letters to nature

10 ml YPD to an OD600 value of 0.2 except pol30-6 and pol30-42, which were diluted to an OD600 value of 0.25, as these two mutants grew more slowly. When the OD600 value of each culture reached 0.81, 0.5 ml of each culture was used for FACS analysis as described15. The rest of the yeast cells were harvested, and washed with 20% cold glycerol plus 2 mM of the protease inhibitor pefabloc. The chromatin-binding assay was performed as described15 except that twice the amounts of PMSF and pefabloc were used for all the buffers containing these two inhibitors. Equivalent amounts of each sample were then loaded onto 12.5% SDSPAGE, transferred to nitrocellulose membranes, and probed with monoclonal antibodies against Cac1-3HA (12CA5, 1:10000), Cac2-13Myc (9E10, 1:10000), Orc3 (SB3, 1:10000) and polyclonal antisera to PCNA (871, 1:2000).
Received 19 April; accepted 19 September 2000. 1. Loo, S. & Rine, J. Silencing and heritable domains of gene expression. Annu. Rev. Cell Dev. Biol. 11, 519548 (1995). 2. Grunstein, M. Yeast heterochromatin: regulation of its assembly and inheritance by histones. Cell 93, 325328 (1998). 3. Lustig, A. J. Mechanisms of silencing in Saccharomyces cerevisiae. Curr. Opin. Genet. Dev. 8, 233239 (1998); erratum Curr. Opin. Genet. Dev. 8, 721 (1998). 4. Pillus, L. & Rine, J. Epigenetic inheritance of transcriptional states in S. cerevisiae. Cell 59, 637647 (1989). 5. Waga, S. & Stillman, B. The DNA replication fork in eukaryotic cells. Annu. Rev. Biochem. 67, 721751 (1998). 6. Verreault, A. De novo nucleosome assembly: new pieces in an old puzzle. Genes Dev. 14, 14301438 (2000). 7. Shibahara, K. & Stillman, B. Replication-dependent marking of DNA by PCNA facilitates CAF-1coupled inheritance of chromatin. Cell 96, 575585 (1999). 8. Moggs, J. G. et al. A CAF-1-PCNA-mediated chromatin assembly pathway triggered by sensing DNA damage. Mol. Cell Biol. 20, 12061218 (2000). 9. Kaufman, P. D., Kobayashi, R. & Stillman, B. Ultraviolet radiation sensitivity and reduction of telomeric silencing in Saccharomyces cerevisiae cells lacking chromatin assembly factor-I. Genes Dev. 11, 345357 (1997). 10. Enomoto, S. & Berman, J. Chromatin assembly factor I contributes to the maintenance, but not the reestablishment, of silencing at the yeast silent mating loci. Genes Dev. 12, 219232 (1998). 11. Henderson, D. S., Banga, S. S., Grigliatti, T. A. & Boyd, J. B. Mutagen sensitivity and suppression of position-effect variegation result from mutations in mus209, the Drosophila gene encoding PCNA. EMBO J. 13, 14501459 (1994). 12. Gottschling, D. E., Aparicio, O. M., Billington, B. L. & Zakian, V. A. Position effect at S. cerevisiae telomeres: reversible repression of Pol II transcription. Cell 63, 751762 (1990). 13. Sussel, L., Vannier, D. & Shore, D. Epigenetic switching of transcriptional states: cis- and trans-acting factors affecting establishment of silencing at the HMR locus in Saccharomyces cerevisiae. Mol. Cell. Biol. 13, 39193928 (1993). 14. Mahoney, D. J., Marquardt, R., Shei, G. J., Rose, A. B. & Broach, J. R. Mutations in the HML E silencer of Saccharomyces cerevisiae yield metastable inheritance of transcriptional repression. Genes Dev. 5, 605615 (1991). 15. Liang, C. & Stillman, B. Persistent initiation of DNA replication and chromatin-bound MCM proteins during the cell cycle in cdc6 mutants. Genes Dev. 11, 33753386 (1997). 16. Kaufman, P. D., Kobayashi, R., Kessler, N. & Stillman, B. The p150 and p60 subunits of chromatin assembly factor I: a molecular link between newly synthesized histones and DNA replication. Cell 81, 11051114 (1995). 17. Krishna, T. S., Kong, X. P., Gary, S., Burgers, P. M. & Kuriyan, J. Crystal structure of the eukaryotic DNA polymerase processivity factor PCNA. Cell 79, 12331243 (1994). 18. Ayyagari, R., Impellizzeri, K. J., Yoder, B. L., Gary, S. L. & Burgers, P. M. A mutational analysis of the yeast proliferating cell nuclear antigen indicates distinct roles in DNA replication and DNA repair. Mol. Cell. Biol. 15, 44204429 (1995). 19. Eissenberg, J. C., Ayyagari, R., Gomes, X. V. & Burgers, P. M. Mutations in yeast proliferating cell nuclear antigen dene distinct sites for interaction with DNA polymerase delta and DNA polymerase epsilon. Mol. Cell. Biol. 17, 63676378 (1997). 20. Miller, A. M. & Nasmyth, K. A. Role of DNA replication in the repression of silent mating type loci in yeast. Nature 312, 247251 (1984). 21. Triolo, T. & Sternglanz, R. Role of interactions between the origin recognition complex and SIR1 in transcriptional silencing. Nature 381, 251253 (1996). 22. Chien, C. T., Buck, S., Sternglanz, R. & Shore, D. Targeting of SIR1 protein establishes transcriptional silencing at HM loci and telomeres in yeast. Cell 75, 531541 (1993). 23. Fox, C. A., Ehrenhofer-Murray, A. E., Loo, S. & Rine, J. The origin recognition complex, SIR1, and the S phase requirement for silencing. Science 276, 15471551 (1997). 24. Murzina, N., Verreault, A., Laue, E. & Stillman, B. Heterochromatin dynamics in mouse cells: interaction between chromatin assembly factor 1 and HP1 proteins. Mol. Cell 4, 529540 (1999). 25. Le, S., Davis, C., Konopka, J. B. & Sternglanz, R. Two new S-phase-specic genes from Saccharomyces cerevisiae. Yeast 13, 10291042 (1997). 26. Tyler, J. K. et al. The RCAF complex mediates chromatin assembly during DNA replication and repair. Nature 402, 555560 (1999). 27. Verreault, A., Kaufman, P. D., Kobayashi, R. & Stillman, B. Nucleosome assembly by a complex of CAF-1 and acetylated histones H3/H4. Cell 87, 95104 (1996). 28. Ehrenhofer-Murray, A. E., Kamakaka, R. T. & Rine, J. A role for the replication proteins PCNA, RF-C, polymerase epsilon and Cdc45 in transcriptional silencing in Saccharomyces cerevisiae. Genetics 153, 11711182 (1999). 29. Laman, H., Balderes, D. & Shore, D. Disturbance of normal cell cycle progression enhances the establishment of transcriptional silencing in Saccharomyces cerevisiae. Mol. Cell. Biol. 15, 36083617 (1995). 30. Scott, M. P. Development: the natural history of genes. Cell 100, 2740 (2000).

Acknowledgements
We thank P. Burgers, P. Kaufman, R. Sternglanz and D. Shore for plasmids and yeast strains used in this study. We thank A. Stenlund for critical reading of the manuscript, T. Tully for statistical analysis of the data presented in Table 1, and members of the Stillman laboratory, especially L. Zou, for helpful discussions. This work is supported by a grant from the National Institutes of Health (to B. S.). Z. Z. is supported by a postdoctoral fellowship from the Cancer Research Fund of the Damon Runyon-Walter Winchell Foundation. K. S. is a Leukemia Society of America Special Fellow. Correspondence and requests for materials should be addressed to B. S. (e-mail: stillman@cshl.org).

.................................................................
A transcription reinitiation intermediate that is stabilized by activator

Natalya Yudkovsky*, Jeffrey A. Ranish* & Steven Hahn*

* Division of Basic Sciences, The Fred Hutchinson Cancer Research Center, and Molecular and Cellular Biology Program, University of Washington, Seattle, Washington 98109, USA The Howard Hughes Medical Institute, Seattle, Washington 98109, USA
..............................................................................................................................................

High levels of gene transcription by RNA polymerase II depend on high rates of transcription initiation and reinitiation. Initiation requires recruitment of the complete transcription machinery to a promoter, a process facilitated by activators and chromatin remodelling factors. Reinitiation probably occurs through a different pathway1. After initiation, a subset of the transcription machinery remains at the promoter, forming a platform for assembly of a second transcription complex24. Here we describe the isolation of a reinitiation intermediate that includes transcription factors TFIID, TFIIA, TFIIH, TFIIE and Mediator. This intermediate can act as a scaffold for formation of a functional reinitiation complex. Formation of this scaffold is dependent on ATP and TFIIH. The scaffold is stabilized in the presence of the activator Gal4VP16, but not Gal4AH, suggesting a new role for some activators and Mediator in promoting high levels of transcription. The rst step in transcription initiation by RNA polymerase II (RNA Pol II) is recruitment of the transcription machinery to a promoter to form a pre-initiation complex (PIC). After initiation, a subset of the factors in the PIC dissociates from the promoter24. To begin a second round of transcription (reinitiation), this subset of factors, along with RNA Pol II, must again be recruited to the promoter. In the yeast Mediator-dependent system, PIC formation can occur in at least two steps5. In the rst step, TFIID and TFIIA bind cooperatively to the promoter. In the second step, the rest of the transcription machinery stably binds to form a complete PIC. This step requires cooperative binding of TFIIB and holopolymerase, a complex composed of RNA Pol II and Mediator6,7. Mediator, which contains Srb, Med and other proteins, associates with RNA Pol II, allows transcription to be responsive to activators, and stimulates RNA Pol II carboxy-terminal domain (CTD) phosphorylation6,7. Although reinitiation probably involves the same complement of transcription factors as initiation, evidence suggests that the pathways for initiation and reinitiation are distinct. In vitro studies using HeLa extracts have shown that the rate of reinitiation at some promoters is fourfold higher than that of initiation8. Other studies showed that activator, TFIID and TFIIA remain at the promoter after initiation24. We used an immobilized promoter template assay and yeast nuclear extracts to isolate a reinitiation intermediate5. Such an
Present address: Institute for Systems Biology, Seattle, Washington 98105, USA.

Supplementary information is available at Nature's World-Wide Web site (http://www.nature.com) or as paper copy from the London editorial ofce of Nature.
NATURE | VOL 408 | 9 NOVEMBER 2000 | www.nature.com

2000 Macmillan Magazines Ltd

225

letters to nature
approach allowed us to monitor the entire transcription machinery in a crude Mediator-dependent transcription system, rather than one using puried factors. In this assay, HIS4 promoter templates immobilized on magnetic beads were incubated with nuclear extract and the activator Gal4AH to allow PIC formation. The PICs were then washed and transcription initiated by addition of nucleotides for 2 min. This procedure allows only a single round of transcription to occur, as the transcription signal detected is equivalent to that seen after incubation of PICs with NTPs for 1 min, followed by addition of sarkosyl to block reinitiation (data not shown). After nucleotide addition, proteins still bound to the templates were isolated. As expected from previous studies24, RNA Pol II, TFIIB and TFIIF dissociated from the templates, but activator, TBP, the TFIID subunits TAFII90 and TAFII67 and TFIIA remained bound to the promoters (Fig. 1, compare lanes 2 and 3; and data not shown). We found that the Mediator complex (Srb4, Srb2, Med6 and Gal11 subunits) and substantial amounts of TFIIH and TFIIE also remained at the promoter. Specically, the level of RNA Pol II was reduced 13-fold, the level of TFIIB was reduced 24-fold and the level of TFIIF was reduced 14-fold, whereas the levels of all other components were reduced less than 2.5-fold. We then investigated whether this complex of activator, TFIID, TFIIA, TFIIH, TFIIE and Mediator could function as a reinitiation intermediate, by acting as a scaffold on which a functional transcription complex would reassemble. The scaffolds were formed and washed as described in Fig. 1. A second nuclear extract was then added along with nucleotides to determine whether a second round of transcription could occur (Fig. 2a). For the second extract, we used extracts made from strains with mutations in Mediator components (DSrb2 or Srb4ts), TFIIB (G41E), TFIIH (Kin28ts), TFIIE (Tfa1ts), TBP (I143N) or TFIIA (Toa1-25). All of these extracts are defective in PIC assembly5 (data not shown) and transcription (Fig. 2b, lanes 17; Fig. 2c, lanes 13; Fig. 2d, lanes 15). As a control to show that few active PICs remained after the rst round of transcription, very little RNA was produced when nucleotides were added to the scaffolds in the absence of a second extract (Fig. 2b, lane 8; Fig. 2c, lane 4; Fig. 2d, lane 6). When supplemented with extracts from DSrb2, Srb4ts, Kin28ts, TBPI143N and Toa1-25 mutants, the scaffolds supported a second round of transcription, conrming the presence of these components

Template:

+ +AMPPNP a
Imm. Temp. +Gal4AH 40' +1st NE Wash +NTPs 1st NE IIB(G41E ) Kin28ts +NTPs 2' Wash Stop Scaffold + 2nd NE IIB(G41E ) Kin28ts +2nd NE +NTPs Stop

+NTPs

+ATP

RPB3 TFIIB

POLII

Srb4ts

TFG2 TBP

TFIIF

Srb4ts

Srb2

Srb2

WT

WT

rIIB: rSrb2:

+ + +

TFIID TAF II 90 TOA2 SRB4 c SRB2 Mediator

WT 1st NE TFA1ts Scaffold + 2nd NE TFA1ts TS

TFIIA
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

d
1st NE TBP (I143N )

Scaffold + 2nd NE TBP (I143N ) +

WT

Toa1-25

GAL11 KIN28 TFA2 GAL4AH 1 2 3 4 5 TFIIH TFIIE

rTFIIE:

+ rTFIIA : rTBP :

+ +

TS

Toa1-25 + TS

MED6

WT

1 2 3

4 5 6 7

1 2 3 4 5

6 7 8 9 10 11

Figure 1 Scaffold contains activator, TFIID, TFIIA, Mediator, TFIIH and TFIIE. The 515 template, comprising a HIS4 promoter with a single Gal4 DNA-binding site upstream, was immobilized on a magnetic Dynabead. Immobilized templates were incubated with the activator Gal4-AH and nuclear extract for 40 min to form PICs. Templates were washed and nucleotides added for 2 min as indicated. Templates were washed again, and bound proteins were isolated by PstI digestion and detected by western blotting. Lane 2 shows a typical PIC. As a control for nonspecic binding to the Dynabeads, the reaction in lane 1 was performed without template.
226

Figure 2 Scaffold supports reinitiation. a, The scaffold reinitiation assay. b, Lanes 17, nuclear extracts were incubated with 515 immobilized templates for 40 min to form PICs. NTPs were added and reactions stopped after 2 min to allow for a single round of transcription. Lanes 815, reactions were performed as described in a. NTPs were added along with the second nuclear extract for 2 min. As a control for residual active complexes, no second nuclear extract was added in lane 8. c, d, Reactions were performed as described in b, but in d all nal NTP incubations occurred for 30 min. Reactions were assayed by primer extension. TS, transcription signal. WT, wild type.
NATURE | VOL 408 | 9 NOVEMBER 2000 | www.nature.com

2000 Macmillan Magazines Ltd

WT

letters to nature
in a functional reinitiation intermediate (Fig. 2b, lanes 812, 15; Fig. 2d, lanes 611). Little transcription was seen with the IIB(G41E) extract, however, conrming that TFIIB is not part of the scaffold (Fig. 2b, lanes 13 and 14). Notably, although recombinant Srb2 restored transcriptional activity to the DSrb2 extract, it had no effect on transcription when the scaffold was used (Fig. 2b, compare lanes 2 and 3 with 10 and 11). Although recombinant TBP and TFIIA did stimulate transcription on the scaffold when the TBPI143N and Toa1-25 extracts were used, transcription in their absence was signicantly higher than that seen in the absence of scaffold (Fig. 2d, compare lanes 2 and 4 with 8 and 10). Recombinant TFIIE stimulated transcription from the Tfa1ts extract fourfold (Fig. 2c, lanes 2 and 3), as compared with twofold stimulation when scaffold templates were used (lanes 6 and 7). From this and from Fig. 1, we conclude that the scaffold contains some functional TFIIE. It is apparent that TFIIE is the least stable component of the scaffold, and that TFIIH and TFIIA also dissociate to some extent on NTP addition. As a control for the experiments of Fig. 2, we used a competition assay to show that most of the transcription observed in the second round of initiation originated from scaffolds, rather than from newly formed PICs. As expected, when a second template was added to scaffolds along with the second nuclear extract, most of the transcription observed originated from the scaffold template (see Supplementary Information). We were able to isolate this functional reinitiation intermediate despite the low percentage of active PICs in our assay. We determined the number of active PICs by measuring the amount of RNA produced in a single round of transcription. Comparing this number to the total number of PICs formed showed that only 5 10% of PICs were active in transcription (data not shown). These data indicate that the scaffold complexes isolated by our assay may be the result of dissociation of both active and inactive PICs. As these complexes can support reinitiation, these results imply that both active and inactive PICs dissociate by the same mechanism on nucleotide addition. We found that adding only ATP to PICs had the same effect as adding all four nucleotides: both resulted in PIC dissociation and loss of active PICs (Fig. 1; and data not shown). The ATP analogue AMPPNP did not promote PIC dissociation (Fig. 1, lane 5), which suggests that ATP hydrolysis, rather than transcription, is necessary for PIC dissociation. We therefore attempted to identify a PIC component with ATP-dependent activity that might be responsible for PIC dissociation. Three subunits of TFIIH were good candidates: the helicases Rad25 and Rad3, and the CTD kinase Kin28. We prepared nuclear extracts from transcriptionally defective strains that contained temperature-sensitive mutations in either Rad3 or Kin28 and tested them for PIC dissociation. These extracts were able to form a PIC intermediate that lacked both the Kin28 and Tfb1 subunits of TFIIH, even though these subunits were present in the mutant extracts (Fig. 3, lanes 2, 4, 6; and data not shown). The lack of these subunits suggests that this PIC intermediate lacks the entire TFIIH complex. After ATP addition, PICs lacking TFIIH were not able to dissociate into scaffolds, indicating that PIC dissociation is dependent on ATP and TFIIH (Fig. 3, lanes 3, 5, 7). Because PICs formed with both TFIIH mutant extracts probably lacked the entire TFIIH complex, we were unable to determine which of the TFIIH subunits is necessary for PIC dissociation. Studies have shown that phosphorylated, elongating RNA Pol II is not associated with Mediator9, suggesting that CTD phosphorylation by Kin28 may be required for the dissociation of Mediator from RNA Pol II during scaffold formation.

a
Activator: _ _
_

GAL4AH +
0' 10' 20' 40'

NTPs : _ + Time: _ 0' 10' 20' 40'

GAL4VP16 _ +
_ 0' 10' 20' 40'

TFIIB RPB3 POLII TBP TFIID TAFII90 TOA2 TFIIA SRB4 SRB2

Wild type

Mediator

Kin28 ts

Rad3 ts

MED6 GAL11 KIN28 TFIIH GAL4AH GAL4VP16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

NE:

ATP: _ _ + Template: _ + +

_ + _ + + + + + GAL4AH

b TBP IID/IIA TOA2 RPB3 TFIIB KIN28 TFB1 SRB4 SRB2 1 2 3 4 5 6 7 Mediator
0 0 10 20 30 40 50 60 Time after NTP addition (min) 1.0 RNA No activator +GAL4AH +GAL4VP16

POLlI/IIB

IIH

Figure 3 Scaffold formation is TFIIH-dependent. The immobilized template assay was performed as described in Fig. 1, using the indicated nuclear extracts. ATP was added to the reactions where indicated. Factors bound to the templates were assayed by SDS PAGE and western blot. A typical wild-type PIC and a typical wild-type scaffold are shown in lanes 2 and 3, respectively. As a control for nonspecic binding to Dynabeads, the reaction in lane 1 was performed without template.
NATURE | VOL 408 | 9 NOVEMBER 2000 | www.nature.com

Figure 4 Gal4VP16 promotes scaffold stability and a higher rate of reinitiation. a, 515 immobilized templates were pre-incubated with Gal4AH, Gal4VP16 or no activator. Scaffolds were formed using wild-type nuclear extract as described in Fig. 1. Scaffolds were incubated in transcription buffer for the times indicated, washed for 1 min, and the bound proteins analysed by western blot. As controls, PICs are shown in lanes 1, 6 and 11. b, Wild-type nuclear extract was incubated with pSH515, a plasmid containing the HIS4 template (Fig. 1) and Gal4AH, Gal4VP16 or no activator to form PICs. NTPs were added and samples removed for primer extension at 2.5, 5, 10, 21.5, 35 and 50 min. The transcription signal is shown plotted against time. The 2.5 min time point is equivalent to a single round of transcription.
227

2000 Macmillan Magazines Ltd

letters to nature
We also analysed the effects of activator on scaffold formation and stability. Although activators function by stimulating PIC formation through transcription factor recruitment, evidence indicates that they also have a role in reinitiation. The heat shock factor4 and oestrogen receptor10 transcription activators, as well as the HIV-1 enhancer11, stimulate reinitiation in vitro. Other in vitro2 and in vivo12 experiments showed that the presence of activator at the promoter is required for continued high levels of transcription. As PIC dissociation can occur in the absence of activator (Fig. 4a, lanes 1 and 2), we measured scaffold stability in either the absence of activator, or the presence of the activators Gal4AH or Gal4VP16. The scaffold was formed as described in Fig. 1 and analysed by western blot after incubation in transcription buffer for up to 40 min (Fig. 4a). In either the absence of activator or the presence of Gal4AH, the levels of TBP, TFIIA, Srb4, Srb2 and Med6 decreased by 35-fold after 40 min. In contrast, when Gal4VP16 was used the levels of all of these factors remained steady after 40 min. These results show that although the scaffold reinitiation intermediate can be formed without activator it is more stable in the presence of Gal4VP16. As some activators can interact with TFIID13, TFIIA14 and various Mediator components15,16, this stabilization is probably due to interactions between activators and scaffold components. We next measured the rate of multiround transcription in the absence of activator and in the presence of Gal4AH or Gal4P16. Nucleotides were added to PICs formed with wild-type nuclear extract, and transcription was measured at various time points (Fig. 4b). When the transcription signal is plotted against time, the resulting curve shows biphasic kinetics of transcription1,17. RNA is rapidly produced from the preformed PICs, followed by a slower rate of RNA synthesis resulting from reinitiation and new initiation events. As, in the presence of activators, the rate of RNA synthesis after the initial burst of transcription is much faster than the rate of initial PIC formation, most of the subsequent RNA synthesis probably results from reinitiation rather than new initiation events17. We found that with Gal4VP16, the rate of transcription after the rst round was 10-fold higher than with no activator and 3fold higher than with Gal4AH. These data show a correlation between scaffold stability and the rate of reinitiation, and support a role in scaffold stability for some activators in reinitiation. Our results suggest a model for reinitiation in which activator, Mediator, TFIID, TFIIA, TFIIH and TFIIE remain at the promoter after RNA Pol II initiates transcription (Fig. 5). These factors are components of a scaffold on which other factors can assemble to form a reinitiation complex. As the binding of TFIID to promoters is a rate-limiting step in transcription initiation in vivo18,19, such a model can account for the observation that rates of reinitiation are higher than those of initiation8. In this model, the activator could play a dual role in promoting high levels of reinitiation. First, after initiation, activator could directly promote the recruitment of the missing components of the transcription machinery. Second, some activators such as VP-16 can directly stabilize the scaffold complex to promote reinitiation. Our model is supported by in vivo articial recruitment assays in which high levels of transcription are achieved by fusing scaffold components to DNA binding domains1921. Although these high levels of transcription have been interpreted as resulting from an increase in factor recruitment, our model suggests that they could also result from an increase in reinitiation owing to greater scaffold stability. This model predicts that the rate of transcription initiation on a scaffold template would be higher than that on a naked template. Indeed, our experiments have shown that the presence of scaffold stimulates initial rates of transcription 23-fold (see Supplementary Information), consistent with results showing that previously transcribed templates are preferentially transcribed22. Our model also suggests that holopolymerase is not involved in reinitiation. Although there is evidence that Mediator subunits are responsible for transcription of most genes in yeast23, the presence of Mediator in the scaffold suggests that holopolymerase participates only in initiation and not in reinitiation. Instead, reinitiation may involve the recruitment of free RNA Pol II, or RNA Pol II in a distinct complex, to the scaffold. M

Preinitiation complex TFIID Activator TFIIA TATA

TFIIF holo Pol II TFIIB TFIIH TFIIE

Methods
Preparation of nuclear extracts
All yeast strains have been described5, except for Rad3ts14 (ref. 24) and Kin28-ts16 (ref. 25). Nuclear extracts were prepared as described previously26 and on the World-Wide Web (www.fhcrc.org/labs/hahn).

+ NTPs Scaffold TFIID Mediator Activator TFIIA TATA TFIIH TFIIE

Immobilized template assay

Immobilized templates were prepared as described5. PIC formation experiments were performed as described on the World-Wide Web (www.fhcrc.org/labs/hahn). The reactions were run on a 412% NuPAGE gel (NOVEX) and transferred to Immobilon membranes (Millipore). Proteins were detected using Pierce ECL kits. Band intensities were determined by densitometry using IQMACv1.2 software (Molecular Dynamics). Scaffold isolation was performed similarly, except that after washing, PICs were resuspended in 100 ml transcription mix, and incubated with 1 mg HaeIII-digested Escherichia coli DNA competitor, and 100 mM NTPs, ATP or AMPPNP for 2 min at room temperature. The templates were washed once with wash buffer, isolated by digestion with 60 units PstI for 30 min at 37 8C, and processed as described above. For the scaffold stability experiment, scaffolds were isolated as described above, except that after being washed, they were resuspended in transcription mix with 1 mg HaeIII-digested E. coli DNA competitor. Aliquots of 100 ml were removed at the indicated times, washed once with wash buffer, and isolated and processed as described above.

+ NTPs Mediator TFIID Activator TFIIA TATA

+Pol II +TFIIF +TFIIB TFIIF

Reinitiating Pol II TFIIB TFIIE TFIIH

Transcription
We carried out plasmid transcription by incubating wild-type nuclear extract with the HIS4-promoter-containing plasmid, pSH515 or pSH559 as described previously27 and on the World-Wide Web (www.fhcrc.org/labs/hahn). pSH559 was made by digesting pSH515 with BamHI and SfoI to delete 50 base pairs of transcribed sequence. In experiments where both pSH515 and pSH559 were used, the Cyc1 primer (59-GAGAGGCGGTTTGCGTAT TGGG-39) was used for primer extension. We carried out transcription on immobilized templates as described5. The RNA was isolated by phenol:chloroform (2:1) extraction and ethanol precipitation. Primer extension was performed on the RNA samples as described using either the LacI primer or the Cyc1 primer27. For scaffold functional assays, scaffolds were formed as described above using wild-type nuclear extracts. After washing, scaffolds were resuspended in transcription mix containing 120180 mg of a second nuclear extract
NATURE | VOL 408 | 9 NOVEMBER 2000 | www.nature.com

Reinitiation complex
Figure 5 Reinitiation model. When NTPs are added to a pre-initiation complex, RNA Pol II initiates transcription. Whereas TFIIB and TFIIF dissociate from the promoter, activator, TFIID, TFIIA, TFIIH, TFIIE and Mediator are left behind in a scaffold complex. RNA Pol II, TFIIF and TFIIB then reassemble onto the scaffold to form a complex capable of reinitiating transcription. TFIIE is shown as being the least stable scaffold component.
228

2000 Macmillan Magazines Ltd

letters to nature
and 500 ng HaeIII-digested E. coli DNA competitor. NTPs were added to 100 mM either immediately, or after a 40-min incubation. The reactions were stopped after either 2 or 30 min, and analysed by primer extension. For the scaffold competion experiment, scaffolds were formed on the 559 immobilized template as described above, using DSrb2 nuclear extract with 100 ng rSrb2. After washing, an equivalent amount of 515 immobilized competitor template was added, and the reactions were resuspended in transcription mix containing DSrb2 nuclear extract either with or without 100 ng rSrb2. Reactions were incubated for either 10 or 20 min at room temperature. NTPs were then added for 2 min, and the reactions were stopped and analysed by primer extension using the Cyc1 primer. All transcription signals were quantied by PhosphorImager (Molecular Dynamics).
enhancer activity in vitro. Genes Dev. 11, 33273340 (1997). 12. Ho, S. N., Biggar, S. R., Spencer, D. M., Schreiber, S. L. & Crabtree, G. R. Dimeric ligands dene a role for transcriptional activation domains in reinitiation. Nature 382, 822826 (1996). 13. Stringer, K. F., Ingles, C. J. & Greenblatt, J. Direct and selective binding of an acidic transcriptional activation domain to the TATA-box factor TFIID. Nature 345, 783786 (1990). 14. Ozer, J. et al. Molecular cloning of the small (g) subunit of human TFIIA reveals functions critical for activated transcription. Genes Dev. 8, 23242335 (1994). 15. Koh, S. S., Ansari, A. Z., Ptashne, M. & Young, R. A. An activator target in the RNA polymerase II holoenzyme. Mol. Cell 1, 895904 (1998). 16. Lee, Y. C., Park, J. M., Min, S., Han, S. J. & Kim, Y. J. An activator binding module of yeast RNA polymerase II holoenzyme. Mol. Cell. Biol. 19, 29672976 (1999). 17. Yean, D. & Gralla, J. Transcription reinitiation rate: a special role for the TATA box. Mol. Cell. Biol. 17, 38093816 (1997). 18. Klein, C. & Struhl, K. Increased recruitment of TATA-binding protein to the promoter by transcriptional activation domains in vivo. Science 266, 280282 (1994). 19. Xiao, H., Friesen, J. D. & Lis, J. T. Recruiting TATA-binding protein to a promoter: transcription activation without an upstream activator. Mol. Cell. Biol. 15, 57575761 (1995). 20. Farrell, S., Simkovich, N., Wu, Y., Barberis, A. & Ptashne, M. Gene activation by recruitment of the RNA polymerase II holoenzyme. Genes Dev. 10, 23592367 (1996). 21. Keaveney, M. & Struhl, K. Activator-mediated recruitment of the RNA polymerase II machinery is the predominant mechanism for transcriptional activation in yeast. Mol. Cell 1, 917924 (1998). 22. Hawley, D. K. & Roeder, R. G. Functional steps in transcription initiation and reinitiation from the major late promoter in a HeLa nuclear extract. J. Biol. Chem. 262, 34523461 (1987). 23. Holstege, F. C. et al. Dissecting the regulatory circuitry of a eukaryotic genome. Cell 95, 717728 (1998). 24. Guzder, S. N. et al. DNA repair gene RAD3 of S. cerevisiae is essential for transcription by RNA polymerase II. Nature 367, 9194 (1994). 25. Cismowski, M. J., Laff, G. M., Solomon, M. J. & Reed, S. I. KIN28 encodes a C-terminal domain kinase that controls mRNA transcription in S. cerevisiae but lacks cyclin-dependent kinase-activating kinase (CAK) activity. Mol. Cell. Biol. 15, 29832992 (1995). 26. Kang, J. J., Auble, D. T., Ranish, J. A. & Hahn, S. Analysis of the yeast transcription factor TFIIA: distinct functional regions and a polymerase II-specic role in basal and activated transcription. Mol Cell Biol 15, 1234-1243 (1995). 27. Ranish, J. A. & Hahn, S. The yeast general transcription factor TFIIA is composed of two polypeptide subunits. J. Biol. Chem. 266, 1932019327 (1991).

PIC and RNA quantication

The total number of PICs formed in an immobilized template assay was quantied by a western blot comparison with known amounts of puried recombinant TBP, TFIIA, TFIIB and TFIIE. Band intensities were determined by densitometry using IQMACv1.2 software (Molecular Dynamics). We determined the number of active PICs by quantifying the amount of RNA produced in a single round of transcription. The amount of RNA produced in a single round of immobilized template transcription was assayed by S1 nuclease protection using a 59-end-labelled DNA oligonucleotide probe (59-GTAAACT ATTGTATTACTATTACACAGCGCAGGGTGTAG-39). S1 nuclease protection using the same probe was also performed on increasing quantities of a 30-nucleotide RNA oligonucleotide standard to quantify the amount of RNA produced in the transcription reaction. This amount was taken to be equivalent to the number of active PICs.
Received 3 August; accepted 25 September 2000. 1. Hahn, S. Activation and the role of reinitiation in the control of transcription by RNA polymerase II. Cold Spring Harb. Symp. Quant. Biol. 63, 181188 (1998). 2. Roberts, S. G., Choy, B., Walker, S. S., Lin, Y. S. & Green, M. R. A role for activator-mediated TFIIB recruitment in diverse aspects of transcriptional regulation. Curr. Biol. 5, 508516 (1995). 3. Zawel, L., Kuman, K. P. & Reinberg, D. Recycling of the general transcription factors during RNA polymerase II transcription. Genes Dev. 9, 14791490 (1995). 4. Sandaltzopoulos, R. & Becker, P. B. Heat shock factor increases the reinitiation rate from potentiated chromatin templates. Mol. Cell. Biol. 18, 361367 (1998). 5. Ranish, J. A., Yudkovsky, N. & Hahn, S. 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, 4963 (1999). 6. Kim, Y. J., Bjorklund, S., Li, Y., Sayre, M. H. & Kornberg, R. D. A multiprotein mediator of transcriptional activation and its interaction with the C-terminal repeat domain of RNA Polymerase II. Cell 77, 599608 (1994). 7. Koleske, A. J. & Young, R. A. An RNA polymerase II holoenzyme responsive to activators. Nature 368, 466469 (1994). 8. Jiang, Y. & Gralla, J. D. Uncoupling of initiation and reinitiation rates during HeLa RNA polymerase II transcription in vitro. Mol. Cell. Biol. 13, 45724577 (1993). 9. Svejstrup, J. Q. et al. Evidence for a mediator cycle at the initiation of transcription. Proc. Natl Acad. Sci. USA 94, 60756078 (1997). 10. Kraus, W. L. & Kadonaga, J. T. p300 and estrogen receptor cooperatively activate transcription via differential enhancement of initiation and reinitiation. Genes Dev. 12, 331342 (1998). 11. Sheridan, P. L., Mayall, T. P., Verdin, E. & Jones, K. A. Histone acetyltransferases regulate HIV-1

Supplementary information is available at Nature's World-Wide Web site (http://www.nature.com) or as paper copy from the London editorial ofce of Nature.

Acknowledgements
We thank members of the Hahn and Reeder laboratories for helpful discussions, and A. Krumm, S. Parkhurst and R. Reeder for comments on the manuscript. We also thank L. Prakash for providing the Rad3ts strain, M. Solomon for providing the Kin28ts strain, D. Reinberg for TFIIE antibodies and H. Sakurai for providing the Tfa1ts strain and antibodies to Gal11. This work was supported by grants from the NIH to S.H. and an NIH training grant to N.Y. S.H. is an associate investigator of the Howard Hughes Medical Institute. Correspondence and requests for materials should be addressed to S.H. (e-mail: shahn@fhcrc.org).

NATURE | VOL 408 | 9 NOVEMBER 2000 | www.nature.com

2000 Macmillan Magazines Ltd

229