CA2649114A1

CA2649114A1 - Compositions and methods for modulating gene expression

Info

Publication number: CA2649114A1
Application number: CA002649114A
Authority: CA
Inventors: Dusan Stanojevic
Original assignee: Individual
Current assignee: Crosslink Genetics Corp
Priority date: 2006-04-12
Filing date: 2007-04-12
Publication date: 2007-10-25
Also published as: EP2015782A2; WO2007121326A2; EP2015782A4; WO2007121326A3

Abstract

A description of novel artificial transcription factors (ATFs) is provided having a non- peptidic DNA-binding domain, flexible linker, and an effector domain based on a small molecule compound. These ATFs are capable of modulating transcription from nucleic acids both in vitro and in vivo. Importantly, these novel ATFs are capable of targeting and modulating the transcription of native (endogenous) genes in vivo. Method for targeted regulation of gene expression and the development of new class of pharmaceuticals are also provided.

Description

COMPOSITIONS AND METHODS FOR MODULATING GENE
EXPRESSION
GOVERNMENT FUNDING
This invention was made in part with government support under grants 1 R41 GM57712-02A and 1 R43GM077863-01 awarded to the inventor by the National Institute of General Medical Sciences. The government has certain rights in the invention.

FIELD
The invention is in the field of molecular biology, specifically gene regulation. More specifically, the invention relates to the field of artificial transcription factors (ATFs).

BACKGROUND
Modular Structure of Transcription Factors The fundamental level of genome regulation involves activation and repression of RNA
synthesis in a precisely coordinated, gene-specific manner. This process is controlled by a distinct class of proteins called transcription factors. Almost all transcription factors described thus far are modular proteins that contain at least two functional parts: a DNA-binding domain (DBD) and an activation or repression (effector) domain". Typically, the DBD
anchors the transcription factor to the promoter through interaction with specific DNA
sequences. The effector domain participates in interactions with other "target" proteins that are directly or indirectly involved in transcription 3A . As a final outcome, this complex array of protein-DNA
and protein-protein interactions either facilitates or inhibits the assembly of the transcriptional apparatus at the promoter, thus resulting in activation or repression of transcription 5,6 A vast body of existing experimental data strongly indicates that effector domains are functionally independent from DNA-binding domains. For example, the activation domain of one transcription factor can be attached to the DNA-binding domain of a different factor'.
Such hybrid molecules retain their characteristic functions across the range of organisms, from yeast to human cells, thus indicating that the basic regulatory mechanisms are common to diverse eukaryotesT8. Therefore, DNA-binding and effector domains can be viewed as completely separate functional and structural entities joined together as a bifunctional molecule. While the structures of DNA-binding domains have been elucidated in great detail over the last two decades, the precise structural features of effector domains are still poorly understood, most likely because their overall structure appears to be highly flexible 9. For example, it has been demonstrated that almost 30% of randomly selected protein sequences can activate transcription when fused to the DNA-binding domain, thus indicating a considerable "promiscuity" of effector-target interactions'o The Recruitment Model for Transcriptional Activation A simplified model for the mechanism of transcriptional activation in eukaryotes can be summarized as follows. Inside the cell, the RNA Polymerase II enzyme (Pol II) exists predominantly in the form of a preassembled multi-protein complex containing at least 50 other polypeptides, commonly referred to as the Pol II holoenzyme 5. The key biochemical function of a typical transcriptional activator is to help recruit (directly or indirectly) the Pol II
holoenzyme to the promoter and/or to stimulate a subsequent step such as transcriptional elongation ",12. The recruitment of the holoenzyme is currently seen as a crucial initial step in transcriptional activation; the DNA-binding-domain of a transcription factor serves as an "anchor," while the activation domain engages in protein-protein interactions that results in the recruitment of the holoenzyme to the promoter and subsequent initiation of transcription 2.
Recently published work reveals that this recruitment can be achieved through the mediator protein complex 3. The mediator complex is considered to be a"loose" part of the holoenzyme that can be recruited to the promoter independently from other holoenzyme components 4.
Activation domains of some transcription factors are also known to recruit other kinds of proteins to the promoter such as chromatin remodeling enzymes13,'a However, these proteins do not contact the transcriptional machinery directly. Instead, they influence transcription indirectly by changing the local chromatin conformation15.

Gene Repression and Activation as Complementary Therapeutic Strateqies Thus far, efforts in artificial manipulation of gene expression have focused predominantly on repression of gene activity at the post-transcriptional level through diverse molecular mechanisms such as antisense, ribozymes, and RNAi, among others16-

2'. In addition, an interesting new approach to gene repression at the level of transcription has been described recently. It involves blocking the transcriptional machinery at the initiation site with anti-gene PNA or RNA 22,23. Conversely, almost all conventional pharmaceuticals are designed to act as repressors or antagonists, namely, inhibitors of specific protein activity. In contrast, we propose that the opposite therapeutic strategies based on agonists (i.e., activators of specific gene/protein function) have a significant potential to address numerous unique medical challenges. For example, increased expression of the normally inactive embryonic hemoglobin (y-globin) gene in adult erythrocytes is able to compensate for the defective j3-globin protein, thus alleviating the symptoms of sickle-cell anemia 24. This raises the possibility of a novel therapy for sickle cell disease based on targeted activation of y-globin gene. Also, various experiments have shown that increasing the expression or activity of certain tumor suppressor genes such as p53 results in inhibition of tumor growth in mice 25"2' As discoveries of molecular mechanisms underlying various diseases are now advancing very rapidly, the number of such cases is bound to increase in the near future.
Despite the great promise for gene activation as a therapeutic strategy, there are still only a few available molecular tools designed to activate or increase the activity of specific genes. For example, a general strategy for gene activation at the level of transcription has been developed on the basis of engineered zinc-finger proteins 28. This approach is basically a variation of gene therapy, whereby an expression construct or an engineered protein is introduced into cells for a desired physiological effect 29. The main problem with this approach lies in the general lack of cell permeability and poor in vivo transfer efficiency of large molecules such as zinc-finger proteins or their parent expression vectors. For these reasons, the pharmaceutical development of engineered zinc-finger proteins faces many of the same obstacles as other forms of gene therapy. A recent report demonstrated that small modulatory dsRNA can be used to activate expression of a battery of neuron-specific genes in neural stem cells 30. In this case, the action of dsRNA is based on binding to the endogenous NRSF/REST transcriptional repressor, thus blocking its activity via a"decoy"
mechanism 30"32 However, since most endogenous transcription factors participate in regulation of many different genes, the decoy-based molecules have limited specificity, with great likelihood of potential side effects.
Accordingly, there remains a need for novel synthetic molecules capable of modulating (i.e., activating or repressing) expression of nucleic acids, especially of complex genetic targets in living cells, namely, endogenous genes in their native chromosomal context.
The ability to target and regulate (activate or repress) native endogenous genes with novel chemical compounds opens up the possibility for the first practical applications in basic research as well as in medicine.
SUMMARY
This application relates to the regulation of gene expression and, in particular, to the design and synthesis of novel artificial transcription factors (ATFs) comprising small molecule effectors. These novel ATFs are capable of regulating expression of nucleic acids such as, but not limited to, endogenous gene targets in vivo. The invention also relates to the use of

3 such molecules as novel pharmaceuticals that regulate gene expression at the level of transcription. Other potential uses of the ATFs of the present invention include target validation in functional genomics and pharmaceutical drug development.
Specifically, the present invention relates, in part, to unique and novel ATF
molecular designs that possess several unique properties. In particular, these novel designs involve the use of a small molecule as an effector (activation or repression) domain, thus resulting in ATFs that are smaller and more compact and potentially much more active than those described previously 33. Most importantly, the novel kinds of ATFs described in this invention are capable of regulation of endogenous, unmodified genes in their native chromosomal setting in vivo, as demonstrated in assays performed in human tissue culture cells. The achievement of this goal constitutes an important milestone in the development of ATF
technology because, thus far, the biological activity of ATFs or other synthetic activators has been demonstrated only on designed, reporter genes in vitro (i.e. cell-free assays), or in tissue culture transient transfection assays 33-3' In one aspect of the invention, artificial transcription factors for modulating transcription of a nucleic acid are provided. In one embodiment, the nucleic acid is a reporter gene. In another embodiment, the nucleic acid is a eukaryotic gene. In a specific embodiment the eukaryotic gene is a tumor suppressor gene. Non-limiting examples of tumor suppressor genes include p53, Rb, APC, BRCA1, BRCA2, CDKN2A, DCC, SMAD4, SMAD2, MADR2, MEN1, MTS1, NF1, NF2, PTEN, VHL, WRN, and WT1. The artificial transcription factors include a non-peptidic DNA-binding domain and a small molecule compound that functions as a transcriptional effector. In sorne embodiments, artificial transcription factors include a non-peptidic DNA-binding domain, a linker, and a small molecule transcriptional effector. In some embodiments, the small molecule compound is amanitin. In one embodiment, the amanitin is selected from the group listed in Example 1 and Figure 11 of this application.
In a specific embodiment, one end of the linker is bound to the DNA-binding domain, and the other end of the linker being bound to the amanitin molecule. In other embodiments, the linker is bound to the 3' or the 5' end of the nucleic acid DNA-binding domain. In some embodiments, the composition for modulating transcription of a eukaryotic gene binds RNA
polymerase II (Pol II), or RNA Polymerase III, that is, the enzymes catalyzing the transcription of a great majority of eukaryotic genes. In other embodiments, the composition for modulating transcription of a eukaryotic gene binds a component of the RNA Polymerase holoenzyme, a component of the mediator protein complex, or another protein involved in transcription such as histone or P-TEFb.

4 In some embodiments, the DNA-binding domain is a nucleic acid. In certain embodiments, the nucleic acid is a modified nucleic acid. In other embodiments, the nucleic acid includes a modified backbone, and / or the nucleic acid contains modified bases. In other ernbodiments, the nucleic acid includes a modified sugar. In still other embodiments, the modified backbone comprises substitutions for at least one phosphodiester bond. In specific embodiments, the nucleic acid backbone is modified to replace at least one of the phosphodiester bonds with a electrostatically neutral (uncharged) internucleoside linkage. In certain embodiments, the internucleoside linkage is selected from the group consisting of phosphorothioates 38, phosphoroamidites (N3'-P5' oligonucleotides) 39, amides, phosphonates, carbamates, methylenmethylimino, heterocycles, acetals, or any combination thereof. In some embodiments, the nucleic acid backbone is modified to contain peptide nucleic acids (PNA) 40 41 PNA analogues bearing phosphate groups, peptide nucleic acid analogs, or any combination thereof. In some embodiments, the nucleic acid molecule is modified to introduce positive charges to the backbone, bases, or sugar rings. In another embodiment, the DNA-binding domain is a triplex forming oligonucleotide. In another embodiment, the DNA-binding domain is a polyamide. In some embodiments, the DNA-binding domain contains a plurality of pyrrole or imidazole groups. In certain embodiments, the DNA-binding domain binds a sequence of 10 or more contiguous purine bases on one strand. In other embodiments, the DNA-binding domain binds to a sequence containing about 5 to about 10 purines interrupted by at least one pyrimidine base.
In certain embodiments, the linker is a polymer of ethylene glycol, alkyl groups, nucleotides, amino acids, amides, and ketones. In some embodiments, the linker is comprised of a plurality of monomer units selected from the group consisting of nucleotides, peptides, and lower alkyls or other kinds of small organic monomers containing alkyl chain derivatives or other moieties. In other embodiments, the linker is composed of a polymerized glycol. In certain embodiments, the linker comprises about 2 to about 30 glycol units. In other embodiments, the linker comprises at least 2, at least 3, at least 4, at least

5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, at least 18, at least 19, at least 20, or at least 21 glycol units. In certain embodiments, the linker has a length in the range of about 5 Angstroms to about 200 Angstroms. In other embodiments, the linker has a length in the range of about 0.1 Angstroms to about 5 Angstroms. In further embodiments, the linker has a length of about 5 Angstroms to about 10 Angstroms. In some embodiments the linker is flexible, while in other embodiments the linker is rigid.

6 PCT/US2007/066571 In at least some embodiments, the transcriptional effector is amanitin or a chemical derivative thereof that can still bind RNA Polymerase, a component of the holoenzyme, or a component of the mediator, and effect transcriptional activation in the context of a DNA-binding domain. In specific embodiments, the amanitin is selected from the group consisting of a-amanitin, j3-amanitin, y-amanitin, and E-amanitin. In some embodiments, the transcriptional effector contains a plurality of amanitin molecules. In at least some embodiments, the effector is a chemically modified amanitin molecule (e.g., Example 1 and Fig. 11). In at least some embodiments, the effector is a synthetic or natural molecule that binds (i.e., has a chemical affinity for) Pol II molecule or other protein component of transcriptional machinery.
In some embodiments both the DNA-binding domain and the effector domain are modified. In certain specific embodiments, the DNA-binding domain and the effector domain are modified by a modification selected from the group consisting of oxidation, hydroxylation, substitution, and reduction.
In another aspect, the invention provides a pharmaceutical composition comprising one or more ATFs of the present invention, and a pharmaceutically acceptable carrier. In sorne embodiments, the pharmaceutically acceptable carrier is selected from the group consisting of water, phosphate buffered saline, bacteriostatic water for injection (BWFI), and sterile water for injection (SWFI). In certain embodiments, the composition further includes a buffer.
In certain aspects, the invention provides a kit comprising one or more ATFs of the present invention. In some embodiments the kits include other substances such as carriers, and a description of the molecules and/or directions for their use.
In another aspect, the invention relates to a method of regulating the expression of a nucleic acid in a eukaryotic cell. In specific embodiments, the nucleic acid is an endogenous gene or a reporter gene. In one embodirnent, the method comprises contacting the cell, or introducing into the cell, an effective amount of one or more ATFs that can regulate the expression of the nucleic acid into the eukaryotic cell. In sorne embodiments, the eukaryotic cell is selected from the group consisting of a hematopoietic stem cell, a neural stem cell, a T
cell, a B cell, a neuronal cell, a tumor cell, a muscle cell, an epithelial cell, a connective tissue cell, a eukaryotic cell line such as HEK293, HeLa, T47D, HepG2, and A549. In certain embodiments, the ATF is introduced into a eukaryotic cell by a method selected frorn the group consisting of free diffusion (no carrier used), lipofection (use of lipophilic carriers such as lipofectamine or oligofectamine), electroporation, particle bornbardment, calcium-phosphate precipitation, and the use of cell-membrane transduction peptides (that can be attached covalently or non-covalently to the ATF molecule). In at least some embodiments, the amount of transcription initiated from the nucleic acid is at least 50% greater compared to a second amount initiated in the absence of the ATF.
In another aspect, the invention relates to a method of regulating the transcription of an endogenous gene in a eukaryotic cell. This could be any gene. In specific embodiments, the endogenous gene is selected from the group consisting of c-myc, p53, y-globin, T-bet, Nf-xB, insulin gene, insulin promoter factor 1(IPF1 or IDX-1), and E-cadherin. In one embodiment, the method involves introducing an effective amount of one or more ATFs that can regulate the expression of the endogenous gene into the eukaryotic cell. In some embodiments, the method further involves administering the eukaryotic cell to a subject in need thereof. In some embodiments, the eukaryotic cell is selected from the group consisting of a hematopoietic stem cell, a neural stem cell, a T cell, a B cell, a neuronal cell, a tumor cell, a muscle cell, an epithelial cell, a connective tissue cell, a eukaryotic cell line such as HEK293, HeLa, T47D, HepG2, and A549. In certain embodiments, the ATF is introduced into a eukaryotic cell by a method selected from the group consisting of free diffusion (no carrier used), lipofection (use of lipophilic carriers such as lipofectamine or oligofectarnine), electroporation, calcium-phosphate precipitation, and the use of cell-membrane transduction peptides (that can be attached covalently or non-covalently to the ATF molecule). In some embodiments, the one or more ATFs activate an endogenous gene. In at least some embodiments, the amount of transcription initiated from an endogenous (native) target gene is at least 50% greater compared to the amount of transcription initiated in the absence of the composition of the invention. In at least some embodiments, the DNA-binding domain of the compositions for effecting (modulating) transcription of a eukaryotic gene has affinity for at least one site in the promoter of the target gene.
In another aspect, the invention relates to a method of regulating the transcription of an endogenous gene in a subject in need thereof. In one embodiment, the method involves administering a composition comprising one or more ATFs to a subject in need thereof. In some embodiments, the subject is a human subject. In other embodiments, the subject is a domesticated animal. In certain embodiments, the ATF is administered to a subject by topical application, intravenous injection, subcutaneous injection, oral delivery, depot formulations, and inhalation through lungs. In some embodiments, the endogenous gene is selected from the group consisting of c-myc, p53, y-globin, T-bet, NF-xB, insulin gene, insulin promoter factor 1(IPF1 or IDX-1), and E-cadherin.

7 BRIEF DESCRIPTION OF DRAWINGS
FIGURE 1: Schematic diagram of the amanitin-based artificial transcription factor (AmATF).
The structure contains three basic parts: A DNA-binding domain, linker domain, and a small-molecule (e.g., amanitin) effector domain. The three domains are linked through any method known in art to form a three-part structure shown in the figure, with the linker being between the DNA-binding domain and the effector domain.

FIGURE 2: A) Chemical structure of R-amanitin. The numbers 1 to 8 denote positions of each residue. Total molecular weight of (3-amanitin is 920 D. The only chemical difference between a- and R-amanitin lies in the side chain at position 1; the former contains amide instead of carboxyl group. Both a- and P-amanitin have very similar overall conformation, toxicity, and stability 42-44. B) Synthesis of amanitin-based ATF (AmATF). The upper portion depicts three steps in amanitin activation. The lower part represents the coupling of activated amanitin with TFO-linker combination to yield AmATF. TFO with attached polyethylene-glycol linker ending with primary amine is shown on the lower left side.

FIGURE 3: AmATF cell permeability assays. The pictures show human embryonic kidney HEK293 cells incubated with AmATF labeled with green fluorescence tag. Both pictures (representing two different areas of the same well) are taken after 4-hour incubation. It is evident that the AmATF is present inside cells, predominantly in nuclei. No carriers of any kind were used in this experiment. Similar results were obtained with other human and mammalian tissue culture cells such as A549, C6, BHK21 and HeLa (data not shown).
FIGURE 4: Transcriptional activation of reporter gene constructs by AmATFs in tissue culture cells. A) Reporter construct having five repeats of GAL4 and five repeats of AmATF binding sites (the sequence of the site shown below the picture; SEQ ID NO: 56) upstream from the basal promoter driving the expression of luciferase (luc) reporter gene. The relative positions of the TATA box and transcription initiation site (+1) are marked. B) Schematic structures of AmATF#1, AmATF#2 and AmATF#3 show the TFO DNA-binding domain (SEQ ID NO: 1), polyethylene glycol (PEG) linkers of different length and/or orientation, and an effector domain derived from amanitin (AM). The AmATF#2 and AmATF#3 linkers incorporate an additional six-member PEG unit, thus being twice as long as the linker of AmATF#1. While AmATF#1 and AmATF#2 had linkers attached to the 3' end of the TFO, AmATF#3 had linker attached to the 5' end of the TFO. C) The results of luciferase reporter assays with AmATFs in stably

8 transfected hamster BHK21 cells. Vertical bars represent the relative luciferase activity normalized to the control (cells treated with "blank" transfection cocktail, with no AmATFs added). AII AmATFs and TFO-linkers were added to cells with the aid of carrier Lipofectamine 2000 (Invitrogen), at the following concentrations: 100 nM (black bars;
labeled as 1:1 on the right side of chart), 20 nM (1:5), 4 nM (1:25) and 0.8 nM (1:125). It is evident that both AmATF#2 and AmATF#3 demonstrated a significant ability to activate the target reporter gene transcription (up to 4-fold), with AmATF#3 showing a slightly higher activity at most concentrations. At the same time, the TFO-linker control molecules (with no amanitin attached) were not able to activate the reporter gene under these conditions, thus indicating that the amanitin effector domain is essential for AmATFs' biological activity. We have also tested the non-specific control AmATF (namely, the M2AmATF (SEQ ID NO: 3) shown in Figure 5) that has the same structure as AmATF#2 except for the TFO sequence.
Being unable to bind to the AmATF binding sites (SEQ ID NO: 56) in the target promoter shown in Figure 4A, the non-specific control AmATF did not show any significant ability to activate transcription under identical experimental conditions (data not shown). This indicates that the interaction between AmATFs and the target promoters is sequence-specific.
Error bars in Figure 4C represent standard deviation.

FIGURE 5: Endogenous Myc oncogene regulation by AmATFs. A) Nucleotide sequence (SEQ ID NO: 4) written in uppercase represents the part of human oncogene myc lying immediately upstream of one of the main transcription initiation sites (designated "+1). The underlined sequences represent polypurine / polypyrimidine stretches that serve as the two AmATF binding sites. The first (or distal) site is designated M1 site, and the second (proximal) site is designated as M2 site (as shown under each site). The oligonucleotide sequences written in lowercase letters above the promoter sequence represent the two TFOs designed to specifically bind M1 site (SEQ ID NO: 2) and M2 site (SEQ ID NO: 3). Hence these two oligonucleotides are designated as M1 and M2 TFOs, respectively. B) Schematic structures of the two AmATF molecules designed for targeting and regulation of endogenous human myc gene. These molecules are designated M1AmATF (SEQ ID NO: 2; shown on left) and M2AmATF (SEQ ID NO: 3; shown on right). The M1AmATF shown on left contains an TFO (SEQ ID NO: 2; described in Figure 4A) as a DNA-binding domain, hence it is designed to bind the M1 site in the myc promoter. Conversely, the M2AmATF shown on right contains M2 TFO (SEQ ID NO: 3), and is therefore designed to bind to the M2 site. The rest of the structure of both AmATFs is identical. They both contain a linker containing a total of 12 PEG
units that is covalently attached to the TFO terminus. The distal part of the linker is covalently

9 bound to a single derivatized amanitin molecule, as described in Figure 2B and further below.
C) AmATF-mediated transcriptional activation of endogenous myc gene in HEK293 tissue culture cells. The vertical bars represent the relative amount of myc mRNA
where level 1 on the Y axis represents the basal transcription (myc mRNA levels in untreated cells). White bars correspond to signals obtained with AmATFs, while grey bars are signals obtained with control molecules lacking the amanitin moiety (i.e TFO-linkers). Control bars represent untreated cells showing the basal (i.e., background) levels of myc mRNA. M1 bars represent cells treated with 6nM M1AmATF (white) or 6nM M1TF0-linker (grey) designed to bind specifically to the M1 promoter site, while M2 bars represent cells treated with 6nM
M2AmATF (white) or 6nM M2TF0-linker (grey) designed to bind specifically to M2 promoter site.
M1+M2 bars represent the signal obtained with the mixture of molecules (M1AmATF + M2AmATF
or M1TF0-linker + M2TF0-linker). Error bars represent standard deviation. It is evident that the addition of TFO-linkers alone results in decrease of myc mRNA synthesis (transcriptional repression) compared to the basal (control) level. However, the treatment of cells with M1AmATF (SEQ ID N0:2) or M2AmATF (SEQ ID N0:3) results in about 20%-30%
activation of myc transcription. If both myc-specific AmATFs are combined (bars labeled M1+M2), the effect of such mixture on cells is much higher, thus resulting in about 2-fold (100%) increase in myc transcription over the basal (control) level.

FIGURE 6: Targeted regulation of the endogenous human tumor suppressor gene p53 with AmATFs. A) A part of the human p53 promoter sequence (SEQ ID NO: 6) is indicated in uppercase letters, with transcription initiation site (+1) indicated in bold letters. The underlined parts of the promoter represent the homopurine / homopyrirnidine stretches serving as the two ArnATF-binding sites designated as P 1 and P2. The sequences of TFOs designed to specifically bind these two sites are indicated in lowercase letters (SEQ ID
NO: 7; SEQ ID NO:
8). B) Schematic structures of the two p53-specific AmATFs that incorporate the TFO
sequences shown above. These molecules are designated P1AmATF (SEQ ID NO: 8;
shown on the right) and P2AmATF (SEQ ID NO: 7; shown on left), and were designed to bind specifically to P 1 and P2 sites, respectively. The double PEG linker was chemically coupled to the 3' end of each TFO, ending with the ainanitin-derived activation domain (as described in Experimental Methods). C) Transcriptional activation of an endogenous p53 gene by AmATFs in HEK293 tissue culture cells. In this particular experiment, the P
IAmATF (SEQ
ID NO: 8) was administered to the medium together with carrier Lipofectamine (Experimental Methods). The vertical bars show relative amount of p53 mRNA
normalized to the control (untreated cells). Error bars represent standard deviation. It is evident that the addition of increasing concentrations of P1AinATF (SEQ ID NO: 8) results in significant transcriptional activation of p53 (up to 3.5 fold at 20 nM). D) The activation of p53 by AmATFs does not require the use of carriers. In this experiment, 20 nM P2AmATF
(SEQ ID
NO: 7) was applied to HEK293 cells with no carriers, resulting in almost 3-fold p53 activation.
At the same time, the corresponding control molecule (TFO-linker 2; SEQ ID NO:
7) did not cause any significant increase in p53 transcription, thus proving again that the amanitin moiety is critical for AmATF biological activity. Under the same conditions the P1AmATF (SEQ ID
NO: 8) also showed similar biological activity, while the non-specific control AmATF
(M1AmATF shown in Figure 513; SEQ ID N0:2) did not show any significant p53 activation under the same conditions (Experimental Methods; data not shown).

FIGURE 7: Activation of the endogenous human y-globin gene by AmATFs. A) Sequence of the portion of human y-globin promoter (SEQ ID NO: 5) (-337 to +1) is indicated in capital letters, and the transcription initiation site is indicated in a bold letter "A" (+1). The potential homopurine / homopyrimidine AmATF-binding sites in the y-globin promoter are underlined.
The sequences written in lowercase letters above the promoter sequence correspond to the TFOs (SEQ ID NO: 9; SEQ ID NO: 10; SEQ ID N0:11;; SEQ ID NO: 12; SEQ ID NO:
13) that are used as DNA-binding domains of AmATFs designed to activate y-globin gene.
B) The same region of the y-globin gene (SEQ ID NO: 5) is being targeted by the PNA-based AmATF.
The PNA (SEQ ID NO: 14) clamp shown in lowercase letters below and above the prornoter sequence forms a sequence-specific complex with the y-globin gene via D-loop formation as described in the Examples. The letter "j" indicates pseudo-isocytosine that replaces cytosine in one of the PNA strands for improved D-loop stability. C) The structure of one of the AmATF
(SEQ ID NO: 12) designed to target and activate y-globin gene. The TFO DNA-binding domain is based on one of the TFO sequences shown in Figure 7A (SEQ ID NO:
12). The linker containing a total of 18 polyethylene glycol (PEG) units is attached to the 3'end of the TFO. The distal part of the linker is terminated with the primary amine group used for coupling to the amanitin moiety (as described in the Examples below).

FIGURE 8: Targeted regulation of human T-bet gene with AmATFs. A) A portion of the human T-bet pramoter (SEQ ID NO: 22) containing an AmATF-binding site is shown in uppercase letters. T-bet specific binding TFO that serves as an AmATF DNA-binding domain is shown above the promoter sequence in lowercase letters (SEQ ID NO: 20). B) The same portion of the T-bet promoter (SEQ ID NO: 22) being targeted by PNA-based AmATF, whose DNA-binding domain is shown above and below the promoter in lowercase letters (SEQ ID
NO: 21).

FIGURE 9: A) The frequency of potential ATF-binding sites in human promoters.
Each square in the matrix corresponds to one of 14,000 human promoters chosen from the genomic database, and circles indicate the beginning of data for the indicated chromosome (the number within each circle corresponds to one of the 23 human chromosomes). The number of ATF-binding sites ("hits") within the proximal 750 bp of each promoter is indicated by the color of each square (the key is at the boftorn of the figure). AII uniform sequences (poly A, poly G, etc.) were discarded as potential artifacts. AII of the promoters with no sites (white squares) are grouped at the end of each chromosome in order to facilitate the visual estimate of their relative proportion. B) The detailed table showing the sequences and precise locations of ATF-binding sites in promoters of a group of genes residing on human chromosome 10. The table shows the name of the gene, unique genomic database identifier, and one or more polypurine/polypyrimidine sequences (at least 10 bp long) within the first 750 bp downstream from the transcription initiation site. The numbers in square brackets on the left of the each sequence designate the exact location of the sequence relative to the transcription initiation site. C) The detailed table showing the sequences and precise locations of all the ATF-binding sites identified on human chromosome 12. The table shows the same kind of information as described in Figure 9B above.

FIGURE 10: Non-limiting examples of possible chemical structures of linker domains. The sphere on the left represents the effector domain. These structures can be easily synthesized from commercially available components and multimerized to yield linkers of any desired length. The linker domain can comprise any possible combination of these or many other possible chemical structures.

FIGURE 11: Non-limiting examples of possible chemical modifications of the amanitin effector domain. The basic structure of amanitin (top) can be easily modified at several different sites.
Four such sites are denoted as R, R1, R2, and R3. The previously published and described chemical modifications of amanitin are depicted below. There are many more possible chemical modifications of amanitin.

DETAILED DESCRIPTION
The patents and scientific literature referred to herein estab(ishes the knowledge that is available to those of skill in the art. The issued U.S. patents, allowed applications, and references cited herein are hereby incorporated by reference. If there is any conflict between the teachings of the instant application and those of the references incorporated by reference, then the present disclosure governs.

Definitions:
The following definitions are used throughout this description and in the claims, unless the context otherwise requires.
As used herein the term "artificial transcription factors (ATFs)" means transcription factors not normally found in nature and includes synthetic transcription factors.
The term "about" is used herein to mean a numeral value having a range of 20%
around the cited value.
The term "modulate" is used herein to mean activatie or repress ( gene transcription).
By "non-peptidic DNA-binding domain," is meant a domain that does not include a substantial amount of a natural amino acid. Substantially excluding peptidic components in the DNA-binding domain does not, however, exclude the possibility of isolated inclusion of amino acids. For the purpose of this invention, substantially non-peptidic shall mean less than 50%, or less than 20% of natural amino acid content.
A"transcriptional effector" refers to a molecule which, when present in the vicinity of a promoter and bound to a DNA-binding domain, causes an increase or decrease in quantity of RNA synthesized from a particular promoter or class of promoters.
Transcription in the absence of an effector is said to be at a"basal" level. Transcription can be activated (also known as induced or up-regulated) by a positive effector or "activator."
Similarly, a basal or activated level can be repressed or down-regulated by a negative effector or "repressor."
A transcriptional effector can act near or at the site of initiation of transcription of a gene, or over a distance. Genes are transcribed when RNA is synthesized in a 5' to 3' direction using a strand of DNA as a template. The site of initiation of transcription, in which a first ribonucleoside triphosphate complexes with the RNA polymerase, occurs complementary to a site on the DNA template known as "+1," with each successive nucleotide addition occurring complementary to "downstream" sites with increasing positive numbers. "Upstream"

of the +1 site are generally found the DNA regulatory signals, such as the promoter, that specify binding of the transcription factors, components of the transcriptional machinery, RNA
polymerase, and associated proteins. A promoter is generally found within a fixed distance upstream of the +1 site, to position the RNA polymerase holoenzyme appropriately for transcription initiation. Regulatory signals in the DNA sequence that modulate the amount of transcription are generally located in the promoter, for example, upstream of or adjacent to the +1 site.
The term "RNA polymerase" includes both RNA polymerase II (Pol II) and RNA
polymerase III (Pol III). Both of these eukaryotic polymerase enzymes are inhibited by amanitin; Pol II is 1000 times more sensitive to amanitin than Pol III.
The term "small rnolecule compound" is used herein to mean any molecule that can function as a transcriptional effector, and that, in its unmodified state, is between about 0.1 kDa and about 1.4 kDa in size. In some cases, multiple "small molecule compounds" are linked together in an ATF. In such a case, each individual small molecule compound, in its unmodified state, is between about 0.1 kDa and about 1.4 kDa in size. A non-limiting example of a small molecule compound is amanitin. Of course, multiple amanitin molecules may be linked together in an ATF.
The term "domain" as used herein means a portion of an artificial transcription factor that perForms a specific function. For example, the ATFs described herein have at least two domains: a DNA-binding domain that binds to a nucleic acid of interest and a small molecule compound effector domain that binds to a component of RNA Polymerase, the holoenzyme, the mediator protein, or other protein involved in transcription such as histone H1, elongation factor P-TEFb. In some instances, the ATFs also have a linker domain that connects the DNA-binding domain and the small molecule compound effector domain.
The term "modified" is used herein to mean a chemical modification or replacement of the molecule or atom normally present at a given site. For example, a"modified sugar" means the nucleic acid sugar moiety (ribose or deoxyribose) is chemically modified by, for example, introducing moieties that introduce positive charges (e.g., 2'-O-(2-aminoethyl) ribose substitutions). Sirnilarly, a"modified backbone" includes modification of the phosphodiester backbone by, for example, alkylation. A"modified backbone" also includes replacement of the phosphodiester bonds, for exarnple, with phosphorothioate bonds, phosphoramidite bonds, or neutral internucleoside linkages (e.g., PNAs). A"modified base" means any base other than the five major naturally occurring bases (i.e., adenosine, cytosine, thymine, guanine, and uracil), such as, but not limited to, 5-methyl cytosine or 5-propyniluracil.

The term "reporter gene" means a coding unit whose product is easily assayed (e.g., luciferase, beta-galactosidase, green fluorescent protein, and chloramphenicol acetyltransferase). The coding unit may be connected to any promoter of interest so that expression of the coding unit can be used to assay promoter function.
The phrase "therapeutically-effective amount" as used herein means that amount of an ATF composition, or composition comprising such an ATF composition, which is effective for the ATF composition to produce its intended function, e.g., the modulation of gene expression.
The effective amount can vary depending on such factors as the type of cell growth being treated or inhibited, the particular type of ATF composition, the size of the subject, or the severity of the undesirable cell growth or activity. One of ordinary skill in the art would be able to study the aforementioned factors and make the determination regarding the effective amount of the ATF composition without undue experimentation.
The phrase "pharmaceutically acceptable" is employed herein to refer to those ATF
compositions containing such compounds, and/or dosage forms which are within the scope of sound medical judgment, suitable for use in contact with the tissues of human beings and animals without excessive toxicity, irritation, allergic response, or other problem or complication, commensurate with a reasonable benefit/risk ratio.

Detailed Description of Embodiments:
Amanitins (also referred to as amatoxins) are small molecule natural products that are commonly used as potent and specific inhibitors of RNA Polymerase II (Pol II) and to a lesser extent, RNA Polymerase III. Amatoxins are one of the three groups of toxic compounds isolated from the death cap mushroom, Amanita phalloides. Amatoxins are slow-acting (couple of days), yet potent toxins whose ingestion results in necrosis of the liver and kidney cells through inhibition of Pol II, hence blocking transcription and, consequently, protein production 45. In effect, amanitins act as general (non-specific) repressors of transcription.
AII amatoxins share a common general chemical structure (Figure 2A). It consists of a bicyclic octapeptide containing unusual hydroxylated amino acids as well as a cysteine sulfoxide trans-annular bridge 46. The natural amatoxins differ from each other in the number of hydroxyl groups and the chemical nature of the side chain in position 1(Figure 2A). For example, the two most abundant forms, a- and j3-amanitin, contain asparagine and aspartic acid at position 1, respectively. Due to the bicyclic (double ring) structure as well as the presence of several intramolecular hydrogen bonds, amatoxin molecules are very compact and rigid. They have proven resistant to hydrolysis by peptidase enzymes, thus able to survive intact the passage through human or animal digestive system 47. In addition, amatoxins are thermostable 45 The recently obtained crystallographic data reveals that a-amanitin binds to the "cleft"
region between the two largest Pol II subunits, Rpb1 and Rpb2 48. More precisely, the binding site spans the "bridge" helix and the adjacent surface of Rpb1, with only a few contacts in the Rpb2 subunit. The amanitin binding site is not close to the Pol II enzymatic active site, which is consistent with the findings that amatoxins do not interfere with the entry of the nucleoside triphosphates or the formation of the phosphodiester bond 48. Instead, they strongly reduce the rate of translocation of Pol 11 along the DNA, from several thousand to only a few nucleotides per minute 49. It has been suggested that flexible, moving bridge helix or Rpb1 plays a key role in translocation of Pol II 4,50.5' Therefore, the biological activity of amanitin can be explained by the mechanism that involves buttressing the bridge helix, thus constraining its movement and consequently causing Pol II to "stalP" along the DNA 48. These unique biological properties make amanitin a very important and commonly used tool in molecular biology, in particular for general inhibition of Pol II-mediated transcription.
We have surprisingly discovered, however, that amanitin, when coupled to a sequence-specific DNA-binding moiety, is able to function as a gene-specific transcriptional activator. In this manner, a novel kind of artificial analogue of natural transcription factors has been invented, referred to herein as an amanitin-based artificial transcription factor (or AmATF).
The structure and synthesis of such AmATFs is described in Examples below. In one embodiment of the present invention, AmATFs contain the following functional parts (domains):
(i) a non-peptidic DNA-binding domain; and (ii) an effector domain based on the amanitin molecule.
In another embodiment of the present invention, AmATFs contain the following domains:
(i) a non-peptidic DNA-binding domain;
(ii) a linker domain; and (iii) a small molecule compound effector domain (e.g., amanitin, flavopiridol, tagetitoxin, UK-118005).
Since AmATFs are functionally analogous to natural protein transcription factors, the term "domain" is used to designate the specific functionality; however, none of these "domains" that constitute the ATF structure need to be a protein domain.

It is to be understood that in certain embodiments, the ATFs of the present invention can be modified to have two or more non-peptidic DNA binding domains that target different ATF-binding sites that are in the regulatory region of a nucleic acid.
Similarly, in certain embodiments the ATFs of the present invention can have two or more small molecule compound effectors (e.g., two or more copies of the same molecule effector, or combinations of different types of small molecule effectors). In these instances, the two or more non-peptidic DNA binding domains and the two or more small molecule effectors may be linked by any method known in the art, including using the linkers described herein.

(i) The DNA-Bindinq Domain:
The DNA-binding domain of the invention serves to bind the promoter regions of genes in a sequence-specific manner, thus allowing the delivery of AmATFs to designated target genes. The binding of one or more AmATF molecules to the target promoter results in the modulation of transcription (activation or repression) from the corresponding gene.
The DNA-binding domain can be any non-peptidic moiety with the ability to bind promoter DNA in a sequence-specific manner. By non-peptidic moiety, it is meant that the domain does not include a substantial amount of a natural amino acid.
Substantially excluding peptidic cornponents in the DNA-binding domain does not, however, exclude the possibility of isolated inclusion of amino acids. For the purpose of this invention, substantially non-peptidic shall mean less than 50%, or less than 20% of natural amino acid content.
In at least some ernbodiments, the DNA-binding domain is an oligonucleotide, such as a triplex-forming oligonucleotide (TFO). The formation of triple-helical DNA
complexes has been studied extensively over the last fifteen years 52,53 TFOs have been shown to bind in the major groove of the DNA through Hoogsteen hydrogen bonding with remarkable specificity and stability under physiological conditions 52"54. The general Hoogsteen base-pairing rules restrict the triplex targeting to DNA sequences containing polypurine stretches, and such sites have been shown to exist in promoters of most, if not all, native human genes as shown in Figure 9 and Examples 4-8 55 56 55. However, the repertoire of potential target DNA sites has been extended through the use of modified bases, linkers, or various other methods. For example, it is now possible to target a polypurine stretch interrupted by several pyrirnidine residues with the "bridged" or clamped TFOs 56. Also, the triplex recognition scheme can be extended by synthesizing TFOs with nonnatural bases and nucleotide analogs 5758. The triple helix formation has been used as a strategy to target unique sites in the mammalian genome with remarkable specificity and affinity in vifro, in tissue culture experiments and in animal model studies 59 60,61 At the same time, TFOs are commercially available and easy to synthesize and modify with a very broad range of nucleotide analogues and adducts.
Moreover, TFOs do not possess complicated secondary or tertiary structure and generally follow very simple and predictable rules for sequence-specific binding in the major groove of double stranded DNA 54 The chemical structure of TFOs can be based not only on natural (phosphodiester) chemistry, but also on many other types of nucleoside analogs. Such examples include, but are not limited to, the following examples:
(a) those types of nucleoside analogs that incorporate neutral internucleoside linkages to reduce or eliminate mutual repulsion between the TFOs and DNA such as:
oligonucleosides linked by phosphoroamidites 39, amides 62 , phosphonates 63, carbamates 64 methylenmethylimino 65, heterocycles 66 and acetals s'.
(b) those analogs that result in a total replacement of the backbone such as PNA 68, a PNA analogue bearing phosphate groups (PHONA)69, or a peptide nucleic acid analog (PNAA)'o (c) modifications that introduce positive charges, either to the bases " or sugar rings72.
Phosphate backbone can be modified by alkylation with alkylamines, thus giving rise to positively-charged phosphate triester linkages73. The internucleotide linkages can also be replaced with guanidinium group74. A particularly interesing approach to modification of sugars involves TFOs containing 2'-O-methoxy (2'-OMe) and 2'-O-(2-aminoethyl) (2'-AE) ribose substitutions. This type of modification has been shown to greatly stabilize the formation of triple helix in vivo'5.
Most of these modifications render TFOs resistant to degradation by intracellular enzymes, which is of utmost importance in in vivo applications. Most of these modifications also facilitate the interaction between TFOs and the DNA and/or increase their binding affinity and/or specificity.
A particularly interesting nucleoside analog that can be used for the design of the ATF
DNA-binding domain is a peptide nucleic acid (PNA). PNA represents a family of molecular analogs of DNA in which the phosphate backbone is replaced with a backbone similar to that found in peptides. Peptide nucleic acids can bind to single-stranded DNA by Watson-Crick base pairing and can form triple helices to DNA/PNA duplexes much in the way of nucleosides 68,76,77 A PNA "clamp consisting of two PNA strands connected with a flexible linker can form a very stable complex with a DNA duplex and can be designed to target similar kinds of polypurine and polypyrimidine DNA sequences as TFOs 47. Recently, a new generation of PNAs called "pseudo-complementary PNAs" (pcPNAs) were synthesized ". These PNAs target the designated sites on DNA that contain mixed sequence of purines and pyrimidines via double duplex invasion mode. Since the backbone of PNA is not charged, the lack of electrostatic repulsion leads to the formation of strong and stable complexes with DNA. Also, PNA has a smaller mass per monomer unit than DNA and is generally resistant to degradation by enzymes that can attack the phosphate backbone of an oligonucelotide. These and other properties make PNA a very attractive choice for an ATF DNA-binding domain.
In another embodiment, the DNA-binding domain may be a peptide analog, such as polyamides, e.g., polypyrroles and polyimidazoles, described in United States Patent No.
5,874,555. The general function of the ATF DNA-binding domain is to recognize and bind promoters of target genes in a sequence-specific manner. The DNA sequences that are specifically recognized and bound by ATFs are called the ATF-binding sites.
The binding of ATF to the promoter delivers the effector domain to the target gene and, consequently, results in modulation (activation or repression) of its transcription. The promoter sequences that are recognized and bound by the ATF are often present downstream from the transcription start site, but sometimes they can be present upstream of the transcription start site as well. As ATF effector domains are generally able to act over considerable distance, the ATF-binding sites can be located close to the transcription start site (within 1000 base pairs) or further away, at distances over 1000 base pairs.

(ii) The Linker pomain:
In one embodiment of the invention, the DNA-binding domain and the effector domain are coupled through a linker domain. The linker domain of an ATF herein is of sufficient length that ensures that the effector domain of the ATF, when the DNA-binding domain is bound to its recognition site in the promoter, is capable of interacting molecularly with a surface of the Pol II or other protein components of the transcriptional machinery. In at least some embodiments of the invention, the linker is of a flexibility and length such that the effector moiety or domain is free to move above the surface of the DNA within the range specified by the length of a linker. The linker facilitates the interaction between the effector domain and its protein targets and plays a role in increasing the biological activity of ATFs.
In at least some embodiments, the linker of the present invention is at least about 5 A, or at least about 10 A, or at least about 20 A, or at least about 30 A, or at least about 45 A in length. It is recognized that the length of the linker may vary depending upon, among other factors, the location and orientation of the DNA-binding domain at the DNA
template or the chemical composition of the effector domain. It is anticipated that in some embodiments of the invention, the linker can be very shork (less than about 5 A), or that the presence of the linker is not necessary for maximum ATF activity.

The linker is generally of low molecular weight, chemically inert (after incorporation into the ATF structure), and water-soluble. In at least some embodiments, the linker is an oxygen-containing moiety, which improves hydrophilicity and is generally desirable for drug development. In some embodiments, the linker is a polymerized glycol (e.g., polyethylene glycol, polybutylene glycol, etc.). In at least some embodiments, the linker is a polymer consisting of multiple monomeric units such as ethylene glycol, alkyl groups, nucleotides, amino acids, amides, ketones, etc. The polymeric linker may contain monomeric units of the same kind, or any combination of different monomeric units. For covalent linkage, various functionalities may be used, such as amides, carbonic acid derivatives, ethers, esters, including organic and inorganic esters, amino, urethane, urea, and the like.
To provide linking, the particular domain, e.g., the DNA-binding domain or the effector domain, may be modified, for example, by oxidation, hydroxylation, substitution, reduction, etc., to provide a site for coupling to the linker. The domains may terminate in a reactive amine, carboxylic acid, hydroxyl, thiol group, or the like, which are susceptible to conventional chemical reactions to form a stable covalent bond.
One or more linkers can be coupled to the DNA-binding domain through a variety of different sites. For example, the linker can be coupled to the TFO at the 3' end, and the 5' end, or at any position in between the two ends. The coupling of the linker to the DNA-binding domain can be done during the automated synthesis, so that the DNA-binding domain and the linker are synthesized as one contiguous polymer. Alternatively, the coupling of the linker and the DNA-binding domain can be achieved at the later stage, after the automated synthesis is completed. The DNA-binding domain and linker can be coupled covalently using any coupling method known in art, provided that it does not interfere with the functions of the two domains.
In at least some embodiments, a bifunctional crosslinking agent is used to join the linker to either the DNA-binding domain or the effector domain. Suitable crosslinking agents include small bifunctional molecules capable of linking two target groups. The target groups typically are the functional groups discussed above. Exemplary thiol-thiol crosslinking groups include dibromobimane. Exemplary amine-amine crosslinking groups include bis(succinimidyl esters), e.g., bis(succinimidyl esters) of 5,5'-dithiobis-(2-nitrobenzoic acid), or ethylene glycol bis(succcinic acid). Exemplary amine-thiol crosslinking agents include amine-reactive maleimide and iodoacetimide derivatives, such as succinimidyl trans-4-(ma lei m idyl m ethyl)cyclo hexa ne- 1 -ca rboxylate, succinimidyl 3-maleimidylbenzoate, succinimidyl 6-maleimidylhexanoate, or 4-nitrophenyl iodoacetate. Coupling of amine and carboxylic acid groups may also be facilitated by "zero length" crosslinks, a crosslinking agent that is not incorporated into the final product. Exemplary agents include 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide and 2-ethoxy-1-ethyxocarbony-1,2-dihydroquinoline. In another embodiment of the invention, the coupling between the linker and effector can be achieved through a simple chernical reaction involving the activation of the reactive centers (such as primary amine or carboxyl) with a good leaving group (such as succinimide) and a simple chemical reaction with free primary amine or other suitable reactive chemical group.
An exemplary ATF structure and synthesis is shown in Figure 2B and is further described in the following sections.

(iii) The Effector pomain:
The effector domain can be any chemical moiety that has binding affinity for RNA
polymerase enzyme, basal transcription factors, or any other component of the holoenzyme, or any other protein involved in transcriptional regulation such as histones, histone-modifying enzymes, elongation factors and others. Upon binding of the ATF to the promoter DNA, the effector domain contacts its designated protein target and consequently modulates transcription (RNA synthesis) from the target gene. The small molecule effector can be a positive (an activator) or a negative (a repressor) modulator of the amount of basal level of transcription. The nature of the effector (activator or repressor) depends on the exact protein it binds (interacts with), and the strength of interaction. In one embodiment, the small molecule effector is a-amanitin. In another embodiment, the small molecule effector is R-amanitin. In other embodiments, the small molecule effector is a chemically-modified a-amanitin or a chemically-modified R-amanitin, wherein the modified amanitin molecule is able to bind RNA Pol II, RNA Pol III, a component of the holoenzyme, and/or a component of the mediator, and in the context of an ATF, modulate transcription. A small-molecule effector based on amanitin is thought to act through the following mechanism: Amanitin-based ATFs mimic natural transcription activators by having a specific binding afFinity for the promoter (through the TFO) as well as for the RNA Polymerase molecule, a key component of the multiprotein complex called the holoenzyme. Therefore, it is very likely that, through these two interactions, amanitin-based ATFs can participate in the recruitment of the holoenzyme to the promoter, thus increasing the rate of transcription initiation reaction. Once transcription starts, RNA Polymerase is pushed along the DNA in a robust biochemical process driven by rNTPs 'a. At the same time, the AmATF molecule remains bound to the promoter through its TFO
DNA-binding domain. This forward movement forces the abolishment of interaction between AmATF and RNA Polymerase, leaving the amanitin moiety free to interact with new RNA
Polymerase molecules, thus repeating the cycle.

Several naturally occurring variants of amanitin (e.g., a-amanitin, a-amanitin, y-amanitin, and E-amanitin), as well as many synthetic chemical derivatives of amanitin are known and may be used in the ATF molecules of the present invention (such as, but not limited to, those listed in Example 1 and Figure 11). In some embodiments, amanitin moiety is modified to include a suitable chemical coupling group (e.g., an amine, carboxyl or thiol group) that allows for the formation of the covalent bond between the effector and the linker (Figure 1). In one embodiment of the invention, the coupling of the effector to the linker can be achieved through position 1 of the P-amanitin molecule, as described in Figure 2. Another embodiment of the invention involves a different coupling chemistry whereby the a-amanitin is attached to the linker through the tryptophan bridge, as described in Examples below.
Persons skilled in the art can easily devise many other such protocols for covalent coupling of various amanitin derivatives to the linker domain of ATF.
In at least some embodiments of the invention, the effectiveness of ATFs is enhanced by combining several different amanitin-based effectors into a single ATF
molecule. In other embodiments of the invention, the amanitin-based effector can be combined with other kinds of effectors (e.g., synthetic peptides described in 33) in a single ATF
molecule.
It is also contemplated that the effector configurations are not limited to amanitin molecules. Any other small molecule compound effector can be used, so long as the small molecule compound can (i) bind to RNA Polymerase, a component of the holoenzyme, or a component of the mediator, or other protein involved in transcription and (ii) in the context of a DNA-binding domain, modulate transcription of a gene or coding region. Methods of determining whether a small molecule can interact with a protein are well known in the art. In at least some embodiments, a polyglycol or other kinds of inert spacer moieties are introduced between individual arnanitin-based or other effectors attached to the single ATF molecule.
This arrangement increases the conformational flexibility and the ability to interact with one or multiple proteins. In other embodiments, a functionally analogous molecule (natural or synthetic) that specifically binds RNA Polymerase enzyme can be used as an effector domain.
In another embodiment of the invention, the effector can be any small molecule (naturally occurring or synthetic) that binds a component of the transcriptional machinery and causes an effect on the transcription of a specific, designated target gene upon being incorporated into the rest of the ATF structure and introduced into the living cell. Examples of such small-molecule effectors are described in Example 9 and include flavopiridol and 5,6-dichloro-1-beta-D-ribofuranosylbenzimidazole (DRB) that bind the elongation factor P-TEFb and also tagetitoxin, a small-molecule inhibitor or Pol II isolated from bacteria.
Also, the first synthetic small-molecule inhibitor of Pol II (compound named UK-118005) was isolated recently79.
Assays The experimental system described herein can be used to test the activation or repression potential of small molecule effectors in vitro; however, additional embodiments of the invention herein include in vivo assays. The experimental system described herein activates transcription from a linear (typically in in vitro assays) or circular plasmid template that could be used both in in vitro and in vivo assays. For example, the binding of ATF to the promoter DNA is performed in vitro using circular (plasmid) DNA transcription templates, and the resulting pre-formed ATF-template complex is introduced into the tissue culture by the various methods of transformation, including transfection, electroporation, particle bombardment, liposome-assisted techniques, etc. The quantification of the resulting RNA
transcript from a reporter gene transcriptionally downstream from the initiation site (i.e., beta-galactosidase, chloramphenicol acetyl transferase, etc.) in vivo reveals the activity of the tested ATF. Alternatively, a sample of cells that contain a stably-incorporated plasmid having the transcriptional reporter gene is contacted with the ATF of interest. In this case, the binding of the ATF to the corresponding sites in the template promoter occurs in vivo.
Being a relatively small, largely non-peptidic molecule, the ATF readily penetrates the cell membranes in an analogous way to penetration by antisense oligonucleotides.
Alternatively, ATFs are enclosed in liposomes to assist them in penetrating the cell membrane.
Reporter gene activity in cells in the presence and absence of the ATF reveals the extent of transcriptional activation or repression.
The ATF of the invention is used in a method for assaying a test small molecule effector compound for activity as a transcriptional modulator. The method includes linking the test compound covalently to a DNA-binding domain, or covalently to a flexible-linker domain which is covalently bound to a DNA-binding domain to provide a test ATF, the DNA-binding domain having affinity for a DNA-binding site on a DNA template sufficient to bind the site and to modulate transcription at a promoter; contacting the test composition with a transcription mixture including a DNA template, a eukaryotic RNA polymerase molecule capable of forming a complex, either directly or indirectly through other proteins, with the test composition and the DNA template, a buffer and substrates under conditions suitable for RNA
synthesis, such that RNA is synthesized; and determining the quantity of the RNA produced in the presence of the test composition compared to a basal level in the absence of the test composition, which is a measure of the activity of the test composition as an ATF composition. In some embodirnents, the DNA-binding site is a plurality of repeats of the binding-site sequence.
The synthesized RNA may be quantified using any conventional detection system. Such systems for quantitation of RNA product are well known in the prior art.
The ATF compositions of the invention could be adapted to develop in vivo screening system for novel ATFs as well as for therapeutic applications as described above (precise regulation of transgenic cells in vivo). For example, instead of transient transfection (as described in in vivo experiments herein below), stably transfected cell lines can be generated with a reporter construct incorporated into the chromosome. Therefore, any new ATF can be tested for the ability to activate or repress this reporter gene. Endogenous genes can also be used as targets, however, in this case, the signal is detected via quantitative PCR or by using DNA array technology ("DNA chips").
The high-density DNA and oligonucleotide microarrays allow monitoring the expression of many different genes simultaneously, which allows extension of the in vivo assay because it can provide a number of clues about the effectiveness of a particular ATF
design. For example, the monitoring of early changes in gene expression pattern following the treatment of tissue culture cells with ATFs reveals which genes are directly affected by the ATF. ATF
targets in genome could be identified without prior knowledge about the sequences in the promoter. The relative levels of gene expression also provides useful information on the activity of a particular effector. In this manner, both DNA-binding and effector domains are characterized in greater detail simultaneously, along with the identification of potential gene targets for possible medical applications in the future. The microarray analysis is not limited to one type of cell; detection kits are commercially available for many different kinds of eukaryotes, from yeast to humans (such as those produced by Affymetrix).

Applications The novel ATFs of the invention are useful in modulating the transcription of target genes in living cells. The target gene can be any eukaryotic gene (including, but not limited to, hurnan, animal, plant, or fungal genes). Being relatively small (compared to natural protein transcription factors), largely or entirely non-peptidic molecules, the novel ATFs described herein readily penetrate the cell membranes in an analogous way to cell penetration by antisense oligonucleotides 80-82 . Alternatively, they can be applied with various kinds of lipophilic carriers to further assist them in penetrating the cell membrane 83, 84. It is also possible to use covalent or non-covalent conjugation with various kinds of transduction peptides to assist in cell penetration 85,86. Upon introduction into living cells, the present ATFs diffuse across membranes and enter the nucleus where they locate and bind the corresponding site(s) lying in the promoter of the target gene. The subsequent interaction of promoter-bound ATFs with RNA Polymerase or other associated enzymes accomplishes the desired effect (activation or repression) of transcription of the target gene.
This effect can be observed either by rnonitoring its physiological consequences, or by measuring the amount of mRNA synthesized (transcribed) from the target gene. This amount is then compared to the amount of the same mRNA produced under the same conditions in the cells that have not been treated with ATFs (this corresponds to the basal level). Alternatively, some target genes may produce a protein that is excreted from cells, or is otherwise easily detected by various enzymatic assays or antibodies (western blots). In these cases, the net effect of ATFs can be observed indirectly by measuring the amount of the protein product (instead of the mRNA).
These methods for measuring the amounts of mRNA or protein are well known in prior art.
For these and many other applications, the DNA-binding domain is designed to be able to bind promoter of the target gene in a sequence-specific manner. To achieve that, the sequence of the target gene is analyzed, the presence of potential ATF-binding sites is determined, and one or more ATFs that bind specifically to the target promoter are synthesized. A non-limiting example of rules for designing the TFO-based DNA-binding dornains is the following: The most suitable promoter sequence has about 10 or more contiguous purine bases (G or A) in one strand and pyrimidine bases (C and T) in the complementary strand. This polypurine/polypyrimidine stretch need not be perfect. The TFO
sequence is designed in such a way as to contain a G residue opposite the GC
base pairs of the promoter and A residue opposite AT base pairs. If a polypurine target sequence is interrupted by a T or C, the incorporation of a T residue in the TFO will maintain the stability of the complex 82 54. An alternative triplex pairing scheme has also been described, where T
residue goes opposite AT base pairs, thus resulting in GT-rich TFOs 33, $' .
Both GA and GT-rich TFOs have been applied successfully by many researchers under a variety of physiological conditions, and they are roughly equivalent in terms of average-binding affinity and specificity. Several specific illustrative examples of application of these rules in practice are provided in Figures 4, 5, 6, 7, 8 and the corresponding Examples below).
In addition to these rules, there are many other available TFO design possibilities based on modified bases that recognize an expanded range of "mixed" DNA sequences 57, 58. Also, various backbone modifications, such as PNA, can also be designed to specifically recognize polypurine/polypyrimidine sites in the promoters. In one such example, PNA is synthesized in two shorter sequences separated by a flexible linker to make a PNA clamp 88 .
This sequence of each of these two sequences are designed to contain a T residue opposite AT
base pairs in the prornoter, and a C residue opposite GC base pairs in the prornoter. The incorporation of two lysine residues at each end of the PNA clamp further facilitates the binding to DNA due to the electrostatic interactions 88. Several specific examples of promoter-specific PNAs are illustrated in Figures 7 and 8. The linker domain is then attached at one or both ends of the PNA clamp via standard methods during automated synthesis or post syntheticallly, by any method known in the art. The distal end of the linker serves for coupling to one or more effector domains in a manner identical to that described in Example 1.
The synthesized ATFs can be purified, lyophilized (dried), or dissolved in water or any type of physiological buffer or medium such as PBS, DMEM, and any other water-based solvent as well in organic solvents mixable with water such as acetonitrile or DMF, and stored at any suitable temperature (e.g., at -80 C, -20 C, or 4 C).
In one aspect of the invention, ATFs are provided in an amount sufficient to, upon introduction into the living cell, diffuse across intracellular membranes, enter the nucleus, bind the specific site on a designated target gene, and specifically modulate the transcription of the designated target gene. The application of ATFs can involve cells in tissue culture, animal models, such as mouse, or human subjects (patients). Some of the diseases and corresponding gene targets for regulation by ATFs include, but are not limited to, the following:
cancers, such as those associated with the oncogenes bcl-2, Fos, Jun, AML-1;
tumor suppressor genes p53, Rb, APC, BRCA1, BRCA2, CDKN2A, DCC, SMAD4, SMAD2, MADR2, MEN1, MTS1, NF1, NF2, PTEN, VHL, WRN, and WT1; cases oftype 2 diabetes (estimated at

10% or more) caused by mutations in genes encoding transcription factors such as HNF-4a, HNF-10 and IPF-1; obesity, which has been linked in some cases to defects in transcription factors PPARy and SIM1; sickle cell anemia and thalassemia in which activation of the inactive copy of the y-globin gene (normally expressed only during early development) may be therapeutic; Hodgkin's disease, which has been linked to mutations affecting Oct 2 transcription factor and/or BOB1/OBF1 co-activator; defects in apoptosis, which contribute to diseases such as cancer and spinal muscular atrophy (SMA); hereditary diseases involving defects in transcription factors (e.g., Rieger Syndrome); and eukaryotic pathogens (e.g., HIV, adenovirus, influenza, Lyme disease, and sleeping sickness).

Pharmaceutical Compositions In another aspect, the present invention provides pharmaceutically-acceptable compositions which comprise a therapeutically-effective amount of one or more of the ATF
compositions of the present invention, formulated together with one or more pharmaceutically acceptable carrier(s). Alternatively, the ATF pharmaceutical compositions can be applied without carrier(s). The pharmaceutical compositions and methods described herein can include one or more ATF compositions of the present invention.
ATF compositions of the present invention can exist in free form or, where appropriate, in salt form. Pharmaceutically-acceptable salts and their preparation are well known to those of skill in the art. The pharmaceutically acceptable salts of such compounds include the conventional non-toxic salts or the quaternary ammonium salts of such compounds which are formed, for example, from inorganic or organic acids of bases. The compounds of the invention may form hydrates or solvates. It is known to those of skill in the art that charged compounds form hydrated species when reconstituted in water from a lyophilized form, or form solvated species when concentrated in a solution with an appropriate organic solvent.
The pharmaceutical preparations can include one or more ATFs in dry form (e.g., lyophilized alone or with a stabilizer) or in liquid solutions or suspensions (e.g., in a pharmaceutically acceptable carrier or diluent). Pharmaceutically-acceptable carriers for parenteral administration of liquids include, without limitation, water, buffered saline, polyols (e.g., glycerol), polyalkylene glycols (e.g., propylene glycol, liquid polyethylene glycol), vegetable oils, hydrogenated napthalenes, or suitable mixtures thereof. The ATFs can also be formulated with buffers or excipients.
In some embodiments, the ATFs are formulated in sustained-release particles or implantable devices. For example, such particles or devices can be formed from biocompatible, biodegradable lactide polymers, lactide/glycolide copolymers, polyoxyethylene-poloxypropylene copolymers, ethylene-vinyl acetate copolymers, and the like, to control the release of the ATF. Other potentially useful parenteral delivery systems include osmotic pumps, implantable infusion systems, and liposomes. In one embodiment, the ATFs are delivered to a patient using DUROS Implant or ALZAMER Depot technology (Alza Corporation).
Methods for formulating pharmaceutical preparations can be found, for example, in Remington's Pharmaceutical Sciences (18th edition), ed. A. Gennaro (1990), Mack Publishing Company, Easton, PA.
The route of administration for compounds, includes, but is not limited to, intravitreal, parenteral, topical, enema, oral, intravenous, ex-vivo, intra-tumoral and as an aerosol.
The amount of compound which will be effective in the treatment or prevention of a particular disorder or condition will depend, in part, on the nature of the disorder or condition, and can be determined by standard clinical techniques. In addition, in vitro or in vivo assays may optionally be employed to help identify optimal dosage ranges. Effective doses may be extrapolated from dose-response curves derived from in vitro or animal model test systems.

The precise dosage level should be determined by the attending physician or other health care provider and will depend upon well known factors, including route of administration, and the age, body weight, sex, and general health of the individual; the nature, severity and clinical stage of the disease; the use (or not) of concomitant therapies; and the nature and extent of genetic engineering of cells in the patient.

Kits The invention also provides a pharmaceutical package or kit comprising one or more containers holding one or more ingredients including a precursor composition having a linker covalently bound to a DNA-binding domain, the DNA-binding domain having affinity for a DNA-binding site on a DNA template sufficient to bind the site and modulate the transcription at a promoter. The precursor composition generally contains a reactive end group that can be used to couple the precursor compound to a test compound of interest for assessing the activity of the composition in transcription. The kit also includes a transcription mixture comprising a DNA template and a eukaryotic RNA polymerase molecule that forms a complex with the DNA template.
In another aspect, the invention provides a kit comprising, in a suitable container, a therapeutically effective amount of one or more substantially-pure ATFs and a second agent.
The ATF and the second agent may be in separate containers or formulated together in one container. The components of the kit may be provided in a liquid solution(s), and/or as a dried powder(s). When the components are provided in liquid solution, the liquid solution is a sterile solution. When the reagents or components are provided as a dry powder, the powder can be reconstituted by the addition of a suitable solvent (e.g., water or buffered saline). The solvent may be provided as part of the kit.
Optionally associated with the kit may be instructions for using the precursor composition according to the methods of the invention.
The invention is further illustrated in the following examples, which are provided for the purpose of illustration only and are not intended to be limiting of the invention, the full scope of which is shown in the claims that follow the specification.

EXAMPLES

Example 1 Design and Synthesis of Amanitin-Based ATFs (ArnATFs) From a practical standpoint, the development of ATFs is greatly facilitated by the ability to combine different "stock" components (effectors, linkers, and DNA-binding domains or DBDs) into new ATF structures. To accomplish this, a simple and general chemical procedure has been developed for coupling diverse effector domains (such as synthetic peptides, peptidomimetics, and non-peptidic moieties) to different DBD-linker combinations. This procedure has been used in the synthesis of amanitin-based ATFs (AmATFs) as described in Figure 2.

Synthesis As a starting material we chose (3-amanitin instead of the more commonly used a-amanitin because the former contains a convenient reactive carboxyl group in position 1. This group has served previously for synthesis of various kinds of chemical derivatives of amanitin43. Recently published structure of amanitin complex with Pol II
reveals that carboxyl 1 does not form any contacts with Pol I I enzyme, and does not form a part of amanitin "active center.48"
The activation of (3-amanitin has been accomplished in three steps:
(i) Activation of carboxyl 1 with N-hydroxysuccinimide (NHSU);
(ii) Reaction of activated amanitin with amino-caproic acid (AC) to yield amanitin-AC; and (iii) Reactivation of Amanitin-AC with NHSU.
In this manner, the original carboxyl group is "extended" by a five-alkyl chain spacer derived from AC. This has proven important because, in the absence of an AC
extension, the activated p-amanitin reacted very poorly in the coupling reaction with TFO-linker (yields below 5%). This is likely due to interference from adjacent internal amide and/or hydroxyl groups, thus making the activated form unstable 43. However, the addition of an AC
spacer markedly improved this reaction. The final coupling of activated amanitin-AC to TFO-linker routinely yielded over 80% of the final product - AmATF.
The first step in AmATF design involved the synthesis of modified oligonucleotides having a general formula such as the following examples:
5'TTGTGGTGGGTGGGGTGTGGGTXY3' or (SEQ ID NO: 1)XY for (AmATF#1) 5'TTGTGGTGGGTGGGGTGTGGGTXXY3' or (SEQ ID NO: 1)XXY for (AmATF#2) 5'YXXTTGTGGTGGGTGGGGTGTGGGT3' or YXX (SEQ ID NO: 1) for (AmATF#3) 5'TGGGGTGGTTGGGGTGGGTGGGGTGGGTXXY3' or (SEQ ID NO: 2) XXY (for M1AmATF) and 5'TGGGTGGGTGGTTTGTTTTTGGGXXY3' (SEQ ID NO: 3) XXY (for M2AmATF) In the preceding sequences, X represents the Spacer Phosphoramidite 18 (Glenn Research) and Y represents the amino modifier residue bearing the primary amine on a short tether (Amino-Modifier C6, Glenn Research). Spacer Phosphoroamidite residue incorporates a polyglycol linker (consisting of 6 ethylene glycol monomer units) at the 3' end or 5' end of the TFO sequence. The primary amine incorporated at the distal end of the linker serves as a reactive group for the attachment of activated [3-amanitin. Both modified residues (X and Y) are introduced into the DNA oligonucleotide using standard automated DNA
synthesis methods (Operon Biotechnologies) 33 The activation of P-amanitin has been accomplished in three steps (described in Figure 2):
Reaction for step #1: 0.5 mg of purified (3-amanitin (Sigma) was mixed with 15 mg of N-hydroxysuccinimide (Acros Organics) in 50 l of dry dimethylformamide (DMF) with 20 l of N.N'-Diisopropylcarbodiimide (TCI America). After 40 minutes at room temperature, the reaction mixture was separated by reverse-phase HPLC [Buffer A: water + 0.1 %
Trifluoroacetic acid (TFA; Fluka); Buffer B: acetonitrile +0.1 % TFA].
Collected fractions containing activated amanitin (Am-OSU) were dried in a lyophilizer.
Reaction for step #2: 0.2 mg of dry Am-OSU was dissolved in 50 l of DMF. 15 mg of E-Aminocaproic acid (AC; Chem-Impex International) was dissolved in 50 l of 0.1 M
phosphate buffer pH7.2. Both solutions were mixed together and incubated at room temperature for 30 minutes. The reaction mixture was separated by HPLC as described above.
Reaction for step #3: The activation of amanitin-AC was essentially a repetition of step #1, performed exactly as described above. The coupling of the activated amanitin-AC (Am-AC-OSU) to the TFO-linker was done as follows: 5 nanornols of TFO-linker was dissolved in 100 l 0.1 M phosphate buffer and about 50 nanomols of Am-AC-OSU was dissolved in 50 l of DMF. The two solutions were mixed together and incubated at room temperature for 30 minutes, followed by reverse-phase HPLC (Buffer A: 0.1 M ammonium-bicarbonate;
buffer B:
acetonitrile) and drying in lyophilizer. The mass of the final product (AmATF) was confirmed by MALDI mass spectroscopy (at MIT Biopolymers Laboratory). The sequence specific DNA

binding ability of the synthesized AmATFs was confirmed by standard gel-mobility shift assays as described previously (data not shown)33 An example of an alternative coupling chemistry that can be used in the synthesis of amanitin-based ATFs is described below: As described in Figure 2, the coupling reaction of P-amanitin is accomplished through the side chain at position 1. An alternative site for amanitin derivatization has been described in literature 43. It involves the tryptophan "bridge" in position 4(Figure 1). The available data indicates that, similar to position 1, tryptophan 4 is also a "non-criticaP' residue for amanitin function. For example, tryptophan 4 does not form any close contacts or hydrogen bonds with RNA Polymerase enzyme 48. The attachment of various spacer moieties, lipophillic residues and radioactive labels at position 4 preserves much of the ability of amanitin to inhibit RNA Polymerase 43. However, the crystal structure of amanitin -RNA Polymerase complex reveals one major difference: residue 1 and residue 4 point in very different, almost opposite, directions 48. For this reason the coupling through position 4 is likely to yield amanitin-based ATFs that assume different overall conformation upon interaction with RNA Polymerase molecule. This may lead to an improved ability to interact with transcriptional apparatus and, therefore, higher activity in transcription assays. The starting material for this reaction is a-amanitin instead of R-amanitin since the carboxyl 1 is bound to interfere with derivatization at position 4. The reaction is accomplished as described in 43. 1 mg of a-amantin (Sigma) will be dissolved in is 0.5 ml of dry ethanol. In another tube 1 mg of sodium is dissolved in 2 ml of dry ethanol and 0.25 ml of the sodium ethylate solution will be added to amanitin solution. The reaction mixture is evaporated in lyophilizer, and the residue is dissolved in 0.2 ml of dry DMF and reacted with 3 equivalents of 6-bromocaproic acid (Merck) for 12 hours at room temperature. Under these conditions the alkylation occurs predominantly at the phenolic hydroxy group. The reaction mixture is resolved by reverse phase HPL.C. The alkylated a-amanitin is activated at the introduced carboxyl group and coupled with TFO-linker exactly as described in preliminary results above and in Figure 2.

Chemical Modifications of DNA-Binding Domain Linker and Effector pomain:
(i) The DNA-Bindincl Domain:
Further chemical modifications designed for improvement of the DNA-binding domain moiety are described below. The TFO plays a role in delivering AmATFs from the extracellular environment to the target promoter incorporated into the chromosome inside the nucleus.
TFOs are very convenient for use as DNA-binding domains of ATFs because: (i) they show intrinsic cell permeability, and (ii) the existence of the vast amount of data on design, synthesis, and application of TFOs accumulated over the last 18 years. Based on the published data, as well as our own research, we believe that unmodified TFOs can be greatly improved with simple chemical modifications. For example, the stability (resistance to degradation) of AmATFs is one of the most important properties, not only with respect to future in vivo applications, but also because both intra- and extra-cellular stability can greatly influence the outcome of tissue culture transcription assays. The more stable AmATF
structures are, the more likely they are to have a substantial biological activity. There are several types of chemical modifications of TFOs that have been proven to increase the stability by rendering the DNA backbone resistant to degradation by nuclease enzymes. For example, the replacement of the phosphodiester DNA backbone with phosphorothioates has been extensively used to increase half-life as well as cellular uptake of antisense oligonucleotides 89. Introduction of phosphrothioates does not affect significantly the binding properties of purine-rich TFOs 90. Other suitable modification involves N3'-P5' phoshporoamidate backbone that has been shown to increase both stability and DNA-binding affinity of TFOss'. The synthesis of AmATFs incorporating each of these two modifications as well as many other possible chemical modifications (such as listed in previous chapters) and their testing in toxicity and transcription assays is a straightforward task, and it will be accomplished by using methods described in examples and well known in the art.
For example, each sequence listed in the table can be synthesized with regular (phosphodiester) DNA monomers, regular (phosphodiester) RNA monomers, phosphorothioate and N3'-P5' phoshporoamidate backbone. Since these and many other types of natural and modified monomers are commercially available, the TFO-linker synthesis and modifications can be readily achieved by methods known in the art.

(ii) The Linker pomain:
The linker dornain can be any relatively inert chemical moiety that physically separates the DNA-binding domain from the effector domain and, thus, facilitating the interaction of each of these two domains with their putative target molecules. In terms of chemical composition, there were several kinds of ATF linkers described in literature such as polyethylene glycol (PEG), polyglycine, polyamide-ethylene g(ycol (AEEA), and alkyl chains 33,36,92 37 There are also many other possibilities for linker chemical design, some of which are listed in Figure 10.
The optimal linker length likely depends on parameters such as the nature of DNA-binding and effector domains and the exact position of the AmATF binding site within the target promoter.
The data shown in Figure 4 implies that, under these experimental conditions, AmATFs having a linker with 12 (polyethylene glycol) PEG units are more biologically active then those having shorter linkers with 6 PEG units. This indicates that varying the length of a linker may yield AmATFs having higher biological activity. For example, the length of the linker can be easily increased by synthesizing AmATFs with linkers comprising 15, 18, 21 or rnore ethylene-glycol monomers. This is not expected to cause any adverse effect on AmATFs' pharmacologically relevant properties because conjugation to PEG molecules is often used to improve properties of many drugs 93. In addition, the overall geometry of the AmATF structure can be easily changed by incorporation of the linker at the 5' terminus of the TFO (as illustrated above in listed example of AmATF#2-5'). AII linkers will be incorporated during the automated TFO
synthesis as described above. Alternatively, the linkers can be attached to the DNA-binding domains at a later stage through the use of bifunctional crosslinkers or any other conjugation method known in the art.

(iii) The Effector pomain The effector domain comprises amanitin, modified amanitin, or any other naturally-ocurring, synthetic or semi-synthetic small molecule that confers the ability to modulate transcription when bound to the linker and DNA-binding domain. There are several naturally occurring chemical variants of amanitin, as well as numerous examples of chemical modifications of amanitin known in the art, not limited to the following list:
a-amanitin, R-amanitin, E-amanitin, O-Methyl-a-amanitin, S-deoxo-a-amanitin, a-amanitin-(S)-sulfoxide, a-amanitin sulfone 47 or O-Methyl-a-amanitin (S) sulfoxide, O-methyl-a-amanitin sulfone, O-methyl-a-amanitin sulfide 94 or y-amanitin, methyl-y-amanitin, (3-amanitin methyl ester, methyl R-amanitin methyl ester, amanullin, amanuilic acid, amaninamide, amanin, 2,3-diacetyl-6'-O-methyl-y-amantin, 3-acetyl-6'-O-methyl-y-amantin, 2-acetyl-6'-O-methyl-7-amantin, methylaldoamanitin, dethiomethyl-a-amanitin (monocyclic), peptide ring opened amanitin (monocyclic) 95, or S-deoxo[g(R)-hydroxy-Ile]-amaninamide, S-deoxo-Ile-amaninamide 96 or (3-amanitin thiophenyl ester, P-amanitin anilide, R-amanitin dodecylamide, O-methyl(dehydroxymethyl)-a-amanitin, [4'-[[(6-aminohexyl)amino]carbonyl]phenyl]azo-a-amanitin trifluoroacetate, [4'[[[6-[(tert-butyloxycarbonyl)amino]hexyl]amino]carbonyl]-phenyl]azo-a-amanitin, 7'iodo-a-amanitin, 0-(4'-tetrazolyl)phenyl-a-amanitin, O-ethyl-a-amanitin, O-propyl-a-amanitin, O-allyl-a-amanitin, O-(n-hexyl)-a-amanitin, O-(n-decyl)-a-amanitin, O-benzyl-a-amanitin, O-acetonyl-a-amanitin, 0-(2-hydroxyropyl)-a-amanitin, 0-(5-carboxypentyl)-a-amanitin, 0-[5-[[(aninoethyl)-amino]carbonyl]-pent-1-yl]-a-amanitin fluoroacetate, O-[5-[[[(succinoyl-amino)ethyl]-amino]carbonyl]-pent-1-yl]-a-amanitin, 0-[5-[[[[i-[(tert-butyloxycarbonyl) -amino]ethyl]-amino]carbonyl]-pent-1-yl]-a-amanitin, and O,N-dimethyl-a-amanitin 43. Some of these examples are further illustrated in Figure 11.
Examples of other kinds of small-molecule effectors are provided in Example 9.

Example 2 Toxicity Assa,ys While the affinity of amanitin for Pol II is important for its use as a putative novel ATF
effector domain, its toxicity may cause a potential problem in biological assays and in vivo applications. The toxicity of various synthetic derivatives of amanitin has been studied previously. For example, it has been shown that the conversion of carboxyl 1 into methyl ester reduces j3-amanitin toxicity by 50% 43. To address the issue of toxicity, quantitative toxicity assays with AmATFs are performed in human tissue culture cells (as described below in Experimental Methods). Thus far, the lack of significant AmATF toxicity was confirmed while performing numerous transcription assays described in the examples following below.
Specifically, we routinely observed all cells treated with AmATFs by an inverted microscope on a daily basis during the course of each experiment. Such direct observations have not revealed thus far any toxicity of AmATFs within the range of concentrations used in cell culture trascription assays.

Experimental Methods Mammalian or human tissue culture cells (HeLa, BHK21, HEK293 and others) are seeded on 96-well assay plates (Corning) at -20% confluency in 90 l of Dulbecco's Modified Eagle's Medium (DMEM; HyClone) supplemented with 10 /o Fetal Bovine Serum (FBS;
HyClone). The following day, lyophilized AmATF (such as the one depticted in Figure 413) is dissolved in water to make 2.5 micro molar stock solution. These stock solutions are serially diluted 1:5 four times to give five different concentrations. A 10 l aliquot from each tube is added directly to the wells containing cells and medium.
Alternatively, the same aliquot is mixed with 0.2 1 Lipofectamine 2000 (Invitrogen) in 90 l of Opti-mem I reduced serum medium (Invitrogen). After a 20 minute incubation, the original DMEM medium is aspirated and AmATF/Lipofectamine mixture is applied to each well.
After 6-hour incubation, a 50 l aliquot of DMEM supplemented with 30% FBS is added to each well. The plates are subsequently incubated for 2-4 days, following which a CeIlTiter-Glo Luminescent Cell Viability Assay is performed exactly according to manufacturer's instructions (Promega).

Example 3 AmATFs Cell Permeability Assays We have conducted cell permeability studies by incubating labeled AmATFs with tissue culture cells under standard conditions (Experimental Methods). These experiments provide evidence of AmATF cell permeability in human tissue culture cells (Figure 3).
The cell permeability of ATFs is explained by the fact that, even though ATFs are composed of three chemically diverse parts (TFO, PEG linker and peptide), the bulk of ATFs' molecular mass consists of an oligonucleotide (80-90 %). Thus, overall physical characteristics of ATFs, including cell permeability, are likely to be close to those of oligonucfeotides. Numerous previous studies have shown that single-stranded DNA oligonucleotides are able to penetrate human cells in tissue culture and in animal models 81,82,97. Mechanisms of oligonucleotide transport across the cell membrane are still poorly understood and may include endocytosis or active transport triggered by receptors on cell surface 54.98 In addition, amanitin molecule by itself possesses cell-permeability, which is necessary for its natural biochemical function as a toxin that binds intracellular protein Pol II. Therefore, the attached amanitin moiety likely poses no hindrance to TFO entry into the cell (Figure 3).

Experimental Methods:
An aliquot of 100 picomoles of AmATF#2 (SEQ ID N0:1) was labeled with AlexaFluor 488 fluorescent tag using the ULYSIS Nucleic Acid Labeling Kit (Molecular Probes) according to the manufacturer's protocol. The labeled AmATF was purified by the filtration through sephadex spin-column, followed by ethanol precipitation. The labeled AmATF
(SEQ ID N0:1) was dissolved in 0.1 mL of sterile water. Various human and mamrnalian tissue culture cells (HEK293, HeLa, BHK21, A549 and C6) were plated on 96-well plates in 0.1 mL of DMEM
medium supplemented with 10 % fetal bovine serum (HyClone). Aliquots of labeled AmATF
were added directly to the medium surrounding the cells to give the final AmATF concentration of 20 -100 nM. After 2 to 12-hour incubation at 37 degrees C the medium covering cells was replaced with phosphate buffered saline (1 X PBS) and cells were observed with Olympus IX71 fluorescence-capable microscope. Images were captured with RS Photometrix Coolsnap digital color camera and analyzed with RS Photometrix software.

Example 4 AmATFs Activate Chromosomally-Located Reporter Gene Constructs in Tissue Culture Cells We have developed a simple and rapid assay to test whether AmATFs can penetrate the "chromosomal barrier" and regulate genes incorporated into a chromosome of mammalian cells. This assay is based on vectors expressing the firefly luciferase (luc) reporter gene used to make a series of stably transfected cell lines. Such experimental gene target was able to mimic the chromosomal context and location of native (endogenous) genes, while at the same time provided a much simpler and controllable promoter, with the capability of rapid signal detection(Figure 4A). As Figure 4A shows, this target vector was designed to have the luc gene under the control of a minimal promoter, multiple upstream ATF binding sequences, as well as GAL4 sites that bind the control hybrid activator GAL4-VP16 33, 99.
This construct was used to generate several stably-transfected baby hamster kidney 21 (BHK21) cell lines, each carrying the target reporter gene integrated into chromosome.
As an ATF DNA-binding domain we used the 22-mer GT-rich TFO sequence (Figure 413) whose DNA-binding and other properties have already been studied in detail 33.59,100 In order to facilitate the initial molecular design of ATFs, we have adopted the 22-mer GT-rich TFO sequence 57TGTGGTGGGTGGGGTGTGGG73 (SEQ ID NO: 1) whose properties are already well known (Figure 413). For example, this TFO sequence has been studied in detail previously $7,100 These articles have described both the DNA-binding assays and functional assays with the same 22-mer TFO sequence as well as with mutated (control) sequences.
Those experiments proved that the effect of the 22-mer TFO is sequence-specific under a variety of different conditions, including those compatible with the in vitro and tissue culture transcription assays. In addition, very similar GT or GA-rich TFOs have been used by other researchers to target DNA sites both in vifro and in tissue culture assays 54.82. In every such study, it has been shown that TFOs are both sequence- and target-specific.
Other research groups have studied the biochemical properties of TFOs incorporating a variety of covalently-attached moieties. For example, it has been shown that the attachment of various adducts, such as chemical DNA cleavers, steroids, peptides, or even proteins such as nucleases, does not significantly affect the specificity of DNA-binding by TFOs 33,53,101-103 The transcription assays were performed with three different AmATFs designed to bind specifically to the corresponding target sites in the promoter of the reporter gene (Figure 413).
As a specificity control, we have also tested a non-specific AmATF (namely, M2AmATF-myc (SEQ ID NO: 3) shown in Figure 5). The only difference between AmATF#1 (SEQ ID
NO: 1) and AmATF#2 (SEQ ID NO: 1) is the length of the linker domain, having 6 PEG
and 12 PEG
units, respectively (Figure 413). The AmATF#3 is the same structure as AmATF#2 except that the linker and the attached amanitin moiety is located at the 5' end of the TFO (Figure 413).
The non-specific AmATF has the same structure as AmATF#2, except for the TFO
sequence that is not specifically matched to the target sites in the reporter promoter.
Increasing concentrations of these AmATFs were introduced into stably-transfected cells carrying the reporter gene with the aid of carrier Lipofectamine 2000 or Oligofectamine (Invitrogen), and after a 36-58-hour incubation, the expression levels of the reporter gene was quantified with a luciferase assay kit in an automated plate reader (as described in Experimental Methods).
Figure 4C shows that bothAmATF#2 and AmATF#3 were able to activate the reporter gene, largely in a concentration dependent manner. The maximum activation signal was about 3-4 fold compared to the controls. Since AmATF#3 shows consistently highest activation potency at every concentration, it is possible that orientation of the linker (i.e the "po(arity" of the AmATF) plays a role in biological activity in the context of this assay. To confirm that the activation is dependent on the amanitin moiety, at the same time we also performed the parallel control assays with TFO-linkers (TFOs with linkers but with no amanitin attached).
The results shown in Figure 4C reveal that the presence of the amanitin moiety in AmATF
structure is crucial for activation because the TFO-linker#2 and TFO-linker#3 alone did not cause any increase in reporter expression levels (Figure 4C). In fact, at high concentrations both of these TFO-linkers caused a slight repression in reporter gene activity (Figure 4C).
Interestingly, the AmATF#1 also had very little effect in this assay, implyingthat the length of the linker may play an importantrole in AmATF biological activity (data not shown). The non-specific AmATF also did not cause any significant effect on transcription of the reporter gene, thus indicating that the interaction between the AmATFs and the target promoter is sequence-specific (data not shown).
To understand these results better, it is important to provide a broader context by comparing the activities of AmATF#2 and control activator GAL4-VP16, an extremely potent hybrid activator derived from yeast GAL4 DBD and Herpes Simplex virus AD 99.
The application of GAL4-VP16 to the cells carrying the target reporter gene was achieved through transient transfection with constitutively-active expression plasmid (described below). The maximum levels of target activation by GAL4-VP16 were -20-fold (D. Stanojevic, unpublished results). This experiment was performed with at least 30 different BHK21 cell lines, and although the background (i.e. basal) transcription and overall activation signal varied significantly from line to line, we were unable to achieve reproducibly more than -20-fold activation with GAL4-VP16 in this experimentai system under any conditions (data not shown).
This limit on the magnitude of the activation signal is probably associated with chromosomal integration of the target gene (chromosomal "barrier") because transient transfection experiments performed with the same constructs in the same cell line yield much higher activation signals (>100-fold; data not shown). It is important to note that direct comparison between GAL4-VP16 and AmATFs is complicated by the fact that GAL4-VP16 is constantly expressed inside the cell while AmATF is only introduced at the start of the assay. However, despite this potential problem, our results suggest that the biological activity of AmATF
structures described here is within one order of magnitude frorn even the most potent transcriptional activators known thus far (such as hybrid protein GAL4-VP16).

Experimental Methods:
The target plasmid was constructed by transferring the 120 bp DNA fragment containing 5 ATF binding sites from the previously described construct 33 into the luciferase expression vector pFR-luc (Stratagene). The insertion Xba I site was between the basal promoter and GAL4 binding sites already present in the commercial pFR-luc vector. Since the pFR-luc vector does not contain a selection marker, we introduced the neo selection marker on another plasmid (pEGFP, Promega) that was introduced into cells at the same time. To minimize the possibility of insertion of multiple copies of the reporter gene into chromosomes, the linearized reporter plasmid constructs were transfected into BHK21 cells along with the marker plasmid by electroporation as described in104. The selection of cell lines carrying sta bly-i nteg rated vectors was done with Geneticin (invitrogen). AII
selected cell lines were tested for luciferase expression and response to GAL4-VP16 (introduced via transient transfection of plasmid pM3-VP16 (Clontech)), and only those showing the highest signal were chosen for AmATF assays. The application of AmATFs was done exactly the same way as described in the experimental methods of Example 2(above), except that either Lipofectamine 2000 or Oligofectamine (Invitrogen) carriers were used (with very similar results). The plates were incubated for 36-58 hours in the standard C02 incubator (Napco) and the level of reporter expression was quantified with Firefly luciferase assay kit (Biotium) according to manufacturer's instructions in an automated plate reader (Wallac-Victor II, Perkin-Elmer). The results of three assays were analyzed and processed and the data were plotted with the aid of Microsoft Excel software.

Example 5 AmATFs are Able to Target and Activate Endogenous Myc Oncocgene in Human Tissue Culture Cells In order to demonstrate the ability of AmATFs to target and activate the transcription of endogenous genes, we have designed two new AmATF molecules designed to recognize and specifically bind two polypurine sites within the myc promoter, designated as M1 and M2 (Figure 5A). These two sites have been previously targeted with TFOs by other researchers.
For example, Catapano et al. have designed and characterized several TFOs able to bind M1 or M2 site with remarkable specificity and affinity in vitro and in vivo 82.
Similar results have also been obtained by Postel et al. 105. Therefore, we used these same, well-characterized TFO sequences as DNA-binding domains for myc-specific AmATFs. The rest of the AmATF
structure was synthesized exactly as described above, thus yielding the two AmATFs depicted in Figure 5B. Each molecule contains one of the myc promoter-specific TFO
sequences, the polyethylene glycol (PEG) linker and activation domain derived from amanitin (Experimental Methods). At the same time, we also made the control molecules having the TFO
and linker (TFO-linkers), but lacking the amanitin moiety (Experimental Methods).
The two myc-specific AmATFs (hereby termed M1AmATF and M2AmATF) were tested on human HEK293 tissue culture cells in the 24-well plate format. The AmATFs were typically applied to the medium surrounding the cells at 6 nM concentration, with the use of carrier lipofectamine 2000 (Invitrogen). Following the 12-hour incubation, the total RNA was isolated and myc mRNA was quantified by quantitative PCR assay under standard conditions (as described in Experirnental Methods below). The results of these experiments are summarized in Figure 5C. The addition of each of the myc-specific AmATFs (M1AmATF or M2AmATF) results in 20-30% increase of myc mRNA transcription compared to the basal (control) level.
Under the same conditions the application of a 1:1 mixture of two AmATFs results in much stronger (over 100%) increase of myc mRNA transcription. The transcriptional activation is critically dependent on the presence of amanitin because the control molecules lacking the amanitin moiety (i.e., TFO-linkers) are clearly unable to activate myc transcription under any circumstances (Figure 5C). In fact, similarly to myc-specific TFOs described previously by Catapano et al., the data in Figure 5C shows that TFO linkers actually repress the myc transcription, likely due to interference with endogenous transcription factors that bind to the overlapping promoter sites 82. We have also tested under these same conditions control AmATFs containing the non-specific TFOs (TFOs unable to bind the myc promoter) that proved to have no effect on myc transcription, thus proving that the interaction between AmATFs and myc promoter is sequence-specific (data not shown). Another proof of specificity of interaction is built in the assay technology itself; namely, the glyceraidehyde-3-phosphate dehydrogenase (GAPDH) gene serves as an internal control, and the addition of myc-specific AmATFs affects only the myc gene, not GAPDH.
These experiments represent the first successful activation of an endogenous human gene with ATFs. Interestingly, under the described conditions the highest activation signal was obtained with the mixture of two different myc-specific AmATFs, thus implying a synergistic mechanism of action. Therefore, it is very likely that even higher activation potency can be achieved by simple addition of more myc-specific AmATFs into the mixture. Another interesting aspect about these experiments lies in the fact that without the attached amanitin, TFO-linkers cause significant repression of myc transcription (Figure 5C).
Despite this, the full-length AmATFs (TFO-linker-amanitin) are able to overcome this repression and cause a significant overall activation of myc transcription. The mechanism for TFO-mediated repression likely involves the competition between TFOs and endogenous transcription factors for binding to overlapping promoter sites 82.105 For example, the transcription factor PuF has been shown to bind a sequence CCCACCC that overlaps with the M2 site106.
However, this arrangement of overlapping binding sites is likely to be rnore of an exception rather then a rule, and there are many other described examples where the TFO binding to the promoter does not cause repression 33. a','o' Therefore, the use of such "non-overlapping" sites will lead to a more efficient activation and, consequently, a higher overall signal.

Experimental Methods:
Myc transcriptional assays: Hek293 cells were seeded in 24-well assay plates (Corning) at -40% confluency in 500 l of Dulbecco's Modified Eagle's Medium (DMEM;
HyClone) supplemented with 10% Fetal Bovine Serum (FBS; HyClone). The following day, AmATFs and TFO-linkers were dissolved in sterile water to make 4 M stock solution of each.
An aliquot from each tube was mixed with 1 l Lipofectamine 2000 (Invitrogen) in 100 L of Opti-mem I reduced serum medium (Invitrogen) according to the manufacturer's instructions.
After 20 minutes of incubation, the 100 L of the mixture was applied to each well to give the final concentration of 6 nM for each AmATF or TFO-linker. The control wells were also treated with the same procedure, except that the Lipofectamine 2000 was mixed with pure sterile water instead of the AmATF stock solution. This insured that all wells containing HEK 293 cells (including the controls) were treated with the same concentration of transfection agent Lipofectamine 2000. After a 12-hour incubation in a tissue-culture incubator, the medium was aspirated and 0.5 ml of Trizole (Invitrogen) was added to each well. RNA was subsequently isolated according to the manufacturer's instructrions, and dissolved in 0.1 rnl of sterile water.
The concentration of each RNA stock was determined by UV spectroscopy (Hewlett-Packard 8452A). An aliquot containing 2 g RNA was taken from each tube and converted into cDNA
by using the ThermoScript RT-PCR System according to manufacturers protocol (Invitrogen).
A 2 L aliquot from each cDNA prep was mixed with myc specific PCR primers (5'TCGGAAGGACTATCCTGCTG3' SEQ ID NO: 16 and 5'GCTTTTGCTCCTCTGCTTGG3' SEQ ID NO: 17, 5 pmols each) or the PCR primers specific for a housekeeping gene GADPH
(5'GGGTGTGGGCAAGGTCATCC3' SEQ ID NO 18 and 5'TCCACCACCCTGTTGCTGTA3' SEQ ID NO: 19, 5 pmols each) and reagents from iQ SYBRGreen Supermix kit according to manufacturers instruction (Bio-Rad). These same PCR primers were previously described in Catapano et al. (ibid). The total reaction volume was 25 L. The reaction mixtures were put into the standard 96-well transparent plate, and each sample was analyzed in duplicate. The quantitative PCR reaction was performed in Bio-Rad iCycler iQ real-time (quantitative) PCR
machine, using the 3-step cycle (95-60-72 degrees, 30 seconds each, 40 repeats). The initial background calibration was performed automatically using the fluorescein contained within the Bio-Rad Supermix kit. The melting curve analysis on the Bio-Rad machine as well as analysis of the quantitative PCR reaction products on the agarose gels revealed the presence of only one PCR product that corresponded exactly to the predicted sizes of the amplicons for both myc and GADPH gene. The threshold was calculated automatically by the Bio-Rad qPCR
software. The resulting Ct values for both myc and GAPDH mRNAs were imported into the Bio-Rad Genex-1 computer program for qPCR data analysis and presentation, and the resulting data was further analyzed with Excel software (Microsoft).

Example 6 AmATF-mediated Activation of Endogenous Human p53 tumor Suppressor Gene Tumor suppresor gene p53 was discovered in 1979 as a 53 kD protein associated with transforming protein (Large T antigen) from Simian Virus 40108. The elucidation of a physiological role of p53 began a decade later with the discovery that many types of cancer contain mutant (defective) p53 protein'o9,"o Since then it has been established that a wild-type p53 is a tetrameric transcription factor acting as a"guardian of the genome" because it stops cell division in response to DNA damage or other kinds of cellular stress such as oncogene activation or hypoxia "' 13. In addition, p53 stimulates the DNA
repair machinery and destroys damaged cells through induction of apoptosis 74-116 A majority of human tumors completely or partially lack the functionality of a wild-type p53. Therefore, the p53 loss-of-function is the most common genetic deficiency in human cancer identified thus far "',"s This phenotype results from a variety of mutations such as:
(i) deletions of one or both p53 alleles;
(ii) nonsense or splice site mutations resulting in the production of a truncated p53 protein without the oligomerization domain;
(iii) dominant-negative missense mutations altering the p53 protein structure, most often affecting the DNA-binding domain;
(iv) mutations in other cellular genes such as HOXA5 or MDM2 that interact with and regulate p53 at the transcriptional or post-transcriptional level 19120. In addition, the normal p53 function can be disrupted through interaction with viral genes such as E6 gene of human papillomavirus121 .
Human p53 has generated a substantial interest as a promising target for the development of cancer therapies. The introduction of a wild-type p53 into a wide range of cancer cells has been shown to inhibit tumor growth and/or tumorigenicity 27.122. At the same time, the overexpression of wild-type p53 in normal, non-malignant cells is likely to have minimal harmful effects122-'2 . Therefore, the ability to restore wild-type p53 function is seen as having a huge potential for treating many, if not most types of cancer "Z.
Several different approaches to this problem have been reported thus far. For example, a wild-type p53 gene can be introduced exogenously into cancer cells expressing mutant p53 through various forms of gene therapy125. Another approach employs small molecules, ribozymes or peptides for restoring the function of defective p53 protein 21,121-128 Also, small molecules have been used to restore normal intracellular levels of p53 protein by disrupting the interaction with inhibitory viral or cell proteins12s"'3'. However, these approaches have not yet produced a viable cancer therapy.
An AmATF-mediated activation of endogenous p53 gene at the level of transcription can form the basis for an entirely novel approach to cancer treatment. This method is especially suitable for the types of cancer that contain the wild-type p53 that is inactivated through promoter mutation, chromosomal translocation, or interaction with other proteins. For example, it has been discovered that 60-70% of human breast cancers have abnormally low levels of p53 protein due to compromised function of HOXA5 gene "s The product of HOXA5 gene acts like a transcriptional activator of p53, and thus the HOXA5 loss of function leads to abnormally low levels of p53 transcription, and consequently to malignant transformation "s.
Another such example involves some types of brain cancer (astrocytomas), where the p53 loss of function is likely to be mediated via transcriptional repression by overabundant protein PAx 132 In this example we provide experimental evidence for activation of endogenous hurnan p53 tumor suppressor gene by AmATFs. The proximal promoter of human p53 contains several extended polypurine sites suitable for targeting via triplex formation. Two such sites (labeled P1 and P2) are shown in Figure 6A. We designed two TFOs to bind specifically to the P1 or P2 site (Figure 6A). Starting with these two TFOs, the rest of the AmATF structure was synthesized exactly as described in the previous sections, thus yielding the two p53-specific AmATFs depicted in Figure 6B. Each AmATF molecule contains one of the p53 promote r-s pecific TFO sequence, the polyethylene glycol (PEG) linker and activation domain derived from amanitin (Figure 66; Experimental Methods). At the same time we also made two kinds of control molecules: (i) the structural controls lacking the amanitin moiety (i.e. TFO-linkers) and (ii) the non-specific control AmATF having a TFO sequence of similar size and composition as the previous two, but unable to match and bind either P1 or P2 site (Experimental Methods).
We tested the two p53-specific AmATFs (hereby termed P1AmATF and P2AmATF) for the ability to activate endogenous p53 gene in human HEK293 tissue culture cells in the 24-well plate format. The AmATFs were typically applied to the medium surrounding the cells at a range of concentrations, with or without the use of carrier lipofectamine (Experimental Methods). Following the 24-48 hour incubation, total RNA was isolated and the arnount of p53 mRNA was measured by quantitative PCR (qPCR) assay under standard conditions (Experimental Methods). The glyceraidehyde-3-phosphate dehydrogenase (GAPDH) housekeeping gene served as an endogenous control (Experimental Methods). The results of these experiments are summarized in Figures 6C and 6D. For example, Figure 6C
shows that the addition of 20 nM P2AmATF results in a 3 to 4-fold increase in p53 transcription.
Furthermore, the effect of AmATFs on p53 transcription is largely con centratio n-d ependent (Figure 6C). Interestingly, similar levels of p53 activation by AmATFs were achieved even in the absence of carrier lipofectamine (Figure 6D). The transcriptional activation by AmATFs is critically dependent on the presence of amanitin within the AmATF structure because the control molecules lacking the amanitin moiety (i.e. TFO-linkers) are clearly unable to activate p53 transcription (Figure 6D). The non-specific control AmATF had no significant effect on p53 transcription under these experimental conditions, thus indicating that the interaction between AmATFs and p53 promoter is sequence-specific (data not shown;
Experimental Methods). To further exclude any possibility of non-specific effects of amanitin moiety on p53 transcription, a series of experiments was performed whereby the level of p53 mRNA was observed in response to increasing concentrations of R-amanitin. We concluded that the 0-amanitin by itself has no significant effect on p53 transcription in HEK293 human cells, even at highly toxic concentrations of over 100nM (data not shown).

Experimental Methods:
-Synthesis of p53-specific AmATFs: The first step involved the synthesis of modified oligonucleotides (SEQ ID NO: 8) XXY3' and (SEQ ID NO: 7)XXY3 where X
represents the Spacer Phosphoramidite 18 (Glenn Research) and Y represents the amino modifier residue bearing the primary amine on a short tether (Amino-Modifier C6, Glenn Research). Each Spacer Phosphoroamidite residue incorporates an 18-atom long polyglycol linker at the 3' end of the TFO sequence. These molecules were used as TFO-linker controls in the p53 targeting assays (Figure 6). AII AmATFs were synthesized exactly as described in previous chapters.
The specificity control molecules contained the modified oligonucletide (SEQ
ID NO: 3)XXY3' of similar size and composition as the ones described above, except that the TFO sequence was not designed to bind specifically to the p53 promoter.
-P53 transcriptional assays: Human embryonic kidney (HEK293) cells were seeded on 24-well assay plates (Corning) at -80 % confluency in 500 L of Dulbecco's Modified Eagle's Medium (DMEM; HyClone) supplemented with 10 % Fetal Bovine Serum (FBS;
HyClone).
The following day, AmATFs and TFO-linkers were dissolved in sterile water to make 5 M
stock solutions of each. For experiments performed with no carriers, an aliquot from each AmATF or TFO-linker stock solution was added directly to the DMEM medium covering the cells to achieve the final concentration of 20 nM. Pure sterile water was added to wells containing control cells. For the experiments performed with the carrier, aliquots of AmATF or TFO-linker stock solutions were mixed with 1 L Lipofectamine 2000 (Invitrogen) in 100 L of Opti-mem I reduced serum medium (Invitrogen) according to the manufacturer's instructions.
After 20 minute incubation, 100 L of mixtures were applied to each well. The control wells were also treated in the same manner, except that the Lipofectamine 2000 was mixed with pure sterile water. This insured that all wells containing HEK 293 cells (including the controls) were treated with the same concentration of transfection agent Lipofectamine 2000. After 24-48 hours of incubation the medium was aspirated and 0.2 mL of Trizole (Invitrogen) was added promptly to each well. RNA was subsequently isolated according to the manufacturer's instructions, and dissolved in 0.1 mL of sterile water. The concentration of each RNA stock was determined by UV sprectroscopy (Hewlett-Packard 8452A), and the quality of RNA preps was confirmed by agarose gel electrophoresis. An aliquot containing 2 g RNA
from each prep was converted into cDNA with ThermoScript RT-PCR System according to manufacturer's protocol (Invitrogen). The TAQman reagents, probes and primers for qPCR
were obtained from Applied Biosystems (ABI assays: Hs00153349_m1 for p53 and Hs99999905_m1 or 4326317E for GAPDH). We verified experimentally that each assay was able to distinguish cDNA from genomic DNA. The qPCR reactions were done in quadruplicate (Figure 5C) or duplicate (Figure 5D) in an ABI 7500 real-time PCR machine using standard procedures and protocols. The resulting data was initially analyzed with ABI
7500 system software, and was subsequently imported into Excel program for further analysis and presentation. Each experiment was reproduced at least twice under slightly different experimental conditions (incubation times, cell density, etc.) with similar results.

Example 7 Targeting y-Globin Gene With TFO and PNA-based ATFs.
Sickle cell disease and R-thalassemia are among the most common inherited human genetic diseases and are very well characterized at the molecular level. They are caused by defective hemoglobin, a tetramer protein complex formed from two a- and two P-globin chains. Sickle cell disease is caused by an A--4T point mutation in the sixth codon of the R-globin gene that replaces glutamic acid with valine. This causes the accumulation of polymerized hemoglobin molecules, thus leading to the alteration of red blood cell shape (i.e.
sickling) and vascular occlusion 41.133. In (3-thalassemia, deletions within the (3-globin gene lead to the imbalance in a and R-chain synthesis. The resulting accumulation of unpaired a-globin protein results in red blood cell damage, ineffective erythropoiesis and anemia133 Beside these two globin genes, human genome also contains y-globin gene that forms part of fetal hemoglobin, and is active only during embryonic development. In adult cells ,v-globin gene is virtually silent, expressed at very low levels (< 1%). A group of mutations called hereditary persistence of fetal hemoglobin (HPFH) causes the presence of y-globin protein in adulthood. Interestingly, this elevated expression of y-globin ameliorates clinical symptoms of both sickle cell disease and R-thalassemia. In sickle cell anemia the presence of y-globin protein inhibits the polymerization of hemoglobin, while in R-thalassemia the precipitation of unpaired a-globin is preciuded133.'3a The elevated level of y-globin in adulthood can be artificially induced by compounds such as 5-azacytidine, hydroxyurea and short chain fatty acids (butyrate and its analogs).
However, despite their initial promise, these drugs generally have low efficacy and specificity, show significant toxicity and rnay even be carcinogenic135 For that reason there is an urgent need to discover novel, more specific pharmacological agents able to induce y-globin without the potentially harmful side effects.
This example describes the design, synthesis and applications of the y-globin-specific AmATFs. The targeting of the human y-globin promoter with TFOs and PNA has already been described in the literature a','o' For example, one of the TFOs depicted in Figure 7(second from the top, in italics) has been shown to bind specifically the y-globin promoter in vitro, with a binding constant Kd= 5X10"707.The principles and methods for the design of TFO-based AmATFs that target and regulate transcription of y-globin are similar to those described in previous examples. For the TFO-based AmATFs, as a first step it is necessary to design TFOs that are specifically matched to the hompurine / homopyrimidine stretches in the y-globin promoter (as shown in Figure 7A). For this purpose both GA-rich and GT-rich TFOs can be used, and Figure 7A illustrates both examples. Also, it is possible to synthesize the two adjacent shorter TFOs as a single long molecule where the gap is "bridged" by incorporation of simple PEG linker by automated synthesis (Figure 7A, top line; Experimental Methods).
The second step involves the synthesis of one or several AmATF that incorporate the y-globin-specific TFO as a DNA-binding domain. For example, one such AmATF structure is depicted in Figure 7C. It contains the DNA binding domain composed of the TFO sequence depicted in Figure 7A (SEQ ID: 12) the linker composed of 9 polyethylene glycol (PEG) units attached to the 5' end of the TFO, and 5' terminal residue comprising a primary amine.
This amine is used for covalent coupling to the activated R-amanitin, as described in previous examples.
Following the same procedures different y-globin-specific AmATFs can be synthesized based on different TFOs depicted in Figure 7A, as well as non-specific control AmATFs that contain unrelated TFO sequences and serve as specificity controls in cell culture or in vivo experiments (Experimental Methods).
An example of targeting and regulation of the same promoter with peptide nucleic acid (PNA) - based AmATF is described in detail in Figure 7B. The PNA clamp design involves two shorter 11-mer strands linked by a PEG or other kind of flexible and inert spacer. This PNA
clamp binds to the specific site in the myc promoter via D-loop formation 41.
The linker terminating with the primary amine residue is added at the C-terminus, and the whole PNA-linker conjugate is synthesized as a single strand via automated synthesis, with the following chemical modifications (i) the N-terminus is converted into an amide to prevent possible coupling of amanitin moiety at the N-terminus and (ii) the second PNA stretch is synthesized using a pseudoisocytosine (labeled as J) instead of cytosine. This modification has been shown to stabilize the PNA binding to the DNA under physiological conditions 41. After the automated synthesis is completed, the amanitin moiety is attached to the linker using the same procedure as described in Example 1.
The application of the all y-globin-specific AmATFs to human cells is accomplished as described in previous examples. The signal detection is perFormed by qPCR
using y-globin specific TAQman probes and primers (Applied Biosystems) and standard techniques described in previous examples.

Example 8 Targeting the T-bet Gene With AmATFs The portion of the T-bet promoter sequence with putative AmATF binding site is shown, along with T-bet specific TFO- and PNA-based AmATFs (Figures 8A and 8B, respectively). The design rules, synthesis and assay procedures are the same as described in previous exarnples.

Example 9 Small-molecule analogs of AmATFs Besides amanitin, there are other natural or artificial small molecules that bind Pol II or other proteins involved in transcription13s'3' Since the basic three-part ATF
structure described here (i.e. the DNA-binding domain, the linker and the effector domain) can accommodate almost any chemical moiety as an effector domain, this invention can be extended easily to include other such small molecules. Even the lack of water solubility would not pose a problem because the ATF synthesis can be performed in organic solvents, and the full-length ATF would retain water solubility because of the predominance (in terms of the overall mass) of the DNA-binding domain and the linker, which are both based on water-soluble compounds such as modified DNA, PEG, etc.
One such example involves Flavopiridol, a compound that has been tested in clinical trials for cancer therapy136 Flavopiridol has been shown to block human immunodeficiency virus Tat transactivation and replication by inhibiting a positive transcription elongation factor b (P-TEFb). Flavopiridol binds human P-TEFb protein very tightly, even in the presence of high salt. The presence of P-TEFb protein is required for Pol II mediated transcription in vivo, and the recruitment of the P-TEFb to the promoter of a targeted gene by Flavopiridol-based ATFs is very likely to stimulate the transcription of the target gene.

Another such small-molecule compound involves nucleoside analogue 5,6-dichloro-(3-D-ribofuranosylbenzimidazole (DRB) which acts through a very similar mechanism as Flavopiridol, but has a lower binding constant to P-TEFb protein138. Also, tagetitoxin (commercial name tagetinT"'; Epicentre Technologies) is a bacterial small-molecule toxic compound (M. W. = 416) that complements the activity of a-amanitin as a potent and selective inhibitor of all eukaryotic RNA polymerases139. In addition, it is possible to set up a simple screening assay to identify other small-molecule compounds that bind to Pol II
or other proteins involved in transcription via standard techniques. For example, a yeast or other simple eukaryotic cell culture is grown in the presence of a library of synthetic or natural chemical compounds, and an initial screen is done using the growth inhibition (measured by OD) as the endpoint. After the selected compound with the highest inhibition constant is identified, it is necessary to generate resistant mutants using the standard genetic methods79.
The analysis of such mutants identifies the target protein for candidate compound. For example, if the mutation conferring the resistance against the selected compound lies in one of the subunits of Pol II, the mechanism of action (inhibition of Pol II) is therefore confirmed.
The selected molecule is subsequently characterized in in vitro Pol II binding assays to determine the binding constants under various conditions. One such compound (UK-118005, having M. W. 214.3) was isolated recently via such a screen as the first example of a synthetic small molecule inhibiting Pol II79. Such newly identified small-molecule compounds that exhibit high binding constants can be incorporated into the ATF structure and tested in biological assays for cell-permeability, toxicity and the ability to activate target genes as described in previous examples.

Example 10 General Method for ldentification of ATF Binding Sites in the Human Genome In general, stable and specific triple helical complexes can be formed primarily with DNA sequences that possess special features such as stretches of purine or pyrimidine bases 54 Such polypurine/polypyrimidine sites are very frequently found in natural eukaryotic promoters such as myc, p53, y-globin, T-bet, bcl-2, and many others 55.10e,140-142 To estimate the total number of gene targets available for manipulation via ATF
technology, it is necessary to estimate the genome-wide distribution of TFO binding sites in promoters.
In order to obtain a rough estimate of the fraction of the human genome that can potentially be targeted by TFOs, we performed a search for polypurine and polypyrimidine sites in the database containing DNA sequences from 14,000 human promoters (Figure 9A).

This group of 14,000 promoters encompasses genes that have been mapped or described in previously-published literature and public databases. Our initial search was focused exclusively on the 750 bp region of each promoter immediately upstream from the transcription initiation sites (+1 to -750). The potential binding site was defined as an uninterrupted stretch containing at least 10 polypurines (or polypyrimidines).
This size of 10 bp was chosen because that is the length of the shortest TFO reported thus far that is able to successfully bind endogenous genes in tissue culture assays'43. The results summarized in Figure 9 show that about 90% of promoters contain at least one potential ATF
binding site and that a great majority of promoters contain multiple sites. However, it is important to consider the following:
(i) Extension of search for TFO sites beyond the 750 bp region will inevitably increase the fraction of genome that can be targeted to more than 90%.
(ii) An increasing body of evidence suggests that the basic polypurine /
polypyrimidine TFO recognition scheme can be extended with simple modifications. For example, it is possible to target a polypurine stretch interrupted by one or even several pyrimidine residues with TFOs having T residues opposite each pyrimidine interruption (as illustrated in Figure 5A, for example). Also, a single TFO can be used to target polypurine stretches on alternate strands144. Finally, the triplex recognition scheme can be extended by synthesizing TFOs with non-natural bases and nucleotide analogues 5$'45. If necessary, all of these techniques and many others are available to extend the range of target sequences and increase both the number and average size of potential ATF binding sites146 In conclusion, the data presented here strongly implies that ATF technology has the potential to become the basis for a comprehensive gene targeting strategy, applicable to a vast majority of eukaryotic genes.

Experimental Methods:
The process of identifying the ATF binding site throughout the human genome involved several steps. First, using publicly available genomic databases we compiled a list of all genes whose promoter regions have been precisely delimited (14,000 promoters is total).
Second, we chose 750 bp upstream from the transcription initiation site as an arbitrary limit for searching for ATF binding sites. Third, we instructed the database search program to compile a list of all polypurine and polypyrimidine stretches longer than 10 bp for every gene on each human chromsome. For example, a small portion of such a list compiled from human chromosome 10 is provided in Figure 9B. The list shows the name of the gene, unique genomic database identifier, and one or more ATF binding sites for each gene.
As another example, a full list of all ATF binding sites identified in this manner on human chromosome 12 is also included (Figure 9C).

Example 11 Animal Models The present invention can be tested in animal models (including human patients) for the purpose of developing new treatments for many different diseases. Two non-limiting examples of such models are described below.

Sickle Cell Disease Sickle cell disease and (3-thalassemia are among the most common inherited human genetic diseases and are very well characterized at the molecular level. They are caused by defective hemoglobin, a tetramer protein complex formed from two a- and two a-globin chains.
Sickle cell disease is caused by an A->T point mutation in the sixth codon of the (3-globin gene that replaces glutamic acid with valine. This causes the accumulation of polymerized hemoglobin molecules, thus leading to the alteration of red blood cell shape (i.e., sickling) and vascular occlusion 4','33 In (3-thalassemia, deletions within the (3-globin gene lead to the imbalance in a- and (i-chain synthesis. The resulting accumulation of unpaired a-globin protein results in red blood cell damage, ineffective erythropoiesis, and anemia133 Beside these two globin genes, hurnan genome also contains y-globin gene that forms the part of fetal hemoglobin, and is active only during embryonic development. In adult cells y-globin gene is virtually silent, expressed at very low levels (<1 %). A group of mutations, called hereditary persistence of fetal hemoglobin (HPFH), causes the presence of y-globin protein in adulthood.
Interestingly, this elevated expression of y-globin ameliorates clinical symptoms of both sickle cell disease and R-thalassemia. In sickle cell anemia the presence of y-globin protein inhibits the polymerization of hemoglobin, while in P-thalasemia the precipitation of unpaired a-globin is precluded 133,134 The elevated levels of y-globin in adulthood can be artificially induced by compounds such as 5-azacytidine, hydroxyurea, and short chain fatty acids (butyrate and its analogs).
However, despite their initial promise, these drugs generally have low efFicacy and specificity, show significant toxicity, and may even be carcinogenic135 For that reason, there is an urgent need to discover novel, more specific pharmacological agents (such as v,-globin specific AmATFs described in Example 7) able to induce y-globin without the potentially harmful side effects.
The y-globin specific AmATFs are tested in a SCD mouse, an animal model developed for sickle cell disease147. This SCD mouse is a transgenic organism created to express only human hemoglobin in adult red blood cells. Similar to many human patients with sickle cell disease, the SCD mice have severe hemolytic anemia and show prominent organ pathology, with many sickled erythrocytes present in peripheral blood. Despite these symptoms that mimic those in afflicted human patients, most animals survived for 2 to 9 months and were fertile148. For these reasons, the SCD mouse is a very suitable model animal for developing the AmATF-based therapy for sickle cell disease. The y-globin-specific AmATFs are designed and synthesized as described in Example 1 and Example 7, and administered to SCD mice by intravenous injection. The aim of this approach is to induce the y-globin expression in bone marrow hematopoietic stem cells (HSCs), thus increasing the amount of "healthy" hemoglobin and alleviating the disease symptoms. The administration of AmATFs can be accomplished with the aid of pharmaceutical compositions such as those described in the section "Pharmaceutical Compositions" above, and many others.
The AmATFs that target different sites within the y-globin promoter can be administered as a mixture to maximize their overall effect, or individually, for comparative studies. The AmATFs can also be administered multiple times over a period of time.
The effects of the AmATF-based therapy can be studied by comparing the treated and untreated (control) animals by using the standard experimental methods. For example, this can be accomplished by analyzing the amount of globin protein directly, by single-cell globin mRNA level analysis, and by measuring the hematologic indices and histopathology as described in147. The induction of y-globin expression is correlated to hematologic indices and with the physiological effects, and compared to control animals as well as healthy (wild-type) animals to reveal the extent of the AmATF-based therapy. Since most of these experimental procedures involve studying of blood via non-invasive methods, they can be applied to human clinical trials as well.

Cancer The turnor suppressor gene p53 is inactivated in almost all forms of human cancer 1z The insufficient p53 expression or the presence of defective (mutated) p53 protein disable an emergency brake on cell proliferation and lead to genetic instability. The p53 protein acts like a"guardian angel" of the genome because it stops cell division in response to DNA damage to allow the DNA to be repaired. In addition, p53 stimulates the DNA repair machinery and, in some cases causes the destruction of damaged cells. Therefore, the ability to restore the function of the p53 gene is crucial for the prevention and treatment of many, if not most, types of cancer. Moreover, the recent findings indicate that p53 is hapio-insufficient for tumor suppression. In other words, even a-50% reduction of p53 expression due to inactivation of only one of the two p53 alleles is sufficient to promote cancer formation 'as Consequently, many tumors contain a p53+/- genotype, with one mutated (inactivated) and one wild-type (active) copy of the gene. This raises the possibility of turnor suppression via the activation of the single copy of wild-type endogenous p53 gene. In addition, this approach would be particularly convenient for treatment of those genetic defects resulting from inactivation of the p53 gene or other tumor suppressor genes due to the chromosomal translocation or mutations affecting the promoter sequences, because these mutations alter the amount of p53 and not the structure of p53 protein. It is expected that only a small effect (-2 fold activation) is sufFicient to restore a normal intracellular amount of p53 and, at the same time, avoid or minimize toxic effect due to the potential increase of p53 in normal tissues.
Mice and other animal models having heterozygous for p53 (p53+/- genotype) have been used extensively in cancer research because such animals are highly cancer prone15o In p53+/- mice tumors in wide variety of tissues appear spontaneously or upon treatment with carcinogenic chemicals such as 2-acetylaminofluorene (2-AAF), benzo(a)pyrene, and many others 150 For this reason, the p53+/- mouse models are used to study the ability of p53-specific AmATFs to inhibit tumor growth.
The AmATFs that specifically target p53 promoter will be synthesized as described in Figure 6 and Examples 1 and 6. The tumors are induced by treating the p53+/-animals with 300 ppm 2-AAF for 39 weeks in the diet, followed by normal diet for two weeks150. Since the tumors in these animal models is widespread among many tissues including skin and internal organs, the administration of AmATFs can be accomplished by any method known in the art, including, but not limited to, topical applications, intravenous injection, subcutaneous injection, oral delivery and inhalation through the lungs. The application of AmATF is done over increasing periods, from two weeks to several months. The effects of the AmATF
treatment are studied by comparing the frequency of tumor formation and the average size of tumors between treated and untreated (control) animals. The data will be subjected to standard statistical analysis as described in15o Example 12 ATF-based Ex Vivo Therapy with Hematopoietic Cells Ex vivo therapy represents one of the emerging innovative therapies based on the administration of cells which have been treated or modified outside of the body. AmATF-based ex vivo therapy in essence involves the application of AmATFs to patient's own cells in culture, followed by the transfer back to the patient to treat grave medical conditions like cancer or genetic diseases.
As described in the previous Examples 7 and 11, the AmATF-based activation of y-globin is a very promising new approach for the treatrnent of sickle cell anemia. One strategy for treatment of patients suffering from this disease involves AmATF-based ex vivo therapy.
For example, hematopoietic stem cells or other hematopoietic cells are collected from the patient's peripheral blood or bone marrow and propagated in cell culture as described previously15'. Such hematopoietic cell cultures are then treated with y-globin specific AmATFs (such as those described in Example 7) according to the protocols and methods described in the previous Examples. Subsequently the AmATF-treated cells are transferred back to the patient. This ex-vivo method, relying on the supply of patient's own cells is therefore likely to avoid potential problems due to the tissue incompatibility as well as the difficulty of finding the right donor, as is often the case with methods involving the transplantation of donor hematopoietic cells.
Another example of ex vivo therapy involves the TGF-R cytokine, the expression of which is necessary to convert precursor lymphocyte cells into T-regulator cells. It has been shown that elevated expression of TGF-(3 has beneficial effect against autoimmune attack.
Therefore, the ATF-mediated activation of human TGF-0 in hematopoietic T-cells ex-vivo can be used to alleviate the symptorns of autoimmune diseases such as lupus or type I diabetes.

Those skilled in the art will recognize, or be able to ascertain, using no more than routine experimentation, numerous equivalents to the specific substances and procedures described herein. Such equivalents are considered to be within the scope of this invention and are covered by the following claims.

REFERENCES CITED

1. Ptashne M. A genetic switch: phage [lambda] and higher organisms (ed 2nd).
Cambridge, Mass.: Cell Press: Blackwell Scientific Publications; 1992.
2. Ptashne M. How eukaryotic transcriptional activators work. Nature.
1988;335:683-689.
3. Hahn S. Structure and mechanism of the RNA polymerase II transcription machinery.
Nat Struct Mol Biol. 2004;11:394-403.
4. Asturias FJ. RNA polymerase II structure, and organization of the preinitiation complex. Curr Opin Struct Biol. 2004;14:121-129.
5. Koleske AJ, Young RA. The RNA polymerase II holoenzyme and its implications for gene regulation. Trends Biochem Sci. 1995;20:113-116.
6. Hampsey M, Reinberg D. RNA polymerase II as a control panel for multiple coactivator complexes. Curr Opin Genet Dev. 1999;9:132-139.
7. Brent R, Ptashne M. A eukaryotic transcriptional activator bearing the DNA
specificity of a prokaryotic repressor. Cell. 1985;43:729-736.
8. Majello B, De Luca P, Lania L. Sp3 is a bifunetional transcription regulator with rnodular independent activation and repression domains. J Biol Chem.
1997;272:4021-4026.
9. Lee H, Mok KH, Muhandiram R, et al. Local structural elements in the mostly unstructured transcriptional activation domain of human p53 [In Process Citation]. J Biol Chem. 2000;275:29426-29432.
10. Ma J, Ptashne M. A new class of yeast transcriptional activators. Cell.
1987;51:113-119.

11. Blau J, Xiao H, McCracken S, O'Hare P, Greenblatt J, Bentley D. Three functional classes of transcriptional activation domain. Mol Cell Biol. 1996;16:2044-2055.

12. Ptashne M, Gann A. Transcriptional activation by recruitment. Nature.
1997;386:569-577.

13. Wolffe AP, Wong J, Pruss D. Activators and repressors: making use of chromatin to regulate transcription. Genes Cells. 1997;2:291-302.

14. Hall DB, Struhl K. The VP 16 activation domain interacts with multiple transeriptional components as determined by protein-protein cross-linking in vivo. J Biol Chem.
2002; 277:46043 -4605 0.

15. Workman JL, Kingston RE. Alteration of nucleosome structure as a meehanism of transcriptional regulation. Annu Rev Biochein. 1998;67:545-579.

16. Carmichael GG. Antisense starts making more sense. Nat Biotechnol.
2003;21:371-372.

17. Brantl S. Antisense-RNA regulation and RNA interference. Biochim Biophys Acta.
2002;1575:15-25.

18. Branch AD, Robertson HD. Efficient trans cleavage and a common structural motif for the ribozymes of the human hepatitis delta agent. Proc Natl Acad Sci U S A.
1991;88:10163-10167.

19. Breaker RR, Joyce GF. A DNA enzyme that cleaves RNA. Chem Biol. 1994;1:223-229.

20. Hannon GJ. RNA interference. Nature. 2002;418:244-251.

21. Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T.
Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature.
2001;411:494-498.

22. Janowski BA, Kaihatsu K, Huffinan KE, et al. Inhibiting transcription of chromosomal DNA with antigene peptide nucleic acids. Nat Chem Biol. 2005;Advanced Online Publication.

23. Janowski BA, I-Iuffrnan KE, Schwartz JC, et al. Inhibiting gene expression at transcription start sites in chromosomal DNA with antigene RNAs. Nat Chem Biol.
2005;Advanced Online Publication.

24. Stamatoyannopoulos G, Neinhuis A, Majerus X, Varmus X. The Molecular Basis of Blood Diseases. Philadelphia, PA.: W. B. Sauders Company; 1994.

25. Blagosklonny MV. P53: an ubiquitous target of anticancer drugs. Int J
Cancer.
2002;98:161-166.

26. Foster BA, Coffey HA, Morin MJ, Rastinejad F. Pharmacological rescue of mutant p53 conformation and function. Science. 1999;286:2507-2510.

27. Ventura A, Kirsch DG, McLaughlin ME, et al. Restoration of p53 function leads to tumour regression in vivo. Nature. 2007;445:661-665.

28. Beerli RR, Segal DJ, Dreier B, Barbas CF, 3rd. Toward controlling gene expression at will: specific regulation of the erbB-2/HER-2 promoter by using polydactyl zinc finger proteins constructed from modular building blocks. Proc Natl Acad Sci U S A.
1998;95:14628-14633.

29. Beerli RR, Dreier B, Barbas CF, 3rd. Positive and negative regulation of endogenous genes by designed transcription factors. Proc Natl Acad Sci U S A.
2000;97:1495-1500.

30. Kuwabara T, Hsieh J, Nakashima K, Taira K, Gage FH. A small modulatory dsRNA
specifies the fate of adult neural stem cells. Cell. 2004;116:779-793.

31. Cho-Chung YS, Park YG, Lee YN. Oligonucleotides as transcription factor decoys.
Curr Opin Mol Ther. 1999;1:386-392.

32. Nakamura H, Morishita R, Kaneda Y. Molecular therapy via transcriptional regulation with double-stranded oligodeoxynucleotides as decoys. In Vivo. 2002;16:45-48.

33. Stanojevic D, Young RA. A highly potent artificial transcription factor.
Biochemistry.
2002;41:7209-7216.

34. Kuznetsova S, Ait-Si-Ali S, Nagibneva I, et al. Gene activation by triplex-forming oligonucleotide coupled to the activating domain of protein VP 16. Nucleic Acids Res.
1999;27:3995-4000.

35. Mapp AK, Ansari AZ, Ptashne M, Dervan PB. Activation of gene expression by small molecule transcription factors. Proc Natl Acad Sci U S A. 2000;97:3930-3935.

36. Liu B, Han Y, Ferdous A, Corey DR, Kodadek T. Transcription activation by a PNA-peptide chimera in a mammalian cell extract. Chem Biol. 2003;10:909-916.

37. Kwon Y, Arndt HD, Mao Q, et al. Small molecule transcription factor mimic.
J Am Chem Soc. 2004;126:15940-15941.

38. Latimer LJ, Hampel K, Lee JS. Synthetic repeating sequence DNAs containing phosphorothioates: nuclease sensitivity and triplex formation. Nucleic Acids Res.
1989;17:1549-1561.

39. Gryaznov SM, Lloyd DH, Chen JK, et al. Oligonucleotide N3'-->P5' phosphoramidates.
Proc Natl Acad Sci U S A. 1995;92:5798-5802.

40. Nielsen PE, Egholm M, Berg RH, Buchardt O. Peptide nucleic acids (PNAs):
potential antisense and anti-gene agents. Anticancer Drug Des. 1993;8:53-63.

41. Wang G, Xu X, Pace B, et al. Peptide nucleic acid (PNA) binding-mediated induction of human gamma-globin gene expression. Nucleic Acids Res. 1999;27:2806-2813.

42. Kostansek EC, Lipscomb WN, Yocum RR, Thiessen WE. Conforrnation of the mushroom toxin beta-amanitin in the crystalline state. Biochemistry.
1978;17:3790-3795.

43. Faulstich H, Trischmatul H, Wieland T, Wulf E. Ether derivatives of alpha-amanitin.
Introduction of spacer moieties, lipophilic residues, and radioactive labels.
Biochemistry.
1981;20:6498-6504.

44. Shoham G, Lipskomb WN, Wieland T. Conformations of Amatoxins in the Crystalline State. J Am Chem Soc. 1989;111:4791-4809.

45. Block SS, Stephens RL, Murrill WA. The Amanita Toxins in Mushrooms.
Agricultural and Food Chemistry. 1955;3:584 - 587.

46. Kostansek EC, Lipscomb WN, Yocum RR, Thiessen WE. The crystal structure of the mushroom toxin beta-amanitin. J Am Chem Soc. 1977;99:1273-1274.

47. Pomilio AB, Battista ME, Vitale AA. Semiempirical AMl and ab initio parameters of the lethal cyclopeptides alpha-amanitin and its related thioether, S-sulphoxide, sulphone, and O-methyl derivative. Journal of Molecular Structure (Theochem). 2001;536:243-262.

48. Bushnell DA, Crainer P, Kornberg RD. Structural basis of transcription:
alpha-amanitin-RNA polymerase Il cocrystal at 2.8 A resolution. Proc Natl Acad Sci U
S A.
2002;99:1218-1222.

49. Gong XQ, Nedialkov YA, Burton ZF. Alpha-amanitin blocks translocation by human RNA polymerase II. J Biol Chem. 2004;279:27422-27427.

50. Cramer P, Bushnell DA, Kornberg RD. Structural basis of transcription: RNA
polymerase II at 2.8 angstrom resolution. Science. 2001;292:1863-1876.

51. Boeger H, Bushnell DA, Davis R, et al. Structural basis of eukaryotic gene transcription. FEBS Lett. 2005;579:899-903.

52. Le Doan T, Perrouault L, Praseuth D, et al. Sequence-specific recognition, photocrosslinking and cleavage of the DNA double helix by an oligo-[alpha]-thymidylate covalently linked to an azidoproflavine derivative. Nucleic Acids Res.
1987;15:7749-7760.

53. Moser HE, Dervan PB. Sequence-specific cleavage of double helical DNA by triple helix formation. Science. 1987;238:645-650.

54. Casey BP, Glazer PM. Gene targeting via triple-helix formation. Prog Nucleic Acid Res Mo1 Bio1. 2001;67:163-192.

55. Brahmachari SK, Sarkar PS, Raghavan S, Narayan M, Maiti AK.
Polypurine/polypyrimidine sequences as cis-acting transcriptional regulators.
Gene.
1997;190:17-26.

56. Helene C, Giovannangeli C, Guieysse-Peugeot AL, Praseuth D. Sequence-specific control of gene expression by antigene and clamp oligonucleotides. Ciba Found Symp.
1997;209:94-102.

57. Rusling DA, Powers VE, Ranasinghe RT, et al. Four base recognition by triplex-forming oligonucleotides at physiological pH. Nucleic Acids Res. 2005;33:3025-3032.

58. Griffin LC, Kiessling LL, Beal PA, Gillespie P, Dervan PB. Recognition of All Four Base Pairs of Double-Helical DNA by Triple-Helix Formation: Design of Nonnatural Deoxyribonucleotides for Pyrimidine-Purine Base Pair Binding. J Am Chem Soc.
1992;114:7976-7982.

59. Maher LJd, Dervan PB, Wold B. Analysis of promoter-specific repression by triple-helical DNA complexes in a eukaryotic cell-free transcription system.
Biochemistry.
1992;31:70-81.

60. Giovannangeli C, Helene C. Progress in developments of triplex-based strategies.
Antisense Nucleic Acid Drug Dev. 1997;7:413-421.

61. Vasquez KM, Narayanan L, Glazer PM. Specific mutations induced by triplex-forming oligonucleotides in inice [In Process Citation]. Science. 2000;290:530-533.

62. Fritsch V, De Mesmaeker A, Waldner A, Lebreton J, Blommers MJ, Wolf RM.
Molecular mechanics and dynamics studies on two structurally related amide-modified DNA
backbones for antisense technology. Bioorg Med Chem. 1995;3:321-335.

63. Levis JT, Butler WO, Tseng BY, Ts'o PO. Cellular uptake of oligodeoxyribonucleoside methylphosphonates. Antisense Res Dev. 1995;5:251-259.

64. Griffin D, Laramee J, Deinzer M, Stirchak E, Weller D. Negative ion fast atom bombardment mass spectrometry of oligodeoxynucleotide carbamate analogs.
Biomed Environ Mass Spectrom. 1988;17:105-111.

65. Morvan F, Sangvhi YS, Perbost M, Vasseur JJ, Bellon L. Oligonucleotide mimics for antisense therapeutics: solution phase and automated solid-support synthesis of MMI linked oligomers. J Am Chem Soc. 1996;118:255-256.

66. Denisov AY, Zamaratski EV, Maltseva TV, et al. The solution conformation of a carbocyclic analog of the Dickerson-Drew dodecamer: comparison with its own X-ray structure and that of the NMR structure of the native counterpart. J Biomol Struct Dyn.
1998;16:547-568.

67. Zhang J, Shaw JT, Matteucci MD. Synthesis and hybridization property of an oligonucleotide containing a 3-thioforrnacetal linked pentathymidylate. Bioorg Med Chem Lett. 1999;9:319-3 22.

68. Nielsen PE. DNA analogues with nonphosphodiester backbones. Annu Rev Biophys Biomol Struct. 1995;24:167-183.

69. Gambari R. Peptide-nucleic acids (PNAs): a tool for the development of gene expression inodifiers. Curr Pharm Des. 2001;7:1839-1862.

70. Fujii M, Yoshida K, Hidaka J. Nucleic acid analog peptide (NAAP) 2.
Synthesis and properties of novel DNA analog peptides containing nucleobase linked beta-aminoalanine.
Bioorg Med Chem Lett. 1997;7:637-640.

71. Ueno Y, Mikawa M, Matsuda A. Nucleosides and nucleotides. 170. Synthesis and properties of oligodeoxynucleotides containing 5-[N-[2-[N,N-bis(2-aminoethyl)-amino] ethyl] carbamoyl]-2'-deoxyuridine and 5-[N-[3-[N,N-bis(3-aininopropyl) amino]propyl]carbamoyl]-2'-deoxyuridine. Bioconjug Chem. 1998;9:33-39.

72. Dande P, Liang G, Chen FX, et al. Regioselective effect of zwitterionic DNA
substitutions on DNA alkylation: evidence for a strong side chain orientational preference.
Biochemistry. 1997; 36:6024-6032.

73. Chaturvedi S, Horn T, Letsinger RL. Stabilization of triple-stranded oligonucleotide complexes: use of probes containing alternating phosphodiester and stereo-uniform cationic phosphorarnidate linkages. Nucleic Acids Res. 1996;24:2318-2323.

74. Dernpcy RO, Almarsson O, Bruice TC. Design and synthesis of deoxynucleic guanidine: a polycation analogue of DNA. Proc Natl Acad Sci U S A.
1994;91:7864-7868.

75. Puri N, Majumdar A, Cuenoud B, et al. Minimum number of 2'-O-(2-aminoethyl) residues required for gene knockout activity by triple helix forming oligonucleotides.
Biochernistry. 2002;41:7716-7724.

76. Kim JH, Kim KH, Mollegaard NE, Nielsen PE, Koo HS. Helical periodicity of (PNA)2.(DNA) triplexes in strand displacement complexes. Nucleic Acids Res.
1999;27:2842-2847.

77. Izvolsky KI, Demidov VV, Nielsen PE, Frank-Kamenetskii MD. Sequence-specific protection of duplex DNA against restriction and methylation enzymes by pseudocomplementary PNAs. Biochemistry. 2000;39:10908-10913.

78. Nedialkov YA, Gong XQ, Hovde SL, et al. NTP-driven translocation by human RNA
polyrnerase lI. J Biol Chem. 2003;278:18303-18312.

79. Wu L, Pan J, Thoroddsen V, et al. Novel small-molecule inhibitors of RNA
polymerase 111. Eukaryot Cell. 2003;2:256-264.

80. Rappaport J, Hanss B, Kopp JB, et al. Transport of phosphorothioate oligonucleotides in kidney: implications for molecular therapy. Kidney Int. 1995;47:1462-1469.

81. Agrawal S, Temsamani J, Tang JY. Pharmacokinetics, biodistribution, and stability of oligodeoxynucleotide phosphorothioates in mice. Proc Natl Acad Sci U S A.
1991;88:7595-7599.

82. Catapano CV, McGuffie EM, Pacheco D, Carbone GM. Inhibition of gene expression and cell proliferation by triple helix-forming oligonucleotides directed to the c-myc gene.
B i o chemistry. 2000; 3 9: 512 6-513 8.

83. Ljungstrom T, Knudsen H, Nielsen PE. Cellular uptake of adamantyl conjugated peptide nucleic acids. Bioconjug Chem. 1999;10:965-972.

84. Ghosh C, Iversen PL. Intracellular delivery strategies for antisense phosphorodiamidate morpholino oligomers. Antisense Nucleic Acid Drug Dev. 2000;10:263-274.

85. Schwarze SR, Hruska KA, Dowdy SF. Protein transduction: unrestricted delivery into all cells? Trends Cell Biol. 2000;10:290-295.

86. Schwarze SR, Ho A, Vocero-Akbani A, Dowdy SF. In vivo protein transduction:
delivery of a biologically active protein into the mouse [see comments].
Science.
1999;285:1569-1572.

87. Maher LJ, 3rd. Inhibition of T7 RNA polymerase initiation by triple-helical DNA
complexes: a model for artificial gene repression. Biochemistry. 1992;31:7587-7594.

88. Wang G, Xu XS. Peptide nucleic acid (PNA) binding-mediated gene regulation. Cell Res. 2004;14:111-116.

89. Baertschi AJ. Antisense oligonucleotide strategies in physiology. Mol Cell Endocrinol.
1994; l O1:R15-24.

90. Hacia JG, Wold BJ, Dervan PB. Phosphorothioate oligonucleotide-directed triple helix formation. Biochemistry. 1994;33:5367-5369.

91. Barre FX, Ait-Si-Ali S, Giovannangeli C, et al. Unambiguous demonstration of triple-helix-directed gene modification. Proc Natl Acad Sci U S A. 2000;97:3084-3088.

92. Arora PS, Ansari AZ, Best TP, Ptashne M, Dervan PB. Design of artificial transcriptional activators with rigid poly-L-proline linkers. J Am Chem Soc.
2002;124:13067-13071.

93. Bhadra D, Bhadra S, Jain P, Jain NK. Pegnology: a review of PEG-ylated systems.
Pharmazie. 2002; 57: 5 -29.

94. Wieland T, Gotzendorfer C, Dabrowski J, Lipscomb WN, Shoham G. Unexpected similarity of the structures of the weakly toxic amanitin (S)-sulfoxide and the highly toxic (R)-sulfoxide and sulfone as revealed by proton nuclear magnetic resonance and X-ray analysis.
Biochemistry. 1983;22:1264-1271.

95. Baumann K, Munter K, Faulstich H. Identification of structural features involved in binding of alpha-amanitin to a monoclonal antibody. Biochemistry. 1993;32:4043-4050.

96. Zanotti G, Wieland T, Benedetti E, Di Blasio B, Pavone V, Pedone C.
Structure-toxicity relationships in the amatoxin series. Synthesis of S-deoxy[gamma(R)-hydroxy-Ile3]-amaninamide, its crystal and molecular structure and inhibitory efficiency.
Int J Pept Protein Res. 1989;34:222-228.

97. Wang H, Cai Q, Zeng X, Yu D, Agrawal S, Zhang R. Antiturnor activity and pharmacokinetics of a mixed-backbone antisense oligonucleotide targeted to the RIalpha subunit of protein kinase A after oral administration. Proc Natl Acad Sci U S
A.
1999;96:13989-13 994.

98. Wu-Pong S, Weiss TL, Hunt CA. Antisense c-myc oligodeoxyribonucleotide cellular uptake. Pharm Res. 1992;9:1010-1017.

99. Sadowski I, Ma J, Triezenberg S, Ptashne M. GAL4-VP16 is an unusually potent transcriptional activator. Nature. 1988;335:563-564.

100. Skoog JU, Maher LJd. Repression of bacteriophage promoters by DNA and RNA
oligonucleotides. Nucleic Acids Res. 1993;21:2131-2138.

101. Ing NH, Beekman JM, Kessler DJ, et al. In vivo transcription of a progesterone-responsive gene is specifically inhibited by a triplex-forming oligonucleotide. Nucleic Acids Res. 1993;21:2789-2796.

102. Durland RH, Kessler DJ, Gunnell S, Duvic M, Pettitt BM, Hogan ME. Binding of triple helix forming oligonucleotides to sites in gene promoters. Biochemistry.
1991;30:9246-9255.

103. Pei D, Corey DR, Schultz PG. Site-specific cleavage of duplex DNA by a semisynthetic nuclease via triple-helix formation. Proc Natl Acad Sci U S A. 1990;87:9858-9862.

104. Baer A, Schubeler D, Bode J. Transcriptional properties of genomic transgene integration sites inarked by electroporation or retroviral infection.
Biochemistry. 2000;39:7041-7049.

105. Postel EH. Modulation of c-myc transcription by triple helix formation.
Ann N Y Acad Sci. 1992;660:57-63.

106. Postel EH, Flint SJ, Kessler DJ, Hogan ME. Evidence that a triplex-forming oligodeoxyribonucleotide binds to the c-myc promoter in HeLa cells, thereby reducing c-myc mRNA levels. Proc Natl Acad Sci U S A. 1991;88:8227-8231.

107. Xu XS, Glazer PM, Wang G. Activation of human gamma-globin gene expression via triplex-fonning oligonucleotide (TFO)-directed mutations in the gamma-globin gene 5' flanking region. Gene. 2000;242:219-228.

108. DeLeo AB, Jay G, Appella E, Dubois GC, Law LW, Old LJ. Detection of a transformation-related antigen in chemically induced sarcomas and other transformed cells of the mouse. Proc Natl Acad Sci U S A. 1979;76:2420-2424.

109. Baker SJ, Fearon ER, Nigro JM, et al. Chromosome 17 deletions and p53 gene mutations in colorectal carcinomas. Science. 1989;244:217-221.

110. Nigro JM, Baker SJ, Preisinger AC, et al. Mutations in the p53 gene occur in diverse human tumour types. Nature. 1989;342:705-708.

111. Lane DP. Cancer. p53, guardian of the genome. Nature. 1992;358:15-16.

112. Vogelstein B, Kinzler KW. Aehilles' heel of cancer? Nature. 2001;412:865-866.

113. Vousden KH. Activation of the p53 tumor suppressor protein. Biochim Biophys Acta.
2002;1602:47-59.

114. Smith ML, Chen IT, Zhan Q, et al. Interaction of the p53-regulated protein Gadd45 with proliferating cell nuclear antigen. Science. 1994;266:1376-1380.

115. Vogelstein B, Kinzler KW. Cancer genes and the pathways they control. Nat Med.
2004;10:789-799.

116. Moll UM, Wolff S, Speidel D, Deppert W. Transcription-independent pro-apoptotic functions of p53. Curr Opin Cell Biol. 2005;17:631-636.

117. Levine AJ, Momand J, Finlay CA. The p53 turnour suppressor gene. Nature.
1991;351:453-456.

118. Soussi T, Beroud C. Assessing TP53 status in human tumours to evaluate clinical outcome. Nature Rev Cancer. 2001;1:233-240.

119. Raman V, Martensen SA, Reisman D, et al. Compromised HOXA5 function can limit p53 expression in human breast tumours. Nature. 2000;405:974-978.

120. Oliner JD, Kinzler KW, Meltzer PS, George DL, Vogelstein B. Amplification of a gene encoding a p53-associated protein in human sarcomas. Nature. 1992;358:80-83.

121. Vogelstein B, Kinzler KW. p53 function and dysfunction. Cell. 1992;70:523-526.

122. Harris CC, Hollstein M. Clinical implications of the p53 tumor-suppressor gene. N Engl J Med. 1993;329:1318-1327.

123. Garcia-Cao I, Garcia-Cao M, Tomas-Loba A, et al. Increased p53 activity does not accelerate telomere-driven ageing. EMBO Rep. 2006;7:546-552.

124. Mendrysa SM, O'Leary KA, McElwee MK, et al. Tumor suppression and normal aging in mice with constitutively high p53 activity. Genes Dev. 2006;20:16-21.

125. Roth JA, Grammer SF, Swisher SG, et al. Gene therapy approaches for the management of non-small cell lung cancer. Semin Oncol. 2001;28:50-56.

126. Selivanova G. Mutant p53: the loaded gun. Curr Opin Investig Drugs.
2001;2:1136-1141.

127. Watanabe T, Sullenger BA. Induction of wild-type p53 activity in human cancer cells by ribozymes that repair mutant p53 transcripts. Proc Natl Acad Sci U S A.
2000;97:8490-8494.

128. Bykov VJ, Issaeva N, Shilov A, et al. Restoration of the tumor suppressor function to rnutant p53 by a low-molecular-weight compound. Nat Med. 2002;8:282-288.

129. Vassilev LT. MDM2 inhibitors for cancer therapy. Trends Mol Med.
2007;13:23-31.

130. Hietanen S, Lain S, Krausz E, Blattner C, Lane DP. Activation of p53 in cervical carcinoma cells by small molecules. Proc Natl Acad Sci U S A. 2000;97:8501-8506.

131. Chen L, Lu W, Agrawal S, Zhou W, Zhang R, Chen J. Ubiquitous induction of p53 in tumor cells by antisense inhibition of MDM2 expression. Mol Med. 1999;5:21-34.

132. Stuart ET, Haffner R, Oren M, Gruss P. Loss of p53 function through PAX-mediated transcriptional repression. Embo J. 1995;14:5638-5645.

133. Vadolas J, Wardan I-I, Orford M, et al. Development of sensitive fluorescent assays for embryonic and fetal hemoglobin inducers using the human beta -globin locus in erythropoietic cells. Blood. 2002;100:4209-4216.

134. Swank RA, Stamatoyannopoulos G. Fetal gene reactivation. Curr Opin Genet Dev.
1998;8:366-370.

135. Vadolas J, Wardan H, Orford M, Williamson R, Ioannou PA. Cellular genomie reporter assays for screening and evaluation of inducers of fetal hemoglobin. Hum Mol Genet.
2004;13:223-233.

136. Chao SH, Price DH. Flavopiridol inactivates P-TEFb and blocks most RNA
polyrnerase II transcription in vivo. J Biol Chem. 2001;276:31793-31799.

137. Chao SH, Fujinaga K, Marion JE, et al. Flavopiridol inhibits P-TEFb and blocks HIV-1 replication. J Biol Chem. 2000;275:28345-28348.

138. te Poele RH, Okorokov AL, Joel SP. RNA synthesis block by 5, 6-dichloro-l-beta-D-ribofuranosylbenzimidazole (DRB) triggers p53-dependent apoptosis in human colon carcinoma cells. Oncogene. 1999;18:5765-5772.

139. Steinberg TH, Burgess RR. Tagetitoxin inhibition of RNA polymerase III
transcription results from enhanced pausing at discrete sites and is template-dependent. J
Biol Chem.
1992;267:20204-20211.

140. Svinarchuk F, Bertrand JR, Malvy C. A short purine oligonucleotide forms a highly stable triple helix with the promoter of the murine c-piin-1 proto-oncogene.
Nucleic Acids Res.
1994;22:3742-3747.

141. Nakanishi M, Weber KT, Guntaka RV. Triple helix formation with the promoter of human alphal(I) procollagen gene by an antiparallel triplex-forming oligodeoxyribonucleotide.
Nucleic Acids Res. 1998;26:5218-5222.

142. Olivas WM, Maher LJ, 3rd. Binding of DNA oligonucleotides to sequences in the promoter of the human bcl-2 gene. Nucleic Acids Res. 1996;24:1758-1764.

143. Vasquez KM, Dagle JM, Weeks DL, Glazer PM. Chromosome targeting at short polypurine sites by cationic triplex-forming oligonucleotides. J Biol Chem.
2001;276:38536-38541.

144. de Bizemont T, Sun JS, Garestier T, Helene C. New junction models for alternate-strand triple-helix formation. Chem Biol. 1998;5:755-762.

145. Mohan V, Cheng YK, Marlow GE, Pettitt BM. Molecular recognition of Watson-Crick base-pair reversals in triple-helix formation: use of nonnatural oligonucleotide bases.
Biopolymers. 1993;3 3 :1317-1325.

146. Gowers DM, Fox KR. Towards mixed sequence recognition by triple helix formation.
Nucleic Acids Res. 1999;27:1569-1577.

147. Levasseur DN, Ryan TM, Pawlik KM, Townes TM. Correction of a mouse model of sickle cell disease: lentiviral/antisickling beta-globin gene transduction of unmobilized, purified hematopoietic stem cells. Blood. 2003;102:4312-4319.

148. Ryan TM, Ciavatta DJ, Townes TM. Knockout-transgenic mouse model of sickle cell disease. Science. 1997;278:873-876.

149. Venkatachalam S, Tyner SD, Pickering CR, et al. Is p53 haploinsufficient for tumor suppression? Implications for the p53+/- mouse model in carcinogenicity testing. Toxicol Pathol. 2001;29:147-154.

150. Hoogervorst EM, van Oostrom CT, Beems RB, et al. p53 heterozygosity results in an increased 2-acetylaminofluorene-induced urinary bladder but not liver tumor response in DNA
repair-deficient Xpa mice. Cancer Res. 2004;64:5118-5126.

151. Luck L, Zeng L, Hiti AL, Weinberg KI, Malik P. Human CD34(+) and CD34(+)CD38(-) hematopoietic progenitors in sickle cell disease differ phenotypically and functionally from normal and suggest distinct subpopulations that generate F
cells. Exp Hematol. 2004; 3 2 :483 -493 .

Claims

1. An artificial transcription factor, comprising:
(a) a non-peptidic DNA-binding domain;
(b) a linker; and (c) an amanitin wherein the linker is located between the DNA-binding domain and the amanitin.

2. The artificial transcription factor of claim 1, wherein the amanitin is selected from the group consisting of .alpha.-amanitin, .beta.-amanitin, .gamma.-amanitin, and .epsilon.-amanitin.

3. The artificial transcription factor of claim 1, wherein the amanitin is a modified amanitin.

4. The artificial transcription factor of claim 3, wherein the amanitin is chemically modified by modification of naturall.gamma.-occurring amanitin or by total chemical synthesis.

5. The artificial transcription factor of claim 3, wherein the modified amanitin is selected from the group consisting of O-Methyl-.alpha.-amanitin, S-deoxo-.alpha.-amanitin, .alpha.-amanitin-(S)-sulfoxide, .alpha.-amanitin sulfone, O-Methyl-.alpha.-amanitin (S) sulfoxide, O-methyl-.alpha.-amanitin sulfone, O-methyl-.alpha.-amanitin sulfide, .gamma.-amanitin, methyl-.gamma.-amanitin, .beta.-amanitin methyl ester, methyl (3-amanitin methyl ester, amanullin, amanullic acid, amaninainide, amanin, 2,3-diacetyl-6'-O-methyl-.gamma.-amantin, 3-acetyl-6'-O-methyl-.gamma.-amantin, 2-acetyl-6'-O-methyl-.gamma.-amantin, methylaldoamanitin, dethiomethyl-.alpha.-amanitin (monocyclic), peptide ring opened amanitin (monocyclic), S-deoxo[g(R)-hydrox.gamma.-Ile]-amaninamide, S-deoxo-Ile-amaninamide, .beta.-amanitin thiophenyl ester, O-amanitin anilide, .beta.-amanitin dodecylamide, O-methyl(dehydroxymethyl)-.alpha.-amanitin, [4'-[ [(6-aminohexyl) amino] carbonyl]phenyl] azo-.alpha.-amanitin trifluoroacetate, [4' [[[6-[(tert-butyloxycarbonyl)amino]hexyl]amino]carbonyl]-phenyl]azo-.alpha.-amanitin, 7'iodo-.alpha.-amanitin, O-(4'-tetrazolyl)phenyl-.alpha.-amanitin, O-ethyl-.alpha.-amanitin, O-propyl-.alpha.-amanitin, O-allyl-.alpha.-amanitin, O-(n-hexyl)-.alpha.-amanitin, O-(n-decyl)-.alpha.-amanitin, O-benzyl-.alpha.-amanitin, O-acetonyl-.alpha.-amanitin, O-(2-hydroxyropyl)-.alpha.-amanitin, O-(5-carboxypentyl)-.alpha.-amanitin, O-[5-[[(aninoethyl)-amino]carbonyl]-pent-1-yl]-.alpha.-amanitin fluoroacetate, O-[5-[[[(succinoyl-amino)ethyl]-amino]carbonyl]-pent-1-yl]-.alpha.-amanitin, O-[5-[[[.beta.-[(tert-butyloxycarbonyl) -amino]ethyl]-amino]carbonyl]-pent-1-yl]-.alpha.-amanitin, and O,N-dimethyl-.alpha.-amanitin.

6. The artificial transcription factor of claim 1, wherein the non-peptidic DNA-binding domain is a nucleic acid molecule.

7. The artificial transcription factor of claim 6, wherein the nucleic acid molecule is modified.

8. The artificial transcription factor of claim 7, wherein the nucleic acid molecule has a modified backbone, a modified base, or a modified sugar.

9. The artificial transcription factor of claim 8, wherein the nucleic acid molecule backbone is modified to contain a peptide nucleic acid (PNA), a PNA analogue bearing phosphate groups (PHONA), a peptide nucleic acid analog (PNAA), or any combination thereof.

10. The artificial transcription factor of claim 1, wherein the DNA-binding domain is a triplex forming oligonucleotide.

11. An artificial transcription factor, comprising:
(a) a non-peptidic DNA-binding domain;
(b) a linker; and (c) a small molecule compound wherein the linker is located between the DNA-binding domain and the small molecule compound.

12. The artificial transcription factor of claim 11, wherein the non-peptidic DNA-binding domain is a nucleic acid molecule.

13. The artificial transcription factor of claim 12, wherein the nucleic acid molecule is modified.

14. The artificial transcription factor of claim 13, wherein the nucleic acid molecule has a modified backbone, a modified base, or a modified sugar.

15. The artificial transcription factor of claim 14, wherein the nucleic acid molecule backbone is modified to replace at least one of the phosphodiester bonds with a neutral internucleoside linkage.

62 `16. The artificial transcription factor of claim 15, wherein the neutral internucleoside linkage is selected from the group consisting of phosphorothioates, amides, phosphonates, carbamates, methylenmethylimino, heterocycles, acetals, phosphoroamidites, and any combination thereof.

17. The artificial transcription factor of claim 14, wherein the nucleic acid molecule backbone is modified to contain a peptide nucleic acid (PNA), a PNA analogue bearing phosphate groups (PHONA), a peptide nucleic acid analog (PNAA), or any combination thereof.

18. The artificial transcription factor of claim 14, wherein the nucleic acid molecule is modified to introduce positive charges to the backbone, bases, or sugar rings.

19. The artificial transcription factor of claim 1, wherein the DNA-binding domain is a triplex forming oligonucleotide.

20. The artificial transcription factor of claim 11, wherein the DNA-binding domain is a polyamide.

21. The artificial transcription factor of claim 11, wherein the DNA-binding domain comprises a plurality of pyrolle or imidazole groups.

22. The artificial transcription factor of claim 11, wherein the DNA-binding domain binds a sequence of 10 or more contiguous purine bases on one strand.

23. The artificial transcription factor of claim 11, wherein the DNA-binding domain binds a sequence of 10 or more purine bases on one strand wherein the purine bases are interrupted by a pyrimidine base.

24. The artificial transcription factor of claim 11, wherein the linker is a polymer comprising a polymer of ethylene glycol, alkyl groups, nucleotides, amino acids, amides, and ketones.

25. The artificial transcription factor of claim 11, wherein the linker is comprised of a plurality of monomer units selected from the group consisting of nucleotides, peptides, and lower alkyls.

26. The artificial transcription factor of claim 11, wherein the linker is composed of a polymerized glycol.

27. The artificial transcription factor of claim 11, wherein the linker comprises about 2 to about 30 glycol units.

28. The artificial transcription factor of claim 11, wherein the linker comprises at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, at least 18, at least 19, at least 20, or at least 21 glycol units.

29. The artificial transcription factor of claim 11, wherein the linker has a length of about 5 angstroms to about 200 angstroms.

30. The artificial transcription factor of claim 11, wherein the linker has a length of about 0.1 angstroms to about 5 angstroms.

31. The artificial transcription factor of claim 11, wherein the linker is attached to the 3' or 5' end of the non-peptidic DNA-binding domain.

32. The artificial transcription factor of claim 11, wherein the small molecule compound binds RNA Polymerase II, RNA Polymerase III, any component of the RNA Polymerase holoenzyme, or any component of the mediator complex.

33. The artificial transcription factor of claim 11, wherein the DNA-binding domain and the effector are modified.

34. The artificial transcription factor of claim 33, wherein the modification is selected from the group consisting of oxidation, hydroxylation, substitution, and reduction.

35. A composition comprising an ATF of any one of the preceding claims, and a carrier.

36. The composition of claim 35, wherein the carrier is selected from the group consisting of water, phosphate buffered saline, bacteriostatic water for injection, and sterile water for injection.

37. The composition of claim 35, further comprising a buffer.

38. A kit comprising an artificial transcription factor of any one of claims 1-34, and directions for use.

39. A kit comprising a composition of any one of claims 35-37, and instructions for use.

40. A method for modulating the expression of a nucleic acid molecule in a eukaryotic cell, comprising contacting the cell or introducing into the cell an effective amount of an artificial transcription factor of claim 11, that can regulate the expression of the nucleic acid, wherein the artificial transcription factor binds to its binding site on the nucleic acid and modulates the expression of the nucleic acid in the eukaryotic cell.

41. The method of claim 40, wherein the nucleic acid molecule is a reporter gene.

42. The method of claim 40, wherein the nucleic acid molecule is an endogenous gene.

43. The method of claim 40, wherein the nucleic acid molecule is introduced into the cell by a method selected from the group consisting of free diffusion, lipofection, electroporation, particle bombardment, calcium-phosphate precipitation, and the use of cell-membrane transduction peptides.

44. A method for modulating the expression of a nucleic acid molecule in a subject in need thereof, comprising administering to the subject an effective amount of an artificial transcription factor of claim 11, that regulates the expression of the nucleic acid.

45. The method of claim 44, wherein the nucleic acid molecule is selected from the group consisting of p53, .gamma.-globin, T-bet, NF-.kappa.B, insulin gene, insulin promoter factor 1(IPF1 or IDX-1), and E-cadherin.