Molecular basis of the human ApoAII exon 3 splicing

Arrisi Mercado, Pablo (2004). Molecular basis of the human ApoAII exon 3 splicing. PhD thesis The Open University.



Among human genes, CFTR intron Vlll/exon 9 and apolipoprotein All intron Il/exon 3 boundaries share the characteristic feature given by the presence of a peculiar tract of alternating pyrimidines and purines close to the 3’ splice site.
In the case of CFTR gene, the pyrimidine rich tract is composed by a stretch of Ts in a row and a stretch of alternating pyrimidines and purines (microsatellite TG dinucleotide repeats) and both tracts are polymorphic for their length. On the other hand, in the case of ApoAII gene the pyrimidine rich tract is made exclusively of alternating pyrimidines and purines (microsatellite TG dinucleotide repeats) and it is also polymorphic for its length in both genes. This apparent sequence similarity, concerning the TG tract, is contrasted by the different splicing pattern exhibited by the two genes. In fact, CFTR exon 9 undergoes alternative splicing to a variable extent, depending on the variations in length of the pyrimidine rich tract, whereas ApoAII exon 3 is constitutively included in mRNA. Previous studies of our group have shown in the CFTR intron Vlll/exon 9 context, the stretch of pyrimidines alternated with purines (within the UG tract) alone is not equivalent to a functional continuous polypyrimidine tract, contrarily to what has been observed for the apolipoprotein All gene. Moreover, the comparison of splice sites strength of human ApoAII exon 3 and human CFTR exon 9 has outlined an apparent contradiction between the splicing behavior of the two exons and the strength of the splice sites. In fact, both the 3’ and the 5’ splice sites of CFTR exon 9 display a good match with the consensus whereas the match of the ApoAII exon 3 splice sites is not good. Altogether these observations prompted us to investigate the mechanisms underlying the constitutive splicing of ApoAII exon 3 and, in particular, to characterize the cis-acting elements and the trans-acting factors involved in ApoAII exon 3 definition to assure its constitutive splicing.
In order to study in vivo the splicing mechanism of ApoAII exon 3, we set up an eukaryotic expression system by cloning the whole ApoAII gene, from its promoter to the poly-A signal. Then, the effects of point mutations, deletions or substitutions on splicing of exon 3 were analyzed by RT-PCR after transient transfection in Hep3B cell line.
Deletion or replacement of the UG repeats at the 3’ splice site of intron 2 resulted in a significant increase in exon 3 skipping, indicating the importance of this alternated arrangement of U and G as a functional polypyrmidine tract or at least as an important sequence able to lead the exon 3 definition. Furthermore, UV-crosslinking assays showed that the (UG)I6 repeats of ApoAII intron 2 are recognized by TDP-43, a protein that binds specifically the UG tract within the context of the 3' end of CFTR intron VIII and that affects negatively CFTR exon 9 splicing. Next, we characterized the exonic cis-acting elements able to affect the splicing efficiency of ApoAII exon 3. Transient transfections of different constructs of the ApoAII gene system carrying deletions or point mutations showed that the region spanning from nucleotide 87 to 113 of human ApoAII exon 3 is important for its close to the 5’ splice site. In order to identify trans-acting factor/s able to bind the 9nt-ESE within ApoAII exon 3, both Electro Mobility Shift Assay (EMSA) and UV-crosslinking coupled to immunoprécipitation assays were carried out. EMSA showed a broad band of shifted material with the ESEwt RNA, whereas no significant shift was seen with mutated ESE RNA. This suggested that one or more proteins interact specifically with the wild type ApoAII exon 3 across the 9nt-sequence. Then, UV-crosslinking followed by immunoprécipitation with monoclonal antibodies anti-SR proteins ASF/SF2 and anti-SC-35 showed that ESEwt but not mutated ESE RNA was able to immunoprecipitate a band whose molecular weight corresponds to that of ASF/SF2 and, even if to a lower extent, also a band whose molecular weight corresponds to that of SC35. Thus, these results provided an evidence that at least two SR proteins are able to interact with the sequence across the ESE sequence.
Subsequently, the relevance of ESE position within an internal intron (and therefore with other flanking cis-acting elements) was also tested. Both in vitro and in vivo experiments showed that the ApoAII exon 3 ESE works only if it is localized within an internal exon and not if it is placed within the first or the last exon. These results also suggested the presence of other splicing regulatory elements within the flanking intronic regions of ApoAII exon 3. Thus, to explore intron 2 and 3 for the presence of cis-acting elements able to affect the splicing of exon 3, a series of deletions within both introns were carried out. Thus, we found at least one regulatory element placed within intron 3 that regulates positively exon 3 inclusion.
In conclusion, the constitutive splicing of ApoAII exon 3 seems to be the result of the balance between positive and negative action of the regulatory elements found in the exon 3 and its flanking introns. Future studies will be aimed at identifying the factors interacting with the intronic regulatory elements and at defining a possible model to explain the mechanism of ApoAII exon 3 constitutive splicing by integrating the network of interactions among the identified cis-acting elements and trans-acting factors.

Viewing alternatives

Download history


Public Attention

Altmetrics from Altmetric

Number of Citations

Citations from Dimensions

Item Actions