|Home | About | Journals | Submit | Contact Us | Français|
The basal RNA polymerase II (RNAPII) transcription machinery is composed of RNAPII and the general transcription factors (TF) TATA binding protein (TBP), TFIIB, TFIIE, TFIIF and TFIIH. Due to the powerful genetic and molecular approaches that can be utilized, the budding yeast S. cerevisiae has proven to be an invaluable model system for studies of the mechanisms of RNAPII transcription. Complementary biochemical studies of the S. cerevisiae basal transcription machinery, however, have been hampered by difficulties in the purification of TFIIF and TFIIH, most notably due to the severe toxicity of the TFIIF Tfg1 subunit in E. coli and the complexity of the purification scheme for native TFIIH. Here, we report the elimination of TFG1-associated toxicity in E. coli, the identification and removal of a functional E.coli promoter and internal translation initiation within the N-terminal coding region of TFG1, and the efficient production and two step purification of recombinant TFIIF complexes. We also report conditions for the efficient two step tandem affinity purification (TAP) of holo-TFIIH, core TFIIH and TFIIK complexes from yeast whole cell extracts.
The general transcription factors TFIIF and TFIIH play essential roles in transcription of eukaryotic protein encoding genes by RNA polymerase II (RNAPII). Studies in both mammalian cells and the budding yeast S. cerevisiae have demonstrated multiple roles for TFIIF during the transcription cycle that include (i) recruitment of RNAPII, TFIIE and TFIIH into the preinitiation complex (PIC) [1, 2]; (ii) wrapping of promoter DNA around RNAPII in the PIC ; (iii) modulating transcription start site utilization [4, 5]; (iv) facilitating efficient promoter escape of RNAPII [6, 7]; (iv) increasing the rate of transcript elongation [8, 9]; (v) stimulating the rescue of paused elongation complexes in conjunction with TFIIS ; and (vi) stimulating the RNAPII C-terminal domain (CTD) phosphatase Fcp1 to facilitate recycling of RNAPII back into the initiation-competent hypo-phosphorylated form . Similarly, the TFIIH complex plays several important roles that include (i) unwinding of promoter DNA through multiple ATPase/helicase activities [12–19]; (ii) phosphorylation of the RNAPII C-terminal domain (CTD) during the transition from initiation to transcript elongation [20, 21]; and (iii) participating in nucleotide excision repair of damaged DNA [22–24].
S. cerevisiae TFIIF comprises three protein subunits designated Tfg1, Tfg2, and Tfg3. The Tfg1 (calculated MW ~82 kDa) and Tfg2 (calculated MW ~47 kDa) subunits are homologous to the RAP74 and RAP30 subunits of mammalian TFIIF, respectively, and are essential for cell viability . The Tfg3 subunit, also designated as Taf14, is associated with at least six yeast nuclear complexes, has no known homologues in higher eukaryotes, and is not required for viability [25–32]. The S. cerevisiae holo-TFIIH complex contains ten polypeptide subunits that include Ssl2, the core TFIIH sub-complex comprising Rad3, Tfb1, Tfb2, Ssl1, Tfb4 and Tfb5, and the TFIIK sub-complex, which contains Ccl1, Tfb3 and Kin28 [12, 33–35] (Fig. 3A).
Despite the powerful genetic approaches that can be utilized in studies with S. cerevisiae, comprehensive structure-function analyses of yeast TFIIF and TFIIH have been hindered by difficulties in the production of the recombinant proteins and/or purification of the native complexes from yeast cells. Although Holo-TFIIH has been purified from a strain containing a hexahistidine-tagged Tfb1 subunit, the purification scheme nonetheless involves five to six chromatographic steps to reach apparent homogeneity [36, 37]. For TFIIF, native Tfg1-Tfg2-Tfg3 complexes can be isolated from yeast extract using tandem affinity purification (TAP), but additional purification is required due to contamination with RNAPII . Importantly, the extreme toxicity of plasmids containing the Tfg1 subunit in E.coli has presented a major obstacle in the production of recombinant TFIIF. We previously reported a relatively inefficient approach for the production of functional recombinant Tfg1-Tfg2 complexes that involved a low-copy vector, induction of Tfg1, Tfg2 co-expression by infection with bacteriophage containing T7 RNA polymerase, and multi-step affinity purification . In the work presented here, we report the elimination of TFG1-associated toxicity in E. coli, the identification of a functional E.coli promoter and internal translation initiation within the N-terminal coding region of TFG1, and two-step purification of recombinant Tfg1-Tfg2 complexes. We also report conditions for efficient TAP-purification of TFIIH complexes from yeast whole cell extracts.
The initial dual-expression plasmid used in these studies (pDt/g1g2) was constructed using pCOLADuet-1 (Novagen) as the starting vector and contains the coding regions for the S. cerevisiae TFIIF Tfg1 and Tfg2 subunits (Tfg2p: myc-epitope tag at the C-terminus; Tfg1p: hexahistidine and myc-epitope tags at the N-terminus, FLAG and TAP tags at the C-terminus) under control of the T7 promoters in the adjacent MCS1 and MCS2 expression cassettes, respectively. The final dual-expression plasmid (pDt/g1*g2) was generated from pDt/g1g2 and contains the base substitutions indicated as mutant 2 (M2) in Figure 1C. S. cerevisiae strains YT062 and YT063 have been described previously  and contain TAP-tagged versions of the TFIIH subunits Tfb3 and Tfb4, respectively.
Overnight cultures of E. coli strain Rosetta™ (DE3) (Novagen) harboring pDt/g1g2 or the subsequent M1 or M2 variants were grown at 37°C in LB media containing 50 g/ml kanamycin, 34 μg/ml chloramphenicol and then diluted to an initial OD600 of 0.1 in 50 ml of LB media containing 50 μg/ml kanamycin, 34 μg/ml chloramphenicol. Cultures were grown at 37°C to an OD600 of 0.6 and 10 ml was harvested immediately (−IPTG) or after 2 additional hrs. growth in the presence of 1 mM IPTG. Total RNA was isolated using hot acidic phenol essentially as described  and μ30 μg was analyzed by primer extension as described  using the oligonucleotide primer 5′-dATATTCATTAGGATCTTCGTCC-3′ (corresponding to TFG1 positions +294 to +273, where +1 is defined as the A in the translation-initiating ATG.
For the analysis of Tfg1 proteins in E. coli, pre- and post-induction cultures (1.5 ml) were harvested by centifugation, resuspended with 80 μl 2X SDS-PAGE sample buffer, and lysed by incubation at 95°C for 5 min. Samples were centrifuged at 14,000 × g for 3 min. and 5 μl of the supernatant resolved on an SDS-10% polyacrylamide gel, transferred to Immobilon-P PVDF membranes (Millipore), and probed with a mouse monoclonal anti-FLAG antibody (Sigma, 1:1000 dilution) followed by goat anti-mouse IgG coupled to horseradish peroxidase (Pierce, 1:1000 dilution). Immune complexes were visualized by the addition of SuperSignal ECL substrate (Pierce) and autoradiography. For the analysis of Tfg1p in purified recombinant Tfg1-Tfg2 complexes, 120 ng protein was analyzed using either the mouse monoclonal anti-FLAG antibody (Sigma, 1:1000 dilution) or a mouse monoclonal anti-His6 antibody (Covance, 1:1000 dilution) for detection of the Tfg1p C-terminus or N-terminus, respectively.
Overnight cultures of E. coli strain Rosetta™ (DE3) harboring pDt/g1*g2 (inoculated from a freshly streaked colony) were grown at 37°C in LB media containing 50 μg/ml kanamycin, 34 μg/ml chloramphenicol and then diluted to an initial OD600 of 0.1 in 1 liter of LB media containing 50 μg/ml kanamycin, 34 μg/ml chloramphenicol. The culture was grown at 30°C to an OD600 of 0.6 – 0.9 and expression of the recombinant Tfg1 and Tfg2 proteins was induced with the addition of 1 ml 1M IPTG (1 mM final concentration). The culture was grown at 30°C for 3 hrs. and the cells were harvested by centrifugation and frozen at −80°C. Frozen cell pellets were thawed on ice, resuspended with 27 ml of ice-cold lysis buffer (50 mM Tris-HCl pH 8.0, 500 mM NaCl, 10% glycerol, 1 mM PMSF, 2 mM Benzamidine, 20 μM Bestatin, 3μM Pepstatin A, 1μM Leupeptin) containing 2 mg/ml lysozyme, and the suspension was incubated on ice for 30 min. with occasional mixing. Cells were lysed by the addition of 3 ml of 20% Triton X-100 (in lysis buffer), 300μl of 1M MgCl2, 300μl of 10 mg/ml RNase A and 300μl of 2 mg/ml DNase I. The suspension was mixed by inversion and incubated on ice for 30 min. with occasional mixing, after which another 300 μl of 2 mg/ml DNase I was added and incubation continued for an additional 30 min. The lysate was centrifuged at 15,000 × g for 20 min. at 4°C, and the cleared supernatant was adjusted to a final concentration of 1 mM imidazole and batch loaded onto 1 ml Ni-NTA resin (Invitrogen) for 90 min. at 4°C. The resin was subsequently packed into a column and washed with 30 ml of lysis buffer containing 1 mM imidazole, (omitting Bestatin, Pepstatin A, Leupeptin) and the protein was eluted with 10 ml of lysis buffer containing 250 mM imidazole (omitting Bestatin, Pepstatin A, Leupeptin), collecting 1 ml fractions. Peak fractions of Tfg1,2 protein were pooled (2–3 ml final), added to 400μl of IgG Sepharose resin (GE Healthcare) pre-equilibrated with Buffer A (20 mM HEPES-KOH pH 7.6, 300 mM potassium acetate, 10% glycerol, 2 mM DTT, 0.5 mM EDTA, 0.01% NP-40, 1 mM PMSF, 2 mM Benzamidine), and incubated at 4°C overnight (or greater than 5 hrs.) with gentle rocking. The resin was harvested by centrifugation for 5 min. at 2500 × g, washed twice with 1.5 ml ice-cold Buffer A, and resuspended with 1.5 ml Buffer A containing 150 units AcTEV protease (Invitrogen). The mixture was incubated at 16°C for 3 hrs. with gentle rocking and then centrifuged as before, collecting the eluant supernatant. The resin was washed once with 0.5 ml Buffer A and the supernatant was pooled with the eluant supernatant. If necessary, the TEV eluant/wash could be concentrated using a Microcon Ultracel YM-50 spin column (Millipore). Typical yield was 1–1.2 mg of TFIIF protein per 1 liter of starting culture.
Cultures (20 L) of yeast strains YT062 or YT063 were grown at 30°C in YPD media to an OD600 of ~ 6.0, harvested by centrifugation, and the cell pellet (~160 g) resuspended in ½ volume (80 ml) of yeast lysis buffer (150 mM Tris-acetate pH 7.9, 100 mM potassium acetate, 20% glycerol, 1 mM DTT, 1 mM EDTA), and the suspension frozen at −80°C. The frozen suspension was thawed on ice, protease inhibitors were added (final concentrations: 2 mM Benzamidine, 1 mM PMSF, 3 μM Chymostatin, 2 μM Pepstatin A, 1 μM Leupeptin), and the cells were disrupted with a BioSpec Bead Beater, using an ice/water bath surrounded stainless steel vessel, 150 ml glass beads (0.5 mm), and 10 cycles of 1 min. beating and a 1.5 min. rest. The supernatant was harvested and the beads were rinsed once with 150 ml lysis buffer and pooled with the supernatant. Potassium acetate was added to a final concentration of 600 mM, the lysate was stirred for 20 min. at 4°C, and then centrifuged at 20,000 × g for 15 min. at 4°C. The supernatant was then centrifuged using a 45 Ti rotor at 100,000 × g for 60 min. at 4°C, or where indicated, adjusted to a final concentration of 0.33% polyethyleneimine (PEI) and stirred for 15 min. at 4°C prior to the high speed clarification. In the absence of PEI, three phases were observed after the high speed spin which included a top lipid phase, a middle clear lysate (harvested), and a loose pellet of precipitated material. In the presence of PEI, the pellet was solid, and the top phase was not observed. The clear lysate was harvested and filtered through Whatman 1 paper. Typical recoveries were approximately 270 ml for the untreated lysate (20–25 mg/ml protein) and 350 ml for the PEI treated lysate (16 mg/ml protein). The lysates were adjusted to a final concentration of 0.01% NP-40, protease inhibitors were again added, and 45 ml aliquots were added to 400 μl IgG Sepharose resin pre-equilibrated with Buffer A (20 mM HEPES-KOH pH 7.6, 300 mM potassium acetate, 10% glycerol, 2 mM DTT, 0.5 mM EDTA, 0.01% NP-40, 2 mM Benzamidine, 1 mM PMSF) and incubated at 4°C overnight (or greater than 2 hrs.) with gentle rocking. For each 45 ml sample, the resin was harvested by centrifugation for 3 min. at 2500 × g, washed once with 8 ml ice-cold Buffer A, and then washed three times with 1.2 ml ice-cold Buffer A in a microcentruge tube. The resin was then resuspended in 1.5 ml Buffer A containing 150 units AcTEV protease (Invitrogen), incubated at 16°C for 3 hrs. with gentle rocking, centrifuged for 2 min. at 2500 × g, and the eluant supernatant was collected. The resin was washed once with 1 ml Buffer A and the supernatant was pooled with the eluant. Each 2.5 ml pool was adjusted to 2 mM CaCl2, raised to 10 ml total volume by the addition of 7.5 ml of calmodulin binding buffer (CBB; 20 mM HEPES-KOH pH 7.6, 300 mM potassium acetate, 10% glycerol, 2 mM CaCl2, 2 mM DTT, 1 mM magnesium acetate, 1 mM imidazole, 0.01% NP-40, 2 mM Benzamidine, 1 mM PMSF), added to 200 μl CBB-equilibrated Calmodulin Affinity Resin (Stratagene), and incubated for 90 min. with gentle rocking at 4°C. The resin was harvested by centrifugation for 3 min. at 2500 × g, washed twice with 0.8 ml ice-cold CBB in a microcentruge tube, and the TFIIH complexes were eluted by the addition of 400 μl of calmodulin elution buffer (CEB; 20 mM HEPES-KOH pH 7.6, 300 mM potassium acetate, 10% glycerol, 3 mM EGTA, 2 mM DTT, 1 mM magnesium acetate, 0.01% NP-40, 2 mM Benzamidine, 1 mM PMSF) and incubation for 15 min. with gentle rocking at 4°C. The resin was harvested by centrifugation for 2 min. at 2500 × g, the eluant was harvested, and the elution procedure was repeated once more using 200 μl CEB. The pooled eluants (600 l) were then concentrated using a Microcon Ultracel YM-30 spin column (Millipore). Total yields from 20 liters of strain YT062 (Tfb3-TAP) were approximately 200 μg of a mixture of holo-TFIIH and TFIIK for - PEI preps, and 90 μg of pure TFIIK for + PEI preps. Total yields from 20 liters of strain YT063 (Tfb4-TAP) were approximately 100 μg of a mixture of holo-TFIIH and core TFIIH for - PEI preps, and 40 μg of pure core TFIIH for + PEI preps.
Reconstituted in vitro transcription reactions were carried out using plasmid template pG5CG- as described , using 20 units of Ribolock RNase inhibitor (Fermentas) substituted for RNasin and the amount of TFIIH indicated in the legend to Figure 4. CTD kinase reactions (15 μl) contained 50 mM HEPES-KOH pH 7.6, 8% glycerol, 80 mM potassium acetate, 10 mM magnesium acetate, 5 mM EGTA, 2.5 mM DTT, 1 pmol (500 ng) purified S. cerevisiae RNAPII (IIA, hypo-phosphorylated form) and the amount of TFIIH indicated in the legend to Figure 4. Reactions were initiated by the addition of dATP or ATP (400 μM final) and incubated at ambient temperature for 10 min. Reactions were terminated by the addition of 15 μl 2X SDS-PAGE sample buffer and 10 μl was resolved on an SDS-8% polyacrylamide gel, transferred to Immobilon-P PVDF membranes (Millipore), and probed with mouse anti-phosphoSer5-CTD antibody (H14, Covance, 1:1000 dilution) followed by goat anti-mouse IgM coupled to horseradish peroxidase (Pierce, 1:1000 dilution). Immune complexes were visualized by the addition of SuperSignal ECL substrate (Pierce) and autoradiography.
We previously reported an approach for the production of functional recombinant Tfg1-Tfg2 complexes that involved a broad host range low-copy vector, induction of Tfg1, Tfg2 co-expression by infection with bacteriophage CE6 carrying T7 RNA polymerase, and multi-step affinity purification . The low-copy vector was required due to the previously identified toxicity of the TFG1 gene in E. coli, and induction by CE6 infection was required due to the inability to establish stable transformants of the co-expression vector in E.coli cells containing inducible T7 RNA polymerase, such as BLR (DE3). In addition, the purification involved the preparation and use of an anti-myc antibody column, making the approach somewhat cost ineffective in light of the yield of Tfg1-Tfg2 complexes obtained. To address this last issue, we sought to determine whether a TAP epitope tag could be substituted for the myc tag that was fused to the C-terminus of Tfg1. Using the plasmid pCOLADuet-1 as a co-expression vector for Tfg1 and Tfg2, we generated an initial construct (designated pDt/g1g2, see Materials and methods) and discovered that the addition of the C-terminal TAP tag significantly reduced the toxicity of the TFG1-containing plasmid in E. coli, suggesting that this toxicity involves the C-terminal region of Tfg1. More importantly, this result allowed for stable transformants of pDt/g1g2 to be established in host strains such as Rosetta™ (DE3), which carries both IPTG-inducible T7 RNA polymerase as well as the plasmid pRARE, which supplies tRNAs for rarely used codons in E. coli. Using Rosetta as the host strain, we initially examined IPTG induction of Tfg1 by western blotting using antibody recognizing the FLAG epitope that was part of the C-terminal TAP tag used (see Materials and methods). The results demonstrated that full-length Tfg1 protein was induced, but that the vast majority of the Tfg1 detected was an N-terminal truncation product that, importantly, was present prior to IPTG induction (Fig. 1A, lanes 1, 4). This result suggested the potential presence of a cryptic functional E. coli promoter somewhere within the first 300 base pairs of the TFG1 coding region. Using primer extension analysis, we identified an IPTG-independent transcript initiating within the N-terminal Tfg1 coding region (Fig. 1B, lane 1; Fig. 1C) and was able to eliminate the E. coli promoter generating this transcript by site-directed mutagenesis (mutant M1, Fig. 1B, lane 2; Fig. 1C). Significantly, the elimination of this E. coli promoter dramatically increased the growth rate of strains carrying the M1 plasmid (data not shown). Although the elimination of this transcript generated by the E. coli promoter abolished the production of the Tfg1 N-terminal truncation in the absence of IPTG, (Fig. 1A, lane 2), it did not abolish the generation of this product after IPTG induction (Fig. 1A, lane 5). This result suggested that, in addition to the presence of an E. coli promoter in the Tfg1 N-terminal coding region, this region might also contain an alternative downstream ribosome binding site for internal translation initiation. Using site-directed mutagenesis, we confirmed that, indeed, internal translation initiation does occur at Tfg1 Met-114 to generate the prominent N-terminal truncation product (mutant M2, Fig. 1A, lane 6; Fig. 1C).
Having eliminated both a cryptic E. coli promoter and the internal translation initiation within the N-terminal Tfg1 coding region, we proceeded to utilize the M2 mutant construct (designated pDt/g1*g2, Fig. 2A) for production and purification of Tfg1-Tfg2 complexes. Briefly, the approach included (i) IPTG induction of Tfg1 and Tfg2 in the host strain Rosetta™ (DE3); (ii) enrichment of Tfg1-Tfg2 complexes using Ni-NTA affinity chromatography; and (iii) purification of the Tfg1-Tfg2 complexes by binding to IgG-Sepharose beads and liberating by TEV protease cleavage (see Materials and methods for details). The complexes were sufficiently pure after the IgG-Sepharose step that further purification using calmodulin beads was unnecessary (Fig. 2B, lane 3). Western blot analysis using antibodies specific to the recombinant Tfg1 C-terminus (anti-FLAG) or N-terminus (anti-His6) confirmed that the single Tfg1 species present in the purified M2 preparation corresponded to full-length protein (Fig. 2C, lanes 2, 4). Importantly, in vitro transcription assays demonstrated that the TAP purified Tfg1-Tfg2 complexes were as functional, if not slightly more active, than Tfg1-Tfg2 complexes purified using the previously reported method (Fig. 4A).
As noted earlier, purification of functional S. cerevisiae holo-TFIIH has previously been achieved using a strain containing a hexahistidine-tagged Tfb1 subunit and five to six chromatographic steps in the purification scheme to reach apparent homogeneity [36, 37]. More recently, Kornberg and colleagues generated S. cerevisiae strains containing TAP-tagged versions of the TFIIH subunits Tfb3 and Tfb4 in order to utilize TAP purification to resolve an apparent discrepancy in the subunit composition of S. cerevisiae and human TFIIH complexes . Using a TAP purification scheme that involved ammonium sulfate precipitation and 500 mM ammonium sulfate washing during the IgG-agarose chromatographic step, the authors successfully determined a revised subunit structure for yeast TFIIH with complete conservation to that of the human factor. Significantly, however, the stringent purification conditions used leads to the underrepresentation of some polypeptides, the dissociation of core TFIIH from TFIIK, and complexes that do not support transcription in reconstituted assays (data not shown). Using these Tfb3-TAP and Tfb4-TAP strains, we were able to establish conditions for the isolation of functional TFIIH and TFIIK complexes (see Materials and methods for details). Using the Tfb3-TAP strain, pure TFIIK was isolated from polyethyleneimine (PEI) treated cell lysates (Fig. 3C), whereas a mixture of holo-TFIIH and TFIIK was isolated from untreated cell lysates (Fig. 3B). Using the Tfb4-TAP strain, pure core TFIIH was isolated from PEI treated cell lysates (Fig. 3E), whereas a mixture of holo-TFIIH and core TFIIH was isolated from untreated cell lysates (Fig. 3D). Importantly, TFIIH complexes isolated from PEI lysates supported robust in vitro transcription in reconstituted assays (Fig. 4B), and exhibited CTD kinase activity comparable to that observed with the purified TFIIK (Fig. 4D).
In this study, we sought to develop improved approaches for (i) the expression and purification of recombinant S. cerevisiae TFIIF (Tfg1-Tfg2) complexes and (ii) the purification of functional native TFIIH complexes from yeast extracts. In regards to the expression and purification of recombinant Tfg1-Tfg2 complexes, a major obstacle has been the extreme toxicity conferred by the TFG1 gene in E. coli. As noted earlier, we previously reported an approach that partially circumvented the toxicity and allowed for the production of functional recombinant Tfg1-Tfg2 complexes through the use of broad host range low-copy vector, induction of Tfg1, Tfg2 co-expression by infection with bacteriophage CE6 carrying T7 RNA polymerase, and subsequent multi-step affinity purification . In an attempt to simplify the purification scheme, we fortuitously discovered that the addition of a TAP tag to the C-terminus of Tfg1 significantly reduced the toxicity and allowed for stable transformants of our new co-expression plasmid to be established in host strains such as Rosetta™ (DE3), which carries both IPTG-inducible T7 RNA polymerase and the pRARE plasmid to provide rare tRNAs for improved translation in E. coli. Utilizing this new system in combination with western blotting and primer extension analysis, we subsequently identified both a cryptic E. coli promoter and an alternative translation initiation site (Met-114) in the N-terminal coding region of TFG1 (Fig. 1). As a result of these two functional elements, a predominant N-terminal Tfg1 truncation product is generated both constitutively (cryptic promoter functioning in combination with the internal initiation at Met-114) as well as after IPTG induction of the T7 expression promoter (internal initiation of the full-length T7 generated transcript at Met-114). Importantly, elimination of both these elements by site-directed mutagenesis coordinately and completely eliminated the production of the N-terminal truncation product and the toxicity of the TFG1 gene (Fig. 1 and data not shown). Since the initial addition of a C-terminal TAP tag partially eliminated the toxicity of a TFG1 gene that was generating the N-terminal truncation product, the combined results strongly suggest that the toxicity of the TFG1 gene in E. coli is due to the N-terminal truncation product that contains a wild-type C-terminus. Although the precise target and mechanism for this toxicity remains unclear, it is noteworthy that the RAP74 human homolog of Tfg1 has been reported to contain a C-terminal RNAPII binding domain that is masked by an intramolecular interaction with the RAP74 N-terminal region . Thus, it is tempting to speculate that Tfg1 may also contain a C-terminal RNAPII binding domain that is masked by an intramolecular interaction with the N-terminal region, that this RNAPII binding domain has the potential to interact with E. coli polymerase, and that the combination of the cryptic E. coli promoter and Met-114 translation initiation constitutively generates an N-terminal truncation product with the potential to do so and confer extreme toxicity. In any event, the elimination of this toxicity and the significantly enhanced production and purification of Tfg1-Tfg2 complexes reported here should prove invaluable for future structure and function studies of S. cerevisiae TFIIF.
In the second part of this work, we sought to develop an improved approach for the purification of functional native TFIIH complexes from yeast extracts. As discussed earlier, TFIIH purification has previously been achieved using five to six traditional chromatographic steps to reach apparent homogeneity [36, 37]. Although S. cerevisiae strains containing TAP-tagged versions of the TFIIH subunits Tfb3 and Tfb4 have been generated and utilized to examine the composition of yeast TFIIH complexes , the stringent purification conditions that were utilized, involving both ammonium sulfate precipitation and high salt column washes, leads to the underrepresentation of some polypeptides and complexes that do not support transcription in reconstituted assays. By using these strains, eliminating the ammonium sulfate precipitation step, and both reducing and changing the salt during the TAP purification scheme, we have established conditions for the two step isolation of TFIIH complexes that are functional in both reconstituted transcription and RNAPII CTD kinase assays (Figs. 3 and and4).4). As is the case for TFIIF, the improved approach for the isolation of functional TFIIH complexes should prove invaluable for future mechanistic studies of this essential and centrally important transcription factor.
The authors thank Yuichiro Takagi and Roger Kornberg for providing yeast strains YT062 and YT063. This work was supported by a Public Health Service grant (GM51124) from the National Institutes of Health to A.S.P.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.