CA2383698A1 - In vitro transcription systems and uses - Google Patents

Abstract

The invention relates to methods of identifying agents that interact with retinoid X receptor dimers. The invention also relates to in vitro chromatin based DNA template transcription systems.

Description

In Vitro Transcription Systems and Uses Background of the Invention Field of the Invention The invention relates to methods of identifying agents that interact with retinoid X receptor dimers. The invention also relates to in vitro chromatin based DNA template transcription systems.
Related Art Retinoic acids (RAs) exert their pleiotropic effects on vertebrate development and homeostasis by binding to nuclear receptors (NRs) (Kastner, P., et al., Cell 83:859-869 (1995), and references therein). These receptors belong to a gene superfamily that includes the receptors for steroid hormones, thyroid hormones, vitamin D3, and a growing number of so-called orphan receptors (for reviews, see Gronemeyer, H., & Laudet, V., Protein Profile 2:1173-1236 (1995);
Perlmann, T. & Evans, R.M., Ce1190:391-397 (1997)). Two families ofreceptors, the retinoic acid receptor isotypes (RARa, RAR~3 and RARy) and the retinoid X
receptors isotypes (RXRa, RXR~3 and RXRy) are implicated in the transduction of the RA signal (Chambon, P., FASEB J. 10:940-954 (1996), and references therein). RARs bind all-traps RA (tRA) and 9-cis RA (9cRA), whereas RXRs respond exclusively to 9cRA (Allenby, G., et al., Proc. Natl. Acad. Sci. USA
90:30-34 (1993), and references therein). The C-terminal region of RARs and RXRs contains both the ligand binding domain (LBD), which functions as a ligand-dependent transactivation domain (activation function 2 (AF-2)), and surfaces for both homo- and hetero-dimerization as well as for interaction with other factors (see below). An additional ligand-independent activation function, AF-1, is present within the N-terminal region (see, Chambon, P., FASEB J.
10:940-954 (1996)).
RARs and RXRs can bind as dimers to RA response elements (R.AREs) consisting of two hexameric motifs [PuG(G/A)(T/A)CA] usually arranged as direct repeats. However, RXRs readily heterodimerize with RARs, and R;XR/RAR heterodimers bind to and transactivate from RAREs made up of direct repeat motifs separated by 5 (DRS) and 2 (DR2) by much more efficiently than RXR homodimers on their own. This indicates that ~ heterodimers might be the functional units transducing the retinoic acid signals in vivo (Chambon, P., FASEB J. 10:940-954 (1996); Leid, M., et al., Trends Biochem.
Sci. 17:427-433 (1992); and references therein). Several lines of evidence support this possibility: (i) genetic studies have established the functionality of RXR/RAR
heterodimers in the RA-responsive F9 embryonal carcinoma cell line (Clifford, J., etal., EMBOJ. 15:4142-4155 (1996); Chiba, H., etal., J. Cell. Science 139:735-747 (1997); Chiba, H., et al., Mol. Cell. Biol. 17:3013-3020 (1997)), as well as in the mouse (Kastner, P., et al., Cell 83:859-869 (1995); Kastner, P., et al., Development 124:313-326 (1997); Kretzel, W., et al., Science 279:863-867 (1998); Mascrez, B., et al., Development 125:4691-4707 (1998); and references therein), and (ii) synergistic effects of RXR- and RAR-selective synthetic retinoid on target gene expression, proliferation, apoptosis and/or differentiation have been observed in a variety of cultured cell lines, including the embryonal carcinoma cell lines F9 and P 19 (Clifford, J., et al., EMBO J. 15:4142-4155 ( 1996); Chiba, H., et al., J. Cell. Science 139:735-747 (1997); Chiba, H., et al., Mol. Cell.
Biol.
17:3013-3020 (1997); Apfel, C.M., et al., J. Biol. Chem. 270:30765-30772 (1995); Lotan, R., et al., Cancer Res. 55:232-236 (1995); Roy, B., et al., Mol.
Cell. Biol. 15:6481-6487 (1995); Chen, J-Y., et al., Nature 382:819-822 (1996);
Horn, V., et al., FASEBJ. 10:1071-1077 (1996); La Vista-Picard, N., et al., Mol.
Cell. Biol. 16:4137-4146 (1996); Taneja, R., et al., Proc. Natl. Acad. Sci.
USA
93:6197-6202 (1996); Taneja, R., etal., EMBOJ. 16:6452-6465 (1997); Botling, J., et al., J. Biol. Chem. 272:9443-9449 (1997); Minucci, S., et al., Mol.
Cell.
Biol. 17:644-655 (1997); Joseph, B., et al., Blood 91:2423-2432 (1998)).
However, in all cases, the liganded RXR was transcriptionally inactive, unless its RAR partner was itself liganded. This intra-heterodimeric subordination of the RXR AF-2 activity to the binding of a RAR agonistic ligand could be caused by an allosteric effect ofthe unliganded RAR on its liganded RXRpartner (Vivat, V., et al., EMBOJ. 16:5698-5709 (1997)).

Transfection studies have suggested that the AF-2 activation function of NRs is mediated through coactivators (intermediary factors) (Tasset, D., et al., Cell 62:1177-1187 (1990)). Numerous proteins that interact directly with NRs in an agonistic ligand-dependent manner have been cloned and characterized, and several of them have been shown to enhance the activity of NR AF-2s when co-expressed in transiently transfected mammalian cells (Chambon, P., FASEB J.
10:940-954 (1996); Glass, C.K., etal., Curr. Opin. Cell. Biol. 9:222-232 (1997);
and references therein). Some of these putative coactivators, SRC-I (Spencer, T.E., et al., Nature 389:191-198 (1997)), CBP/p300 (B nnister, A.J. &
Kouzarides, T., Nature 384:641-643 (1996); Ogryzko, V.V., et al., Cel187:953-959 (1996)) and ACTR (Chen, H., et al., Cell 90:569-580 (1997)) can interact with the histone acetyltransferase p/CAF (Yang, X-J., et al., Nature 382:319-( 1996)) and also possess an intrinsic histone acetyltransferase activity.
Moreover, CBP and p300 also interact with RNA helicase A, which in turn binds RNA
polymerase II (Nakajima, T., et al., Cell 90:1107-1112 (1997)).
It has been shown that RXRs form heterodimers in solution with either RARs, TRs or VDR and that the receptor domains required for heterodimeric interactions overlap with the LBD of each receptor. Ligand dependent transcription activation by the RXRIVDR heterodimer has been shown (Rachez, C. et al., Nature 398:824-828 (1999)). The formation of heterodimers between RXRs and PPARs was also demonstrated (Kliewer, S.A. et al. Nature 358:771-774 (1992); Bardot, O. etal., Biochem. Biophys. Res. Comm. 192:37-45 (1993)).
RXR also heterodimerizes with liver X receptors (LXRs; Apfel et al., Mol. Cell Biol. 14:7025-7035 (1994), farnesoid X receptor (FXR; Forman et al., Cell 81:687-693 (1995)), benzoate X receptor (BXR; Blumberg et al., Genes Dev.
12:1269-1277 (1998)), constitutively active receptor or constitutive androstane receptors (CARS; Choi et al., J. Biol. Chem. 272:23565-23571 (1997); Forman et al., Nature 395:612-615 (1998)), and steroid and xenobiotic receptor (SXR;
Blumberg et al., Genes Dev. 12:3195-3205 (1998)).
In contrast to prokaryotes, genomes in eukaryotes are packaged into chromatin. Biochemical and genetic studies in vivo and in vitro, often performed with yeast systems, have demonstrated that transcriptional initiation is severely inhibited on chromatin templates due to the presence of nucleosomes that limit access of sequence-specific DNA-binding activators, coactivators and polymerase II (pol II) basic transcriptional machinery to the DNA template. Thus, alleviating nucleosomal repression through remodeling chromatin structure is of critical importance to increase the accessibility of DNA for protein interaction, thereby potentiating gene expression (reviews include Kadonaga, J.T., Cell 92:307-313 (1998); Kornberg, R.D. and Lorch, Y., Cell 98:285-294 (1999); Struhl, K., Cell 98:1-4 (1999)). To this end conserved mechanisms have evolved that involve at least two classes of chromatin modifying activities. The first class consists of several ATP-driven chromatin remodeling complexes that modify chromatin structure by affecting the position andlor stability of nucleosomes. The second class corresponds to a variety of core histone acetyltransferases (HAT) (recent reviews on chromatin modifying activities include Varga-Weisz, P.D. and Becker, P.B., Curr. Opin. Cell Biol. 10:346-353 (1998); Workman, J.L. and Kingston, R.E., Ann. Rev. Biochem. 67:545-579 (1998); Kingston, R.E. and Narlikar, G.J., Genes Dev. 13:2339-2352 (1999); Muchardt, C. and Yaniv, M., J. Mol. Biol.
293:187-198 (1999); Brown, C.E., et al., Trends Biochem. Sci. 25:15-19 (2000);
Strahl, B.D. and Allis, C.D., Nature 403:41-45 (2000)).
Summary of the Invention The invention is directed to methods of identifying agents that interact with retinoid X receptor dimers. The invention is also directed to in vitro chromatin based DNA template transcription systems.
Brief Description of the Figures FIG. 1. DNA templates and S 1 nuclease probe. The structures of the (DRS)5(32G, (17m)5(32G and internal control pGl reporter templates are schematically represented with the positioning of the response elements relative to the transcription start site.

FIG. 2. Analysis of RARa/RXRa heterodimers and chromatin structure.
FIG. 2(A). Purification of RARa/RXRa heterodimers: FhRARa and HmRXRa were co-expressed in Sf9 cells and affinity-purified using a Niz+ column followed by anti-Flag agarose column that bind the hImRXR moiety and the FhRAR moiety of the heterodimer, respectively. Purified heterodimers ( 100 ng of protein) were separated on a 10% SDS-PAGE gel before staining with Comassie Blue (lane 1) or Western blot analysis using monoclonal antibodies recognizing either human RARa (lane 2) or mouse RXRa (lane 3). FIG. 2(B). Overall chromatin structure was not affected by RARa/RXRa heterodimers: chromatin or naked (DRS)5 ~32G
templates (200 pM) incubated in the presence or absence of FhR.ARa/HmRXRa (1 nM) and tRA (10-6 M) were digested with varying concentrations of micrococcal nuclease in a final volume of 80 ~1, separated on a 1.5% agarose gel, and Southern blotted using a [32P] probe corresponding to the -40 to +5 region of the (DRS)5~32G promoter. DNA supercoiling was estimated as described (Becker, P.B., et al., Methods Cell Biol. 44:207-223 (1994)) on DNA (200 ng) treated (or not treated) by topoisomerase I ( 10 units; final volume of 45 ~1). DNA
was separated on a 1 % agarose gel in the presence or absence of 1.2 ~M
chloroquine. Migration of relaxed and supercoiled template DNA is indicated.
FIG. 3. RARa/RXRa heterodimers activate transcription from chromatin templates in a ligand- and template-specific manner. FIG. 3(A). tRA-induced derepression of transcription from chromatin templates by RARa/RXRa. In vitro transcription was performed on chromatin or naked (DRS)5(32G templates (200 pM) using a HeLa cell nuclear extract (100 fig) for 45 min in the presence or absence of FhRARa/HmRXRa ( 1 nM) and tRA ( 1 ~IVI) in a final reaction volume of 50 ~l as indicated. S 1 nuclease analysis was carried out after deproteinization. FIG. 3(B). Template specificity of activated transcription.
Activation of transcription on chromatin (DRS)S~i2G or (17M)5(32G templates was determined in the presence of 1 nM of activator (either Gal4(1-147), Gal4-VP 16 or FhRARa/HmRXRa) with or without tRA ( 1 ~.M) as above. S 1 nuclease digestion of RNA transcripts originating from ~32G and pG 1 templates generated 179- and 60-nt fragments, respectively (see FIG. 1 ).

FIG. 4. RARa/RXRa heterodimers bind all five R.AREs in the promoter region ofthe (DRS)5~32G chromatintemplate, irrespective ofthe presence oftRA.
Chromatin or naked (DRS)5 ~i2G templates (250 ng) were incubated in the presence or absence of FhR.ARa/HmRXRa and tRA (10-6 M) (under the conditions described above for transcription reactions) for 30 min, subjected to DNase I digestion (5 units; final volume 50 ~1), then analyzed by primer extension foot printing (see Materials and Methods, Example 1 ). Sites of increased (closed triangle) or decreased (open triangle) sensitivity to DNase I are shown.
FIG. 5. Dose-dependent synergistic effects of specific retinoids on activation of transcription by RARa/RXRa heterodimers. FIG. 5(A). Dose-dependent activation by tRA and 9cRA. Transcription reactions were performed as described in FIG. 3 on (DRS)5(32G template by using FhRARa/HmRXRa in presence of varying concentrations (5x10-'° to 10-6 M) of tRA (open circles) or 9cRA (closed squares). FIG. 5(B). Receptor-selective and synergistic activation of transcription. Transcription reactions were performed as described above using synthetic retinoid agonists and antagonists at the concentrations indicated.
The receptor specificity of retinoids used are as follows: tRA (panRAR-specific ligand), 9cRA (panR.AR- and panRXR-ligand), Compound I (R.ARa-specific agonist), Compound IV (RARE-specific agonist), SR11237 (panRXR-specific agonist), and Compound II (RARa-specific antagonist). Transactivation by FhR.ARa/HmRXRa is expressed relative to that observed from the internal control template (pGl). Induction by tRA (10-6 M) was arbitrarily set to 100%.
All points are the average of at least two independent experiments run in duplicate.
FIG. 6. p300 enhances transactivation by RARa/RXRa heterodimers in vitro. FIG. 6(A). Addition of exogenous acetylCoA (AcCoA) does not effect ligand-dependent transactivation by RARa/RXRa. Transcription reactions were performed in the presence or absence of acetylCoA ( 1 ~M) on naked or chromatin (17M)5~32G or (DRS)S~i2G templates plus or minus 1 nM activator (either Gal4(1-147), Gal4-VP16 or FhRARa/HmRXRa) and/or tRA (1 ~M), as described in FIG. 3. FIG. 6(B). Addition of acetylCoA does not further enhance p300-activated transcription. Transcription was performed on (DRS)5~32G

-7_ templates in the presence or absence of FhRARa/1=ImRXRa and/or tRA
(5x10-8 M). Where indicated, the co-activator p300 (0.5 nM) and Acetyl CoA
( 1 g.M) were added.
FIG. 7. FIG. 7(A). Histone H1 increases nucleosome repeat length from 170 to 200 by in a chromatin structure that is more resistent to micrococcal nuclease (MNase I) digestion. Chromatin (DRS)5~32G template (160 ng) with (+H 1 ) or without (-H 1 ) histone H 1 were digested with 6.5 units MNaseI
(final volume 55 g1, 24°C) for the time indicated in seconds (sec.). Following deproteinization and separation on a 1.2% agarose gel in O.~x TBE, DNA was stained with ethidium bromide. FIG. 7(B). Incorporation of Histone H1 into "crude" chromatin (DRS)5(32G templates does not render transactivation by liganded RARa/RXRa heterodimers Acetyl CoA-dependent. In vitro transcription was performed (Example 2, Experimental Procedures) using a HeLa cell nuclear extract and purified RARa/RXRa heterodimers on "crude" chromatin (DRS)5~32G templates assembled in the presence (+H1) or absence (-H1) of histone H1. TIF2, p300, tRA/SR 11237, and Acetyl CoA were added as indicated. S 1 nuclease digestion of RNA transcripts originating from the (DRS)5(32G and internal "control" pGl templates generated 179 and 60 nt-long fragments, respectively (see Dilworth et al., 1999). FIG. 7(C). Histone H1 remains associated with chromatin template after size-exclusion chromatography.
"Purified" chromatin (DRS)5~32G template ( 160 ng) assembled in the presence of histone H1 was digested with 0.35 units of MNasel and analyzed as described above. FIG. 7(D). SWI/SNF endogenous to the chromatin assembly extract is removed during size exclusion chromatography. "Crude" or "purified" chromatin (DRS)5~32G templates (70 ng) assembled in presence or absence of histone Hl were applied to a nitrocellulose membrane using a slot blot apparatus, and then probed for the presence of dSWI/SNF using antibodies directed against dBRM.
FIG. 7(E). Purification of human SWI/SNF complexes. Human SWI/SNF
S WI/SNF complexes were purified from HeLa cells expressing Flag-tagged Ini 1.
Purified SWI/SNF was separated on a 10% SDS-PAGE gel before staining with Coomassie Blue (lane 1 ) or Western blot analysis using mAbs recognizing SNF2b or Flag-Ini 1 (lane 2). FIG. 7(F). Chromatin purification removes some HAT

_g_ activity associated with the "crude" chromatin preparation. "Crude" or "purified"
chromatin (DRS)5~32G templates (30 ng) assembled in presence or absence of histone H1 were incubated with free core histones (2 fig) and 0.1 ~Ci of [Acetyl-1-'4C] Acetyl CoA at 30°C. Ovalbumin and p300 (1 pmol of each) were used as a negative and positive controls, respectively. After 1 hr, radiolabelling of histones was visualized (Example 2, Experimental Procedures). FIG. 7(G). Both TIF2 and p300 possess intrinsic HAT activity. HAT assays were as described in panel D using 2 ~.g of free core histones or purified nucleosomes in the presence of either TIF2 (1 or 3 pmol), p300 (1 or 3 pmol), or ovalbumin (1 pmol), as indicated. A 25-fold excess of cold Acetyl CoA was added to reactions as indicated to confirm that transfer of'4C to the histones was specific.
FIG. 8. Requirements for activation of transcription on purified chromatin templates. FIG. 8(A). Transcriptional flow chart. FIG. 8(B). Ligand-dependent activation of transcription by RARa/RXRa heterodimers on purified chromatin templates required coactivator acetyltransferase and ATP-dependent chromatin remodeling activities. Transcription was performed using a HeLa nuclear extract on "crude" or "purified" chromatin (DRS)5~32G templates assembled in the presence or absence ofhistone H1. RARa/RXRa heterodimers, tRA/SR 11237, p300, TIF2, hSWI/SNF, ATP, and/or Acetyl CoA were included as indicated.
Transcription reactions was analyzed as described in Figure 7(B). FIG. 8(C).
Acetyl CoA and ATP are required for transactivation on "purified" chromatin.
Transcription was performed on "purified" (DRS)S~i2G chromatin templates assembled in the presence of histone H1 using a HeLa nuclear extract, RARa/RXRa heterodimers, and tRA/SR 11237 ligands. Different combination of p300, TIF2, hSWI/SNF, ATP and Acetyl CoA were included as indicated.
FIG 8(D). TIF2 and p300 stimulate activation of transcription by RARa/RXRa heterodimers. Transcription was performed on "purified" (DRS)5(32G chromatin templates assembled in the presence of histone H 1 using a HeLa nuclear extract, RARa/RXRa heterodimers, ATP, and hS WI/SNF. Acetyl CoA, p300, TIF2, and tRA/SR 11237 were included as indicated.
FIG. 9. Both ATP-driven chromatin remodeling factors and coactivator acetyltransferase activities synergistically contribute to stimulation of transcription by enhancing the formation of productive initiation complexes, rather than by increasing the relative frequency of reinitiation events. Transcription was performed on "purified" (DRS)S~i2G chromatin templates assembled in the presence ofhistone H 1. RARa/RXRa heterodimers, tRA/SR 11237, p300, TIF2, hSWI/SNF, ATP, and/or Acetyl CoA were included as indicated. Sarkosyl (0.5%
final) was added 1 min after rNTPs to restrict transcription to one round.
Relative transcription is that observed in the absence of sarkosyl (multiple rounds) and is expressed relative to that achieved when all components of the system were present (lane 12). Rounds oftranscription was calculated by dividing the intensity of the signal in the absence of sarkosyl (multiple rounds) by that in the presence of sarkosyl (single round) and is rounded off to a whole number. Similar results were obtained in several experiments carried out with different chromatin preparations and sarkosyl concentrations.
FIG.10. ATP and ISWI-containing complexes are required for efficient 1 S binding of RARa/RXRa heterodimers to their response elements (RARE) on chromatintemplates. FIG.10(A)."Purified"(DRS)5~32Gchromatintemplates(1 nM) assembled in the presence of histone H1 were incubated as indicated in the presence or absence of RARa/RXRa (5 nM), tRA (1 ~.M), SR 11237 (1 pM), TIF2 (2.5 nM), p300 (2.5 nM), hSWI/SNF (10 ng/~1), ATP (0.1 mM), and/or Acetyl CoA (2 ~,M) for 30 min at 27°C, and then processed for DNAse I
footprinting (Experimental Procedures). FIG. 10(B). Same as (A) except that hISWI (5 ng/~1) was added to reactions were indicated.
FIG.11. ATP-dependent remodeling of (DRS)5(32G chromatin templates in the presence of RAR/RXR heterodimers is limited to the region containing the REs. "Purified" (DRS)5(32G chromatin templates (1 nM) were incubated as indicated in the presence or absence of RARa/RXRaa (S nM), and/or ATP (0.1 mM) at 27°C. After 30 min chromatin templates were digested with 1 unit MNaseI (final volume 60 ml, 24°C) for the time indicated in seconds (sec.).
Following deproteinization and separation on a 1.2% agarose gel in O.Sx TBE, DNA was transfered to a nylon membrane and the chromatin structure at specific regions was examined using [32P]-labelled oligonucleotides recognizing either the DRS binding elements (-205 to -56), the transcriptional start site (-8 to +37), or the 3' untranslated region (+2249 to +2277) of the (DRS)5~32G template.
FIG.12. RAR/RXR heterodimers target histone acetyltransferase activity to (DRS)5(32G promoter region in a ligand-dependent manner. "Purified"
chromatin (DRS)5(32G (200 pM) and pSGS (200 pM) templates were incubated in the presence of RARa/RXRa heterodimers (1 nM), hSWI/SNF (2 ng/~1), ATP
(0.1 mM), Acetyl CoA (2 ~M), tRA and SR 11237 (1 ~M each), and/or TIF2 and p300 (500 pM each) as indicated. ChIP assays were then performed (Example 2, Experimental Procedures). The data displayed is representative of three independent experiments.
FIG.13. ATP-driven chromatin remodeling activities and coactivators act sequentially to mediate the ligand-dependant stimulatory effect of RARa/Rxita heterodimers on transcription initiation. FIG.13(A). Transcriptional timeline and schematic diagram indicating the "normal" time of addition of the different components of the in vitro transcription system. FIG. 13(B). Effect of varying the time of addition of RARa/RXRa heterodimers and tRA/SR 11237.
Transcriptionwas performed on "purified" (DRS)5~32GH1-containing chromatin templates using a HeLa nuclear extract, ATP, hSWI/SNF, AcetylCoA, p300, and TIF2. RARa/RXRa heterodimers and/or tRA/SR 11237 were then added at the times indicated. FIG.13(C). Effect of varying the time of addition of p300/TIF2 and hSWI/SNF, as indicated. FIG.13(D). Effect of varying the time of Acetyl CoA addition as indicated. FIG. 13(E). Effect of varying the time of ATP
addition as indicated. The activators, cofactors, coactivators or agonistic ligands were introduced as indicated at either the same time as the chromatin template (-40 min relative to transcription initiation), 20 min later (-20 min), at the same time as (-10 min) or 5 min after (-5 min) the HeLa nuclear extract, 0.5 min before the rNTPs (-0.5 min), or not at all (none or -). The experiments displayed in panels B-E were repeated at least three times with different chromatin preparations and yielded similar results.

Detailed Description of the Preferred Embodiments It has been discovered that an in vitro chromatin based DNA template transcription system, in the presence of R:XR/RAR heterodimers, mimics the effects of retinoids on gene transactivation as observed in vivo. Activation of transcription by RXR/RAR heterodimers depends on packaging of the template into a nucleosomal structure and that it is specific, in that it requires the heterodimer, a cognate ligand, and a cognate response element. Moreover, it has been discovered that the agonist-bound transcription activation function of RXR
can act synergistically with that of RAR but that the binding of an agonist to RAR
is a prerequisite for effective activation of transcription by agonist-bound RXR.
It has further been discovered that RAR/RXR heterodimers cannot efficiently initiate transcription from a chromatin template from which transcription co-regulators have been removed, unless the chromatin template is exposed to two types of chromatin modifying activities that synergize to activate transcription:
ATP-driven chromatin remodeling activities and histone acetyltransferase activities.
The invention is directed to a method of identifying an agent which interacts with a retinoid X receptor (RXR) dimer, the method comprising: (a) adding an agent to a chromatin based DNA template in the presence of the RXR
dimer; and (b) measuring activation of transcription, thereby determining whether the agent interacts with the RXR dimer. In the method, activation of transcription can be compared to the method performed in the absence of the agent or in the presence of a known agent. In one embodiment, transcription co-regulators are not removed from the chromatin based DNA template prior to adding an agent.
In another embodiment, the method further comprises removing one or more or all transcription co-regulators. In the above embodiments, a co-activator can be added before or after adding the agent. The agent can be a co-activator or an agent that mediates an interaction between a RXR dimer and a co-activator.
Thus, this method is useful in identifying agents that regulate interactions between RXR dimers and co-activators.

Another embodiment of the invention is directed to a method of identifying a retinoic acid receptor (RAR) agonist, the method comprising: (a) adding an agent to a chromatin based DNA template in the presence of an R;~~t/RAR dimer;
and (b) measuring activation of transcription, thereby determining whether the agent is an RAR agonist. In the method, activation of transcription can be compared to the method performed in the absence of the agent or in the presence of a known RAR agonist. In one embodiment, transcription co-regulators are not removed from the chromatin based DNA template prior to adding an agent. In another embodiment, the method further comprises removing one or mofe or all transcription co-regulators. In the above embodiments, a co-activator can be added before or after adding the agent. The agent can be a co-activator or an agent that mediates an interaction between a RXR dimer and a co-activator.
Thus, this method is useful in identifying agents that regulate interactions between RXR dimers and co-activators.
The invention is also directed to a method of identifying an RXR agonist, the method comprising: (a) adding an agent to a chromatin based DNA template in the presence of an R~~/RAR dimer and an RAR agonist; and (b) measuring activation of transcription, thereby determining whether the agent is an RXR
agonist. In the method, activation of transcription can be compared to the method performed in the absence of the agent or in the presence of a known RXR
agonist.
In one embodiment, transcription co-regulators are not removed from the chromatin based DNA template prior to adding an agent. In another embodiment, the method further comprises removing one or more or all transcription co-regulators. In the above embodiments, a co-activator can be added before or after adding the agent. The agent can be a co-activator or an agent that mediates an interaction between a RXR dimer and a co-activator. Thus, this method is useful in identifying agents that regulate interactions between RXR dimers and co-activators.
The invention is ftwther directed to a method of identifying an RAR
antagonist, the method comprising: (a) adding an agent to a chromatin based DNA template in the presence of an R~~R/RAR dimer and an RAR agonist; and (b) measuring activation of transcription, thereby determining whether the agent is an RAR antagonist. In the method, activation of transcription can be compared to the method performed in the absence of the agent or in the presence of a known RAR antagonist. In one embodiment, transcription co-regulators are not removed from the chromatin based DNA template prior to adding an agent. In another embodiment, the method further comprises removing one or more or all transcription co-regulators. In the above embodiments, a co-activator can be added before or after adding the agent. The agent can be a co-activator or an agent that mediates an interaction between RXR dimers and a co-activator.
Thus, this method is useful in identifying agents that regulate interactions between RXR
dimers and co-activators. In the above embodiments, the method further comprises adding a co-repressor. A co-repressor can be added before or after adding the agent. The agent can be a co-repressor or an agent that mediates an interaction between a RXR dimer and a co-repressor. Thus, this method is useful in identifying agents that regulate interactions between RXR dimers and co-repressors.
The invention is directed to a method of identifying an RXR antagonist, the method comprising: (a) adding an agent to a chromatin based DNA template in the presence of a RXR/RAR dimer, an RAR agonist, and an RXR agonist; and (b) measuring activation of transcription, thereby determining whether the agent is an RXR antagonist. In the method, activation of transcription can be compared to the method performed in the absence of the agent or in the presence of a known RXR antagonist. In one embodiment, transcription co-regulators are not removed from the chromatin based DNA template prior to adding an agent. In another embodiment, the method further comprises removing one or more or all transcription co-regulators. In the above embodiments, a co-activator can be added before or after adding said agent. The agent can be a co-activator or an agent that mediates an interaction between a RXR dimer and a co-activator.
Thus, this method is useful in identifying agents that regulate interactions between RXR dimers and co-activators. In the above embodiments, the method further comprises adding a co-repressor. A co-repressor can be added before or after adding said agent. The agent can be a co-repressor or an agent that mediates an interaction between a RXR dimer and a co-repressor. Thus, this method is useful in identifying agents that regulate interactions between RXR dimers and co-repressors.
The invention is directed to a method of identifying a co-activator of an RXR dimer, the method comprising: (a) adding a first agent to a chromatin based DNA template in the presence of the RXR dimer and an agonist of the RXR
dimer; and (b) measuring activation of transcription, thereby determining whether the first agent is a co-activator of the RXR dimer. In the method, activation of transcription can be compared to the method performed in the absence of the first agent or in the presence of a known co-activator. In another embodiment, this method can be used for identifying a co-repressor of the RXR dimer. In the method, activation of transcription can be compared to the method performed in the absence of the first agent or in the presence of a known co-repressor. In one embodiment, transcription co-regulators are not removed from the chromatin based DNA template prior to adding the first agent and a RXR dimer agonist. In another embodiment, the chromatin based DNA template is purified such that one or more or all transcription co-regulators are removed prior to adding the first agent and the agonist of the RXR dimer.
The invention is further directed to a method of identifying a modulator which modulates interactions between a RXR dimer and a co-activator of the RXR dimer, the method comprising: (a) adding an agent to a chromatin based DNA template in the presence of the RXR dimer, an agonist of the RXR dimer, and a co-activator of the RXR dimer; and (b) measuring activation of transcription, thereby determining whether the agent modulates interactions between the RXR dimer and the co-activator of the RXR dimer. In the method, activation of transcription can be compared to the method performed in the absence of the agent. In another embodiment, this method can be used for identifying a co-repressor of the RXR dimer. In the method, activation of transcription can be compared to the method performed in the absence of the agent. In one embodiment, transcription co-regulators are not removed from the chromatin based DNA template prior to adding an agent, RXR dimer agonist, and a co-activator. In another embodiment, the chromatin based DNA template is purified such that one or more or all transcription co-regulators are removed prior to adding the first agent, the agonist of the RXR dimer, and a co-activator.
Agents identified by the above methods can be useful in treating a variety of conditions and diseases including, but not limited to, lung cancer, mesothelioma, photodamaged skin, fine and course wrinkling, improper skin pigmentation, skin roughness, premalignant skin growths such as actinic keratoses, diseases of the nervous system, improper regulation of cellular growth and differentiation, visual impairment, acute promyelocytic leukemia, and basal cell carcinoma.
The invention is directed to an in vitro chromatin based DNA template transcription system comprising: (a) a chromatin based DNA template; and (b) an RXR dimer. In one embodiment, transcription co-regulators are not removed from the chromatin based DNA template. In another embodiment, the chromatin based DNA template is purified such that one or more or all transcription co-regulators are removed. The system can further comprise a co-activator and/or a co-repressor. The invention is also directed to a kit comprising the in vitro chromatin based DNA template transcription system.
Each of the terms and elements of the invention as described in the above embodiments are detailed as follows.
By a "nuclear receptor" or "nuclear receptor superfamily receptor" or "steroid/thyroid hormone receptor superfamily" is intended a ligand-dependent transcription factor that regulates the expression of target genes involved in metabolism, development, and reproduction. Nuclear receptors include receptors for which specific ligands have not yet been identified (termed "orphan receptors"). These hormone binding proteins can bind to specific DNA sequences to modulate transcriptional activity of a target gene, upon binding of a ligand to the receptor. Exemplary nuclear receptors include, but are not limited to, retinoic acid receptors (RARs; a, ~3 and 'y), retinoid X receptors (RXRs; a, (3 and y), vitamin D3 receptor (VDR), thyroid receptors (TRs; a and Vii), peroxisome proliferator activated receptors (PPARs; a, Vii, 8 and y), liver X receptors (LXRs;
a and ~3) (Willy, P.J. et al., Genes Dev. 9:1033-1045 (1995); Willy, P.J. et al., GenesDev.11:289-298 (1997); Mukherjee, R. etal., Nature 386:407-410 (1997);
Peet, D.J. et al., Curr. Opin. Genet. Dev. 8:571-575 (1998); Janowski, B.A. et al., Proc. Natl. Acad. Sci. USA 96:266-271 (1999)), farnesoid X receptor (FXR) (Makishima, M. etal., Science 284:1362-1365 (1999)); Wang, H. etal., Mol. Cell 3:543-553 (1999)), benzoate X receptor (BXR) (Blumberg et al., Genes Dev.
12:1269-1277 (1998)), constitutively active receptor or constitutive androstane receptors (CARs; a and ~3) (Choi etal., J. Biol. Chem. 272:23565-23571 (1997);
Forman et al., Nature 395:612-615 (1998)), and steroid and xenobiotic receptor (SXR) (Blumberg et al., Genes Dev. 12:3195-3205 (1998)), HNF4 (Sladek et al., Genes Dev. 4:2353-2365 (1990)), the COUP family of receptors (Miyajima et al., Nucl. Acids Res. 16:11057-11074 (1988)), nerve growth factor-induced receptor (NGFI-B) (Crawford, P.A. et al., Mol. Cell. Biol. 15:4331-4336 (1995)), ultraspiracle receptor (Oro et al., Nature 347:298-301 (1990)), and the like.
By an "RXR dimer" is intended a dimer formed by an RXR (a, ~3 or y) and a second nuclear receptor, and includes an RXR homodimer and RXR
heterodimer. By an "RXR homodimer" is intended a dimer of an RXR (a, (3 or y) and another RXR (a, (3 or y). By an "RXR heterodimer" is intended a dimer of an RXR (a, ~3 or y) and a non-RXR nuclear receptor capable of dimerizing with an RXR, including, but not limited to, an RAR (a, ~i or y), VDR, TR (a or Vii), PPAR (a, (3, b or y), LXR (a or ~3), BXR, CAR (a or Vii), SXR and FXR.
Preferred non-RXR nuclear receptors capable of dimerizing with an RXR include RARs, TRs, PPARs, LXRs, BXR, CARs, SXR and FXR. More preferred non-RXR nuclear receptors capable of dimerizing with an RXR include RARs, TRs and PPARs.
Generally, the nuclear receptor structure contains an amino-terminal activation function (AF-l; A/B domain), the DNA-binding domain (DBD;
C domain), a hinge region (D domain), and a carboxy-terminal ligand-binding domain, LBD (E domain), which includes the activation function AF-2, required for ligand-dependent activation by nuclear receptors.
By an "agent which interacts with an RXR dimer" is intended a compound which binds to an RXR dimer to mediate transcription of a target gene, i.e., "RXR
dimer mediated transcription." The agent can mediate an interaction between RXR dimers and co-regulators. The agent can activate or repress transcription of the target gene. Such agents can be, but are not limited to, peptides, carbohydrates, steroids and vitamin derivatives, which may each be natural or synthetic (prepared, for example, using methods of synthetic organic and inorganic chemistry that are well-known in the art).
For example, such an agent includes a "retinoid" which is a compound which binds to one or more of the retinoid receptors (RARa, RAR(3, RARy, RXRa, RXR~3 and RXRy). Compounds can be either "RAR retinoids" or "RXR
retinoids" depending on their binding characteristics (R.AR retinoids bind to one or more RARs; RXR retinoids bind to one or more RXRs (also referred to as "rexinoids")). Of course, some of such compounds can bind to both RARs and RXRs.
Many RAR and RXR agonists and antagonists are known in the art, such as, for example, 4-[[(2,3-Dihydro-1,1,3,3-tetramethyl-2-oxo-1H-inden-5-yl) carbonyl]amino]benzoic acid (Compound I; WO 98/47861 ), 4-[[[5,6-Dihydro-5,5-dimethyl-8-(3-quinolinyl)-2-naphthalenyl]carbonyl)amino]benzoic acid (Compound II; U.S. PatentNo. 5,559,248; U.S. Patent No. 5,849,923), 3-Fluoro-4[[(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)hydroxyacetyl]
amino]benzoic acid (Compound IV; U.S. Patent No. 5,624,957), 4-[2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)-1,3-dioxolan-2-yl)benzoic acid (SR 11237; U.S. Patent No. 5,552,271).
Agents that interact with RXR dimers, including RAR and RXR agonists and antagonists, can be screened using the methods of the present invention.
In the invention, agents that interact with only one partner or both partners of the RXR dimer can be identified. In fact, because the invention mimics the effects of retinoids on gene transactivation in vivo, the invention provides more accurate methods of identifying RXR dimer agonists and antagonists. Moreover, the action of the identified agent can be further confirmed by binding assays known in the art to determine which partner of the RXR dimer is bound by the identified agent (Rochel, N. et al., Biochem. Biophys, Res. Comm. 230:293-296 (1997)).
Generally, agents which cause transactivation via their receptors are examples of "agonists," while agents which do not cause transactivation, but instead block the transactivation caused by other agonists, are examples of "antagonists." However, because CARs are constitutively expressed, a reverse agonist is needed to deactivate transcription and a reverse antagonist is needed to activate transcription.
Agents can have the ability to bind to multiple receptors. By agents that are "specific" for a nuclear receptor are intended compounds that only bind to one, two or three particular nuclear receptors) and not to others. By agents that are "selective" for a nuclear receptor are intended compounds that preferably bind to one, two or three particular nuclear receptors) over others by a magnitude of approximately five-fold or greater than to other retinoid receptors, preferably eight-fold or greater, more preferably, ten-fold or greater.
By a "ligand for a member of the nuclear receptor" is intended an agent, compound or hormone that binds to a nuclear receptor, which in turn can activate an appropriate hormone response element. Thus, a ligand acts to modulate transcription of a gene maintained under the control of a hormone response element.
Ligands include hormones, steroid or steroid-like compounds, retinoids, thyroid hormones, pharmaceutically active compounds, and the like. Exemplary ligands include ligands for retinoid receptors (e.g., all-traps retinoic acid, 9-cis retinoic acid, etc.), ligands for thyroid hormone receptors (e. g., thyroid hormone), and ligands for vitamin D3 receptor (e.g., 1,25-dihydroxyvitamin D3). Other ligands which bind to nuclear receptors can be identified by the present invention.
By a "hormone" is intended a substance produced in a gland of an animal, human and nonhuman, which exerts specific effects on other parts of the body.
By a "co-regulator" is intended a "co-activator" or a "co-repressor." By a "co-activator" is intended a molecule or factor, generally a protein or RNA, that interacts with nuclear receptors (e.g., RARs, RXRs) and enhance their transactivation. The co-regulator can complex with other molecules or factors to interact with nuclear receptors. Exemplary co-activators include, but are not limited to, ERAP-160 (GRIP-170; p160), ERAP-140, RIP-140, RIP-160, TBP/TAFI,s, SRC-1 (hSRC-I;NCoA-1/mSRC-1),hSRC-3,Trip-1 (Sug-1), Trips, TIF 1 a, TIF 1 (3, y, ARA-70, TRAPS (DRIPs), CBP, p300, PCAF (hGCNS), TIF2/hSRC-2 (GRIP-1/mSRC-2; NCoA-2, p160), mSRC-3/hSRC-3, TRIP230, L7/SPA, p/CIP/mSRC-3 (ACTR/hSRC-3; RAC3/hSRC-3; AIB-L/hSRC; TR.AM-1/hSRC-3; p160; SRC-3), E6-AP, RPF-1 (hRSPS), BRG-1 (SWI2/SNF2), Brahma, NSD-1, PGC-1, HMG-1, HMG-2, NCoA-62, BX42, TSC-2 (Tuberin), PBP (TRAP220; TRIP2; mPIP9), positive cofactors (PC2; PC4), ADA, SMCC, SRA, SNURF, ARIP3, Mizl, PIAS3, GBP (reviewed in, McKenna, E.J. et al., Endocrine Reviews 20:321-344 (1999); Torchia, J. et al., Curr. Opin. Cell Biol.
10:373-383 (1998)), TIF-2, ATP, and Acetyl CoA.
By a "co-repressor" is intended a molecule or factor, generally a protein or RNA, that interacts with nuclear receptors (e.g. RARs, RXRs) and lowers the transcription rate at their target genes. Exemplary co-repressors include, but are not limited to, NCoR (RIP-13), SMRT (silencing mediator for retinoic acid and thyroid receptors; TRAC2), repressor domains of SMRT (e.g., SRD-1, SRD-2, amino acids 1-981 thereof, etc.), TRUP (SURF-3; PLA-X; L7a), SUNCoR, NURD, mSin3A, protein-protein interaction domains of mSin3A (e.g., PAH-1, PAH-2, PAH-3, PAH-4, combinations thereof, etc.), N-CoR, Mad/Mxi-1, mSin3B, Sin3, etc. (reviewed in, McKenna, E.J. et al., Endocrine Reviews 20:321-344 (1999); Torchia, J. et al., Curr. Opin. Cell Biol. 10:373-383 (1998)).
In the invention, methods are provided for identifying a "modulator" which promote dissociation of the co-activator or co-repressor complex from the nuclear receptors (e.g., retinoid and/or thyroid hormone receptors) orpromote association of co-activator or co-repressor complexes with the nuclear receptors.
As used herein, an "agent" is alternatively intended a molecule, factor, substance or compound which is screened for an intended function, such as co-activator, co-repressor, or modulator function, as it will be clear from the context in which the term is used.
The RXR dimers of the invention can be obtained by expressing the receptors proteins in eukaryotic or bacteria cells and purifying the receptors. In one embodiment, the receptor is purified from tissues or cells which naturally produce the receptor. Alternatively, the receptor can be expressed recombinantly, for example, by inserting the gene encoding the receptor into the baculovirus or vaccinia virus genome and infecting the baculovirus or vaccinia virus, respectively, into insect or human cells, respectively. The receptors can also be expressed in yeast. Exemplary constructs for production of the receptor can be obtained from, for example, human, mouse or chicken, and include, but are not limited to, human Flag-tagged RARa and mouse His-tagged RXRa (Dilworth, et al., Proc. Natl.
Acad. Sci. USA 96:2000-2004 ( 1999)), human Flag-tagged VDR and human Flag-tagged RXRa (Rachez et al., Nature 398:824-828 ( 1999)), and human Flag-tagged TRa (Fondell et al., Proc. Natl. Acad. Sci. USA 96:1959-1964 (1996)).
Other constructs can be generated by subcloning the cDNA from existing DNA
vectors of, for example, LXRs (Willy et al., Genes Dev. 9:1033-1045 (1995)), PPARa (Isseman and Green, Nature 347:645-650, FXR (Forman et al., Cell 81:687-693 (1995)), LXRa (Apfel et al., Mol. Cell Biol. 14:7025-7035 (1994)), BXR (Blumberg et al., Genes Dev. 12:1269-1277 (1998)), CARa and (3 (Choi et al., J. Biol. Chem. 272:23565-23571 (1997); Forman et al., Nature 395:612-615 (1998)), and SXR (Blumberg et al., Genes Dev. 12:3195-3205 (1998)).
1 S A variety of methodologies are known in the art that can be used to obtain, isolate or purify the nuclear receptors, including, but not limited to, immunochromatography, HPLC, size-exclusion chromatography, ion-exchange chromatography, and affinity chromatography.
Nuclear receptors bind to specific DNA sequences known as response elements (REs) or hormone response elements (HREs). Those of skill in the art can readily determine suitable hormone response elements (HREs) for use in the practice of the present invention, such as, for example, the response elements described in U.S. Patent No. 5,091,518 and WO 92/16546. The recognition of REs by a given RXR dimer is dependent on the actual sequence, orientation and spacing of the repeated motifs.
Naturally occurring HREs are composed of direct repeats (DRs; Umesono et al., Cell 65:1255-1266 (1991)), and inverted repeats (IRs; Umesono et al., Nature 336:262-265 (1988)).
Direct repeats and inverted repeats can have a gap which separates the two core-binding sites. Thus, for example, spacers of 1, 3, 4 and S nucleotides serve as preferred DR response elements for heterodimers of RXR with PPAR, VDR, T3R and RAR, respectively (Naar et al., Cell 65:1267-1279 (1991 ); Kliewer et al., Nature 358:771-774 (1992); and Issemann etal., Biochimie 75:251-256 (1993)).
The optimal gap length for each heterodimer is determined by protein-protein contacts which appropriately position the DNA binding domains (DBDs) of RXR
and its partner (Kurokawa et al., Genes Dev. 7:1423-1435 (1993); Perlmann S etal., GenesDev. 7:1411-1422 (1993); Towers etal., Proc. Natl. Acad. Sci.
USA
90:6310-6314 (1993); and Zechel et al., EMBO J. 13:1414-1424 (1994)).
Exemplary DR1 is provided in Vivat et al., EMBO J. 16:5697-5709 (1997).
Exemplary DR3 is provided in Rachez et al., Nature 398:824-828 (1999).
Exemplary DR4 is provided in Fondell et al., Proc. Natl. Acad.. Sci. USA
96:1959-1964 (1996). Exemplary DRS is provided herein and in Dilworth et al., Proc.
Natl. Acad Sci. USA 96:1995-2000 (1999).
The preferential RXI~/RAR heterodimer binding repertoire in vitro to DNA (DR1, DR2, and DRS, in order of increasing efficiency) is similar to the "natural" RARE repertoire, which suggest that R~~/RAR heterodimers are the 1 S functional units that transduce the retinoid signal in vivo. Similarly, RXR/TR, RXR/VDR, and RXR/PPAR bind preferentially to DR4, DR3, and DR 1 elements, respectively (Giguere, V. Endocr. Rev. 15:61-79 (1994); Glass, C.K. Endocr.
Rev. 15:391-407 (1994); and Mader, S. et al. J. Biol. Chem. 268:591-600 (1993)). RXRs also bind as homodimers to a DR1 element (Nakshatri, H., and Chambon, P. J. Biol. Chem. 269:890-902 (1994)).
RXR/LXR binds to DR4, RXR/BXR binds to modified DR4, RXR/CAR
binds to DRS, RXR/SXR binds to DR4, and RXR/FXR binds to IR1 (inverted repeat with a 1 by spacer).
Direct repeat hormone response elements (HREs) contemplated for use in the practice of the invention are composed of at least one direct repeat of two or more half sites, optionally separated by one or more spacer nucleotides (with spacers of 1-S preferred). The spacer nucleotides can be selected from any one of A, C, G or T. Each half site of direct repeat HREs contemplated for use in the practice of the invention comprises the sequence -RGBNNM- wherein R is selected from A or G; B is selected from G, C, or T; each N is independently selected from A, T, C, or G; and M is selected from A or C; with the proviso that at least 4 nucleotides of said -RGBNNM- sequence are identical with the nucleotides at corresponding positions of the sequence -AGGTCA-. Response elements employed in the practice of the invention can optionally be preceded by NX, wherein x falls in the range of 0 up to 5. Exemplary hormone response elements include, but are not limited to, direct repeats of -PuG(G/A)(T/A)CA-S (Mader, S. et al., J. Biol. Chem. 268:591-600 (1993)).
Response elements are operatively linked to a reporter or target gene, whereby expression of the reporter or target gene indicates the action of a ligand, RXR dimer and/or the response element. Exemplary reporter genes include, but are not limited to, chloramphenicol acetyl transferase (CAT), ~3-galactbsidase (~3-gal), luciferase (LUC), and (3-globin.
In a steady state, eukaryotic chromosomes ("chromatin") are organized into a repeating protein DNA unit, the nucleosome. The basic protein unit of the nucleosome is the histone, a small, highly basic, globular moiety. A
nucleosome core particle contains a histone octamer, made up of two copies of each of histones H2A, H2B, H3 and H4, around which is wrapped 1.7 turns of a left-handed DNA superhelix 0200 by of DNA).
By a "chromatin based DNA template" or "chromatin template" or "chromatin assembled template" is intended an in vitro nucleosomal array generated by complexing an oligonucleotide (linear DNA or plasmid) with histone octamers (H2A, H2B, H3 and H4) and/or histone H1. The oligonucleotide sequence or DNA template comprises a hormone response element, at least a minimal promoter element (including a TATA box and a transcription start site, i.e., -35 to +80 of any natural eukaryotic or viral gene promoter), and a reporter gene, as described above. A "naked" oligonucleotide sequence or DNA template is not complexed with histone octamers.
A chromatin based DNA template is prepared by adding a chromatin assembly extract to the oligonucleotide (linear DNA or plasmid) in the presence of histones. A chromatin assembly extract contains the proteins and factors necessary for assembly of the DNA template around the histones into nucleosomes and for movement of the nucleosome along the DNA template to allow transcriptionally repressive and permissive states.

Methods for preparing chromatin assembly extracts useful in the present invention are known in the art (Becker, P.B. et al., Meth. Cell Biol. 44:207-(1994); Pazin, M.J. et al., Science 266:2007-2011 (1994); Kamakaka, R.T. et al., Genes Dev. 7:1779-1795 (1993)).
Chromatin assembly extracts can be prepared, for example, from tissue culture cells (Banerjee, S. and Cantor, C.R., Mol. Cell. Biol. 10:2863-2873 (1990)), Xenopus eggs and oocytes (Almouzni, G. and Mechali, M., EMBO J.
9:573-582 (1988)); Shimamura, A. et al., Mol. Cell. Biol. 8:4257-4269 (1988)), Drosophila ISWI (Ito et al., Genes Dev. 13:1529-1539 (1999); Carona et al.
Mol.
Cell. 3:239-245 (1999)), human SNF2h (Leroy et al., Science 282:1900-1904 (1998)), and preferably, Drosophila embryos (Becker, P.B. and Wu, C., Mol.
Cell. Biol. 12:2241-2249 (1992); Becker, P.B. et al., Methods Cell Biol.
44:207-223 (1994)).
For example, a method of preparation of S-190 Drosophila chromatin 1 S assembly extracts is provided in the "Materials and Methods" section in Example 1, infra. A method for preparing S-150 chromatin assembly extracts is provided in Becker, P.B. et al., Meth. Cell Biol. 44:207-223 (1994).
As indicated above, in one embodiment, the chromatin based DNA
template can be purified such that one or more or all transcription co-regulators are removed. Methods of removing a transcription co-regulator include, but are not limited to, chromatography and immunoprecipitation. Chromatographic methods for removing a transcription co-regulator include, but are not limited to, size-exclusion chromatography, affinity chromatography, heparin chromatography, DEAF chromatography, ion exchange chromatography, phenyl sepharose chromatography, phosphocellulose chromatography, and hydroxy-apatite.
Removal of a transcription co-regulator from a chromatin based DNA
template by size-exclusion chromatography, for example through a spin-column, is a function of the resin employed, the size of the template, and the size of the co-regulator(s) to be removed. The resin is selected such that factors greater in size than the exclusion limit of the resin are isolated. Thus, in order to remove a transcription co-regulator from a chromatin based DNA template, the template must be larger in size than the exclusion limit of the resin used, and the transcription co-regulator must be smaller in size than the exclusion limit of the resin used. The size of the template is a function of the size of the DNA, the chromatin extract complexed thereto, and the histones present. The size of the DNA is itself a function of the length (in base pairs) of the hormone response element, the minimal promoter element, and the reporter gene. The transcription co-regulators to be removed in the present invention include, but are not limited to, SWI/SNF [2000 kilodaltons (kDa)], p300 (300 kDa), TIF2 (160 kDa), ATP
[molecular weight = 551.1 atomic mass units (amu)] and Acetyl CoA (molecular weight = 809.6 amu) (see Example 2, infra). For example, in order to remove these co-regulators, assuming constancy of the resin, the chromatin based DNA
template must be larger than 2000 kDa. Since the molecular weight of a single nucleosomal unit [200 base pairs (bp) of DNA and a histone octamer] is approximately 110 kilodaltons (kDa), the DNA to which the chromatin extract is complexed must be at least 3.6 kb (2000 kDa -110 kDa =18; 18 x 200 by = 3.6 kb). Other co-regulators that can be removed are provided above.
In preparing a chromatin based DNA template from which a transcription co-regulator has been removed by size-exclusion chromatography (see Example 2, infra), the pre-equilibration and elution buffer can contain KCl or a sodium monovalent salt such as NaCI in a concentration of 1-100 mM, preferably 1-50 mM, more preferably 10-40 mM, still more preferably 10-30 mM, and even more preferably 10-20 mM. The concentration of this salt in the final in vitro transcription reaction (see Example 2, infra) can be 40-100 mM, preferably 50-100 mM, more preferably 50-75 mM, and still more preferably 50-60 mM. It has been discovered that RARa/RXRa heterodimers cannot efficiently initiate transcription from chromatin templates from which transcription co-regulators have been removed unless initiation of transcription is preceded by a preincubation period during which the chromatin template is exposed to ATP-driven remodeling activities of complexes containing SNF2 family members including, but not limited to, SWI/SNF, CHRAC, ACF, NURF, WCRF, and RSF.
By "ATP-driven remodeling activity" or "ATP-driven chromatin remodeling activity" is intended an ATP-requiring activity that modifies chromatin and/or nucleosomal structure such that ligand-dependent transcriptional activation is enhanced. Examples of compounds that confer such activity include, but are not limited to, SWI/SNF, CHRAC, ACF, NURF, WCRF, and RSF.
By "HAT activity" or "histone acetyltransferase activity" is meant a histone acetyltransferase-requiring activity that modifies chromatin and/or nucleosomal structure by interacting with DNA-binding proteins that regulate transcription.
Examples of compounds that confer such activity include, but are not limited to, p300, TIF2, p160, CBP, pCAF, GCNS, ACTR, and SRC-1.
Chromatin can be assembled on relaxed or supercoiled circular DNA by preincubating the extract with histories to assemble histone octamers and-adding the template of interest. Core histories can be purified according to the method of Simon, R.H., & Felsenfeld, G., Nucl. Acids. Res. 6:689-696 (1979), or calf thymus histories are commercially available (Boehringer Mannheim). The appropriate amount of histories can be determined empirically, using as a guide a stoichiometry of histories to DNA of 0.8:1 (w/w) (Albright, S.C. et al., J.
Biol.
Chem. 254:1065-1073 (1979)). Details of a method for chromatin assembly on a DNA template are provided in Example 1, infra.
DNA supercoiling assay is based on topological changes that accompany the wrapping of DNA around a nucleosome core (Becker, P.B. et al., Meth. Cell Biol. 44:207-223 (1994)). Winding ofDNAaroundanucleosome core introduces one positive superhelical turn in the plasmid DNA, which is relaxed by topoisomerase I activity present in the embryo extracts. When nucleosomes are removed by proteinase K digestion and DNA purification, one negative superhelical turn corresponding to each assembled nucleosome appears in the closed circular DNA. The superhelical density of a plasmid, i.e., the absolute number of superhelical turns, can be directly counted by visualization of the plasmid topoisomers on two-dimensional agarose gels or by resolving duplicate samples on multiple agarose gels containing different chloroquine concentrations.
The introduction of supercoils into a plasmid can simply be visualized by agarose gel electrophoresis as a rapid indicator of nucleosome reconstitution.
Generally, supercoiling of the chromatin can be assayed by incubating the assembled chromatin with the heterodimer (e.g., Flag-tagged human RARa/His-tagged mouse RXRa) in the presence of ligand. Supercoiling can be determined by adding topoisomerase I and/or chloroquine and resolving the DNA on an agarose gel. Details of a supercoiling assay are provided in the "Materials and Methods" section in Example 1, infra.
DNA supercoiling measures the wrapping of DNA around a particle but does not necessarily indicate the reconstitution of a full octamer of core histones (Becker, P.B. et al., Meth. Cell Biol. 44:207-223 (1994)). The winding of DNA
around a complete histone octamer or a tetramer of histones H3 and H4 cannot be distinguished by the supercoiling assay. Therefore, nuclease digestion assay is used to provide information on the nature of the nucleosome core particle'as well as on the average distance between particles.
Generally, no more than 20% of the genome is organized as active chromatin in a given cell type. Active chromatin is less compact than bulk chromatin, and is more accessible to enzymes. Nuclease digestions can be used to investigate changes in nucleosome organization and positioning around a given gene in different cell types and stages. Such nucleases include, for example, DNase I, DNase II, micrococcal nuclease, S 1 nuclease, copper/phenanthroline, and restriction enzymes.
Micrococcal nuclease (MNase) assay relies on the ability of MNase to preferentially cleave the linker DNA between nucleosome core particles. After the initial endonucleolytic attack of linker DNA, the trimming activity associated with the enzyme progressively removes the linker DNA. Extensive digestion of chromatin with MNase can bring the size of the mononucleosome from 160-220 by to the 147 by DNA fragment protected by the nucleosome core particle whereas a partial digest results in a ladder of fragments representing oligonucleosomal DNAs. Details of a MNase assay is provided in the "Materials and Methods" section in Example 1, infra.
The invention lends itself readily to the preparation of kits containing the elements necessary to carry out the methods disclosed herein. Such a kit can comprise a carrier being compartmentalized to receive in close confinement therein one or more contain means, such as tubes or vials. One of the container means can contain the DNA template. One of the container means can contain the chromatin extract. One or more of the container means can contain the histones.

One or more of the containers can contain known agonists, antagonists, co-activators, co-repressors or modulators which can be used as controls. In addition, the kit can also include a "catalog" defined broadly as a booklet, book pamphlet, computer disk or the like, which can assist in carrying out the invention.
The kit can contain all of the additional elements necessary to carry out the method of the invention, such as buffers, enzymes, pipettes, tubes, nucleic acids, nucleoside triphosphates, and the like.
As described herein, by a "compound" is intended a protein, nucleic acid, carbohydrate, lipid or a small molecule.
It will be readily apparent to one of ordinary skill in the relevant arts that other suitable modifications and adaptations to the methods and applications described herein can be made without departing from the scope of the invention or any embodiment thereof.
The following Examples serve only to illustrate the invention, and are not to be construed as in any way to limit the invention.
Example 1 The supercoiled plasmid (DRS)5~32G that contains the RAR(32 core promoter (-35 to +85) and five copies ofthe RA response element (RARE) ofthe RAR~32 gene was used to study activation of transcription by RAR/RXR
heterodimers (see Materials and Methods, Example 1; see FIGS. 1 and 2B). To determine whether a chromatin-assembled template was important, the transcriptional activity of purified RARa/RXRa heterodimers was analyzed (see, "Materials and Methods" section and FIG. 2A), using both naked and chromatin based DNA templates. Periodic nucleosomal arrays (FIG. 2B) were generated using supercoiled (DRS)5(32G plasmid and a chromatin-assembly extract (see, "Materials and Methods" section). Note that the nucleosomal organization of the chromatin template was not grossly affected by the addition of RARa/RXRa heterodimers and tRA (FIG. 2B).
When expressed relative to basal transcription from the internal control naked pGl template (FIG. 1), "constitutive" transcription on the naked (DRS)5(32G template was not affected by the presence of RA ligand and/or receptor heterodimers (FIG. 3A). In marked contrast, very little transcription was observed from the corresponding chromatin template in the absence of RARa/RXRa heterodimers, irrespective of the presence of RA. However, the addition of both heterodimers and tRA resulted in a potent activation of transcription (between 30- to 100-fold; FIG. 3A). Little ligand-dependent activation of transcription by RARa/RXRa heterodimers was observed when exogenous histones were not added to the Drosophila extract during chromatin assembly on the (DRS)5 ~32G plasmid. Optimal activation of transcription from the chromatin template was achieved using 1 nM RARa/RXRa heterodimer that corresponds to approximately five heterodimers per (DRS)5~32G template molecule (200 pM), i.e., one heterodimer per DRS response element (FIG. 3A).
Consistent with this observation, DNase I foot printing analysis showed that all five DRS RAREs were bound by RARa/RXRa heterodimers at these concentrations, with no RARE being particularly favored (FIG. 4). Note that in contrast to activation of transcription, the binding of the heterodimers to the chromatin template was not dependent on the presence of tRA (FIG. 4). Binding of unliganded RAR/RXR heterodimers to chromatin is therefore clearly not sufficient for transcriptional activation, thus suggesting that the critical step in transactivation is a ligand-dependent transconformation of DNA-bound heterodimers.
The response element specificity of transcriptional activation by RARa/RXR heterodimers was examined by comparing transcription from the cognate (DRS)S(32G template and the (17M)5(32G template, in which the five DRS RAREs have been replaced by five copies of the 17-mer binding site for the DNA binding domain [GAL(1-147)] of the yeast transactivator Gal4 (FIG. 1).
RARa/RXRa heterodimers did not activate transcription from the chromatin-assembled ( 17M)5 (32G template, whereas under similar conditions, the chimeric acidic transactivator GAL-VP16 (Sadowski, L, Nature 335:563-564 (1988)) efficiently activated transcription from that template, but not from the chromatin-assembled (DRS)5~32G template (FIG. 3B).

The above results demonstrate that activation of transcription by RAR/R~~R heterodimers is dependent on packaging of the template into a nucleosomal structure and that it is specific, in that it requires the heterodimer, a cognate ligand and a cognate response element. As tRA binds RARs, but not RXRs, the effect of ligands that bind to RXRs was then investigated. Of interest, 9cRA that binds both RARs and RXRs was more efficient than tRA at limiting concentrations, with EDS° of approximately 9x10-'° M and 4x10'9 M for 9cRA
and tRA, respectively (FIG. 5A). Because these differential effects of 9cRA
and tRA suggested that synergistic activation of transcription might occur when both RARa and RXRa are liganded, transcriptional activation by RARa/RXRa heterodimers upon addition of receptor-specific synthetic retinoids (FIG. 5B) was investigated. As expected, a stimulation was observed in the presence of the RARa-specific agonist Compound I, but not on addition of either the RARy-specific agonist Compound IV or the RARa antagonist Compound II (Chen, J-Y., et al., Nature 382:819-822 (1996)) (FIG. 5B). Of interest, the RXR-specific pan-agonist SR11237 (Chen, J-Y., et al., Nature 382:819-822 (1996)) did not activate transcription on its own (FIG. 5B). However, a synergistic stimulation was observed upon concomitant addition of SRl 1237 and limiting concentrations of RAR agonists (FIG. 5B; compare 10-8 and SxlO-g M tRA in the presence and absence of SR11237, and also Compound I in the presence and absence of SRl 1237). In contrast, no stimulation resulted from the simultaneous addition of the RARa antagonist Compound II and the RXR agonist SR11237. It appears therefore that the AF-2 activation function of RXRa can act synergistically with that of RARa, but that the binding of arl agonist to RARa is a prerequisite for effective activation oftranscription by agonist-bound RXRa. This conclusion was further supported by the observation that the RARa-specific antagonist Compound II abrogated the synergistic effect of the RARa-specific agonist Compound I and RXR agonist SR11237 (FIG. 5B). Similarly, Compound II
abrogated the 9cRA-induced transcriptional activation by RARa/RXRa heterodimers (FIG. 5B), even though 9cRA binds to both RARs and RXRs.
Acetylation and deacetylation of nucleosomal histones in transcriptionally active (euchromatin) and inactive (heterochromatin) chromatin, respectively, is well documented (reviewed in Kuo, M-H. & Allis, C.D., BioEssays 20:615-626 (1998)). The facilitating role of histone acetylation in transcriptional activation has also been recently demonstrated in vitro (Nightingale, K.P., et al., EMBO
J.
17:2865-2876 (1998); Utley, R.T., et al., Nature 394:498-502 (1998)). The effect of the addition of acetylCoA to the in vitro transcription system was examined. No effect could be evidenced using either RARa/RXRa heterodimers or the Gal-VP 16 activator, in the presence of either naked or chromatin-assembled cognate templates (FIG. 6A). Because certain coactivators are thought to mediate transactivation by nuclear receptors at least in part through their intrinsic histone acetyltransferase activities (e.g. SRC-1, ACTR, CBP and p300, etc.), it was investigated whether addition of purified baculovirus-expressed p300 could stimulate transcriptional activation by RARa/RXRa in the in vitro system. p300 enhanced the activation of transcription by the heterodimers ~-4-fold in the presence of tRA, while transcription of the chromatin template remained repressed in the absence of the agonistic ligand, irrespective of the presence of the heterodimers (FIG. 6B). No p300 effect was seen on naked DNA templates. The further addition of acetylCoA had no effect on the extent of transcriptional enhancement, even though the purified p300 coactivator exhibited histone acetyltransferase activity. However, the in vitro system contains some endogenous histone acetyltransferase activity that was not further enhanced by the addition of p300 to the transcription reaction.
Materials and Methods DNA and Chromatin Templates. The plasmids (DRS)S~i2G and (17M)5~32G (~5.2 kb) were constructed by inserting five copies of the DRS RA
response element from the mouse RAR(32 promoter or the 17-mer GAL4 binding site, respectively, upstream of the mouse RAR(32 core promoter [-35 to +85]
which had been previously linked to the -9 to +1516 chicken ~3-globin gene sequence (FIG. 1). The most 3' DRS element is positioned at approximately the same distance from the TATA box as the DRS RARE found in the natural RAR[i2 promoter (Zelent, A., et al., EMBO J. 9:71-81 (1991)).

Chromatin assembly extracts were prepared from Drosophila embryos (0-6 hr) as described in Kamakaka, R.T. et al., Genes Dev. 7:1779-1795 ( 1993)).
Canton-S wild-type flies were grown at 25 °C at 70-80% humidity in population cages. The embryos were collected on apple juice-agar plates covered with yeast.
S Four batches of embryos [typically, 30-50 grams, collected (every 6 hr over a 24-hr period) between 0 and 6 hr after fertilization and then stored for s 18 hr at 4 ° C], were harvested in nylon mesh with water and then dechorionated by immersion for 90 sec in 1:1 (vol/vol) bleach [5.25% (wt/vol) sodium hypochlorite]/water [final concentration of sodium hypochlorite is 2.63%
(vvtJvol)]
at room temperature. The embryos were quickly rinsed with embryo wash buffer I [1 liter; 0.7% (wtlvol) NaCI and 0.04% (vol/vol) Triton X-100; buffer at room temperature], washed with water (at room temperature), and transferred to a ml beaker in an ice bath. Embryo wash buffer II [500 ml, buffer at 4°C;
0.7%
(wt/vol) NaCI and 0.05% (vol/vol) Triton X-100] was added to the beaker. The embryos were suspended with a glass rod and allowed to settle to the bottom of the beaker (for ~2 min). The cloudy suspension above the embryos, which contained chorion particles and other debris, was removed by aspiration. This wash/settling/aspiration procedure was repeated once with embryo wash buffer II (500 ml; at 4°C), twice with 0.7% (wt/vol) NaCI solution (500 ml for each wash; buffer at 4°C), and once with buffer R [500 ml; buffer at 4°C; 10 mM
HEPES (K+) (pH 7.5), 10 mM KC1, 1.5 mM MgCl2, 0.5 mM EGTA, 10%
(vollvol) glycerol, 10 mM (3-glycerophosphate, 1 mM DTT, 0.2 mM
phenylmethylsulfonyl fluoride (PMSF)]. In the final wash, the embryos take longer to settle (~l 0 min), and the final volume of the embryo suspension before homogenization is roughly twice of that of the loosely packed volume of dechorionated embryos. The embryos were then transferred to a Wheaton Dounce homogenizer (40 ml) and disrupted by 15 strokes with the B pestle followed by 40 strokes with the A pestle. The homogenate was subjected to centrifugation in a Falcon 2059 tube in a Sorvall SS-34 rotor at 8000 rpm for min. at 4 ° C. The cloudy, yellow cytoplasmic fraction was collected with a syringe (the white layer at the top and the pellet at the bottom of the tube were avoided).
MgCl2 (from a 1 M stock solution) was added to increase the Mg(II) concentration from 1.5 mM to a final concentration of 7 mM. The extract was then subjected to centrifugation in a Beckman SW55 rotor at 45,000 rpm (192,000g) for 2 hr at 4°C. After centrifugation, the white upper layer was removed with a spatula and the yellow-brown liquid was collected. This supernatant fraction was frozen in liquid nitrogen, thawed in water (at room temperature), and then subjected to a second centrifugation in the Beckman SWS

rotor at 45,000 rpm for 2 hr at 4°C. The resulting chromatin reconstitution extract (also referred to as the Drosophila S-190 extract) was frozen in liquid nitrogen and stored at - 80°C. The extracts remain active for >1 year at = 80°C.
Chromatin was assembled on supercoiled circular DNA (see FIG. 2B) as follows. The chromatin assembly extract was preincubated with 3 ~g of calf thymus core histones (Boehringer Mannheim) at room temperature to assemble histone octamers. After 30 minutes, 1 ~g of (DRS)5~32G (or (17m)5~32G) and an ATP regeneration solution (3 mM MgCl2,1 mM DTT, 30 mM creatine phosphate, 1 S 3 mM ATP,1 ~g/mL creatine kinase) were added and allowed to incubate for 4 hr at 27°C (Becker, P.B., et al., Methods Cell Biology 44:207-223 (1994)).
Supercoiling assays were performed essentially as previously described (Pazin, M.J., et al., Science 266:2007-2011 (1994)). Two hundred nanograms of assembled chromatin (or naked control template) was incubated with 1 nM Flag-tagged human RARa/His-tagged mouse RXRa (FhRARa/HmR~~a) heterodimer in the presence of ligand or vehicle for 30 min at 27°C. The concentration of MgCl2 was increased to 11.5 mM immediately before the addition of 10 units of topoisomerase I (Life Technologies, Cergy Pontoise, France), and allowed to incubate at 37°C for 30 min. DNA was resolved on a 1% agarose gel in the presence or absence of 1.2 ~M chloroquine for 18 hr at 2 volts/cm.
Determination of supercoiling within (DRS)5~32G chromatin template using topoisomerase I and/or chloroquine (Becker, P.B., et al., Methods Cell Biology 44:207-223 (1994)) indicated the presence of at least 25 nucleosomes.
Micrococcal digestion analysis of reconstituted chromatin was performed essentially as previously described (Bellard, M., et al., in Methods Enzymol.
170:317-346 (1989)). Assembled chromatin was incubated with 1 nM
FhR.ARa/HmRXRa heterodimer or vehicle in the presence of 10-6 M tRA for 30 min at 27°C. Chromatin was then digested with varying concentrations of micrococcal nucleate (1-20 units) for 1 min at 27°C. Proteins were removed by proteinase K treatment followed by an extraction with phenol-chloroform.
Digested DNA was separated on a 1.5% agarose gel for 4 hr at 4 volts/cm, transferred to a nitrocellulose membrane and analyzed by southern blotting using a probe corresponding to the promoter region of the (DRS)5[iG plasmid.
Micrococcal nuclease digestion (Bellard, M., et al., in Methods Enzymol.
170:317-346 ( 1989)) showed that they had a periodicity of approximately 160 by (FIG. 2B).
DNase I footprinting was performed essentially as previously described (Pazin, M.J., et al., Science 266:2007-2011 (1994)). Chromatin was assembled on 250 ng of (DRS)S(32G plasmid then incubated alone or with 1 nM
FhR.ARa/HmR~~a heterodimer in the presence or absence of 10-6 M RA for 30 min at 27°C. CaClz was added to a final concentration of 3 mM along with S U of DNase I (Boehringer Mannheim) for 90 sec. DNA fragments were amplified with VentR (exo-) [New England Biolabs, Beverley, MA] using a 30 by primer complementary to a sequence located between -280 and -250 upstream of the RAR(32 promoter start site.
Protein Expression and Purification. The spodoptera frugipenda cell line S~ was co-infected with baculoviruses expressing His-tagged mouse RXRa (HmRXRa) and Flag-tagged human RARa (FhRARa) for 48 hr. S~ cells expressing the heterodimeric proteins were lysed by homogenization in a low salt buffer (20 mM Hepes pH 7.6, 100 mM KCI, 10 mM imidazole, lx PIC
[2.5 ~,g/mL leupeptin, 2.5 p.g/mL pepstatin, 2.5 ~g/mL aprotinin, 2.5 ~g/mL
antipain, 2.5 ~g/mL chymostatin], and 1 mM PMSF). The FhR.ARa/HmRXRa heterodimer was partially purified by chromatography using a Niz+ column (Amersham Pharmacia) and eluted with a low salt buffer containing 300 mM
imidazole. The heterodimer was then further purified from the Ni2+ column eluate by affinity purification using agarose-coupled M2 anti-Flag antibodies (Sigma), as specified in the manufacturer's instructions. The purified heterodimer was eluted from the resin in a buffer consisting of 20 mM Hepes pH 7.6, 100 mM KCI, 1.5 mM MgCl2, 0.5 mM EGTA, 50 ~M ZnCl2, 15% glycerol, 500 p.g/mL

competitor peptide (DYKDDDDK) (SEQ ID NO:1 ), 1 mM DTT, 1 mM PMSF
and lx PIC.
Western blot analysis was performed by using the monoclonal antibodies anti-RARa 9a-9A6 (Gaub, M.P., et al., Exp. Cell Res. 201:335-346 (1992)) and anti-RXRa 1RX-6612 (Rochette-Egly, C., et al., Biochem. Biophys. Res.
Commun. 204:525-536 (1994)) (FIG. 2A).
The DNA binding properties of the heterodimer were examined by electrophoretic mobility shift analysis (Kumar, V. & Chambon, P., Cell 55:145-156 ( 1988)). Briefly, the purified heterodimer was incubated with 10~ M tRA
on ice. After 15 min., the [32P]-labeled DRS oligonucleotide 5'-TCGGGAGGGTTCACCGAAAGTTCACTCGCC-3' (SEQ ID N0:2) hybridized to its unlabeled compliment were added to the reaction in the presence of 20 mM Hepes pH 7.6, 10 mM KCI, 10% Glycerol, 1.5 mM MgCl2, 0.5 mM
EGTA, 1 mg/mL BSA, and 2 ~,g poly dIdC. The reaction was then allowed to proceed for another 15 min at 22°C. Reaction mixtures were resolved by polyacrylamide gel electrophoresis (S% in O.Sx TBE) for 4 hr at 150 volts, dried and visualized by autoradiography.
The integrity of the ligand binding domain of RARa was examined using a ligand binding assay (Rochel, N., et al., Biochem. Biophys. Res. Comm.
230:293-296 (1997)). The RARa/RXRa heterodimer (10 fmol) was diluted to 200 ~1 in a buffer of 10 mM Tris-HCl (pH 8.0), 1 SO mM KCl then incubated in the presence of varying concentrations of diluted [3H]tRA (5x10-'° -SxlO-8 M) on ice. After 4 hr, unbound RA was separated from RARa-bound RA by washing through GF/C glass fibre filters (Whatman, Maidstone, England) using a buffer consisting of 50 mM Tris-HCl pH 7.5,154 mM NaCI, 0.01 % Triton X-100. The amount of [3H]tRA bound to the heterodimer was determined by liquid scintillation counting.
Full-length p300 was prepared from Sf~ cells infected with a p300-expressing baculovirus (Kraus, W.L. & Kadonaga, J.T., Genes Dev. 12:331-342 ( 1998)), its purification was monitored using a rabbit polyclonal anti-p300 (C-20) antibody (Santa Cruz Biotechnology), and the histone acetyltransferase activity of the purified protein was confirmed as described (Ogryzko, V.V., Cell 87:953-959 (1996)). His-tagged Gal(1-147) and GAL-VP16 (Tora, L., et al., Cell 59:477-487 (1989)) were expressed from pET3 expression vectors in the BL-21 pLysS bacterial strain and purified by Niz+ column chromatography.
In vitro Transcription. Transcription was performed using a HeLa cell nuclear extract (Dignam, J.D., et al., Nucleic Acids Res. 11:1475-1489 (1983)) as described (Pazin, M.J., et al., Science 266:2007-2011 (1994)). Chromatin or naked templates were incubated with 1 nM FhR.ARa~HmRXRa heterodimers (in the presence of ligand or vehicle) for 30 min at 27°C prior to transcription initiation. Transcription reactions were initiated by the addition of the '100 pg HeLa nuclear extract (Dignam, J.D. et al., Nucleic Acids Res. 11:1475-1489 (1983)), pGl (internal control plasmid) (Sassone-Corsi, P., et al., Cold Spring Harbor Symp. Quant. Biol. 50:747-752 (1985); see FIG.1), and rNTPs (0.5 mM) and incubated at 30°C for 45 min. Transcription was quantitated by Sl nuclease analysis (Tora, L., et al., Cell 59:477-487 (1989)) using a [32P]-labeled probe (S 1 probe) that hybridizes with transcripts from the (DRS)5~32G, (17 m)5[32G and pG 1 plasmids through their transcription start sites to yield fragments of 179,179 and 60 nt, respectively (FIG. 1 ).
Retinoids. 4-[[(2,3-Dihydro-1,1,3,3-tetramethyl-2-oxo-1 H-inden-5-yl) carbonyl]amino]benzoic acid (Compound I; WO 98/47861 ), 4-[[[5,6-Dihydro-5,5-dimethyl-8-(3-quinolinyl)-2-naphthalenyl]carbonyl]amino]benzoic acid (Compound II; U.S. PatentNo. 5,559,248; U.S. PatentNo. 5,849,923), 3-Fluoro-4[[(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)hydroxyacetyl]
amino]benzoic acid (Compound IV; U.S. Patent No. 5,624,957), SRl 1237 (LT.S.
Patent No. 5,552,271) were obtained from Bristol-Myers Squibb Company.
Discussion Numerous studies aimed at reproducing transactivation by nuclear receptors in vitro have been reported over the last years. Most of these studies were performed with naked DNA templates and the reported transcriptional activations were either ligand-independent (Shemshedini, L., et al. , J. Biol.
Chem.
267:1834-1839 (1992); Schmitt, J. & Stunnenberg, H.G., Nucleic Acids Res.

21:2673-2681 (1993); De Vos, P., et al., Nucleic Acids Res. 22:1161-1166 (1994)) or only modestly dependent on addition of agonistic ligands (Valcarcel, R., et al., Genes Dev. 8:3068-3079 (1994); Fondell, J.D., et al., Proc. Natl.
Acad.
Sci. USA 93:8329-8333 (1996); Lemon, B.D., et al., Mol. Cell. Biol. 17:1923-1937 (1997)). Moreover, these studies did not or only poorly reproduced the effects of agonistic and antagonistic ligands, as observed on responsive genes in vivo. In one case (Schild, C., et al., EMBO J. 12:423-433 (1993)), a greater stimulation was observed with salt dialysis-reconstituted chromatin templates than with naked DNA templates, but the ligand-dependency of transcriptional activation was not established. With the recent finding that several putative transcriptional coactivators that interact in an agonist-ligand-dependent manner with NRs possess histone acetyltransferase activity, it became evident that more physiologically relevant templates might be required to faithfully reproduce in vitro the essential features of ligand-dependent transcriptional activations as observed in vivo. Estrogen- and anti-estrogen-regulated transcriptional activation by the estrogen receptor a (ERa), resembling the natural mechanism of action of ERa in vivo, was in fact recently achieved in vitro with chromatin, but not with naked DNA templates (Kraus, W.L. & Kadonaga, J.T., Genes Dev. 12:331-342 (1998)). Using a similar approach, it is shown by the present invention that "constitutive" transcription from a naked DNA template containing a RA-responsive promoter [(DRS)5 ~32G] is not affected by RARa/RXRa heterodimers, irrespective of the presence of the tRA agonist. In contrast, there is very little transcription when the same promoter is present in chromatin-assembled templates, unless tRA is bound to the RAR/RXR heterodimers, which results in activation of transcription to a level similar to that achieved with naked DNA
templates. These observations clearly establish that the tRA-induced transcriptional activation mediated by RXR/RAR heterodimers corresponds to the relief of a repression generated by the chromatin organization of the template.
Moreover, these results show that this relief does not correspond to the binding of RXR/RAR heterodimers to the chromatin template, because it is clear from DNase I footprinting data (FIG. 4) that unliganded heterodimers specifically bind to the RA response elements. Thus, the critical events underlying ligand-induced transcriptional activation by RXR/R.AR heterodimers must occur subsequent to their binding to the chromatin template.
Previous studies of the effects of retinoids on transcription of RA
responsive genes, differentiation and apoptosis of mouse embryonal carcinoma cells (Clifford, J., etal.,EMBOJ. 15:4142-4155 (1996); Chiba, H., etal.,J.
Cell.
Science 139:735-747 (1997); Chiba, H., et al., Mol. Cell. Biol. 17:3013-3020 (1997); Roy, B., et al., Mol. Cell. Biol. 15:6481-6487 (1995); Horn, V., et al., FASEB J. 10:1071-1077 (1996); Taneja, R., et al., Proc. Natl. Acad. Sci. USA
93:6197-6202 (1996); Taneja, R., etal., EMBOJ.16:6452-6465 (1997); Minucci, S., et al., Mol. Cell. Biol. 17:644-655 (1997)) and human acute promyelocytic leukemia cells (Chen, J-Y., et al. , Nature 382:819-822 ( 1996)), as well as genetic studies in the mouse (Kastner, P., et al., Development 124:313-326 (1997);
Mascrez, B., et al., Development 125:4691-4707 (1998); and references therein), have established that RAR/RXR heterodimers are the main functional units mediating the effect of retinoids in vivo (see "Related Art" section for additional references). Furthermore, these studies have shown that the ligand-dependent activation function AF-2 of both RAR and RXR partners are instrumental in this mediation. However, in all cases there is a subordination of the activity of RXR
AF-2 to the binding of an agonistic ligand to the RAR partner. Most remarkably, the present in vitro system reproduces these in vivo features of retinoid action.
Although ligands binding to RARa but not to RXRa (tRA, Compound I) can activate transcription on their own, a ligand binding to RXRa but not to RARa (SR11237) is inactive on its own. Consistent with the subordination of RXR
AF-2 activity to that of RAR AF-2, the transcriptional activation brought about by 9-cis RA that induces both RAR and RXR AF-2s, is abrogated by the addition of the RARa antagonist Compound II. Furthermore, in agreement with previous in vivo observations (Clifford, J., et al., EMBOJ. 15:4142-4155 (1996); Roy, B., et al., Mol. Cell. Biol. 15:6481-6487 (1995); Taneja, R., et al., Proc. Natl.
Acad.
Sci. USA 93:6197-6202 (1996); Taneja, R., et al., EMBO J. 16:6452-6465 (1997); Botling, J., et al., J. Biol. Chem. 272:9443-9449 (1997); Minucci, S., et al., Mol. Cell. Biol. 17:644-655 (1997); Durand, B., et al., EMBO J.
13:5370-5382 (1994)), synergistic effects between limiting amounts of RAR ligands and a RXR-specific ligand were observed.
Structural studies (Wurtz, J-M., et al., Nature Struct. Biol. 3:87-94 (1996); Moras, D. & Gronemeyer, H., Curr. Opin. Cell. Biol. 10:384-391 (1998);
Nolte, R.T., et al., Nature 395:137-144 (1998); Darimont, B.D., et al., Genes Dev. 12:3343-3356 (1998); and references therein) have demonstrated that binding of an agonistic ligand triggers a transconformation of the ligand binding domain. This generates an interaction surface for coactivators of the activation function AF-2, that are thought to recruit factors of the general transcription machinery and/or act on chromatin remodeling through histone acetyltransferase activities. Interestingly, Kraus and Kadonaga (Kraus, W.L. & Kadonaga, J.T., Genes Dev. 12:331-342 (1998)) have recently reported that the p300 coactivator (Hanstein, B., et al., Proc. Natl. Acad. Sci. USA 93:11540-11545 (1996);
Chakravarti, D., et al. , Nature 383:99-103 ( 1996)) acts synergistically with ligand-activated ERa to stimulate transcription in vitro from a cognate chromatin template. Similarly, upon addition of exogenous p300, a further 4-fold enhancement was observed in ligand-dependent transcription from chromatin templates in the presence of RXRa/RARa heterodimers. Note that the actual level of enhancement by p300 is likely to be higher, because endogenous p300/CBP is already present in the HeLa cell extract used in the transcription system. However, though the purified p300 exhibits intrinsic histone acetyltransferase activity, the addition of acetylCoA to the present transcription system has no further effect on p300-activated transcription. A similar observation was made by Naar et al. (Naar, A.M., et al., Genes Dev. 12:3020-3031 (1998)) in a study of Spl/SREBP-1-activated transcription on a chromatin template in the presence of CBP. Thus, p300 may further enhance ligand-induced activation of transcription on chromatin templates by bridging RXRa/RARa heterodimers to RNA polymerase II through its interaction with RNA helicase A
(Nakajima, T., et al., Cell 90:1107-1112 (1997)), rather than by locally remiodelling the chromatin structure through histone acetylation.

Example 2 A. p300 and TIF2 coactivators, SWIlSNF remodeling complexes, Acetyl CoA and ATP are all required for efficient retinoid dependent activation of transcription by RAR alRXR a heterodimers on a S "purified" cognate chromatin template, irrespective of the presence of histone HI
Using a crude chromatin template assembled in vitro in the absence of histone H1, conditions for ligand-dependent activation of transcription by RARa/RXRa heterodimers have previously been established that mimicked transactivation by retinoids in vivo (Dilworth, F.J., et al., Proc. Natl.
Acad. Sci.
USA 96:1995-2000 (1999); see also FIG. 8(B) and Example 1). As histone H1 may reduce the transcription efficiency of chromatin templates (reviewed in Paranjape, S.M., et al., Ann. Rev. Biochem. 63:265-297 (1994)), it was first investigated whether the apparent dispensability of histone acetylation could be related to its absence. Chromatin templates were assembled in the absence or presence of stoichiometric amounts ofhistone H 1. Efficient Hl incorporation was evidenced by a clear shift of the nucleosomal repeat length from 170 to 200 by (FIG. 7(A)). Transcription assays carned out with "crude" (DRS)5~32G cognate chromatin templates assembled in absence (-H 1 ) and presence (+H 1 ) of histone Hl showed very little difference relative to the internal control naked pGl template [FIG. 7(B), compare signals from (DRS)S~i2G templates (~32G) and pGl]. In both cases activation of transcription was crucially dependent on the presence of all-traps RA (tRA) that binds RARa but not RXRa, and further enhanced by the addition of the RXR-specific ligand SR 11237 (FIG. 7(B) and Dilworth, F.J., et al., Proc. Natl. Acad. Sci. USA 96:1995-2000 (1999)).
Furthermore, addition of p300 and TIF2 coactivators during the preincubating period that precedes initiation of transcription resulted in an additional 2-3 fold enhancement of transcription, whereas the addition of Acetyl CoA had very little effect, irrespective of the presence of the coactivators (FIG. 7(B), compare lanes 1-4 with S-8, and 9-12 with 13-16).

To exclude the possibility that the Drosophila chromatin assembly extract which remained associated with the "crude" chromatin template could be a source of Acetyl CoA that would mask its requirement, the (DRS)5(32G template was purified by size-exclusion chromatography. This separated the chromatin template and its associated proteins away from "low" molecular weight components, including ATP that is required for chromatin assembly (Pazin, M.J. and Kadonaga, J.T., "Transcriptional andStructuralAnalysisofChromaticAssembledln Vitro,"
in Chromatin: A Practical Approach, Gould, ed., Oxford University Press, Oxford, UK (1998), pp. 173-194). The regular spacing of the nucleosomal array and the binding of histone H 1 were not compromised during this purification, as evidenced by micrococcal digestion (FIG. 7(C)), and in the absence of histone H1.
However, there was a marked loss ofDrosophila S WI/SNF complex (dS WI/SNF) and HAT activity, as judged from western blotting (FIG. 7(D)) and acetylation assays (FIG. 7(F)), respectively. It was therefore examined whether, upon chromatin purification, the ligand-dependent transcriptional activation by RARa/RXRa may have been rendered dependent on ATP-driven chromatin remodeling activities and coactivator HAT activities and their cofactors. In marked contrast with the transactivation observed on "crude" chromatin templates, and irrespective of the presence of H1, the addition of tRA/SR

did not activate transcription by RARa/RXRa heterodimers on "purified"
chromatin templates (FIG. 8(B), compare lanes 1-3). Interestingly, the re-addition of Drosophila embryo chromatin assembly extract during the preincubation period partially restored transactivation, while the addition of Acetyl CoA and/or ATP
had little effect on their own (FIG. 8(B), lane 4; Fig. 8(C), lanes 2, 4 and 7).
Under all conditions tested, similar levels of transcription were detected on the pGl "naked" internal control template [or "naked" (DRS)5~32G template], thus suggesting that factors required to overcome the repressive effect of nucleosomes on transcription were removed during chromatin purification. As the purification of the "crude" chromatin was accompanied by a decrease in both dSWI/SNF complex and HAT activity (see above), the effect of the addition of human S WI/SNF (hS WI/SNF - FIG. 7(E)) and/or NR coactivators p300 and TIF2 during the preincubation period that precedes initiation of transcription was examined. The intrinsic HAT activity of purified p300 has been established using either free or nucleosomal core histones, as substrates (Ogryzko et al. 1996;
see also FIG. 7(G)). Furthermore, the present data (FIG. 7(G)) show that purified TIF2 also exhibits an intrinsic HAT activity similar to that reported for the other NR coactivators (SRC-1, ACTR) that belong to the same p160 family (Spencer, T.E., et al., Nature 389:191-198 (1997); Chen, H., et al., Cell 90:569-580 (1997)). The addition of p300/TIF2 alone did not stimulate transcription (FIG.
8(B), lane 9; FIG. 8(C), lane 5), while the concomitant addition of Acetyl CoA
resulted in a weak activation (FIG. 8(C), lane 10), which was further enhanced by the addition of ATP during the preincubation period (FIG. 8(B), lane 7; FIG.
8(C), compare lanes 8, 10 and 13). As this ATP requirement suggested the involvement of ATP-driven chromatin remodeling activities, it was examined whether the addition of purified human SWI/SNF (hSWI/SNF) complexes may enhance transcription. Very little enhancement, was observed upon addition of either hSWI/SWF or ATP on their own (FIG. 8(B), lanes 8 and 10; FIG. 8(C), lanes 2, 3 and 6). However, the addition of both hSWI/SNF and ATP markedly enhanced the activation observed in the presence of Acetyl CoA and p300/TIF2 to reach levels similar to those achieved with "crude" chromatin templates (FIG.
8(C), compare lanes 10, 13, 14 and 15; see also FIG. 8(B), lanes 6 and 7).
Collectively, the above data indicate that RARa/RXRa heterodimers cannot efficiently initiate transcription from cognate "purified" chromatin templates, unless initiation of transcription is preceded by a preincubation period during which the chromatin template must be exposed to two types of chromatin modifying activities that synergize to activate transcription. The requirement for ATP-driven remodeling activities is suggested by the strong dependence of transcription activation on the presence of ATP during the preincubation period (FIG. 8(C), compare lanes 14 and 15), and by the further enhancing effect of the addition of hSWI/SNF (FIG. 8(B), compare lanes 6 and 7; FIG. 8(C), compare lanes 13 and 15). The clear stimulatory effect of ATP in the absence of hSWI/SNF (FIG. 8(C), compare lanes 10 and 13) most probably reflects the effect of Drosophila ISWI-containing nucleosomal remodeling complexes that are known to be associated with purified in vitro assembled chromatin templates (Di Croce, L., et al., Mol. Cell 4:45-54 (1999); Di Croce, L., et al., Nucleic Acids Res. 27:e1 l (1999)). Note that the reduced, but still apparent stimulatory activity of hSWI/SNF, when ATP was omitted during the preincubation period, may be due to nucleosome remodeling occurring during the "transcription" step (FIG.
8(C), compare lanes 14 and 15; see also FIG. 13(E)). The almost absolute requirement for a HAT activity is demonstrated by the dependence of transcriptional activation on addition of Acetyl CoA (FIG. 8(B), compare lanes 12 and 14; FIG. 8(C), compare lanes 12 and 15; FIG. 8(D), compare lanes 14 and 16), as well as by a drastic decrease of activation in the absence of both p300 and TIF2 (FIG. 8(C), compare lanes 11 and 15; FIG. 8(D), compare lanes 4 and 16).
The activations brought about by p300 and TIF2 were clearly additive, and in both cases synergistic with that resulting from the presence of ATP-driven remodeling activities (FIG. 8(D), compare lanes 6 to 8, 1 0 to 12 and 14 to 16; FIG.
8(C), compare lanes 14 and 15).
Interestingly, under all conditions tested, the presence of histone H1 in either "crude" or "purified" chromatin templates had no effect on transcriptional activation by RARa/RXRa heterodimers, which in all cases was strictly ligand-dependent (FIG. 8(B) and (D)). Thus, in agreement with Sandaltzapoulos et al.
(EMBO J. 13:373-379 (1994)), histone Hl does not appear to further contribute to transcriptional repression when incorporated into chromatin assembled in vitro with a Drosophila embryo extract.
B. BothATP driven chromatin remodelingactivities and coactivatorHAT
activities act to enhance initiation ojtranscription To determine the transcription steps) activated by ATP-driven chromatin remodeling activities and coactivator HAT activities, single-round transcription reactions were performed. To this end, preinitiation complexes (PICs) were assembled on chromatin templates under various conditions (FIG. 9), and transcription was initiated by addition of rNTPs. Sarkosyl (final concentration of 0.5%) was added one minute later, as it dissociates histones from chromatin templates (Izban, M.G. and Luse, D. S., Genes Dev. 5:683-696 ( 1991 )) and allows elongation of transcriptionally engaged RNA polymerases, while inhibiting reinitiation of transcription (Hawley, D.K., and Roeder, R.G., J. Biol. Chem.
260:8163-8172 (1985)). Thus, the ratio between the amount of RNA synthesized in the absence and presence of sarkosyl yields an estimate of the number of rounds of transcription. Clearly, the assembly of productive PICs was synergistically enhanced by addition of p300/TIF2 and hSWI/SNF in the presence of Acetyl CoA
and ATP, but there was approximately 5 rounds of transcription in all instances where transcription was detectable (FIG. 9; similar results were obtained at 0.1 sarkosyl). Thus, both ATP-utilizing chromatin remodeling activitics and coactivator HAT activities appear to synergistically contribute to stimulation of transcription by enhancing the initial formation of productive PICs, rather than by increasing the relative frequency of reinitiation events. This raises the question of which steps preceding initiation of transcription are actually affected to increase the efficiency of PIC formation by RXRa/RARa heterodimers on chromatin 1 S templates.
C. An ATP utilizing activity is required for efficient binding of RARcrlRXRa heterodimers to their chromatin cognate response elements The requirements for efficient binding of heterodimers to "purified"
(DRS)5 ~i2G chromatin template were investigated by DNase I footprint analysis.
Similar results were obtained with chromatin assembled in the presence (FIG.10) and in the absence of histone H1. Upon addition of heterodimers alone, DNAse I hypersensitive sites were generated on the border of each of the five DRS
response elements (REs), but there was little protection of residues located within these REs (compare lanes 3 and 4 with l and 2). The addition of agonistic ligands had no detectable effect on this "weak" footprint, indicating that the heterodimers were "loosely associated" with their REs, regardless of whether their ligands were present. In marked contrast, the addition of ATP (lane 8) resulted in a "much stronger" footprint, which was also ligand-independent (lane 7). This "tighter,"
ATP-dependent, association of the heterodimers was not significantly affected by the addition of hS WI/SNF (compare lanes S-10). Thus, the tight, ATP-dependent binding of RARa/RXRa heterodimers to their cognate REs most likely results from ATP-driven chromatin remodeling by complexes that are known to be associated with purified chromatin templates assembled in vitro using Drosophila embryo extracts. It has previously been demonstrated that ISWI-containing chromatin remodeling activities remain associated with the template after purification by size-exclusion chromatography (Tsukiyama, T. and Wu, C., Cell 83:1011-1020 (1995)) 1995). As such, it was examined whether the "tight", ATP-dependent binding of RAR/RXR heterodimers on the DRS elements could be observed on a greater percentage of chromatin templates in the presence of exogenously added human SNF2h (hIS WI)-containing complexes (which includes the WCRF complex - Bochar, D.A. et al., Proc. Nat'l. Acad. Sci. USA 97:1038-43 (2000). Indeed, the addition of hISWI-containing complexes resulted in a much stronger footprint (FIG.10(B), lane 6) suggesting that "tight" binding of the RAR/RXR heterodimers to their response elements in an ATP-dependent manner is being mediated by ISWI-type chromatin remodeling complexes. This "tight"
ATP-dependent binding of the heterodimers to their response element appears to be due to a localized remodeling of the nucleosomal stucture on the chromatin template. MNase I digestion demonstrates that in the presence of both RAR/RXR
heterodimers and ATP, the favoured dinucleosomes shifts to a mononucleosome in the region surrounding the DRS elements (FIG. 5 Probe A compare lanes 1-6 with 7 and 8) while the nucleosomal structure appears unchanged around the transcription start site (FIG. 5 Probe B compare lanes 1-6 with 7 and 8), or at the 3' end of the globin reporter gene. Interestingly, the addition of p300/TIF2 and Acetyl CoA did not result in "tight" binding of the heterodimers in the absence of ATP (lanes 1 l and 12), and did not detectably affect that observed in the presence of ATP. Thus, the synergism between hSWI/SNF chromatin remodeling complexes and the Acetyl CoA-dependent effect of the p300/TIF2 coactivators appears to occur at a step subsequent to "tight" binding of RARa/RXRa heterodimers.

D) The nuclear receptor coactivators p300 and TIF2 mediate retinoid dependent nucleosomal acetylation To investigate whether the stimulatory effect of Acetyl CoA could be related to acetylation of nucleosomal histones by p300/TIF2 bound to liganded heterodimers, chromatin immunoprecipitation (ChIP) assays were performed using a monoclonal antibody directed against acetylated histone H4. The "purified"
cognate (DRS)5(32G chromatin template and a "purified control" chromatin template similarly assembled in vitro with the pSGS plasmid that contains the SV40 early promoter (Green, S., et al., Nucleic Acids Res. 16:369 (1988)) were incubated in the presence of RARa/RXRa heterodimers under a variety of conditions: presence or absence of either ATP, Acetyl CoA, tRA/SR 11237, p300/TIF2 or hSWI/SNF complexes. At the end of incubation, chromatin templates were fragmented using the restriction endonuclease HaeIII, and ChIP
was carned out. DNA was prepared from "acetylated" and "nonacetylated"
nucleosomal clusters, and its amount was estimated by Southern blotting using [3zP]_labelled oligonucleotides surrounding the promoter regions of (DRS)5(32G
(Dilworth, F.J., et al., Proc. Natl. Acad. Sci. USA 96:1995-2000 (1999)) and pSGS templates. Very little acetylation was detected in the absence ofp300/TIF2 or Acetyl CoA on either of the two chromatin templates (FIG. 12, lanes 1 and 2).
Both cognate (DRS)5~32G and "control" pSGS chromatin templates were similarly "weakly" acetylated upon addition of p300/TIF2 and Acetyl CoA (FIG. 12, lane 3), as expected from the known HAT activity of these coactivators (see above).
Interestingly, upon further addition of tRA/SR 11237, there was an increase (~S-fold) in nucleosomal acetylation of the cognate (DRS)5~32G, but not of the pSGS
"control" chromatin (FIG. 12, lane 4). However, this increased nucleosomal acetylation was also observed further along the reporter gene (+2249 to +2277).
The acetylation of the (DRS)5~32G template at a distance may be due to the fact that a nucleosomal array on a template is not as rigid as a chromosome in vivo, and may permit distal regions to be in close proximity to the heterodimer-bound co-activators.

The above results demonstrate that RARa/RXRa heterodimers bound to cognate REs recruit in a ligand-dependent manner the p300 and TIE2 coactivators whose HAT activity preferentially acetylate nucleosomes located in cis. As expected from the"tighter"association of RARa/RXRa heterodimers to their cognate REs in the presence of ATP (see above), this preferential nucleosomal histone acetylation was strongly decreased in the absence of ATP, while the stimulatory effect of ATP on ligand-dependent nucleosomal acetylation could be overcome by pre-binding the receptor to the chromatin template before purification.
E. Time and interdependence of action of ATP driven chromatin remodeling activities and coactivators in the ligand dependent stimulatory effect of RARalRXRa heterodimers on transcription initiation Order of addition experiments were performed to determine the sequence in which RARa/RXRa heterodimers and their ligands, ATP-driven chromatin remodeling activities, coactivators, as well as transcription factors present in the HeLa nuclear extract, act during the preincubation period that precedes initiation of transcription. The association of RARa/RXRa heterodimers with their chromatin template response elements and their ATP-dependent "tight" binding appear to be initial steps, as activation of transcription was maximal only when heterodimers and ATP were added at the start of the preincubation period (time -40 min; FIG. 13). Addition of heterodimers at -20 min resulted in a marked decrease in transcription, while only very little stimulation of transcription occurred when they were added at the same time as HeLa nuclear extract (- 10 min) (FIG. 13(B)). Interestingly, the addition of ATIP at - 20 min still activated transcription, while its addition at a later stage was inefficient (FIG.
13(E)). Note that, in agreement with the data of DNaseI footprint analyses (see above), these heterodimer binding steps do not appear to require ligands, as their addition at -20 min still resulted in maximal transcription (FIG. 13(B)).
The next steps most probably correspond to acetylation reactions catalyzed by p300/TIF2 recruited by heterodimers tightly bound to their REs.

Indeed, there was no decrease in transcription when either p300/TIF2 (FIG.
13(C)), tRA/SR 11237 ligands (FIG. 13(B)) or Acetyl CoA (FIG. 13(D)), were added at - 20 min, whereas their addition at - 10 min (at the same time as HeLa nuclear extract) or at later times, resulted in severe decreases in activation of transcription. Taken together with the ATP-dependence of nucleosome histone acetylation by p300/TIF2 recruited through liganded heterodimers (see above), these observations indicate that the acetylation step catalyzed by heterodimer--bound coactivators must precede the addition of HeLa nuclear extract.
Finally, the hSWI/SNF remodeling complexes must act at -a step subsequent to transacetylation, as their full stimulatory effect was still achieved upon addition at the same time as HeLa nuclear extract (- 10 min.), whereas no stimulation was observed upon addition, just before initiation of transcription ( 0.5 min) (FIG. 13(C), compare lanes 5 to 8). Thus two distinct chromatin remodeling activities appear to be required at two steps during the course of enhanced PIC formation by RARa/RXRa heterodimers.
An in vitro chromatin template-based transcription system has previously been established that, upon addition of RARa/RXRa heterodimers, mimics transactivation by retinoids in vivo (Dilworth, F.J., et al., Proc. Natl.
Acad. Sci.
USA 96:1995-2000 (1999). A wealth of recent studies have revealed that at least two distinct classes of chromatin modifying activities consisting of ATP-driven chromatin remodeling complexes and histone acetyltransferases (HATs) have evolved to alleviate the repressive nucleosomal organization of chromatin templates. However, a recent study (Dilworth, F.J., et al., Proc. Natl. Acad.
Sci.
USA 96:1995-2000 ( 1999)) did not provide evidence supporting the involvement of these chromatin modifying activities. Using a "purified" cognate chromatin template, it is demonstrated here that, irrespective of the presence of the "linker"
histone H1, both ATP-driven chromatin remodeling activities and coactivators exhibiting intrinsic HAT activities are involved in transcriptional initiation triggered by liganded heterodimeric RARa/RXRa. Furthermore, these two classes of chromatin modifying activities act in a temporally-ordered and interdependent manner.

Experimental Procedures Chromatin Templates Nucleosomal arrays were assembled on (DRS)5~32G (Dilworth, F.J., et al. , Proc. Natl. Acad. Sci. USA 96:1995-2000 (1999)) or pSGS (Green, S., et al., Nucleic Acids Res. 16:369 (1988)) DNA using core histones and assembly extracts prepared from Drosophila embryos (Pazin, M.J. and Kadonaga, J.T., "Transcriptional and Structural Analysis of Chromatic Assembled In Vitro," in Chromatin: A Practical Approach, Gould, ed., Oxford University Press, Oxford, UK (1998), pp. 173-194). Bovine thymus histone Hl (Roche) was added to assembly reactions at 1:4 (w/w) histone H1:DNA ratio. "Crude" assembled (DRS)5 ~32G chromatin templates were purified on a 1.2 mL S-3 OOHR (Pharmacia) spin column that had been pre-equilibrated in EX buffer (10 mM HEPES KOH, pH 7.6, 1.5 mM MgC 12, 10 mM glycerophosphate, 0.5 mM EGTA, and 10%
glycerol) containing 10 mM KCl and spun for 5 min at 11 OOxg (Alexiadis, V-W., et al., EMBO J. 17:3428-3438 (1998)). Approximately 70% of the "crude"
chromatin was recovered in the "purified" fraction. Micrococcal nuclease digestion was used as described (Dilworth, F.J., et al., Proc. Natl. Acad.
Sci. USA
96:1995-2000 (1999)) to examine the periodicity of the nucleosomal arrays.
DNase I Footprinting DNase I footprinting analysis was according to Pazin, M.J. and Kadonaga, J.T., "Transcriptional and Structural Analysis of Chromatic Assembled In Vitro,"
in Chromatin: A Practical Approach, Gould, ed., Oxford University Press, Oxford, UK (1998), pp. 173-194. Purified (DRS)S~i2G chromatin templates (1 nM) assembled in the presence or absence of histone H1 were incubated in a 50 ~1 volume with or without RARa/RXRa (5 nM), tRA (1 ~gM), SR 11237 (1 ~M), TIF2 (2.5 nM), p300 (2.5 nM), hSWI/SNF (10 ng/~1), hSNF2h complexes (5 ng/~,1), ATP (0. 1 mM) and Acetyl CoA (2 ~M) at 270C. After 30 min, CaClz was added to 3 mM and chromatin was digested using 0.75 units (5 ~1) of DNase I (Roche) for 2 min at room temperature. After removal of RNA and proteins, DNase I-generated template fragments were analyzed by PCR-based primer extension footprinting using a [32P]-labelled oligonucleotide situated between and -250 by upstream of the (DRS)S(32G promoter transcription startsite.
Purification of RARalRXRaHeterodimers and Cofactors Purified RARa/RXRa heterodimers were as described Dilworth, F.J., et al., Proc. Natl. Acad. Sci. USA 96:1995-2000 (1999)). A cDNA baculovirus vector was used to express full length human TIF2 containing an amino-terminal His-tag (Voegel et al., EMBO J. 17:507-519 (1998)) in Sf~ cells. TIF2 was purified from whole cell extracts using DEAF-sephacryl (Pharmacia) and Ni2+
NTA resins (Qiagen). Human p300 containing a carboxy-terminal His-tag was expressed and purified as described (Kraus, W.L. and Kadonaga, J.T., Genes Dev.
12:331-342 (1998)). Human SWI/SNF (hSWI/SNF) complexes [containing either BRM (SNF2a) or BRG1 (SNF2~3) - Muchardt, C. and Yaniv, M., J. Mol. Biol.
293:187-198 (1999)] were purified from HeLa cells stably expressing the SNFS
(Ini 1 ) protein with a carboxy-terminal Flag-tag as described (Sif, S., et al., Genes Dev. 12:2842-2851 (1998)). Immunopurification of human SNF2h-containing complexes using a specific monoclonal antibody is to be described elsewhere.
In vitro Transcription Transcription reactions were essentially as described (Dilworth, F.J., et al., Proc. Natl. Acad. Sci. USA 96:1995-2000 (1999); see FIG. 8(A). "Crude" or "purified" chromatin (DRS)5~32G templates (200 pM) assembled with or without histone H 1 were preincubated in the presence or absence of RARa/RXRa heterodimers (1 nM), the RAR-specific agonist tRA (1 ~cM), the RXR-specific agonist SR 11237 (1,uM), hSWI/SNF (2 ng/tcl), p300 (500 pM) , TIF2 (500 pM), ATP (0.1 mM), and Acetyl CoA (2 ~M) for 30 min at 27°C. PICs were formed by adding 60 ~g of HeLa cell nuclear extract (10 ~1) (Dignam, J.D., et al., Nucleic Acids Res. 11:1475-1489 (1983)) for 10 min at 30°C. Transcription was then initiated by the addition of 0.5 mM rNTPs and 50 pM of the internal control "naked" plasmid pGl, and was carned out for 45 min at 30°C in a final volume of 50 ~1. After deproteinization, transcription was quantitated as described (Dilworth, F.J., et al., Proc. Natl. Acad. Sci. USA 96:1995-2000 (1999)).
Histone Acetyltransferase Activity HAT assays were performed as described (Brand, M., et al., J. Biol.
Chem. 274:18285-18289 (1999)). Briefly, recombinant p300 or TIF2 -(1 or 3 pmol) were incubated with 2 ~g of core histones or nucleosomes isolated from HeLa cells and 0.1 ~Ci of [acetyl-1-'4C] Acetyl CoA (65 mCi/mmol - ICN) at 30 ° C. Where indicated, cold Acetyl CoA (2 mM) was added to reactions to confirm the specificity of acetylation. After 1 hr, reactions were run on 12%
SDS-PAGE gels, which were then treated with diphenyloxazol, dried and exposed to autoradiographic film for 5 days.
Chromatin immunoprecipitation (ChIPs) Purified chromatin (DRS)5 ~i2G and pSGS templates (200 pM) assembled in the presence of histone Hl were incubated with or without RARa/RXRa heterodimers (1 nM), p300 (500 pM), TIF2 (500 pM), SWI/SNF (10 ng/~,l) , ATP
(0.1 mM), Acetyl CoA (2 ~M), tRA (1 ~,M), and SR 11237 (1 ~.M) at 27°C
as described above for transcription reactions. After 60 min, the histone deacetylase inhibitor TSA (10 nM) was added to all reactions, MgCl2 was adjusted to 10 mM, and (DRS)S~i2G and pSGS templates were fragmented (10 units of endonuclease Hae III for 1 hr at 37 ° C). Chromatin immunoprecipitation was then carried out essentially as described (Kuo, M.H. and Allis, C.D., Methods 19:425-433 ( 1999)) using the monoclonal antibody 3HH4-3C10 (3C10) generated against a peptide corresponding to the human histone H4 amino-terminal tail (aa 1-24) acetylated at lysines 5, 8, 12, and 16. Nucleosomal arrays were incubated O/N at 4 ° C with ~g of purified 3C10 antibody. Antibody-bound nucleosomes were recovered by incubation with 30 ~1 of protein G-sepharose and washed with EX buffer (see above) containing 500 mM KCI. Acetylated nucleosomes were eluted by incubating with 1% SDS for 30 min, and deproteinized using proteinase K.
Acetylated chromatin templates were quantitated by Southern blotting using [32P]
labelled, 45-base-long, oligonucleotides recognizing the promoter regions of the (DRS)5~32G (-37 to +8) and pSGS (-121 to -77) templates, respectively.
Discussion A. Distinct ATP driven remodeling activities are required for efficient binding of RARalRXRa heterodimers and preinitiation complex formation The requirement for ATP during the preincubation period that precedes the addition of HeLa nuclear extract (FIG. 13(E)) indicates the involvement of ATP-driven chromatin remodeling activity(ies) at an early stage in the process leading to R:I~R/RXR triggered transcriptional initiation. This involvement is strongly supported by the stimulatory effect of ATP on binding of heterodimers to their chromatin template cognate REs in the absence of any of the other components required to achieve efficient initiation of transcription (FIG.
10A).
This is further supported by MNase I digestions which demonstrate an ATP-dependent disruption of chromatin structure localized to the region containing the DRS response elements in the presence of RAR/RXR heterodimers (FIG. 11 ).
Several ATP-driven chromatin remodeling complexes are present in "crude" chromatin preparation assembled in vivo using Drosophila embryo extracts. NURF (Tsukiyama, T. and Wu, C., Cell 83:1011-1020 (1995)), CHRAC (Varga-Weisz, P.D., et al., Nature 388:598-602 (1997)) and ACF (Ito, M., et al., Mol. Cell 90:145-155 (1997)) complexes contain the dISWI ATPase subunit, while the Brahma ATPase subunit (Tamkun, J.W., et al., Cell 68:561-(1992)) is present in the dSWI/SNF complex (reviewed in Varga-Weisz, P.D. and Becker, P.B., Curr. Opin. Cell Biol. 10:346-353 (1998); Kingston, R.E. and Narlikar, G.J., Genes Dev. 13:2339-2352 (1999); see Introduction for additional refs). This latter complex is known to be removed during purification of "crude"

chromatin templates (FIG. 7(D) and Di Croce, L., et al., Mol. Cell 4:45-54 (1999); Di Croce, L., et al., Nucleic Acids Res. 27:e11 (1999)). Similar DNaseI
footprints were obtained in the presence of RARa/RXRa heterodimers with "crude" (which contains ATP) and "purified" (to which only ATP was added) S chromatin preparations (FIG. 10). The ability of hIS WI-containing complexes to further enhance this ATP-dependent footprint suggests that ISWI-containing complexes could mediate "tight" binding of RAIR/RXR heterodimers to their REs.
This is consistent with previous studies which have demonstrated a role for the dISWI-containing complex NURF in assisting the binding of -several transactivators to their cognate REs [progesterone receptor (Di Croce, L., et al., Mol. Cell 4:45-54 (1999a)), GAL4 derivatives (Mizuguchi, G., et al., Mol. Cell 1:141-1 SO (1997)), GAGA factor (Tsukiyama, T. and Wu, C., Cel183:1011-1020 (1995); Okada, M. and Hirose, S., Mol. Cell Biol. 18:2455-2461 (1998)]. The observation that there is a "weak", but clear footprint of the heterodimers on their REs in the absence of ATP, suggests that a NURF-catalyzed nucleosome remodeling results in a greater RE accessibility, and therefore in a "tighter"
binding. The initial "weak" binding of the heterodimers to their REs is most probably the limiting step, as maximal transcriptional initiation is achieved only upon heterodimer addition at the start of the preincubation period, whereas ATP
can be added up to 20 min later with only a moderate decrease in transcription efficiency (FIGs.13 (B) and (E)). Note in this respect that IS WI complex-induced nucleosome remodeling through short-range sliding is known to occur within minutes in vivo, whether or not histone H 1 is associated with the chromatin template (Varga-Weisz, P.D., et al., EMBO J. 14:2209-2216 (1995); Hamiche, A., et al., Cell 97:833-842 (1999)). Alternatively or concomitantly the limiting step may correspond to the possible recruitment targeting of ISWI-containing complexes through interaction with RARa/RXRa heterodimers, as recently suggested in the case of progesterone receptor (Di Croce, L., et al., Mol.
Cell 4:45-54 ( 1999)), and demonstrated for the S WI/SNF complexes that are recruited/targeted by a number of transactivators (see below).
Human SWI/SNF chromatin remodeling complexes are further required at a later stage in the process leading to receptor-mediated transactivation, as optimal transcription is still achieved upon hS WI/SNF addition at the same time as HeLa nuclear extract that provides the transcription machinery required for formation of preinitiation complexes (PICs) (FIG.13 (C). In contrast, the addition of hSWI/SNF just before NTPs is ineffective, indicating that some further chromatin remodeling is indeed required for PIC formation, but does not exclude additional effects on transcriptional elongation. Whether the effect of hS
WI/SNF
on PIC formation involves its targeting to the promoter region through direct or indirect recruitment by liganded heterodimers bound to their REs remains to be seen. However, this possibility is strongly suggested by a number of in vfvo and in vivo observations indicating that S WI/SNF complexes can be recruited through interaction with yeast transcriptional activators (Ryan, M.P., et al. , Mol.
Cell Biol.
18:1774-1782 (1998); Cosma, M.P., et al., Cell 97:299-311 (1999); Gregory, P.D., et al., EMBO J. 18:6407-6414 (1999); Krebs, J.E., et al., Genes Dev.
13:1412-1421 (1999); Natarajan, K., Mol. Cell 4:657-664 (1999); Neely, K.E., et al., Mol. Cell 4:649-655 (1999); Yudkovslcy, et al., Genes Dev. 13:2369-( 1999)), as well as with animal activators (Armstrong, J.A., et al., Cell 95:93-104 (1998); Kowentz-Leutz, E. and Leutz, A., Mol. Cel14:735-743 (1999); Lee, C.-H., et al., Proc. Natl. Acad. Sci. USA 96:12311-12315 (1999)), including several members of the nuclear receptor superfamily (Yoshinaga, S.K., et al., Science 258:1598-1604 (1992); Ichinose, H., et al., Gene 188:95-100 (1997); Fryer, C.J.
and Archer, T.K., Nature 393:88-91 (1998); reviewed in Muchardt, C. and Yaniv, M., J. Mol. Biol. 293:187-198 (1999)). A second posibility is that SWI/SNF may be targeted to the nucleosomes acetylated by p300/TIF2 within the proximal promoter region through bromodomain of SNF2a/b, as bromodomains domains have previously been shown to have high affinity for acetylated lysine residues (Jacobson, R.H. et al., Science 288:1422-S (2000) and references therein).

B. Nucleosomal histone acetylation by receptor-bound coactivators is a prerequisite to enhanced retinoid induced formation of preinitiation complexes The second step in the process leading to transcriptional initiation S triggered by RAR/RXR heterodimers corresponds to the ligand-dependent recruitment/targeting of coactivators (p300 and TIF2) that acetylates histones through their intrinsic HAT activities. To occur optimally this step requires the presence of ATP in addition to liganded heterodimers, coactivators and Acetyl CoA (FIG.12), indicating that it has to be preceded by the ATP-dependent ligand-independent receptor "tight" binding step. On the other hand, this histone acetylation step is a prerequisite for subsequent efficient PIC formation upon HeLa nuclear extract addition, as activation of transcription is strongly decreased when either Acetyl CoA, ligands, and/or p300/TIF2, are added at the same time as HeLa nuclear extract (FIGS. 13(D), (B), (E)). This shows that the observed activation of transcription can be attributed to histone acetylation, but does not exclude that acetylation of nonhistone chromatin proteins, components of the basal transcription machinery and activators by coactivator acetyltransferases (reviewed in Bergen S.L., Curr. Opin. Cell Biol. 11:336-341 (1999)) may play a role at a later stages) for enhanced PIC formation (see below). Note that this histone acetylation step occurs efficiently in the absence of hS WI/SNF (FIGS.

and 13), indicating that no further ATP-driven chromatin remodeling is required.
Furthermore, no stimulation of transcription by HeLa nuclear extract was observed on "naked" cognate DNA templates in the presence of p300/TIF2, Acetyl CoA and liganded heterodimers. It is therefore concluded that nucleosomal histone acetylation by coactivators targeted to the promoter region through liganddependent recruitment by the receptors play a crucial role in transcriptional activation in vivo. Interestingly, Chen, H., et al., Cell 98:675-686 (1999) have recently reported that both estrogens and retinoic acid induce hyperacetylation of histones; at promoters of target genes in cultured cells, and have provided evidence indicating that this hyperacetylation is mediated by receptors and their coactivators. Similarly, Parekh, B.S. and Maniatis, T., Mol.

Cell 3:125-129 (1999) have shown that p300/CBP-mediated localized hyperacetylation of histones may play a crucial role in transcriptional activation at the IFN-~i promoter of virus-infected cells.
Previous failures to reveal a role for nucleosomal acetylation in transcription activated in vivo by nuclear receptors in the presence of coactivators with known intrinsic HAT activities (Kraus, W.L. and Kadonaga, J.T., Genes Dev.
12:331-342 (1998); Dilworth, F.J., et al., Proc. Natl. Acad. Sci. USA 96:1995-2000 (1999); Kraus, W.L., et al., Mol. Cell Biol. 19:8123-8135 (1999); Liu, Z., et al., Proc. Natl. Acad. Sci. USA 96:9485-9490 (1999); see also Naar, A.M., et al., Nature 398:828-832 (1999) and Rachez, C., et al., Nature 398:824-828 (1999)), are most probably due to the use of "crude" chromatin templates assembled in vivo with Drosophila embryo extracts that may contain Acetyl CoA
and HAT activities. In this respect, it is noteworthy that a recent study by Kraus, W.L., et al., Mol. Cell Biol. 19:8123-8135 (1999) has demonstrated the importance of the p300 HAT domain in transcription activated by the estrogen receptor in vivo, although no Acetyl CoA requirement was reported. On the other hand, it has been clearly established that several yeast HAT activities, through recruitement/targeting by transcriptional activators, facilitate transcription in vivo from nucleosomal templates in an Acetyl CoA-dependent fashion (Ikeda, K., et al., Mol. Cell Biol. 19:855-863 (1999) and refs therein).
C. Multiple sequential and interdependent steps in the process leading to ligand dependent transcriptional preinitiation complex formation mediated by RARlRXR heterodimers Consistent with recent yeast in vivo data, showing the ordered recruitment of ATP-driven chromatin remodeling and HAT activities (Cosma, M.P., et al., Cell 97:299-311 (1999); Krebs, J.E., et al., Genes Dev. 13:1412-1421 (1999); Syntichaki, P. et al., Nature 404:414-7 (2000), the present data clearly establish that the events leading to ligand dependent enhancement of transcriptional PIC formation mediated by retinoid receptors on chromatin templates in vivo are both temporally-ordered and interdependent. The first step corresponds to the recognition of cognate REs by unliganded RAR/RXR
heterodimers that results in a "weak" interaction. In the second step, "tight"
binding of the receptors is achieved through nucleosomal remodeling by an ATP-driven machinery, presumably an ISWI-containing complex. The third step is the ligand-dependent recruitment of coactivators (p300 and TIF2) by the receptors.
The fourth step that occurs upon addition of Acetyl CoA corresponds to histone acetylation by the intrinsic HAT activity of recruited coactivators. These first four steps take place in the absence of any of the transcription machinery components present in HeLa nuclear extract. Thus, the "raison d'etre" of these chromatin remodeling steps appears to be to create the proper environment for the next steps that lead to the formation of PICs.
Histone acetylation has been shown to increase the accessibility of regulatory molecules to nucleosomal DNA (Nightingale, K.P., et al., EMBO J.
17:2865-2876 (1998); Steger et al., 1998; for additional refs see Kingston, R.E.
and Narlikar, G.J., Genes Dev. 13:2339-2352 (1999)). In subsequent steps, the liganded receptors bound to their cognate REs and associated with p300/CBP and p160 coactivators (e.g. TIF2) may then recruit the general transcription factors (GTFs), TFIID complexes and RNA polymerase II to form PICs. Direct and indirect interactions between a number of GTFs, TFIID TAF subunits and RNA
polymerase II, and either nuclear receptors (including retinoid receptors;
Rochette-Egly, C., et al., Cell 90:97-107 (1997)) or p300/CBP (Nakajima, T., Cell 90:1107-1112 (1997)), have indeed been shown to occur in vivo, in transfected cells or in yeast two-hybrid assays (reviewed in McKenna, N.J., et al. , J. Steroid Biochem. Mol. Biol. 69:3-12 (1999); Torchia et al., 1998 ; see also Mengus, G., et al., Genes Dev. 11:1381-1395 (1997)), but their physiological relevance for PIC formation in vivo is unknown. Importantly, novel large-multisubunit complexes, that appear to be required for activator-enhanced initiation of transcription in vivo on both "naked" and "chromatin" templates, have been recently characterized and shown to be related in their subunit composition (reviewed in Kingston, R.E., Nature 399:199-200 (1999)). In particular, the so-called SMCC/TRAP (Ito., M., et al., Mol. Cell 3:361-370 (1999)), DRIP
(Rachez, C., et al., Nature 398:824-828 (1999)) and ARC (Naar, A.M., et al., Nature 398:828-832 (1999)) complexes appear to be closely related, if not identical. Most interestingly, the SMCC/TRAPlDRIP/ARC complex (called hereafter SMCC/DRIP) may associate with most, if not all, nuclear receptors through a direct interaction between one of its subunits (TRAP220/DRIP205) and S liganded LBDs (Yuan, C.X., et al., Proc. Natl. Acad. Sci. USA 95:7939-7944 (1998)). As other SMCC/DRIP subunits belong to human and mouse mediator complexes and are homologs of components of the yeast SRB/mediator complex that is found associated with RNA polymerase II holoenzyme (Rachez, C., et al., Nature 398:824-828 (1999); Ito, M., et al., Mol. Cell 3:361-370 (i999));
Kingston, R.E., Nature 399:199-200 ( 1999) and refs therein), one of the function of the SMCC/DRIP complex might be to recruit Pol II holoenzyme to promoters during PIC formation.
Thus, the fifth step in the process leading to ligand-dependent enhancement of transcriptional PIC formation mediated by retinoid receptors on chromatin templates may correspond to the binding of SMCC/DRIP complexes to liganded receptor LBDs through interaction with their TRAP220 subunit. The next step would then be the recruitment of Pol II holoenzyme. Alternatively the liganded receptor might directly recruit preformed SMCC/DRIP-Pol II
holoenzyme complexes. Further in vivo studies using isolated SMCC/DRIP
complexes and purified components from HeLa nudear extract are required to discriminate between these possibilities and to reveal whether nucleosomal acetylation is a prerequisite for these recruitments. As the present data have revealed a need for hS WI/SNF complexes during PIC formation, such studies will also reveal when and how ATP-driven nucleosomal remodeling at the promoter region is involved in PIC formation. Furthermore, as TIF2, and presumably other p160 coactivators, as well as p300/CBP, are apparently bound to the same liganded-LBD surface as TRAP220 (Treuter, E., et al., J. Biol. Chem. 274:6667-6677 ( 1999)), these studies will indicate whether the release of these coactivators from liganded LBDs is a prerequisite for SMCC/DRIP binding, and whether it is achieved through their acetylation, as suggested by the recent in vivo studies of Chen, H., et al., Cell 98:675-686 (1999) using hormone-treated cultured cells.
Further in vivo and in vivo studies are clearly required to characterize the presently unresolved temporally-ordered and interdependent molecular events that underlie enhanced transcriptional initiation mediated by liganded receptors.
United States Application No. 60/151,919, filed September 1, 1999, is hereby incorporated by reference in its entirety. All other documents, e.g., scientific publications, patents and patent publications recited herein are hereby incorporated by reference in their entirety to the same extent as if each individual document was specifically and individually indicated to be incorporated by reference in its entirety. Where the document cited only provides the first page of the document, the entire document is intended, including the remaining pages of the document.

SEQUENCE LISTING
<110> Institut National De La Sante Et De La Recherche Medicate Centre National De La Recherche Scientifique Universite Louis Pasteur Bristol-Myers Squibb Company Chambon, Pierre Dilworth, F. Jeffrey Fromental-Ramain, Catherine <120> In Vitro Transcription Systems and Uses <130> 1383.024PC03 <140>
<141>
<160> 2 <170> PatentIn Ver. 2.0 <210> 1 <211> 8 <212> PRT
<213> Unknown <220>
<223> Description of Unknown Organism: competitor peptide <400> 1 Asp Tyr Lys Asp Asp Asp Asp Lys <210> 2 <211> 30 <212> DNA
<213> Artificial Sequence <220>
<223> Description of Artificial Sequence:
oligonucleotide <400> 2 tcgggagggt tcaccgaaag ttcactcgcc 30

Claims (95)

What Is Claimed Is:

1. A method of identifying an agent which interacts with a retinoid X
receptor (RXR) dimer, said method comprising:
(a) adding an agent to a chromatin based DNA template in the presence of said RXR dimer; and (b) measuring activation of transcription, thereby determining whether said agent interacts with said RXR dimer.

2. The method of claim 1, wherein said chromatin based DNA
template is purified by removing a transcription co-regulator prior to adding said agent.

3. The method of claim 2, wherein said chromatin based DNA
template is purified by chromatography.

4. The method of claim 3, wherein said chromatography is size-exclusion chromatography.

5. The method of claim 1, wherein the second receptor of said RXR
dimer is selected from the group consisting of retinoic acid receptors (RARs), RXRs, thyroid receptors (TRs), vitamin D3 receptor (VDR), peroxisome proliferator activated receptors (PPARs), liver X receptors (LXRs), farnesoid X
receptor (FXR), benzoate X receptor (BXR), constitutive androstane receptors (CARS) and steroid and xenobiotic receptor (SXR).

6. The method of claim 5, wherein said RXR dimer is an RXR.alpha./RAR.alpha. dimer.

7. The method of claim 1, wherein said chromatin based DNA
template comprises a hormone response element (HRE) which is capable of binding by said RXR dimer.

8. The method of claim 1, further comprising adding a co-activator.

9. The method of claim 8, wherein said co-activator is selected from the group consisting of p300, CBP, SRC-1, ACTR, TIF2, PCAF, SWI/SNF, GCN5, PC2, RSF, NURF, CHRAC, ACF, ATP, and Acetyl CoA.

10. The method of claim 9, wherein PC2 is selected from the group consisting of ARC, TRAP, DRIP, and SMCC.

11. The method of claim 8, wherein said co-activator is p300.

12. The method of claim 2, further comprising adding a co-activator after removing said transcription co-regulator.

13. The method of claim 12, wherein said co-activator is selected from the group consisting of p300, CBP, SRC-1, ACTR, TIF2, PCAF, SWI/SNF, GCN5, PC2, RSF, NURF, CHRAC, ACF, ATP and Acetyl CoA.

14. The method of claim 12, wherein said co-activator is p300.

15. A method of identifying a retinoic acid receptor (RAR) agonist, said method comprising:

(a) adding an agent to a chromatin based DNA template in the presence of a retinoid X receptor (RXR)/RAR dimer; and (b) measuring activation of transcription, thereby determining whether said agent is an RAR agonist.

16. The method of claim 15, wherein said chromatin based DNA
template is purified by removing a transcription co-regulator prior to adding said agent.

17. The method of claim 16, wherein said chromatin based DNA

template is purified by chromatography.

18. The method of claim 17, wherein said chromatography is size-exclusion chromatography.

19. The method of claim 15, wherein said chromatin based DNA
template comprises a retinoic acid response element (RARE).

20. The method of claim 15, wherein said RXR/RAR dimer is an RXR.alpha./RAR.alpha. dimer.

21. The method of claim 15, further comprising adding a co-activator.

22. The method of claim 21, wherein said co-activator is selected from the group consisting of p300, CBP, SRC-1, ACTR, TIF2, PCAF, SWI/SNF, GCNS, PC2, RSF, NURF, CHRAC, ACF, ATP, and Acetyl CoA.

23. The method of claim 22, wherein PC2 is selected from the group consisting of ARC, TRAP, DRIP, and SMCC.

24. The method of claim 21, wherein said co-activator is p300.

25. The method of claim 16, further comprising adding a co-activator after removing said transcription co-regulator.

26. The method of claim 25, wherein said co-activator is selected from the group consisting of p300, CBP, SRC-1, ACTR, TIF2, PCAF, SWI/SNF, GCN5, PC2, RSF, NURF, CHRAC, ACF, ATP and Acetyl CoA.

27. The method of claim 25, wherein said co-activator is p300.

28. A method of identifying a retinoid X receptor (RXR) agonist, said method comprising:

(a) adding an agent to a chromatin based DNA template in the presence of an RXR/retinoic acid receptor (RAR) dimer and an RAR agonist; and (b) measuring activation of transcription, thereby determining whether said agent is an RXR agonist.

29. The method of claim 28, wherein said chromatin based DNA
template is purified by removing a transcription co-regulator prior to adding said agent.

30. The method of claim 29, wherein said chromatin based DNA
template is purified by chromatography.

31. The method of claim 30, wherein said chromatography is size-exclusion chromatography.

32. The method of claim 28, wherein said chromatin based DNA
template comprises a retinoic acid response element (RARE).

33. The method of claim 28, wherein said RXR/RAR dimer is an RXR.alpha./RAR.alpha. dimer.

34. The method of claim 28, further comprising adding a co-activator.

35. The method of claim 34, wherein said co-activator is selected from the group consisting of p300, CBP, SRC-1, ACTR, TIF2, PCAF, SWI/SNF, GCN5, PC2, RSF, NURF, CHRAC, ACF, ATP, and Acetyl CoA.

36. The method of claim 35, wherein PC2 is selected from the group consisting of ARC, TRAP, DRIP, and SMCC.

37. The method of claim 34, wherein said co-activator is p300.

38. The method of claim 29, further comprising adding a co-activator after removing said transcription co-regulator.

39. The method of claim 38, wherein said co-activator is selected from the group consisting of p300, CBP, SRC-1, ACTR, TIF2, PCAF, SWI/SNF, GCN5, PC2, RSF, NURF, CHRAC, ACF, ATP and Acetyl CoA.

40. The method of claim 38, wherein said co-activator is p300.

41. A method of identifying a retinoic acid receptor (RAR) antagonist, said method comprising:

(a) adding an agent to a chromatin based DNA template in the presence of a retinoid X receptor (RXR)/RAR dimer and an RAR agonist; and (b) measuring activation of transcription, thereby determining whether said agent is an RAR antagonist.

42. The method of claim 41, wherein said chromatin based DNA
template is purified by removing a transcription co-regulator prior to adding said agent.

43. The method of claim 42, wherein said chromatin based DNA
template is purified by chromatography.

44. The method of claim 43, wherein said chromatography is size-exclusion chromatography.

45. The method of claim 41, further comprising adding said agent in the presence of an RXR agonist.

46. The method of claim 41, wherein said chromatin based DNA
template comprises a retinoic acid response element (RARE).

47. The method of claim 41, wherein said RXR/RAR dimer is an RXR.alpha./RAR.alpha. dimer.

48. The method of claim 41, further comprising adding a co-repressor.

49. The method of claim 48, wherein said co-repressor is selected from the group consisting of SMRT, N-COR and NURD.

50. The method of claim 42, further comprising adding a co-activator after removing said transcription co-regulator.

51. The method of claim 50, wherein said co-activator is selected from the group consisting of p300, CBP, SRC-1, ACTR, TIF2, PCAF, SWI/SNF, GCN5, PC2, RSF, NURF, CHRAC, ACF, ATP and Acetyl CoA.

52. The method of claim 50, wherein said co-activator is p300.

53. The method of claim 42, further comprising adding a co-repressor after removing said transcription co-regulator.

54. The method of claim 53, wherein said co-repressor is selected from the group consisting of SMRT, N-COR and NURD.

55. A method of identifying a retinoid X receptor (RXR) antagonist, said method comprising:

(a) adding an agent to a chromatin based DNA template in the presence of a RXR/retinoic acid receptor (RAR) dimer, an RAR agonist, and an RXR agonist; and (b) measuring activation of transcription, thereby determining whether said agent is an RXR antagonist.

56. The method of claim 55, wherein said chromatin based DNA
template is purified by removing a transcription co-regulator prior to adding said agent.

57. The method of claim 56, wherein said chromatin based DNA
template is purified by chromatography.

58. The method of claim 57, wherein said chromatography is size-exclusion chromatography.

59. The method of claim 55, wherein said chromatin based DNA
template comprises a retinoic acid response element (RARE).

60. The method of claim 55, wherein said RXR dimer is an RXR.alpha./RAR.alpha. dimer.

61. The method of claim 55, further comprising adding a co-repressor.

62. The method of claim 61, wherein said co-repressor is selected from the group consisting of SMRT, N-COR and NURD.

63. The method of claim 56, further comprising adding a co-activator after removing said transcription co-regulator.

64. The method of claim 63, wherein said co-activator is selected from the group consisting of p300, CBP, SRC-1, ACTR, TIF2, PCAF, SWI/SNF, GCN5, PC2, RSF, NURF, CHRAC, ACF, ATP and Acetyl CoA.

65. The method of claim 63, wherein said co-activator is p300.

66. The method of claim 56, further comprising adding a co-repressor after removing said transcription co-regulator.

67. The method of claim 66, wherein said co-repressor is selected from the group consisting of SMRT, N-COR and NURD.

68. A method of identifying a co-activator or co-repressor of a retinoid X receptor (RXR) dimer, said method comprising:

(a) adding a first agent to a chromatin based DNA template in the presence of said RXR dimer and an agonist of said RXR dimer; and (b) measuring activation of transcription, thereby determining whether said first agent is a co-activator or co-repressor of said RXR dimer.

69. The method of claim 68, wherein said chromatin based DNA
template is purified by removing a transcription co-regulator prior to adding said first agent and said RXR dimer agonist.

70. The method of claim 69, wherein said chromatin based DNA
template is purified by chromatography.

71. The method of claim 70, wherein said chromatography is size-exclusion chromatography.

72. The method of claim 68, wherein the second receptor of said RXR
dimer is selected from the group consisting of retinoic acid receptors (RARs), RXRs, thyroid receptors (TRs), vitamin D3 receptor (VDR), peroxisome proliferator activated receptors (PPARs), liver X receptors (LXRs), farnesoid X
receptor (FXR), benzoate X receptor (BXR), constitutive androstane receptors (CARS) and steroid and xenobiotic receptor (SXR).

73. The method of claim 68, wherein said RXR dimer is an RXR.alpha./RAR.alpha. dimer.

74. The method of claim 68, wherein said chromatin based DNA
template comprises a hormone response element (HRE) which is capable of binding by said RXR dimer.

75. A method of identifying a modulator which modulates interactions between a retinoid X receptor (RXR) dimer and a co-activator or co-repressor of said RXR dimer, said method comprising:

(a) adding an agent to a chromatin based DNA template in the presence of said RXR dimer, an agonist of said RXR dimer, and a co-activator or co-repressor of said RXR dimer; and (b) measuring activation of transcription, thereby determining whether said agent modulates interactions between an RXR dimer and said co-activator or co-repressor.

76. The method of claim 75, wherein said chromatin based DNA
template is purified by removing a transcription co-regulator prior to adding said agent.

77. The method of claim 76, wherein said chromatin based DNA
template is purified by chromatography.

78. The method of claim 77, wherein said chromatography is size-exclusion chromatography.

79. The method of claim 75, wherein said chromatin based DNA
template comprises a hormone response element (HRE) which is capable of binding by said RXR dimer.

80. The method of claim 75, wherein the second receptor of said RXR
dimer is selected from the group consisting of retinoic acid receptors (RARs), RXRs, thyroid receptors (TRs), vitamin D3 receptor (VDR), peroxisome proliferator activated receptors (PPARs), liver X receptors (LXR), farnesoid X
receptor (FXR), benzoate X receptor (BXR), constitutive androstane receptors (CARs) and steroid and xenobiotic receptor (SXR).

81. The method of claim 80, wherein said RXR dimer is an RXR.alpha./RAR.alpha. dimer.

82. The method of claim 75, wherein said co-activator is selected from the group consisting of p300, CBP, SRC-1, ACTR, TIF2, PCAF, SWI/SNF, GCN5, PC2, RSF, NURF, CHRAC, ACF, ATP, and Acetyl CoA.

83. The method of claim 82, wherein PC2 is selected from the group consisting of ARC, TRAP, DRIP, and SMCC.

84. The method of claim 75, wherein said co-repressor is selected from the group consisting of SMRT, N-COR and NURD.

85. An in vitro chromatin based DNA template transcription system comprising:

(a) a chromatin based DNA template; and (b) a retinoid X receptor (RXR) dimer.

86. The system of claim 85, wherein said chromatin based DNA
template is purified by removing a transcription co-regulator.

87. The system of claim 86, wherein said chromatin based DNA
template is purified by chromatography.

88. The system of claim 87, wherein said chromatography is size-exclusion chromatography.

89. The system of claim 86, further comprising:

(c) a co-activator.

90. The system of claim 86, further comprising:

(c) a co-repressor.

91. The system of claim 85, wherein the second receptor of said RXR
dimer is selected from the group consisting of retinoic acid receptors (RARs), RXRs, thyroid receptors (TRs), vitamin D3 receptor (VDR), peroxisome proliferator activated receptors (PPARs), liver X receptors (LXR), farnesoid X
receptor (FXR), benzoate X receptor (BXR), constitutive androstane receptors (CARS) and steroid and xenobiotic receptor (SXR).

92. The system of claim 91, wherein said RXR dimer is an RXR.alpha./RAR.alpha. dimer.

93. The system of claim 91, wherein said DNA template further comprises a hormone response element (HRE) which is capable of binding by said RXR dimer.

94. The system of claim 91 which is contained in a kit.

95. The system of claim 94, wherein each of (a) and (b) are contained m separate compartments.