Limits...
AG-dependent 3'-splice sites are predisposed to aberrant splicing due to a mutation at the first nucleotide of an exon.

Fu Y, Masuda A, Ito M, Shinmi J, Ohno K - Nucleic Acids Res. (2011)

Bottom Line: RNA-EMSA revealed that wild-type FECH requires U2AF(35) but wild-type LPL does not.Our studies suggest that a mutation at the AG-dependent 3'-splice site that requires U2AF(35) for spliceosome assembly causes exon skipping, whereas one at the AG-independent 3'-splice site that does not require U2AF(35) gives rise to normal splicing.The AG-dependence of the 3'-splice site that we analyzed in disease-causing mutations at E(+1) potentially helps identify yet unrecognized splicing mutations at E(+1).

View Article: PubMed Central - PubMed

Affiliation: Division of Neurogenetics, Center for Neurological Diseases and Cancer, Nagoya University Graduate School of Medicine, 65 Tsurumai, Showa-ku, Nagoya 466-8550, Japan.

ABSTRACT
In pre-mRNA splicing, a conserved AG/G at the 3'-splice site is recognized by U2AF(35). A disease-causing mutation abrogating the G nucleotide at the first position of an exon (E(+1)) causes exon skipping in GH1, FECH and EYA1, but not in LPL or HEXA. Knockdown of U2AF(35) enhanced exon skipping in GH1 and FECH. RNA-EMSA revealed that wild-type FECH requires U2AF(35) but wild-type LPL does not. A series of artificial mutations in the polypyrimidine tracts of GH1, FECH, EYA1, LPL and HEXA disclosed that a stretch of at least 10-15 pyrimidines is required to ensure normal splicing in the presence of a mutation at E(+1). Analysis of nine other disease-causing mutations at E(+1) detected five splicing mutations. Our studies suggest that a mutation at the AG-dependent 3'-splice site that requires U2AF(35) for spliceosome assembly causes exon skipping, whereas one at the AG-independent 3'-splice site that does not require U2AF(35) gives rise to normal splicing. The AG-dependence of the 3'-splice site that we analyzed in disease-causing mutations at E(+1) potentially helps identify yet unrecognized splicing mutations at E(+1).

Show MeSH

Related in: MedlinePlus

RT–PCR analysis of nine disease-causing mutations at E+1. (A) Sequences at the intron/exon junctions of nine pairs of wild-type and mutant constructs. The longest polypyrimidine stretches are underlined. (B) RT–PCR of minigenes transfected into HEK293 cells. Five mutant constructs are aberrantly spliced, whereas the remaining four mutants are normally spliced. Numbers in the parentheses indicate exon numbers. In PKHD1, a cryptic 3′-splice site (open arrowhead in panel A) at 55 nt upstream of the native site is activated (asterisk). Mean and SD of three independent experiments of the densitometric ratios of the exon-skipped product is shown at the bottom.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3105431&req=5

Figure 6: RT–PCR analysis of nine disease-causing mutations at E+1. (A) Sequences at the intron/exon junctions of nine pairs of wild-type and mutant constructs. The longest polypyrimidine stretches are underlined. (B) RT–PCR of minigenes transfected into HEK293 cells. Five mutant constructs are aberrantly spliced, whereas the remaining four mutants are normally spliced. Numbers in the parentheses indicate exon numbers. In PKHD1, a cryptic 3′-splice site (open arrowhead in panel A) at 55 nt upstream of the native site is activated (asterisk). Mean and SD of three independent experiments of the densitometric ratios of the exon-skipped product is shown at the bottom.

Mentions: We next sought for parameters that differentiate normal and aberrant splicings in these minigenes. Analysis of parameters that potentially dictate the strength of the PPT indicated that the length of pyrimidine stretch, the number of pyrimidines in 25 or 50 nt at the 3′-end of an intron correlated with the ratio of exon skipping with correlation coefficients of more than 0.6 (Supplementary Table S1). The number of pyrimidines in 25 or 50 nt at the 3′-end of an intron, however, failed to predict splicing consequences of nine other constructs shown in Figure 6, and is likely to be overfitted parameters unique to the 35 constructs in Figure 4. Coolidge and colleagues report that (GU)11 in PPT is partly functional, but we did not observe alternative purine and pyrimidine residues in our PPTs and did not quantify effects of alternative nucleotides (10). We thus took the length of pyrimidine stretch as a best parameter to dictate the strength of the PPT (Figure 5A). The native GH1, FECH and EYA1 carry a stretch of 6–10 pyrimidines, whereas the native LPL and HEXA harbor a stretch of 14 and 13 pyrimidines, respectively (arrows in Figure 5A). For highly degenerate PPTs in the artificial constructs, the total number of pyrimidines in a stretch of 25 nt at the 3′-end of an intron well predicts the ratio of exon skipping (Figure 5B). These analyses revealed that the length of the polypyrimidine stretch should be at least 10–15 nt to ensure normal splicing even in the presence of a mutation at E+1.Figure 5.


AG-dependent 3'-splice sites are predisposed to aberrant splicing due to a mutation at the first nucleotide of an exon.

Fu Y, Masuda A, Ito M, Shinmi J, Ohno K - Nucleic Acids Res. (2011)

RT–PCR analysis of nine disease-causing mutations at E+1. (A) Sequences at the intron/exon junctions of nine pairs of wild-type and mutant constructs. The longest polypyrimidine stretches are underlined. (B) RT–PCR of minigenes transfected into HEK293 cells. Five mutant constructs are aberrantly spliced, whereas the remaining four mutants are normally spliced. Numbers in the parentheses indicate exon numbers. In PKHD1, a cryptic 3′-splice site (open arrowhead in panel A) at 55 nt upstream of the native site is activated (asterisk). Mean and SD of three independent experiments of the densitometric ratios of the exon-skipped product is shown at the bottom.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3105431&req=5

Figure 6: RT–PCR analysis of nine disease-causing mutations at E+1. (A) Sequences at the intron/exon junctions of nine pairs of wild-type and mutant constructs. The longest polypyrimidine stretches are underlined. (B) RT–PCR of minigenes transfected into HEK293 cells. Five mutant constructs are aberrantly spliced, whereas the remaining four mutants are normally spliced. Numbers in the parentheses indicate exon numbers. In PKHD1, a cryptic 3′-splice site (open arrowhead in panel A) at 55 nt upstream of the native site is activated (asterisk). Mean and SD of three independent experiments of the densitometric ratios of the exon-skipped product is shown at the bottom.
Mentions: We next sought for parameters that differentiate normal and aberrant splicings in these minigenes. Analysis of parameters that potentially dictate the strength of the PPT indicated that the length of pyrimidine stretch, the number of pyrimidines in 25 or 50 nt at the 3′-end of an intron correlated with the ratio of exon skipping with correlation coefficients of more than 0.6 (Supplementary Table S1). The number of pyrimidines in 25 or 50 nt at the 3′-end of an intron, however, failed to predict splicing consequences of nine other constructs shown in Figure 6, and is likely to be overfitted parameters unique to the 35 constructs in Figure 4. Coolidge and colleagues report that (GU)11 in PPT is partly functional, but we did not observe alternative purine and pyrimidine residues in our PPTs and did not quantify effects of alternative nucleotides (10). We thus took the length of pyrimidine stretch as a best parameter to dictate the strength of the PPT (Figure 5A). The native GH1, FECH and EYA1 carry a stretch of 6–10 pyrimidines, whereas the native LPL and HEXA harbor a stretch of 14 and 13 pyrimidines, respectively (arrows in Figure 5A). For highly degenerate PPTs in the artificial constructs, the total number of pyrimidines in a stretch of 25 nt at the 3′-end of an intron well predicts the ratio of exon skipping (Figure 5B). These analyses revealed that the length of the polypyrimidine stretch should be at least 10–15 nt to ensure normal splicing even in the presence of a mutation at E+1.Figure 5.

Bottom Line: RNA-EMSA revealed that wild-type FECH requires U2AF(35) but wild-type LPL does not.Our studies suggest that a mutation at the AG-dependent 3'-splice site that requires U2AF(35) for spliceosome assembly causes exon skipping, whereas one at the AG-independent 3'-splice site that does not require U2AF(35) gives rise to normal splicing.The AG-dependence of the 3'-splice site that we analyzed in disease-causing mutations at E(+1) potentially helps identify yet unrecognized splicing mutations at E(+1).

View Article: PubMed Central - PubMed

Affiliation: Division of Neurogenetics, Center for Neurological Diseases and Cancer, Nagoya University Graduate School of Medicine, 65 Tsurumai, Showa-ku, Nagoya 466-8550, Japan.

ABSTRACT
In pre-mRNA splicing, a conserved AG/G at the 3'-splice site is recognized by U2AF(35). A disease-causing mutation abrogating the G nucleotide at the first position of an exon (E(+1)) causes exon skipping in GH1, FECH and EYA1, but not in LPL or HEXA. Knockdown of U2AF(35) enhanced exon skipping in GH1 and FECH. RNA-EMSA revealed that wild-type FECH requires U2AF(35) but wild-type LPL does not. A series of artificial mutations in the polypyrimidine tracts of GH1, FECH, EYA1, LPL and HEXA disclosed that a stretch of at least 10-15 pyrimidines is required to ensure normal splicing in the presence of a mutation at E(+1). Analysis of nine other disease-causing mutations at E(+1) detected five splicing mutations. Our studies suggest that a mutation at the AG-dependent 3'-splice site that requires U2AF(35) for spliceosome assembly causes exon skipping, whereas one at the AG-independent 3'-splice site that does not require U2AF(35) gives rise to normal splicing. The AG-dependence of the 3'-splice site that we analyzed in disease-causing mutations at E(+1) potentially helps identify yet unrecognized splicing mutations at E(+1).

Show MeSH
Related in: MedlinePlus