Limits...
The consensus 5' splice site motif inhibits mRNA nuclear export.

Lee ES, Akef A, Mahadevan K, Palazzo AF - PLoS ONE (2015)

Bottom Line: The motif, however, does not disrupt splicing or the recruitment of UAP56 or TAP/Nxf1 to the RNA, which are normally required for nuclear export.This motif is also depleted from the beginning and ends of the 3'terminal exons of spliced mRNAs, but less so for lncRNAs.Our data suggests that the presence of the 5'splice site motif in mature RNAs promotes their nuclear retention and may help to distinguish mRNAs from misprocessed transcripts and transcriptional noise.

View Article: PubMed Central - PubMed

Affiliation: Department of Biochemistry, University of Toronto, 1 King's College Circle, MSB Room 5336, Toronto, ON, M5S 1A8, Canada.

ABSTRACT
In eukaryotes, mRNAs are synthesized in the nucleus and then exported to the cytoplasm where they are translated into proteins. We have mapped an element, which when present in the 3'terminal exon or in an unspliced mRNA, inhibits mRNA nuclear export. This element has the same sequence as the consensus 5'splice site motif that is used to define the start of introns. Previously it was shown that when this motif is retained in the mRNA, it causes defects in 3'cleavage and polyadenylation and promotes mRNA decay. Our new data indicates that this motif also inhibits nuclear export and promotes the targeting of transcripts to nuclear speckles, foci within the nucleus which have been linked to splicing. The motif, however, does not disrupt splicing or the recruitment of UAP56 or TAP/Nxf1 to the RNA, which are normally required for nuclear export. Genome wide analysis of human mRNAs, lncRNA and eRNAs indicates that this motif is depleted from naturally intronless mRNAs and eRNAs, but less so in lncRNAs. This motif is also depleted from the beginning and ends of the 3'terminal exons of spliced mRNAs, but less so for lncRNAs. Our data suggests that the presence of the 5'splice site motif in mature RNAs promotes their nuclear retention and may help to distinguish mRNAs from misprocessed transcripts and transcriptional noise.

No MeSH data available.


Related in: MedlinePlus

The 5’SS motif is depleted in the 3’exons and 3’UTRs of genes across the human genome.Various regions across the human genome were analyzed for the presence of the consensus 5’SS motif. As controls, we analyzed regions 1kb regions that were 3kb upstream (“US”) from the transcriptional start sites of and downstream (“DS”) from the 3’cleavage site of protein-coding genes as seen in the schematic in (A). Note that introns are known to be rarely present in the 3’UTRs and as a result the 3’UTR is normally present in its entirety in the 3’terminal exon (B). (C) The frequency of the consensus 5’SS motif was analyzed in various loci (see Table 1 for a full description). The expected frequencies were determined using the tri-nucleotide frequencies in each region. Regions analyzed included 3’UTRs from all protein coding genes, 3’UTRs that do not contain introns (“3’UTR (3UI-)”), the reverse compliment of 3’UTRs (“3’UTR RC”), the 3’terminal exon of mRNAs, intronless mRNAs, the 3’terminal exon of lncRNAs, intronless lncRNAs, intronless eRNAs, and upstream and downstream intergenic regions as defined in (A). List of lncRNAs was taken from LNCipedia. List of eRNAs was taken from the FANTOM5 consortia. (D-E) Distribution of the consensus 5’SS motif along the 3’terminal exon for mRNAs (D) and lncRNAs (E). The lengths of each RNA were divided into 20 bins and the number of 5’SS motifs in each bin was plotted. (F-H) The number of 5’SS motifs was plotted as a function of distance from the last exon-exon junction (F), stop codon (G), or 3’cleavage site (H) of mRNAs. The number of 5’SS motifs in each bin of 50 nucleotides was normalized for the total number of mRNAs remaining as distance increases.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4380460&req=5

pone.0122743.g006: The 5’SS motif is depleted in the 3’exons and 3’UTRs of genes across the human genome.Various regions across the human genome were analyzed for the presence of the consensus 5’SS motif. As controls, we analyzed regions 1kb regions that were 3kb upstream (“US”) from the transcriptional start sites of and downstream (“DS”) from the 3’cleavage site of protein-coding genes as seen in the schematic in (A). Note that introns are known to be rarely present in the 3’UTRs and as a result the 3’UTR is normally present in its entirety in the 3’terminal exon (B). (C) The frequency of the consensus 5’SS motif was analyzed in various loci (see Table 1 for a full description). The expected frequencies were determined using the tri-nucleotide frequencies in each region. Regions analyzed included 3’UTRs from all protein coding genes, 3’UTRs that do not contain introns (“3’UTR (3UI-)”), the reverse compliment of 3’UTRs (“3’UTR RC”), the 3’terminal exon of mRNAs, intronless mRNAs, the 3’terminal exon of lncRNAs, intronless lncRNAs, intronless eRNAs, and upstream and downstream intergenic regions as defined in (A). List of lncRNAs was taken from LNCipedia. List of eRNAs was taken from the FANTOM5 consortia. (D-E) Distribution of the consensus 5’SS motif along the 3’terminal exon for mRNAs (D) and lncRNAs (E). The lengths of each RNA were divided into 20 bins and the number of 5’SS motifs in each bin was plotted. (F-H) The number of 5’SS motifs was plotted as a function of distance from the last exon-exon junction (F), stop codon (G), or 3’cleavage site (H) of mRNAs. The number of 5’SS motifs in each bin of 50 nucleotides was normalized for the total number of mRNAs remaining as distance increases.

Mentions: We rationalized that if the 5’SS motif, inhibits proper 3’end processing, promotes decay and causes nuclear retention, that it should be depleted in most naturally intronless protein-coding genes from the human genome. Moreover, these motifs should also be eliminated from the 3’exons within human mRNAs. We also reasoned that since 3’UTRs are usually present in their entirety in the 3’exon (see schematic, Fig 6A), they should also be depleted of the 5’SS motifs. Indeed we found that the consensus motif ([C/A]AGGU[C/A]AG) was present in both 3’exons and 3’UTRs at about half the rate at which it appeared in either intergenic regions (upstream, “US”, or downstream, “DS” of protein coding mRNAs, see schematic Fig 6B) or in the reverse compliment of 3’UTRs (“3’UTR RC”; Fig 6C, Table 1). About 5% of all protein-coding genes have introns in their 3’UTRs, however even if these are eliminated (see Fig 6A, “3’UTR 3UI-”; list of human genes with 3’UTR introns was obtained from the Blencowe lab), the remaining 3’UTRs are still depleted of the consensus 5’SS motif (Fig 6C, Table 1). When naturally intronless genes were analyzed (list obtained from the intronless gene database [42], http://www.bioinfo-cbs.org/igd/) these were even more depleted of 5’SS motifs (Fig 6C, Table 1). Since 3’UTRs are AU-rich, they may contain fewer 5’SS motifs due to their nucleotide composition. To control for this effect we compared the frequency that the motif is present to the expected rate that they would appear given the trinucleotide frequencies of these genomic regions. Again 3’exons and 3’UTRs are depleted of the 5’SS consensus motif, but to our surprise the rate of depletion, in comparison to the expected frequency (~30%), was not as dramatic as we thought it would be. The reduction was much more pronounced for intronless genes (>50% depletion).


The consensus 5' splice site motif inhibits mRNA nuclear export.

Lee ES, Akef A, Mahadevan K, Palazzo AF - PLoS ONE (2015)

The 5’SS motif is depleted in the 3’exons and 3’UTRs of genes across the human genome.Various regions across the human genome were analyzed for the presence of the consensus 5’SS motif. As controls, we analyzed regions 1kb regions that were 3kb upstream (“US”) from the transcriptional start sites of and downstream (“DS”) from the 3’cleavage site of protein-coding genes as seen in the schematic in (A). Note that introns are known to be rarely present in the 3’UTRs and as a result the 3’UTR is normally present in its entirety in the 3’terminal exon (B). (C) The frequency of the consensus 5’SS motif was analyzed in various loci (see Table 1 for a full description). The expected frequencies were determined using the tri-nucleotide frequencies in each region. Regions analyzed included 3’UTRs from all protein coding genes, 3’UTRs that do not contain introns (“3’UTR (3UI-)”), the reverse compliment of 3’UTRs (“3’UTR RC”), the 3’terminal exon of mRNAs, intronless mRNAs, the 3’terminal exon of lncRNAs, intronless lncRNAs, intronless eRNAs, and upstream and downstream intergenic regions as defined in (A). List of lncRNAs was taken from LNCipedia. List of eRNAs was taken from the FANTOM5 consortia. (D-E) Distribution of the consensus 5’SS motif along the 3’terminal exon for mRNAs (D) and lncRNAs (E). The lengths of each RNA were divided into 20 bins and the number of 5’SS motifs in each bin was plotted. (F-H) The number of 5’SS motifs was plotted as a function of distance from the last exon-exon junction (F), stop codon (G), or 3’cleavage site (H) of mRNAs. The number of 5’SS motifs in each bin of 50 nucleotides was normalized for the total number of mRNAs remaining as distance increases.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4380460&req=5

pone.0122743.g006: The 5’SS motif is depleted in the 3’exons and 3’UTRs of genes across the human genome.Various regions across the human genome were analyzed for the presence of the consensus 5’SS motif. As controls, we analyzed regions 1kb regions that were 3kb upstream (“US”) from the transcriptional start sites of and downstream (“DS”) from the 3’cleavage site of protein-coding genes as seen in the schematic in (A). Note that introns are known to be rarely present in the 3’UTRs and as a result the 3’UTR is normally present in its entirety in the 3’terminal exon (B). (C) The frequency of the consensus 5’SS motif was analyzed in various loci (see Table 1 for a full description). The expected frequencies were determined using the tri-nucleotide frequencies in each region. Regions analyzed included 3’UTRs from all protein coding genes, 3’UTRs that do not contain introns (“3’UTR (3UI-)”), the reverse compliment of 3’UTRs (“3’UTR RC”), the 3’terminal exon of mRNAs, intronless mRNAs, the 3’terminal exon of lncRNAs, intronless lncRNAs, intronless eRNAs, and upstream and downstream intergenic regions as defined in (A). List of lncRNAs was taken from LNCipedia. List of eRNAs was taken from the FANTOM5 consortia. (D-E) Distribution of the consensus 5’SS motif along the 3’terminal exon for mRNAs (D) and lncRNAs (E). The lengths of each RNA were divided into 20 bins and the number of 5’SS motifs in each bin was plotted. (F-H) The number of 5’SS motifs was plotted as a function of distance from the last exon-exon junction (F), stop codon (G), or 3’cleavage site (H) of mRNAs. The number of 5’SS motifs in each bin of 50 nucleotides was normalized for the total number of mRNAs remaining as distance increases.
Mentions: We rationalized that if the 5’SS motif, inhibits proper 3’end processing, promotes decay and causes nuclear retention, that it should be depleted in most naturally intronless protein-coding genes from the human genome. Moreover, these motifs should also be eliminated from the 3’exons within human mRNAs. We also reasoned that since 3’UTRs are usually present in their entirety in the 3’exon (see schematic, Fig 6A), they should also be depleted of the 5’SS motifs. Indeed we found that the consensus motif ([C/A]AGGU[C/A]AG) was present in both 3’exons and 3’UTRs at about half the rate at which it appeared in either intergenic regions (upstream, “US”, or downstream, “DS” of protein coding mRNAs, see schematic Fig 6B) or in the reverse compliment of 3’UTRs (“3’UTR RC”; Fig 6C, Table 1). About 5% of all protein-coding genes have introns in their 3’UTRs, however even if these are eliminated (see Fig 6A, “3’UTR 3UI-”; list of human genes with 3’UTR introns was obtained from the Blencowe lab), the remaining 3’UTRs are still depleted of the consensus 5’SS motif (Fig 6C, Table 1). When naturally intronless genes were analyzed (list obtained from the intronless gene database [42], http://www.bioinfo-cbs.org/igd/) these were even more depleted of 5’SS motifs (Fig 6C, Table 1). Since 3’UTRs are AU-rich, they may contain fewer 5’SS motifs due to their nucleotide composition. To control for this effect we compared the frequency that the motif is present to the expected rate that they would appear given the trinucleotide frequencies of these genomic regions. Again 3’exons and 3’UTRs are depleted of the 5’SS consensus motif, but to our surprise the rate of depletion, in comparison to the expected frequency (~30%), was not as dramatic as we thought it would be. The reduction was much more pronounced for intronless genes (>50% depletion).

Bottom Line: The motif, however, does not disrupt splicing or the recruitment of UAP56 or TAP/Nxf1 to the RNA, which are normally required for nuclear export.This motif is also depleted from the beginning and ends of the 3'terminal exons of spliced mRNAs, but less so for lncRNAs.Our data suggests that the presence of the 5'splice site motif in mature RNAs promotes their nuclear retention and may help to distinguish mRNAs from misprocessed transcripts and transcriptional noise.

View Article: PubMed Central - PubMed

Affiliation: Department of Biochemistry, University of Toronto, 1 King's College Circle, MSB Room 5336, Toronto, ON, M5S 1A8, Canada.

ABSTRACT
In eukaryotes, mRNAs are synthesized in the nucleus and then exported to the cytoplasm where they are translated into proteins. We have mapped an element, which when present in the 3'terminal exon or in an unspliced mRNA, inhibits mRNA nuclear export. This element has the same sequence as the consensus 5'splice site motif that is used to define the start of introns. Previously it was shown that when this motif is retained in the mRNA, it causes defects in 3'cleavage and polyadenylation and promotes mRNA decay. Our new data indicates that this motif also inhibits nuclear export and promotes the targeting of transcripts to nuclear speckles, foci within the nucleus which have been linked to splicing. The motif, however, does not disrupt splicing or the recruitment of UAP56 or TAP/Nxf1 to the RNA, which are normally required for nuclear export. Genome wide analysis of human mRNAs, lncRNA and eRNAs indicates that this motif is depleted from naturally intronless mRNAs and eRNAs, but less so in lncRNAs. This motif is also depleted from the beginning and ends of the 3'terminal exons of spliced mRNAs, but less so for lncRNAs. Our data suggests that the presence of the 5'splice site motif in mature RNAs promotes their nuclear retention and may help to distinguish mRNAs from misprocessed transcripts and transcriptional noise.

No MeSH data available.


Related in: MedlinePlus