Shao, Shen, Zhao, Guo, Niu and Liu15} and venous thromboembolism, ^{Reference Ko, Kim and Kim16} but it also appears to decrease the risk of coronary artery disease. ^{Reference Borghini, Pulignani and Mercuri10}

To our knowledge, no previous studies associating these specific single-nucleotide polymorphisms with CHD are available. In our work, this polymorphic variant resulted associated with an increased risk of disease revealing a potential role of XPO5 miRNA-machinery gene in CHD susceptibility. An increased risk of disease was also found for the DROSHA rs10719 AA genotype, although the results did not reach statistical significance.

DROSHA is an RNAse III enzyme, which mediates the processing of pri-miRNAs into pre-miRNAs inside the nucleus. In a previous in vitro functional investigation, a reduction in miRNA processing efficacy induced by the knockdown of DROSHA was found to reduce the levels of mature forms of tumour-suppressive miRNAs. ^{Reference Noh, Hong, Dong, Pan, Jing and Huang17} In addition, the loss of DROSHA leads to vascular smooth muscle cells disorder followed by hypoplastic blood vessel walls, cardiomyopathy, and embryonic lethality in mice. ^{Reference Fan, Chen and Tian18}

The 3′-UTR rs10719 polymorphism alters DROSHA expression levels. ^{Reference Saeki, Watanabe and Inoue19} It has been associated with different kinds of cancers ^{Reference Wen, Lv, Ding, Fang and Sun20} as well as recurrent implantation failure. ^{Reference Lee, Ryu and Kim21} Interestingly, these single-nucleotide polymorphisms may affect the risk of non-syndromic orofacial clefts which represent major craniofacial congenital abnormalities characterised by unsuccessful fusion of normal facial processes. ^{Reference Xu, Ma and Lou22}

Previous studies have already proven the important role of DICER in the craniofacial development. ^{Reference Barritt, Miller and Scheetz23} DICER is essential for the survival of the tissues derived from neural crest cells which exerts an essential influence on the formation of a majority of embryo tissues, including craniomaxillofacial structures. ^{Reference Zehir, Hua, Maska, Morikawa and Cserjesi24}

Deletion of cardiac DICER also results in defective heart morphogenesis and embryonic lethality. ^{Reference Zhao, Ransom and Li25} Cardiac outflow defects were observed in mice with a conditional knockout of the DICER gene in the developing heart. In particular, embryos lacking DICER displayed a highly penetrant phenotype of double outlet right ventricle with a concurrent ventricular septal defect. ^{Reference Saxena and Tabin26} Nevertheless, the analysis of DICER mutations in sporadic and familial cases did not appear to play a major role in the transposition of the great arteries. ^{Reference Sabbaghian, Digilio, Blue, Revil, Winlaw and Foulkes27}

DICER is an enzyme responsible for the cleavage of pre-miRNAs into their mature form. The rs1057035 T>C polymorphism is located within 3′-UTR and the C allele may lead to reduced expression levels of the enzyme as compared to the T allele. ^{Reference Ma, Yuan and Yuan28}

To date, conflicting results remain about the association between DICER rs1057035 and cancer risk in Caucasian populations, ^{Reference Yu, Kuang and Yin29} whereas it seems to be involved in coronary artery disease. ^{Reference Borghini and Andreassi5} In our study, no significant association was observed between this genetic variant and CHD susceptibility.

To overcome the limited effects of individual risk variants, we further constructed a genetic risk score by summing the risk alleles of the selected variants present in each study participant. The genetic risk score approach allows combining multiple loci with low-modest effects into a global risk score facilitating the identification of high-risk patients.

Interestingly, in the analysis of the combined effects of single-nucleotide polymorphisms in miRNA-machinery genes, the resultant genetic risk score predicted a 1.73 risk for CHD per risk allele. In addition, individuals in the top tertile score had a three-fold higher CHD risk, thus supporting the utility of the genetic risk score approach to increase understanding of the genetic basis of CHD.

The results of this case–control study should be approached and interpreted bearing in mind some limitations. Firstly, the number of enrolled patients and controls is limited. This does not allow to focus on the impact of the single-nucleotide polymorphisms on different types of CHD. Secondly, the study did not evaluate the relationship between other genetic variants in miRNA-machinery genes and CHD risk. Finally, the molecular mechanisms underlying the impact of these polymorphisms in CHD have not been investigated.

In conclusion, this study is the first to identify an association between the 3′-UTR XPO5 rs11077 polymorphism and CHD susceptibility. Additionally, our findings indicate that the combined analysis of single-nucleotide polymorphisms in miRNA-machinery genes might participate in the development of CHD.

Further studies are necessary to validate our results in larger populations and other ethnic groups. Experimental studies should also be conducted to understand the biological relevance of DICER rs1057035, DROSHA rs10719, and XPO5 rs11077 polymorphisms in the pathogenesis of cardiac defects. The characterization of these variants and the identification of their functional impact may provide a good starting point for the development of individually tailored miRNA-based therapy. Therapeutic modulation of miRNA expression during cardiac malformations by either inhibition by anti-miRNAs or overexpression by miRNA-mimics may represent a promising approach to enhance clinical outcomes in CHD patients.