• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. Oct 2012; 32(20): 4206–4214.
PMCID: PMC3457339

Posttranscriptional Regulation of Cell-Cell Interaction Protein-Encoding Transcripts by Zfs1p in Schizosaccharomyces pombe

Abstract

Members of the tristetraprolin (TTP) family of CCCH tandem zinc finger proteins can bind directly to AU-rich elements in mRNAs and promote transcript deadenylation and decay. The yeast Schizosaccharomyces pombe expresses a single TTP family member, Zfs1p. In this study, we identified probable Zfs1p target mRNAs by comparing transcript levels in wild-type yeast and zfs1Δ mutants, using deep sequencing and microarray approaches. We also used direct RNA sequencing to determine polyadenylation site locations and to confirm the presence of potential Zfs1p target sequences within the target mRNA. These studies identified a set of transcripts containing potential Zfs1p binding sites that accumulated significantly in the zfs1Δ mutants; a subset of these transcripts decayed more slowly in the zfs1Δ mutants and bound directly to Zfs1p in coimmunoprecipitation assays. One apparent direct target encodes the transcription factor Cbf12p, which is known to increase cell-cell adhesion and flocculation when overexpressed. Studies of zfs1Δ cbf12Δ double mutants demonstrated that the increased flocculation seen in zfs1Δ mutants is due, at least in part, to a direct effect on the turnover of cbf12 mRNA. These data suggest that Zfs1p can both directly and indirectly regulate the levels of transcripts involved in cell-cell adhesion in this species.

INTRODUCTION

Zfs1p is a member of the tristetraprolin (TTP) family of CCCH tandem zinc finger (TZF) proteins and the only member of this protein family expressed in Schizosaccharomyces pombe (15). Most yeast genomes sequenced to date contain a single orthologue of zfs1, although two members of this family, CTH1 and CTH2, are present in the Saccharomyces cerevesiae genome (15, 31, 36). Mammalian TTP is a well-characterized CCCH TZF domain-containing RNA binding protein that can promote the decay of its target mRNAs after binding to AU-rich elements (AREs) found in the 3′ untranslated regions (UTRs) of those mRNAs (2, 4, 5, 9, 20). The preferred binding site for TTP family members is the UUAUUUAUU 9-mer, often found multiple times within the 3′-UTRs of target transcripts (3, 18, 19). TTP binds directly to this sequence and then destabilizes the transcript through mechanisms that are not fully understood but involve the initial, stimulated, processive removal of the poly(A) tail, or deadenylation (6).

In S. pombe, Zfs1p is thought to be involved in the mating response pathway, sexual development, and septum formation with exit from mitosis (1, 15). zfs1Δ mutants exhibit a reduced rate of cell division, with decreased cell size, and do not respond properly to pheromone (1, 15). Our group previously searched for potential mRNA targets of Zfs1p by measuring transcripts elevated in zfs1Δ mutants, and we identified several potential target transcripts, including one encoded by arz1, a gene encoding a protein of unknown function (8). We showed that arz1 transcripts were stabilized in the absence of Zfs1p. This transcript contained three optimal 9-mer TTP family binding sites in its 3′-UTR (UUAUUUAUU); A to G mutations within the two 3′-most binding sites completely abrogated the Zfs1p effect on transcript decay, as well as direct binding of Zfs1p to the RNA. These data strongly suggested that the ideal Zfs1p binding sequence is similar to that seen with mammalian TTP. This conclusion is supported by the model of the TZF domain of S. pombe (8), based on the nuclear magnetic resonance structure of the human ZFP36L2 protein bound to the 9-mer UUAUUUAUU (14). In this model, the TZF domain of S. pombe Zfs1p was predicted to form a very similar complex with the same RNA 9-mer as ZFP36L2. Similar conclusions concerning the binding of one of the Zfs1p orthologues in Saccharomyces cerevisiae were reached previously by Puig et al., who used a mutagenesis approach (31).

Since our initial study, the development of a complete deletion series of mutants in S. pombe (17), as well as improved methods for random mutagenesis (12, 23), has made it possible to contemplate genome-wide screens for genetic modifiers in this species. Before embarking on such screens, we decided to use deep RNA sequencing to develop a more complete catalog of transcripts affected by the loss of Zfs1p. Transcripts identified by this means as potential direct Zfs1p targets were further examined for potential Zfs1p binding sites, and the presence of these binding sites within the transcripts was confirmed by direct RNA sequencing to identify sites of polyadenylation; this technique was also used to assess the possibility that the Zfs1p might affect polyadenylation site usage, as suggested recently for S. cerevisiae and Cth2p (30).

Unexpectedly, we found that many transcripts with increased expression in zfs1Δ mutants encoded cell surface glycoproteins involved in cell-cell adhesion. Some of these appeared to be direct targets of Zfs1p, as determined by the presence of Zfs1p binding sites in the transcripts, effects on mRNA decay rates, and coimmunoprecipitation of the transcripts with tagged Zfs1p. One novel target identified by these means was the transcription factor Cbf12p, which is proposed to regulate the biosynthesis of proteins involved in cell-cell adhesion and causes increased cell-cell adhesion and flocculation upon overexpression (29). We found that zfs1Δ mutants exhibited increased flocculation that appeared to be due in part to direct effects on transcripts encoding other adhesin proteins, but also to secondary effects on transcripts regulated by Cbf12p. These data provide a comprehensive list of direct Zfs1p target transcripts that can be used in future studies of genetic modifiers and also suggest that a major physiological function of Zfs1p in this species is to regulate the levels of surface proteins involved in cell-cell interactions.

MATERIALS AND METHODS

Yeast strains and media.

All strains were constructed using standard yeast molecular genetics techniques and are listed in Table S2 of the supplemental material; primers for strain construction are listed in Table S3 of the supplemental material. zfs1 was deleted (zfs1Δ) by using the NAT cassette in parental strain PR110 (referred to here as the wild type [WT]), as described in reference 11, using plasmid pAG25. cbf12 was deleted (cbf12Δ) using the KANMX6 cassette in parental strain PR110 and the zfs1Δ strain as described in reference 22, using plasmid pFa6aKANMX6. PCR was performed to confirm proper chromosomal integration of all mutants, using primers flanking the integration site of the selection cassette. The Zfs1p:HA strain was constructed as described in reference 22, using the pFa6aHAKANMX6 plasmid. In this strain, the hemagglutinin (HA) tag was added in frame to the C terminus of Zfs1p by targeting the endogenous zfs1 locus; isolates were initially screened with PCR using the primers listed in Table S3 and ultimately by Western blotting against the HA epitope (HA antibody sc-805; Santa Cruz Biotechnology Inc., Santa Cruz, CA). Unless otherwise indicated, yeast cultures were grown in yeast extract with supplements medium (YES; MP Biomedicals, Solon, OH) at 30°C.

RNA isolation.

Total cellular RNA was isolated from cells growing at mid-log phase by using the RiboPure yeast RNA purification kit (Life Technologies, Grand Island, NY) following the manufacturer's protocol. Poly(A)+ RNA was isolated from the total RNA samples using the Oligotex Direct mRNA kit (Qiagen, Valencia, CA). Total and poly(A)+ RNA samples were reverse transcribed into cDNAs by using the High Capacity cDNA kit (Life Technologies, Grand Island, NY).

Microarray analysis.

Microarray analysis of gene expression was conducted using Yeast Genome 2.0 gene chip arrays (Affymetrix, Santa Clara, CA). For these analyses, 25 ng of poly(A)+ RNA was amplified as directed for the NuGEN Ovation Pico WTA system protocol and labeled using the NuGEN Encore biotin module. Five micrograms of biotin-amplified RNAs was fragmented and hybridized to each array for 18 h at 45°C in a rotating hybridization oven, as directed for the Nugen FL-Ovation cDNA biotin module V2 protocol, using the Affymetrix eukaryotic target hybridization controls. Array slides were stained with streptavidin-phycoerythrin by utilizing a double-antibody staining procedure and then washed for antibody amplification according to the user manual for the GeneChip hybridization, wash, and stain kit, following protocol FS450-0003. Arrays were scanned in an Affymetrix Scanner 3000, and data were obtained using the GeneChip Command Console software (AGCC, version 1.1). Data preprocessing, normalization, and error modeling were performed with the Rosetta Resolver system (version 7.2).

Direct RNA sequencing analysis.

Direct RNA sequencing was performed as described in reference 27. Briefly, total RNA was isolated from two biological replicates each of WT and zfs1Δ mutants. A 3′ blocking reaction was performed with the poly(A) tailing kit (Ambion) and 3′deoxy-ATP (cordycepin triphosphate; Jena Biosciences, Germany). The blocked RNA was hybridized to flow cell surfaces for sequencing as described previously (27, 28). Sequence read alignments were conducted with the indexDPgenomic software (Helisphere, release 1.2.717) available on the Helicos website and included standard parameters (seed_size, 18; num_erors, 1; weight, 16; best_only, min_norm_score, 4.0; strands both, alignment_type, GL). The S. pombe genome (Schizosaccharomyces_pombe.EF1.55.dna.full_genome.fa file) fasta and .gtf annotation files were downloaded from the Ensembl Fungi resource (http://fungi.ensembl.org/info/data/ftp/index.html) and were used as references for alignment of reads generated with the Helicos BioSciences HeliScope software. Candidate transcripts that were shown to be differentially expressed by a separate mRNA and sequencing (mRNA-Seq) analysis (see below) were examined for read hit density along the genome sequence. The .gtf file-defined protein coding region (CDS) ends of these genes were extracted, and regions from genomic position −1000 to +2000 relative to the stop codon of the CDS were examined for mRNA sense-strand read hit density. These hits were binned in 300 10-bp bins, with counts within each bin derived using the positions of the 5′ end of the read, such that each read was defined as a single point. Total reads were determined for each bin, graphed, and examined manually for the 3′-most differentially expressed sequences.

mRNA-Seq analysis.

Total RNA was isolated from two independent isolates each of WT and zfs1Δ mutants by using the RiboPure yeast RNA purification kit, as described above. The RNA was quantitated with a Qubit fluorometer, and 10 μg of each sample was transcribed to generate cDNA libraries and sequenced at the NIH Intramural Sequencing Center (http://www.nisc.nih.gov) by using 50-bp paired-end reads on a Genome Analyzer IIx (Ilumina Inc., San Diego, CA). MAQ (21) was used to map paired-end reads from each replicate separately to all annotated cDNA sequences of the S. pombe genome (ftp://ftp.sanger.ac.uk/pub/yeast/pombe/GFF/pombe_09052011.gff), with the maximum allowed mapping distance between a read pair set at <500 bp. After read mapping, the EpiCenter program (http://www.niehs.nih.gov/research/resources/software/biostatistics/epicenter) (13) and the statistical computing language R (http://www.r-project.org/) were used for statistical analyses.

ARE identification and base composition.

Based on the latest version of the S. pombe genome annotation (pombe_09052011.gff) from the Sanger Institute, RefSeq from hg19 for human, and Mm10 for mouse, we extracted all annotated 3′-UTR sequences. We created a customized program to count the number of nonoverlapping 7-mers (UAUUUAU) in each 3′-UTR by the simple exact match approach. We also calculated the frequencies of all four nucleotides in the 3′-UTR sequences of the entire S. pombe, human, and mouse transcriptomes.

Real-time RT-PCR analysis.

For each sample, approximately 1 μg of total cellular RNA was reverse transcribed using the ABI high-capacity reverse transcription kit, following the manufacturer's protocol. Real-time reverse transcription-PCR (RT-PCR) analysis was performed using the ABI Prism 7900 HT system. All cDNAs were diluted and then subjected to real-time PCR with the TaqMan Sybr green PCR master mix (Life Technologies, Grand Island, NY) with transcript-specific primers, according to the manufacturer's protocol. Transcript-specific primers were designed using the Bähler website (http://www.bahlerlab.info/microarray/) and are listed in Table S3 of the supplemental material. Relative abundance was determined by normalizing to an internal control, Spbc32f12.11 (gapdh).

Flocculation assays.

Flocculation in mid-log-phase yeast cells was measured as described previously (35). Briefly, 5 ml of logarithmically growing culture was centrifuged at 4,500 × g for 5 min at 4°C. Cells were then washed twice in 250 mM EDTA, once in 250 mM NaCl (pH 2), and finally in 250 mM NaCl (pH 4.5). Washed cells were resuspended to a final concentration of 1 × 108 cells/ml in 25 ml of 250 mM NaCl (pH 4.5) and placed into a 25-ml graduated cylinder. The cell suspension was adjusted to 4 mM CaCl2 with 100 mM CaCl2 (pH 4.5) and inverted 18 times. At defined intervals, 100 μl of cell suspension was removed from a fixed position in the graduated cylinder and diluted with 900 μl of water, and absorbance was then immediately measured at 600 nm. All values were normalized to 100% cells in suspension at time point zero.

mRNA decay assays.

The protein coding region of each candidate target transcript, plus 500 bp of downstream sequence, was PCR amplified using the primers listed in Table S3 of the supplemental material; the products were then cloned into the pSLF273 vector (10). Each target vector was then transformed into WT and zfs1Δ cells by the lithium-acetate method (26), and positive transformants were verified by Western blot analysis against the HA epitope. To induce expression, the indicated strains were grown in Edinburgh minimal medium (EMM; MP Biomedicals, Solon, OH) lacking uracil and thiamine until the mid-log growth phase; cells were then diluted to an A600 of 0.5 and grown for a further 1 h. At this point, 10 μM thiamine was added to repress expression from the nmt promoter. At the indicated time points, 1.5-ml samples of each culture were taken and flash-frozen, and RNA was extracted and used for real-time RT-PCR as described above.

RNA immunoprecipitation.

Zfs1p:HA and an isogenic no-tag control (PR110) were grown in 100 ml of YES at 30°C until mid-log phase and processed for RNA immunoprecipitation analysis as described previously (16, 40). Cell lysates were prepared exactly as described in reference 40, with the exception of the exclusion of EDTA from the lysis buffer. Protein concentrations were determined for each of the cleared lysates, and 1 μg of lysate protein was used for each immunoprecipitation. Ten percent of the lysate was removed for isolation of total input RNA. The remaining lysate was rotated for 4 h at 4°C with 50 μl of preblocked HA-protein A–Sepharose beads (sc-7392; Santa Cruz Biotechnology Inc., Santa Cruz, CA). The beads were then washed 5 times with ice-cold TMG100 buffer [10 mM Tris (pH 8.0), 1 mM MgCl2, 100 mM NaCl, 10% glycerol, 0.1 mM dithiothreitol, 2 mM phenylmethanesulfonylfluoride, 0.2 mM (4-amidinophenyl)-methanesulfonyl fluoride, 1 U/μl RNasin (Promega, Madison, WI), and Complete protease inhibitor cocktail] and treated with proteinase K as described in reference 40. Coimmunoprecipitated RNA was extracted with acidic phenol-chloroform-isoamyl alcohol and ethanol precipitated. Purified RNA was digested with 20 U of DNase I (Life Technologies, Grand Island, NY), ethanol precipitated, and subjected to real-time RT-PCR analysis with the indicated primers (see Table S3 in the supplemental material).

Microarray data accession numbers.

The microarray analysis results were deposited in the NCBI GEO database and are accessible through GEO series accession number GSE35009. Results of the mRNA-Seq analysis were deposited in the NCBI GEO and are accessible through GEO series accession number GSE35376.

RESULTS

Transcripts elevated in the absence of zfs1.

Previous studies from our group utilized microarray analysis of total RNA to examine transcripts that were elevated in zfs1Δ mutants. We identified several potential target transcripts by this means, including a transcript encoded by arz1, a gene encoding a protein of unknown function (8). We have extended these studies, as well as confirmed our initial analysis, using mRNA-Seq analysis, as well as additional microarrays of poly(A)+-selected RNA. Using mRNA-Seq analysis, we found 185 transcripts that were increased by at least 1.5-fold in the zfs1Δ mutant strain compared to the parental WT strain and were significantly different (P < 0.05). Similarly, 195 transcripts were significantly downregulated by 1.5-fold or more. A search for target sequences within the most recently updated S. pombe transcriptome, which contains annotated 3′-UTRs, identified 119 of the 185 (64%) significantly upregulated transcripts that contained at least one potential binding site 7-mer (UAUUUAU) in their 3′-UTRs, with overlapping 7-mers only counted once (see Table S1 in the supplemental material). In contrast, only 18% of the downregulated transcripts contained a single 7-mer binding site. Of the 119 upregulated transcripts, 68/119 (59%) were not observed in the previous microarray analysis (8), for which similar cutoffs were used; however, 20 of the 25 most upregulated transcripts were in both data sets.

To determine whether the putative target transcripts contained Zfs1p binding sites within their 3′-UTRs and to determine whether the Zfs1p deficiency resulted in the use of novel polyadenylation sites, we performed direct RNA sequencing in WT and zfs1Δ mutant strains. Direct RNA sequencing produces single-strand sequence reads that average 33 to 34 nucleotides in length without the need for cDNA amplication (28). In addition, direct RNA sequencing allows identification of poly(A) tail attachment sites at a resolution of ±2 nucleotides (19). By comparing sequence read alignments in the 3′-UTRs of the top prospective Zfs1p target transcripts, we determined that there were no apparent differences in polyadenylation site selection between WT and zfs1Δ mutants (Fig. 1A to toH).H). Many of these prospective target transcripts contained several polyadenylation sites, although the use of these sites did not appear to differ between WT and zfs1Δ mutants (Fig. 1C to toI).I). Moreover, the direct RNA sequencing confirmed most of the 7-mer putative Zfs1p binding sites identified in the updated S. pombe transcriptome (see Table S1).

Fig 1
Polyadenylation site determinations for the WT and zfs1Δ mutants. Direct RNA sequencing was performed using total RNA purified from two independent isolates of WT and zfs1Δ mutants. Data shown are from 7 transcripts elevated for the zfs1Δ ...

Annotated 3′-UTRs in S. pombe are relatively A/T rich, with 31% A, 37% T, 16% C, and 17% G, compared to 27% A, 30% T, 21% C, and 22% G in humans, and 27% A, 29% T, 22% C, and 22% G in the mouse. Overall, 22% of the annotated S. pombe 3′-UTRs (1,076 of 4,805) contained at least a single 7-mer potential binding site, compared to frequencies of 23% in mice and 26% in humans. In other words, the frequency of 7-mer potential binding sites was very similar in all three organisms, ranging from 22 to 26%.

We also examined the potential S. pombe targets to determine whether they clustered into particular cellular functions or processes. By using the DAVID database (Database for Annotation, Visualization, and Integrated Discovery) to perform GO analysis, we found 16 transcripts in the S. pombe transcriptome that encode proteins involved in biological adhesion. Of those 16 transcripts, four were upregulated in zfs1 mutants. Many of these transcripts encoded cell surface glycoproteins and other proteins suspected of being involved in cell-cell adhesion. One of these transcripts encoded the transcription factor Cbf12p, known to promote cell-cell adhesion or flocculation when overexpressed (29). We therefore examined whether zfs1Δ mutants flocculated abnormally and whether this was dependent on the presence of cbf12.

Flocculation of zfs1Δ and cbf12Δ mutants.

In a standard flocculation assay, we found that the zfs1Δ mutants rapidly settled out of suspension after the addition of calcium, whereas WT cells did not flocculate under these conditions (Fig. 2A). When cbf12 was deleted in the zfs1Δ mutants, the abnormal flocculation observed in the zfs1Δ mutants was abolished (Fig. 2B). These data suggest that the flocculation observed in the zfs1Δ mutant was due to the overexpression of Cbf12p, which should follow from the abnormal accumulation of its transcript in the zfs1Δ mutant.

Fig 2
Flocculation analysis of zfs1Δ and cbf12Δ mutants. (A) Flocculation of WT and zfs1Δ cells (A) and flocculation of WT, zfs1Δ, cbf12Δ, and zfs1Δ cbf12Δ cells (B) was initiated by the addition of CaCl ...

Identification of targets of Zfs1p.

To identify which of the transcripts elevated in the zfs1Δ mutant were likely to be indirectly elevated because of Cbf12p overexpression, we performed a microarray analysis, comparing gene expression in the WT, zfs1Δ, cbf12Δ, and zfs1Δ cbf12Δ strains. When we focused on transcripts that were elevated in both zfs1Δ and zfs1Δ cbf12Δ strains, we found only 16 transcripts that were significantly increased 1.5-fold or more in both the single and double mutant strains (P < 0.05), suggesting that a large number of the previously identified potential target transcripts were elevated secondary to increases in Cbf12p. Of these 16 transcripts, 11 contained at least a single potential Zfs1p 7-mer binding site; these 11, plus cbf12 and ecl3, are listed in Table 1. Seven of these were chosen for further analysis, as indicated in Table 1. These changes in transcript levels were confirmed by real-time RT-PCR (Fig. 3). We also confirmed the increased expression of two transcripts, Spac1F8.06 and Spcc188.09c, which were upregulated in the zfs1Δ mutants but not in the zfs1Δ cbf12Δ double mutants (Fig. 3). Although most of the transcripts encoding cell adhesion proteins were no longer elevated in the zfs1Δ cbf12Δ double mutant, the Spbc359.04c transcript, encoding a cell surface glycoprotein, remained significantly increased in the zfs1Δ cbf12Δ double mutant, although to a slightly lesser extent than in the zfs1Δ mutant alone (Table 1 and Fig. 3).

Table 1
Potential direct Zfs1p targetsa
Fig 3
Real-time RT-PCR analysis of target transcripts. Poly(A)+-specific mRNA was extracted from WT, zfs1Δ, cbf12Δ, and zfs1Δ cbf12Δ cells and subjected to real-time RT-PCR with gene-specific primers. Shown are the mean values ...

Zfs1p-dependent decay of target transcripts.

Using the nmt expression system (7), we measured the stability of a subset of the significantly increased transcripts that were at least 1.5-fold increased in expression in the zfs1Δ cbf12Δ mutants and contained at least a single 9-mer in the 3′-UTR (Table 1). In addition, we measured the stability of one of the transcripts, Spbc3E7.02c, that exhibited decreased expression in the zfs1Δ mutants. For these measurements, WT and zfs1Δ mutants were transformed with plasmids bearing the identified Zfs1p target mRNAs, including 500 bp of the 3′-UTR, under the control of the nmt promoter. The transformants were grown in the absence of thiamine to permit expression of the target mRNAs, and then thiamine was added to repress transcription of the nmt promoter. RNA was then harvested at intervals, and the abundance of the target transcripts was measured relative to the levels of an internal control transcript encoding Gapdhp. The decay of all seven proposed Zfs1p targets was delayed in the zfs1Δ mutants compared to WT cells (Fig. 4A to toG),G), while the decay of the single downregulated transcript, Spbc3E7.02c, was not affected by the loss of Zfs1p (Fig. 4H). These data suggest that Zfs1p can directly regulate the turnover of these target transcripts.

Fig 4
Decay analysis of Zfs1p target transcripts. WT and zfs1Δ strains were transformed with plasmids containing the indicated genes under the control of the nmt promoter, followed by 500 bp of the mRNA 3′-UTR. The strains were grown in medium ...

Direct binding of Zfs1p to target transcripts.

To test whether Zfs1p could directly interact with the proposed target transcripts, we performed RNA immunoprecipitation (RIP) with a strain that expressed a fusion protein of Zfs1p with HA, linked to the Zfs1p C terminus. This tag was integrated into the endogenous chromosomal locus rather than expressed from a plasmid, and thus the tagged Zfs1p fusion protein was expressed from its own promoter. After immunoprecipitation of the Zfs1p fusion protein with an HA antibody, we found that the top seven putative target transcripts were relatively enriched in the immunoprecipitated RNA isolated from the Zfs1p:HA strain compared to the control no-tag strain (Fig. 5). We also examined Zfs1p binding to two transcripts, Spac186.01 and Spcc1742.01, which were initially identified as potential Zfs1p targets in the mRNA-Seq analysis but were not significantly increased in the zfs1Δ cbf12Δ double mutant. There was no significant enrichment of these transcripts in the Zfs1p:HA strain compared with the no-tag strain (Fig. 5), confirming that they are not likely direct targets of Zfs1p.

Fig 5
RIP analysis of Zfs1p binding to target transcripts. RNA immunoprecipitation was performed with strains expressing a C-terminal HA-tagged Zfs1p and a no-tag control. Proteins and associated RNA were precipitated with anti-HA antibodies, analyzed using ...

Phylogenetic considerations.

The Schizosaccharomyces genus is thought to have split off from the ancestral fungal lineage approximately 220 million years ago (32). In addition to S. pombe, the genomes from the other three known members of the genus (S. cryophilus, S. octosporus, and S. japonicus) have been sequenced. A single Zfs1p-like protein appears to be expressed in each of these three species, with remarkable lack of amino acid alignment outside the extreme C-terminal tandem zinc finger domain (see Fig. S1 in the supplemental material). Overall, there was 54% amino acid identity between the proteins from S. pombe and S. cryophilus and 52% identity between S. pombe and S. octosporus, with S. japonicus showing much lower similarity.

To determine whether these four related species might share Zfs1p target transcripts, we searched 3′-UTR sequences within the orthologues from the other species for potential Zfs1p binding motifs. As examples, we chose the seven high-likelihood mRNA targets of S. pombe Zfs1p shown in Table 1. The readout was the number of UAUUUAU 7-mers within the 3′-UTRs of the putative targets (see Table S4 of the supplemental material). There was a remarkable lack of conservation of the number of 7-mer binding sites, even within these members of the same genus, with some not containing any of these minimal binding sites (see Table S4).

DISCUSSION

In the present study, we attempted to generate a complete catalog of direct Zfs1p target transcripts to use as readouts for future studies of genetic interactions in this species. We used several technical improvements in the present study. First, we used deep sequencing analysis of poly(A)+-specific mRNA. This method allowed for the confirmation of some transcripts from our earlier study (8), and it also allowed for the identification of novel potential targets that were not identified by microarray. Second, we performed additional microarrays using poly(A)+ RNA and used these to support the identification of many of the transcripts highlighted by deep sequencing. Third, we used direct RNA sequencing to confirm the presence or absence of the putative Zfs1p binding site sequences within the 3′-UTRs, rather than in flanking DNA regions. Finally, we measured mRNA decay rates and direct Zfs1p binding to the final set of proposed target transcripts. These techniques, coupled with the use of the cbf12 deletion mutants discussed below, allowed us to propose a discrete set of seven transcripts that we believe to be direct binding and decay targets of Zfs1p in this species under these growth conditions.

Several potential target transcripts identified by these means encoded proteins that are thought to be involved in cell-cell interactions or flocculation. Flocculation is defined as a reversible condition in which cells aggregate spontaneously and form “flocs” due to the exhaustion of nutrients during the stationary phase of growth (34). It is important in certain industrial applications, such as in the brewing industry, where flocculation is a desirable trait that allows the separation of yeast during the brewing process (37, 38). Cell-cell interactions and cell adhesion to other surfaces also are important traits that affect the pathogenicity of certain human fungal pathogens (33, 39).

One of the proposed direct targets identified by our experiments is the transcript encoded by cbf12. This transcript was consistently elevated in the zfs1Δ mutant yeast. It has been shown that overexpression of the transcription factor Cbf12p results in self-aggregation or flocculation, possibly due to transcriptional upregulation of a number of genes encoding cell surface glycoproteins, or adhesins (29). In the cell cycle, cbf12 expression peaks during stationary phase, possibly to allow flocculation to occur as a stress response under unfavorable conditions (29). We had noted that zfs1Δ mutants often flocculated during stationary phase, a finding that was confirmed in the present study in a calcium-induced flocculation assay (35). We found that the disruption of cbf12 abolished the calcium-induced flocculation observed in zfs1Δ mutants; this is therefore likely to be due to the intermediacy of Cbf12p, as shown by results with the double zfs1Δ cbf12Δ mutant.

The same set of single and double mutant strains was examined to determine whether Cbf12p plays a role in the increased expression of the adhesin transcripts observed to be elevated in the zfs1Δ mutant. Most of the identified adhesin mRNAs were not upregulated in the double mutant, suggesting that they were indirectly elevated in the zfs1Δ mutant due to increased expression of Cbf12p. One exception, besides the cbf12 transcript itself, was the transcript encoding Spbc359.04c, a known cell surface adhesion glycoprotein. This transcript was consistently the most highly elevated of all potential targets in the zfs1Δ cells and thus might represent a useful indicator of Zfs1p activity in genetic modifier experiments.

One of the other novel Zfs1p targets identified in these studies was a transcript that was not represented on the microarray chips, i.e., Spbc8E4.12c, or ecl3, a gene thought to be involved in the extension of chronological life span (25). Little is known about the function of the encoded protein and its involvement in aging in S. pombe, but it will be of interest to further investigate a potential role for Zfs1p in the aging process.

We examined the sites of poly(A) tail attachment in the WT and zfs1Δ mutants for two reasons. First, from a practical standpoint, we needed to know whether the potential Zfs1p binding sites in each identified putative target were actually present in the transcripts rather than in flanking DNA sequences. Second, previous studies in S. cerevisiae had suggested that one of the Zfs1p orthologues in that species, Cth2p, had an influence on 3′-end processing and extension of specific target mRNAs (30). Direct RNA sequencing allowed us to identify the sites of poly(A) tail attachment for our potential target sequences, and the differences in mRNA target abundance between WT and zfs1Δ mutants confirmed that the target transcript was the one represented in the sequence traces. We used this method to confirm the presence of potential Zfs1p binding sites in the 3′-UTRs of our proposed targets. However, we did not find any differences in the relative usage of poly(A) tail attachment sites between the WT and zfs1Δ mutant strains, either in our proposed target list or in other mRNAs whose 3′ ends were directly sequenced.

In S. cerevisiae there are two TTP family member genes, CTH1 and CTH2, which have been shown to be involved in iron metabolism through the regulated turnover of a subset of iron regulatory transcripts (31). However, previous studies showed that the loss of zfs1 had no effect on the expression of iron metabolism genes in S. pombe (24). The present studies again failed to identify a link between the loss of zfs1Δ and transcripts related to iron metabolism. It is fascinating that the divergent evolution of these two organisms since their common ancestor has resulted in apparently different physiological roles for the TTP family member proteins in the two species, despite their evidently similar biochemical roles in binding to ARE-containing mRNAs and promoting their decay. Even within the four members of the Schizosaccharomyces genus, there was lack of conservation of both the number or, in some cases, even the presence of the core 7-mer Zfs1p binding sites in the orthologues of the S. pombe Zfs1p target transcripts in these other species. It seems possible that even these relatively closely related yeast species will have evolved separate sets of mRNA targets, with different physiological consequences.

Supplementary Material

Supplemental material:

ACKNOWLEDGMENTS

We thank Danica Andrews and Rick Fannin of the NIEHS Microarray Group for their help in obtaining the microarray data and the NIH Intramural Sequencing Center for the mRNA-Seq data. We also thank Shay Covo and Jessica Williams for critical comments on the manuscript and Paul Russell and Jessica Williams for providing the PR110 strain.

This work was supported by the Intramural Research Program of the National Institutes of Health, National Institute of Environmental Health Sciences.

Footnotes

Published ahead of print 20 August 2012

Supplemental material for this article may be found at http://mcb.asm.org/.

REFERENCES

1. Beltraminelli N, Murone M, Simanis V. 1999. The S. pombe zfs1 gene is required to prevent septation if mitotic progression is inhibited. J. Cell Sci. 112:3103–3114. [PubMed]
2. Blackshear PJ. 2002. Tristetraprolin and other CCCH tandem zinc-finger proteins in the regulation of mRNA turnover. Biochem. Soc. Trans. 30:945–952. [PubMed]
3. Brewer BY, Malicka J, Blackshear PJ, Wilson GM. 2004. RNA sequence elements required for high affinity binding by the zinc finger domain of tristetraprolin: conformational changes coupled to the bipartite nature of AU-rich mRNA-destabilizing motifs. J. Biol. Chem. 279:27870–27877. [PubMed]
4. Carballo E, Lai WS, Blackshear PJ. 2000. Evidence that tristetraprolin is a physiological regulator of granulocyte-macrophage colony-stimulating factor messenger RNA deadenylation and stability. Blood 95:1891–1899. [PubMed]
5. Carballo E, Lai WS, Blackshear PJ. 1998. Feedback inhibition of macrophage tumor necrosis factor-alpha production by tristetraprolin. Science 281:1001–1005. [PubMed]
6. Chen CY, Shyu AB. 2011. Mechanisms of deadenylation-dependent decay. Wiley Interdiscip. Rev. RNA 2:167–183. [PMC free article] [PubMed]
7. Cuthbertson BJ, Blackshear PJ. 2008. Evaluating the control of mRNA decay in fission yeast. Methods Enzymol. 449:73–95. [PubMed]
8. Cuthbertson BJ, Liao Y, Birnbaumer L, Blackshear PJ. 2008. Characterization of zfs1 as an mRNA-binding and -destabilizing protein in Schizosaccharomyces pombe. J. Biol. Chem. 283:2586–2594. [PMC free article] [PubMed]
9. De J, et al. 1999. Identification of four CCCH zinc finger proteins in Xenopus, including a novel vertebrate protein with four zinc fingers and severely restricted expression. Gene 228:133–145. [PubMed]
10. Forsburg SL, Sherman DA. 1997. General purpose tagging vectors for fission yeast. Gene 191:191–195. [PubMed]
11. Goldstein AL, McCusker JH. 1999. Three new dominant drug resistance cassettes for gene disruption in Saccharomyces cerevisiae. Yeast 15:1541–1553. [PubMed]
12. Guo Y, Levin HL. 2010. High-throughput sequencing of retrotransposon integration provides a saturated profile of target activity in Schizosaccharomyces pombe. Genome Res. 20:239–248. [PMC free article] [PubMed]
13. Huang W, et al. 2011. Efficiently identifying genome-wide changes with next-generation sequencing data. Nucleic Acids Res. 39:e130 doi:10.1093/nar/gkr592. [PMC free article] [PubMed]
14. Hudson BP, Martinez-Yamout MA, Dyson HJ, Wright PE. 2004. Recognition of the mRNA AU-rich element by the zinc finger domain of TIS11d. Nat. Struct. Mol. Biol. 11:257–264. [PubMed]
15. Kanoh J, Sugimoto A, Yamamoto M. 1995. Schizosaccharomyces pombe zfs1+ encoding a zinc-finger protein functions in the mating pheromone recognition pathway. Mol. Biol. Cell 6:1185–1195. [PMC free article] [PubMed]
16. Keene JD, Komisarow JM, Friedersdorf MB. 2006. RIP-Chip: the isolation and identification of mRNAs, microRNAs and protein components of ribonucleoprotein complexes from cell extracts. Nat. Protoc. 1:302–307. [PubMed]
17. Kim DU, et al. 2010. Analysis of a genome-wide set of gene deletions in the fission yeast Schizosaccharomyces pombe. Nat. Biotechnol. 28:617–623. [PMC free article] [PubMed]
18. Lai WS, Carrick DM, Blackshear PJ. 2005. Influence of nonameric AU-rich tristetraprolin-binding sites on mRNA deadenylation and turnover. J. Biol. Chem. 280:34365–34377. [PubMed]
19. Lai WS, Kennington EA, Blackshear PJ. 2002. Interactions of CCCH zinc finger proteins with mRNA: non-binding tristetraprolin mutants exert an inhibitory effect on degradation of AU-rich element-containing mRNAs. J. Biol. Chem. 277:9606–9613. [PubMed]
20. Lai WS, Parker JS, Grissom SF, Stumpo DJ, Blackshear PJ. 2006. Novel mRNA targets for tristetraprolin (TTP) identified by global analysis of stabilized transcripts in TTP-deficient fibroblasts. Mol. Cell. Biol. 26:9196–9208. [PMC free article] [PubMed]
21. Li H, Ruan J, Durbin R. 2008. Mapping short DNA sequencing reads and calling variants using mapping quality scores. Genome Res. 18:1851–1858. [PMC free article] [PubMed]
22. Longtine MS, et al. 1998. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14:953–961. [PubMed]
23. Majumdar A, Chatterjee AG, Ripmaster TL, Levin HL. 2011. Determinants that specify the integration pattern of retrotransposon Tf1 in the fbp1 promoter of Schizosaccharomyces pombe. J. Virol. 85:519–529. [PMC free article] [PubMed]
24. Mercier A, Pelletier B, Labbe S. 2006. A transcription factor cascade involving Fep1 and the CCAAT-binding factor Php4 regulates gene expression in response to iron deficiency in the fission yeast Schizosaccharomyces pombe. Eukaryot. Cell 5:1866–1881. [PMC free article] [PubMed]
25. Ohtsuka H, Ogawa Y, Mizuno H, Mita S, Aiba H. 2009. Identification of Ecl family genes that extend chronological lifespan in fission yeast. Biosci. Biotechnol. Biochem. 73:885–889. [PubMed]
26. Okazaki K, et al. 1990. High-frequency transformation method and library transducing vectors for cloning mammalian cDNAs by trans-complementation of Schizosaccharomyces pombe. Nucleic Acids Res. 18:6485–6489. [PMC free article] [PubMed]
27. Ozsolak F, et al. 2010. Comprehensive polyadenylation site maps in yeast and human reveal pervasive alternative polyadenylation. Cell 143:1018–1029. [PMC free article] [PubMed]
28. Ozsolak F, et al. 2009. Direct RNA sequencing. Nature 461:814–818. [PubMed]
29. Prevorovsky M, et al. 2009. Cbf11 and Cbf12, the fission yeast CSL proteins, play opposing roles in cell adhesion and coordination of cell and nuclear division. Exp. Cell Res. 315:1533–1547. [PubMed]
30. Prouteau M, Daugeron MC, Seraphin B. 2008. Regulation of ARE transcript 3′ end processing by the yeast Cth2 mRNA decay factor. EMBO J. 27:2966–2976. [PMC free article] [PubMed]
31. Puig S, Askeland E, Thiele DJ. 2005. Coordinated remodeling of cellular metabolism during iron deficiency through targeted mRNA degradation. Cell 120:99–110. [PubMed]
32. Rhind N, et al. 2011. Comparative functional genomics of the fission yeasts. Science 332:930–936. [PMC free article] [PubMed]
33. Silva S, et al. 2011. Adherence and biofilm formation of non-Candida albicans Candida species. Trends Microbiol. 19:241–247. [PubMed]
34. Soares EV. 2011. Flocculation in Saccharomyces cerevisiae: a review. J. Appl. Microbiol. 110:1–18. [PubMed]
35. Soares EV, MM 1997. Quantification of yeast flocculation. J. Inst. Brew 103:93–98.
36. Thompson MJ, Lai WS, Taylor GA, Blackshear PJ. 1996. Cloning and characterization of two yeast genes encoding members of the CCCH class of zinc finger proteins: zinc finger-mediated impairment of cell growth. Gene 174:225–233. [PubMed]
37. Van Holle A, Machado MD, Soares EV. 2012. Flocculation in ale brewing strains of Saccharomyces cerevisiae: re-evaluation of the role of cell surface charge and hydrophobicity. Appl. Microbiol. Biotechnol. 93:1221–1229. [PubMed]
38. Verstrepen KJ, Derdelinckx G, Verachtert H, Delvaux FR. 2003. Yeast flocculation: what brewers should know. Appl. Microbiol. Biotechnol. 61:197–205. [PubMed]
39. Verstrepen KJ, Klis FM. 2006. Flocculation, adhesion and biofilm formation in yeasts. Mol. Microbiol. 60:5–15. [PubMed]
40. Webb CJ, Zakian VA. 2008. Identification and characterization of the Schizosaccharomyces pombe TER1 telomerase RNA. Nat. Struct. Mol. Biol. 15:34–42. [PMC free article] [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...