Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/unpaywall/the-atf1-transcription-factor-is-a-target-for-the-sty1-stress-X0g6uhTa4J
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- Unpaywall
- ISSN
- 0890-9369
- DOI
- 10.1101/gad.10.18.2289
- Publisher site
- See Article on Publisher Site
Abstract
Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press The Atfl transcription factor is a target for the Styl stress-activated MAP kmase pathway in fission yeast Marc G. Wilkinson, 1'4 Michael Samuels, 2'4 Tadayuki Takeda, 2'5 W. Mark Toone, 2 Jia-Ching Shieh, 1 Takashi Toda, 3 Jonathan B.A. Millar, 1'6 and Nic Jones 2'6 ~Division of Yeast Genetics, National Institute for Medical Research, London NW7 1AA UK; 2Laboratory of Gene Regulation and 3Laboratory of Cell Regulation, Imperial Cancer Research Fund, London WC2A 3PX UK The atfl ÷ gene of Schizosaccharomyces pombe encodes a bZIP transcription factor with strong homology to the mammalian factor ATF-2. ATF-2 is regulated through phosphorylation in mammalian cells by the stress-activated mitogen-activated protein (MAP) kinases SAPK/JNK and p38. We show here that the fission yeast Atfl factor is also regulated by a stress-activated kinase, Styl. The Styl kinase is stimulated by a variety of different stress conditions including osmotic and oxidative stress and heat shock. Deletion of the atfl + gene results in many, but not all, of the phenotypes associated with loss of Styl, including sensitivity to environmental stress and inability to undergo sexual conjugation. Furthermore, we identify a number of target genes that are induced rapidly in a manner dependent upon both the Styl kinase and the Atfl transcription factor. These genes include gpdl +, which is important for the response of cells to osmotic stress, the catalase gene ~ important for cells to combat oxidative stress, and pyp2 +, which encodes a tyrosine-specific MAP kinase phosphatase. Induction of Pyp2 by Atfl is direct in that it does not require de novo protein synthesis and results in a negative feedback loop that serves to control signaling through the Styl/Wisl pathway. We show that Atfl associates stably and is phosphorylated by the Styl kinase in vitro. Taken together, these results indicate that the interaction between Atfl and Styl is direct. These findings highlight a remarkable level of conservation in transcriptional control by stress-activated MAP kinase pathways between fission yeast and mammalian cells. [Key Words: S. pombe; stess response; MAP kinase; Atfl transcription factor] Received June 26, 1996; revised version accepted August 2, 1996. The mitogen-activated protein kinase (MAPK) signaling merer 1994; Herskowitz 1995; Wasiewicz and Cooper 1995). In addition, it is clear that multiple independently pathways are critical for the response of cells to changes in their external environment (for review, see Marshall acting cascades can function in any one cell. The most 1994; Herskowitz 1995; Wasiewicz and Cooper 1995; striking example of such multiplicity can be found in the Treisman 1996). They serve to transduce signals gener- budding yeast Saccharomyces cerevisiae, where at least ated at the cell surface to the nucleus, where changes in five distinct pathways have been identified that serve to the gene expression pattern of the cell occur. The re- control very different physiological processes such as sporulation, mating, cell wall biosynthesis and response sponses that follow can be varied, including alterations to osmotic stress (Ammerer 1994; Herskowitz 1995; in the proliferation or differentiation state of the cell or Levin and Errede 1995). In mammalian cells three dis- changes in protective mechanisms that serve to combat different stress conditions. At the heart of such signaling tinct pathways have been identified, although there are pathways lies the MAPK module, a kinase cascade that clear indications that more pathways do exist. The best- culminates in the activation of a MAPK through phos- characterized pathway activates the extra-cellular-sig- phorylation by a MAPK kinase (MAPKK), which itself is nal-regulated kinases ERK1 and ERK2 in response to a phosphorylated and activated by a MAPKK kinase variety of growth factors and mitogens and has been (MAPKKK). Such MAPK cascades are evolutionarily shown to be involved in the control of cell proliferation conserved, having been identified and characterized in and differentiation (Marshall 1994). The physiological function of the other two pathways that activate the numerous organisms ranging from yeast to mammals (Am- MAPKs JNK/SAPK and p38 kinase is less well under- stood. Both pathways are activated by a variety of stress 4These authors contributed equally to this paper. conditions (e.g., osmotic stress and UV radiation) as well Spresent address: Institute of Medical Science, University of Tokyo, To- as inflammatory cytokines (e.g., interleukin-1 and tumor kyo, 108 Japan. 6Corresponding authors. necrosis factor) and consequently have been implicated GENES & DEVELOPMENT 10:2289-2301 © 1996 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/96 $5.00 2289 Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Wilkinson et al. in playing a role in the response of cells to stress and in Dam et al. 1995) and p38 (Raingeaud et al. 1995). In all T-cell activation and inflammation (Davies 1994; Wask- three cases phosphorylation results in an increase in the iewicz and Cooper 1995). The p38 kinase is related activation potential of the factor, although the mecha- highly to the HOG1 kinase of S. cerevisiae (57% iden- nisms that underlie these increases are not understood. tity; Han et al. 1994; Freshney et al. 1994; Lee et al. A central question concerning the conservation of the 1994; Rouse et al. 1994), which functions in the response MAPK signaling pathways is whether it is restricted to of cells to high extracellular osmolarity; thus, deletion of the MAPK module or whether it extends downstream of HOG1 results in cells sensitive to osmotic stress (Brew- the pathway to common transcription factors and target ster et al. 1993). The conservation that is evident from genes that are regulated by the pathways. Structural ho- the structural similarity between these stress-activated mologs of ATF-2 and cJun have been identified in S. MAPKs is strikingly clear from the finding that HOG1 pombe. The atfl + gene encodes a factor containing a can be replaced functionally by p38 (Han et al. 1994). bZip domain at its carboxyl terminus that is related A Hogl homolog from Schizosaccharomyces pombe, closely to the bZip domain of ATF-2 (Takeda et al. 1995). Styl, has been described recently (Millar et al. 1995), and Furthermore, it has a DNA-binding specificity that is is also known as Spcl (Shiozaki and Russell 1995) and similar to ATF-2 (TG/TACGTC/AC/A) and can acti- Phhl (Kato et al. 1996). It is phosphorylated and acti- vate transcription from a reporter containing ATF sites vated by the Wisl MAPKK. This pathway is involved within its promoter. Although disruption of the atfl + intimately in the stress response of the cell; it is acti- gene is not lethal, it does result in rapid cell death in vated by a variety of stress conditions including osmotic stationary phase as well as failure in the initation of the stress, and deletion of either wisl or sty l results in in- sexual differentiation pathway. Here we show that Atfl creased sensitivity to heat and high osmotic conditions is downstream of the Wisl/Styl pathway and that the (Warbrick and Fantes 1991; Millar et al. 1995; Shiozaki expression of multiple targets shown previously to be and Russell 1995; Degols et al. 1996; Kato et al. 1996). dependent upon Styl and/or Wisl is also dependent However, the pathway has additional physiological upon Atfl. One such target is pyp2 +, which encodes a roles. Both wisl or styI deletion mutants are profoundly tyrosine-specific phosphatase that dephosphorylates and sterile, indicating that signaling through this pathway is hence inactivates Styl. Therefore, through the Atfl fac- required for one or more steps in the sexual differentia- tor, a negative feedback loop operates to regulate the tion pathway. In addition, the Sty 1/Wisl pathway plays extent of signaling through this pathway. Furthermore, a role in the control of mitotic initiation, because cells we show that the link between Styl and Atfl is direct; deleted for wisl or styl are considerably longer at cell Atfl binds specifically to, and is phosphorylated by, the division than wild-type cells, and mutations in the path- Styl kinase. These results demonstrate remarkable way display strong genetic interactions with mutations structural and functional conservation in the stress re- in other genes such as cdc25 + that result in mitotic de- sponse between mammalian cells and fission yeast. fects (Millar et al. 1992, 1995; Shiozaki and Russell 1995). However, how the Styl kinase mediates these dif- ferent activities is at present unknown and will depend Results absolutely upon the identification of relevant Styl phos- Disruption of either atfl + or wisl + results in phorylation targets. sensitivity to osmotic stress and suppression of Because the activation of MAPK pathways results in uncontrolled meiosis driven by mutation in patl + changes in the transcriptional pattern of the cell, a key step in understanding their function is the characteriza- We had shown previously that disruption of the atfl + tion of transcription factors whose activity is modulated gene resulted in sterility and loss of viability in station- by MAPK phosphorylation. In mammalian cells, a num- ary phase (Takeda et al. 1995). These properties are ber of factors now have been identified (for review, see shared by mutants disrupted in either styl + (Millar et al. Treisman 1996). Elk-l, a member of the TCF family of 1995; Shiozaki and Russell 1995; Kato et al. 1996) or ETS domain proteins that synergizes with the serum re- wisl + (Warbrick and Fantes 1991), indicating the possi- sponse factor (SRF) to mediate activation of certain im- bility that the Atfl factor might be a downstream target mediate early genes such as the c-fos gene, was demon- of the Wisl/Styl signaling pathway. Activation of this strated to be regulated through phosphorylation by pathway can be triggered by a variety of conditions, in- ERK1/2 (Gille et al. 1992; Marais et al. 1993). More re- cluding an increase in osmolarity, heat shock, UV radi- cently, however, it has been shown to be a target of JNK/ ation, oxidative stress, and the addition of DNA damag- ing agents such as bleomycin (Millar et al. 1995; SAPK and p38 (Gille et al. 1995; Whitmarsh et al. 1995; Zinck et al. 1995), indicating that all three mammalian Shiozaki and Russell 1995; Degols et al. 1996; Kato et al. 1996). This suggests that the pathway is vital for the pathways can converge at a common factor and presum- ably, therefore, common promoters. Other targets show protective response of cells to such stress conditions, a greater degree of specificity. The cJun factor is regu- supported by the finding that disruption of styl + or lated by JNK/SAPK (Hibi et al. 1993; Derijard et al. 1994; wisl + results in hypersensitivity to high extracellular Kyriakis et al. 1994) but not by ERK1/2 or p38, whereas osmotic conditions. If Atfl is an essential downstream ATF-2 is phosphorylated and regulated by both JNK/ target of the pathway, then its absence should also result SAPK (Gupta et al. 1995; Livingstone et al. 1995; van in stress hypersensitivity. As shown in Figure 1A, this 2290 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Regulation of At[1 by stress-activated MAPK et al. 1995); suppression on minimal medium was ob- YPD+O.5M NaCI tained in the presence of moderate to high levels but not YPD on low levels of cAMP and caffeine. Similar observations were made with pat1-114 zlstyl cells (data not shown). Mutations in either sty1 or wisl result in a delay in the <:%: timing of mitotic initiation (Warbrick and Fantes 1991; Millar et al. 1995; Shiozaki and Russell 1995). This phe- notype is not shared by zlatfl cells. As the results in Z~atf l WT Z~wis l Z~atf l WT zlwis 1 Table 1 demonstrate, unlike zlstyl or zlwisl cells, Aatfl cells display no defect in the timing of mitotic initiation, their cell size at division being close to that seen with wild-type cells. Furthermore, mutants in either styl or wisl are synthetically lethal with cdc25-22 cells that express a temperature-sensitive mutant of the Cdc25 mi- totic inducer. No such synthetic interaction was ob- served between 3atfl and cdc25-22 (Table 1). Transcription from an A TF-site-dependent reporter is regulated by osmotic stress \~ . :j We have described previously a reporter gene containing three tandem copies of the E4 ATF-binding site upstream of the basic promoter of the fission yeast p25 gene (Takeda et al. 1995). Expression of this reporter was 250C 330C shown to be dependent upon the Atfl factor. Since the results described above indicated that Atfl could be a Figure 1. Overlapping phenotypes caused by loss of Atfl or components of the Styl/Wisl signaling pathway. (A) Atfl is downstream target of the Styl/Wisl pathway, we asked required for resistance to osmotic stress. Wild-type (HM123), whether expression of the reporter could be stimulated 3wisl (JM544) and zlatfl (NT146) cells were grown on YPD by an increase in osmolarity. As shown in Figure 2A, a plates and then streaked onto either YPD medium alone or YPD time-dependent 2.5- to 3-fold increase was observed fol- containing 0.5 M NaC1 and incubated at 30°C for 3 days. (B) lowing treatment with 1.2 M sorbitol; expression from a Deletion of wisl or styl rescues the patl-114 phenotype, patl- control reporter that lacked the ATF sites was unaf- 114 (JM836) or Awisl patl-ll4 (JM1266) or Astyl patl-ll4 awisl cells and fected. This increase was abolished in (JM1267) cells were plated and streaked on rich medium at ei- enhanced in Apypl cells (Fig. 2B). Pypl, like Pyp2, is a ther 25°C (left) or 33°C (right). tyrosine-specific phosphatase that can inhibit the Styl MAPK pathway (Millar et al. 1995; Shiozaki and Russell 1995; Kato et al. 1996). Tyrosine phosphorylation of Sty1 kinase is increased in Apypl cells. The results show that was found to be the case for osmotic stress. Unlike wild- the expression of the reporter can be controlled by sig- type cells that proliferate normally on medium contain- naling through the Styl pathway. ing 0.5 M NaC1, both aatfl and zlwisl cells fail to do so. An identical result was observed on medium containing 1.2 M sorbitol (data not shown). Conversely, it might be expected that phenotypes as- Table 1. Aatfl cells do not have a cell size at division defect cribed to loss of Atfl would also be seen in zlwisl or astyl cells. We had shown previously that the 3pat1 Temperature Cell size at phenotype could be suppressed by atfl + inactivation Strain (°C) division (l~m) (Takeda et al. 1995). The pat1 + gene encodes a kinase wt 30 14.2 + 0.3 that negatively regulates entry into meiosis. Strains car- cdc25-22 26 19.3 + 0.3 rying a temperature-sensitive allele of pat1 + can initiate wisl::ura4 30 23.4 + 1.3 meiosis at the restrictive temperature in haploid cells styl::ura4 30 23.2 + 1.6 (Iino and Yamamoto 1985a); as a result of this meiotic atfl::ura4 30 13.1 + 1.6 induction, cells fail to grow at 33°C (Iino and Yamamoto cdc25-22 wisl ::ura4 26 cdc- 1985b). However, as shown in Figure 1B, both pat1-114 cdc25-22 styl ::ura4 26 cdc- cdc25-22 atfl::ura4 26 18.1 + 1.6 Awisl and patl-ll4 Astyl cells grew well at this tem- perature on rich medium, indicating that mutation of Loss of atfl does not delay the timing of mitosis. wisl + and styl +, like mutation of atfl +, suppresses the aCell size measurements of septated cells grown in liquid syn- phenotype that results from Patl inactivation. We also thetic minimal medium (EMM). {cdc) Cell division cycle phe- showed previously that suppression of the patl-114 phe- notype. Measurements were the mean of 30 individual determi- notype by Aatfl was influenced by cAMP levels (Takeda nations [ +- S.D.}. GENES & DEVELOPMENT 2291 Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Wilkinson et al. 13-gal units 13-gal units 500 , 4OO Figure 2. Atfl activity is regu- • + sorbitol • pATFLacZ lated by environmental stress. 17":1 . sorbitol [] pLacZ {A) Wild-type ceils containing ei- 400- ther a promoter-less lacZ re- porter {pLacZ; stippled bars) or a lacZ reporter fused to an Atfl- dependent promoter (pATFLacZ; 300- solid bars) (Takeda et al. 1995) were assayed for 13-galactosidase activity following growth in min- imal medium containing 1.2 M sorbitol for the indicated times. {B) Wild-type {HM123), Awisl (JM544), or apypl (NT132) cells containing a lacZ reporter fused to an ATF-dependent promoter (Takeda et al. 1995) were grown for 3 hr in minimal medium {stippled bars) or minimal me- dium containing 1.2 M sorbitol {solid bars), and subsequently Oh lh 2h 3h 4h 5 h WT A wis I z~pyp 1 assayed for 13-galactosidase ac- Time after osmotic stress tivity. Atfl is required for the induction of pyp2 and cultures of wild-type cells or zaatfl cells were treated mediates feedback inhibition of the Styl MAPK in with 0.9 M KC1 and the expression of pyp2 was observed response to osmotic stress over time. As shown in Figure 4A, the large stimulation of pyp2 mRNA expression was completely lost in Aatfl Previous results demonstrated that transcription of the cells. Reintroduction of the atfl + gene expressed from pyp2 + gene encoding a tyrosine-specific phosphatase the thiamine-repressible nmtl promoter efficiently re- was induced by increased extracellular osmolarity in a versed this loss. Similar results were observed when Styl- and Wisl-dependent manner (Millar et al. 1995). wild-type or Aatfl cells were shocked with 1.2 M sorbi- Furthermore, this induction results in the initiation of a tol, 1 mM H202, or elevated temperature (data not negative feedback loop; the newly synthesized Pyp2 de- shown). All of these conditions stimulate the Styl path- phosphorylates, and hence inactivates, the Styl kinase, way (Degols et al. 1996; Kato et al. 1996; M.G. Wilkin- resulting in transient tyrosine phosphorylation of Styl son and J.B.A. Millar, unpubl.). following its activation. In support for such a regulatory Because expression of pyp2 is dependent upon Atfl pathway, we show in Figure 3A that the loss of Sty 1- and because the Pyp2 phosphatase has been shown to be associated phosphotyrosine following prolonged expo- responsible for feedback inhibition of the Styl kinase, sure of cells to osmotic stress is dependent on protein then it should follow that in ziatfl cells activation of synthesis. When cells are pretreated with cycloheximide Sty l is prolonged upon exposure to osmotic stress. To and then challenged with 0.9 M KC1, tyrosine phospho- test this prediction, either wild-type cells or Aatfl cells rylation of Styl remains high for up to 3 hr. Importantly, bearing an epitope-tagged version of Styl were chal- however, such treatment did not prevent the induction lenged with 0.9 M KC1 and the phosphotyrosine content of pyp2 mRNA; indeed, elevated levels of pyp2 + tran- on the Styl protein was assessed. As shown in Figure 4B, scription persisted over the time course of the experi- unlike wild-type cells where tyrosine phosphorylation is ment, whereas in wild-type cells, induction was tran- maximal between 15 and 30 min after exposure but then sient (Fig. 3B). These results strongly suggest that the declines, phosphorylation on Styl remains high in Aatfl induction of pyp2 + mRNA by activation of the Styl cells. These results show that Atfl mediates feedback MAP kinase occurs via a pre-existing transcription factor inhibition of the Styl kinase in response to osmotic acting on the pyp2 + promoter. stress. We considered Atfl to be a strong candidate for such a factor. Not only was there an overlap in phenotypes that derive from mutations in atfl + or in the Styl/Wisl Atfl is required for the reduction of gpdl and catalase pathway, but the activity of Atfl as measured by the mRNAs in response to environmental stress reporter assay appeared to be regulated by the Styl ki- nase. Therefore, we asked whether the induction of The expression of gpdl +, a gene that encodes glycerol- pyp2 + transcription by stress required Atfl. Log-phase 3-phosphate dehydrogenase required for glycerol biosyn- 2292 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Regulation of Attl by stress-activated MAPK A induction of catalase mRNA following the addition of 1.2 M sorbitol was lost in Aatfl cells. Therefore, with Time (hours) respect to the expression of pyp2 +, gpdl +, and catalase, 015'30'1 1.522.53 015'30'1 1.522.53 there is a similar dependence on Atfl and the stress- activated pathway. Control As shown previously (Takeda et al. 1995) and in Figure + Cyclohex. 6A, using the E4 ATF site as a probe, Atfl-dependent complexes can be detected readily by the electrophoretic (x-PTyr ~-HA mobility shift assay {EMSA) following incubation of the probe with whole-cell extracts from wild-type cells. The Time (hours) complex is not seen with extracts from Aatfl cells and is 0 30' I 1.5 2 2.5 3 competed efficiently with unlabeled probe but consider- ably less well with a probe that has point mutations Control + Cyclohex. Aatf l WT z~atf l +pREP81 -atfl Figure 3. Feedback inhibition of Styl kinase requires protein o 30 60 12o o 30 60 12o o 30 6o12o synthesis. {A) Dephosphorylation of Sty1 in response to envi- ronmental stress requires de novo protein synthesis. Log phase cultures of wild-type cells bearing an epitope-tagged version of the Styl kinase [IpREP41-styllHA-6His)} were grown in mini- mal medium lacking leucine and thiamine either in the pres- ence {+ Cyclohex.) or absence of {Control) of 0.1 mg/ml cyclo- heximide for 2 hr and then incubated for the times indicated in the same medium containing 0.9 M KC1. Styl was isolated by Ni 2 +-NTA affinity precipitation and probed by Western blot for the presence of phosphotyrosine (a-PTyr) or Hemagglutinin epitope tag (a-HA). (B) Inhibition of protein synthesis inhibits disappearance but not induction of Pyp2 mRNA. Log phase cul- Time (hours) tures of wild-type cells were grown in minimal medium either 015'30'1 1.522.53 015'30'1 1.522.53 in the presence (+ Cyclohex.)or absence of (Control} of 0.1 mg/ml cycloheximide for 2 hr and then incubated for the times indicated in the same medium containing 0.9 M KC1. The level WT of pyp2 mRNA was assessed by Northern blot analysis using a pyp2 specific probe. Ethidium staining of ribosomal RNA was used to illustrate approximate equal loading in each lane. Aatf l o~-PTyr o~-HA Figure 4. The Atfl transcription factor is required for induc- thesis, is induced by osmotic stress in a Sty1- (M.G. tion of pyp2 feedback inhibition of the Styl kinase in response Wilkinson and J.B.A. Millar, unpubl.) and Wisl-depen- to environmental stress. {A) Induction of the Pyp2 MAP kinase dent (Aiba et al. 1995) manner. We predicted, therefore, phosphatase requires Atfl. Log phase cultures of either wild- that expression of gpdl + would also require Atfl. That type cells (PR109) bearing the empty vector pREP81 (WT), Llatfl this is indeed the case is demonstrated in Figure 5A. cells {NT146) bearing pREP81 {Aatfl) or Aatfl cells bearing Treatment of wild-type cells by the addition of 0.9 M KC1 pREP81-atfl (Aatfl + pREP81-atfl) were grown in minimal me- resulted in a modest increase in gpdl mRNA levels; the dium lacking leucine and thiamine and then incubated for the increase was -3- to 5-fold, which corresponds closely to times indicated (in minutes) in the same medium containing 0.9 M KC1. The level of pyp2 mRNA was assessed by Northern blot previous observations (Aiba et al. 1995 I. It is striking that analysis using a pyp2 specific probe. Ethidium staining of ribo- although the stimulation by stress is small in compari- somal RNA was used to illustrate approximate equal loading in son to pyp2 +, deletion of wisl + or atfl + results in an each lane. (B) Feedback inhibition of Sty1 phosphorylation re- almost complete loss in expression {Fig. 5A). This indi- quires Atfl. Wild-type cells or cells deleted for Atfl (Aatfl) were cated that the uninduced levels of gpdl + were also de- transformed with an epitope-tagged version of Styl and grown pendent upon the Sty1 pathway and the Atfl factor. in minimal medium lacking leucine and thiamine at 30°C. They The catalase gene in fission yeast is induced by os- were then incubated in the same medium containing 0.9 M KC1 motic and oxidative stress and by UV radiation {Naka- for the times indicated. The Styl protein was extracted using gawa et al. 1995}. Since the same signals activate the Ni2+-NTA agarose beads from the cellular lysates and probed Styl pathway, we asked whether this factor was in- by Westem blot for the presence of either phosphotyrosine {~- volved in catalase expression. As shown in Figure 5B, the Ptyr) or hemagglutinin tag (a-HA). GENES & DEVELOPMENT 2293 Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Wilkinson et ai. A link between the pathway and Atfl was direct. Log- WT Awls 7 Aatf l phase cultures containing an HA-epitope and His 6- tagged Styl expressed ectopically from a thiamine-re- 001.5 ] 1~52 I 00,.5] 1,52, 00.5j 111:52 pressible nmtl promoter were subjected to osmotic gpdl stress, and the Styl protein was immunoprecipitated at various times thereafter. Kinase assays were performed with the immunocomplexes using GST-Atfl as a sub- strate. Atfl was found to be an excellent Styl substrate (Fig. 7A). Efficient phosphorylation was obtained within 15 min of stress treatment; at later time points there was a gradual decrease, in keeping with the transient nature WT Aatfl of Styl induction. The presence of multiple phosphory- 0 1 2 3 0 1 2 3 lated species was due to breakdown of the GST-Atfl I I I I I I I I fusion protein (data not shown). There was no phospho- rylation with immunocomplexes isolated from cells ,~-~ m ~Catalase treated in the presence of thiamine; under these condi- tions no Sty 1 protein was synthesized as judged by West- ern analysis (Fig. 7B). Recently, a second ATF-like protein in fission yeast has been described to be encoded by the pcrl + gene (Wa- Figure 5. The Atfl factor is required for induction of gpdl and catalase mRNAs in response to osmotic stress. (A) Log phase tanabe and Yamamoto 1996). Furthermore, it appears cultures of either wild-type cells (PR109) or Aatfl cells (NT146) that Atfl and Pcrl can form heterodimers both in vitro were grown in rich medium and then incubated for the times and in vivo. Therefore, we tested whether Pcrl was a indicated (in hours) in the same medium containing 0.9 M KC1. substrate for Styl using the immunocomplex kinase as- The level of gpdl mRNA was assessed by Northern blot anal- say. As shown in Figure 7C, no inducible phosphoryla- ysis using a gpdl specific probe. {B) The level of catalase mRNA tion was seen. In mammalian cells, the stress-activated was assessed by Northern blot analysis using a catalase specific kinases phosphorylate and activate members of both the probe as described above, except that osmotic stress was in- duced with 1.2 M sorbitol. Extract: <~ WT within the ATF binding sequence (Takeda et al. 1995). Furthermore, there was no significant difference in com- Competitor: - - - wt ml plex formation with extracts from sorbitol-treated cells, 1.2M Sorb: - - + - suggesting that activation of Atfl does not entail an in- wr Extract: crease in its DNA binding activity (Fig. 6A). Examina- 1.2M Sorb: 4- tion of the sequences upstream of the pyp2 +, gpd 1 +, and . l[, catalase transcriptional start sites revealed the presence of multiple potential Atfl binding sites. In the case of the gpdl + gene, three potential sites are evident at positions - 420 to - 413 (TTACGTAA), - 371 to - 364 (TGACGTTT), and - 324 to - 317 (TTACGTCA) rela- tive to the translation initiation codon. To test whether this upstream region of gpdl + could bind to Atfl, EMSA was carried out using a probe extending from -459 to - 248. As shown in Figure 6B, a slow migrating complex was detected with wild-type extracts that was missing with extracts from Aatfl cells. Again, no difference in the degree of complex formation was seen with extracts from sorbitol-treated cells. These results indicate that Figure 6. Atfl binding activity does not change in response to Atfl can bind to sequences upstream of the gpdl + tran- osmotic stress. (A) EMSA was performed with the labeled E4 scriptional start site, suggesting that the regulation of ATF probe described previously {Takeda et al. 1995) using ex- gpdl + expression by Atfl is direct. tracts derived from wild-type {HM 123) or zaatfl (NT146) cells or from wild-type cells incubated in medium containing 1.2 M sor- bitol. Where indicated, unlabeled competitor DNAs were added at a molar ratio of 100:1; the competitors were either the olig- Atfl is phosphorylated in vitro by Styl kinase onucleotides E4ATF (wt) or E4M1 (ml)(Takeda et al. 1995). (B) Taken together, the results described above strongly in- EMSA was performed as above except that the probe used rep- dicate that the activity of Atfl is regulated by the Styl/ resented sequences -459 to -248 with respect to the transla- Wis 1 signaling pathway. We next addressed whether the tion initiation site of the gpdl + gene. 2294 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Regulation of Atfl by stress-activated MAPK GST-Atfl -I~ 0' 15' 30' 45' 0' 15' 30' 45' Figure 7. Atfl is phosphorylated by the Styl kinase in vitro. (A) Cells mutated for styl (JMl144) were transformed with a - Thiamine + Thiamine pREP41 vector containing an HA-epitope and His6-tagged ver- sion of Styl. Expression of the modified Styl is regulated by the thiamine-repressible promoter present in this vector; expres- sion is permissible in the absence but not the presence of thia- mine. The transformed cells were grown in minimal medium either lacking or containing thiamine; KC1 was added to a con- centration of 0.9 M; and aliquots subsequently harvested at the times indicated. Extracts were prepared and Styl protein was HA-Styl immunoprecipitated from each extract using a hemagglutinin- specific antibody. The immune complexes were tested for ki- nase activity using GST-Atfl fusion protein as a substrate. The kinase assays were carried out for 30 min at 30°C and the sam- 0' 15' 30' 45' 0' lS' 30' 45' ples separated by SDS--PAGE and autoradiographed. The arrow indicates the position of the full-length fusion protein. The - Thiamine + Thiamine smaller species represent truncated products of the fusion pro- tein that accumulate following expression in bacteria. (B) The immunoprecipitated complexes described above were assayed GST-Atfl GST-Pcrl GST-Papl directly for the presence of Styl protein by Western analysis 0,9M NaCI + using the hemagglutinin-specific antibody. The position of HA- His6-Styl is indicated by the arrow. Expression of the modified i ¸~ i %~: • Sty 1 protein is repressed in the presence of thiamine. (C} Trans- formed cells containing HA-His6-Styl were grown in the ab- sence of thiamine and aliquots harvested before (-) and 15 min after ( + the ) addition of 0.9 M KC1. Extracts were prepared, Styl protein immunoprecipitated, and kinase assays performed as described in Fig. 7A. In addition to GST-Atfl, GST-Pcrl and GST-Papl were used as substrates. Only the GST-Atfl fusion was phosphorylated. The arrows indicate the migration position of each of the full-length fusion proteins. ATF family (ATF-2) and the AP-1 family (cJun, junD}. the GST-Atfl fusion was phosphorylated after preincu- The papl + gene of S. pombe encodes a transcription fac- bation in lysates from stressed cells but not from un- tor that has AP-1 DNA binding specificity (Toda et al. stressed cells or cells that were grown in the presence of 19911. We tested whether the Papl protein was also a thiamine and therefore did not express the Styl protein. substrate for Styl. As shown in Figure 7C, no phospho- In addition, the kinase activity that binds to the GST- rylation was observed. These results highlight the spec- Atfl beads is Wisl-dependent (data not shown). These ificity of the Styl-dependent phosphorylation of Atfl. results are consistent with Styl binding to its substrate The mammalian SAPKs, both SAPK/JNK and p38, Atfl to form a stable complex in vitro. form stable complexes with their substrates in vitro A physical interaction between Styl and Atfl was con- (Hibi et al. 1993; Derijard et al. 1994; Kyriakis et al. firmed by two additional assays. The first used the two- 1994; Gille et al. 1995; Raingeaud et al. 1995}. Such com- hybrid assay IFig. 8B). Styl was fused to the Gal4 DNA plex formation is essential for phosphorylation to occur binding domain and Atfl to a strong activation domain. and provides at least one mechanism by which targets Coexpression of the fusion proteins resulted in activa- can be recognized and phosphorylated by specific mem- tion of the Gal4-driven reporter, whereas expression of bers of the MAPK family. In order to investigate whether either alone failed to do so. The second involved the Styl could form a stable complex with Atfl, lysates from incubation of GST-Atfl with extracts from cells ex- osmotically stressed cells expressing HA-His6-Styl pressing epitope-tagged Styl followed by Western analy- were incubated with GST-Atfl protein prebound to glu- sis of bound proteins. As shown in Figure 8C, Styl bound tathione-agarose beads. Following extensive washing, to GST-Atfl but not to GST alone or a GST fusion con- kinase assays were performed. As shown in Figure 8A, taining the carboxy-terminal region of Atfl. GENES & DEVELOPMENT 2295 Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Wilkinson et al. Discussion Our main conclusion is that the Atfl transcription factor is regulated and directly phosphorylated by the Styl stress-activated kinase. As such, this demonstrates a re- markable degree of conservation between the function of GST-Atfl ~. stress-activated MAPK pathways in fission yeast and mammalian cells where the related ATF-2 factor is reg- ulated by SAPK/JNK and p38 kinases. In addition, it rep- resents the first biochemical demonstration in a yeast system of a direct link between a specific transcription factor and an MAPK signaling pathway. The conclusion is supported by three lines of evidence: (1) There is considerable overlap in the phenotypes that 0 15 30 45 0 15 30 45 result from the deletion of atfl or components of the - Thiamine + Thiamine pathway such as sty1 or wisl. Thus in both cases the resulting mutant cells are sterile, defective in survival at r~ r~ r~ ~, <, < stationary phase, hypersensitive to stress, and sup- pressed in the uncontrolled meiotic induction that nor- (3 + + + ,..- mally accompanies loss of Patl kinase activity. (2) The expression of a number of target genes is regulated by both the stress-activated pathway and Atfl. These in- clude the genes pyp2 +, gpdl +, and catalase (this study) I 96-- and stell + and fbpl + (Takeda et al. 1995; Kato et al. 8J 1996; Stettler et al. 1996). (3) Atfl is a substrate for the Sty l kinase in vitro. In addition, it can form a complex m~ 6 with Styl that is sufficiently stable to be detected by the ~1- Sty1 two-hybrid interaction assay and to withstand stringent washing conditions. Such complex formation is a com- mon feature of the mammalian stress-activated MAPKs 9 z e~ and their physiologically relevant target proteins. 0 a-HA Regulation of Atfl Figure 8. Interaction between Styl and Atfl. (A) styl mutant cells (IM1144) containing HA-His6-Styl as described in Fig. 7A The mechanism by which phosphorylation of Atfl by were grown in minimal medium either lacking or containing Sty l facilitates Atfl-dependent transcription is presently thiamine to control Styl expression. KC1 was added to a con- unknown. An important step will be to identify the sites centration of 0.9 M and aliquots harvested at the times indi- on Atfl that are phosphorylated by Styl. Preliminary cated. Extracts were prepared and incubated with GST-Atfl experiments suggest that multiple phosphorylation sites protein prebound to GSH-Sepharose beads for 2 hr at 4°C, as exist and that the elimination of single sites is not suf- described in Materials and Methods. The beads were exten- sively washed, kinase reaction buffer added, and a kinase reac- ficient to render the factor inactive (M. Samuels and N. tion assay allowed to proceed for 30 min at 30°C. The phospho- Jones, unpubl.). In the case of the mammalian ATF-2 rylation of the GST-Atfl protein was assessed by SDS-PAGE factor, regulation is mediated at the level of transcription and autoradiography. {B) Two-hybrid analysis of Atfl-Styl in- activation; the factor has an activation domain that is teraction. S. cerevisiae strain Y187 was transformed with: (1) stimulated by SAPK/INK or p38 phosphorylation (Gupta pAS-CYH2 expressing the GAL4 DNA binding domain (BD- et al. 1995; Livingstone et al. 1995). Current evidence cont.} and pACTII-Atfl containing the Atfl sequence fused to a would suggest that it is also the transactivation proper- strong activation domain (Atfl-AD); or I2) pAS1-CYH2-Styl ties of Atfl that are stimulated by Styl phosphorylation, containing the GAL4 DNA binding domain fused to Styl {BD- as there is no evidence that either the degree of DNA Styl) and pACTII containing the unfused activation domain- binding of Atfl or its specificity is altered following in- (Cont-ADb or {3) pAS1-CYH2-Styl and pACTII-Atfl. Expres- sion of a lacZ reporter containing GAL4 binding sites within its duction {Fig. 6B). Recent evidence suggests that Atfl is promoter was measured by assessing [3-galactosidase activity in present in cells as a heterodimer with another bZip pro- extracts as described in Materials and Methods. Measurements tein Pcrl, and deletion of the pcrl + gene results in a are expressed as the mean of three independent determinations phenotype similar to that of zlatfl cells {Watanabe and [ + S.D.J [C) Extracts of wild-type cells expressing HA-tagged Styl Yamamoto 1996). This suggests that it is the heterodi- were incubated in the presence of 2 ~g of GST, GST-Atfl (a mer that is the subject of regulation by the stress-acti- fusion containing full-length Atfl) or GST-Atfl{Cterm} (a fu- vated pathway. However, Pcrl is not phosphorylated by sion containing the carboxy-terminal half of Atfl }. After 30 min Styl and so we suggest that it is the Atfl component of at 4°C, proteins were precipitated with glutathione beads, ex- the heterodimeric complex that is the target of regula- tensively washed, seperated by SDS-PAGE, and probed for the tion. It will be important in the future to determine the presence of Styl using a monoclonal anti-HA antibody. 2296 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Regulation o| At[1 by stress-activated MAPK individual activities associated with each of the subunits ual differentiation pathway. However, it is likely that of the complex. Preliminary evidence indicates that the Atfl plays more than one role in the regulation of this expression of only a subset of the Atfl-dependent target pathway, because ectopic expression of Stel 1 suppresses genes is also dependent upon Pcrl, suggesting differen- the conjugation defect of AatfI cells without an accom- tial roles for Atfl homodimeric and Atfl/Pcrl het- panying increase in spore formation {data not shown). rodimeric complexes (W.M. Toone and N. Jones, un- Very similar observations were made in Astyl cells ex- publ.). pressing stell + from the SV40 promoter (Kato et al. 1996). The additional role of Atfl, therefore, presumably involves the expression of another target required in Genes that are regulated by Atfl meiosis. It is important for cells to maintain intracellular os- A model forthe different biological roles that have been molarity in response to changes in the extracellular wa- attributed to Atfl is presented in Figure 9. At least five ter potential. One way to accomplish this is to adjust the different genes that are dependent upon the Atfl factor internal concentratio~as of an osmolyte such as glycerol. have now been described (Takeda et al. 1995; this study). The gpdl + gene is involved in the synthesis of glycerol, In three of these cases {stell +, gpdl +, and pyp2+), acti- and its expression has been shown to be important for vation by stress conditions in a Styl/Wisl-dependent the response of the cell to high osmolarity conditions. manner has been demonstrated. The known function of The dependence on Atfl of its expression explains why the products encoded by these genes helps to explain the aatfl cells are osmosensitive. The expression of gpdl is defective phenotypes associated with deletion of atfl + induced upon osmotic shock. The level of induction, The sterile phenotype of Aatfl cells is, at least partially, however, is rather modest (three- to fivefold), because of explained by the loss of expression of the stel 1 + (Takeda a relatively high level of basal, uninduced expression; et al. 1995} gene that encodes a transcription factor re- nevertheless, this basal expression is dependent upon quired for the expression of genes necessary for the sex- Atfl and Wisl. This may reflect a persistent basal level of signaling through the Styl/Wisl pathway that can be activated acutely upon imposition of stress. Both the astyl (Degols et al. 1996) and Aatfl (data not Osmotic stress Oxidative stress shown) cells are mildly sensitive to oxidative stress. This Heat shock phenotype is very likely explained by the requirement of Ultraviolet light the Styl/Wisl pathway and the Atfl factor for the in- duced expression of the catalase gene following a variety of stress conditions. Catalase is a potent free-radical scavenger and plays an important role in the protection of cells against oxidant injury. Accordingly, the induc- tion of catalase activity by H202 is a common feature of 1 ,T! yeast and mammalian cells. The pyp2 + gene is another important downstream tar- get of the Styl kinase and the Atfl factor and is respon- [ at division I ° sible, at least in part, for feedback inhibition of the Sty1 P'tase pathway by the Pyp2 tyrosine-specific phosphatase. As a result, down-regulation of Sty1 kinase activity observed normally on prolonged exposure to osmotic stress is lost in Aatfl cells. One caveat is that the Sty1 kinase is over- stel 1 fbpl gpdl pyp2 expressed in these experiments and that the stoichio- catalase metric quantity of MAP kinase phosphatase relative to l l MAP kinase may be important for the timing of inacti- anti- Mei2 glycerol oxidant l synthesis vation. Nevertheless, we believe this provides the first full example of a feedback loop controlling an MAPK MEIOSIS OSMOTIC pathway to be described. Considerable attention has TOLERANCE been focused on the mechansims of MAPK pathway ac- Figure 9. The stress activated Styl kinase pathway in fission tivation, whereas relatively little is known of the yeast. This model depicts the interactions described in this pa- mechansims by which the signal is attenuated, despite per and previous reports {Millar et al. 1995; Shiozaki and Russell the fact that the duration of signal flow through the cas- 1995; Takeda et al. 1995; Degols et al. 1996; Kato et al. 1996; cade may be critical for the ultimate cellular response. Stettler et al. 1996). Activation of Styl kinase by a signal trans- Indeed, more recent experiments have uncovered Pyp2- duction pathway triggered by numerous stress conditions re- and Atfl-independent mechanisms operating to attenu- sults in the phosphorylation and activation of the transcription ate signaling through the Sty1 pathway that are yet un- factor Atfl. Active Atfl promotes the transcription of a number characterized (Degols et al. 1996; M.G. Wilkinson and of target genes that are involved in the stress response of the cell J.B.A. Millar, unpubl.). (gpdl +, fbpl +, catalase}, in the sexual differentiation pathway (stel 1 + ) or in a negative feedback control loop {pyp2 + ). Although to date many of the targets that we have GENES & DEVELOPMENT 2297 Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Wilkinson et al. identified as being Atfl-dependent are also dependent naling pathway. On the other hand, it may mean that upon the Styl/Wisl pathway, there are likely to be ad- there are common features that are difficult to recognize ditional targets regulated by the pathway but indepen- until careful structure and function analyses have been dent of Atfl. This conclusion comes from the observa- carried out. Support for this possibility comes from a tion that Aatfl cells do not display any cell size at divi- comparison of the ATF-2 and cJun factors. Outside of the sion defect and show no synthetic interaction with binding domain, very little homology is apparent. Yet it cdc25-22 cells; both phenotypes are associated with loss is now clear that they both contain activation domains of Styl or Wisl. Therefore, the control of the initiation of that are regulated in a similar fashion; they are depen- mitosis by the Styl pathway cannot be explained by its dent upon phosphorylation by stress-activated MAPKs control of Atfl activity. Perhaps a different transcription and contain a region involved in stable complex forma- factor controlled by Styl is involved. tion with these kinases. Furthermore, ATF-2 has a Hob- l-like motif that has been shown to be an important component of the cJun activation domain, and muta- Regulation of ATF activity by stress-activated MAPK tions within this motif of ATF-2 cripple its activity (A. pathways is conserved Clark and N. Jones, unpubl.). Thus ATF-2 and cJun Atfl shows homology to the mammalian factor ATF-2 clearly overlap at the functional level despite the fact (Takeda et al. 1995). This homology is almost entirely that there is little similarity at the primary amino acid restricted to the bZip domain; it would appear to be sig- sequence level. We suggest that a similar situation may nificant because it is greater than the sequence similar- exist between ATF-2 and Atfl. ity seen between bZip domains of many of the different The situation appears to be quite different in the bud- mammalian members of the ATF family. The activity of ding yeast S. cerevisiae. An element has been identified ATF-2 is regulated by the stress-activated MAPK path- (STRE) that mediates the activation of many genes in ways through the phosphorylation of critical threonine response to a variety of stress conditions including os- residues in the amino-terminal activation domain motic and oxidative stress. This element, however, is (Gupta et al. 1995; Livingstone et al. 1995). Our finding not an ATF binding site, and two recently described fac- that Atfl is regulated by the equivalent pathway in fis- tors that recognize and bind to it are zinc-finger-contain- sion yeast highlights a remarkable level of conservation ing proteins rather than bZip proteins (Martinez-Pastor in the function of the stress-activated pathways. Al- et al. 1996). One budding yeast bZip-containing factor though it has been clear for some time that the signaling that does play a significant role in stress response is the pathways in yeast and mammalian cells show a degree of Yap 1 p activator, which mediates the induction of certain conservation, this is the first time that the conservation genes by oxidative stress (Ruis and Schuller 1995). How- extends to a downstream transcription factor target. It ever, this factor is homologous to the fission yeast factor would appear, therefore, that the Atfl factor is both a Pap1 rather than Atfl; the DNA binding specificities of structural and functional homolog of ATF-2, and, indeed, Papl and Atfl are clearly different. Furthermore, the preliminary experiments indicate that ectopic expres- Styl homolog in budding yeast, Hoglp, is activated by sion of ATF-2 in Aatfl cells results in partial suppression osmotic stress only, and so it appears that the HOG path- of the defective mating phenotype (T. Takeda and N. way, unlike the Styl pathway in fission yeast or the Jones, unpubl.). It should be noted, however, that outside JNK/SAPK or p38 pathways in mammalian cells, is not of the bZip domain there is no obvious homology be- a general stress-signaling pathway. Therefore, with re- tween the two factors. In particular, features of the mam- spect to the signaling pathways and the transcription malian ATF-2 transactivation domain such as the zinc- factors that are involved in the stress response, it appears finger and kinase interaction domain cannot be found in that fission yeast more closely resembles what has been Atfl. This may mean that the two factors are regulated observed in mammalian cells. As such it offers a useful differently even though they both require the stress-sig- model for studying these stress-related events. Table 2. Strains used in this study Strain no. Genotype Reference/source HM 123 h-leu-32 Our stock PR 109 h-leul-32ura4-D18 P. Russell (The Scripps Research Institute, La Jolla, CA) NT146 h - leul-32ura4-D18atf l :ura4 : + Takeda et al. 1995 JM544 h- leul-32ura4-D18wis l :ura4 : ÷ Millar et al. 1995 JM1144 h- sty l- l leu l -32ura4-D18 Millar et al. 1995 SOP13 h- leu l-32ura4-D18ade6-216pyp l :ura4 : + P. Russell (The Scripps Research Institute, La Jolla, CA) JM 836 h -leu1-32 ura4-D18 patl- 114 This study JM 1266 h -leul-32 ura4-D18 wisl ::ura4 +patl-I 14 This study JM 1267 h -leu1-32 ura4-D18 styl ::urta4 +patl-114 This study JM1352 h - leu 1-32ura4-D 18ade6-M210cdc25-22 This study 2298 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Regulation of At[1 by stress-activated MAPK Materials and methods was stopped by adding 5 ~1 of 5 x Lammeli buffer and heated to 95°C for 5 min. Samples were then run on 10% PAGE gels, Yeast strains, media, and general methods transferred onto PVDF membranes, and autoradiographed. The yeast strains used in this study are described in Table 2. Media and general experimental methods have been described elsewhere (Moreno et al. 1991; Alfa et al. 1993). Western blot analysis The PVDF membranes were blocked for 1 hr with 3% BSA Plasmids containing 0.1% {vol/vol) Tween 20. Anti-HA antibody was added at 1 / 1000 dilution in blocking buffer for 1 hr and the blot To express various GST proteins, the plasmids pGST-Atfl, was washed 3x for 15 min with TBS plus 0.5% NP40. Anti- pGST-Atfl(C-term), pGST-Pcrl and GST-Papl were con- mouse HRP at 1/10000 was applied for 1 hr and the blot was structed by inserting PCR-generated fragments containing full- washed 3 x for 15 rain. ECL (Amersham) was performed to vi- length Atfl, Pcrl, or Papl or truncated Atfl (amino acids 286- sualize the bound antibody. 586) into the vector pGEX-KG (Guan and Dixon 1991). The proteins were expressed and purified on glutathione-agarose beads as described previously (Guan and Dixon 1991). For the two-hybrid interaction assay, full-length Styl and Atfl se- Electrophoretic mobility shift assay quences were cloned into the vectors pAS1-CYH2 and PACTII, Cells were grown to mid-log phase, harvested, washed with 1 ml respectively (Durfee et al. 1993). of H20, resuspended in 40 ~1 of lysis buffer [25 mM HEPES {pH 7.6), 0.1 mM EDTA, 150 mM KC1, 0.1% TritonX100, 25% Glyc- RNA isolation and hybridization erol, 1 M urea, 1 mM DTT, 1 mM PMSF, 10 p~g/ml aprotinin, 10 p~g/ml leupeptin, 1 mM EGTA, 1 mM tetrasodium pyrophos- The isolation of mRNA and subsequent Northern analysis was phate, 100 mM ~-glycerophosphate, 10 mM NaF, and 1 mM so- performed as described previously (Millar et al. 1995). A 1-kb dium orthovanadate] and lysed as above. Then 30 ~g lysate was EcoRV fragment from pGEM-3Z-pyp2 was used to probe for incubated in 20 ~1 binding buffer containing 25 mMHEPES (pH pyp2 mRNA iMillar et al. 1995), a 1.1-kb EcoRI fragment was 7.6), 34 mM KC1, 5 mM MgC12, and 2 ~g of poly[d(I-C)] for 10 min used to probe for gpdl mRNA (Pidoux et al. 1990), and a 1.5-kb at room temperature, and then for 20 min with 1 ng 32p-labeled PCR-derived fragment covering the whole open reading frame probe. Reactions were run out on a 4% acrylamide gel in 0.5 M was used to probe for catalase mRNA {Nakagawa et al. 1995). TBE, dried down, and exposed for autoradiography. Styl kinase assays AStyl cells containing the plasmid pRep41 HA/6xHis styl Two-hybrid interaction assay (Millar et al. 1995) were grown in EMM containing 75 ~g/ml The atfl gene was amplified by PCR using the 5' oligonucle- uracil, and where indicated, 2 ~M thiamine until they reached otide 5'-ACATACCATGGCATCCCCGTCTCCCGTCAATA- midlog phase. Then 40 ml of cells were mixed with either 17 ml CTTCCA-3' incorporating an NcoI site (shown italicized) and fresh (Edinburgh minimal medium (EMM) or EMM containing 3 the 3' oligonucleotide 5'-ACGTCTCGAGAGATCAAAAA- M KC1 and incubated for the appropriate length of time. The CAGTCTAGTACCT-3' incorporating a XhoI site (shown itali- cells were rapidly cooled and harvested 4°C. The resulting pel- cized) to produce a 1734-bp fragment that was digested with lets were washed with ice-cold H20 and resuspended with 40 ~1 NcoI and XhoI and ligated into the NcoI and XhoI sites of pAC- of lysis buffer [20 mM Tris (pH 8.0}, 137 mM NaC1, 15% glycerol, TII (Durfee et al. 1993) to form pACTII-atfl expressing a fusion 0.1% TritonX-100, 1 mM DTT, 5 mM EDTA, 1 mM PMSF, 10 protein of the GAL4 activation domain with the atfl ORF under ~g/ml aprotinin, 10 g/ml leupeptin, 1 mM EGTA, 1 mM tetra- the control of the adh promoter. The plasmid pRep41HA/6xHis sodium pyrophosphate, 100 mM beta-glycerophosphate, 10 mM styl (Millar et al. 1995)was digested with NdeI and BamHI and NaF, and 1 mM sodium orthovanadate] and broken by vortexing a fragment containing the styl ORF ligated into the NdeI and with 1 ml of chilled acid-washed glass beads. The resulting su- BamHI sites of pAS1-CYH2 (Durfee et al. 1993) to form pAS1- pernatant was removed and assayed for protein concentration CYH2-Styl. Expression of both fusion proteins in S. cerevisiae by use of the Bio-Rad Protein Assay Kit. strain Y187 (Durfee et al. 1993) was verified by Western blot For the immunecomplex kinase assay, 170 ~g of lysate was using a monoclonal antibody to the hemaglutinin tag (12CA5). incubated with 5 ~1 of monoclonal anti-HA antibody for 1 hr at 4°C. Then 20 Izl of protein G-Sepharose (dry volumet was added for an additional 1 hr. The beads were collected, and washed 3 x with 400 ~1 of lysis buffer. Then 2 ~g of GST-Atfl protein Acknowledgments was added in wash buffer. The buffer was removed and 20 ~1 of kinase reaction buffer [20 mM HEPES (pH 7.5), 20 mM MgCI~, 2 We thank Lee Johnston, Steve Sedgwick, Richard Treisman, and mM DTT, 20 ~M ATP, 5 ~Ci [~/_a2p] ATP] was added. The reac- members of the Division of Yeast Genetics INational Institute tion was carried out for 30 rain at 30°C and stopped by adding 5 for Medical Research) and Laboratory of Gene Regulation [Im- ~.1 of 5 x Lammeli buffer and heated to 95°C for 5 min. Samples perial Cancer Research Fund (ICRF)] for helpful advice and dis- were then run on 10% PAGE gels, transferred onto PVDF mem- cussions and Paul Russell for strains. This research was sup- branes, and autoradiographed. ported by the ICRF (N.J. and T. Toda.), the Medical Research For the binding kinase assay, 300 ~g of lysate was incubated Council (J.B.A.M.}, and the Human Frontier Science Program with 2 ~g of GST-Atfl protein prebound to glutathione-Seph- Organization (N.J.). arose for 2 hr at 4°C. The beads were collected, washed 3 x with The publication costs of this article were defrayed in part by 400 ~1 lysis buffer, and 20 ~1 kinase reaction buffer [20 mM payment of page charges. This article must therefore be hereby HEPES (pH 7.5), 20 mM MgC12, 2 mM DTT, 20 ~M ATP, 5 ~Ci marked "advertisement" in accordance with 18 USC section [~/_32p] ATP] was added. The reaction was for 30 min at 30°C and 1734 solely to indicate this fact. GENES & DEVELOPMENT 2299 Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Wilkinson et al. Reierences Kyriakis, J.M., P. Banerjee, E. Nikolakaki, T. Dai, E.A. Rubie, M.F. Ahmad, J. Avruch, and J.R. Woodgett. 1994. The stress- Aiba, H., H. Yamada, R. Ohmiya, and T. Mizuno. 1995. The activated protein kinase subfamily of c-Jun kinases. Nature osmoinducible gpdl + gene is a target of the signaling path- 369: 156-160. way involving Wisl MAP-kinase kinase in fission yeast. Kuge, S. and N. Jones. 1994. YAP1 dependent activation of FEBS Lett. 376: 199-201. TRX2 is essential for the response of Saccharomyces cerevi- Alfa, C., P. Fantes, J. Hyams, M. McLeod, and E. Warbrick. 1993. siae to oxidative stress by hydroperoxides. EMBO J. 13: 655- Experiments with fission yeast. Cold Spring Harbor Labora- tory Press, Cold Spring Harbor, N.Y. Lee, J.C., J.T. Laydon, P.C. McDonnell, T.F. Gallagher, S. Ku- Ammerer, G. 1994. Sex, stress and integrity: The importance of mar, D. Green, D. McNulty, M.J. Blumenthal, J.R. Heys, MAP kinase pathways in yeast. Curr. Opin. Genet. Dev. S.W. Landvatter, J.E. Strickler, M.M. McLaughlin, I.V. Sie- 4: 90-95. mens, S.M. Fisher, G.P. Livi, J.R. White, J.L. Adams, and P.R. Brewster, J.L., T. de Valoir, N.D. Dwyer, E. Winter, and M.C. Young. 1994. A protein kinase involved in the regulation of Gustin. 1993. An osmosensing signal transduction pathway inflammatory cytokine biosynthesis. Nature 372: 739-746. in yeast. Science 259: 1760-1763. Levin, D.E. and B. Errede. 1995. The proliferation of MAP kinase Davies, R.J. 1994. MAPKs: A new JNK expands the group. signaling pathways in yeast. Curt. Opin. Ceil Biol. 7: 197- Trends Biochem. Sci. 19: 470--473. Degols, G., K. Shiozaki, and P. Russell. 1996. Activation and Livingstone, C., G. Patel, and N. Jones. 1995. ATF-2 contains a regulation of the Spcl stress-activated protein kinase in phosphorylation-dependent transcriptional activation do- Schizosaccharomyces pombe. Mol. Cell Biol. 16: 2870- main. EMBO J. 14: 1785-1797. 2877. Marais, R., J. Wynne, and R. Treisman. 1993. The SRF accessory Derijard, B., M. Hibi, I.H. Wu, T. Barrett, B. Su, T. Deng, M. protein Elkl contains a growth factor-regulated transcrip- Karin, and R.J. Davies. 1994. JNKI: A protein kinase stim- tional activation domain. Cell 73: 381-393. ulated by UV light and Ha-Ras that binds and phosphorylates Marshall, C.J. 1994. MAP kinase kinase kinase, MAP kinase the c-Jun activation domain. Cell 76: 1025-1037. kinase, and MAP kinase. Curr. Opin. Genet. Dev. 4: 82-89. Durfee, T., K. Becherer, P.-L. Chen, S.-H. Yeh, Y. Yang, A. Martinez-Pastor, M.T., G. Marchler, C. Sculler, A. Marchler- Kilbum, W.-H. Lee, and S.J. Elledge. 1993. The retinoblasto- Bauer, H. Ruis, and F. Estruch. 1996. The Saccharomyces ma protein associates with the protein phosphatase type 1 cerevisiae zinc finger proteins Msn2p and Msn4p are re- catalytic subunit. Genes & Dev. 7: 555-569. quired for transcriptional induction through the stress- re- Freshney, N.W., L. Rawlinson, F. Guesdon, E. Jones, S. Cowley, sponse element (STRE 1. EMBO J. 15: 2227-2235. J. Hsuan, and J. Saklatvala. 1994. Interleukin-1 activates a Maundrell, K. 1993. Thiamine-repressible expression vectors novel protein kinase cascade that results in the phosphory- pREP and pRIP for fission yeast. Gene 123: 127-130. lation of hsp27. Cell 78: 1039-1049. Millar, J.B.A., P. Russell, J.E. Dixon, and K.L. Guan. 1992. Neg- Gille, H., T. Strahl, and P.E. Shaw. 1995. Activation of ternary ative regulation of mitosis by two functionally overlapping complex factor elk-1 by the stress-activated protein kinases. PTPases in fission yeast. EMBO J. 11: 4943--4952. Curr. Biol. 5: 1191-1200. Millar, J.B.A., V. Buck, and M.G. Wilkinson. 1995. Pypl and Guan, K.L. and J.E. Dixon. 1991. Eukaryotic proteins expressed pyp2 PTPases dephosphorylate an osmosensing MAP kinase in Escherichia coli: an improved thrombin cleavage and pu- controlling cell size at division in fission yeast Genes & Dev. rification procedure of fusion proteins with glutathione 9: 2117-2130. S-transferase. Anal. Biochem. 192: 262-267. Moreno, S., A. Klar, and P. Nurse. 1991. Molecular genetic anal- Gille, H., A.D. Sharrocks, and P.E. Shaw. 1992. Phosphorylation ysis of fission yeast Schizosaccharomyces pombe. Method. of transcription factor p62 Tcv by MAP kinase stimulates Enzymol. 194: 795-823. temary complex function at the c-fos promoter. Nature Nakagawa, C.W., N. Mutoh, and Y. Hayashi. 1995. Transcrip- 358: 414-417. tional regulation of catalase gene in the fission yeast Gupta, S., D. Campbell, B. Derijard, and R.J. Davies. 1995. Tran- Schizosaccharomyces pombe: Molecular cloning of the cat- scription factor ATF2 regulation by the JNK signal transduc- alase gene and northern blot of the transcript. J. Biochem. tion pathway. Science 267: 389-393. 118: 109-116. Han, J., J.-D. Lee, L. Bibbs, and R.L. Ulevitch. 1994. A MAP Pidoux, A.L., E.H. Fawell, and J. Armstrong. 1990. Glycerol-3- kinase targeted by endotoxin and hyperosmolarity in mam- phosphate dehydrogenase homologue from Schizosaccharo- malian cells. Science 265:808-811. myces pombe. Nucleic Acids Res. 18: 7145. Herskowitz, I. 1995. MAP kinase pathways in yeast: For mating Raingeaud, J., S. Gupta, J.S. Rogers, M. Dickens, J. Han, R.J. and more. Cell 80: 187-197. Ulevitch, and R.J. Davies. 1995. Pro-inflammatory cytokines Hibi, M., A. Lin, T. Smeal, A. Minden, and M. Karin. 1993. and environmental stress cause p38 mitogen-activated pro- Identification of an oncoprotein- and UV-responsive protein tein kinase activation by dual phosphorylation on tyrosine kinase that binds and potentiates the c-Jun activation do- and threonine. J. Biol. Chem. 270: 7420-7426. main. Genes & Dev. 7: 2135-2148. Rouse, J., P. Cohen, S. Trigon, M. Morange, A. Alonso-Llamaz- Iino, Y. and M. Yamamoto. 1985a. Mutants of Schizosaccharo- ares, D. Zamanillo, T. Hunt, and A. Nebrada. 1994. A novel myces pombe which sporulate in the haploid state. Mol. kinase cascade triggered by stress and heat shock that stim- Gen. Genet. 118: 416-421. ulates MAPKAP kinase-2 and phosphorylation of the small ~. 1985b. Negative control for the initiation of meiosis in heat shock proteins. Cell 78: 1027-1037. Schizosaccharomyces pombe. Proc. Natl. Acad. Sci. Ruis, H. and C. Schuller. 1995. Stress signaling in yeast. BioEs- 82: 2447- 2451. says 17: 959-965. Kato, T., K. Okazaki, H. Murakami, S. Stettler, P. Fantes, and H. Shiozaki, K. and P. Russell. 1995. Cell cycle control linked to Okayama. 1996. Stress signal, mediated by a HOGl-like extracellular environment by MAP kinase pathway in fis- MAP kinase, controls sexual development in fission yeast. sion yeast. Nature 378: 739-743. FEBS Lett. 378: 207-212. Stettler, S., E. Warbrick, S. Prochnik, S. Mackie, and P. Fantes. 2300 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press Regulation of Atfl by stress-activated MAPK 1996. The wisl signal transduction pathway is required for expression of cAMP-repressed genes in fission yeast. I. Cell Sci. 109: 1927-1935. Sugimoto, A., Y. Iino, Y. Watanabe, and M. Yamamoto. 1991. Schizosaccharomyces pombe stel 1 + encodes a transcription factor with an HMG motif that is a critical regulator of sex- ual development. Genes & Dev. 5: 1990-1999. Takeda, T., T. Toda, K. Kominami, A. Kohnosu, M. Yanagida, and N. lones. 1995. Schizosaccharomyces pombe atfl + en- codes a transcription factor required for sexual development and entry into stationary phase. EMBO ]. 14: 6193-6208. Toda, T., M. Shimanuki, amd M. Yanagida. 1991. Fission yeast genes that confer resistence to staurosporine encode an AP1- like transcription factor and a protein kinase related to the mammalian ERK 1/MAP2 and budding yeast FUS3 and KSS 1 kinases. Genes & Dev. 5: 60-73. Treisman, R. 1996. Regulation of transcription by MAP kinase cascades. Curt. Opin. Cell Biol. 8: 205-215. van Dam, H., D. Wilhelm, I. Herr, A. Steffen, P. Herrlich, and P. Angel. 1995. ATF-2 is preferentially activated by stress-acti- vated protein kinases to mediate c-jun induction in response to genotoxic agents. EMBO ]. 14: 1798-1811. Warbrick, E. and P. Fantes. 1991. The Wisl protein kinase is a dose-dependent regulator of mitosis in Schizosaccharomy- ces pombe. EMBO J. 10: 4291--4299. Wasiewicz, A.J. and J.A. Cooper. 1995. Mitogen and stress re- sponse pathways: MAP kinase cascades and phosphatase regulation in mammals and yeast. Curr. Opin. Cell. Biol. 7: 798-805. Watanabe, Y. and M. Yamamoto. 1996. Schizosaccharomyces pombe pcrl + encodes a CREB/ATF protein involved in reg- ulation of gene expression for sexual development. Mol. Cell Biol. 16: 704-711. Whitmarsh, A.J., P. Shore, A.D. Sharrocks, and R.I. Davies. 1995. Integration of MAP kinase signal transduction path- ways at the serum response element. Science 269: 403-407. Zinck, R., M.A. Cahill, M. Kracht, C. Sachsenmaier, R.A. Hip- skind, and A. Nordheim. 1995. Protein synthesis inhibitors reveal differential regulation of mitogen-activated protein- kinase and stress-activated protein kinase pathways that converge on elk-1. Mol. Cell. Biol. 15: 4930-4938. GENES & DEVELOPMENT 2301 Downloaded from genesdev.cshlp.org on November 1, 2021 - Published by Cold Spring Harbor Laboratory Press The Atf1 transcription factor is a target for the Sty1 stress-activated MAP kinase pathway in fission yeast. M G Wilkinson, M Samuels, T Takeda, et al. Genes Dev. 1996, 10: Access the most recent version at doi:10.1101/gad.10.18.2289 This article cites 47 articles, 15 of which can be accessed free at: References http://genesdev.cshlp.org/content/10/18/2289.full.html#ref-list-1 License Receive free email alerts when new articles cite this article - sign up in the box at the top Email Alerting right corner of the article or click here. Service Copyright © Cold Spring Harbor Laboratory Press
Journal
Genes & Development – Unpaywall
Published: Sep 15, 1996