Limits...
Integrative genome-wide analysis reveals HLP1, a novel RNA-binding protein, regulates plant flowering by targeting alternative polyadenylation.

Zhang Y, Gu L, Hou Y, Wang L, Deng X, Hang R, Chen D, Zhang X, Zhang Y, Liu C, Cao X - Cell Res. (2015)

Bottom Line: We show HLP1 is significantly enriched at transcripts involved in RNA metabolism and flowering.A distal-to-proximal poly(A) site shift in the flowering regulator FCA, a direct target of HLP1, leads to upregulation of FLC and delayed flowering.Our results elucidate that HLP1 is a novel factor involved in 3'-end processing and controls reproductive timing via targeting APA.

View Article: PubMed Central - PubMed

Affiliation: State Key Laboratory of Plant Genomics and National Center for Plant Gene Research, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China.

ABSTRACT
Alternative polyadenylation (APA) is a widespread mechanism for gene regulation and has been implicated in flowering, but the molecular basis governing the choice of a specific poly(A) site during the vegetative-to-reproductive growth transition remains unclear. Here we characterize HLP1, an hnRNP A/B protein as a novel regulator for pre-mRNA 3'-end processing in Arabidopsis. Genetic analysis reveals that HLP1 suppresses Flowering Locus C (FLC), a key repressor of flowering in Arabidopsis. Genome-wide mapping of HLP1-RNA interactions indicates that HLP1 binds preferentially to A-rich and U-rich elements around cleavage and polyadenylation sites, implicating its role in 3'-end formation. We show HLP1 is significantly enriched at transcripts involved in RNA metabolism and flowering. Comprehensive profiling of the poly(A) site usage reveals that HLP1 mutations cause thousands of poly(A) site shifts. A distal-to-proximal poly(A) site shift in the flowering regulator FCA, a direct target of HLP1, leads to upregulation of FLC and delayed flowering. Our results elucidate that HLP1 is a novel factor involved in 3'-end processing and controls reproductive timing via targeting APA.

No MeSH data available.


HLP1 regulates global APA. (A) Distribution of poly(A) clusters. (B) Distribution of A-rich and U-rich PAC motifs relative to the poly(A) site (PAS) are indicated by blue and red curves, respectively. (C) Scatter plots show poly(A) site shifts in hlp1-1 mutant. The x axis and y axis show the ratio of PAC counts on a log2 scale between hlp1-1 and Col at proximal and distal poly(A) sites, respectively. Genes with significant proximal-to-distal poly(A) site shifts (P-to-D) are indicated by blue dots and red dots; genes with significant distal-to-proximal shifts (D-to-P) are colored in green and purple. Red and purple dots represent APA shifts with HLP1 binding. Bar graphs indicate the number of these dots. Grey dots represent genes without significant changes in APA. The poly(A) site shifts were evaluated using Fisher's exact test (P < 0.02). (D) Case studies of transcripts with APA shift. Wiggle plots on the left panels show P-to-D shift at At3g15450, At3g23030, At5g11070 and D-to-P shift at At1g72645. CDS regions are boxed in black. The 5′-UTR and 3′-UTR are boxed in green and grey, respectively. Introns are indicated as lines. The x axis indicates genome site in chromosome. The y axis indicates normalized HITS-CLIP/CLIP-seq or PAS-seq abundance. HITS-CLIP/CLIP-seq or PAS-seq tag counts were normalized to tag per 10 million (TP10M) to adjust for differences of two libraries (wild-type and mutant) in sequencing depth. Right panels show RT-qPCR assessments of the shift.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4493284&req=5

fig4: HLP1 regulates global APA. (A) Distribution of poly(A) clusters. (B) Distribution of A-rich and U-rich PAC motifs relative to the poly(A) site (PAS) are indicated by blue and red curves, respectively. (C) Scatter plots show poly(A) site shifts in hlp1-1 mutant. The x axis and y axis show the ratio of PAC counts on a log2 scale between hlp1-1 and Col at proximal and distal poly(A) sites, respectively. Genes with significant proximal-to-distal poly(A) site shifts (P-to-D) are indicated by blue dots and red dots; genes with significant distal-to-proximal shifts (D-to-P) are colored in green and purple. Red and purple dots represent APA shifts with HLP1 binding. Bar graphs indicate the number of these dots. Grey dots represent genes without significant changes in APA. The poly(A) site shifts were evaluated using Fisher's exact test (P < 0.02). (D) Case studies of transcripts with APA shift. Wiggle plots on the left panels show P-to-D shift at At3g15450, At3g23030, At5g11070 and D-to-P shift at At1g72645. CDS regions are boxed in black. The 5′-UTR and 3′-UTR are boxed in green and grey, respectively. Introns are indicated as lines. The x axis indicates genome site in chromosome. The y axis indicates normalized HITS-CLIP/CLIP-seq or PAS-seq abundance. HITS-CLIP/CLIP-seq or PAS-seq tag counts were normalized to tag per 10 million (TP10M) to adjust for differences of two libraries (wild-type and mutant) in sequencing depth. Right panels show RT-qPCR assessments of the shift.

Mentions: Although HLP1 shares low similarity to the yeast Hrp1, which has been shown to bind the AU-rich efficiency element (EE) and has been implicated in correct positioning of the cleavage site and regulating 3′-end formation of pre-mRNAs33,34, it binds to the U-rich far upstream element (FUE) and A-rich near upstream element (NUE), the plant cis-element equivalent to the yeast EE and A-rich positioning element (PE), respectively. Therefore, HLP1 may function as the Hrp1 ortholog determining the ploy(A) site in plants. To address this, poly(A) site sequencing (PAS-Seq) was used to quantitatively profile poly(A) site usage35,36 in hlp1-1 mutant and the wild-type plant Col (Supplementary information, Figure S8 and Table S3). As expected, the poly(A) clusters (PACs) are predominantly located in the 3′-UTR (∼70%), but also were found in CDS (∼23%), 5′-UTR (∼5%) and introns (∼1%), suggesting APA in CDS, and to a lesser extent in 5′-UTR, as potential regulatory mechanisms (Figure 4A). However, we cannot exclude the possibility that the high percentage of PACs in CDS could be artifacts caused by internal priming11,37. The overrepresented 5′-AAAGAAAA-3′ and 5′-UGUUUC-3′ motifs surrounding the poly(A) site are very similar to the HLP1-binding motifs (Figure 4B). Notably, out of 2 691 HLP1 binding sites at the 3′-UTR, 78% (2 088/2 691) overlap with 1 777 PACs at the 3′-UTR (∼13% of PACs at this region) in wild-type plant (standard score Z = 82), further supporting the role of HLP1 in 3′-end formation (Supplementary information, Figure S9A). By analyzing overlapping APA profiles from two APA biological replicates, we found that HLP1 mutation caused proximal-to-distal poly(A) site shifts in 2 274 transcripts compared with Col (P < 0.02, Fisher's exact test), suggesting that HLP1 is a 3′-end factor predominantly suppressing the usage of distal poly(A) sites (Figure 4C). The single-molecule direct RNA sequencing (DRS) is a newly developed method and is believed to have less or no artifacts for PAS-seq analysis37. Comparison of transcripts with poly (A) site shifts in our data with the DRS data show that 94% switched PACs were also detected by DRS, suggesting that these switched PACs are reliable (Supplementary information, Figure S9B). Both proximal-to-distal and distal-to-proximal polyadenylation shifts in genes were validated by q-PCR or shown as wiggle plots (Figure 4D and Supplementary information, Figure S10).


Integrative genome-wide analysis reveals HLP1, a novel RNA-binding protein, regulates plant flowering by targeting alternative polyadenylation.

Zhang Y, Gu L, Hou Y, Wang L, Deng X, Hang R, Chen D, Zhang X, Zhang Y, Liu C, Cao X - Cell Res. (2015)

HLP1 regulates global APA. (A) Distribution of poly(A) clusters. (B) Distribution of A-rich and U-rich PAC motifs relative to the poly(A) site (PAS) are indicated by blue and red curves, respectively. (C) Scatter plots show poly(A) site shifts in hlp1-1 mutant. The x axis and y axis show the ratio of PAC counts on a log2 scale between hlp1-1 and Col at proximal and distal poly(A) sites, respectively. Genes with significant proximal-to-distal poly(A) site shifts (P-to-D) are indicated by blue dots and red dots; genes with significant distal-to-proximal shifts (D-to-P) are colored in green and purple. Red and purple dots represent APA shifts with HLP1 binding. Bar graphs indicate the number of these dots. Grey dots represent genes without significant changes in APA. The poly(A) site shifts were evaluated using Fisher's exact test (P < 0.02). (D) Case studies of transcripts with APA shift. Wiggle plots on the left panels show P-to-D shift at At3g15450, At3g23030, At5g11070 and D-to-P shift at At1g72645. CDS regions are boxed in black. The 5′-UTR and 3′-UTR are boxed in green and grey, respectively. Introns are indicated as lines. The x axis indicates genome site in chromosome. The y axis indicates normalized HITS-CLIP/CLIP-seq or PAS-seq abundance. HITS-CLIP/CLIP-seq or PAS-seq tag counts were normalized to tag per 10 million (TP10M) to adjust for differences of two libraries (wild-type and mutant) in sequencing depth. Right panels show RT-qPCR assessments of the shift.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4493284&req=5

fig4: HLP1 regulates global APA. (A) Distribution of poly(A) clusters. (B) Distribution of A-rich and U-rich PAC motifs relative to the poly(A) site (PAS) are indicated by blue and red curves, respectively. (C) Scatter plots show poly(A) site shifts in hlp1-1 mutant. The x axis and y axis show the ratio of PAC counts on a log2 scale between hlp1-1 and Col at proximal and distal poly(A) sites, respectively. Genes with significant proximal-to-distal poly(A) site shifts (P-to-D) are indicated by blue dots and red dots; genes with significant distal-to-proximal shifts (D-to-P) are colored in green and purple. Red and purple dots represent APA shifts with HLP1 binding. Bar graphs indicate the number of these dots. Grey dots represent genes without significant changes in APA. The poly(A) site shifts were evaluated using Fisher's exact test (P < 0.02). (D) Case studies of transcripts with APA shift. Wiggle plots on the left panels show P-to-D shift at At3g15450, At3g23030, At5g11070 and D-to-P shift at At1g72645. CDS regions are boxed in black. The 5′-UTR and 3′-UTR are boxed in green and grey, respectively. Introns are indicated as lines. The x axis indicates genome site in chromosome. The y axis indicates normalized HITS-CLIP/CLIP-seq or PAS-seq abundance. HITS-CLIP/CLIP-seq or PAS-seq tag counts were normalized to tag per 10 million (TP10M) to adjust for differences of two libraries (wild-type and mutant) in sequencing depth. Right panels show RT-qPCR assessments of the shift.
Mentions: Although HLP1 shares low similarity to the yeast Hrp1, which has been shown to bind the AU-rich efficiency element (EE) and has been implicated in correct positioning of the cleavage site and regulating 3′-end formation of pre-mRNAs33,34, it binds to the U-rich far upstream element (FUE) and A-rich near upstream element (NUE), the plant cis-element equivalent to the yeast EE and A-rich positioning element (PE), respectively. Therefore, HLP1 may function as the Hrp1 ortholog determining the ploy(A) site in plants. To address this, poly(A) site sequencing (PAS-Seq) was used to quantitatively profile poly(A) site usage35,36 in hlp1-1 mutant and the wild-type plant Col (Supplementary information, Figure S8 and Table S3). As expected, the poly(A) clusters (PACs) are predominantly located in the 3′-UTR (∼70%), but also were found in CDS (∼23%), 5′-UTR (∼5%) and introns (∼1%), suggesting APA in CDS, and to a lesser extent in 5′-UTR, as potential regulatory mechanisms (Figure 4A). However, we cannot exclude the possibility that the high percentage of PACs in CDS could be artifacts caused by internal priming11,37. The overrepresented 5′-AAAGAAAA-3′ and 5′-UGUUUC-3′ motifs surrounding the poly(A) site are very similar to the HLP1-binding motifs (Figure 4B). Notably, out of 2 691 HLP1 binding sites at the 3′-UTR, 78% (2 088/2 691) overlap with 1 777 PACs at the 3′-UTR (∼13% of PACs at this region) in wild-type plant (standard score Z = 82), further supporting the role of HLP1 in 3′-end formation (Supplementary information, Figure S9A). By analyzing overlapping APA profiles from two APA biological replicates, we found that HLP1 mutation caused proximal-to-distal poly(A) site shifts in 2 274 transcripts compared with Col (P < 0.02, Fisher's exact test), suggesting that HLP1 is a 3′-end factor predominantly suppressing the usage of distal poly(A) sites (Figure 4C). The single-molecule direct RNA sequencing (DRS) is a newly developed method and is believed to have less or no artifacts for PAS-seq analysis37. Comparison of transcripts with poly (A) site shifts in our data with the DRS data show that 94% switched PACs were also detected by DRS, suggesting that these switched PACs are reliable (Supplementary information, Figure S9B). Both proximal-to-distal and distal-to-proximal polyadenylation shifts in genes were validated by q-PCR or shown as wiggle plots (Figure 4D and Supplementary information, Figure S10).

Bottom Line: We show HLP1 is significantly enriched at transcripts involved in RNA metabolism and flowering.A distal-to-proximal poly(A) site shift in the flowering regulator FCA, a direct target of HLP1, leads to upregulation of FLC and delayed flowering.Our results elucidate that HLP1 is a novel factor involved in 3'-end processing and controls reproductive timing via targeting APA.

View Article: PubMed Central - PubMed

Affiliation: State Key Laboratory of Plant Genomics and National Center for Plant Gene Research, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China.

ABSTRACT
Alternative polyadenylation (APA) is a widespread mechanism for gene regulation and has been implicated in flowering, but the molecular basis governing the choice of a specific poly(A) site during the vegetative-to-reproductive growth transition remains unclear. Here we characterize HLP1, an hnRNP A/B protein as a novel regulator for pre-mRNA 3'-end processing in Arabidopsis. Genetic analysis reveals that HLP1 suppresses Flowering Locus C (FLC), a key repressor of flowering in Arabidopsis. Genome-wide mapping of HLP1-RNA interactions indicates that HLP1 binds preferentially to A-rich and U-rich elements around cleavage and polyadenylation sites, implicating its role in 3'-end formation. We show HLP1 is significantly enriched at transcripts involved in RNA metabolism and flowering. Comprehensive profiling of the poly(A) site usage reveals that HLP1 mutations cause thousands of poly(A) site shifts. A distal-to-proximal poly(A) site shift in the flowering regulator FCA, a direct target of HLP1, leads to upregulation of FLC and delayed flowering. Our results elucidate that HLP1 is a novel factor involved in 3'-end processing and controls reproductive timing via targeting APA.

No MeSH data available.