ecl1+, ecl2+ and ecl3+ genes encode highly homologous small proteins, and their over-expressions confer both H2O2 stress resistance and chronological lifespan extension on Schizosaccharomyces pombe. However, the mechanisms of how these Ecl1 family proteins function have not been elucidated. In this study, we conducted microarray analysis and identified that the expression of genes involved in sexual development and stress responses was affected by the over-expression of Ecl1 family proteins. In agreement with the mRNA expression profile, the cells over-expressing Ecl1 family proteins showed high mating efficiency and resistant phenotype to H2O2. We showed that the H2O2-resistant phenotype depends on catalase Ctt1, and over-expression of ctt1+ does not affect chronological lifespan. Furthermore, we showed that six genes, ste11+, spk1+, hsr1+, rsv2+, hsp9+ and lsd90+, whose expressions are increased in cells over-expressing Ecl1 family proteins are involved in chronological lifespan in fission yeast. Among these genes, the induction of ste11+ and hsr1+ was dependent on a transcription factor Prr1, and we showed that the extensions of chronological lifespan by Ecl1 family proteins are remarkably diminished in prr1 deletion mutant. From these results, we propose that Ecl1-family proteins conduct H2O2 stress resistance and chronological lifespan extension in ctt1+- and prr1+-dependent manner, respectively.
In the past years, simple organisms such as yeasts and worms have contributed a great deal to aging research (Roux et al. 2010). Especially, studies used in Saccharomyces cerevisiae elucidated a significant number of molecular mechanisms underlying cellular aging and discovered novel longevity genes, such as Ras2, Tor1 and Sch9 (Wei et al. 2008). Recently, in fission yeast Schizosaccharomyces pombe, it was reported that disruption of pka1+ and sck2+ also increases the chronological lifespan (Roux et al. 2006; Ohtsuka et al. 2008).
The chronological lifespan of yeast cells is measured by viability after entry into the stationary phase. Previously, we characterized the genes involved in chronological lifespan in Schizosaccharomyces pombe (Oshiro et al. 2003; Fujita et al. 2007; Ohtsuka et al. 2008, 2009). Both lcf1+ and lcf2+ encode a long-chain fatty acyl-CoA synthetase. The mutant of lcf1+ shows the rapid loss of viability after entry into the stationary phase, whereas the mutant of lcf2+ lives longer than wild type in SD medium. ecl1+, ecl2+ and ecl3+ genes encode highly homologous small proteins, and they extend chronological lifespan of fission yeast when over-expressed, albeit their functions remain unclear. Interestingly, it is reported that many long-lived mutants were more resistant to oxidative stress than wild type (Parkes et al. 1998; Chen et al. 2005; Schriner et al. 2005; Roux et al. 2006; Ito et al. 2010), and in accordance with this, ecl1+-, ecl2+- and ecl3+-over-expressed cells also showed resistant phenotype to H2O2.
ecl1+, ecl2+ and ecl3+ encode proteins with weak sequence homology to protein encoded by YGR146C in S. cerevisiae (Azuma et al. 2009). YGR146C gene also extends chronological lifespan of S. cerevisiae when over-expressed (Azuma et al. 2009). At present, these Ecl1 family proteins have been identified only in fungi groups but not in higher eukaryotes.
In this study, microarray analysis was carried out to know the mechanisms how Ecl1 family proteins extend chronological lifespan. And the results suggested that the Ecl1 family proteins of S. pombe have similar function each other. Based on the array data, we developed our study and showed that the increased resistance to H2O2 stress by ecl1+, ecl2+ and ecl3+ over-expression depends on ctt1+ and the chronological lifespan extension by them depends on prr1+. It was suggested that the ability to extend lifespan is independent on the ability to resist oxidative stress in fission yeast.
DNA microarray analysis of cells over-expressing Ecl1, Ecl2 and Ecl3
We previously identified new genes, ecl1+, ecl2+ and ecl3+, which encode the small proteins composed of 80, 84 and 89 amino acids, respectively, in S. pombe (Ohtsuka et al. 2008, 2009). When these genes are over-expressed from plasmids pEcl1, pEcl2 and pEcl3, the survival rate in stationary phase remains higher than that of control. However, the mechanisms of lifespan extension by them have not been elucidated. To get a clue how ecl1+, ecl2+ and ecl3+ affect the chronological lifespan and to clarify the changes in gene expressions in chronologically long-lived cells, microarray analysis was carried out.
mRNA expression profiles of S. pombe carrying pEcl1, pEcl2 and pEcl3, respectively, at the logarithmic growth phase were analyzed by means of microarray analysis. As a result, it was shown that some genes whose expressions are changed by pEcl1 are also changed in both cells carrying pEcl2 and pEcl3 (Fig. 1). There were 83, 140 and 122 genes that were increased more than 1.7-fold by pEcl1, pEcl2 and pEcl3, respectively. Among 83 genes increased by pEcl1, 71 genes were also increased by pEcl2 (86% identical), whereas 73 genes by pEcl3 (88% identical). And among 122 genes increased by pEcl3, 96 genes were also increased by pEcl2 (79% identical). Furthermore, 65 genes were commonly increased in each cell carrying pEcl1, pEcl2 and pEcl3.
The number of genes decreased by pEcl1, pEcl2 and pEcl3 over-expression was less than the number of increased. There were 21, 154 and 23 genes whose expressions were decreased more than 1.7-fold by pEcl1, pEcl2 and pEcl3-over-expression, respectively. Among 21 genes decreased by pEcl1, 13 genes were also decreased by pEcl2 (62% identical), whereas 12 genes were also decreased by pEcl3 (52% identical). And among 23 genes decreased by pEcl3, 15 genes were also decreased by pEcl2 (65% identical). Then, nine genes were commonly decreased in each cell carrying pEcl1, pEcl2 and pEcl3.
Based on these results, we concluded that the cells carrying pEcl1, pEcl2 and pEcl3 show similar mRNA expression patterns, suggesting that ecl1+, ecl2+ and ecl3+ may have a common, if not identical, physiological role in S. pombe. The list of entire genes identified in the microarray analysis is provided in the Table S1 in Supporting information.
The genes involved in mating and stress response were induced by pEcl1, pEcl2 and pEcl3
Next, we analyzed the characters of genes whose expressions were affected by Ecl1 family genes. As a result, it was shown that the expressions of genes involved in mating and stress response were up-regulated in ecl1+-, ecl2+- and ecl3+-over-expressed cells (Table 1). Among them, mfm1+, mfm2+ and mfm3+ were the most induced genes, and the ratios were 35.70-, 7.21- and 8.44-fold by pEcl1, 34.80-, 12.83- and 32.26-fold by pEcl2 and 36.36-, 8.61- and 18.27-fold by pEcl3, respectively. mfm1+, mfm2+ and mfm3+ encode precursors of mating pheromone, M-factor. On the contrary, the most repressed gene was tlh1+, which encodes RecQ type DNA helicase, and the ratios were 0.01, 0.00 and 0.01 by pEcl1, pEcl2 and pEcl3, respectively.
Table 1. List of genes up- or down-regulated more than 1.7-fold in common
65 genes induced more than 1.7-fold in pEcl1, pEcl2 and pEcl3
Among the common 65 genes that were up-regulated by pEcl1, pEcl2 and pEcl3, 25 genes (38% of 65 genes) have been characterized experimentally to date. However, tlh1+ is the only gene that has been characterized experimentally among the common nine genes down-regulated by pEcl1, pEcl2 and pEcl3.
Over-expression of spk1+ and ste11+ slightly extends chronological lifespan
ecl1+, ecl2+ and ecl3+ induced some genes involved in mating and sexual development. Among them, we focused on two genes, spk1+ and ste11+. In S. cerevisiae, it was reported that KSS1 affects the chronological lifespan (Yan et al. 2007), that is, budding yeast lacking Kss1 protein shows shorter lifespan than wild type. In addition, they have also shown that chronological lifespan of budding yeast over-expressing human ERK2 (KSS1 homologue gene) is longer than that of wild type. KSS1 encodes mitogen-activated protein kinase involved in signal transduction pathways that control filamentous growth and pheromone response. In S. pombe, Spk1 is most similar to the Kss1 protein of S. cerevisiae. On the other hand, ste11+ encodes a key transcription factor for sexual development that activates the transcription of many genes involved in meiosis. Based on these facts, we chose two genes, spk1+ and ste11+, and characterized them as follows.
First, we conducted the Northern blot assay to verify the results of microarray analysis. The cells over-expressing pEcl1, pEcl2 and pEcl3 certainly increase the expression of ste11+ and spk1+ mRNA (Fig. 2A). If the lifespan extension by pEcl1, pEcl2 and pEcl3 is caused by the over-expression of ste11+ or spk1+, it is expected that over-expression of them extends chronological lifespan. So we measured the chronological lifespan of cells over-expressing ste11+ and spk1+. For this, each ste11+ and spk1+ gene was cloned on high copy number plasmid, and the resultant plasmid was named pSte11 and pSpk1, respectively (See Experimental procedures, Tables 2 and 4). The cells carrying pSte11 and pSpk1 extended their chronological lifespan slightly (Fig. 2B,C). However, deletion mutant of each ste11+ and spk1+ showed similar chronological lifespan to wild-type cell, and over-expression of Ecl1 in these mutants extended chronological lifespan (Fig. 2D,E). Over-expression of Ecl2 and Ecl3 showed the same effect as Ecl1 in the deletion mutants of ste11+ and spk1+ (unpublished observations).
Table 2. Genes that are cloned into multicopy plasmid, pLB-Dblet
*The values from microarray data that indicate the fold of expression of each gene when over-expressed Ecl1, Ecl2 or Ecl3, respectively.
Mating, sexual development, meiosis
Mating, sexual development, meiosis
Carbon, nitrogen, sulfur, phosphorous metabolism
Mating, sexual development, meiosis
Carbon, nitrogen, sulfur, phosphorous metabolism
Mating, sexual development, meiosis
Carbon, nitrogen, sulfur, phosphorous metabolism
Carbon, nitrogen, sulfur, phosphorous metabolism
Stress response/sexual development, meiosis
Based on these results, we concluded that although over-expression of ste11+ and spk1+ extends chronological lifespan slightly, the lifespan extension by ecl1+, ecl2+ and ecl3+ is independent on ste11+ and spk1+.
Based on the results from microarray analysis, in which some genes involved in the mating and sexual development were increased, we expected that ecl1+, ecl2+ and ecl3+ might affect mating efficiency and sporulation rate. This was the case as shown in Fig. 3. When ecl1+, ecl2+ and ecl3+ genes were over-expressed, mating efficiency was somewhat increased, although it is not as sufficient as that of ste11+ (Fig. 3A). Sporulation rate was markedly increased by the over-expression of ecl1+ and ecl2+, but the effect of ecl3+ was slight (Fig. 3B). Over-expression of spk1+ did not affect, if any, both the mating efficiency and sporulation rate. As expected, over-expression of ste11+ increased in both mating efficiency and sporulation rate. These phenotypes were consistent with the microarray data.
ctt1+ is required for increased resistance to H2O2 but not for extension of chronological lifespan by pEcl1, pEcl2 and pEcl3
ctt1+ encodes catalase, which serves to protect cells from the toxic effects of hydrogen peroxide. pEcl1, pEcl2 and pEcl3 made cells resistant to H2O2 (Ohtsuka et al. 2008, 2009), and the result of microarray indicated that ctt1+ mRNA is increased in ecl1+-, ecl2+- and ecl3+-over-expressed cells. Then, we examined the relationship between H2O2 resistance and ctt1+ expression in ecl1+-, ecl2+- and ecl3+-over-expressed cells. The involvement of ctt1+ in chronological lifespan was also analyzed.
First, we analyzed the induction of ctt1+ mRNA in ecl1+-, ecl2+- and ecl3+-over-expressed cells by Northern blot analysis. The induction by pEcl1, pEcl2 and pEcl3 was confirmed (Fig. 4A, lanes 1–4). Next, we examined whether ctt1+ affects chronological lifespan or not. For this, pCtt1 that carries ctt1+ on plasmid pLB-Dblet was introduced and the chronological lifespan was measured (Fig. 4B). Although cells carrying pCtt1 showed prominent H2O2-resistant phenotype (Fig. 4D), it did not affect the chronological lifespan (Fig. 4B). The level of ctt1+ mRNA expressed from cells carrying pCtt1 was more abundant than that from cells carrying pEcl1, pEcl2 or pEcl3 (Fig. 4A). The quantified ratio of ctt1+ mRNA level in cells carrying vector plasmid, pEcl1, pEcl2, pEcl3 and pCtt1 was estimated as 1 : 1.8 : 2.2 : 1.5 : 4.9. This ratio is reasonably in agreement with microarray data and plasmid copy number of vector plasmid, pLB-Dblet (six copies).
Next, we examined the effect of pEcl1, pEcl2 and pEcl3 on the chronological lifespan in the disruption mutant of ctt1+. As shown in Fig. 4C, ctt1Δ mutant did not affect chronological lifespan, and pEcl1 extended the chronological lifespan sufficiently with or without ctt1+. The effect of pEcl2 and pEcl3 was similar to that of pEcl1 (unpublished observations). These results suggest that ctt1+ does not involve in the chronological lifespan, and Ecl1 family proteins extend chronological lifespan independently on the ctt1+. Finally, we examined whether ctt1+ is involved in the H2O2-resistant phenotypes conferred by Ecl1 family proteins. As shown in Fig. 4D, the H2O2-resistant phenotypes conferred by Ecl1 family proteins were completely diminished in ctt1Δ mutant, although Ecl1 family proteins sufficiently extended the chronological lifespan in ctt1Δ mutant. These results indicated that ctt1+ is required for H2O2-resistant phenotypes but not for the elongation of chronological lifespan by pEcl1, pEcl2 and pEcl3.
Over-expressions of hsr1+, lsd90+, rsv2+ and hsp9+ extend chronological lifespan
Among the genes induced by pEcl1, pEcl2 and pEcl3, spk1+, ste11+ and ctt1+ did not seem to be the important factors for the chronological lifespan because their over-expressions did not extend the chronological lifespan sufficiently and deletion mutants of them did not affect the extension of chronological lifespan by pEcl1, pEcl2 and pEcl3. Therefore, to know how ecl1+, ecl2+ and ecl3+ extend the chronological lifespan, we attempted to identify the genes involved in the extension of chronological lifespan by pEcl1, pEcl2 and pEcl3. First, we expected that if there were the genes, over-expression of such genes might contribute to extend chronological lifespan. Therefore, we cloned several genes induced by pEcl1, pEcl2 and pEcl3 on the multicopy plasmid, pLB-Dblet. The genes are listed in Table 2. Using the plasmids, we measured the chronological lifespan of cells over-expressing such genes.
From these analyses, we found four genes, lsd90+, hsp9+, rsv2+ and hsr1+, whose over-expression resulted in the extension of chronological lifespan (Fig. 5). First, we conducted the Northern blot assay and confirmed that all of these genes were increased in cells over-expressing Ecl1 family proteins (Fig. 5A). lsd90+ was identified as a gene that is induced by stress and is involved in long-chain fatty acid biosynthetic process (Yokoyama et al. 2008). hsp9+ encodes heat shock protein that is induced by several stresses (Chen et al. 2003). rsv2+ encodes zinc finger transcription factor that induces stress-related genes during spore formation (Mata et al. 2007). It was also reported that rsv2+ has similar function to rsv1+ in stationary phase. rsv1+ is required for stationary-phase viability (Hao et al. 1997). This implicates that rsv1+ also appears to be involved in chronological lifespan, as well as rsv2+. Unfortunately, we could not detect the signal for rsv1+ transcripts in our microarray assay. hsr1+ also encodes zinc finger transcription factor, which has low similarity to S. cerevisiae Msn2. It is known that Msn2/Msn4 are involved in chronological lifespan in budding yeast (Wei et al. 2009). It has been reported that the expression of hsr1+ is regulated by transcription factors, Pap1, Prr1 and Atf1 (Chen et al. 2008).
Over-expression of the genes induced by pEcl1, pEcl2 and pEcl3 brought to identify several genes that are related to chronological lifespan of fission yeast. Among them, hsr1+ showed prominent extension of lifespan (Fig. 5B). This suggests that lifespan extension by pEcl1, pEcl2 and pEcl3 may be caused by increase in the expression of these genes.
Lifespan extension by ecl1+, ecl2+ and ecl3+ partially depends on hsr1+
As described earlier, among genes in Table 2, hsr1+ showed the most prominent positive effect on lifespan extension when over-expressed. Therefore, we expected that hsr1+ might be a major factor of lifespan extension by Ecl1 family proteins. If hsr1+ functions as a main mediator of lifespan extension, it would be supposed that the extension of lifespan by Ecl1 family proteins does not occur in hsr1Δ mutant. To investigate this, we measured the chronological lifespan of hsr1Δ mutant carrying pEcl1, pEcl2 and pEcl3. As shown in Fig. 6A, lifespan extension by pEcl1 was partially diminished in hsr1Δ mutant (Fig. 6A). As same as pEcl1, the lifespan extension by pEcl2 and pEcl3 was also diminished in hsr1Δ mutant (Fig. 6B).
Next, to help clarify how much hsr1+ mRNA is required to contribute to lifespan, we analyzed the relevance of hsr1+ mRNA level and lifespan extension ability of cells carrying pEcl1 and pHsr1. Based on the results presented in Fig. 5A, the ratio of hsr1+ mRNA level cells carrying vector plasmid, pEcl1, pEcl2, pEcl3 and pHsr1 was quantified as 1 : 2 : 3 : 2 : 13. However, lifespan extension by pEcl1 and pHsr1 was same (Fig. 6D). This meant that the amount of hsr1+ mRNA is not directly relevant to lifespan extension ability. Altogether, these results suggest that the functioning of Ecl1 family proteins on chronological lifespan extension partially depends on hsr1+.
Lifespan extension by ecl1+, ecl2+ and ecl3+ depends on prr1+
Because hsr1+ is partially required for the extension of lifespan by Ecl1 family proteins, we focused on this gene, especially how Ecl1 family proteins induce hsr1+ expression. Because Ecl1 protein localized mainly in nucleus (Ohtsuka et al. 2008) and Ecl1 protein has intrinsic transcription activation activity when assayed by yeast two hybrid assay (Azuma, Ohtsuka and Aiba, unpublished results), first we supposed that Ecl1 directly regulates hsr1+ expression as transcription factor. To confirm this possibility, we performed chromatin immunoprecipitation analysis. However, the binding of Ecl1 protein to the upstream DNA region of hsr1+ was not observed (unpublished observations). Next, we considered the possibility that Ecl1 family proteins regulate hsr1+ expression indirectly, that is, there may be other factor(s) that links Ecl1 family proteins to hsr1+ expression.
For this, we focused on the regulation of hsr1+. As described earlier, transcription of hsr1+ is regulated by transcription factors, Pap1, Prr1 and Atf1 (Chen et al. 2008). Therefore, we analyzed whether the expression of hsr1+ is increased by pEcl1 in pap1Δ, prr1Δ and atf1Δ mutants. Although pEcl1 increased the expressions of hsr1+ mRNA in pap1Δ and atf1Δ mutants (unpublished observations), the induction of hsr1+ mRNA was remarkably diminished in prr1Δ mutant (Fig. 7A). Furthermore, it was shown that pEcl1 did not induce the expressions of not only hsr1+- but also ste11+ mRNA in prr1Δ mutant. However, the induction of spk1+ and ctt1+ mRNA was occurred, if slightly diminished, by the Ecl1 over-expression. Over-expression of neither Ecl2 nor Ecl3 induced hsr1+ and ste11+ mRNA in prr1Δ mutant (Fig. 7B). This suggested that Ecl1 family proteins regulate hsr1+ expression through Prr1 transcription factor.
Next, we measured the chronological lifespan of prr1Δ mutant when pEcl1, pEcl2 and pEcl3 are introduced. Because it is expected that Prr1 has important role to induce several genes such as hsr1+ and ste11+ by the over-expression of Ecl1 family proteins, we forecasted that the lifespan extension by Ecl1 family proteins should be canceled in prr1Δ mutant. As expected, lifespan extension by the over-expression of Ecl1 family proteins was diminished in prr1Δ mutant (Fig. 7C). The prr1Δ mutant, itself, showed short-lived phenotype as compared to wild-type cells. Based on these results, we concluded that lifespan extensions by Ecl1 family proteins are dependent on prr1+. It should be mentioned that the over-expression of Ecl1 family proteins did not affect the transcription of prr1+ and transcription of Ecl1 family genes was not affected by prr1 deletion (unpublished observations).
The DNA microarray analysis using the strains that have long chronological lifespan had been carried out in S. cerevisiae (Wei et al. 2009). In that study, sch9Δ, tor1Δ and ras2Δ mutants were used as long-lived mutants, and the array data suggested that genes involved in stationary-phase survival, sporulation, meiosis and stress response were up-regulated in the long-lived mutants. The result is similar to our results, that is, over-expression of Ecl1 family proteins resulted in the increase in genes involved in meiosis (such as ste11+ and spk1+), stress response (ctt1+ and hsp9+) and stationary-phase survival (hsr1+ and lsd90+).
In consistent with the increased expression of genes related to mating and sporulation, over-expression of Ecl1 family proteins resulted in the increase in mating efficiency and sporulation rate. In our assay, pSte11 dramatically increased both mating efficiency and sporulation rate, whereas pSpk1 slightly increased them.
In this study, we found that increased resistance to H2O2 and extension of chronological lifespan by Ecl1 family are independent phenotypes. ctt1+ is necessary to the H2O2 resistance but not to the extension of chronological lifespan. Recently, it has been reported that reactive oxygen species that have been thought as a causing factor of aging could function as a factor of extending chronological lifespan in fission yeast (Zuin et al. 2010). The relevance between oxidative stress and aging has been discussed, as the free radical theory of aging, for a long time (Muller et al. 2007). However, clear conclusion has not yet been obtained. Needless to say, the results in our assay suggest that increased stress resistance to H2O2 through ctt1+ by Ecl1 family genes is independent on the extension of chronological lifespan; however, we do not exclude the possibility that antioxidants activated by Ecl family proteins, if any, affect the lifespan extension.
It was observed that there are similarities among Ecl1 family proteins not only in their amino acids sequences and over-expression phenotypes but also in the gene expression profiles affected by them. In addition, Ecl1 family proteins require prr1+ for both the induction of ste11+ and hsr1+ mRNA and the extension of chronological lifespan. Based on these results, it is speculated that ecl1+, ecl2+ and ecl3+ genes have similar or the same roles. However, the expression profiles of Ecl1 family genes seem to be different. The expression of ecl1+ mRNA during growth phase changes dramatically (Ohtsuka et al. 2008), whereas that of ecl3+ does not (Ohtsuka et al. 2009). The ecl1+ mRNA increased transiently when growth phase was shifted from logarithmic to stationary phase. In addition, the induction by thermal stress is observed only in ecl2+ (Ohtsuka et al. 2011). Furthermore, nitrogen starvation induces the transcription of ecl1+ only (Miwa et al. 2011). This suggests that Ecl1 family proteins might be evolved to respond to various stress signals to protect cell viability. The Ecl1 family proteins should be necessary for the proper adaptation to various situations. For this, it should be mentioned that ecl1Δecl2Δecl3Δ triple mutant showed short-lived phenotype (Ohtsuka et al. 2011).
Among the genes induced by pEcl1, pEcl2 and pEcl3, over-expression of ste11+, spk1+, lsd90+, hsp9+, rsv2+ and hsr1+ is shown to extend chronological lifespan of fission yeast; especially, over-expression of hsr1+ dramatically increased the chronological lifespan, and the effect of lifespan extension by Ecl1 family proteins partially diminished in hsr1Δ mutant. This suggests that hsr1+ is an important factor to extend chronological lifespan by Ecl1 family proteins. However, the lifespan extension by Ecl1 family proteins did not diminish completely in hsr1Δ mutant. Based on these results, it is expected that lifespan extension by Ecl1 family proteins may be dependent not only on hsr1+ but also on other factors that are induced by Ecl1 family proteins. In addition, based on the analysis of prr1Δ mutant, it was shown that lifespan extension and expression of some genes induced by Ecl family proteins remarkably depend on Prr1. Therefore, there may be other important factor(s) for lifespan extension among genes that are regulated by Prr1.
Then, we analyzed the genes whose expressions are decreased in prr1Δ mutant and increased in Ecl1-, Ecl2- and Ecl3-over-expressing cells by using array data (Chen et al. 2008) and identified the following genes as candidates: ste11, mfm2, ctt1, rds1, pho1, dak2, mae2, SPBC8E4.04, SPCP20C8.03, SPAC2H10.01 and SPAC212.09c. Among these genes, we have characterized ste11, ctt1, dak2 and SPBC8E4.04 in this study (Table 2). There might be crucial gene(s) for prolonged cell survival in the stationary phase in the remaining genes. Such analysis is waiting in the future.
Prr1 is a response regulator composing two-component signal transduction pathway. Prr1 is known to be involved in the oxidative stress and sexual differentiation, but the regulation of it was obscure (Ohmiya et al. 1999, 2000; Nakamichi et al. 2003). Deducing from the scenario of two-component signal transduction, the activity of Prr1 is regulated by the phosphorylation of Asp residue by histidine kinase. Fission yeast has three histidine kinases (Phk1/Mak2, Phk2/Mak3 and Phk3/Mak1) and one HPt protein (Spy1/Mpr1), so the analysis of functioning of Ecl1 family proteins in the mutants of two-component factors should contribute to the understanding of relevance between Ecl1 family proteins and Prr1 (Aoyama et al. 2000, 2001; Nguyen et al. 2000; Buck et al. 2001). These studies are underway in our laboratory.
If the ectopic expressions of Ecl1 family proteins would somehow enhance the Prr1 activity, Prr1 might bind to the promoter region of the target genes (e.g., hsr1) in cells over-expressing Ecl1. To test this possibility, ChIP analysis of Prr1 protein was carried out by employing strain (FY14737) that carries prr1-GFP-HA construct on the chromosome. However, unfortunately Prr1-GFP-HA protein was not immunoprecipitated by anti-HA antibody. So, we need to establish the more sufficient ChIP condition for the detection of Prr1 binding to target genes in the future.
In summary, we identified Prr1 as one major transcription factor that mediates the functions of Ecl1 family proteins on the extension of chronological lifespan in fission yeast. This finding gave us a hint for better understanding of the fundamental cellular event, i.e., chronological lifespan in fission yeast.
Strains and media
Schizosaccharomyces pombe strains used in this study are listed in Table 3. The strains were grown in SD medium [0.67% yeast nitrogen base without amino acids (Difco), 2% glucose] supplemented with necessary growth requirements in standard amounts. For the measurements of mating efficiency and sporulation rate, EMM medium (Moreno et al. 1991) was used.
Table 3. List of Schizosaccharomyces pombe strains used in this work
h−leu1 ade6-M216 spk1::kanr
h90leu1 ade6-M210 ura4-D18 ste11::ura4+
h−leu1 ade6-M216 ctt1::kanr
h−leu1 ade6-M210 hsr1::HygB
h−leu1 ade6-M216 ura4-D18 gad7(atf1)::ura4+
h−leu1 his2 ura4-D18 pap1::ura4+
h−leu1 ade6-M216 ura4-D18 prr1::ura4+
Analysis of viability
To determine the cell viability, cells were grown in SD liquid medium, sampled in each growth phase and then plated on YE agar plates after suitable dilution. After 4 days of incubation at 30 °C, the number of colonies derived from 1 mL of culture was counted and then divided the number by the cell turbidity at the sampling time. Cell growth was monitored by turbidity using a Bactomonitor (BACT-550) equipped with a 600-nm filter (Nissho Denki Co., Tokyo, Japan). Data are represented as mean with standard deviation.
To clone the gene, DNA fragment encompassing the gene with its upstream and downstream regions was amplified from S. pombe chromosome by using two primers. The amplified DNA fragment was digested with appropriate restriction enzymes and then cloned into pLB-Dblet. The pLB-Dblet is a derivative of pDblet (Brun et al. 1995). pLB-Dblet was made by replacing ura4+ marker of pDblet with LUE2 and by inserting BglII linker into SmaI site at multiple cloning site of pDblet. Primers and restriction enzymes used in this work are listed in Table 4 with precise cloned regions. The plasmid, pEcl1 (pSht3), that was used for the over-expression of ecl1+ was described in the study by Ohtsuka et al. (2008). Plasmids, pEcl2 and pEcl3, that were used for the over-expressions of ecl2+ and ecl3+, respectively, were described in the study by Ohtsuka et al. (2009).
Table 4. List of primers that are used to clone following genes
Restriction enzyme sites used are underlined. Cloned regions of each gene are indicated as upstream (−) from initiation codon and downstream (+) from termination codon.
−723 from start
+971 from stop
−972 from start
+964 from stop
−895 from start
+871 from stop
−770 from start
+587 from stop
−900 from start
+955 from stop
−969 from start
+1007 from stop
−977 from start
+655 from stop
−940 from start
+955 from stop
−950 from start
+746 from stop
−990 from start
+754 from stop
−960 from start
+1922 from stop
−821 from start
+903 from stop
−967 from start
+830 from stop
−994 from start
+835 from stop
−831 from start
+954 from stop
−860 from start
+883 from stop
−990 from start
+642 from stop
−847 from start
+880 from stop
Constructions of spk1Δ, ctt1Δ and hsr1Δ mutants
For the spk1+ disruption, the ORF region of the Spk1 protein was replaced with a kanR cassette by the methods described by Krawchuk & Wahls (1999). Both the upstream and downstream regions of spk1+ were PCR-amplified by using F1 and F2 primers and R1 and R2 primers, respectively, and both fragments were purified. After mixing both DNA fragments with pFA6a-kanMX6, a PCR was performed with the F1 and R1 primers. JY333 was transformed with the amplified DNA fragment, and stable G418-resistant transformants were selected. Then, the spk1::kanR construct on the chromosome was confirmed by PCR using appropriate primers. The primers used were as follows: F1, GACTAGTCCAACCACACTAC; F2, TTAATTAACCCGGGGATCCGATAGTAGGCGTGGAAGTAGC; R1, GGGTTCACGAAGAACAGCTT; and R2, GTTTAAACGAGCTCGAATTCCTCCCAACCTCGTCGATTTA.
JY333Δctt1 mutant was constructed as described above. The primers used were as follows: F1, CGGTATGAGCGCATCGTAAA; F2, TTAATTAACCCGGGGATCCGTGGTGTAAACAGGGACGGTA; R1, TAAACCGGCCAAACCTCAAC; and R2, GTTTAAACGAGCTCGAATTCAACAAGGGCGAACCCCATAT.
HM3802Δhsr1 mutant was made from original strain obtained from J. Bähler.
Northern hybridization analysis
Northern hybridization analysis was carried out as described previously (Ohtsuka et al. 2008). Cells were grown in SD medium until OD600 = 1.0. Then, cells were collected and total RNA fractions were prepared. After denaturation with formamide–formaldehyde, the RNA (10 μg) was analyzed on a 1.2% agarose gel containing formaldehyde. Hybridization was carried out with a 32P-labeled probe, which specifically encompasses the genes listed in Table 2 at 65 °C for 2 h in Rapid-hyb buffer, as recommended by the supplier (GE Healthcare).
DNA microarray construction
DNA microarrays were prepared by Life Science Group, Hitachi, Ltd. (Kawagoe, Japan). Sequence-optimized 70-base DNA probes representing each 4906 open reading frames in S. pombe chromosomes (GeneDB, http://old.genedb.org//genedb/pombe/) were chemically synthesized for use on glass slide arrays designed to monitor the gene expression. Each of these probes was designed to minimize the cross-hybridization of homologous regions and to allow a Tm normalization of 70–75 °C. After polyacrylamide gel electrophoresis purification, the probe set was spotted onto glass slide partitioned into 32 blocks and was fixed by covalent binding. Serially diluted act1+ probe, pUC18 DNA and M13mp18 DNA were also spotted onto each block as for normalization of Cy3 and Cy5 signal intensity and background subtraction.
RNA labeling and hybridization
For each experiment, 50 μg of total RNA was converted to either Cy3-labeled (32 °C) and Cy5-labeled (37 °C) cDNA using reverse transcriptase reaction kit (Takara Bio Inc., Otsu, Japan) with oligo-dT as primer. The labeled cDNA products were purified using gel filtration column. Labeled Cy3 and Cy5 targets were then combined, dried and resuspended in 35 μL of hybridization buffer (3.4 × SSC, 0.3%-SDS) containing 10 μg of baker’s yeast tRNA (Sigma Chemical Co., St. Louis, USA). Hybridization on the microarray with coverslip (22 mm × 50 mm) was performed at 62 °C for 16 h. Following hybridization, glass slides were washed twice in 1 × SSC, 0.1%-SDS for 5 min, once in 0.1 × SSC, 0.1%-SDS and finally twice in 0.1 × SSC for 5 min; then slides were immediately dried by slide centrifuge (Wakenyaku, Kyoto, Japan). Hybridized microarrays were scanned for Cy3- and Cy5-labeled targets, using GenePix 4000B scanner (Axon Instruments, Union City, USA) with a resolution of 10 μm. Signal quantification for each probes on the microarray was performed using GenePix acquisition software (version 4.0; Axon Instruments). Normalization of Cy3 and Cy5 signal intensity was performed using act1+ gene as the expression level is same in both degrees.
We thank M. Kawamukai (Shimane University, Japan), J. Bähler (University College London, UK), M. Yamamoto (The University of Tokyo, Japan) and The National BioResource Project/Yeast Genetic Resource Center and for strains. This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan [to HA]. The part of this work was also supported by A Research for Promoting Technological Seeds from JST, Nagase Science and Technology Foundation and The Asahi Glass Foundation [to HA] and by a grant from the Japan Society for the Promotion of Science for Young Scientists [to HO].