Splicing dysregulation contributes to the pathogenicity of several F9 exonic point variants

Abstract

Background: Pre-mRNA splicing is a complex process requiring the identification of donor site, acceptor site, and branch point site with an adjacent polypyrimidine tract sequence. Splicing is regulated by splicing regulatory elements (SREs) with both enhancer and suppressor functions. Variants located in exonic regions can impact splicing through dysregulation of native splice sites, SREs, and cryptic splice site activation. While splicing dysregulation is considered primary disease-inducing mechanism of synonymous variants, its contribution toward disease phenotype of non-synonymous variants is underappreciated.

Methods: In this study, we analyzed 415 disease-causing and 120 neutral F9 exonic point variants including both synonymous and non-synonymous for their effect on splicing using a series of in silico splice site prediction tools, SRE prediction tools, and in vitro minigene assays.

Results: The use of splice site and SRE prediction tools in tandem provided better prediction but were not always in agreement with the minigene assays. The net effect of splicing dysregulation caused by variants was context dependent. Minigene assays revealed that perturbed splicing can be found.

Conclusion: Synonymous variants primarily cause disease phenotype via splicing dysregulation while additional mechanisms such as translation rate also play an important role. Splicing dysregulation is likely to contribute to the disease phenotype of several non-synonymous variants.

Keywords: hemophilia B; in silico splicing analysis; in vitro minigene assay; splicing dysregulation; synonymous variants.

Figure 1

F9 exonic variants predicted to affect splicing by in silico splice site tools. Venn diagram shows the overlap between F9 exonic variants predicted to affect splicing by in silico splice site prediction tools employed in this study; NNsplice, MaxEntScan (MES), SpliceSiteFinder‐like (SSF‐like), and Human Splicing Finder (HSF)

Figure 2

Location of F9 exonic splicing variants predicted to affect splicing by all in silico splice site tools. Image shows the location of F9 exonic variants that were predicted to affect splicing by all four in silico splice site tools (NNsplice, MaxEntScan, SpliceSiteFinder‐like, and Human Splicing Finder). Bars indicate the number of variants identified at each location

Figure 3

In vitro minigene analysis of select F9 variants in exons 4, 5, and 6. Representative agarose gel image showing the PCR amplification products of the studied F9 variants. Reporter minigene constructs were transfected into HEK293 cells and splicing pattern was assessed by PCR amplification of cDNA generated from total RNA. Expected sizes of PCR amplified products for exon 4, 5, and 6 reporter constructs were 360, 375, and 449 bp, respectively, in the event of normal splicing (exon inclusion) and 246 bp in the event of exon skipping. Exon skipping induced by individual variants was calculated using ImageJ software and numbers were shown in Table 5. The characteristics of tested variants with regard to disease‐causing (+) or not (−), synonymous (+) or not (−), number of tools (out of four in silico splice site tools; NNsplice, MaxEntScan, SpliceSiteFinder‐like, and Human Splicing Finder) that predicted splice site dysregulation, the extent of exon skipping and/or mis‐splicing in minigene assays, and their exonic locations are indicated below

Figure 4

In vitro translation analysis of c.484C>A (p.R162R) variant. Panel (a) shows the representative autoradiogram of FIX wild‐type (WT) and FIX p.R162R (c.484C>A) in vitro translation products. Panel (b) shows the graphical representation of intensities of in vitro translation products. Error bars represent SD values