(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Alternative TSS use is widespread in Cryptococcus fungi in response to environmental cues and regulated genome-wide by the transcription factor Tur1 [1]

['Thi Tuong Vi Dang', 'Université Paris Cité', 'Institut Pasteur', 'Unité Biologie Des Arn Des Pathogènes Fongiques', 'Département De Mycologie', 'Paris', 'Corinne Maufrais', 'Hub Bioinformatique Et Biostatistique', 'Usr Ip Cnrs', 'Jessie Colin']

Date: 2024-07

Alternative transcription start site (TSS) usage regulation has been identified as a major means of gene expression regulation in metazoans. However, in fungi, its impact remains elusive as its study has thus far been restricted to model yeasts. Here, we first re-analyzed TSS-seq data to define genuine TSS clusters in 2 species of pathogenic Cryptococcus. We identified 2 types of TSS clusters associated with specific DNA sequence motifs. Our analysis also revealed that alternative TSS usage regulation in response to environmental cues is widespread in Cryptococcus, altering gene expression and protein targeting. Importantly, we performed a forward genetic screen to identify a unique transcription factor (TF) named Tur1, which regulates alternative TSS (altTSS) usage genome-wide when cells switch from exponential phase to stationary phase. ChiP-Seq and DamID-Seq analyses suggest that at some loci, the role of Tur1 might be direct. Tur1 has been previously shown to be essential for virulence in C. neoformans. We demonstrated here that a tur1Δ mutant strain is more sensitive to superoxide stress and phagocytosed more efficiently by macrophages than the wild-type (WT) strain.

In the present study, we re-analyzed these Cryptococcus TSS-seq data to first characterize the structure of a genuine TSS cluster in these yeasts. We then described alternative TSS usage and demonstrated that it represents a major means to regulate transcriptome and proteome structure in these fungi. More importantly, we screened a TF mutant collection to identify genes regulating altTSS usage in C. neoformans. We showed that the transcription factor Tur1 is necessary for genome-wide altTSS usage regulation during the exponential to stationary phase transition. We also performed ChiP-Seq and DamID-seq analyses to study the binding of Tur1 at the regulated loci. Finally, we showed that Tur1 regulates superoxide stress resistance and interaction with macrophage linking altTSS usage and virulence in this major fungal pathogen.

Cryptococcus neoformans is a pathogenic basidiomycete yeast that is responsible for 180,000 deaths every year worldwide [ 23 ]. In addition to its ability to grow at 37°C, its main virulence factors are a polysaccharide capsule, the production of melanin and its ability to replicate in macrophages [ 24 , 25 ]. In recent years, we produced several sets of RNA-seq data that were used to produce detailed coding gene annotation of the genomes of several pathogenic Cryptococcus species [ 26 – 28 ]. Our analysis also revealed that nearly all the genes contain several short introns, which are essential for gene expression [ 29 , 30 ]. As expected, alternative splicing is prominent in Cryptococcus, although its impact on proteome structure is limited primarily to regulating gene expression [ 27 ]. More recently, we produced TSS-seq and 3UTR-seq data from C. neoformans and its sibling species Cryptococcus deneoformans grown under 4 conditions (i.e., exponential phase and stationary phase at 30 and 37°C) [ 14 ]. We used this dataset to re-annotate the transcript leader (TL) and 3′ UTR sequences in these species. Our analysis revealed that, in contrast to S. cerevisiae, Cryptococcus TL sequences frequently contain upstream open reading frames (uORFs). These yeasts use the strength of a Kozac-like consensus to determine translation start codon usage, thus regulating both gene expression and protein localization [ 14 ]. This analysis also revealed thousands of additional TSS clusters associated with coding genes, suggesting that alternative TSS (altTSS) usage might be widespread in Cryptococcus [ 14 ]. This hypothesis was supported by several recent studies reporting gene-specific examples of altTSS usage regulation in these yeasts. For instance, Pum1, an RNA-binding protein, is known to positively regulate the expression of ZNF2, a master regulator of filamentation and virulence in C. neoformans [ 31 ]. Under filamentation-inducing conditions, the TF Znf2 favors PUM1 transcription from a downstream TSS, which excludes from the transcript the Pum1-binding site that is normally found in the TL. This shorter form being immune to the negative autoregulation, this leads to the accumulation of the Pum1 protein which in turn activates ZNF2 expression [ 31 ]. Moreover, when exposed to UV light, C. neoformans switches off TSS at the UVE1 gene and uses an upstream altTSS. This promotes the transcription of a longer mRNA coding for the mitochondrial isoform of the DNA damage repair endonuclease Uve1, thus protecting the mitochondrial genome from potentially lethal UV-induced DNA damage [ 32 ]. Finally, a recent study reports that, under copper-limited conditions, C. neoformans cells promote the usage of a downstream altTSS at both SOD1 and SOD2 superoxide dismutase genes, thus regulating transcript stability and protein subcellular localization of these proteins, respectively [ 33 ].

In fungi, although TSS sequencing data have been produced in several species [ 12 – 14 ], the analysis of TSS structure and usage is limited to the 2 model yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe [ 15 – 18 ]. Nevertheless, these studies also revealed a significant number of alternative TSS (altTSS) usage patterns, with some being specific to growth conditions or meiotic stages [ 12 , 19 ] and others revealed by mutation of genes encoding chromatin modifiers or remodelers [ 20 – 22 ].

In eukaryotes, gene transcription begins at the core promoter by the assembly of a pre-initiation complex comprised of general transcription factors (TFs) that recruit the DNA-dependent RNA polymerase II and melt DNA to create the transcription bubble [ 1 ]. In recent years, different RNA sequencing strategies have been used in metazoans to define 2 types of core promoters characterized by specific chromatin structure, DNA sequence and distribution of transcription start site (TSS) [ 2 ]. These analyses have shown that TSS are by nature heterogeneous and organized in clusters. These TSS clusters can be sharp or broad, sharp ones being associated with TATA-box genes [ 1 , 3 ]. These studies also revealed that transcription can be initiated from multiple TSS clusters in most genes with a very dynamic pattern of usage [ 4 – 8 ] that can be specific to different cell types and development stages [ 6 , 9 – 11 ].

Results

Alternative TSS cluster identification in Cryptococcus To evaluate alternative TSS usage in Cryptococcus, we applied the same subclusterization procedure to all previously identified TSS clusters [14]. We then merged the overlapping TSS clusters obtained from the 4 growth conditions in triplicate. The resulting TSS cluster reference GFF file defines the genomic coordinates of 7,213 TSS clusters associated with 4,931 coding genes in C. neoformans (72.6% of the coding genes) (S4 Table). We defined a TSS cluster as annotated TSS clusters if it satisfies all these 3 criteria: it is the most upstream cluster within a gene, the distance between the 5′ boundary of the cluster and the annotated 5′ extremity of the gene does not exceed 50 bp, and the 3′ border of the cluster is at least 20 bp upstream of the annotated ATG. Every other TSS clusters were considered as alternative TSS clusters. Here, we identified 2,431 genes associated with 2 or more TSS clusters. In total, 3,817 TSS clusters were considered as alternative. We performed the same analysis using the C. deneoformans data and defined 4,064 alternative TSS clusters associated with 2,581 genes (S4 Table).

altTSS within the TL sequence A large proportion of these alternative TSS clusters (37% (n = 1,405) in C. neoformans and 31% (n = 1,256) in C. deneoformans) are positioned within the TL sequence. Their usage regulates the length of the TL sequence (Fig 3) and can result in the inclusion or exclusion of protein binding sites, secondary structures or uORFs, potentially impacting mRNA subcellular localization, stability, or translation efficiency [14,15,40]. We previously reported that Cryptococcus TL sequences are very rich in uORFs and that uORF containing mRNAs are more prone to be degraded by the NMD pathway than uORF-free mRNAs, thus reducing translation efficiency [14]. Here, we identified 542 and 633 genes in C. neoformans and C. deneoformans, respectively, for which the sequence between the annotated TSS cluster and the alternative TSS cluster located in the TL contains 1 or several uORFs. In these genes, alternative TSS usage can regulate the presence or absence of uORFs within the mRNA. In these cases, we reasoned that the production of a short transcript devoid of uORFs would be a way to stabilize it by preventing its degradation by the NMD pathway. Accordingly, among the genes containing 1 uORF within their TL sequence, the ones containing an alternative TSS potentially skipping it are less prone to be up-regulated by the deletion of the major NMD factor UPF1 than those devoid of alternative TSS within their TL sequence (Fig 3A) (Wilcoxon rank sum test, p-value = 1.246 × 10−10). For instance, usage of an alternative TSS within the TL sequence of the CNA08250 locus skips 11 uORFs and results in the production of a short isoform immune to the NMD, whereas the long isoform can only be revealed by the deletion of UPF1 (Fig 3B). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. altTSS usage regulates the presence or absence of uORFs within the mRNA impacting its sensitivity to NMD. Among 3,055 uAUG containing genes in C. deneoformans, genes having alternative TSS within their TL sequence are more resistant to NMD. (A) Cumulative distribution plot of Log fold change between upf1Δ mutant and WT at 30°C are generated for genes which use an altTSS at 30°C (blue line) and genes that do not use an altTSS at 30°C (red line). A one-sided Kolmogorov–Smirnov test shows that these differences are significant (p-value = 1.95 × 10−10). (B) IGV visualization of RNA-seq and TSS-seq at the CNA08250 locus. The gene CNA08250 possesses an alternative TSS that skips the uORFs of the transcript leader sequence. At 30°C, the alternative TSS is prominently used, resulting in the short mRNA isoform devoid of uORF. The data underlying this figure can be found in S1 Data. TL, transcript leader; TSS, transcription start site; WT, wild type; uORF, upstream open reading frame. https://doi.org/10.1371/journal.pbio.3002724.g003

altTSS close to the annotated start codon We identified 220 alternative TSS clusters associated with 213 genes in C. neoformans (208 TSS clusters and 199 genes in C. deneoformans) positioned between −20 and +90 nt from the aATG, when counting bases on the spliced mRNA. The usage of each of these altTSSs results in the transcription of an mRNA which can code a protein truncated from its N-termini, thereby potentially lacking an N-terminal targeting motif. Accordingly, DeepLoc 2.0 [41] analysis predicts that whereas for 169 of these genes the long isoform has a defined subcellular localization, for 87 of them the usage of an altTSS is predicted to result in the production of a shorter isoform with a different subcellular localization (S5 Table and Fig 4). For instance, we identified an altTSS associated with the gene MAE102 (CNAG_06374) that encodes a putative mitochondrial malate dehydrogenase (Fig 5A). Depending on the TSS used, the predicted translated protein contains a mitochondrial targeting signal (MTS) or not. Interestingly, the usage of this altTSS is regulated by the growth condition, the long isoform encoding the putative mitochondrial protein being transcribed in stationary phase at 30°C, whereas the exponential phase condition triggers mostly the transcription of the short isoform predicted to code a protein devoid of MTS. Accordingly, a C-terminal mNeonGreen-tagged version of Mae102 protein revealed a regulated localisation of this protein by the phase of growth. It seems to accumulate into mitochondrial-like particles in stationary phase (Fig 5B), whereas in exponential phase it localizes in the cytosol and the nucleus (Figs 5B and S4). Moreover, mutation of the first ATG into CGT (M1R) or deletion of the sequence between the 2 ATGs (MTSΔ) increase the percentage of cells having a nucleus localization of the protein in exponential phase. In stationary phase, the Mae102 protein mitochondrial-like localization seems to be lost confirming the functionality of the MTS sequence. Overall, these results are consistent with the alternative localization of the Mae102 isoforms depending on the alternative TSS usage regulated by the growth phase. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. altTSS usage can result in the production of proteins shorter than the annotated ones, lacking N-terminal targeting motif and targeted to a different organelle. Alluvial diagram depicting the predicted localization of long and short protein isoforms as predicted by DeepLoc 2.0 [41]. TSS, transcription start site. https://doi.org/10.1371/journal.pbio.3002724.g004 PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. Mae102 subcellular localization is regulated through alternative TSS usage. (A) IGV visualization of RNA-seq and TSS-seq at the MAE102 locus. In exponential phase at 30°C cells mainly use an altTSS (TSS2) located 19 bp downstream of the annotated ATG. The shorter mRNA produces a protein isoform lacking the MTS (red box) as predicted by DeepLoc 2.0 and MitoFates [41,98]. In stationary phase at 30°C, the annotated TSS (TSS1) is mainly used to produce a full-length transcript that is translated in a protein containing an MTS at its N-terminal. In exponential phase at 30°C, usage of the downstream TSS2 produces a protein expected to go mainly to cytosol. The percentages indicate the proportion of each TSS cluster used in each condition. Detailed numbers are given in S8 Table. Alternative usage of the 2 TSS clusters (B) DIC and fluorescent microscopy images of C. neoformans cells expressing a Mae102-mNeonGreen fusion protein (green) in different mutants: WT, mae102-MTSΔ, mae102-M1R, and tur1Δ. Mitochondria are stained using MitoTracker (red). The percentages refer to the proportion of cells with a Mae102 nuclear localization when are grown under exponential phase. MTS, mitochondrial targeting signal; TSS, transcription start site; WT, wild type. https://doi.org/10.1371/journal.pbio.3002724.g005

altTSS downstream of the start codon The 2,181 remaining TSS clusters (2,576 in C. deneoformans) are positioned at least 90 bp after the annotated start codon and promote the transcription of shorter RNAs than the annotated ones (S4 Table). We named this last category of transcripts TRASS for Transcript Resulting from Alternative Start Sites. One example of TRASS is given in Fig 6. At the CNC03460 locus in C. deneoformans, cells growing in exponential phase at 30°C use mainly an altTSS located within the sixth intron of the gene and positioned within a CDS-intron 1,395 bp downstream of the annotated TSS. In contrast, cells in stationary phase mainly use the annotated altTSS as confirmed by northern blot analysis (Fig 6B). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 6. altTSS usage regulates the production of TRASS in Cryptococcus. (A) IGV visualization of RNA-seq and TSS-seq at the CNC03460 locus when C. deneoformans cells were cultivated at 30°C under exponential (E30) or stationary (S30) phase, respectively. The percentages indicate the proportion of each TSS cluster used in each condition. Detailed numbers are given in S8 Table. (B) Northern blot validation of the alternative RNA molecule production. For CNC03460, the probe was amplified using primers specific of the second half of the gene (arrows). The actin gene (ACT1) was used as reference. TSS, transcription start site. https://doi.org/10.1371/journal.pbio.3002724.g006 Although the function of these RNAs was unknown, we noticed that nearly all the TRASS have coding capacity. Overall, 1,523 potential new proteins could be encoded by these RNAs in C. neoformans. Most of these new proteins would be in frame with the annotated ones and would thus be completely ignored though classical proteomic analysis. To gain insights into the coding capacities of the TRASS, we performed N-terminomic analysis [42] using proteins extracted from cells growing in stationary phase at 30°C. We identified 844 peptide sequences corresponding to N-terminal sequences associated with 12% of the coding genes (n = 810) in C. neoformans. As expected, most of the N-terminal peptides (97%; n = 818) correspond to the annotated N-terminal sequence of 784 proteins (Table A in S6 Table). Also as expected, we identified 19 peptides corresponding to 19 proteins presenting an alternative N-terminal sequence produced from an altTSS close to the annotated ATG (Table B in S6 Table). Finally, 7 N-terminal sequences are likely the products of translation of 7 TRASS (Table B in S6 Table). A striking case is the gene CNAG_04307 (URO1) coding Urate oxidase for which we identified a peptide that is out-of-frame with the main protein. Of course, these represent only a few examples of peptides translated from these RNAs and it is probable that a large part of the TRASS are lncRNAs. Nevertheless, N-terminomic analysis, although powerful and efficient in selecting protein N-termini, is not powerful enough for low abundance alternative N-termini suggesting that much more TRASS encoded peptides are produced in these cells.

Alternative TSS usage in Cryptococcus To further explore the dynamics of alternative TSS usage in Cryptococcus, we evaluated how each TSS is employed depending on the growth phase (Exponential or Stationary) and temperature (30°C or 37°C). TSS cluster usage was evaluated considering an”expression level” in each replicate of each condition measured by the number of TSS-seq reads associated within a given TSS cluster. These numbers were then normalized by the total amount of TSS-seq reads aligned to each coding gene. To limit the potential bias associated with this type of normalization, we limited our analysis to genes for which we could count at least 20 TSS reads in each considered condition. Overall, we considered 3,648 TSS clusters belonging to 1,366 genes in C. neoformans under 4 growth conditions (4,311 TSS clusters belonging to 1,621 genes in C. deneoformans) for this analysis. We confidently identified 1,478 altTSS usage regulations by growth condition in 627 C. neoformans genes (Fig 7A and S7 Table). Alternative TSS usage seems to be more dynamic in C. deneoformans with 2,369 significant regulations in 1,070 genes. Interestingly, 25% of the altTSS usage regulations are specific to the considered comparison (Fig 7A). Explicit examples of these regulations are given on the Fig 7B. At the gene PKP1 (CNAG_00047), 2 TSS are used alternatively depending on the phase of growth. In stationary phase at 30°C, an altTSS within the second intron is mainly used and promotes the transcription of a TRASS. Strikingly, we also identified SOD1 and SOD2 genes as regulated by altTSS usage during this phase transition (S5 Fig). In that case, the pattern of altTSS usage in stationary phase is similar to the one previously reported in copper-limiting condition using gene-specific experiments [33]. Alternative TSS usage at these 2 genes has been reported to be regulated by copper and is dependent on the transcription factor Cuf1 which can recognize a Cu-responsive element (CuRE) found in their promoters [33]. It is possible that the stationary phase condition somehow mimics the copper shortage as used by the Thiele laboratory thus altering altTSS usage at these loci in the same way. We also observed regulation of altTSS by the temperature. For instance, at the locus CNAG_00812, the usage of a downstream TSS is favored in stationary phase at 37°C thus promoting the production of a TRASS. Moreover, at the loci CNAG_01272 and CNAG_03239, a change in temperature in stationary and exponential phase, respectively, alters the usage of an altTSS located within the TL sequence (Fig 7B). Our original idea was to compare the results obtained with the 2 studied sibling species of Cryptococcus. We observed an apparent low conservation of these regulations with only 16.21% and 10.12% of the altTSS usage regulations observed in C. neoformans upon the transition of the growth phase and temperature change, respectively, conserved in C. deneoformans (S4 Table). However, visual examination of some of the “non-conserved” events revealed that the apparent non-conservation was mainly due to the poor level of one or the other TSS in one species which did not pass the different thresholds of our bioinformatic pipelines (S6 Fig). Indeed, in some cases, the position of the altTSS is conserved but regulation is reversed. In others, the altTSS exists but its position is not conserved (S6 Fig). These observations suggest that the degree of conservation of these regulations might be more significant than they appear at first glance. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 7. Differential analysis of altTSS usage in response to the growth phase and temperature changes. (A) Venn diagram illustration the overlap between the altTSS usage regulation (FDR < 0.05) in 4 conditions: exponential phase (E30), stationary phase (S30) at 30°C, exponential phase (E37), and stationary phase (S37) at 37°C in C. neoformans. (B) IGV visualization of RNA-seq and TSS-seq of examples altTSS usage regulated by the temperature and the growth phase. The percentages indicate the proportion of each TSS cluster used in each condition. Detailed numbers are given in S8 Table. qRT-PCR confirmation of these regulations was performed using primers specific of one or both isoforms. Y-axis shows 1,000*fold-change compared to the number of ACT1 mRNA molecules. Experiments were performed in biological triplicates and technical duplicates. Error bars are shown median +/− standard deviation. Green and red arrows indicate the position of the primers used for these experiments. FDR, false discovery rate; TSS, transcription start site. https://doi.org/10.1371/journal.pbio.3002724.g007

Tur1 regulates altTSS usage at the PKP1 gene Our analysis revealed a widespread regulation of altTSS usage in Cryptococcus regulating both transcriptome and proteome structures in response to modifications of the growth condition. Although the mechanisms associated with these regulations remain unknown, the specificity of some of these regulations suggests that alternative TSS usage is mediated by precise regulators. We reasoned that at least some of these regulators might be TFs as previously observed in mammals and in yeast [9,10,33,43]. To identify altTSS usage regulators in Cryptococcus, we screened a library of 155 TF mutant strains [44] using an RT-qPCR assay with primers specific to the long or both the long and the short isoforms identified at the PKP1 gene (CNAG_00047). This gene encodes a protein that shares homology with the S. cerevisiae mitochondrial protein kinase Pkp1 [45]. In C. neoformans, PKP1 transcription can start at 2 altTSS clusters with different expression patterns in exponential and stationary phase at 30°C (Fig 7B). RNA was extracted from the wild-type (WT) strain and the 155 TF mutant strains grown in stationary phase at 30°C. The level of the long and the short mRNA isoforms was then evaluated by RT-qPCR. We identified a single mutant strain altered for altTSS usage at this locus. This mutant strain displayed no growth defect and is mutated at the locus CNAG_05642 encoding a zinc cluster TF previously designated Fzc37 [44]. Sequence conservation analysis revealed that this TF is basidiomycete specific with no predicted function. In this mutant strain, the short isoform is strongly down-regulated in stationary phase compared to the WT strain (t test p-value <0.01), whereas the expression of the long isoform is not altered (t test p-value >0.05) (Fig 8A). We therefore renamed this gene TUR1 (TSS Usage Regulator 1). As expected, complementation of the tur1Δ mutation restored the WT expression of the short isoform (Fig 8A). We confirmed the role of Tur1 in the regulation of altTSS usage at the PKP1 gene by constructing a conditional mutant in which a 2xFlag-CBP tagged version of TUR1 is expressed under the control of the C. neoformans GAL7 promoter [46] (Fig 8C). As expected, the expression of the PKP1 short isoform was only observed in stationary phase when the cells were grown with galactose as the sole source of carbon. No regulation was observed when cells were grown in glucose (Fig 8B), a condition in which the GAL7 promoter is repressed and no Tur1 protein is produced (Fig 8B). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 8. Tur1 regulates the altTSS usage at PKP1 during the transition from exponential phase to stationary phase at 30°C. (A) RT-qPCR measuring the level of the long and short mRNA isoforms of PKP1 in WT, tur1Δ, and the complemented tur1Δ+TUR1 strains grown in stationary or exponential phase at 30°C. Y-axis shows 1,000*fold-change compared to the number of ACT1 mRNA molecules (B) RT-qPCR measuring the level of the long and short mRNA isoforms of PKP1 in a P GAL7 -2xFlag-CBP-TUR1 strain in exponential phase and stationary phase at 30°C, cultured in YPD (glucose) or YPGAL (galactose). Y-axis shows 1,000*fold-change compared to the number of ACT1 mRNA molecules. Experiments were performed in biological triplicates and technical duplicates. Error bars are shown median +/− standard deviation. (C) Western blot analysis of TUR1 conditional expression. Tur1 was detected using an anti-flag antibody and actin was used as control. The data underlying this figure can be found in S1 Data. TSS, transcription start site; WT, wild type. https://doi.org/10.1371/journal.pbio.3002724.g008

A Tur1-dependent Mae102 subcellular targeting As seen above, in a WT context, a growth phase-dependent altTSS usage regulation was observed at the MAE102 gene. In that case, the second TSS is mostly used in exponential phase, whereas in stationary phase, the long isoform targeted to the mitochondria is expressed using the most upstream TSS (Fig 5A). TUR1 deletion completely reversed this pattern (Fig 11A). Indeed, tur1Δ mutant grown in stationary phase preferentially expresses the short isoform, whereas in exponential phase it mostly uses the upstream TSS, only expressing the long isoform albeit at a low level (Fig 11B). To study the impact of TUR1 deletion on Mae102 protein subcellular targeting, we deleted TUR1 in a strain expressing a C-terminal mNeonGreen-tagged version of the Mae102 protein. In this genetic background, in exponential phase, Mae102 does not localize in the nucleus in agreement with the absence of production of the short isoform (Fig 5B). In stationary phase, under which the short isoform is strongly induced in tur1Δ, Mae102 seems to be poorly targeted to the mitochondria. It is noteworthy that Mae102 is not targeted to the nucleus either suggesting that additional factors specifically present in exponential phase are necessary for targeting this protein to the nucleus. Overall, this analysis exemplifies how Tur1 regulates altTSS usage to control protein subcellular targeting during the transition between exponential to stationary phase (Fig 5B). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 11. Tur1 regulates altTSS usage at MAE102. (A) IGV visualization of RNA-seq and TSS-seq of CNAG_06374 in WT and tur1Δ strain grown in exponential phase (E30) and stationary phase (S30) at 30°C. The percentages indicate the proportion of each TSS cluster used in each condition. Detailed numbers are given in S9 Table. (B) RT-qPCR analysis of the level of the long and short MAE102 mRNA isoforms in WT, tur1Δ, and complemented tur1Δ + SH1::TUR1 strains grown in exponential or stationary phase at 30°C. Y-axis shows 1,000*fold-change compared to the number of ACT1 mRNA molecules. Experiments were performed in biological triplicates and technical duplicates. Error bars are shown median +/− standard deviation. The data underlying this figure can be found in S1 Data. TSS, transcription start site; WT, wild type. https://doi.org/10.1371/journal.pbio.3002724.g011

Tur1 could act directly on a subset of genes To determine whether the effect of Tur1 on the regulation of aTSS usage during the transition from exponential to stationary phase could be direct, we constructed a strain expressing a 2× flag tagged version of Tur1 under the control of the actin gene promoter. After phenotypically validating the strain, we used it to perform ChIP-Seq analysis on chromatin purified from cells grown in exponential and stationary phases in triplicate. After peak calling using MASC2 [48], we considered positions where a peak was identified in all 3 stationary phase replicates (151 peaks), or all 3 exponential replicates (29 peaks), or both (400 peaks) to be robust Tur1 binding sites (S10 Table). These 580 Tur1 binding sites were associated with 713 genes. Nearly 60% (n = 414) of these genes were down-regulated in WT during the transition from exponential to stationary phase, but not in the tur1Δ strain, suggesting a direct regulation by Tur1 in this process. Finally, we found that 13% (n = 20) of the genes regulated in a Tur1-dependent manner by alternative TSSs during this phase transition were also bound by Tur1, suggesting a direct role of Tur1 in this regulation at these loci. However, the patterns of ChIP-seq read alignment were not easy to interpret and overall, these ChIP-seq experiments remained partly inconclusive regarding the mechanism by which Tur1 regulates alternative TSS usage in Cryptococcus. We reasoned that the use of a strong promoter to express the tagged version of Tur1 might have confounded our ChIP-Seq results by masking potential regulation. To overcome this problem, we constructed a strain expressing a Dam-Tur1 fusion protein under the control of the native TUR1 promoter and performed DamID-Seq analysis [49] using DNA purified from cells grown in exponential and stationary phases in triplicate. Analysis of the data revealed 1,698 adenine residues associated with 1,283 genes that were specifically methylated in either exponential or stationary phase, suggesting genome-wide regulated Tur1 binding (S11 Table). As for our ChIP-Seq analysis, nearly 60% (n = 731) of these genes were down-regulated in WT during the transition from exponential to stationary phase but not in the tur1Δ strain, suggesting direct regulation. However, the overlap between the 2 lists was only 13% (n = 106). We found that 21% of the genes (n = 32) regulated by aTSS in a Tur1-dependent manner contained methylated adenine residues. Interestingly, a highly regulated adenine residue was identified very close to the second TSS in PKP1, suggesting regulated Tur1 binding (Fig 10B). Overall, both strategies indicate that Tur1 does indeed bind DNA of some genes regulated by alternative TSSs, suggesting that at least for some genes the regulation may be direct. However, the pattern of Tur1 binding regulation during the transition from exponential to stationary phase suggested by the 2 strategies did not provide a straightforward model for Tur1-dependent aTSS regulation.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002724

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.