WO2009004632A2

WO2009004632A2 - Methods of identifying components of a biological pathway and use of said components in regulating diseases associated with altered cell proliferation

Info

Publication number: WO2009004632A2
Application number: PCT/IL2008/000920
Authority: WO
Inventors: Yitzhak Pilpel; Reut Shalgi; Moshe Oren
Original assignee: Yeda Research And Development Co. Ltd.
Priority date: 2007-07-05
Filing date: 2008-07-03
Publication date: 2009-01-08
Also published as: WO2009004632A3

Abstract

A method of identifying components of a biological pathway is disclosed. The method comprises selecting a transcription factor and a microRNA pair which regulate a common gene, the transcription factor and the microRNA being the components of the biological pathway. Methods of treating diseases associated with cell proliferation using components identified according to the method above and pharmaceutical compositions comprising same are also disclosed.

Description

METHODS OF IDENTIFYING COMPONENTS OF A BIOLOGICAL PATHWAY AND USE OF SAID COMPONENTS IN REGULATING DISEASES ASSOCIATED WITH

ALTERED CELL PROLIFERATION

FIELD AND BACKGROUND OF THE INVENTION

The present invention, in some embodiments thereof, relates to a method of identifying components of a biological pathway. In addition, the present invention relates to use of identified components for regulating diseases associated with altered cell proliferation.

MicroRNAs (miRs) are short RNAs that post transcriptionally regulate messenger RNAs. Two main mechanisms for such effects are degradation of the target mRNA, and inhibition of its translation. In recent years considerable progress within multiple genomes was obtained in the experimental identification of genes encoding for miRs, and in tools for the identification of target genes of miRs, based on miR sequences and the sequence of the targets' 3' untranslated regions (UTRs). Compared with the regulation of transcription, the study of the regulatory networks spanned by miRs is only at its beginning. A lot is known about the main players of transcriptional regulation and the interactions between them. Transcription factors (TFs) are well characterized and promoter binding motifs are available in a diversity of species. The combinatorial interactions between TFs have been explored as well as the global level properties of the transcription regulatory network. In addition, the local structures of the network have been intensively investigated. It was found in several species that the transcription regulatory network may be decomposed into elementary building blocks, or network motifs, that recur in the network more than expected by chance, and that these motifs likely perform local "computations," such as the detection of signal persistency or the coordinated gradual activation of a set of genes.

When it comes to posttranscriptional regulation, and in particular to the miR world, most of the parallel knowledge is lacking. While evidence exists for the occurrence of many miRs in multiple genomes, their targets are predicted with relatively limited accuracy. Furthermore, there is a lack of knowledge about the structure of the miR regulatory network, and about the potential interface between this network and the transcriptional one. Similarly to TFs, miRs are expected to work in combinations on their target genes. The target specificity-determining site of the miRs is often short (seven to eight nucleotides), hence some genes that contain a match to a single miR in their 3' UTRs may represent false positive assignments. Thus combinatorial interactions among the miRs are probably necessary to allow specific targeting of genes targeted by each miR. As in the realm of transcription regulators, combinatorics may also have the advantage of allowing multiple sources of information, each represented by a single miR, to be integrated into the regulation of individual transcripts. Since TFs regulate mRNA production, and miRs regulate transcript stability and its translation, an attractive possibility is that miRs and TFs cooperate in regulating shared target genes. This may be advantageous since a gene that is regulated through multiple mechanisms may be tuned at a level of precision that is higher than what may be obtained by either mechanism alone. The tumor suppressor p53 is a sequence-specific transcription factor (TF) that exerts many of its downstream effects by activating gene transcription. P53 is considered a central regulator of cell fate decisions. Activation of p53 can induce several cellular responses, including cell-cycle arrest, senescence and apoptosis. Thus, absence of functional p53 predisposes cells to neoplastic transformation. Accordingly, mutations of this gene are highly common in human cancers. Even though p53 is known as a transcription factor, additional transactivation-independent functions of p53 contribute to its tumor suppressive activity, including protein-protein interactions with additional transcription factors and other cell fate regulators. The importance of transcription regulation by p53 is exemplified by the fact that most p53 tumor-derived mutants are defective in DNA binding and incapable of transactivation. In addition to its capability to induce gene transcription, p53 activation results in extensive gene repression. Direct and indirect transcriptional repression by p53 is considered important for its tumor suppressive functions, such as induction of cell-cycle arrest and apoptosis.

As well as transcription factors, microRNAs are also known to regulate cancer- related processes such as apoptosis, proliferation and differentiation. Deregulated miRs were suggested to exert their function in cancer via silencing of key cell fate regulators, as shown for let-7 and Ras (Johnson et al, 2005, Cell 120: 635-647), as well as for miR-106b and ρ21 (Ivanovska et al, 2008, MoI Cell Biol; Petrocca et al, 2008, Cancer Cell 13: 272- 286). Several studies have implicated p53 in the regulation of miR expression (Chang et al, 2007, MoI Cell; He et al, 2007, Nature; Kumamoto et al, 2008, Cancer Res, In Press; Raver-Shapira et al, 2007, MoI Cell; Tarasov et al, 2007, Cell Cycle 6; Xi et al, 2006, Clin Cancer Res 12: 2014-2024). These studies exploited various high throughput methods to identify p53-regulated miRs in several cellular systems with differential p53 status. Although the resulting candidate lists from each study differed considerably, probably due to differences in the cellular contexts and p53 activation signals, all studies identified members of the miR-34 family as direct transactivation targets of p53. In line with p53 function, induction of miR-34 family members was suggested to mediate cell-cycle arrest, apoptosis and senescence. Importantly, none of these studies focused on miRs whose expression negatively correlated with p53 activation, and who are likely repressed by this tumor suppressor.

Many miRNAs have been shown to be overexpressed in various tumors; and some have been shown to possess oncogenic functions. For example, miR-106a, miR-17-5p, miR-20a and miR-155, as well as miR-92, were reported to be commonly overexpressed in solid tumors (Volinia et al, 2006, Proc Natl Acad Sci U S A 103: 2257-2261). Members of the miR- 17-92 polycistron are overexpressed in lymphomas as well as in lung and colorectal carcinomas (He et al, 2005, Nature 435: 828-833; Schetter et al, 2008, Jama 299: 425-436), and were shown to accelerate tumor growth (O'Donnell et al, 2005, Nature 435: 839-843). Finally, miR-155 was reported to be specifically overexpressed in several types of B-cell lymphomas (Eis et al, 2005, Proc Natl Acad Sci U S A 102: 3627-3632) and to predict poor prognosis in lung cancer (Yanaihara et al, 2006, Cancer Cell 9: 189- 198).

SUMMARY OF THE INVENTION

According to an aspect of some embodiments of the present invention there is provided a method of identifying components of a biological pathway, the method comprising selecting a transcription factor and a microRNA pair which regulate a common gene, the transcription factor and the microRNA being the components of the biological pathway.

According to an aspect of some embodiments of the present invention there is provided a method of treating a hyperproliferative disease in a subject, the method comprising administering to the subject a therapeutically effective amount of an oligonucleotide agent capable of down-regulating at least one microRNA selected from the group consisting miR-106b, miR-93, miR-25, miR-17, miR-18a, miR-19a, miR-20a, miR- 19b-l, miR-92a-l, miR-106a, miR-18b, miR-20b, miR-19b-2, miR-92-2*, hsa-miR-l9b- 1* miR-15b, miR-16, hsa-miR-92a-2*, and hsa-miR-92a-l* into the subject, thereby treating the hyperproliferative disease. According to an aspect of some embodiments of the present invention there is provided a method of treating a degenerative disease in a subject, the method comprising administering to the subject a therapeutically effective amount of at least one microRNA selected from the group consisting of miR-106b, miR-93, miR-25, miR-17, miR-18a, miR- 19a, miR-20a, miR-19b-l, miR-92a-l, miR-106a, miR-18b, miR-20b, miR-19b-2, miR-92- 2*, hsa-miR-19b-l* miR-15b, miR-16, hsa-miR-92a-2*, and hsa-miR-92a-l* into the subject, thereby treating the degenerative disease.

According to an aspect of some embodiments of the present invention there is provided a pharmaceutical composition comprising a pharmaceutically acceptable carrier and as an active agent at least one microRNA selected from the group consisting of miR-

106b, miR-93, miR-25, miR-17, miR-18a, miR-19a, miR-20a, miR-19b-l, miR-92a-l, miR-106a, miR-18b, miR-20b, miR-19b-2, miR-92-2, miR-15b and miR-16.

According to some embodiments of the invention, the transcription factor and the microRNA pair are listed in Tables 5-8. According to some embodiments of the invention, a gene encoding the microRNA comprises a binding site for the transcription factor and/or a gene encoding the transcription factor comprises a binding site for the microRNA.

According to some embodiments of the invention, the transcription factor and the microRNA pair are listed in Table 9-11. According to some embodiments of the invention, the transcription factor and the microRNA pair are listed in Figure 4.

According to some embodiments of the invention, the transcription factor regulates a transcription of an additional transcription factor, the microRNA comprising a binding site for the additional transcription factor. According to some embodiments of the invention, the transcription factor and the microRNA are listed in Tables 12-14.

According to some embodiments of the invention, the hyperproliferative disease is cancer.

Unless otherwise defined, all technical and/or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of embodiments of the invention, exemplary methods and/or materials are described below. In case of conflict, the patent specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and are not intended to be necessarily limiting.

BRIEF DESCRIPTION OF THE DRAWINGS Some embodiments of the invention are herein described, by way of example only, with reference to the accompanying drawings. With specific reference now to the drawings in detail, it is stressed that the particulars shown are by way of example and for purposes of illustrative discussion of embodiments of the invention. In this regard, the description taken with the drawings makes apparent to those skilled in the art how embodiments of the invention may be practiced. In the drawings:

FIGs. IA-B are graphs illustrating miRs and Target Genes in the TargetScan Dataset. (Figure IA) Distribution of the number of different miRs regulating each target gene in the TargetScan dataset. The thick red line represents the distribution in the original dataset, while each of the thin blue lines represents the distribution in one of the column- randomized matrices. The matrix contains only genes with at least one predicted site in their 3' UTR. In each randomization, the assignment of miRs to their targets was shuffled, keeping constant the number of targets per miR. (Figure IB) Distribution of number of targets per miR in the TargetScan dataset. The original distribution is depicted on thick red line, while each blue thin line represents the distribution in one of the 100 row-randomized matrices, which preserves the distribution of number of miRs targeting each gene.

FIGs. 2A-B are graph illustrating the distribution of the density of miRs in the 3' UTRs of target hubs (thick red line) and all the genes (thin blue line) in the TargetScan dataset (Figure 2A) and in the PicTar dataset (Figure 2B). All genes included in this figures have at least one miR site predicted in their 3' UTR. The log 10 densities were binned into bins of 0.1, and relative frequencies were plotted.

FIG. 3 A is a representation of the TargetScan miR co-occurrence network, at FDR level of 0.05. A node represents a miR and an edge connects between pairs of miRs with significant rate of co-occurrence. The nodes in the figure are arranged from most highly connected on the top, to most lowly connected, on the bottom.

FIGs. 3B-C are graphs illustrating the degree distribution in the TargetScan (Figure 3B) and PicTar (Figure 3C) miR combinatorial regulation network (co-occurring miR pairs that passed FDR of 0.05). FIG. 4 is a representation of network designs in the miR-TF Coregulation

Network. The figure depicts the analyzed network motifs in the TargetScan and PicTar dataset, with the use of TF binding sites in RefSeq genes promoters of 10 kb for both networks, and 5 kb for the PicTar network. The figure depicts, for each network motif, its architecture, the number of times it appears in each of the networks, the p- value and z- score for its over representation in the network (as described in Materials and Methods), the total number of RefSeq genes that are regulated by this type of network design, and an example. *For the first design, the coregulating miR-TF pair, the range of hypergeometric p- values of pairs that passed FDR and are considered significant are stated, and in brackets the FDRp- value of these pairs using the randomization co-occurrence test. **in addition, z-scores for significant pairs were calculated based on the co-occurrence edge-swapping randomization model (see Materials and Methods of Example 1).

FIGs. 5A-C are graphs illustrating tissue Expression Correlations between miRs and TFs. miR tissue expression in brain, liver, thymus, testes, and placenta were taken from Barad O, et al. (2004), Genome Res 14: 2486-2494; mRNA tissue expression was taken from Su et al., (2004) Proc Natl Acad Sci U S A 101 : 6062-6067. (Figure 5A) Background distribution of all possible miR-TF pairs for which expression profiles can be derived. (Figures 5B, C) Normalized histograms of correlation coefficients; the same distribution as in (A) was made, yet only for significantly co-occurring miR-TF pairs (red), and FFLs (green) in the PicTar (B) and TargetScan (C) networks. The figure shows the proportion of the various correlation coefficients divided by the background distribution depicted in Figure 5 A.

FIGs. 6A-B are graph illustrating miR Binding Sites in Target Hub Genes in the TargetScan and PicTar Datasets. Mean number of miRs targeting each of the genes that are target hubs (red bar), in the entire set of analyzed genes (green) and a distribution of that mean in random gene sets with the same (or very similar, see Materials and Methods) distribution of 3' UTR lengths as the target hubs (blue) in (Figure 6A) the TargetScan dataset and (Figure 6B) the PicTar dataset. For elaborated procedure see Materials and Methods. FIG. 7 is a graph illustrating the distribution of number of miRs per cluster. As seen, ~82 % of the 301 clusters contain a single miR.

FIGs. 8A-C are graphs illustrating an analysis of miR clusters in the Human Genome. (Figure 8A) Distribution of distances between all neighboring pre-miR genes in the human genome. (Figure 8B) Distribution of tissue expression correlations between pairs of miRs: all possible pairs in the data (thin blue line) and pairs of miRs which reside in shared clusters (thick red line). In the inset are shown tissue expression correlations between pairs of miRs in the same genomic clusters vs. distances between them. (Figure 8C) Distribution of number of conserved TFBS 30 kb upstream of the 5' most nucleotide in each miR clusters. Conserved TFBSs were taken from UCSC hgl7.

FIG. 9 is a graph illustrating the distribution of a number of conserved TFBS 30 kb Upstream of TSS of RefSeq Protein-Coding Genes.

FIGs. 10A-F are graphs and photographs illustrating the establishment of the WI- 38 system. WI-38 primary human fibroblasts were infected with a retrovirus encoding for the p53 -inactivating peptide, GSE56. These cells (GSE) and their active p53 counterparts (NEO) were treated with the DNA damaging agent doxorubicin as well as grown until the onset of replicative senescence. Figure 1OA: Western blot depicting p53 and p21 following doxorubicin treatment. p53 was stabilized and activated its target gene p21 in the NEO cells upon treatment. In the GSE cells, p53 was stabilized in its inactive form already at basal levels by GSE56, a 15 kDa peptide detected by the H-47 anti-p53 polyclonal antibody. The p21 protein was not detectable in the GSE cells at basal levels, nor was it induced upon DNA damage, indicating of a complete inactivation of p53 transactivation activity. Figures lOB-C: Introduction of the GSE56 peptide at passage 20 resulted in accelerated growth rate (Figure 10C) and delay in replicative senescence, as indicated by the reduction of senescence-associated β-galactosidase (SA-β-Gal) staining (Figure 10B). Figure 1OD: Cell cycle analysis demonstrates that both DNA damage and replicative senescence resulted in a sharp p53 -dependent cell cycle arrest. Figure 1OE: QRT-PCR analyses of p21 mRNA levels demonstrated that ρ53 transactivation activity was significantly induced upon both DNA damage and replicative senescence, and was completely abolished by the introduction of GSE56. Figure 1OF: QRT-PCR analyses of cdc20, a p53 -repressed gene that participates in cell-cycle progression. To conclude, an isogenic pair of primary human cell cultures was created that display p53 -dependent application of p53 -activating stress.

FIG. 11 is a representation of the 'p53 -repressed miR cluster'. Primary WI-38 cells (Con) and WI-38 cells were infected with the p53 -inactivating peptide GSE56 (GSE) and analyzed for miRNA expression at early passage (Young), after doxorubicin treatment (0.2μg/ml, 24 hours) of early passage cells (Dox), and at the onset of replicative senescence (Old). FIGs. 12A-B are graphs illustrating that inactivation of ρ53 by GSE56 (GSE) or shRNA (p53i) in three different human primary fibroblasts delays replicative senescence and attenuates the repression of miRs and their hosts upon senescence. Figure 12 A: Growth curves for the human primary fibrablasts WI-38 and IMR90 and for the prostate cancer-associated fibroblasts (CAFs) PF179. Figure 12B: QRT-PCR for miR-106b and miR-17-5p, and their hosts MCM7 and cl3orf25, respectively, in early passage (Young) versus late passage (Old) fibroblasts. Data are represented as mean ± SD.

FIGs. 13A-D are graphs and photographs illustrating the establishment of the IMR90 and CAFs systems. Lung primary human fibroblasts IMR90 and prostate-cancer associated fibroblasts (CAFs) PFCAl 79 were infected with a retrovirus encoding for either a small hairpin RNA targeting p53 (ρ53i) or a control RNAi (Con), and grown until the onset of replicative senescence. Figure 13 A: A Western blot depicting p53 and p21 downregulation upon the stable expression of the p53 small hairpin RNA. Figure 13B: QRT-PCR analyses of p53 and p21 mRNA levels. Figure 13C: SA-β-Gal staining for late passage IMR90 and CAFs. Figure 13D: QRT-PCR analysis for the non-coding RNA BIC and its resident miRNA miR-155 in WI-38, CAFs and IMR90 cells. Samples were collected from early passage cultures (Young) and from late passage cultures (Old).

FIGs. 14A-B are graphs illustrating that E2F induces miR-106b/93/25. Figure 14A: WI-38 cells were stably infected with ER-E2F1 and treated with 4-OHT. QRT-PCR analyses demonstrated upregulation of a known E2F1 target, Cyclin E, as well as of host mRNAs and miRNAs representatives of the three paralogous polycistrons miR- 106b/93/25, miR-17-92 and miR-106a-92. Figure 14B: WI-38 cells were infected with the oncoprotein ElA or a control vector and QRT-PCR revealed upregulation of the genes described above. Data are represented as mean ± SD. FIGs. 15A-B are graphs illustrating that miR-106b/93/25 are induced by E2F in cancer cell lines. E2F induces the levels of miR-106b/93/25 polycistron in Hl 299 lung carcinoma cell line (Figure 15A) and U2OS osteosarcoma cell line (Figure 15B) and treated with 4-OHT for the indicated time periods. QRT-PCR analyses demonstrate upregulation of a known E2F1 target, Cyclin E; as well as of MCM7 and its resident miRNAs.

FIGs. 16A-D are graphs and photographs illustrating that MCM7 and miR-106b are repressed by Nutlin-activated p53 in an E2F-dependent manner. Figures 16A-B: WI-38 were infected with a retrovirus encoding for either a small hairpin RNA targeting p53 (p53i) or a control shRNA (Con) and treated with lOμM Nutlin-3 for 24 or 48 hours. QRT-PCR (Figure 16A) and Western blotting (Figure 16B) analyses demonstrate p53 stabilization that resulted in a robust activation of p21 and repression of E2F1 mRNA and protein levels. MCM7 and its resident miR-106b were repressed in a p53-dependent manner upon Nutin-3 treatment. Figures 16C-D. WI-38 cells were infected with ElA or a control vector and treated with lOμM Nutlin-3 for 24 hours. ElA elevated E2F transactivation activity, resulting in the induction of Cyclin E and E2F1 itself as well as of MCM7 and miR-106b. Nutlin treatment of the control-infected cells repressed transcription of E2F1 and its targets. ElA abolished this repression, indicating that the repression of E2F1 by p53 is necessary for the p53-depedent downregulation of MCM7 and miR-106b. GAPDH protein levels serve as loading controls in Figures 16B and D. QRT-PCR data are represented as mean ± SD.

FIGs. 17A-D are graphs and photographs illustrating that overexpression of miR- 106b/93/25 polycistron results in silencing of cell-cycle related genes. WI-38 primary fibroblasts and MCFlOA mammary cells were infected with a retrovirus encoding for either the genomic region that contains miR-106b/93/25 or an empty vector control. Figure 17A: Western blot analysis of reported and novel cell-cycle regulating targets of the overexpressed miRs. Overexpression of miR-106b/93/25 reduced the protein levels of E2F1, pRb, pl30, E2F1 and p21 in both cell types and of p57 in WI-38 cells. GAPDH and β-tubulin serve as loading controls. Figure 17B: QRT-PCR analysis of the mRNA levels of the genes presented in (Figure 17A). Values represent the fold change of each mRNA relative to the empty vector infected cells. Figure 17C: Expression pattern of predicted targets of at least five miRs from the 'p5S-repressed miR cluster'. mRNA expression levels are derived from WI-38 cells that underwent immortalization and gradual in-vitro transformation (the cell status is indicated below) as described in (Milyavsky et al, 2003, Cancer Res 63: 7147-7157). This expression pattern was found to be significantly coherent (EC score=0.14, EC p-value=5xlθ^"3). Figure 17D: Promoter analysis performed on the genes from (Figure 17C) (using AMADEUS) revealed enrichment for E2F motif (p- value=2.2xlθ^"13). Genes with E2F motif in their promoter are indicated in black in the bar on the right. FIG. 18 is a graphical representation of the miRNAs from the three paralogous polycistrons and their cell-cycle associated targets. Targeting of cell-cycle associated genes by miRNAs that belong to the miR- 17-93, miR-106a-93 and miR-106b-25 polycistrons as predicted by PicTar. Black areas indicate predicted targeting. FIGs. 19A-D are graphs and photographs illustrating that miR-106b/93/25 polycistron promotes proliferation. Overexpression of miR-106b/93/25 polycistron in WI- 38 cells promotes proliferation as evident from enhanced growth rate (Figure 19A), increased proportion of S-phase cells (Figure 19B), increased colony formation efficiency (Figure 19C), and decreased SA-β-Gal staining (Figure 19D).

FIG. 20 is a schematic model for the cell-cycle regulatory model comprising E2F, p53, miRs, and other cell-cycle regulators. Arrows correspond to direct transcriptional activation, while bar-headed lines represent direct or indirect inhibition mediated by the following mechanisms: post-transcription gene silencing (miRs and their targets), protein binding and inactivation (pocket proteins and E2F; as well as CDK inhibitors and CDKs, that in turn inhibit pocket proteins by phosphorylation).

DESCRIPTION OF SPECIFIC EMBODIMENTS OF THE INVENTION

Regulatory RNAs, constitutes a considerable portion of mammalian genomes, and these genes serve as key players in the regulatory network of living cells. Among these regulatory RNAs are the microRNAs, small RNAs that mediate posttranscriptional gene silencing through inhibition of protein production or degradation of mRNAs. So far little is known about the extent of regulation by miRs, and their potential cooperation with other regulatory layers in the network.

The present inventors set out to uncover local and global architectural features of the mammalian miR regulatory network. Using evolutionarily conserved potential binding sites of miRs in human targets, and conserved binding sites of TFs in promoters, two regulation networks were uncovered. The first depicts combinatorial interactions between pairs of miRs with many shared targets. The network reveals several levels of hierarchy, whereby a few miRs interact with many other lowly-connected miR partners (Figure 3A). The present inventors revealed hundreds of "target hubs" genes, many of which are transcription regulators (see Table 1, herein below), each potentially subject to massive regulation by dozens of miRs. The second network which was uncovered by the present inventors consists of miR-TF pairs that coregulate large sets of common targets (see Tables 9-11 herein below). Using bioinformatic analysis and data mining techniques, the present inventors discovered that the network consists of several recurring motifs. Most notably, in a significant fraction of the miR-TF coregulators the TF appears to regulate the miR, or to be regulated by the miR, forming a diversity of feed forward loops (FFL). Together these findings provide new insights on the architecture of the combined transcriptional-post transcriptional regulatory network.

One of the FFLs that came out of the present analysis is a composite loop in which the TF regulates the miR and the miR appears to regulate the TF. The circuit consists of the transcription factor E2F1 and the miR106b/93/25 polycistron. From this analysis, the present inventors predicted that E2F1 may regulate (and be regulated by) miRs in this cluster.

In order to corroborate the above prediction, the present inventors established isogenic cell cultures that differ in their p53 status and analyzed their miRNA profiles both under normal conditions as well as in contexts that involve p53 activation. The present inventors showed that the miRs of the miR106b/93/25 polycistron and paralogs thereof were part of a set of miRs that were transcriptionally repressed by the tumor suppressor, p53 in primary cells (Figure 11) and that this repression was E2Fl-mediated (Figures 14A- B, 15A-B and 16A-D). Whilst further reducing the present invention to practice, the present inventors showed that these microRNAs silence antiproliferative genes, which themselves are E2F1 targets (Figures 17 A-D). Finally, the present inventors showed that similarly to p53 inactivation, overexpression of representative microRNAs promotes proliferation and delays senescence, manifesting the detrimental phenotypic consequence of perturbations in this circuit (Figures 19 A-D). Together these findings position microRNAs as novel key players in the context of mammalian cell proliferation regulatory network. Thus, according to one aspect of the present invention, there is provided a method of identifying components of a biological pathway, the method comprising selecting a transcription factor and a microRNA pair which regulate a common gene, the transcription factor and the microRNA being the components of the biological pathway.

As used herein, the phrase "biological pathway" refers to a discrete cell function or process that is carried out by a gene product (RNA, protein or small molecule metabolite) or a subset of gene products, such as a signaling pathway, a metabolic pathway or a regulatory pathway including pathways involved in cell motility, cell morphology, cellular transformation, cell growth and death and cell communication. Exemplary biological pathways include anabolic, catabolic, enzymatic, biochemical and metabolic pathways as well as pathways involved in the production of cellular structures such as cell walls. Biological pathways that are usually required for proliferation of cells or microorganisms include, but are not limited to, cell division, DNA synthesis and replication, RNA synthesis (transcription), protein synthesis (translation), protein processing, protein transport, fatty acid biosynthesis, electron transport chains, cell wall synthesis, cell membrane production, synthesis and maintenance, and the like.

The biological pathway may be in any organism, including but not limited to prokaryotic and eukaryotic organisms (e.g. plants, animals including mammals and yeast). The phrase "components of a biological pathway", as used herein, refers to at least one transcription factor and one microRNA. It will be appreciated that after illucidation of the transcription factor and microRNA involved in the pathway, the method of this aspect of the present invention may also be used to indirectly uncover other components (including receptors, target genes, signaling molecules, second messenger molecules etc.) belonging to the same pathway, such components being known to interact or be involved with the uncovered transcription factor and microRNA.

As used herein, the phrase "transcription factor" refers to a polypeptide that binds to DNA and regulates gene transcription, and includes regulators that have a positive or a negative effect on transcription initiation or progression. Information concerning transcription factors (e.g. lists thereof and sequences of transcription factor binding domains) may be found on databases such as for example the protein lounge transcription factor database (wwwdotprotemloungedotcom) and the jaspar database (wwwdotaspardevdotgeneregdotnet) .

Target sites in the mRNA being regulated by the transcipriton factor may be in the 5' UTR or the 3 ' UTR region.

Exemplary classes of transcription factors are summarized herein below.

1. Superclass: Basic Domains (Basic-helix-loop-helix):

Class: Leucine zipper factors (bZIP) - including Family: AP-I (-like) components; includes (c-Fos/c-Jun), Family: CREB; Family: C/EBP-like factors; Family: bZIP / PAR; Family: Plant G-box binding factors; Family: ZIP only

Class: Helix-loop-helix factors (bHLH) - including Ubiquitous (class A) factors; Myogenic transcription factors (MyoD); Achaete-Scute; Tal/Twist/ Atonal/Hen. Class: Helix-loop-helix / leucine zipper factors (bHLH-ZIP) - including Ubiquitous bHLH-ZIP factors; includes USF (USFl, USF2); SREBP (SREBP); Family: Cell-cycle controlling factors; includes c-Myc.

Class: NF-I - including Family: NF-I (NFIQ. Class: RF-X - including RF-X (NFX2, NFX3, NFX5).

Class: bHSH.

2 Superclass: Zinc-coordinating DNA-binding domains:

Class: Cys4 zinc finger of nuclear receptor type - including Steroid hormone receptors, Thyroid hormone receptor-like factors Class: diverse Cy s4 zinc fingers including GATA-Factors

Class: Cys2His2 zinc finger domain - including Ubiquitous factors, includes TFIIIA, SpI, Developmental / cell cycle regulators; includes Krϋppel, Large factors with NF-6B-Iike binding properties.

Class: Cys6 cysteine-zinc cluster; Class: Zinc fingers of alternating composition.

3 Superclass: Helix-turn-helix

Class: Homeo domain - including Homeo domain only; includes Ubx; POU domain factors; includes Oct; Homeo domain with LIM region and homeo domain plus zinc finger motifs. Class: Paired box - including Paired plus homeo domain and Paired domain only.

Class: Fork head / winged helix - including Developmental regulators; includes forkhead; Tissue-specific regulators; Cell-cycle controlling factors and other regulators. Class: Heat Shock Factors - including HSF.

Class: Tryptophan clusters - including Myb, Ets-type, Interferon regulatory factors. Class: TEA domain - including TEA (TEADl, TEAD2, TEAD3, TEAD4).

4 Superclass: beta-Scaffold Factors with Minor Groove Contacts

Class: RHR (ReI homology region) - including Rel/ankyrin; NF-kappaB, ankyrin only and NF-AT (NFATCl, NFATC2, NFATC3).

Class: STAT - including STAT. Class : p53 - including p53.

Class: MADS box - including Regulators of differentiation; includes (Mef2) and Responders to external signals, SRF (serum response factor) (SRF).

Class: beta-Barrel alpha-helix transcription factors

Class: TATA binding proteins Class: Heteromeric CCAAT factors:

Class: Grainyhead

Class: Cold-shock domain factors

Class: Runt 0 Superclass: Other Transcription Factors - including: Copper fist proteins:

HMGI(Y) (HMGAl), HMGI(Y), Pocket domain, ElA-like factors, AP2/EREBP-related factors, AP2, EREBP, AP2/B3, ARF, ABI and RAV.

The term "microRNA" as used herein refers to a small RNA transcribed from genes encoding primary transcripts of various sizes. MicroRNAs have been identified in both animals and plants. The primary transcript (termed the "pri-miRNA") is processed through various nucleolytic steps to a shorter precursor miRNA, or "pre-miRNA." The pre-miRNA is present in a folded form so that the final (mature) miRNA is present in a duplex, the two strands being referred to as the miRNA (the strand that will eventually basepair with the target). The pre-miRNA is a substrate for a form of dicer that removes the miRNA duplex from the precursor, after which, similarly to siRNAs, the duplex can be taken into the

RNA-induced silencing complex (RISC) complex.

The resulting siRNA-like duplex, which may comprise mismatches, comprises the mature miRNA and a similar-sized fragment known as the miRNA*. The miRNA and miRNA* may be derived from opposing arms of the pri-miRNA and pre-miRNA. MiRNA* sequences may be found in libraries of cloned miRNAs but typically at lower frequency than the miRNAs.

Although initially present as a double-stranded species with miRNA*, the miRNA eventually becomes incorporated as a single-stranded RNA into the RISC. Various proteins can form the RISC, which can lead to variability in specifity for miRNA/miRNA* duplexes, binding site of the target gene, activity of miRNA (repress or activate), and which strand of the miRNA/miRNA* duplex is loaded in to the RISC.

When the miRNA strand of the miRNA:miRNA* duplex is loaded into the RISC, the miRNA* is typically removed and degraded. The strand of the miRNA:miRNA* duplex that is loaded into the RISC may be the strand whose 5' end is less tightly paired. In cases where both ends of the miRNA:miRNA* have roughly equivalent 5' pairing, both miRNA and miRNA* may have gene silencing activity.

The RISC acts to identify target nucleic acids based on high levels of complementarity between the miRNA and the mRNA, especially by nucleotides 2-8 of the miRNA. A number of studies have looked at the base-pairing requirement between miRNA and its mRNA target for achieving efficient inhibition of translation (reviewed by

Bartel 2004, Cell 116-281). In mammalian cells, the first 8 nucleotides of the miRNA are thought to be important (Doench & Sharp 2004 GenesDev 2004-504). However, other parts of the microRNA may also participate in mRNA binding. Moreover, sufficient base pairing at the 3' can compensate for insufficient pairing at the 5' (Brennecke et al, 2005 PLoS 3-e85). Computation studies, analyzing miRNA binding on whole genomes have suggested a specific role for bases 2-7 at the 5' of the miRNA in target binding but the role of the first nucleotide, found usually to be "A" was also recognized (Lewis et at 2005 Cell 120-15). Similarly, nucleotides 1-7 or 2-8 were used to identify and validate targets by Krek et al (2005, Nat Genet 37-495).

It will be appreciated that multiple miRNAs may regulate the same mRNA target by recognizing the same or multiple sites.

MiRNAs may direct the RISC to downregulate gene expression by either of two mechanisms: mRNA cleavage or translational repression. The miRNA may specify cleavage of the mRNA if the mRNA has a certain degree of complementarity to the miRNA. When a miRNA guides cleavage, the cut may be between the nucleotides pairing to residues 10 and 11 of the miRNA. Alternatively, the miRNA may repress translation if the miRNA does not have the requisite degree of complementarity to the miRNA. Translational repression may be more prevalent in animals since animals may have a lower degree of complementarity between the miRNA and binding site.

Of note, there may be variability in the 5' and 3' ends of any pair of miRNA and miRNA*. This variability may be due to variability in the enzymatic processing of Drosha and Dicer with respect to the site of cleavage. Variability at the 5' and 3' ends of miRNA and miRNA* may also be due to mismatches in the stem structures of the pri-miRNA and pre-miRNA. The mismatches of the stem strands may lead to a population of different hairpin structures. Variability in the stem structures may also lead to variability in the products of cleavage by Drosha and Dicer.

Micro-RNAs can be identified via various databases including for example the micro-RNA registry (wwwdotsangerdotacdotuk/Software/ Rfam/mirna/index.shtml) and sequences of miRNAs may be obtained from micrornadotsangerdotacdotuk/sequences/. Methods of identifying novel miRNAS are known in the art - see for example Bentwich et al., Nat Genet. 2005 Jul;37(7):766-70. Epub 2005 Jun 19.

As mentioned, the present method is effected by selecting a transcription factor and a microRNA pair which regulate a common gene. The term "regulate" as used herein refers to either up-regulation or down-regulation.

According to one embodiment of this aspect of the present invention the selection process is effected by analyzing gene sequences and determining if a known or predicted binding site for both a microRNA and a transcription factor is present. For miRNAs, the target sites in a mRNA of a transcribed gene may be in the 5' UTR, the 3' UTR or in the coding region. For TFs, the target sites are generally found upstream of the transcription start sites. Typically, TFs bind to the promoter region of a gene. The length of the gene sequence analyzed is selected such that it is long enough to include all potential targets, yet short enough to eliminate false positives. Accordingly, a typical length of DNA between about 5 kb and 10 kb upstream of the transcription start site (TSS) may be analyzed for transcription factor targets. Putative regulatory regions for miRNAs may be determined as further described in the Examples section below. For example, the sequence that lies about 10 kb upstream of the 5' most pre-miR in each miR cluster may be analyzed.

Various bioinformatic tools are available for analyzing gene sequences and determining if they comprise TF or miRNA binding sites (i.e. targets).

For TFs, tools such as TRANSFAC [Matys V, Fricke E, Geffers R, Gossling E, Haubrock M, et al. (2003) Nucleic Acids Res 31: 374-378]version 8.3, defined by the UCSC hgl7 genome assembly, in the tfbsConsSites (genomedotucscdotedu/) and tfbsConsFactors may be used. For miRNAs, tools such as TargetScan [Lewis BP, Burge CB, Bartel DP (2005)

Cell 120: 15-20; Lewis BP, Shih IH, Jones-Rhoades MW, Bartel DP, Burge CB (2003) Cell 115: 787-798] (wwwdottargetscandotorg) and PicTar [Krek A, Grun D, Poy MN, WoIf R, Rosenberg L, et al. (2005) Nat Genet 37: 495-500] (genomedotucscdotedu ) may be used. In order to ensure that false positive assignments of miRs or TFs to targets does not occur, targets may be selected based on evolutionary conservation in at least two species (e.g. human and mouse) or more - see the Examples section herein below.

Other methods may be used to increase the accuracy of results obtained using bioinformatic methods e.g. a noise tolerance analysis as described in the Examples section herein below.

According to one embodiment of this aspect of the present invention the selection process is effected by experimentally determining whether a specific TF or miRNA regulate a specific gene. It will be appreciated that this form of experimentation may be effected in place of the bioinformatic searches or as a corroboration of a result obtained from a bioinformatic search.

Thus, for example a method of determining whether a specific transcription factor binds to and regulates transcription of a gene may be effected by transfecting a polynucleotide encoding the promoter region of a particular gene linked to a detectable protein (i.e. reporter protein) into a cell - i.e. a reporter based assay. The method further comprises introducing the transcription factor into the cell (e.g. by transfection of an expression vector encoding the agent) and detecting the detectable protein whereby the amount of the detectable protein reflects the transcriptional activity of the promoter. It will be appreciated that the polynucleotide sequence of any protein that may be readily detected may be transcriptionally linked to the promoter. Thus for example, the protein may be a phosphorescent protein such as luciferase, a fluorescent protein such as green fluorescent protein, a chemiluminescent protein or may be a non-directly detectable protein for which an antibody is available for detection thereof. Cells for analyzing transcriptional activity are typically selected to ensure the presence of necessary cofactors and the absence factors which may potentially down-regulate the tested promoter.

A similar assay may be performed for analyzing whether a particular miRNA regulates a gene. Typically a level of reporter polypeptide is measured prior to and following transfection with a polynucleotide encoding the miRNA. A down-regulation of the reporter polypeptide indicates that the miRNA regulates the gene.

Using bioinformatic techniques, the present inventors have uncovered transcription factor and microRNA pairs that regulate common genes. These pairs are listed in Tables 5-8 of the Examples section herein below.

Pairs uncovered according to the teachings of the present invention can be stored as a database on storage devices, preferably in one or more computer-readable media, which contains information for each stored pair.

The information record may comprise, in fields or subfields, information relating to identification of the pair (including an identification code), date of identification of the pair, tools used to uncover the pair, the biological pathway with which it interacts and other potential components of that pathway.

The present inventors have also determined that as well as transcription factors and miRNAs regulating common genes other levels of crosstalk and regulation exist between these two regulatory agents. Thus for example, the present inventors have shown that subset of the pairs listed in Tables 5-8, comprise miRNAs whose gene contains a binding site for its paired transcription factor (see Figure 4; Tables 9-11 of the Examples section herein below). As another example, the present inventors have shown that a subset of the pairs listed in Tables 5-8, comprise transcription factors whose gene contains a binding site for its paired microRNA (see Figure 4; Table 9-11 of the Examples section herein below). In addition the present inventors have shown that a subset of the pairs listed in Tables 5-8 of the Examples section herein below, comprise transcription factors whose gene comprises a binding site for its paired microRNA and comprise miRNAs whose gene comprises a binding site for its paired transcription factor (see Figure 4; Tables 9-11 of the Examples section herein below). For example, the present inventors have shown that miR-93 comprises an E2F1 binding site and E2F1 comprises a miR-93 binding site and furthermore that miR-93 and E2F1 regulate common genes.

Another level of regulation uncovered by the present inventors is where the transcription factor (of the miRNA/transcription factor pair) regulates a transcription of an additional transcription factor, and the microRNA comprises a binding site for that additional transcription factor. The components involved in this type of regulation are listed in Tables 12-14 of the Examples section herein below.

It will be appreciated that characterization of components of biological pathways and their targets can potentially be utilized to treat diseases associated with such biological pathways. Thus, for example since miR-93 was shown to be regulated by E2F1 and vice versa, this miRNA (including paralogues thereof and other members of its polycistron) may be used to regulate pathways which are known to involve E2F1 (such as cell proliferation and apoptosis, controlling genes regulating S phase entry and DNA synthesis).

In corroboration of the above in silico result, the present inventors have shown that the above identified group of microRNAs silence antiproliferative genes, (Figures 17A-D) and overexpression therof promotes cell proliferation (Figures 19A-D).

Thus, according to another aspect of the present invention there is provided a method of treating a hyperproliferative disease in a subject. The method comprises administering to the subject a therapeutically effective amount of a polynucleotide agent capable of down-regulating at least one microRNA selected from the group consisting of miR-106b (SEQ ID NO: 29), miR-93 (SEQ ID NO: 30), miR-25 (SEQ ID NO: 31), miR- 17 (SEQ ID NO: 32), miR-18a (SEQ ID NO: 33), miR-19a (SEQ ID NO: 34), miR-20a (SEQ ID NO: 35), miR-19b-l (SEQ ID NO: 36), miR-92a-l (SEQ ID NO: 37), miR-106a (SEQ ID NO: 38), miR-18b (SEQ ID NO: 39), miR-20b (SEQ ID NO: 40), miR-19b-2 (SEQ ID NO: 41), miR-92-2* (SEQ ID NO: 42), hsa-miR-19b-l* (SEQ ID NO: 43) miR-

15b (SEQ ID NO: 44) and miR-16 (SEQ ID NO: 45) hsa-miR-92a-2* (SEQ ID NO: 46), and hsa-miR-92a-l* (SEQ ID NO: 47) into the subject.

As used herein, the term "treating" includes abrogating, substantially inhibiting, slowing or reversing the progression of a condition, substantially ameliorating clinical or aesthetical symptoms of a condition or substantially preventing the appearance of clinical or aesthetical symptoms of a condition.

As used herein, the term "subject" refers to an animal, preferably a mammal, most preferably a human being, including both young and old human beings of both sexes who suffer from or are predisposed to a pathology listed herein below. The subject according to this aspect of the present invention may suffer from a pathology associated with abnormal cell proliferation.

Hyperproliferative conditions that can be treated according to the present invention are, but not limited to, brain, skin (such as melanoma), bladder, gastric, pancreatic, breast, head, neck, oesophageal, prostate, colorectal, lung, renal, gynaecological (such as ovarian) or thyroid cancer; other epitheliomas; cysts in various organs; warts and wart-like tumours induced by virus infection; fibrosarcoma and its metastases. In another embodiment, the present invention relates to treatment of non-cancerous hyperproliferative disorder, such as benign hyperplasia of skin or prostate (e.g. benign prostatic hypertrophy), synovial hyperplasia in rheumatoid arthritis, inflammatory bowel disease, restenosis, atherosclerosis, thrombosis, scleroderma or fibrosis.

As used herein the term "cancer" refers to the presence of cells possessing characteristics typical of cancer-causing cells, for example, uncontrolled proliferation, loss of specialized functions, immortality, significant metastatic potential, significant increase in anti-apoptotic activity, rapid growth and proliferation rate, and certain characteristic morphology and cellular markers. In some circumstances, cancer cells will be in the form of a tumor; such cells may exist locally within an animal, or circulate in the blood stream as independent cells, for example, leukemic cells.

Specific examples of cancer which can be treated using the compositions of the present invention include, but are not limited to, adrenocortical carcinoma, hereditary; bladder cancer; breast cancer; breast cancer, ductal; breast cancer, invasive intraductal; breast cancer, sporadic; breast cancer, susceptibility to; breast cancer, type 4; breast cancer, type 4; breast cancer- 1; breast cancer-3; breast-ovarian cancer; Burkitt's lymphoma; cervical carcinoma; colorectal adenoma; colorectal cancer; colorectal cancer, hereditary nonpolyposis, type 1; colorectal cancer, hereditary nonpolyposis, type 2; colorectal cancer, hereditary nonpolyposis, type 3; colorectal cancer, hereditary nonpolyposis, type 6; colorectal cancer, hereditary nonpolyposis, type 7; dermatofibrosarcoma protuberans; endometrial carcinoma; esophageal cancer; gastric cancer, fibrosarcoma, glioblastoma multiforme; glomus tumors, multiple; hepatoblastoma; hepatocellular cancer; hepatocellular carcinoma; leukemia, acute lymphoblastic; leukemia, acute myeloid; leukemia, acute myeloid, with eosinophilia; leukemia, acute nonlymphocytic; leukemia, chronic myeloid; Li-Fraumeni syndrome; liposarcoma, lung cancer; lung cancer, small cell; lymphoma, non-Hodgkin's; lynch cancer family syndrome II; male germ cell tumor; mast cell leukemia; medullary thyroid; medulloblastoma; melanoma, meningioma; multiple endocrine neoplasia; myeloid malignancy, predisposition to; myxosarcoma, neuroblastoma; osteosarcoma; ovarian cancer; ovarian cancer, serous; ovarian carcinoma; ovarian sex cord tumors; pancreatic cancer; pancreatic endocrine tumors; paraganglioma, familial nonchromaffin; pilomatricoma; pituitary tumor, invasive; prostate adenocarcinoma; prostate cancer; renal cell carcinoma, papillary, familial and sporadic; retinoblastoma; rhabdoid predisposition syndrome, familial; rhabdoid tumors; rhabdomyosarcoma; small-cell cancer of lung; soft tissue sarcoma, squamous cell carcinoma, head and neck; T-cell acute lymphoblastic leukemia; Turcot syndrome with glioblastoma; tylosis with esophageal cancer; uterine cervix carcinoma, Wilms' tumor, type 2; and Wilms' tumor, type 1, and the like.

Polynucleotide agents capable of downregulating the miRNAs of the present invention (also referred to herein as anti-miRNAs) typically comprise a sequence that is capable of blocking the activity of a miRNA or miRNA*, such as by binding to the pri- miRNA, pre-miRNA, miRNA or miRNA* (e.g. antisense or RNA silencing), or by binding to the target binding site. The sequence of the anti-miRNA may comprise (a) at least 5 nucleotides that are substantially identical or complimentary to the 5' of a miRNA and at least 5-12 nucleotides that are substantially complimentary to the flanking regions of the target site from the 5' end of the miRNA, or (b) at least 5-12 nucleotides that are substantially identical or complimentary to the 3' of a miRNA and at least 5 nucleotide that are substantially complimentary to the flanking region of the target site from the 3' end of the miRNA.

The term "polynucleotide" refers to a single-stranded or double-stranded oligomer or polymer of ribonucleic acid (RNA), deoxyribonucleic acid (DNA) or mimetics thereof. This term includes polynucleotides and/or oligonucleotides derived from naturally occurring nucleic acids molecules (e.g., RNA or DNA)₅ synthetic polynucleotide and/or oligonucleotide molecules composed of naturally occurring bases, sugars, and covalent internucleoside linkages (e.g., backbone), as well as synthetic polynucleotides and/or oligonucleotides having non-naturally occurring portions, which function similarly to respective naturally occurring portions.

The length of the polynucleotide of the present invention is optionally of 200 nucleotides or less, optionally 175 nucleotides or less, optionally 150 nucletoides or less, optionally 125 nucleotides or less, optionally of 100 nucleotides or less, optionally of 90 nucleotides or less, optionally 80 nucleotides or less, optionally 70 nucleotides or less, optionally 60 nucleotides or less, optionally 50 nucleotides or less, optionally 40 nucleotides or less, optionally 30 nucleotides or less, e.g., 29 nucleotides, 28 nucleotides, 27 nucleotides, 26 nucleotides, 25 nucleotides, 24 nucleotides, 23 nucleotides, 22 nucleotides, 21 nucleotides, 20 nucleotides, 19 nucleotides, 18 nucleotides, 17 nucleotides, 16 nucleotides, 15 nucleotides, optionally between 12 and 24 nucleotides, optionally between 5-15, optionally, between 5-25, or optionally about 20-25 nucleotides.

The polynucleotides (including oligonucleotides) designed according to the teachings of the present invention can be generated according to any oligonucleotide synthesis method known in the art, including both enzymatic syntheses or solid-phase syntheses. Equipment and reagents for executing solid-phase synthesis are commercially available from, for example, Applied Biosystems. Any other means for such synthesis may also be employed; the actual synthesis of the oligonucleotides is well within the capabilities of one skilled in the art and can be accomplished via established methodologies as detailed in, for example: Sambrook, J. and Russell, D. W. (2001), "Molecular Cloning: A Laboratory Manual"; Ausubel, R. M. et al., eds. (1994, 1989), "Current Protocols in Molecular Biology," Volumes I-III, John Wiley & Sons, Baltimore, Maryland; Perbal, B. (1988), "A Practical Guide to Molecular Cloning," John Wiley & Sons, New York; and Gait, M. J., ed. (1984), "Oligonucleotide Synthesis"; utilizing solid- phase chemistry, e.g. cyanoethyl phosphoramidite followed by deprotection, desalting, and purification by, for example, an automated trityl-on method or HPLC. It will be appreciated that a polynucleotide comprising an RNA molecule can be also generated using an expression vector as is further described hereinbelow.

Preferably, the polynucleotide of the present invention is a modified polynucleotide. Polynucleotides can be modified using various methods known in the art. For example, the oligonucleotides or polynucleotides of the present invention may comprise heterocylic nucleosides consisting of purines and the pyrimidines bases, bonded in a 3'-to-5' phosphodiester linkage.

Preferably used oligonucleotides or polynucleotides are those modified either in backbone, internucleoside linkages, or bases, as is broadly described hereinunder. Specific examples of preferred oligonucleotides or polynucleotides useful according to this aspect of the present invention include oligonucleotides or polynucleotides containing modified backbones or non-natural internucleoside linkages.

Oligonucleotides or polynucleotides having modified backbones include those that retain a phosphorus atom in the backbone, as disclosed in U.S. Pat. Nos.: 4,469,863; 4,476,301; 5,023,243; 5,177,196; 5,188,897; 5,264,423; 5,276,019; 5,278,302; 5,286,717; 5,321,131;

5,399,676; 5,405,939; 5,453,496; 5,455,233; 5,466,677; 5,476,925; 5,519,126; 5,536,821;

5,541,306; 5,550,111; 5,563,253; 5,571,799; 5,587,361; and 5,625,050.

Preferred modified oligonucleotide or polynucleotide backbones include, for example: phosphorothioates; chiral phosphorothioates; phosphorodithioates; phosphotriesters; aminoalkyl phosphotriesters; methyl and other alkyl phosphonates, including 3'-alkylene phosphonates and chiral phosphonates; phosphinates; phosphoramidates, including 3 '-amino phosphoramidate and aminoalkylphosphoramidates; thionophosphoramidates; thionoalkylphosphonates; thionoalkylphosphotriesters; and boranophosphates having normal 3'-5' linkages, 2'-5' linked analogues of these, and those having inverted polarity wherein the adjacent pairs of nucleoside units are linked 3'-5' to 5'-

3' or 2'-5^f to 5'-2'. Various salts, mixed salts, and free acid forms of the above modifications can also be used.

Alternatively, modified oligonucleotide or polynucleotide backbones that do not include a phosphorus atom therein have backbones that are formed by short-chain alkyl or cycloalkyl internucleoside linkages, mixed heteroatom and alkyl or cycloalkyl internucleoside linkages, or one or more short-chain heteroatomic or heterocyclic internucleoside linkages. These include those having morpholino linkages (formed in part from the sugar portion of a nucleoside); siloxane backbones; sulfide, sulfoxide, and sulfone backbones; formacetyl and thioformacetyl backbones; methylene formacetyl and thioformacetyl backbones; alkene-containing backbones; sulfamate backbones; methyleneimino and methylenehydrazino backbones; sulfonate and sulfonamide backbones; amide backbones; and others having mixed N, O, S and CH₂ component parts, as disclosed in U.S. Pat. Nos.: 5,034,506; 5,166,315; 5,185,444; 5,214,134; 5,216,141;

5,235,033; 5,264,562; 5,264,564; 5,405,938; 5,434,257; 5,466,677; 5,470,967; 5,489,677; 5,541,307; 5,561,225; 5,596,086; 5,602,240; 5,610,289; 5,602,240; 5,608,046; 5,610,289; 5,618,704; 5,623,070; 5,663,312; 5,633,360; 5,677,437; and 5,677,439.

Other oligonucleotides or polynucleotides which may be used according to the present invention are those modified in both sugar and the internucleoside linkage, i.e., the backbone of the nucleotide units is replaced with novel groups. The base units are maintained for complementation with the appropriate polynucleotide target. An example of such an oligonucleotide mimetic includes a peptide nucleic acid (PNA). A PNA oligonucleotide refers to an oligonucleotide where the sugar-backbone is replaced with an amide-containing backbone, in particular an aminoethylglycine backbone. The bases are retained and are bound directly or indirectly to aza-nitrogen atoms of the amide portion of the backbone. United States patents that teach the preparation of PNA compounds include, but are not limited to, U.S. Pat. Nos. 5,539,082; 5,714,331; and 5,719,262; each of which is herein incorporated by reference. Other backbone modifications which may be used in the present invention are disclosed in U.S. Pat. No. 6,303,374. Oligonucleotides or polynucleotides of the present invention may also include base modifications or substitutions. As used herein, "unmodified" or "natural" bases include the purine bases adenine (A) and guanine (G) and the pyrimidine bases thymine (T), cytosine (C), and uracil (U). "Modified" bases include but are not limited to other synthetic and natural bases, such as: 5-methylcytosine (5-me-C); 5-hydroxymethyl cytosine; xanthine; hypoxanthine; 2-aminoadenine; 6-methyl and other alkyl derivatives of adenine and guanine; 2-propyl and other alkyl derivatives of adenine and guanine; 2-thiouracil, 2- thiothymine, and 2-thiocytosine; 5-halouracil and cytosine; 5-propynyl uracil and cytosine; 6-azo uracil, cytosine, and thymine; 5-uracil (pseudouracil); 4-thiouracil; 8-halo, 8-amino, 8-thiol, 8-thioalkyl, 8-hydroxyl, and other 8-substituted adenines and guanines; 5-halo, particularly 5-bromo, 5-trifluoromethyl, and other 5-substituted uracils and cytosines; 7- methylguanine and 7-methyladenine; 8-azaguanine and 8-azaadenine; 7-deazaguanine and 7-deazaadenine; and 3-deazaguanine and 3-deazaadenine. Additional modified bases include those disclosed in: U.S. Pat. No. 3,687,808; Kroschwitz, J. L, ed. (1990),"The Concise Encyclopedia Of Polymer Science And Engineering," pages 858-859, John Wiley & Sons; Englisch et al. (1991), "Angewandte Chemie," International Edition, 30, 613; and Sanghvi, Y. S., "Antisense Research and Applications," Chapter 15, pages 289-302, S. T. Crooke and B. Lebleu, eds., CRC Press, 1993. Such modified bases are particularly useful for increasing the binding affinity of the oligomeric compounds of the invention. These include 5-substituted pyrirnidines, 6-azapyrimidines, and N-2, N-6, and O-6-substituted purines, including 2-aminopropyladenine, 5-propynyluracil, and 5-propynylcytosine. 5- methylcytosine substitutions have been shown to increase nucleic acid duplex stability by 0.6-1.2⁰C (Sanghvi, Y. S. et al. (1993), "Antisense Research and Applications," pages 276- 278, CRC Press, Boca Raton), and are presently preferred base substitutions, even more particularly when combined with 2'-O-methoxyethyl sugar modifications.

Polynucleotide agents capable of down-regulating miRNAs are known in the art — see for example Weiler et al., Gene Therapy (2006) 13, 496-502, Davis et al., Nucleic Acids Res. 2006; 34(8): 2294-2304, USPTO Application No: 20070287179.

The present invention also contemplates treating degenerative diseases by administration of the miRNAs of the present invention.

Since as is mentioned hereinabove, micro-RNAs are processed molecules derived from specific precursors (i.e., pre-miRNA), upregulation of a specific miRNA function can be effected using a specific miRNA precursor molecule.

Pre-miR designs exist to all of the known miRNAs listed in the miRNA Registry and can be readily designed for any research.

The phrase "degenerative disease" refers to a disease or disorder resulting from a decrease in cellular proliferation.

Exemplary degenerative diseases neurodegenerative diseases, including but not limited to Parkinson's, Multiple Sclerosis, Huntington's disease, action tremors and tardive dyskinesia, panic, anxiety, depression, alcoholism, insomnia and manic behavior, Alzheimer's, ALS and epilepsy.

It will be appreciated that the subject can be treated in vivo (i.e., inside the organism) or ex vivo (e.g., in a tissue culture). In case the cells are treated ex vivo, the method preferably includes a step of administering such cells back to the individual (ex vivo cell therapy). In vivo and ex vivo therapies are further discussed hereinbelow.

As mentioned hereinabove, the polynucleotides of the present invention (e.g., an RNA molecule such as those set forth by SEQ ID NOs: 1-16) can be generated using an expression vector.

To express an exogenous polynucleotide (i.e., to produce an RNA molecule) in mammalian cells, a nucleic acid sequence encoding the polynucleotide of the present invention (e.g., SEQ ID NO:1-16) is typically ligated into a nucleic acid construct suitable for mammalian cell expression. Such a nucleic acid construct includes a promoter sequence for directing transcription of the polynucleotide sequence in the cell in a constitutive or inducible manner. Constitutive promoters suitable for use with the present invention are promoter sequences which are active under most environmental conditions and most types of cells such as the cytomegalovirus (CMV) and Rous sarcoma virus (RSV). Inducible promoters suitable for use with the present invention include for example the tetracycline-inducible promoter (Zabala M, et al., Cancer Res. 2004, 64(8): 2799-804).

The nucleic acid construct (also referred to herein as an "expression vector") of the present invention includes additional sequences which render this vector suitable for replication and integration in prokaryotes, eukaryotes, or preferably both (e.g., shuttle vectors). In addition, typical cloning vectors may also contain a transcription and translation initiation sequence, transcription and translation terminator and a polyadenylation signal.

Eukaryotic promoters typically contain two types of recognition sequences, the TATA box and upstream promoter elements. The TATA box, located 25-30 base pairs upstream of the transcription initiation site, is thought to be involved in directing RNA polymerase to begin RNA synthesis. The other upstream promoter elements determine the rate at which transcription is initiated.

Preferably, the promoter utilized by the nucleic acid construct of the present invention is active in the specific cell population transformed. Examples of cell type- specific and/or tissue-specific promoters include promoters such as albumin that is liver specific [Pinkert et al., (1987) Genes Dev. 1:268-277], lymphoid specific promoters [Calame et al., (1988) Adv. Immunol. 43:235-275]; in particular promoters of T-cell receptors [Winoto et al., (1989) EMBO J. 8:729-733] and immunoglobulins; [Banerji et al. (1983) Cell 33729-740], neuron-specific promoters such as the neurofilament promoter [Byrne et al. (1989) Proc. Natl. Acad. Sci. USA 86:5473-5477], pancreas-specific promoters [Edlunch et al. (1985) Science 230:912-916] or mammary gland-specific promoters such as the milk whey promoter (U.S. Pat. No. 4,873,316 and European Application Publication No. 264,166).

Enhancer elements can stimulate transcription up to 1,000 fold from linked homologous or heterologous promoters. Enhancers are active when placed downstream or upstream from the transcription initiation site. Many enhancer elements derived from viruses have a broad host range and are active in a variety of tissues. For example, the SV40 early gene enhancer is suitable for many cell types. Other enhancer/promoter combinations that are suitable for the present invention include those derived from polyoma virus, human or murine cytomegalovirus (CMV), the long term repeat from various retroviruses such as murine leukemia virus, murine or Rous sarcoma virus and

HIV. See, Enhancers and Eukaryotic Expression, Cold Spring Harbor Press, Cold Spring

Harbor, N. Y. 1983, which is incorporated herein by reference.

In the construction of the expression vector, the promoter is preferably positioned approximately the same distance from the heterologous transcription start site as it is from the transcription start site in its natural setting. As is known in the art, however, some variation in this distance can be accommodated without loss of promoter function.

Polyadenylation sequences can also be added to the expression vector in order to increase RNA stability [Soreq et al., 1974; J. MoI Biol. 88: 233-45). Two distinct sequence elements are required for accurate and efficient polyadenylation: GU or U rich sequences located downstream from the polyadenylation site and a highly conserved sequence of six nucleotides, AAUAAA, located 11-30 nucleotides upstream. Termination and polyadenylation signals that are suitable for the present invention include those derived from SV40. In addition to the elements already described, the expression vector of the present invention may typically contain other specialized elements intended to increase the level of expression of cloned nucleic acids or to facilitate the identification of cells that carry the recombinant DNA. For example, a number of animal viruses contain DNA sequences that promote the extra chromosomal replication of the viral genome in permissive cell types. Plasmids bearing these viral replicons are replicated episomally as long as the appropriate factors are provided by genes either carried on the plasmid or with the genome of the host cell.

The vector may or may not include a eukaryotic replicon. If a eukaryotic replicon is present, then the vector is amplifiable in eukaryotic cells using the appropriate selectable marker. If the vector does not comprise a eukaryotic replicon, no episomal amplification is possible. Instead, the recombinant DNA integrates into the genome of the engineered cell, where the promoter directs expression of the desired nucleic acid.

Examples of mammalian expression vectors include, but are not limited to, ρcDNA3, ρcDNA3.1 (+/-), pGL3, pZeoSV2(+/-), ρSecTag2, pDisplay, pEF/myc/cyto, pCMV/myc/cyto, pCR3.1, pSinRep5, DH26S, DHBB, pNMTl, pNMT41, pNMT81, which are available from Invitrogen, pCI which is available from Promega, pMbac, pPbac, pBK-RSV and pBK-CMV which are available from Strategene, pTRES which is available from Clontech, and their derivatives. Expression vectors containing regulatory elements from eukaryotic viruses such as retroviruses can be also used. SV40 vectors include pSVT7 and pMT2. Vectors derived from bovine papilloma virus include pBV- IMTHA, and vectors derived from Epstein Bar virus include pHEBO, and p2O5. Other exemplary vectors include pMSG, pAV009/A⁺, pMTO10/A⁺, pMAMneo-5, baculovirus pDSVE, and any other vector allowing expression of proteins under the direction of the SV-40 early promoter, SV-40 later promoter, metallothionein promoter, murine mammary tumor virus promoter, Rous sarcoma virus promoter, polyhedrin promoter, or other promoters shown effective for expression in eukaryotic cells. As described above, viruses are very specialized infectious agents that have evolved, in many cases, to elude host defense mechanisms. Typically, viruses infect and propagate in specific cell types. The targeting specificity of viral vectors utilizes its natural specificity to specifically target predetermined cell types and thereby introduce a recombinant gene into the infected cell. Thus, the type of vector used by the present invention will depend on the cell type transformed. The ability to select suitable vectors according to the cell type transformed is well within the capabilities of the ordinary skilled artisan and as such no general description of selection consideration is provided herein. For example, bone marrow cells can be targeted using the human T cell leukemia virus type I (HTLV-I) and kidney cells may be targeted using the heterologous promoter present in the baculovirus Autographa californica nucleopolyhedrovirus (AcMNPV) as described in Liang CY et al., 2004 (Arch Virol. 149: 51-60).

Recombinant viral vectors are useful for in vivo expression of the polynucleotide of the present invention since they offer advantages such as lateral infection and targeting specificity. Lateral infection is inherent in the life cycle of, for example, retrovirus and is the process by which a single infected cell produces many progeny virions that bud off and infect neighboring cells. The result is that a large area becomes rapidly infected, most of which was not initially infected by the original viral particles. This is in contrast to vertical-type of infection in which the infectious agent spreads only through daughter progeny. Viral vectors can also be produced that are unable to spread laterally. This characteristic can be useful if the desired purpose is to introduce a specified gene into only a localized number of targeted cells.

Various methods can be used to introduce the expression vector of the present invention into cells. Such methods are generally described in Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Springs Harbor Laboratory, New York (1989, 1992), in Ausubel et al., Current Protocols in Molecular Biology, John Wiley and Sons, Baltimore, Md. (1989), Chang et al., Somatic Gene Therapy, CRC Press, Ann Arbor, Mich. (1995), Vega et al., Gene Targeting, CRC Press, Ann Arbor Mich. (1995), Vectors: A Survey of Molecular Cloning Vectors and Their Uses, Butterworths, Boston Mass. (1988) and Gilboa et at. [Biotechniques 4 (6): 504-512, 1986] and include, for example, stable or transient transfection, lipofection, electroporation and infection with recombinant viral vectors. In addition, see U.S. Pat. Nos. 5,464,764 and 5,487,992 for positive-negative selection methods.

Introduction of nucleic acids by viral infection offers several advantages over other methods such as lipofection and electroporation, since higher transfection efficiency can be obtained due to the infectious nature of viruses.

Currently preferred in vivo nucleic acid transfer techniques include transfection with viral or non-viral constructs, such as adenovirus, lentivirus, Herpes simplex I virus, or adeno-associated virus (AAV) and lipid-based systems. Useful lipids for lipid-mediated transfer of the gene are, for example, DOTMA, DOPE, and DC-Choi [Tonkinson et al., Cancer Investigation, 14(1): 54-65 (1996)]. The most preferred constructs for use in gene therapy are viruses, most preferably adenoviruses, AAV, lentiviruses, or retroviruses. A viral construct such as a retroviral construct includes at least one transcriptional promoter/enhancer or locus-defining element(s), or other elements that control gene expression by other means such as alternate splicing, nuclear RNA export, or post- translational modification of messenger. Such vector constructs also include a packaging signal, long terminal repeats (LTRs) or portions thereof, and positive and negative strand primer binding sites appropriate to the virus used, unless it is already present in the viral construct. Other vectors can be used that are non- viral, such as cationic lipids, polylysine, and dendrimers.

As mentioned hereinabove, a variety of prokaryotic or eukaryotic cells can be used as host-expression systems to express the polynucleotides of the present invention. These include, but are not limited to, microorganisms, such as bacteria transformed with a recombinant bacteriophage DNA, plasmid DNA or cosmid DNA expression vector containing the coding sequence; yeast transformed with recombinant yeast expression vectors containing the coding sequence; plant cell systems infected with recombinant virus expression vectors (e.g., cauliflower mosaic virus, CaMV; tobacco mosaic virus, TMV) or transformed with recombinant plasmid expression vectors, such as Ti plasmid, containing the coding sequence. Mammalian expression systems can also be used to express the polynucleotides of the present invention.

Examples of bacterial constructs include the pET series of E. coli expression vectors [Studier et al. (1990) Methods in Enzymol. 185:60-89). In yeast, a number of vectors containing constitutive or inducible promoters can be used, as disclosed in U.S. Pat. Application No: 5,932,447. Alternatively, vectors can be used which promote integration of foreign DNA sequences into the yeast chromosome.

In cases where plant expression vectors are used, the expression of the coding sequence can be driven by a number of promoters. For example, viral promoters such as the 35S RNA and 19S RNA promoters of CaMV [Brisson et al. (1984) Nature 310:511- 514], or the coat protein promoter to TMV [Takamatsu et al. (1987) EMBO J. 6:307-311] can be used. Alternatively, plant promoters such as the small subunit of RUBISCO [Coruzzi et al. (1984) EMBO J. 3:1671-1680 and Brogli et al., (1984) Science 224:838- 843] or heat shock promoters, e.g., soybean hspl7.5-E or hspl7.3-B [Gurley et al. (1986) MoI. Cell. Biol. 6:559-565] can be used. These constructs can be introduced into plant cells using Ti plasmid, Ri plasmid, plant viral vectors, direct DNA transformation, microinjection, electroporation and other techniques well known to the skilled artisan. See, for example, Weissbach & Weissbach, 1988, Methods for Plant Molecular Biology, Academic Press, NY, Section VIII, pp 421-463. Other expression systems such as insects and mammalian host cell systems which are well known in the art and are further described hereinbelow can also be used by the present invention.

For ex vivo therapy, cells are preferably treated with the polynucleotides of the present invention (e.g., anti micro-RNA or microRNA), following which they are administered to the subject (individual) which is in need thereof.

Administration of the ex vivo treated cells of the present invention can be effected using any suitable route of introduction, such as intravenous, intraperitoneal, intra-kidney, intra-gastrointestinal track, subcutaneous, transcutaneous, intramuscular, intracutaneous, intrathecal, epidural, and rectal. According to presently preferred embodiments, the ex vivo treated cells of the present invention may be introduced to the individual using intravenous, intra-kidney, intra-gastrointestinal track, and/or intraperitoneal administration.

The cells used for ex vivo treatment according to the present invention can be derived from either autologous sources, such as self bone marrow cells, or from allogeneic sources, such as bone marrow or other cells derived from non-autologous sources. Since non-autologous cells are likely to induce an immune reaction when administered to the body, several approaches have been developed to reduce the likelihood of rejection of non- autologous cells. These include either suppressing the recipient immune system or encapsulating the non-autologous cells or tissues in immunoisolating, semipermeable membranes before transplantation.

Encapsulation techniques are generally classified as microencapsulation, involving small spherical vehicles, and macroencapsulation, involving larger flat-sheet and hollow- fiber membranes (Uludag, H. et al. (2000). Technology of mammalian cell encapsulation. Adv Drug Deliv Rev 42, 29-64). Methods of preparing microcapsules are known in the art and include for example those disclosed in: Lu, M. Z. et al. (2000). Cell encapsulation with alginate and alpha- phenoxycinnamylidene-acetylated poly(allylamine). Biotechnol Bioeng 70, 479-483; Chang, T. M. and Prakash, S. (2001) Procedures for microencapsulation of enzymes, cells and genetically engineered microorganisms. MoI Biotechnol 17, 249-260; and Lu, M. Z., et al. (2000). A novel cell encapsulation method using photosensitive poly(allylamine alpha-cyaiiocinnamylideneacetate). J Microencapsul 17, 245-521.

For example, microcapsules are prepared using modified collagen in a complex with a ter-polymer shell of 2-hydroxyethyl methylacrylate (HEMA), methacrylic acid (MAA), and methyl methacrylate (MMA), resulting in a capsule thickness of 2-5 μm. Such microcapsules can be further encapsulated with an additional 2-5 μm of ter-polymer shells in order to impart a negatively charged smooth surface and to minimize plasma protein absorption (Chia, S. M. et al. (2002). Multi-layered microcapsules for cell encapsulation. Biomaterials 23, 849-856).

Other microcapsules are based on alginate, a marine polysaccharide (Sambanis, A. (2003). Encapsulated islets in diabetes treatment. Diabetes Thechnol Ther 5, 665-668), or its derivatives. For example, microcapsules can be prepared by the polyelectrolyte complexation between the polyanions sodium alginate and sodium cellulose sulphate and the polycation pory(methylene-co-guanidme) hydrochloride in the presence of calcium chloride. It will be appreciated that cell encapsulation is improved when smaller capsules are used. Thus, for instance, the quality control, mechanical stability, diffusion properties, and in vitro activities of encapsulated cells improved when the capsule size was reduced from 1 mm to 400 μm (Canaple, L. et al. (2002). Improving cell encapsulation through size control. J Biomater Sci Polym Ed 13, 783-96). Moreover, nanoporous biocapsules with well-controlled pore size as small as 7 nm, tailored surface chemistries, and precise microarchitectures were found to successfully immunoisolate microenvironments for cells (See: Williams, D. (1999). Small is beautiful: microparticle and nanoparticle technology in medical devices. Med Device Technol 10, 6-9; and Desai, T. A. (2002). Microfabrication technology for pancreatic cell encapsulation. Expert Opin Biol Ther 2, 633-646).

The polynucleotides and/or the expression vectors of the present invention can be administered to the individual per se or as part of a pharmaceutical composition where it is mixed with suitable carriers or excipients. As used herein, a "pharmaceutical composition" refers to a preparation of one or more of the active ingredients described herein with other chemical components such as physiologically suitable carriers and excipients. The purpose of a pharmaceutical composition is to facilitate administration of a compound to an organism.

As used herein, the term "active ingredient" refers to the agent, the polynucleotide and/or the expression vector of the present invention accountable for the intended biological effect.

Hereinafter, the phrases "physiologically acceptable carrier" and "pharmaceutically acceptable carrier," which may be used interchangeably, refer to a carrier or a diluent that does not cause significant irritation to an organism and does not abrogate the biological activity and properties of the administered compound. An adjuvant is included under these phrases.

Herein, the term "excipient" refers to an inert substance added to a pharmaceutical composition to further facilitate administration of an active ingredient. Examples, without limitation, of excipients include calcium carbonate, calcium phosphate, various sugars and types of starch, cellulose derivatives, gelatin, vegetable oils, and polyethylene glycols.

Techniques for formulation and administration of drugs may be found in the latest edition of "Remington's Pharmaceutical Sciences," Mack Publishing Co., Easton, PA, which is herein fully incorporated by reference.

Suitable routes of administration may, for example, include oral, rectal, transmucosal, especially transnasal, intestinal, or parenteral delivery, including intramuscular, subcutaneous, and intramedullary injections, as well as intrathecal, direct intraventricular, intravenous, inrtaperitoneal, intracardiac, intranasal, or intraocular injections. Alternately, one may administer the pharmaceutical composition in a local rather than systemic manner, for example, via injection of the pharmaceutical composition directly into a tissue region of a patient.

Pharmaceutical compositions of the present invention may be manufactured by processes well known in the art, e.g., by means of conventional mixing, dissolving, granulating, dragee-making, levigating, emulsifying, encapsulating, entrapping, or lyophilizing processes.

Pharmaceutical compositions for use in accordance with the present invention thus may be formulated in conventional manner using one or more physiologically acceptable carriers comprising excipients and auxiliaries, which facilitate processing of the active ingredients into preparations that can be used pharmaceutically. Proper formulation is dependent upon the route of administration chosen.

For injection, the active ingredients of the pharmaceutical composition may be formulated in aqueous solutions, preferably in physiologically compatible buffers such as Hank's solution, Ringer's solution, or physiological salt buffer. For transmucosal administration, penetrants appropriate to the barrier to be permeated are used in the formulation. Such penetrants are generally known in the art.

For oral administration, the pharmaceutical composition can be formulated readily by combining the active compounds with pharmaceutically acceptable carriers well known in the art. Such carriers enable the pharmaceutical composition to be formulated as tablets, pills, dragees, capsules, liquids, gels, syrups, slurries, suspensions, and the like, for oral ingestion by a patient. Pharmacological preparations for oral use can be made using a solid excipient, optionally grinding the resulting mixture, and processing the mixture of granules, after adding suitable auxiliaries as desired, to obtain tablets or dragee cores. Suitable excipients are, in particular, fillers such as sugars, including lactose, sucrose, mannitol, or sorbitol; cellulose preparations such as, for example, maize starch, wheat starch, rice starch, potato starch, gelatin, gum tragacanth, methyl cellulose, hydroxypropylmethyl-cellulose, and sodium carbomethylcellulose; and/or physiologically acceptable polymers such as polyvinylpyrrolidone (PVP). If desired, disintegrating agents, such as cross-linked polyvinyl pyrrolidone, agar, or alginic acid or a salt thereof, such as sodium alginate, may be added.

Dragee cores are provided with suitable coatings. For this purpose, concentrated sugar solutions may be used which may optionally contain gum arabic, talc, polyvinyl pyrrolidone, carbopol gel, polyethylene glycol, titanium dioxide, lacquer solutions, and suitable organic solvents or solvent mixtures. Dyestuffs or pigments may be added to the tablets or dragee coatings for identification or to characterize different combinations of active compound doses.

Pharmaceutical compositions that can be used orally include push-fit capsules made of gelatin, as well as soft, sealed capsules made of gelatin and a plasticizer, such as glycerol or sorbitol. The push-fit capsules may contain the active ingredients in admixture with filler such as lactose, binders such as starches, lubricants such as talc or magnesium stearate, and, optionally, stabilizers. In soft capsules, the active ingredients may be dissolved or suspended in suitable liquids, such as fatty oils, liquid paraffin, or liquid polyethylene glycols. In addition, stabilizers may be added. AU formulations for oral administration should be in dosages suitable for the chosen route of administration.

For buccal administration, the compositions may take the form of tablets or lozenges formulated in conventional manner.

For administration by nasal inhalation, the active ingredients for use according to the present invention are conveniently delivered in the form of an aerosol spray presentation from a pressurized pack or a nebulizer with the use of a suitable propellant, e.g., dichlorodifluoromethane, trichlorofluoromethane, dichloro-tetrafluoroethane, or carbon dioxide. In the case of a pressurized aerosol, the dosage may be determined by providing a valve to deliver a metered amount. Capsules and cartridges of, for example, gelatin for use in a dispenser may be formulated containing a powder mix of the compound and a suitable powder base, such as lactose or starch.

The pharmaceutical composition described herein may be formulated for parenteral administration, e.g., by bolus injection or continuous infusion. Formulations for injection may be presented in unit dosage form, e.g., in ampoules or in multidose containers with, optionally, an added preservative. The compositions may be suspensions, solutions, or emulsions in oily or aqueous vehicles, and may contain formulatory agents such as suspending, stabilizing, and/or dispersing agents.

Pharmaceutical compositions for parenteral administration include aqueous solutions of the active preparation in water-soluble form. Additionally, suspensions of the active ingredients may be prepared as appropriate oily or water-based injection suspensions. Suitable lipophilic solvents or vehicles include fatty oils such as sesame oil, or synthetic fatty acid esters such as ethyl oleate, triglycerides, or liposomes. Aqueous injection suspensions may contain substances that increase the viscosity of the suspension, such as sodium carboxymethyl cellulose, sorbitol, or dextran. Optionally, the suspension may also contain suitable stabilizers or agents that increase the solubility of the active ingredients, to allow for the preparation of highly concentrated solutions.

Alternatively, the active ingredient may be in powder form for constitution with a suitable vehicle, e.g., a sterile, pyrogen-free, water-based solution, before use. The pharmaceutical composition of the present invention may also be formulated in rectal compositions such as suppositories or retention enemas, using, for example, conventional suppository bases such as cocoa butter or other glycerides.

Pharmaceutical compositions suitable for use in the context of the present invention include compositions wherein the active ingredients are contained in an amount effective to achieve the intended purpose. More specifically, a "therapeutically effective amount" means an amount of active ingredients (e.g., the agent, the polynucleotide and/or the expression vector of the present invention) effective to prevent, alleviate, or ameliorate symptoms of the pathology [e.g., a pathology related to increased or decreased cell proliferation such as cancer or prolong the survival of the subject being treated. Determination of a therapeutically effective amount is well within the capability of those skilled in the art, especially in light of the detailed disclosure provided herein.

For any preparation used in the methods of the invention, the dosage or the therapeutically effective amount can be estimated initially from in vitro and cell culture assays. For example, a dose can be formulated in animal models to achieve a desired concentration or titer. Such information can be used to more accurately determine useful doses in humans.

Toxicity and therapeutic efficacy of the active ingredients described herein can be determined by standard pharmaceutical procedures in vitro, in cell cultures or experimental animals. The data obtained from these in vitro and cell culture assays and animal studies can be used in formulating a range of dosage for use in human. The dosage may vary depending upon the dosage form employed and the route of administration utilized. The exact formulation, route of administration, and dosage can be chosen by the individual physician in view of the patient's condition. (See, e.g., Fingl, E. et al. (1975), "The Pharmacological Basis of Therapeutics," Ch. 1, p.l.) Dosage amount and administration intervals may be adjusted individually to provide sufficient plasma or brain levels of the active ingredient to induce or suppress the biological effect (i.e., minimally effective concentration, MEC). The MEC will vary for each preparation, but can be estimated from in vitro data. Dosages necessary to achieve the MEC will depend on individual characteristics and route of administration. Detection assays can be used to determine plasma concentrations.

Depending on the severity and responsiveness of the condition to be treated, dosing can be of a single or a plurality of administrations, with course of treatment lasting from several days to several weeks, or until cure is effected or diminution of the disease state is achieved.

The amount of a composition to be administered will, of course, be dependent on the subject being treated, the severity of the affliction, the manner of administration, the judgment of the prescribing physician, etc. Compositions of the present invention may, if desired, be presented in a pack or dispenser device, such as an FDA-approved kit, which may contain one or more unit dosage forms containing the active ingredient. The pack may, for example, comprise metal or plastic foil, such as a blister pack. The pack or dispenser device may be accompanied by instructions for administration. The pack or dispenser device may also be accompanied by a notice in a form prescribed by a governmental agency regulating the manufacture, use, or sale of pharmaceuticals, which notice is reflective of approval by the agency of the form of the compositions for human or veterinary administration. Such notice, for example, may include labeling approved by the U.S. Food and Drug Administration for prescription drugs or of an approved product insert. Compositions comprising a preparation of the invention formulated in a pharmaceutically acceptable carrier may also be prepared, placed in an appropriate container, and labeled for treatment of an indicated condition, as further detailed above.

It is expected that during the life of a patent maturing from this application many novel miRNAs will be developed and the scope of the term miRNA is intended to include all such new technologies a priori.

As used herein the term "about" refers to ± 10 %

The terms "comprises", "comprising", "includes", "including", "having" and their conjugates mean "including but not limited to".

The term "consisting of means "including and limited to". The term "consisting essentially of means that the composition, method or structure may include additional ingredients, steps and/or parts, but only if the additional ingredients, steps and/or parts do not materially alter the basic and novel characteristics of the claimed composition, method or structure. Throughout this application, various embodiments of this invention may be presented in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the invention. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual numbers within that range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the breadth of the range. Whenever a numerical range is indicated herein, it is meant to include any cited numeral (fractional or integral) within the indicated range. The phrases "ranging/ranges between" a first indicate number and a second indicate number and "ranging/ranges from" a first indicate number "to" a second indicate number are used herein interchangeably and are meant to include the first and second indicated numbers and all the fractional and integral numerals therebetween.

As used herein the term "method" refers to manners, means, techniques and procedures for accomplishing a given task including, but not limited to, those manners, means, techniques and procedures either known to, or readily developed from known manners, means, techniques and procedures by practitioners of the chemical, pharmacological, biological, biochemical and medical arts.

It is appreciated that certain features of the invention, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable subcombination or as suitable in any other described embodiment of the invention. Certain features described in the context of various embodiments are not to be considered essential features of those embodiments, unless the embodiment is inoperative without those elements.

Various embodiments and aspects of the present invention as delineated hereinabove and as claimed in the claims section below find support in the following examples. EXAMPLES

Reference is now made to the following examples, which together with the above descriptions illustrate some embodiments of the invention in a non limiting fashion. Reference is now made to the following examples, which together with the above descriptions, illustrate the invention in a non limiting fashion.

Generally, the nomenclature used herein and the laboratory procedures utilized in the present invention include molecular, biochemical, microbiological and recombinant DNA techniques. Such techniques are thoroughly explained in the literature. See, for example, "Molecular Cloning: A laboratory Manual" Sambrook et al., (1989); "Current Protocols in Molecular Biology" Volumes I-III Ausubel, R. M., ed. (1994); Ausubel et al., "Current Protocols in Molecular Biology", John Wiley and Sons, Baltimore, Maryland (1989); Perbal, "A Practical Guide to Molecular Cloning", John Wiley & Sons, New York (1988); Watson et al., "Recombinant DNA", Scientific American Books, New York; Birren et al. (eds) "Genome Analysis: A Laboratory Manual Series", VoIs. 1-4, Cold Spring Harbor Laboratory Press, New York (1998); methodologies as set forth in U.S. Pat. Nos. 4,666,828; 4,683,202; 4,801,531; 5,192,659 and 5,272,057; "Cell Biology: A Laboratory Handbook", Volumes I-III Cellis, J. E., ed. (1994); "Culture of Animal Cells - A Manual of Basic Technique" by Freshney, Wiley-Liss, N. Y. (1994), Third Edition; "Current Protocols in Immunology" Volumes I-III Coligan J. E., ed. (1994); Stites et al. (eds), "Basic and Clinical Immunology" (8th Edition), Appleton & Lange, Norwalk, CT (1994); Mishell and Shiigi (eds), "Selected Methods in Cellular Immunology", W. H. Freeman and Co., New York (1980); available immunoassays are extensively described in the patent and scientific literature, see, for example, U.S. Pat. Nos. 3,791,932; 3,839,153; 3,850,752; 3,850,578; 3,853,987; 3,867,517; 3,879,262; 3,901,654; 3,935,074; 3,984,533; 3,996,345; 4,034,074; 4,098,876; 4,879,219; 5,011,771 and 5,281,521; "Oligonucleotide Synthesis" Gait, M. J., ed. (1984); "Nucleic Acid Hybridization" Hames, B. D., and Higgins S. J., eds. (1985); "Transcription and Translation" Hames, B. D., and Higgins S. J., eds. (1984); "Animal Cell Culture" Freshney, R. L, ed. (1986); "Immobilized Cells and Enzymes" IRL Press, (1986); "A Practical Guide to Molecular Cloning" Perbal, B., (1984) and "Methods in Enzymology" Vol. 1-317, Academic Press; "PCR Protocols: A Guide To Methods And Applications", Academic Press, San Diego, CA (1990); Marshak et al., "Strategies for Protein Purification and Characterization - A Laboratory Course Manual" CSHL Press (1996); all of which are incorporated by reference as if fully set forth herein. Other general references are provided throughout this document. The procedures therein are believed to be well known in the art and are provided for the convenience of the reader.

All the information contained therein is incorporated herein by reference.

EXAMPLE 1 Connectivity Distributions in the miR-Gene Network

Materials and Methods

MiRs and their predicted targets: miRs and their predicted targets were taken from two previously published studies: TargetScan [Lewis BP₅ et al., (2005) Cell 120: 15-20; Lewis BP, et al., (2003) Cell 115: 787-798] (wwwdortargetscandotorg) and PicTar [Krek A, et al. (2005) Nat Genet 37: 495-500] (genomedotucscdotedu). Both resources predict and assign target genes to miRs based on evolutionary conservation between human, mouse, rat and dog. TargetScan targets were downloaded and gene symbols were converted to RefSeq IDs using UCSC mysql databases. PicTar targets were downloaded from the UCSC hgl7 database where they are presented as the picTarMiRNA4Way track. Target hubs analysis: Target hubs were defined as genes which are targeted by more miRs than the 99th percentile of the maximal value in 100 randomizations of the columns in the miR to gene assignment matrix, each preserved the total number of targets per miR. According to this procedure, in the TargetScan dataset, target hubs were defined as genes which are targeted by more than 15 miRs (there were 470 such genes), and in the PicTar dataset, target hubs were defined as genes targeted by more than 20 miRs (834 genes). For original and randomized distributions see Figure IA.

To check whether the target hubs contain many miR target sites merely because they have, on average, longer 3' UTRs, the length of 3' UTRs for all RefSeq genes was retrieved from UCSC hgl7. A randomization test was performed on this 3' UTR length data, in which sets of genes were randomly picked from the data with distribution of 3' UTR length that was as similar as possible (see below) to that of the target hubs. For each such set of genes the average number of different miRs predicted to target them was calculated. This randomization procedure was repeated 100 times, and the distribution of average number of miRs was derived (Figures 6A-B). The figure shows that these values are significantly lower than the average of the original target hubs, indicating that the length is neither necessary nor sufficient for a gene to be a target hub.

100 random sets of genes were generated with length distributions similar to that of the target hubs by the following procedure. For each target hub with UTR length, LTH, a set of genes with similar UTR length was defined, which included all the genes in the dataset with a UTR length equal to LTH₅ or longer up to an additional 5 % of LTH (genes which did not have such sets were excluded from the analysis). Then, a representative from each set to be included in the randomized version of target hubs was randomly selected. miR density in genes 3' UTRs was calculated as the number of miRs targeting a gene divided by its 3 ' UTR length. 3 ' UTR length was extracted from the UCSC database.

When defining high density target hubs, the density cutoffs were selected to be the top 85th percentile of the entire distribution of densities. Of note, this distribution included only genes that participated in the present analyses and thus does not contain genes with a density of zero (i.e., zero predicted sites in the UTR). Degree-preserving matrix randomization: To determine a p- value on the cooccurrence rate of a pair of two miRs, a co-occurrence score was first defined. The Meet/Min score was selected [Goldberg DS, Roth FP (2003) Proc Natl Acad Sci U S A 100: 4372-4376; Ravasz E, et al., (2002) Science 297: 1551-1555], which is formulated in the main text, and it was calculated on the matrix of miR to target genes. For the purpose of p- value calculations, a null model of randomized matrices was defined, which preserves the matrix statistics such that for each gene the number of miRs targeting it, and for each miR the number of genes it targets remains the same as in the original data. This model was first introduced as a randomization model for networks [Shen-Orr SS, et al., (2002) Nat Genet 31: 64-68], which preserved all in- and out-degrees in a given network, and thereby controlling for the possibility that significance of a phenomenon may be merely attributed to the degree distribution in the network. Randomized matrices were created by the edge-swapping procedure, starting from the original matrix of miR to target gene predictions. Two pairs of miR and target gene were randomly picked, miRα-gene_ji and miRi₂-genej₂, and, after verifying that miRπ does not already target genej₂ and miRi₂ does not already target geneμ, the switch of an edge in the matrix was performed, so that after the swap there is a "0" instead of "1" in the positions iiji and i₂,J2 in the matrix, and a "1" instead of a "0" in the positions ^j₂ and i₂ji in the matrix. To decide how many swapping events were needed before the matrix was "well randomized," the number of edges that were actually swapped were monitored and compared with the number of changed edges in a randomly shuffled matrix. This number was followed during the swapping steps and it was realized that it plateaued at about 100,000 steps. Thus in all subsequent analyses the swapping procedure was repeated for 100,000 steps. During the calculation of the Meet/Min score for a pair of miRs in the original data, genes that contained a match to the two miRs were excluded if the two sites physically overlapped on the target's 3' UTR. In addition, the present inventors filtered out from the analysis, pairs of miRs whose seeds were identical (overlap of seven out of seven nucleotides, positions 2-8 of the miR). These two precautions were taken to eliminate the possibility of overestimating the significance of the rate of miR co-occurrence due to seed sequence similarity between different miRs.

After having calculated the co-occurrence p- values and avoidance p- values for all possible miR pairs, the present inventors controlled for multiple hypotheses using FDR and only pairs that passed FDR of 0.05 were considered to be significantly co-occurring or avoiding.

Significant miR— TF co-occurring pairs: For the task of identifying miR— TF pairs that significantly co-occur in a high number of target genes, a p-value was calculated (using a cumulative hypergeometric test) on each pair of regulators as was performed before for pairs of TFs [Pilpel Y, et al., (2001) Nat Genet 29: 153-159]. The hypergeometric p- value was calculated after the RefSeq genes were mapped to a unique set of Gene IDs, to reduce redundancy in the set. hi the miR-TF p-value calculations, the total number of genes in the hypergeometric analysis was calculated as the number of genes that appeared (i.e., had at least one binding site) in both datasets. Genes that appeared only in the TF dataset or in the miR dataset were excluded and were not counted. FDR was used to correct for multiple hypotheses testing, and the set of significant pairs of coregulators was determined.

Co-occurrence p-values were also calculated for all possible miR-TF pairs using the new randomization method presented above. Specifically, both the matrix which assigns TFs to genes and the matrix with assignments of miRs to genes were subjected to 100,000 iterations of the edge-swapping procedure. In total 1,000 such pairs of randomized matrices were generated. The co-occurrence p-value of a given TF-miR pair is the fraction of the randomized matrix pairs in which this pair's Meet/Min score was higher than the pair's Meet/Min score in the original matrices, and the corresponding z-score is the difference between the original Meet/Min score and the mean of the score in the randomized matrices, divided by their standard deviation.

Most reassuringly, when checking the overlap of these significant pairs with the significant pairs that passed FDR cutoff of 0.3 using the hypergeometric model, it was noticed that the overlap was very high; it was more than 72 % for PicTar and 92 % for TargetScan. For subsequent analyses of network motifs (FFLs and indirect FFL search), all the pairs that passed FDR of 0.3 in the hypergeometric test in the three datasets (see Transcription Factor binding sites section below), and that passed FDR of 0.3 (p-value < 6 x 10^"3) in the PicTar 10 kb set, and minimal p- value (< 10^"3) in the PicTar 5 kb and TargetScan sets were selected, as these already had an extremely high overlap (> 93%) in the hypergeometric derived set.

The final set of significant pairs in the miR-TF network is presented in FDR q- value cutoffs of 0.1, 0.2, and 0.3. With q- value, of 0.1, 20 TF-miR pairs were obtained with significant j>-value using the TargetScan dataset, and 267 using the PicTar 10 kb dataset, and 70 using the PicTar 5 kb dataset. With a q-value of 0.2, 60 TF-miR pairs were obtained with significant j?-value using the TargetScan dataset, and 555 using the PicTar 10 kb dataset, and 261 using the PicTar 5 kb dataset. With 0.3 104 TF-miR pairs were obtained with significant p- value using the TargetScan dataset, and 916 using the PicTar 10 kb dataset, and 497 using the PicTar 5 kb dataset. miRs clusters and regulatory regions: As was shown in the past [Tanzer A, Stadler

PF (2004) J MoI Biol 339: 327-335], miRs may be clustered on the genome, and are often transcribed as one unit. Therefore, to predict regulatory regions of miRs (i.e., proximal as well as potentially more distant promoters or enhancers) miRs first had to be clustered on the human genome. All 461 pre-miRs in miRBase (micrornadotsangerdotacdotuk) were mapped onto the human genome and clustered according to physical proximity (genomic locations of miRs were taken from UCSC hgl7 and some miRs were mapped from hgl8 back to hgl7 using the UCSC "lift genome" web service). Two pre-miRs, that are consecutive on the genome, were considered belonging to the same cluster if the distance between them was shorter than a cutoff, provided that they are transcribed from the same strand. miRs kept being added to clusters until the first distance that was larger than the cutoff was hit. To learn a meaningful cutoff from the data, the distribution of distances between all neighboring pre-miRs in the genome was plotted. Interestingly, the distribution was found to be bimodal — distances below and above 10 kb (on a log scale, Figure 6A) were highly represented in contrast to a lower representation about 10 kb. This indicated that a reasonable cutoff on the distance between two adjacent miRs that still belong to the same cluster may be 10 kb. Using this clustering procedure, 301 clusters were generated, the majority of which (-82.39 %) consisted of a single miR; the cluster with the highest number of miRs contains 43 miRs (see Figure 7 for the distribution of number of miRs per cluster). In a previous study, which was based on 207 miRs (compared with the 461 used here), miRs were clustered using different a cutoff [Alruvia Y, et al, (2005) Nucleic Acids

Res 33: 2697-2706]. When the cluster analysis was repeated with the current set of miRs, with the previous cutoff, a similar clustering was obtained — 94 % of the present clusters are identical to the clusters generated with the alternative cutoff and average cluster lengths are very similar (unpublished data).

Reassuringly, using expression data of miRs across tissues it was found that miRs that belong to the same cluster have a significant tendency to be coexpressed compared with miRs that do not map to shared clusters (Figure 8B). This tendency is preserved even in cases where miRs that belong to the same cluster are relatively far from each other on the genome (Figure 8B, inset).

The present inventors then defined, as a putative regulatory region of miRs, the sequence that lies 10 kb upstream of the 5' most pre-miR in each miR cluster. The 10 kb promoter length was determined from the data as follows. A distribution of number of conserved TFBS upstream of clusters was generated (Figure 8C). It was found that the number of conserved TFBS gradually declined as a function of the distance from the putative 5' end of the cluster, with a plateau obtained at about 10 kb upstream. The distribution was rather noisy, probably due to the fact that primary-miR transcripts are much longer than the precursor miR that were related to (e.g., the primary transcript of the miR-17-92 cluster is C13orf25, which is 6,795 bp long), and thus the transcription start site (TSS) taken here is only crudely defined. The presence of a TFBS in a miR promoter was considered only if such occurrence was conserved in mouse and rat, as taken from the UCSC hgl7 conserved track in the relevant regions.

Transcription factor binding sites: Predicted binding sites for all human mouse and rat PSSMs from TRANSFAC [Matys V, et al. (2003) Nucleic Acids Res 31: 374-378] version 8.3 were used, as they are defined by the UCSC hgl7 genome assembly, in the tfbsConsSites (genomedotucscdotedu/) and tfbsConsFactors. All RefSeq genes genomic locations were taken from hgl7. To determine the length of upstream regulatory regions, the number of conserved TFBS upstream RefSeq genes as a function of distance from TSS was measured (see Figure 9). The result shows that the signal decays and plateaus between 5 kb and 10 kb upstream of the TSS. The present inventors hence chose to work with two alternative cutoffs of promoter length, 5 kb and 10 kb. The regulatory regions thus defined probably consist of proximal promoters as well as distant enhancers. The recent Affymetrix promoter chip for detection of ChIP experiments with TF binding in human promoters also consists of probes that span 10 kb of regulatory regions, and future experiments with this chip and as many TFs as possible will allow a better delineation of regulatory regions boundaries. Although regulatory regions which were longer than the common definition were used, use of evolutionary conservation filter gives confidence in the present regulatory region definitions. Feed forward loop statistics: FFL TF^miR: For all the significant pairs of coregulators (i.e., TF-miR partners that co-occur in a significantly high number of targets) the present inventors investigated whether the TF has a binding site in the putative promoter of the miR cluster from which the miR partner is transcribed. In some cases in which the mature miR sequence is transcribed from more than one genomic locus, all possible regulatory regions of the relevant miR clusters were examined. In addition, each PSSM may belong to a family of PSSMs, with similar binding sites, representing the same TF (a family was defined as several PSSMs representing the same TF, as determined from the UCSC hgl7 tfbsConsFactors track). Thus, PSSM-miR pairs are treated as TF-miR pair, and given a pair of PSSM-miR partners, it may be said that the PSSM's TF regulates the miR if at least one of the PSSMs that corresponds to that TF has a match in the regulatory region of the miR partner (the same procedure was carried out in the randomizations described below).

For testing the FFL miR-^TF configuration, the present inventors had to connect first between TRANSFAC PSSMs and the genes encoding the TFs that bind these PSSMs. For that, PSSMs were mapped to the TF they represent which in turn was mapped to a SwissProt ID, these two mappings were done using the UCSC hgl7 tfbsConsFactors track. These SwissProt IDs were then mapped to RefSeq IDs, for which the data on miR targets was maintained. This information served also in the process of indirect FFL search; for each of the TF- miR partners, the present inventors checked whether the miR is regulated by another mediator TF, which in turn is regulated by the partner TF. It was noted that not all TFs had a corresponding SwissProt ID in the UCSC hgl7 tfbsConsFactors track, and therefore not all pairs served as candidates for the FFL miR^TF and the indirect FFL; only in 74 of the 104 (71 %) TargetScan significant pairs, and in 680 of 916 (74 %) of the PicTar pairs the PSSM could be mapped to a RefSeq gene. The following procedure was used for the calculation of the significance of the

FFLs and indirect FFL in the PicTar and TargetScan miR-TF networks. Since there were 104 and 916 pairs of miR-TF partners in the two respective networks, the present inventors have drawn 10,000 times the same number of random pairs of TFs and miRs out of all the possible pairs in each network. The number of each FFL and indirect FFL was recorded in each randomization and a p-value (and a corresponding z-score) on the hypothesis that a given network motif is over-represented in the network was taken to be the number of random sets with a greater or equal number of motifs in it. miR and niRNA tissue expression data: The expression profiles of 150 miRs across five healthy human tissues and organs (brain, liver, thymus, testes, and placenta) were previously measured using miR-dedicated microarrays [Barad O, et al. (2004) Genome Res 14: 2486-2494]. miRs from the chips were mapped to PicTar and TargetScan, they cover 154 and 87 of the miRs in the two respective datasets. In addition, data was used from Su AI, et al. (2004) Proc Natl Acad Sci U S A 101 : 6062-6067 for human mRNAs expression across the same set of tissues. Both sets of expression data were column centered (chip- wise centering: each chip's values were divided by the chip mean to account for differences in chip intensities) and then Iog2 transformed. Regarding mRNA expression chips, the present inventors particularly focused on genes coding for the TFs that participated in the present analysis. Using the above mapping of PSSMs to their corresponding TF genes, a total of 127 TFs were identified that could be matched to at least one probe set in the mRNA expression dataset [Su AI, et al. (2004) Proc Natl Acad Sci U S A 101: 6062-6067]. The tissue expression correlation of all significantly co- occurring miR and TF pairs was examined for which there was an expression profile. When more than one gene was attributed to the same TF, the present inventors chose for each pair of TF and miR the one with the highest absolute value of correlation coefficient out of all options. This was consistently done both for the background statistics of all possible TF-miR pairs and for the predicted TF-miR partners. In total, correlation coefficients for 361 such TF-miR partners out of 916 partners in PicTar, and for 30 out of 104 partners in TargetScan were calculated. The miR expression data [Barad O, 2004, Genome Res 14: 2486-2494] consisted of five healthy tissues, and HeLa cells, while the mRNA study that was focused on [Su AI, et al. (2004) Proc Natl Acad Sci U S A 101: 6062-6067] overlapped with the miR data only in the five tissues. Therefore when expression between miRs and TFs was compared, only the five healthy tissues were used, and when expression of miR pairs was compared, all six samples were used. Noise-tolerance analysis: The assignments of miRs to targets are known to be of limited accuracy. The present inventors thus wanted to assess the noise tolerance of the present results. A procedure previously utilized for the case of network motifs in the bacterial transcription network was adopted [Shen-Orr SS, (2002) Nat Genet 31: 64-68]. The present inventors experimented with different percentages of the connections in the network that were randomly removed or added and the significance of the present FFL motifs was assessed for each case. Similarly to the findings in the E. coli network, it was found that up to 20 %-30 % of the edges can be added or removed without appreciable effect on the FFL significance. RESULTS

Two datasets of miRs and their predicted target genes were used: TargetScan [Lewis BP, et al., (2005) Cell 120: 15-20; Lewis BP, (2003) Cell 115: 787-798] and PicTar [Krek A, et al. (2005) Nat Genet 37: 495-500]. The miRs used in this analysis are characterized by being evolutionarily conserved and, in addition, their targets were defined based on conservation in orthologous genes in four species (human, mouse, rat and dog). This evolutionary conservation criterion was assumed to constitute a good filter for false positive assignments of miRs to genes. Altogether, 8,672 and 9,152 human (RefSeq) genes were analyzed in the TargetScan and PicTar datasets, respectively, that have at least one predicted miR binding site in their 3' UTR, and a total of 138 miRs and 178 miRs in the respective datasets.

A matrix was constructed whose rows are genes and columns are miRs, in which the ij* element is "1" if gene i contains a predicted binding site for miR j in its 3' UTR, and "0" otherwise. One such matrix was created for each of the two miR target prediction datasets. For the sake of clarity, from here on "a miR targets a gene" and "a gene contains in its 3' UTR a predicted binding site for a miR" are used interchangeably. First, the matrix was characterized by the distribution of degree connectivity of each gene and of each miR. Figure IA shows the distribution of the number of miRs assigned per gene, while Figure IB shows the distribution of number of genes assigned to each miR. Each distribution was compared to a set of distributions, each derived by randomization of the original matrix according to two alternative null models. Along with the distribution of number of miRs per gene (Figure IA) 100 distributions were plotted, obtained after randomizing each of the columns in the matrix. In this randomization, the number of genes per miR was preserved, yet assigned genes at random to each miR. The distributions obtained after the randomization differed markedly from the original distribution, both in terms of width and shape. While in the randomized distributions genes rarely have binding sites for more than ten different miRs in their 3' UTR, in the original distribution there are hundreds of genes subjected to extensive predicted miR regulation. In Figure IB the distribution of number of genes per miR is also shown. Along with it is shown a set of distributions obtained by randomizing each of the rows in the matrix, namely by randomly assigning miRs to each gene, preserving the original number of miRs predicted to target each gene, as in the original matrix. Here, too, the randomized distributions differed from the original one both in shape and width; the original data contains multiple miRs which appear to target more than 400 genes, significantly higher than the number that would be obtained by merely preserving the statistics of number of miR sites in genes UTRs. These observations lead the present inventors to highlight some special properties that seem to be unique to the miR regulatory network.

The distribution of number of miRs regulating each target gene (Figure IA) has a long right tail in contrast to the distributions in the randomized matrices that looked Gaussian (as befits a sum of independent random variables). The present inventors thus focused on the genes in that tail of the distribution (which are targeted by more than 15 miRs and 20 miRs in the TargetScan and PicTar datasets, respectively, see Materials and Methods for further details and cut-off justification). These genes were named target hubs. There are 470 such genes in the TargetScan dataset. Similar observations were made with the PicTar dataset and identified 834 target hubs — the set of target hubs based on the TargetScan dataset has an 81 % overlap with the target hubs defined by PicTar dataset.

Inspecting the target hubs genes' annotations (using Gene Ontology, GO), it was found that they are highly enriched for developmental processes, specifically for muscle development and nervous system development, as well as for TFs and transcription regulators (see Table 1 herein below).

Table 1

Table 1. TargetScan Target Hubs GO Functional Enrichment

Target hubs were defined by two alternative definitions: target hubs with high number of miR binding site (more than 15 in the case of TargetScan and more than 20 in the case of PicTar), or as high density target hubs (genes with high density of miRs in their 3' UTRs). The standard method of hypergeometric p-value was used to test for functionally enriched GO annotations in each gene set. The results were corrected for multiple hypotheses and annotations were considered significantly enriched if they passed FDR of 0.05. The table presents the union of significant annotations for the high density target hubs and the high miR number target hubs.

Among the transcription regulators in the set of target hubs are included RUNXl, E2F-3, N-MYC, and SP3. Another very intriguing fact is that the Agol gene, one of the key components of the human RISC (RNAi induced silencing complex), is also a target hub, as in the dataset it appears to be potentially regulated by multiple miRs.

It was suspected, however, that the fact that target hubs host many miR binding sites may result from potentially longer 3' UTRs. Although it was found that target hubs have a distribution of 3' UTR lengths that is significantly longer than that of the rest of the genes in the current analysis (p- value = 4 * 10^~85 and p- value = 3 x ICf¹⁰¹ for TargetScan and PicTar target hubs, respectively, using the Kolmogorov-Smirnov test), that many of them have relatively short 3' UTRs. To test whether the high number of miR binding sites in the target hubs is a simple reflection of their 3' UTR lengths a randomization test was performed, in which the present inventors sampled 100 times random gene sets from the entire dataset with the same or very similar length distributions as that of the target hubs (see Materials and Methods). It was found that such gene sets always have a significantly lower average number of miR sites per gene compared with the target hubs. The density of different miRs in the 3' UTRs was further calculated. Density was defined as number of different miRs targeting a gene divided by 3' UTR length. Remarkably it was found that the miR binding site density in the target hubs is significantly higher than in the rest of the genes in the dataset (p-value = 2 x 10^~85 and p-value = 6 x 10^~124 for the TargetScan and PicTar target hubs, respectively, using the Kolmogorov-Smirnov test; the means are 2.84 and 1.80 times higher in the TargetScan and PicTar dataset means, respectively, see Figures 2A-B). It was concluded that target hubs are rich in binding sites for different miRs to an extent that cannot be explained solely by their 3' UTRs lengths.

Realizing that density of miR binding sites may be an important property by itself, an alternative definition for target hubs was also used — genes with particularly high density of miRs in their 3' UTRs. Genes in the top 85th percentile of the miR binding site density spectrum were collected, and a similar GO enrichment analysis was performed to see whether particular functionalities were enriched among the genes with a high density of miRs. Reassuringly, most of the functionalities that were enriched among the set of target hubs defined by number of different miRs were also significant in the set of high density target hubs (see Table 1). Moreover, it was found that genes that were target hubs according to only one of the two definitions (i.e., genes that are not in the overlap of the two sets) were still significantly enriched for functionalities such as transcription regulator activity and development (unpublished data).

Combinatorial interactions are a fundamental property of the transcription networks. It may be anticipated that, similarly to TFs, miRs may work in combinations. One way to predict pairs of coregulating miRs is to ask which pairs show a high rate of cooccurrence in the same target genes' 3' UTRs. A common statistical test in the field, previously used in the context of promoter motifs and TF binding site, is the cumulative hypergeometric statistic. According to this model, given the rate of occurrence of each of the regulators alone, and the total number of genes in the analysis, a p- value is computed on the size of the set of genes that are shared between the two regulators. The main assumption of this model, that assignment of a gene to the first regulator is independent of the assignment to the second one, is likely fulfilled in the context of fixed-length promoters. Yet when it comes to 3' UTRs of varying length, the assumption does not hold anymore. Some genes, e.g., those with long 3' UTRs, have a higher chance to contain predicted binding sites for miRs, hence a p- value calculated based on the hypergeometric model may overestimate the significance of the co-occurrence rate.

Thus, an alternative, randomization-based test for identifying significantly co- occurring miR pairs was devised. The model was designed such that it will capture the underlying distributions in Figures 1 A and B, and test whether a given pair of miRs co- occurs at a higher rate, considering the above distributions as a background. For each pair of miRs, i and j, with their set of targets, Targets(i) and Targets(j) respectively, the "Meet/Min" score was calculated [Goldberg DS, Roth FP (2003) Proc Natl Acad Sci U S A 100: 4372-4376; Ravasz E, et al., (2002) Science 297: 1551-1555] defined in the present case as:

Targets(j) H Targets(/)| min(|Targets(j)|, |Targets(/)|) namely, the size of the set of genes that contain sites for the two miRs together, divided by the smaller of the two sets of targets (the present inventors filtered from the calculation for each ij pair, 3' UTRs in which the sites for i and j are physically overlapping to avoid overestimation of significance of miR pairs with an overlapping or similar seed, see Materials and Methods for details). Yet this score is not a statistic, i.e., it lacks an estimate of the probability to obtain such score (or better) by chance given an appropriate null model. Following previous works [Shen-Orr SS, MiIo R, Mangan S, Alon U (2002) Nat Genet 31: 64-68] a null model that preserves for each gene the number miRs assigned to it was used, and for each miR the number of genes assigned to it in the input data. The present inventors generated 1,000 randomized matrices according to this null model. In each such matrix the original matrix was randomized in 100,000 steps, using an edge- swapping algorithm [Shen-Orr SS, MiIo R, Mangan S, Alon U (2002) Nat Genet 31: 64-

68]. For each such randomized matrix the Meet/Min score for all pairs of miRs was recomputed. The co-occurrence jp-value for a pair of miRs was computed according to the pair's Meet/Min score and the population of 1,000 Meet/Min scores obtained for that same pair in each of the 1,000 edge-swapped matrices. The p- value for the pair is defined as the fraction of the 1,000 randomized matrices in which the Meet/Min score of that pair is greater than or equal to the Meet/Min score of the pair in the original matrix.

In addition to calculating a score of co-occurrence, using the same formalism, a score was also calculated that captures the tendency of every two miRs to avoid residing within shared 3' UTRs. A pair of miRs that co-occur in the original matrix significantly less frequently than in the edge-swapped matrices was regarded as avoiding each other. Given the Meet/Min score of co-occurrence for a pair of miRs, and the Meet/Min scores obtained for that pair in the 1,000 edge-swapped matrices, the fraction of randomized scores that were lower than or equal to that obtained in the original matrix for that pair, was calculated as the avoidance/*- value of a miR pair.

In both cases of co-occurrence and avoidance, false discovery rate (FDR) was used to control for the testing of multiple hypotheses. In the case of co-occurring miR pairs, using a restrictive FDR threshold (q- value = 0.05), 107 pairs were obtained with significant a p- value in the TargetScan dataset, and 199 pairs in the PicTar dataset (interestingly, the ratio between the number of interactions in the two datasets (-0.54) is very close to the ratio expected based on the square of relative number of miRs in each dataset (~0.6)). A combinatorial network was created based on the significant co-occurring miR pairs. The top miR pairs are given in Table 2 and are also depicted in Figure 3 A.

Table 2

Table 2 above depicts the number of targets each miR has in the specific database, and the number of targets which contain sites for both miRs. Note that in each pairing, genes were filtered out where binding sites for the two miRs physically overlapped, so this

£>-value is not biased by miRs with highly similar seeds. For this reason, the number of target genes may be slightly different for the same miR in two different pairings.

The full list of significant pairs is provided in Table 3 herein below (Full list of TargetScan co-occurring miR pairs that passed FDR of 0.05) and Table 4 herein below (Full list of PicTar co-occurring miR pairs that passed FDR of 0.05). Table 3

Table 4

This combinatorial network consists of several levels of hierarchy. At the top (Figure 3A) are a handful of miRs which interact with a relatively large number of miR partners, while at the bottom are "end-nodes" with very few miR partners each. Examination of the degree distribution in the miR combinatorial network revealed a power-law with slope of about -1.5 and R = -0.89 in TargetScan and R =0.94 in PicTar (Figures 3B-C), indicating that the network of coregulating miRs is scale-free (alternative FDR cut-offs also resulted in scale free networks with R always bigger than 0.72). Interestingly, expression data of the miRs provides some support for the predicted regulatory interactions between them. It was found that coexpressed miRs tended to have relatively high co-occurrence scores, and significant co-occurrence p-values, while miR pairs with negatively correlated expression tended to avoid residing in shared 3' UTRs (see below). EXAMPLE 2

Coordinated Regulation of Target Genes by miRs and TFs

A potential regulatory design in the gene expression network is that genes belonging to the same regulon will be coregulated not only at the transcriptional level, but also posttranscriptionally. One potential realization of this design may be that a particular miR and a particular TF would regulate common targets. A simple means to identify some of the cases of regulatory cooperation between a miR and a TF may be to find TF-miR pairs that co-occur in a large set of shared targets compared to the size expected by chance. Similar to the case of miRs sites in 3' UTRs, the present inventors considered a TF to be present in a human gene's promoter only if its occurrence in the promoter is conserved in the promoters of orthologous genes from mouse and rat (as taken from UCSC, see Materials and Methods). A matrix was then created whose rows are the genes and columns are TFs, with a "1" for the i-th gene and the j-th TF if the TF binding site (TFBS) occurs in the gene's promoter and "0" otherwise. To identify pairs of TFs and miRs that cooperate in regulating shared target genes the present inventors looked for TF-miR pairs with a high rate of co-occurrence in the promoters and 3' UTRs of the regulated genes. The co-occurrence was tested in shared genes of each of the 409 position specific scoring matrices (PSSMs) representing TF binding sites in TRANSFAC [Matys V, et al. (2003) Nucleic Acids Res 31: 374-37] with each of the 138 and 178 miRs in the TargetScan and PicTar databases respectively. A PSSM and a miR are said to co-occur in the same gene if the PSSM has a conserved binding site in the promoter of the gene and the miR has a conserved predicted site in the gene's 3' UTR. Two statistical models were used to calculate the significance of rate of TF-miR co-occurrence, and ultimately considered TF-miR pairs that were found to be significant according to both tests. First, a hypergeometricp- value was calculated based on the number of genes that contain a TFBS in their promoter, the number of genes that contain a miR site in their 3' UTR, and the number of genes that contain both the TF and the miR sites (see Materials and Methods for details). Such p- values were computed on all TF-miR pairs and a threshold was set on the ^-values obtained to account for the multiplicity of hypotheses, using FDR. Using an FDR q-value of 0.3, 111 miR-TF pairs were obtained with significant p- values using the TargetScan dataset and 1,263 miR-TF pairs with significant p- values using the PicTar dataset (see Materials and Methods for number of pairs with more stringent q- values). Reassuringly there is a high overlap between the TargetScan and PicTar networks (68.7 % of the TargetScan miR-TF network pairs were also found to be significant pairs in the PicTar network). The hypergeometric p- value has the advantage of being an analytical model with essentially unlimited resolution. Also, unlike the above situation of miR co-occurring pairs, which exhibited inherent dependency between the two regulators, the present case of TF-miR interaction does not present such limitation. Nevertheless, it was decided to also backup the hypergeometric- based predictions with a randomization test, very similar to the one presented above for the case of miR co-occurrence, that preserves the distribution of number of regulators of each gene, the number of targets of each TF, and the number of targets of each miR in the input datasets. The co-occurrence rates and ^-values of all TF-miR pairs Matys V, Fricke E, Geffers R, Gossling E, Haubrock M, et al. (2003) TRANSFAC: transcriptional regulation, from patterns to profiles. Nucleic Acids Res 31: 374-37, and FDR was used as above to account for the multiplicity of hypotheses (see Materials and Methods for details). Reassuringly, 9 3% and 72 % of the hypergeometric-based TF-miR interactions from the TargetScan and PicTar datasets, respectively, were also supported by this alternative model. The rest of the analyses were based on TF-miR pairs that passed the two statistical tests using FDR; there were 104 pairs in the TargetScan dataset and 916 pairs in the PicTar dataset. For simplicity, a TF and a miR that significantly co-occur are termed "partners". Table 5 herein below lists the top TF-miR partners in the TargetScan and PicTar Networks. Table 5

miR-362 88 V CDPCRl 87 9.29E-06 5.09

For Table 5, The p- value is a hypergeometric p- value for the co-occurrence of the a and a TF in the 3' UTRs and promoters of the same genes, and the Z-score is assigned according to the randomization based co-occurrence method. The Table depicts the number of targets each of miR and each TF, and the number of targets which contain sites for both miR and TF.

*In the PicTar table, the pairs of duplicated miRs (a,b etc.) were unified when they appeared more than once as significant.

The full networks of TF-miR partners are presented in Tables 6-8

Table 6

Table 6 - miR-TF significantly co-occuring pairs in the TargetScan network.

AU the pairs in this table passed FDR of 0.3 or less in the hyper-geometric p-value for co-occurrence, and also received a co-occurrence p-value <10E-3 in the randomization- based test, if a pair also constirues an FFL, the following data is provided in the last columns: TF^==:> miR: we provide the PSSMS and the genomic miR cluster for which a binding site was found for the PSSMs indicated, if more than one connection exists, all are specified. miR==>TF: we indicate which miR (out of the clusters from which the partner miR is transcribed) was found to have a binding site, and the RefSeq ID of the partner PSSM, in which the miR site is found. "TF unmapped" indicates that the PSSM could not be mapped to a SwissProt ID using the UCSC hgl 7 tfbsConsFactors track.

Table 7

hsa- miR- V$CHX 0.00176 0.00520

320 382 10 01 196 23 7 2.6106 0.3 27

V$

MY

B_Q

>hsa miR-

22 in clust er clust erlO

O_hs hsa- a- miR- V$CM 0.00177 0.00520 mir-

22 265 YB 01 54 27 4.2956 0.3 27 22 hsa-miR-

372==>

TF cluster

58:

V$E2F1_

Q3_01,

V$E2F1_

Q4,

V$E2F1_

Q4_01,

V$E2F1_

Q6,

V$E2F1_

Q6_01,

V$E2F_0

I, V$E2F_0

2,

V$E2F_0

3,

V$E2F_Q

3_01,

V$E2F_Q

4_01,

V$E2F_Q

6_01, hsa- miR-

372==>

TF cluster

62:

V$E2F1_

Q4_01,

V$E2F_0

2,

V$E2F_Q

3_01,

V$E2F_Q

4_01, hsa- V$E2F__Q miR- V$E2F_ 0.00177 0.00520 6_01,

372 560 Q3_01 92 5.7473 0.3 27 (NMJ)Ol

Table 7 - miR-TF significantly co-occuring pairs in the PicTar network, 10kb promoters.

All the pairs in this table passed FDR of 0.3 or less in the hyper-geometric p- value for co-occurrence, and also received a co-occurrence p-value <6*10E-3 (FDR q-value=0.3) in the randomization-based test, if a pair also constitues an FFL, the following data is provided in the last columns: TF==> miR: we provide the PSSMS and the genomic miR cluster for which a binding site was found for the PSSMs indicated, if more than one connection exists, all are specified. miR=>TF: we indicate which miR (out of the clusters from which the partner miR is transcribed) was found to have a binding site, and the RefSeq ID of the partner PSSM, in which the miR site is found. "TF unmapped" indicates that the PSSM could not be mapped to a SwissProt ID using the UCSC hgl7 tfbsConsFactors track.

Table 8

Table 8: miR-TF significantly co-occuring pairs in the PicTar network, 5kb promoters.

All the pairs in this table passed FDR of 0.3 or less in the hyper-geometric p- value for co- occurrence, and also received a co-occurrence p- value <10E-3 in the randomization-based test, if a pair also constitues an FFL, the following data is provided in the last columns: TF==> miR: we provide the PSSMS and the genomic miR cluster for which a binding site was found for the PSSMs indicated, if more than one connection exists, all are specified. miR=>TF: we indicate which miR (out of the clusters from which the partner miR is transcribed) was found to have a binding site, and the RefSeq ID of the partner PSSM, in which the miR site is found. "TF unmapped" indicates that the PSSM could not be mapped to a SwissProt ID using the UCSC hgl7 tfbsConsFactors track.

The Network of miR-TF Coregulation Reveals Recurring Local Architectures - Network Motifs It was suggested that in circuits composed of a miR and a TF, in which these two regulators target the same genes, the TF may also exert a regulatory effect on the miR with which it coregulates the target genes [Hornstein E, Shomron N (2006) Nat Genet 38 Suppl: S20-S24]. It was suggested that such a feed-forward loop (FFL), a well known local feature of many biological networks, may have a beneficial function. An FLL consisting of a TF and a miR could act as a switch for developmental and other programs in cells, since it may acquire biological systems with robustness to noise by means of canalization of perturbations. The present inventors wanted to check whether in any of the significant miR-TF partners discovered above, the miR and its partner TF may regulate each other. Accordingly, the present inventors determined how many of the TF-miR partner pairs (out of 104 pairs in the TargetScan dataset and 916 pairs in the PicTar dataset) had a conserved TF binding site of the partner TF, in the putative upstream regulatory region of the partner miR (see Materials and Methods for definition of miRs' up-stream putative regulatory regions). Interestingly, it was found that ten of the TF-miR pairs in the TargetScan dataset (9.6 % of the pairs), and 75 out of 916 pairs in the PicTar dataset (8.2 %) fulfilled that additional requirement (see Figure 4). To establish whether this rate was significant, a randomization test was carried out (see Materials and Methods) in which the present inventors computed, in 10,000 randomized sets of TF-miR pairs, the rate of formation of a regulatory interaction between the TF and the miR. In the TargetScan network a modest p- value of 0.024 was obtained, however in both PicTar networks the present inventors obtained the minimal possible p- value, <10^"4 , i.e., in all 10,000 randomizations a rate of direct regulatory interaction between a TF and the miR was obtained, which was lower than the to the original data (see corresponding z-scores in Figure 4). Thus, the cases in which a TF and a miR co-occur in a highly significant number of target genes was associated more often than random with a direct regulation between the TF and the miR's promoter. This motif was named "FFL TF-^miR". The significance of this motif is robust to "noise" in the input, assessed by the method originally used for network motifs in Escherichia coli [Shen-Orr SS, et al., 2002 Nat Genet 31: 64-68] (see Materials and Methods).

The present inventors were also interested in the opposite interaction — i.e., the case in which the miR regulates its partner TF. This was named "FFL miR^TF." The present inventors determined how many of the TF-miR partners had a predicted binding site of the partner miR in the 3' UTR of the partner TF; it occurred five times in the TargetScan network, and 42 and 48 times in the PicTar networks, using two cutoffs on gene regulatory region lengths. This rate was not significant in the TargetScan network (p- value = 0.16), yet it was significant in the PicTar networks (p-values 0.0038 and <10^"4). Interestingly, a composite loop network motif was also found, which was termed "FFL miR 4r ">TF", in which the pair of partners regulate each other, to be significantly over- represented in the PicTar network; it appeared seven times in the PicTar network (see Figure 4).

Information on FFLs is summarized herein below in Tables 9-12. Table 9

Table 9 - miR-TF FFLs in the TargetScan network.

All the pairs in this table passed FDR of 0.3 or less in the hyper-geometric p-value for cooccurrence, and also received a co-occurrence p-value <10E-3 in the randomization-based test, if a pair also constitues an FFL, the following data is provided in the last columns: TF=-> miR: we provide the PSSMS and the genomic miR cluster for which a binding site was found for the PSSMs indicated, if more than one connection exists, all are specified. miR==>TF: we indicate which miR (out of the clusters from which the partner miR is transcribed) was found to have a binding site, and the RefSeq ID of the partner PSSM, in which the miR site is found. "TF unmapped" indicates that the PSSM could not be mapped to a SwissProt ID using the UCSC hgl7 tfbsConsFactors track.

Table 10

Table 10 - miR-FFLs in the PicTar network, 10kb promoters.

All the pairs in this table passed FDR of 0.3 or less in the hyper-geometric p-value for co-occurrence, and also received a co-occurrence p-value <6*10E-3 (FDR q-value=0.3) in the randomization-based test, if a pair also constitues an FFL, the following data is provided in the last columns: TF=> miR: we provide the PSSMS and the genomic miR cluster for which a binding site was found for the PSSMs indicated, if more than one connection exists, all are specified. miR==>TF: we indicate which miR (out of the clusters from which the partner miR is transcribed) was found to have a binding site, and the RefSeq ID of the partner PSSM, in which the miR site is found. ¹¹TF unmapped" indicates that the PSSM could not be mapped to a SwissProt ID using the UCSC hgl7 tfbsConsFactors track.

Table 11

Table 11 - miR-TF FFLs in the PicTar network, 5kb promoters.

All the pairs in this table passed FDR of 0.3 or less in the hyper-geometric p- value for co-occurrence, and also received a co-occurrence p- value <10E-3 in the randomization- based test, if a pair also constitues an FFL, the following data is provided in the last columns: TF=> miR: we provide the PSSMS and the genomic miR cluster for which a binding site was found for the PSSMs indicated, if more than one connection exists, all are specified. miR==>TF: we indicate which miR (out of the clusters from which the partner miR is transcribed) was found to have a binding site, and the RefSeq ID of the partner PSSM, in which the miR site is found. "TF unmapped" indicates that the PSSM could not be mapped to a SwissProt ID using the UCSC hgl7 tfbsConsFactors track.

In the next step, the present inventors looked for another type of network motif, that was termed an "indirect FFL", in which the TF's regulation on its partner miR is exerted via another mediator TF. The present inventors looked to see if any of the miR-TF partners in the network had a conserved TF binding site in a promoter of at least one other TF, which in turn has a conserved binding site in the promoter of the partner miR. Significantly, this architecture was very common in the present networks; 30 of the TF- miR partners in the TargetScan network (28 %) and 201 partners in the PicTar network (22 %) were connected in a regulatory path between the TF and the miR via another TF. The significance of these results was tested by a randomization test, similar to that described above (see Materials and Methods), and received a p- value of 1.3 x 10^"3 for the appearance of the indirect FFL in the TargetScan network, and p- value < 10^~4 for the PicTar network (see Figure 4). Table 12 herein below lists pairs of indirect FFLs in the TargetScan database. Table 12

Table 13 herein below lists pairs of indirect FFLs in the PicTar database taking 0kb regulatory regions for protein coding genes.

Table 13

Table 14 herein below lists pairs of indirect FFLs in the PicTar database taking 5 kb regulatory regions for protein coding genes.

Table 14

Expression Analyses Supports miR-TF and miR-miR Predicted Regulatory Interactions: Next, the expression profiles of TF-miR partners were analyzed. Expression data across human tissues and organs has recently become available for miRs [Barad O, et al. (2004) Genome Res 14: 2486-2494] and is also available for protein coding niRNAs [Su AI, et al. (2004) Proc Natl Acad Sci U S A 101: 6062-6067]. Fortunately, for all the five healthy tissues (brain, liver, thymus, testes, and placenta) for which miRs expression was assayed, mRNAs were measured too. Thus the correlation coefficient was calculated between the expression profiles of each mRNA and each miR, and in particular between all TF-miR partners. For background statistics, correlations were first calculated between all pairs of miRs and TFs in the expression dataset (i.e., not necessarily the TF-miR partners identified above) and their distribution was obtained. It was found, as may be expected, a distribution that is centered on zero (Figure 5A). On this background the distribution of correlation coefficients between expression profiles of TF-miR partner pairs are shown

(Figure 5B and 5C). Strikingly, it was found that TF-miR partner pairs tended to have high correlation coefficients between them, but curiously, there was also a tendency for strong negative correlations in some of these pairs. These two tendencies were further enhanced when only the TF-miR pairs were insepected that are connected through an FFL. Given that some TFs can act as activators and others as repressors, and given that miRs may act at the level of translation inhibition or transcript degradation, both negative and positive correlations between TF-miR partners may be mechanistically rationalized.

The same miR tissue expression data was further used to shed light on the co- occurrence and avoidance of miR pairs. Pairs of miRs were tested that are either highly correlated in their expression levels or anticorrelated to each other across human samples have particularly high co-occurrence or avoidance p- values. An encouraging correspondence was found, whereby miR pairs that were positively correlated in expression had a significant tendency for high co-occurrence, whereas miRs with negative correlation in tissue expression typically tended to deliberately avoid residing in shared 3' UTRs. These observations provide experimental support for miR pairs and TF-miR regulatory interactions that were initially predicted based on sequence information alone. DISCUSSION The present example provides a comprehensive characterization of both global and local structural properties of the network of combinatorial regulatory interactions spanned by miRs and TFs. Extensive interactions were discovered between miRs and between miRs and TFs, and it was realized that thousands of human genes are subject to their regulatory effects. Inspection of the distributions of predicted miR sites in human genes' 3' UTRs revealed hundreds of target hubs in the human genome, genes that appear to be controlled by multiple regulators — miRs in the present case. Curiously, the current target hubs show highly nonrandom representation of specific gene functionalities. Particularly, genes related to development and genes that regulate transcription are enriched among the set of target hubs. These findings constitute another demonstration of the recent concept [Borneman AR, et al. (2006) Genes Dev 20: 435—448] that suggests that genes that exert extensive regulation on crucial processes are themselves often heavily regulated. So far this has been discussed in the context of the yeast transcription network; this study extends the scope of this concept to the case of miRs in mammalian genomes. In addition, given that each method of target prediction has its own rate of false positives, target hubs, which are predicted to be targeted by multiple miRs, are more likely to actually represent true targets of miR silencing.

The network of extensive regulatory interactions observed here between transcriptional regulators (TFs) and post-transcriptional regulators (miRs), is another interesting global feature. Altogether it was estimated that the number of human genes that are under combined regulation at the transcriptional and posttranscriptional silencing levels is between -1,000 and -4,000 (i.e., -12 % to -43 % of the -9,000 analyzed genes, according to the TargetScan and PicTar networks, respectively). Overall -9,000 genes were included in the present analyses, these are genes that are currently predicted to have at least one binding site for a known miR. Considering the fact that the collection of mammalian miRs is yet incomplete, and the fact that human specific miRs were not included in the analysis, it can be anticipated that the true number of human genes that are subject to a dual TF-miR regulation was underestimated in this study. For comparison, it was recently estimated that in the Saccharomyces cerevisiae genome about 13 % of the genes are subject to regulation at the combined transcriptional and posttranscriptional level, albeit with different mechanisms of posttranscriptional regulation operating in this organism, which does not have the miR silencing pathway.

The present inventors also examined local properties of the regulatory network, the network motifs. The network motifs described here are different from those originally described [Mangan S, Alon U (2003); Proc Natl Acad Sci U S A 100: 11980-11985; MiIo R, et al. (2002) Science 298: 824-827;Shen-Orr SS, (2002) Nat Genet 31: 64-68] in that they are composed of a TF and a miR instead of two TFs, as in the original case. It has been shown here that network motifs are not only significantly abundant, but also that, according to their current definition, each of them is involved in the regulation of large set of targets. Interestingly, TF and miR pairs that participate in network motifs show a significant tendency towards high tissue expression correlations or anticorrelations of the two regulators, providing essential experimental support to combinations predicted solely based on sequence information.

Motifs in which the miR regulates its partner TF constitute a type II coherent FFL [Mangan S, Alon U (2003) Proc Natl Acad Sci U S A 100: 11980-11985]. In this case it seems that a miR that silences a set of genes posttranscriptionally also silences the transcriptional regulator of these genes, presumably to also prevent de novo transcription of its target genes. This design may be used to minimize leaky transcription of genes in space and time when their expression is undesired. For example, this mechanism could be useful in determining developmental fate in differentiation boundaries.

The motifs in which the TF has a binding site in the promoter of its partner miR corresponds to the incoherent type I FFL (assuming that the TF is a positive regulator). Interestingly, in the S. cerevisiae transcription network this circuit is the second most highly abundant FFL [Mangan S, Alon U (2003) Proc Natl Acad Sci U S A 100: 11980- 11985]. An intriguing question is what may be the reason for the observed abundance of this circuit in which a TF regulates its partner miR? On the face of it, such regulation appears wasteful if the TF is a positive regulator, since the TF activates an entire set of genes and also a miR that may shut those target genes down. However, if a temporal gap in the activation time of the target genes and the miR exists, then the circuit may be utilized for useful regulatory purposes. For instance if the TF activates first the target genes and only later the miR (e.g., due to higher affinity, during a process in which the TF's concentration builds up, the activation of the miR may be timed to obtain a desired delayed shutdown of the regulated genes. Similar wiring in the cases of antisense RNAs, another type of regulatory transcripts, and TFs that regulate them in conjunction with their overlapping sense transcripts have also been considered. The opposite situation, in which the TF positively activates the miR first and only later the target gene, may also be of interest as it can act as a buffer for noisy fluctuations in the levels of the targets; as long as the mRNA level of the target gene is below the inhibition capacity of the miR, fluctuations in its expression levels would not be further propagated. Further, in cases where the miR works predominantly as a translation inhibitor, a controlled mechanism for "just in time" translation for multiple genes is needed for certain functionalities. For example, the miR translation inhibition mechanism was suggested to facilitate localized translation in mammalian dendrites, and to play a crucial role in synaptic plasticity. Such a circuit of coregulating TF-miR in an FFL, where the miR is transcribed by the TF in parallel to the set of mutual targets, could function in featuring localized translation to a whole pathway of regulated genes. Interestingly though, an example of one indirect FFL can be pointed out, where a brain related TF, CREB (CREB ATF), partners with a miR that is known to be expressed in the brain, miR-125b. CREBATF was predicted to regulate miR-125b through STAT3, which interestingly is also within the list of mutual targets of both miR125b and CREBATF, indicating an even more complex design.

One of the FFLs that came out of the present analysis is a composite loop in which the TF regulates the miR and the miR appears to regulate the TF (i.e., a TF^"*miR motif). The circuit consists of the TF E2F and miR-93. miR-93 is part of a cluster of three miRs, miR-106b, miR-93 and miR-25, which lie in close proximity to each other inside an intron of the MCMJ gene. This network motif was found as an FFL TF -^ miR in the TargetScan network and as a composite loop in the PicTar network, where all three miRs in the cluster were predicted to target E2F (specifically E2F1 and E2F3). miR-93 cluster members are also homologous to two other genomic miR clusters, one of which is miR cluster 17/92. Recent evidence suggests a tight regulatory connection of cluster miR- 17/92 and E2F. E2F1, 2, and 3 were shown to directly upregulate the expression of the miRs encoded in this cluster, while these miRs in turn were shown to act in a feedback loop and target E2F1-3 mRNAs. It was suggested that this feedback may play a role in the major decision mediated by E2F (induction of cellular proliferation or apoptosis). The present data suggests that this intricate regulatory circuit might have another layer to it; in addition to being targeted by the miR- 17/92 cluster, E2F family genes might also be targeted by miR-93 cluster members, which share similar seeds. In turn, the miR-93 cluster is transcribed from an intron of the MCMl host gene, which is a verified target of the E2F family. Moreover, here the architecture is more complex, as it also includes a set of mutual target genes, through which E2F and the miR-93 cluster may exert their regulatory roles.

EXAMPLE 3 p53-repressed miRNAs are involved with E2F in a Feed Forward Loop promoting proliferation

MATERIALS AND METHODS

Cell culture: WI-38, MRC5, IMR90 (Obtained from the ATCC), and PFCAl 79 cells were cultured in MEM with 10 % FCS, 1 mM sodium pyruvate, 2 mM L-glutamine, and antibiotics. U2OS and Hl 299 cell lines were cultured in DMEM and RPMI, respectively, with 10 % FCS and antibiotics. MCFlOA cells were maintained in DMEM

F12 supplemented with 5 % horse serum, 0.5 μg/ml hydrocortisone, 0.1 mg/ml insulin, 0.1 μg/ml cholera toxin, and 10 ng/ml EGF. All cells were maintained in a humidified incubator at 37 ⁰C and 5 % CO₂. Primary fibroblasts were passaged every 5-6 days. PDLs were calculated using the formula: PDLs=log[cell output/cell input]/log2. For colony formation assays, IxIO³ cells were seeded in 100 mm dishes, grown for 14 days and stained with crystal violet. Plasmids and retroviral infections: GSE56 was subcloned from pBabe-GSE56- puro (Ossovskaya et al, 1996, Proc Natl Acad Sci U S A 93: 10309-10314) into pLXSN- Neo. Small hairpin RNAs (shRNAs) targeting p53 (p53i) or mouse NOXA (Control shRNA) were stably expressed using pRetroSuper (Berkovich and Ginsberg, 2003, Oncogene 22: 161-167). ER-E2F1 was described in (Vigo et al, 1999, MoI Cell Biol 19: 6379-6395). ElA was expressed from pBabe-puro-ElA12S. For expression of miR- 106b/93/25, a 1 kb human genomic fragment was cloned with the primers 5'- ggatcctatcctgcgcctttcc-3' (SEQ ID NO: 1) and 5'-cacatggccacagaagac-3' (SEQ ID NO: 2) into miR-Vec (Voorhoeve et al, 2006, Cell 124: 1169-1181). Retrovirus infection procedures were described in (Milyavsky et al, 2003, Cancer Res 63: 7147-7157).

RNA preparation and quantitative real-time PCR (QRT-PCR): RNA was extracted with TRI-Reagent (Molecular Research Center, Inc.). For mRNA quantification, a 2 μg aliquot of total RNA was reverse transcribed using Bio-RT (BIO LAB) and random hexamers. QRT-PCR was performed using Platinum SYBR Green qPCR SuperMix (invitrogen). mRNAs levels were normalized to GAPDH. Primer sequences are listed in Table 15 herein below. For miRNA quantification, TaqMan miRNA assays (Applied Biosystems) were used according to manufacturer protocol. Levels were normalized to the U6 control. AU QRT-PCR reactions were performed on ABI7300 machine. Results are presented as mean and standard deviation for two duplicate runs. Table 15

miRNA microarrays, data analysis and clustering: The miRNA profiling presented in figure IA was performed as follows: RNA was extracted from WI-38 cells using TRI-Reagent as described above, labeled with Hy5 and hybridized on Exiqon's miRCURY™ LNA Array (v.8.1) with a common reference Hy3-labled RNA pool. Two biological replicates were performed for each sample type. Hy5/Hy3 ratios were Iog2 transformed and filtered such that miRs which were undetected in 11 or 12 samples were discarded. Duplicates were averaged, such that each miR was represented by six values, corresponding to the six different samples. For each miR, a credibility value was calculated as one minus the average of the six standard deviations (SD) between the duplicates. A duplicate that had one missing value was set as the detected value and was assigned with high SD. The 5 % most non-credible miRs were discarded. Data was clustered using hierarchical clustering (average linkage), with 20 clusters.

Analysis of miRNA targets expression coherence: The entire set of miRNA expression profiles was clustered into 20 clusters based on the above expression data (WI- 38 young vs. senescent, along with p53 inactivation). Then, a set of predicted target for the miRs from each cluster using PicTar was compiled (Krek et al, 2005, Nat Genet 37: 495- 500). Specifically, for each of the 20 miR clusters a series of potential sets of targets was created. The first set consisted of mRNAs predicted to be targeted by at least one miR from the cluster. The second set consisted of mRNAs predicted to be targeted by at least two miRs from the cluster, and so on. The expression coherence (EC) score, a measure of expression similarity (Pilpel et al, 2001, Nat Genet 29: 153-159), was then computed for each set of targets according to their expression described in Milyavsky et al. (2005), Cancer Res 65: 4530-4543. The most significantly coherent expression pattern belonged to the set of genes that had target sites for at least five miRs from the 'p53 -repressed miR cluster' (EC p-value = 5x10-3).

Immunoblot analysis: Western blots were performed as described in (Milyavsky et al, 2005, Cancer Res 65: 4530-4543). The following antibodies were used: α-p53 pAb H- 47, α-p21 sc-377 (Santa Cruz), α-E2Fl sc-193 (Santa Cruz), α-GAPDH MAB374 (Chemicon), α-pl30 sc-317 (Santa Cruz), α-p57 sc-8298 (Santa Cruz), α-pRb 554136

(Pharmingen), α-β-tubulin T7816 (Sigma).

Cell-cycle analysis: Cells were labeled for 30 minutes with 10 μM BrdU (Sigma), fixed with 70 % EtOH/HBSS (2 hours, -20 ⁰C), treated with 2M HCl/0.5 % Triton, washed and treated with 0.1M Na₂B₄O₇ pH 8.5, and stained with FITC-conjugated anti-BrdU (Becton Dickinson) and 10 μg/ml propidium iodide. Samples were analyzed using a FACSort machine (Becton Dickinson). At least IxIO⁴ events were recorded per sample.

Senescence-associated beta-galactosidase (SA-β-Gal) activity assay: Cells were fixed with 3 % formaldehyde/PBS for 5 minutes, washed with PBS and incubated for 16 hours at 37 ⁰C with a solution containing 1 mg/ml X-gaI/40 niM citric acid, sodium phosphate, pH 6.0/5 niM potassium ferrocyanide/5 mM potassium ferricyanide/150 niM NaCl/2 mM MgCl₂. RESULTS In order to identify novel p53-regulated miRs two isogenic cell cultures were established that differ in their p53 status and analyzed their miRNA profiles both under normal conditions as well as in contexts that involve p53 activation. WI-38 primary human fibroblasts were infected with a retrovirus encoding for the p53-inactivating peptide, GSE56 (Ossovskaya et al, 1996, Proc Natl Acad Sci U S A 93: 10309-10314). These cells (GSE) and their active p53 counterparts (Con) were treated with the DNA damaging agent doxorubicin or grown until the onset of replicative senescence (the establishment of the system is depicted in Figures 1 OA-F). Analysis of miRNA expression patterns revealed several expression clusters (see Materials and Methods). Notable among these was a cluster populated with miRs whose expression was negatively regulated by p53 at basal levels. The cluster showed additional downregulation in senescent cells, which was attenuated upon p53 inactivation. This cluster was named the ^cp53 -repressed miR cluster' (Figure 11). Notably, doxorubicin treatment, which resulted in a considerable activation of p53 and its mRNA targets (Figures 1 OA-F), did not significantly affect the levels of these miRs.

The presented cluster contain families ofparalogous cancer-related miRNΛs Some of the miRs represented in the cluster (Figure 11) are transcribed from three homologous genomic loci, previously reported as paralogs that evolved from a common evolutionary origin (Tanzer and Stadler, 2004). These include miR-106b/93/25 that reside within an intron of the cell-cycle gene 'minichromosome maintenance protein T (MCMT); miR-17/18a/19a/20a/19b-l/92a-l (miR-17-92 polycistron) that are transcribed as the non- coding RNA cl3orf25; and miR-106a/18b/20b/19b-2/92-2 (miR-106a-92 polycistron) that are clustered on chromosome X. This data indicates that not only were the miRNA sequences and genomic organization conserved during evolution, but also their transcriptional regulation. Another polycistron, comprising miR-15b and miR-16, is also represented in the cluster.

Many members of the cluster are overexpressed in various tumors; consistent with the frequent p53 loss of function in cancer, and some were shown to possess oncogenic functions. For example, miR-106a, miR-17-5p, miR-20a and miR-155, were reported to be commonly overexpressed in solid tumors (Volinia et al, 2006, Proc Natl Acad Sci U S A 103: 2257-2261). Members of the miR-17-92 polycistron are overexpressed in lymphomas as well as in lung and colorectal carcinomas and were shown to accelerate tumor growth. Interestingly, the MCM7 gene that contains three of the clusters' miRs in its intron (miR-106b/93/25) is amplified or overexpressed in diverse types of cancers, as are its resident miRs. Consistently, the miR-106b/93/25 polycistronic members were suggested to promote cell cycle progression. Finally, miR-155 and its host non-coding RNA (BIC) were reported to be specifically overexpressed in several types of B-cell lymphomas and to predict poor prognosis in lung cancer.

Representative miRNAs show p53-dependent repression during senescence in many cell types: To further validate the present data, two additional human isogenic cell culture pairs were generated from IMR90 lung primary fibroblasts and from prostate- cancer associated fibroblasts (CAFs). Each culture was infected with a retrovirus encoding for either a small hairpin RNA targeting p53 (ρ53i) or a control RNAi (Con), and grown until the onset of replicative senescence. p53 knock-down, which significantly reduced the mRNA and protein levels of both p53 and its target p21, delayed the onset of senescence by approximately ten population doublings (Figure 12A and Figures 13 A-D). For these cell types, as well as for the WI-38 cells, the levels of representative miRs were compared using TaqMan miRNA assays. Analyses of miR-106b and miR-17-5p, as well as of their host transcripts MCM7 and cl3orf25, respectively, revealed transcriptional repression upon replicative senescence in all three tested cell cultures in a manner that was partially or completely p53-dependent (Figure 12B). Additionally, the non-coding RNA BIC and its resident miR-155 were also transcriptionally repressed in a p53-dependent manner upon replicative senescence (Figure 13D). The miRNAs are associated with p53 and E2F in a proliferation-related regulatory network: A mRNA 'proliferation cluster' was previously reported that consists mainly of cell-cycle related genes (Milyavsky et al, 2005, Cancer Res 65: 4530-4543). This cluster emerged from an mRNA profiling of an in-vitro transformation process in which primary WI-38 cells were gradually transformed, resulting in tumorigenic cells. Importantly, the 'proliferation cluster^'' is one of the most prominent expression signatures revealed when tumors are compared to normal tissues or when highly proliferating cells are compared to slow growing cells, and contains many cell-cycle periodic genes. The expression pattern of the 'proliferation cluster" is highly similar to that of the 'p53- repressed miR cluster '; i.e. the 'proliferation cluster' mRNAs display p53-dependent downregulation. The similarity in expression patterns prompted the present inventors to hypothesize that both clusters share a common transcriptional program. It was previously shown that the p53 -mediated repression of the 'proliferation cluster' was mediated via E2F (Tabach et al, 2005, MoI Syst Biol 1: 2005 0022). Providing further support, a conserved E2F binding site is found upstream of the three polycistronic miRs. Recently, it was shown that the miR- 17-92 and the miR-106b/93/25 polycistrons are transcriptionally activated by E2F family members (O'Donnell et al, 2005, Nature 435: 839-843; Petrocca et al, 2008, Cancer Cell 13: 272-286; Sylvestre et al, 2007, J Biol Chem 282: 2135-2143; Woods et al, 2007, J Biol Chem 282: 2130-2134). In view of all the above, it appears conceivable that these miRs are transcriptionally activated by E2F, and that p53 exerts its repression via E2F inhibition. In agreement with the observed downregulation of the 'p53 -repressed miR cluster' in senescence, it was shown that E2F activity is significantly downregulated in senescent cells (Campisi and d'Adda di Fagagna, 2007, Nat Rev MoI Cell Biol 8: 729-740). In addition, the miRs presented here are proposed to be novel members of the well established 'proliferation cluster'.

The p53-dependent repression of miRNAs from the cluster is mediated via E2F1: To experimentally test whether the miRs and their host mRNAs are regulated by E2F1 in primary cells, WI-38 cells were infected with a retrovirus encoding for an E2F1 protein fused to a modified estrogen receptor ligand binding domain (ER). Treatment of ER-E2F1 expressing cells with 4-OHT permits ER-E2F1 translocation to the nucleus, thereby inducing its transcriptional activity. As depicted in Figure 14A, following 4-OHT treatment an upregulation of candidate miRNAs and host mRNAs which were part of the cluster were observed and together represent all three paralogous polycistrons. Specifically, the MCM7 gene and its resident miRNAs miR-106b, miR-93 and miR-25, the non coding RNA cl3orf25 and its resident miR-17-5p as well as miR-106a that represents the miR-106a-92 cluster. Upregulation of MCM7 and its resident miRs following 4-OHT treatment was also observed in ER-E2F1 expressing lung carcinoma cells (H1299) and osteosarcoma cells (U2OS) (Figures 15 A-B). Next, WI-38 cells were infected with ElA, a viral oncoprotein that disrupts pRb-E2F complexes and leads to an upregulation of the endogenous E2F activity. As expected, ElA overexpression resulted in elevated levels of all the above mentioned representative miRNAs (Figure 14B). Of note, the levels of miR- 155, which belongs to the immune response co-cluster, were not affected by E2F activation.

Having shown that representative miRs are activated by E2F1 in the present system, p53 -dependent repression was tested to analyze whether it is mediated via modulation of E2F1 activity. To that end, WI-38 cells were infected with a retrovirus encoding for either a small hairpin RNA targeting p53 (p53i) or a control RNAi (Con) and treated them with Nutlin-3, a small molecule that stabilizes the p53 protein by inhibiting its Mdm2-dependent ubiquitylation and degradation. Nutlin treatment resulted in a robust p53 protein accumulation, accompanied by p21 mRNA and protein induction (Figures 16 A-B), which was completely abrogated in the p53i cells. Remarkably, E2F1 mRNA and protein levels were downregulated upon Nutlin treatment in a p53 -dependent manner. Cyclin E showed a similar pattern, supporting the notion that E2F1 downregulation was accompanied by a reduction in E2F activity. Accordingly, MCM7 and its resident miR- 106b were both downregulated in a p53 -dependent manner (Figure 16A) along with other miRs from the cluster but not with the immune-response related miR-155 (data not shown). Thus, Nutlin treatment, a non genotoxic p53 activating signal, resulted in a p53- dependent transcriptional repression of mRNAs and miRNAs with associated cell-cycle functions. To substantiate the causal relationship between the p53-depednent reduction of E2F1 activity and the repression of the miRs and their hosts control and ElA expressing WI-38 cells were treated with Nutlin. As depicted in Figures 16C-D, ElA induced the expression and prevented the Nutlin-dependent repression of E2F1 as well as of its target Cyclin E. Most significantly, ElA overexpression abolished the downregulation of MCM7 and miR-106b upon Nutlin treatment. A similar pattern was observed for miR-17-5p and its host cl3orf25 (data not shown). It may therefore be concluded that E2F1 repression by p53 is necessary for the downregulation of MCM7 and miR-106b. The same mechanism may underlie the p53 -dependent downregulation of additional miRs from the cluster and, more specifically the three paralogous polycistrons.

The miRNAs target key cell-cycle regulators and affect pivotal characteristics of proliferation: Next, the present inventors set out to identify the functions of the p53- repressed miRs. The miR-106b/93/25 polycistron was focused on as a representative member of the large family of miRs that includes also the miR- 17-92 and miR-106a-92 polycistrons. The genomic region encoding miR- 106b, miR- 93 and miR-25 was overexpressed, which corresponds to an intron of the MCM7 gene in young WI-38 cells and in MCFlOA mammary epithelial cells, both characterized by low basal expression of these miRs. Following the prediction of E2F and miR-106b/93/25 involvement in a feed forward loop, in which they both target a mutual set of genes (Example 1), a list of their mutual predicted targets was compiled as set forth in Table 16 herein below.

Table 16

Interestingly, many of these predicted targets participate in cell-cycle regulation (p- value for "cell cycle" annotation=1.4xlO^"10). The protein levels of selected predicted targets in the miR-106b/93/25 overexpressing cells were measured (Figures 17 A-D). A downregulation of p21, was observed, as well as of pRB and pi 3O₅ which were previously suggested, based on reporter assays, as potential targets of miR-106a and the miR- 17-92 cluster, respectively. Interestingly, E2F1, which was shown to be a target of miR-17-5p and miR-20a, and is predicted by PicTar to be a target for both miR- 106b and miR-93, was significantly downregulated as well. A downregulation of p57 was also observed in WI-38 cells, in agreement with PicTar predictions (Figure 17A and Figure 18). Notably, these proteins have defined functions in the regulation of the cell-cycle, most of them being negative regulators of proliferation. Since the mRNA levels of p57, p21, pRb and pi 30 did not decrease in WI-38 cells and only marginally in MCFlOA cells (Figure 17B), the reduction in their protein levels most likely stems from translational inhibition and not mRNA degradation. In contrast, E2F1 mRNA levels were reduced in both cell lines that express the miR- 106b/93/25 polycistron.

Next, the present inventors set out to obtain a global view of the behavior of the predicted targets of the 'p53-repressed miR cluster'. The expression profiles of these targets in the system described above were analyzed, where primary WI-38 cells were gradually transformed into tumorigenic cells. Interestingly, targets harboring predicted sites in their 3'-UTR for multiple miRs had a significant coherent expression pattern (Figure 17C, EC p-value=5xlθ^"3, see Table 17, herein below for the target list).

Note - Table 17 lists only mRNAs that were detected by the microarrays published by Milyavsky et al, 2005, Cancer Res 65: 4530-4543. Target predictions are based on PicTar. Table 17

The expression of this target set was decreased when cells gained the accelerated proliferation phenotype (designated as "Fast Growing"), i.e. most of these targets showed a clear antiproliferative expression pattern. Promoter analysis of these predicted target set revealed that one of the most highly enriched motifs was E2F (p-value=2.2xlθ^"u). Hence, genes heavily targeted by the p53-repressed miRs are characterized by anti-proliferative behavior. This may suggest that the feed-forward loop that consists of E2F and miR- 106b/93/25 may be de-regulated during cancer progression, by downregulation of these anti-proliferative targets both transcriptionally and post-transcriptionally. Having shown the molecular effects of the overexpression of miR-106b/93/25, the present inventors tested whether proliferation-related parameters such as growth rate, colony formation efficiency and replicative senescence are affected by these miRs. As these miRs are significantly repressed by p53 during senescence, and considering the fact that they target several anti-proliferation regulators, it was predicted that their overexpression, similarly to p53 inactivation, would accelerate growth rate and delay senescence. Indeed, as depicted in Figure 19A-D, the miR-106b/93/25 overexpressing WI- 38 cells demonstrated a moderate acceleration in proliferation rate and an increased fraction of S-phase cells (24 % compared to 18 %). Strikingly, these cells displayed a pronounced increase in the efficiency of young cells to form colonies when seeded at low density and reduced SA-β-Gal staining at late passages, indicating a delay of replicative senescence. These observed phenotypes, which mimic the effect of p53 inactivation in primary cells (Figures 10A-F and Figures 13 A-D), suggest that the transcriptional repression of miR-106b/93/25 and their paralogs mediates part of the anti-proliferative effects of p53. Although the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art. Accordingly, it is intended to embrace all such alternatives, modifications and variations that fall within the spirit and broad scope of the appended claims. All publications, patents and patent applications mentioned in this specification are herein incorporated in their entirety by into the specification, to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated herein by reference. In addition, citation or identification of any reference in this application shall not be construed as an admission that such reference is available as prior art to the present invention. To the extent that section headings are used, they should not be construed as necessarily limiting. 238 Although the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art. Accordingly, it is intended to embrace all such alternatives, modifications and variations that fall within the spirit and broad scope of the appended claims.

All publications, patents and patent applications mentioned in this specification are herein incorporated in their entirety by into the specification, to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated herein by reference. In addition, citation or identification of any reference in this application shall not be construed as an admission that such reference is available as prior art to the present invention. To the extent that section headings are used, they should not be construed as necessarily limiting.

Claims

239 WHAT IS CLAIMED IS:

1. A method of identifying components of a biological pathway, the method comprising selecting a transcription factor and a microRNA pair which regulate a common gene, said transcription factor and said microRNA being the components of the biological pathway.

2. The method of claim 1 , wherein said transcription factor and said microRNA pair are listed in Tables 5-8.

3. The method of claim 1, wherein a gene encoding said microRNA comprises a binding site for said transcription factor and/or a gene encoding said transcription factor comprises a binding site for said microRNA.

- 4. The method of claim 3, wherein said transcription factor and said microRNA pair are listed in Table 9-11.

5. The method of claim 3, wherein said transcription factor and said microRNA pair are listed in Figure 4.

6. The method of claim 1, wherein said transcription factor regulates a transcription of an additional transcription factor, said microRNA comprising a binding site for said additional transcription factor.

7. The method of claim 6, wherein said transcription factor and said microRNA are listed in Tables 12-14.

8. A method of treating a hyperproliferative disease in a subject, the method comprising administering to the subject a therapeutically effective amount of an oligonucleotide agent capable of down-regulating at least one microRNA selected from the group consisting miR-106b, miR-93, miR-25, miR-17, miR-18a, miR-19a, miR-20a, miR- 19b-l, miR-92a-l, miR-106a, miR-18b, miR-20b, miR-19b-2, miR-92-2*, hsa-miR-19b- 1* miR-15b, miR-16, hsa-miR-92a-2*, and hsa-miR-92a-l* into the subject, thereby treating the hyperproliferative disease. 240

9. The method of claim 8, wherein the hyperproliferative disease is cancer.

10. A method of treating a degenerative disease in a subject, the method comprising administering to the subject a therapeutically effective amount of at least one microRNA selected from the group consisting of miR-106b, miR-93, miR-25, miR-17, miR-18a, miR-19a, miR-20a, miR-19b-l, miR-92a-l, miR-106a, miR-18b, miR-20b, miR- 19b-2, miR-92-2*, hsa-miR-19b-l* miR-15b, miR-16, hsa-miR-92a-2*, and hsa-miR-92a- 1* into the subject, thereby treating the degenerative disease.

11. A pharmaceutical composition comprising a pharmaceutically acceptable carrier and as an active agent at least one microRNA selected from the group consisting of miR-106b, miR-93, miR-25, miR-17, miR-18a, miR-19a, miR-20a, miR-19b-l, miR-92a- 1, miR-106a, miR-18b, miR-20b, miR-19b-2, miR-92-2, miR-15b and miR-16.