| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Molecular and Cellular Biology, International Graduate School of Arts and Sciences, Yokohama City University, Yokohama 230-0045, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionExperimental proceduresReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
In yeast, TFIID is composed of the TATA box-binding protein (TBP) and 14 TBP-associated factors (TAFs) (Thomas & Chiang 2006). The TAF1 N-terminal domain, TAND, binds to the concave and convex surfaces of TBP, thereby inhibiting TBP binding to the TATA element (Kotani et al. 2000). TBP loading onto the promoter is a key regulatory step for activation. Thus, we previously proposed that transcriptional activators reverse the TAND–TBP interaction, enabling the formation of a productive TFIIA–TBP–TATA complex (Kotani et al. 2000). While characterizing TAND, we found that the taf1
TAND mutation impairs HIS4 transcription in rich media but not under starvation conditions (Takahata et al., unpublished observations). This suggests that TAND may be required for transcriptional activation by Bas1/Bas2 but not by Gcn4. Interestingly, the same phenotype had been reported for the med9 strain that lacks a gene encoding one of nonessential components of Mediator (Han et al. 1999). The fact that taf1TAND and med9 interact genetically (Takahata et al., unpublished observations) raises the possibility that TAND and Med9 may carry similar or overlapping functions in each co-activator complex.
Mediator was originally identified in yeast as a co-activator that appeared to bridge between transcriptional activators and the carboxy-terminal domain (CTD) of the largest subunit of pol II (Kornberg 2005). Yeast Mediator is comprised of 25 subunits, whose mammalian counterparts were all recently identified (Conaway et al. 2005; Bourbon 2008). The yeast Mediator subunits reside in four structurally distinct modules, named Head (Med6, Med8, Med11, Med17, Med18, Med20, Med22), Middle (Med1, Med4, Med7, Med9, Med10, Med19, Med21, Med31), Tail (Med2, Med3, Med5, Med14, Med15, Med16), and Cyc-C (Med12, Med13, Cdk8, CycC) (Baidoobonso et al. 2007; Bourbon 2008). Each module appears to be constituted by a few submodules, such as Med2–Med3–Med15 (Tail) (Li et al. 1995; Zhang et al. 2004), Med2–Med3–Med5–Med15 (Tail) (Beve et al. 2005), Med1–Med4–Med9 (Middle) (Han et al. 2001), minimal Head (Med11, Med17, Med22), core Head (Med6, Med8N, Med11, Med17, Med22), and Med8C–Med18–Med20 (Head) (Takagi et al. 2006; Lariviere et al. 2008). Genome-wide expression analyses have shown that these four modules are required for the transcription of different sets of genes in vivo (van de Peppel et al. 2005). Consistently, mutations in different modules differentially affect the three in vitro activities of Mediator: the abilities to facilitate activated transcription, to enhance basal transcription, and to enhance TFIIH-dependent CTD phosphorylation (Lee et al. 1999; Myers et al. 1999).
Med9 is a small protein (aa 1–149) that is identical to Cse2 (Han et al. 1999). cse2 was originally isolated as a mutation that affects chromosome segregation (Xiao et al. 1993). Earlier studies showed that Med9 forms a stable modular structure with Med1 and Med4, and is required for the transcription of a relatively small number of genes, that is, 175 positively and 209 negatively (Han et al. 2001). Furthermore, a partial Middle module (Med9/Med10 module) reconstituted with recombinant HA-tagged Med1, Med4, Med7, Med9, Med10, and Med21 was shown to interact with TBP, TFIIB, CTD of pol II, and the negative regulatory heterodimer Cdk8-CycC (Kang et al. 2001). Interactions between the Med9/10 module and Cdk8-CycC and/or between Med21 (Middle) and Tup1 (Gromoller & Lehming 2000) may explain why Med9 is required for transcriptional repression by LexA-Cdk8 or LexA-Ssn6, and why it represses a larger number of genes (209) than it activates (175) (Han et al. 2001).
med19 reportedly destabilizes the attachment of the Middle module to the rest of Mediator, thereby generating not only Mediator lacking Med19 (Mediator-Med19) but also Mediator lacking the entire Middle module (i.e. Middle-less Mediator) (Baidoobonso et al. 2007). Using in vitro experimental systems, Mediator-Med19 was shown to be partially defective for both basal and activated transcription (Baidoobonso et al. 2007). Importantly, Middle-less Mediator cannot stimulate activated transcription, but retains, in part, its ability to enhance basal transcription similar to Mediator-Med19 (Baidoobonso et al. 2007). In addition, neither type of Mediator enhances TFIIH-dependent CTD phosphorylation (Baidoobonso et al. 2007). These observations indicate that the Middle module plays an essential role in transcriptional activation. Consistent with this, the Head module alone, when reconstituted with recombinant proteins, could weakly enhance basal transcription (c. twofold to threefold), but failed to stimulate activated transcription or TFIIH-dependent CTD phosphorylation (Kang et al. 2001; Takagi et al. 2006).
Med9 is reportedly required, at least by some specific transcriptional activators, for transcriptional activation in vivo (Han et al. 1999; Leroy et al. 2006), although its requirement for transcriptional activation in vitro remains in dispute (Han et al. 1999; Liu et al. 2001). It is also unclear how Med9 is assembled into the Middle module, and to what extent Med9 contributes to structural integrity maintenance of the Middle module and/or the entire Mediator, although yeast two-hybrid assays suggest that it interacts directly with Med4 and Med7 (Guglielmi et al. 2004). To address these issues and gain novel insights into the structural and/or functional roles of Med9, especially those carried out by its non-conserved amino-terminal and well-conserved carboxy-terminal regions, we constructed a series of med9 mutants and analyzed them in various in vivo and in vitro experimental systems. We compare our novel findings with previous observations regarding Med9 and/or other Mediator components.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
A previous study showed that the amino-terminal regions of Med9 diverge significantly even within mammals, whereas the carboxy-terminal (c. 80 aa) region is highly conserved from mammals to insects (Tomomori-Sato et al. 2004) (Fig. S1 in Supporting Information). The same study also suggested that two short regions, 64–80 aa (S1, Fig. 1A) and 102–120 aa (S2, Fig. 1A), of Saccharomyces cerevisiae Med9 exhibit sequence similarities to their counterparts in higher eukaryotes (Tomomori-Sato et al. 2004). Therefore, to define the function of each domain of S. cerevisiae Med9, we constructed 11 med9 deletion mutants (designated M1–M11) that lacked or contained different combinations of the two short (S1, S2), amino-terminal (N), middle (M), and/or carboxy-terminal (C) regions (Fig. 1A).
|
med9) could not grow at 37 °C (Fig. 1B), which is consistent with previous results (Xiao et al. 1993; Han et al. 1999). Strains carrying M1, M2, M9, and M10 could grow at 37 °C as efficiently as the wild-type (WT) strain (Fig. 1B). While strains carrying M4, M5, M6, M7, and M8 showed severe growth defects similar to the med9 strain, those carrying M3 and M11 showed more modest defects under the same conditions (Fig. 1B). These observations indicate that the N-S1 region is dispensable, but that the S2 and C regions are important and indispensable for growth at 37 °C, respectively. Interestingly, the M region is dispensable by itself (M10), but becomes more vital in the absence of the N-S1 region (compare M3 with M2). As each of these mutant proteins was expressed at a similar level (data not shown, Fig. 3A), the N-S1-M, S2, and C regions may be crucial for interactions with other factors, including components of Mediator itself, as described below.
|
med9 strain cultured at the non-permissive temperature (Fig. 1C, lanes 1–4). HIS4 and ARG4 expression was more or less affected by all med9 mutations, but particularly strongly by M5, M6, M7, and M10 (37 °C), in a manner that was not completely consistent with their growth phenotypes. Interestingly, HXK1 expression was dramatically increased in M1, M2, and M3 strains (Fig. 1C, lanes 5–10), although the expression levels were not as high as in other med9 strains, including med9. Therefore, we propose that the N region probably plays a pivotal role in the repression of HXK1, whereas the S2-C region is probably required for the enhanced expression of HXK1 in the absence of the N region. Notably, all the med9 mutations affected the expressions of PGK1 and ADH1 much less severely than they affected those of HIS4, ARG4 or HXK1. Taken together, these data indicate that distinct regions of Med9 may differentially regulate (i.e., positively or negatively) the transcription of different gene subsets. Med4 and Med7 interact directly with different regions of Med9
The Middle module is composed of Med9 and seven other polypeptides, that is, Med1, Med4, Med7, Med10, Med19, Med21, and Med31 (Guglielmi et al. 2004; Beve et al. 2005; Baidoobonso et al. 2007). Previous studies have shown that Med4 and Med7 interact with Med9 in yeast two-hybrid assays (Guglielmi et al. 2004), and that a stable modular structure can be formed by the co-expression of Med1, Med4, and Med9 in insect cells (Han et al. 2001). Furthermore, recombinant proteins of mammalian Med4 and Med9 interact with each other in several in vitro binding assays (Tomomori-Sato et al. 2004). Therefore, we sought to determine whether Med7 also binds to Med9 directly, and if so, which regions of Med9 are required for its interaction with Med4 and Med7.
The interaction between Med4 and Med9 was examined by the co-expression of these two polypeptides in bacterial cells, as the WT (full-length) Med9 protein was insoluble when produced alone (data not shown). Non-tagged Med9 (WT or M1) proteins could be purified with the His-tagged Med4 protein by using a Ni2+-agarose resin (Fig. 2A), indicating that they both directly bound to Med4. In addition, as the amounts of WT and M1 that co-purified with Med4 were equivalent, at least the N region of Med9 would appear not to be required for interactions with Med4. Subsequent gel filtration analyses showed that the Med4–Med9 [WT or M1] binary complex may contain more than one molecule of each polypeptide, as its apparent molecular mass (>p;200 kDa) was larger than that calculated for a simple dimer (c. 50 kDa) (Fig. 2B).
|
Attempts to express a soluble WT (full-length) Med7 protein failed (data not shown) (Baumli et al. 2005). As yeast two-hybrid assays suggest that the amino-terminal region (aa 1–100) of Med7 is involved in its interaction with Med9 (Guglielmi et al. 2004), we expressed a slightly larger region (Med7-N; aa 1–101) in bacterial cells and subjected the cell extracts to GST pull-down assays, as described above (Fig. 2E). Med7-N bound to GST-Med9 [M5 and M6] strongly (lanes 5 and 6), to GST-Med9 [M2 and M7] weakly (lanes 2 and 7), and to GST-Med9 [M3, M4, M8, aa 64–80, and aa 81–101] at marginally detectable levels (lanes 3, 4, 8–10), but did not bind to GST-Med9 [aa 102–118, M9, M10, and M11] (lanes 11, 13–15). These results indicate that the S1, M and S2 regions are individually necessary (M9–11), but not sufficient (aa 64–80, aa 81–101, and aa 102–118), for interaction with Med7-N. The requirements for S1 (compare M2 and M9), M (compare M7 and M10), and S2 (compare M6 and M11) seem to vary depending on the context. In addition, it remains unknown how important the N and C regions are for this interaction, as GST-Med9 [WT or M1] could not be produced as a soluble protein (data not shown). To our knowledge, this is the first evidence to show a direct interaction between Med7 and Med9. As the binding profiles of Med7-N with GST-Med9 derivatives are different from those of Med4 (compare Fig. 2D,E), these two proteins probably interact with different regions of Med9.
The amino-terminal region (aa 1–70) of Med4 seems to be involved in its interaction with Med9 (Guglielmi et al. 2004). To confirm this, the amino- (Med4-N; aa 1–70) and carboxy- (Med4-C; aa 71–284) terminal regions of Med4 were expressed separately and were subjected to GST pull-down assays (Fig. 2F). GST-Med9 [M2] could bind to Med4-N, but not to Med4-C (lanes 2 and 4). Thus, we conclude that the amino-terminal region (aa 1–70) of Med4 interacts with the carboxy-terminal half of Med9 (M2; aa 81–149).
Regions of Med9 required for the integrity of Mediator
Holo-RNA polymerase II purified from the med9 strain reportedly contains reduced amounts of Med1 and Med4 (Han et al. 2001). However, it remains largely unclear how Med9 contributes to the maintenance of the structural integrity of Mediator. Therefore, we first examined whether several components of the Tail (Med2 and Med15), Head (Med6, Med11, Med17 and Med20), Middle (Med4 and Med7), and Cyc-C (Cdk8) modules could be immunopurified with FLAG-tagged Med9 [WT and M1–M11] (Fig. 3A). The results show that none of these components, including even the direct binding partners (i.e., Med4 and Med7), were co-purified with FLAG-Med9 [M6, M7, and M8] (Fig. 3A, lanes 8–10), implying that these Med9 mutant proteins could not be incorporated into Mediator. On the other hand, these Mediator components could be co-purified in equivalent (M1) (lane 3), lower (M2, M3, M9–11) (lanes 4, 5, 11–13), or minimal (M4, M5) (lanes 6, 7) amounts compared with the WT (lane 2). Thus, it is likely that the N region of Med9 is not necessary (compare WT and M1), but that either the S2 or C region is essential (compare M4, M5 and M6), for the incorporation of Med9 into the Mediator complex. Furthermore, the S1 and M regions probably play certain roles in the assembly of Mediator, as their deletions decreased the amounts of co-purified Mediator components (compare M1, M2, M9, and M10). Consistent with previous studies (Koleske & Young 1994; Liu et al. 2001), RNA polymerase II (Rpb3), TFIIF (Tfg2), and TFIIB (Sua7) were co-purified with FLAG-Med9 [WT]. The binding profiles of these factors were similar to those of other components of Mediator.
med19 reportedly does not affect the integrity of the Middle module itself, but destabilizes its attachment to the rest of the Mediator complex, thereby generating Middle-less (i.e. Head–Tail) Mediator (Baidoobonso et al. 2007). To investigate whether med9 mutations affect the integrity of Mediator similarly or differently, we compared the amounts of several components of the Tail (Med2 and Med15), Head (Med6), and Middle (Med4, Med7 and Med9) modules that could be co-immunoprecipitated with HA-tagged Med17 (Head) from the cell extracts of MED9 (WT) or med9 (med9 and M1–M11) strains (Fig. 3B). Consistent with the results described in Fig. 3A, only minor effects were observed for M1 (compare lanes 2 and 3). However, most of the Mediator components, except Med6 (Head), were barely detectable for other med9 mutants (lanes 1 and 4–13). These results suggest that although the Head module itself is mostly intact, it is easily dissociated from the Tail and Middle modules in cell extracts of med9 strains, with the exception of M1.
In Fig. 3A, signals for the Head and Tail components were detected more easily when Mediator was purified on the basis of its affinity for Med9 [M2–M5 and M9–M11]. This apparent discrepancy could be due to the more harsh conditions used in the experiments in Fig. 3B (e.g., the buffer contained detergent and a higher salt concentration). In any case, the observation that the Tail components were barely detectable in the co-purified fraction with Med17 in most med9 mutants (Fig. 3B) was unexpected, as the Head–Tail Mediator generated by med19 is so stable that it is resistant even to 1 M urea treatment (Baidoobonso et al. 2007). Therefore, we propose that, for some as yet unknown reason, a stable connection between the Tail and Head modules can be formed or maintained when Med19 (thereby the entire Middle module) is lost, but not when the S1-M-S2-C region of Med9 is mutated.
Finally, two-dimensional blue-native (BN) electrophoresis analyses (Wittig et al. 2006) were performed to inspect further the structural integrity of Mediator. Mediator that had been purified on the basis of its affinity for Med9 [WT, M1, M2, M5 or M9–11] was resolved in native form in the first dimension, subjected to SDS-PAGE in the second dimension to denature the complexes, and immunoblotted to identify the positions of the components of the Tail (Med2 and Med15), Head (Med6, Med11, Med17, Med20 and Med22), and Middle (Med4, Med7, Med9 and Med19) modules (Fig. 4). At least four distinct complexes with different mobilities in the first dimension were apparent (I, II, III, and IV) when Mediator [WT] was analyzed by this method (Fig. 4A). Judging from the polypeptide composition of each complex, we propose that complexes I and IV represent the intact Mediator and the Middle module, respectively, and that the latter may dissociate from the former during BN-PAGE. Interestingly, complex II seemed to be defective in the Tail module, as it contained Med15 but not Med2. As Med2 migrated faster (complex III) than Med15, it is likely that the dissociation of complex III from the intact Mediator (complex I) may generate complex II during BN-PAGE.
|
Essential roles of Med9 in transcriptional activation and pre-initiation complex assembly in vitro
Med9 is reportedly required for transcriptional activation in vivo by Bas2 (Han et al. 1999) and Met4 (Leroy et al. 2006). However, it still remains unclear whether Med9 is required for transcriptional activation in vitro. One study using an in vitro transcription system reconstituted with pure general transcription factors and holo-pol II showed that Med9 is not required for transcriptional activation by Gal4-VP16 and Gcn4 (Han et al. 1999). However, another study using crude nuclear extracts showed that Med9 is essential for enhanced levels of transcription in the presence of Gal4-VP16 (Liu et al. 2001). Here, we sought to examine whether Med9 is required for transcriptional activation by Gal4-VP16 in an in vitro transcription system using whole cell extracts (Woontner et al. 1991) and an immobilized template (Ranish et al. 1999) (Fig. 5A). Our results clearly showed that transcriptional activation by Gal4-VP16 was dependent on Med9 (compare lanes 1–4). We also found that this activation was decreased by c. 50% in the med9 [M1] mutant (lanes 5–6) and was reduced to background levels in the other med9 mutants [M2–M11] (compare lanes 1–2 and lanes 7–26). Thus, we conclude that the N and S1-M-S2-C regions of Med9 play an ancillary or obligatory role, respectively, in transcriptional activation by Gal4-VP16, at least in our in vitro transcription system.
|
Consistent with the transcriptional activities (Fig. 5A), PIC assembly was almost normal in M1 (Fig. 5B, lanes 5, 6), but was defective in other med9 mutants (lanes 1–2 and 7–26). For instance, only the binding of TBP, Spt3, Med15, and Med17 was evident in med9 (lanes 1, 2). Surprisingly, TBP binding was constitutive, as previously described for the WT (Ranish et al. 1999; Yudkovsky et al. 2000), although the binding of the other three factors was activator-dependent (lanes 1, 2). Similar or less pronounced phenotypes to that of med9 were observed for most of other med9 mutants [M2–M8 (lanes 7–20) and M11 (lanes 25, 26)]. Intriguingly, binding of nearly all of the transcription factors was activator-dependent in M9, as seen for the WT, except for TBP and Rpb3 which showed constitutive and almost no binding, respectively (lanes 21, 22). The lack of Rpb3 (pol II) binding might explain the low transcriptional activity in this mutant (Fig. 5A), although it is rare that TFIIF can be assembled into PIC without the incorporation of pol II (Baek et al. 2006). M10 showed defective levels of PIC assembly that were between those of M9 and the other med9 mutants (lanes 23, 24).
We next asked whether the defects in PIC assembly observed for most of the med9 mutants (e.g. the constitutive binding of TBP and the severely weakened binding of Taf11, Rpb3, Sua7, Med4, and Tfg2) were specific to med9 or were caused more generally by the loss of Mediator activity. To address this issue, we removed Mediator from the WT cell extracts by immunodepletion of Med9 and Med17 (Fig. 5C, Med), and then examined PIC assembly (Fig. 5D) and transcription activities (Fig. 5E) of Med extracts in the presence or absence of immunopurified Mediator [WT or M5] (Fig. S2 in Supporting Information).
First, we found that the defects of PIC assembly in Med extracts (Fig. 5D) were different from those of med9 or other med9 extracts (Fig. 5B), although they showed similar impairment of activator-dependent transcription (Fig. 5A,E). Namely, Med could support activator-dependent assembly of TBP and Taf11, but not those of Rpb3, Sua7, Med4, or Tfg2 (Fig. 5D, lanes 3, 4). This result is consistent with a previous study using extracts of Head module mutants (i.e. med17, med18 and med20), which showed that Mediator is not required for the assembly of TFIIA and TFIID, but is required for the assembly of holoenzyme components (Ranish et al. 1999).
Second, as expected, the transcriptional activity of the Med extract could be restored by adding back Mediator [WT] (Fig. 5E, lanes 5, 6), but not by adding Mediator [M5] (Fig. 5E, lanes 7, 8). Similarly, the defects in PIC assembly in the Med extract could be restored by Mediator [WT] (Fig. 5D, lanes 5, 6), but not by Mediator [M5] (Fig. 5D, lanes 7, 8). Notably, even after adding back excess amounts of Mediator [M5] to the Med extracts, it was impossible to observe the original defects in PIC assembly caused by M5 (data not shown). Therefore, immunopurified Mediator [M5] does not carry any specifically impaired or dominantly acting function. Rather, it has no apparent Mediator activity, at least when tested in the in vitro system used here. Collectively, we postulate that some as yet unidentified factor(s) other than Mediator itself may be expressed specifically in med9 or in most of med9 strains, generating med9-specific defects in PIC assembly, such as constitutive TBP binding and severely weakened Toa1/Taf11 binding.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Third, we showed that BN-PAGE is a useful technique for analyzing Mediator integrity (Fig. 4). In fact, we found that while the absence of the N (aa 1–63) or S1 (aa 64–80) region alone does not affect the integrity of the Tail module, the absence of the N-S1 (aa 1–80), M (aa 81–101), S2 (aa 102–120), or C (aa 121–149) region destabilizes it (Fig. 4B). Note that these mutations particularly destabilize a submodule structure in the Tail, that is, the Med2–Med3–Med15 triad, which has been shown to recruit TBP and pol II and activate the ARG1 promoter by binding to Gcn4 (Zhang et al. 2004). Assuming that Med2–Med3 or Med2–Med3–Med5–Med15 can be expressed as soluble recombinant complexes in insect cells and that Med2–Med3–Med15 cannot (Beve et al. 2005), Med2 and Med15 may actually exist in two distinct, easily dissociable submodules in the Tail, thereby allowing Med2 to migrate faster than Med15 in BN-PAGE (Fig. 4B). In addition, the absence of any region (N, S1, M, S2 or C) in Med9 destabilized the peripheral structure of the Head module (i.e., caused the dissociation of Med6 and Med20), but not the minimal Head complex (Med11, Med17, and Med22) (Takagi et al. 2006) (Fig. 4C). These observations indicate that certain defects of Med9 in the Middle module could destabilize the integrity of the Head and/or Tail modules. This appears to be contrary to the previous report that the entire Middle module is not necessary to make a stable Head–Tail structure in the med19 strain (Baidoobonso et al. 2007). However, another report demonstrated that lower amounts of Med6 are precipitated with Med17 in the med3 or med16 strain, indicating that the Head structure is destabilized by mutations of the Tail module (Reeves & Hahn 2003). Hence, we postulate that although Med9 may be located at the periphery of the Middle module, certain defects in, or the absence of Med9 could alter the Middle module so that it acts dominantly to destabilize other parts of Mediator.
Fourth, we found that the S1-M-S2-C (aa 64–149) region of Med9 is essential for transcriptional activation in vitro by Gal4-VP16 (Fig. 5A). This finding seems to contradict a previous conclusion that Med9 is not required for activation by Gal4-VP16 (Han et al. 1999). This discrepancy may be attributed to differences in the transcription systems used in these two studies, such as pure reconstitution (Han et al. 1999) vs. crude (Fig. 5A) systems. It is well-known that the requirements of human Mediator for basal and activated transcription differ depending on transcription systems. In a purified system reconstituted with GTFs, pol II, and PC4, Mediator is not essential and only enhances transcription levels. However, transcription is absolutely dependent on Mediator in crude extracts, as shown by immunodepletion experiments (Mittler et al. 2001; Baek et al. 2002). In humans, Gdown1 (Hu et al. 2006) and DSIF (hSpt4–hSpt5 heterodimer) (Malik et al. 2007) have been identified as factors that regulate the dependence on Mediator in crude transcription systems. Despite the absence of a clear counterpart of Gdown1 in yeast, its functional homologue and/or Spt4–Spt5 may be responsible for observed differences in the roles of Med9 in transcriptional activation by Gal4-VP16.
Fifth, in vitro experiments using an immobilized template also showed that Med9 is required for activator-dependent recruitment of TFIIA, TFIIB, Taf11/TFIID, TFIIE, TFIIF and pol II, but not of Spt3 (SAGA) (Fig. 5B). Interestingly, the activator-dependent recruitment of Med4 (Middle) only occurred when Med9 was intact or present as the M1 or M9 mutant form. In contrast, the activator-dependent recruitment of Med17 (Head) and Med15 (Tail) occurred even when Med9 was entirely absent (). Thus, a partial Mediator must be recruited to the template by an activator in some med9 extracts. This is analogous to a previous observation that Med17 (Head) is recruited normally, even though Med3 (Tail) and Med15 (Tail) are not, in the absence of Med16 (Tail) (Reeves & Hahn 2003). On the other hand, although the recruitment of TBP is activator-dependent in the WT extract, it is activator-independent when Med9 is absent () or impaired (M2–M11). Before discussing the reason for this puzzling effect of med9, it is necessary to outline several differences between the results obtained here and those obtained in previous studies.
In previous studies using yeast or human nuclear extracts, significant amounts of TBP were shown to bind constitutively to the promoter, that is, in a manner independent of the TATA element and/or an activator (Ranish et al. 1999; Yudkovsky et al. 2000; Baek et al. 2006). In addition, and contrary to our study, Med17 and Med15 were not recruited to the promoter in the med9 nuclear extract (Liu et al. 2001). Furthermore, the observation that a mutation of the Head module (med20) decreased the recruitment of Ada1 (SAGA) by Gal4-AH (Kim et al. 2007) is in stark contrast to our observation that med9 does not affect the recruitment of Spt3 (SAGA) by Gal4-VP16. Although the reasons for these apparent discrepancies remain unclear, they may be ascribed to differences in the cell-free extracts (i.e., nuclear extracts vs. whole-cell extracts), the number of binding sites for activators on the template (i.e., one vs. three), and/or the sort of activator used (i.e. Gal4-AH vs. Gal4-VP16). In any case, our observation that Med9 is essential for the activator-dependent recruitment of TFIIA and TFIID argues against the prevailing view that Mediator is dispensable for such recruitment (Ranish et al. 1999; Yudkovsky et al. 2000).
The immunodepletion and add-back experiments (Fig. 5C,D) partly addressed these issues, for example, that in some med9 extracts, TBP binds to the template even in the absence of an activator and, conversely, that Taf11 does not bind to the template even in the presence of an activator. These uncommon defects were not observed in the WT extract depleted of the entire Mediator (Med) by immunodepletion, and were not reproduced even when the Mediator containing the Med9 [M5] protein was added back. Namely, TBP and Taf11 bind to the template in a manner that depends on an activator but not on Mediator, at least in the Med extract. Thus, we conclude that the aforementioned prevailing view could still be effective under our experimental conditions, and that certain unidentified factor(s) specifically present in certain med9 extracts may be primarily responsible for such uncommon defects of med9 extracts.
Finally, the in vivo expression data showed that the N (aa 1–63) region represses HXK1 expression, whereas the S2-C region (aa 102–149) is required for its enhanced expression (Fig. 1C). The S2-C region is also important for the expression of HIS4 and ARG4. In general, these results are consistent with those obtained in the in vitro studies described above. However, there remain differences between the in vivo and in vitro analyses; for instance, although Med9 is dispensable for ADH1 expression in vivo (Fig. 1C), it is essential for transcriptional activation by Gal4-VP16 in vitro (Fig. 5A). Therefore, requirements for Med9 function may be altered depending on the assay conditions, for example, in vivo vs. in vitro, and/or on differences in the target genes. In fact, there are precedents for differences in the requirements for Mediator function. For instance, although Med16 (Tail) is required for activation by Gcn4 in vitro (Myers et al. 1999), and med16 decreases Gcn4-dependent HIS4 expression in vivo (Jiang & Stillman 1995), med16 increases the expression of another Gcn4-dependent gene, HIS3, in vivo (Jiang & Stillman 1995). Similarly, med6 (Head) decreases activation by Gal4-VP16 in vitro as well as the expression of GAL1, PYK1, and SUC2, but does not decrease the expression of TRP3 or MED6, in vivo (Lee et al. 1997). Recent studies have shown that med6 (Head) affects activation by Gal4 and Met4, but not by Gcn4, whereas med18 (Head) affects activation by Gcn4, but not by Gal4 or Met4 (Swanson et al. 2003; Leroy et al. 2006). These results argue that different activators may depend on the function of different subunits of Mediator, especially in vivo. Furthermore, it has also been suggested that the same activator may use different functions of Mediator, depending on the promoter structure (Leroy et al. 2006).
A recent study identified three highly conserved regions (SSM; signature sequence motif) in Med9 among opisthokonts: SSM#1 (aa 64–76), SSM#2 (aa 86–109), and SSM#3 (aa 126–143) (Bourbon 2008). SSM#1 and #3 are included in the S1 (aa 64–80) and C (aa 121–149) regions, respectively, while SSM#2 overlaps with both the M (aa 81–101) and S2 (aa 102–120) regions (Fig. S1 Supporting Information). The functional significance of these three SSMs is well-supported by our results, especially those obtained with the in vitro immobilized template assays where only the N (aa 1–63) region was dispensable for transcriptional activation and PIC assembly. As an absence of the N region decreased transcription in vitro (c. 50%) and also dramatically increased HXK1 expression in vivo, this species-specific region may play a regulatory role in Mediator-dependent transcription.
In summary, this study has revealed a modular structure for Med9. The species-specific N region plays a regulatory role, while the highly conserved S1-M-S2-C region is equipped with a more fundamental function that involves direct binding to Med4-N and Med7-N, assembly into the Middle module and Mediator, and transcriptional activation in vitro and in vivo. Assuming that Mediator lacking only the Med9 protein could not be detected, and that the integrity of other modules are also affected by med9, the defects described in this study may include not only the direct effects of med9, but also indirect effects, such as the destabilization of the entire Mediator. To determine the roles of Med9 in transcription more precisely, further studies should be carried out to identify the factor(s) present specifically in certain med9 extracts and to investigate their functional relationships with Mediator.
|
Experimental procedures |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Standard techniques were used to grow and transform yeast (Amberg et al. 2005). The yeast strains used in this study are listed in Table S1 in Supporting Information. Construction of the yeast strains and plasmids is described in Appendix S1 in Supporting Information.
Preparation of whole-cell extract (WCE)
Large-scale WCE used for immobilized template assays and immunopurification of FLAG-tagged Mediator was prepared according to the method of Woontner et al. (1991) with some modifications as described in Appendix S1 in Supporting Information.
To prepare small-scale WCE for immunoprecipitation of HA-tagged Mediator, yeast strains were grown in 100 mL of SC medium (-uracil) at 25 °C to an OD600 value of 1.5, harvested by centrifugation, and washed once with lysis buffer B (20 mM Hepes-KOH [pH 7.6], 150 mM potassium acetate, 20% [v/v] glycerol, 3 mM EDTA, 3 mM DTT). Collected cells were resuspended in 800 µL of lysis buffer B supplemented with 1 mM PMSF. Cell suspensions were transferred to a 2 mL plastic tube suited for a Multi-Beads Shocker (Yasui Kikai), and the tube was filled with glass beads (Sigma #G-8772). The cells were vigorously shaken for 60 s at 2700 rpm, followed by 60 s of rest at 4 °C. After 20 cycles, the cell lysate was removed and clarified by centrifugation at 18 800 g for 45 min. The supernatant (WCE) was used for immunoprecipitation.
Immobilized template assays
Immobilized template assays were conducted according to the method of Ranish et al. (1999) with some modifications as described in Appendix S1 in Supporting Information.
Immunoblot analyses and antibodies
Two dimensional blue-native PAGE (BN-PAGE) (Wittig et al. 2006) was conducted with precast Native PAGE 3%–12% gel in the first dimension and precast NuPAGE 4%–12% Bis-Tris Zoom gel in the second dimension, according to manufacturer's instructions (Invitrogen). Immunoblotting and preparation of the polyclonal antibodies directed against Spt3, Taf11, and TBP were described previously (Tsukihashi et al. 2000; Takahata et al. 2004). Polyclonal antibodies directed against Toa1 (aa 1 to 286), Sua7 (aa 1 to 345), Med4 (aa 1 to 284), Med6 (aa 1 to 295), Med7 (aa 1 to 222), Med9 (aa 1 to 149), Med11 (aa 1 to 132), Med15 (aa 1 to 300), Med17 (aa 1 to 300), Med19 (aa 1 to 220), Med20 (aa 1 to 210), Med22 (aa 1 to 121), Cdk8 (aa 1 to 555), Rpb3 (aa 1 to 318), Tfa1 (aa 1 to 482), and Tfg2 (aa 1 to 400) were raised in rabbits using recombinant, gel-purified His-tagged polypeptides expressed in E. coli as antigens.
Antibodies against Med2 (#sc-28058) and HA (#sc-7392) epitopes were purchased from Santa Cruz Biotechnology, Inc. Antibodies against FLAG epitope (#F3165) were purchased from Sigma. Antibodies against GST (#RPN1236) and His (#27-4710-01) epitopes were purchased from GE Healthcare.
Expression of recombinant proteins in bacterial cells
Gal4-VP16, Med4, Med7, Med9 and GST-Med9 were produced as recombinant proteins in bacterial cells as described in Appendix S1 in Supporting Information.
GST pull-down assays
To study the interactions between Med9 (WT or mutant), Med4 (WT or mutant), and Med7 (mutant), purified His-Med4 or His-Med7 was mixed with a bacterial lysate containing GST-Med9 (mutant) or GST, in 150 µL of buffer F (20 mM Hepes-KOH [pH 7.6], 150 mM KCl, 1 mM MgCl2, 1 mM EDTA, 10% [v/v] glycerol, 1 mM DTT), and then incubated at 4 °C for 30 min. Following the addition of 10 µL glutathione-SepharoseTM 4B (GE Healthcare) and 0.1% NP-40, the incubation was continued for a further 30 min, after which the beads were washed three times with 400 µL of buffer F containing 0.1% NP-40. The beads were boiled in SDS sample buffer to elute the bound protein, and the eluates were separated using 12.5% (Med4/Med9) or 15% (Med7/Med9) SDS-PAGE, followed by immunoblotting with polyclonal anti-His, anti-GST, anti-Med4, and anti-Med7 antibodies.
Immunopurification, immunoprecipitation and immunodepletion
Immunopurification, immunoprecipitation and immunodepletion were performed as described in Appendix S1 in Supporting Information.
Northern blot analyses
Northern blot analyses of several endogenous genes were performed as described previously (Tsukihashi et al. 2000). To detect HIS4, ARG4, HXK1, PGK1, and ADH1, DNA fragments were PCR-amplified from yeast genomic DNA, purified, and 32P-labeled using random priming. The PCR primers used for PGK1 and ADH1 were described previously (Tsukihashi et al. 2000). Other primer pairs used included ARG4, TK6813/TK6814; HIS4, TK3483/TK3484; HXK1, TK10537/TK10538.
|
Acknowledgements |
|---|
This work was supported by grants from the Japan Society for the Promotion of Science, the Ministry of Education, Culture, Sports, Science and Technology of Japan, CREST of the Japan Science and Technology Corporation.
|
Footnotes |
|---|
* Correspondence: kokubo{at}tsurumi.yokohama-cu.ac.jp
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Baek, H.J., Kang, Y.K. & Roeder, R.G. (2006) Human Mediator enhances basal transcription by facilitating recruitment of transcription factor IIB during preinitiation complex assembly. J. Biol. Chem. 281, 15172–15181.
Baek, H.J., Malik, S., Qin, J. & Roeder, R.G. (2002) Requirement of TRAP/mediator for both activator-independent and activator-dependent transcription in conjunction with TFIID-associated TAF(II)s. Mol. Cell. Biol. 22, 2842–2852.
Baidoobonso, S.M., Guidi, B.W. & Myers, L.C. (2007) Med19(Rox3) regulates Intermodule interactions in the Saccharomyces cerevisiae mediator complex. J. Biol. Chem. 282, 5551–5559.
Baumli, S., Hoeppner, S. & Cramer, P. (2005) A conserved mediator hinge revealed in the structure of the MED7.MED21 (Med7.Srb7) heterodimer. J. Biol. Chem. 280, 18171–18178.
Beve, J., Hu, G.Z., Myers, L.C., Balciunas, D., Werngren, O., Hultenby, K., Wibom, R., Ronne, H. & Gustafsson, C.M. (2005) The structural and functional role of Med5 in the yeast Mediator tail module. J. Biol. Chem. 280, 41366–41372.
Bourbon, H.M. (2008) Comparative genomics supports a deep evolutionary origin for the large, four-module transcriptional mediator complex. Nucleic Acids Res. 36, 3993–4008.
Conaway, R.C., Sato, S., Tomomori-Sato, C., Yao, T. & Conaway, J.W. (2005) The mammalian Mediator complex and its role in transcriptional regulation. Trends Biochem. Sci. 30, 250–255.[CrossRef][Medline]
Gromoller, A. & Lehming, N. (2000) Srb7p is essential for the activation of a subset of genes. FEBS Lett. 484, 48–54.[Medline]
Guermah, M., Tao, Y. & Roeder, R.G. (2001) Positive and negative TAF(II) functions that suggest a dynamic TFIID structure and elicit synergy with traps in activator-induced transcription. Mol. Cell. Biol. 21, 6882–6894.
Guglielmi, B., van Berkum, N.L., Klapholz, B., Bijma, T., Boube, M., Boschiero, C., Bourbon, H.M., Holstege, F.C. & Werner, M. (2004) A high resolution protein interaction map of the yeast Mediator complex. Nucleic Acids Res. 32, 5379–5391.
Han, S.J., Lee, J.S., Kang, J.S. & Kim, Y.J. (2001) Med9/Cse2 and Gal11 modules are required for transcriptional repression of distinct group of genes. J. Biol. Chem. 276, 37020–37026.
Han, S.J., Lee, Y.C., Gim, B.S., Ryu, G.H., Park, S.J., Lane, W.S. & Kim, Y.J. (1999) Activator-specific requirement of yeast mediator proteins for RNA polymerase II transcriptional activation. Mol. Cell. Biol. 19, 979–988.
Hu, X., Malik, S., Negroiu, C.C., Hubbard, K., Velalar, C.N., Hampton, B., Grosu, D., Catalano, J., Roeder, R.G. & Gnatt, A. (2006) A Mediator-responsive form of metazoan RNA polymerase II. Proc. Natl. Acad. Sci. USA 103, 9506–9511.
Jiang, Y.W. & Stillman, D.J. (1995) Regulation of HIS4 expression by the Saccharomyces cerevisiae SIN4 transcriptional regulator. Genetics 140, 103–114.[Abstract]
Johnson, K.M. & Carey, M. (2003) Assembly of a Mediator/TFIID/TFIIA Complex Bypasses the Need for an Activator. Curr. Biol. 13, 772–777.[CrossRef][Medline]
Johnson, K.M., Wang, J., Smallwood, A., Arayata, C. & Carey, M. (2002) TFIID and human mediator coactivator complexes assemble cooperatively on promoter DNA. Genes Dev. 16, 1852–1863.
Kang, J.S., Kim, S.H., Hwang, M.S., Han, S.J., Lee, Y.C. & Kim, Y.J. (2001) The structural and functional organization of the yeast mediator complex. J. Biol. Chem. 276, 42003–42010.
Kim, B., Nesvizhskii, A.I., Rani, P.G., Hahn, S., Aebersold, R. & Ranish, J.A. (2007) The transcription elongation factor TFIIS is a component of RNA polymerase II preinitiation complexes. Proc. Natl. Acad. Sci. USA 104, 16068–16073.
Koleske, A.J. & Young, R.A. (1994) An RNA polymerase II holoenzyme responsive to activators. Nature 368, 466–469.[CrossRef][Medline]
Kornberg, R.D. (2005) Mediator and the mechanism of transcriptional activation. Trends Biochem. Sci. 30, 235–239.[CrossRef][Medline]
Kotani, T., Banno, K., Ikura, M., Hinnebusch, A.G., Nakatani, Y., Kawaichi, M. & Kokubo, T. (2000) A role of transcriptional activators as antirepressors for the autoinhibitory activity of TATA box binding of transcription factor IID. Proc. Natl. Acad. Sci. USA 97, 7178–7183.
Lariviere, L., Seizl, M., van Wageningen, S., Rother, S., van de Pasch, L., Feldmann, H., Strasser, K., Hahn, S., Holstege, F.C. & Cramer, P. (2008) Structure-system correlation identifies a gene regulatory Mediator submodule. Genes Dev. 22, 872–877.
Lee, Y.C., Min, S., Gim, B.S. & Kim, Y.J. (1997) A transcriptional mediator protein that is required for activation of many RNA polymerase II promoters and is conserved from yeast to humans. Mol. Cell. Biol. 17, 4622–4632.
Lee, Y.C., Park, J.M., Min, S., Han, S.J. & Kim, Y.J. (1999) An activator binding module of yeast RNA polymerase II holoenzyme. Mol. Cell. Biol. 19, 2967–2976.
Leroy, C., Cormier, L. & Kuras, L. (2006) Independent recruitment of mediator and SAGA by the activator Met4. Mol. Cell. Biol. 26, 3149–3163.
Li, Y., Bjorklund, S., Jiang, Y.W., Kim, Y.J., Lane, W.S., Stillman, D.J. & Kornberg, R.D. (1995) Yeast global transcriptional regulators Sin4 and Rgr1 are components of mediator complex/RNA polymerase II holoenzyme. Proc. Natl. Acad. Sci. USA 92, 10864–10868.
Liu, Y., Ranish, J.A., Aebersold, R. & Hahn, S. (2001) Yeast nuclear extract contains two major forms of RNA polymerase II mediator complexes. J. Biol. Chem. 276, 7169–7175.
Malik, S., Barrero, M.J. & Jones, T. (2007) Identification of a regulator of transcription elongation as an accessory factor for the human Mediator coactivator. Proc. Natl. Acad. Sci. USA 104, 6182–6187.
Mittler, G., Kremmer, E., Timmers, H.T. & Meisterernst, M. (2001) Novel critical role of a human Mediator complex for basal RNA polymerase II transcription. EMBO Rep. 2, 808–813.[CrossRef][Medline]
Myers, L.C., Gustafsson, C.M., Hayashibara, K.C., Brown, P.O. & Kornberg, R.D. (1999) Mediator protein mutations that selectively abolish activated transcription. Proc. Natl. Acad. Sci. USA 96, 67–72.
Ranish, J.A., Yudkovsky, N. & Hahn, S. (1999) Intermediates in formation and activity of the RNA polymerase II preinitiation complex: holoenzyme recruitment and a postrecruitment role for the TATA box and TFIIB. Genes Dev. 13, 49–63.
Reeves, W.M. & Hahn, S. (2003) Activator-independent functions of the yeast mediator sin4 complex in preinitiation complex formation and transcription reinitiation. Mol. Cell. Biol. 23, 349–358.
Swanson, M.J., Qiu, H., Sumibcay, L., Krueger, A., Kim, S.J., Natarajan, K., Yoon, S. & Hinnebusch, A.G. (2003) A multiplicity of coactivators is required by Gcn4p at individual promoters in vivo. Mol. Cell. Biol. 23, 2800–2820.
Takagi, Y. & Kornberg, R.D. (2006) Mediator as a general transcription factor. J. Biol. Chem. 281, 80–89.
Takagi, Y., Calero, G., Komori, H., Brown, J.A., Ehrensberger, A.H., Hudmon, A., Asturias, F. & Kornberg, R.D. (2006) Head module control of mediator interactions. Mol. Cell 23, 355–364.[CrossRef][Medline]
Takahata, S., Kasahara, K., Kawaichi, M. & Kokubo, T. (2004) Autonomous function of the amino-terminal inhibitory domain of TAF1 in transcriptional regulation. Mol. Cell. Biol. 24, 3089–3099.
Thomas, M.C. & Chiang, C.M. (2006) The general transcription machinery and general cofactors. Crit. Rev. Biochem. Mol. Biol. 41, 105–178.[CrossRef][Medline]
Tomomori-Sato, C., Sato, S., Parmely, T.J., Banks, C.A., Sorokina, I., Florens, L., Zybailov, B., Washburn, M.P., Brower, C.S., Conaway, R.C. & Conaway, J.W. (2004) A mammalian mediator subunit that shares properties with Saccharomyces cerevisiae mediator subunit Cse2. J. Biol. Chem. 279, 5846–5851.
Tsukihashi, Y., Miyake, T., Kawaichi, M. & Kokubo, T. (2000) Impaired core promoter recognition caused by novel yeast TAF145 mutations can be restored by creating a canonical TATA element within the promoter region of the TUB2 gene. Mol. Cell. Biol. 20, 2385–2399.
van de Peppel, J., Kettelarij, N., van Bakel, H., Kockelkorn, T.T., van Leenen, D. & Holstege, F.C. (2005) Mediator expression profiling epistasis reveals a signal transduction pathway with antagonistic submodules and highly specific downstream targets. Mol. Cell 19, 511–522.[CrossRef][Medline]
Wittig, I., Braun, H.P. & Schagger, H. (2006) Blue native PAGE. Nat. Protoc. 1, 418–428.[CrossRef][Medline]
Woontner, M., Wade, P.A., Bonner, J. & Jaehning, J.A. (1991) Transcriptional activation in an improved whole-cell extract from Saccharomyces cerevisiae. Mol. Cell. Biol. 11, 4555–4560.
Xiao, Z., McGrew, J.T., Schroeder, A.J. & Fitzgerald-Hayes, M. (1993) CSE1 and CSE2, two new genes required for accurate mitotic chromosome segregation in Saccharomyces cerevisiae. Mol. Cell. Biol. 13, 4691–4702.
Yudkovsky, N., Ranish, J.A. & Hahn, S. (2000) A transcription reinitiation intermediate that is stabilized by activator. Nature 408, 225–229.[CrossRef][Medline]
Zhang, F., Sumibcay, L., Hinnebusch, A.G. & Swanson, M.J. (2004) A triad of subunits from the Gal11/tail domain of Srb mediator is an in vivo target of transcriptional activator Gcn4p. Mol. Cell. Biol. 24, 6871–6886.
Received: 1 September 2008
Accepted: 7 October 2008
This article has been cited by other articles:
|
|
 |
|
  T. Koschubs, K. Lorenzen, S. Baumli, S. Sandstrom, A. J. R. Heck, and P. Cramer Preparation and topology of the Mediator middle module Nucleic Acids Res., January 31, 2010; (2010): gkq029v1 - gkq029. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
