The deep intronic c.903+469T>C mutation in the MTRR gene creates an SF2/ASF binding exonic splicing enhancer, which leads to pseudoexon activation and causes the cblE type of homocystinuria

Abstract

Deep intronic mutations are often ignored as possible causes of human diseases. A deep intronic mutation in the MTRR gene, c.903+469T>C, is the most frequent mutation causing the cblE type of homocystinuria. It is well known to be associated with pre-mRNA mis-splicing, resulting in pseudoexon inclusion; however, the pathological mechanism remains unknown. We used minigenes to demonstrate that this mutation is the direct cause of MTRR pseudoexon inclusion, and that the pseudoexon is normally not recognized due to a suboptimal 5' splice site. Within the pseudoexon we identified an exonic splicing enhancer (ESE), which is activated by the mutation. Cotransfection and siRNA experiments showed that pseudoexon inclusion depends on the cellular amounts of SF2/ASF and in vitro RNA-binding assays showed dramatically increased SF2/ASF binding to the mutant MTRR ESE. The mutant MTRR ESE sequence is identical to an ESE of the alternatively spliced MST1R proto-oncogene, which suggests that this ESE could be frequently involved in splicing regulation. Our study conclusively demonstrates that an intronic single nucleotide change is sufficient to cause pseudoexon activation via creation of a functional ESE, which binds a specific splicing factor. We suggest that this mechanism may cause genetic disease much more frequently than previously reported.

Figure 1

Recognition of the MTRR pseudoexon with weak 5′ splice site is dependent on the presence of the c.903+469T>C mutation. A: Splicing minigene assay. Upper panel depicts the β-globin and HIV-tat minigenes harboring the wild-type or mutant MTRR pseudoexon. Minigenes were used to transiently transfect COS-7 cells. After RNA isolation the splicing products were analyzed by RT-PCR using minigene-specific primers. The lower bands represent correctly spliced minigene exons, the upper bands represent MTRR pseudoexon inserted between minigene exons. B: 5′ splice site optimization. The suboptimal pseudoexon 5′ splice site (AAG/gtcagc) was converted to an optimal 5′ splice site (variant +3A: CAG/gtaagt) or to the nearly optimal 5′ splice site (variant 13G: CAG/gtgagt), and wild-type and mutant minigenes were analyzed by transfection/RT-PCR. The scores of the different 5′ splice sites assessed by the MaxEntScan program (http://genes.mit.edu/burgelab/maxent/Xmaxentscan_scoreseq.html) [Yeo and Burge, 2004] are shown in the left panel. Right panel shows results from RT-PCR analysis of splicing products.

Figure 2

SF2/ASF regulates MTRR pseudoexon inclusion by binding to an MTRR ESE. A: SF2/ASF overexpression. HEK293 cells were cotransfected by the wild-type or c.903+469T>C mutant MTRR β-globin, and vectors expressing SF2/ASF, SRp40 or SRp55, respectively. After RNA isolation the splicing products were analyzed by RT-PCR using minigene specific primers. The lower band represents correctly spliced minigene exons, the upper band represents the MTRR pseudoexon inserted between the minigene exons. B: RNA interference. HEK-293 cells were transfected with double-stranded RNA oligonucleotides directed toward the SF2/ASF mRNA or a negative control, followed by transfection with the MTRR β-globin or HIV-tat minigenes (left panels). The degree of SF2/ASF downregulation was tested by Western blotting (right panel). The results of the Western blot is shown below the corresponding lanes.

Figure 3

SF2/ASF exclusively binds to the mutant MTRR ESE. Pull-down assay. In vitro transcribed wild-type or mutant MTRR pseudoexon RNAs coupled to biotinylated magnetic beads, or oligonucleotides harboring either the wild-type or mutant MTRR ESE sequence were incubated with HeLa nuclear extract, and the interacting proteins were identified by Western blotting with the use of SF2/ASF or hnRNPA1 antibodies. Sequences of the used oligonucleotides are listed below the pictures.