Compositions for downregulating ZEB2 in macrophages and uses thereof

Abstract

Compositions for downregulating Zeb2 in macrophages are provided. Accordingly, there is provided a composition comprising an agent capable of downregulating expression and / or activity of Zeb2 attached to a macrophage-targeting moiety or encapsulated in a particle comprising a macrophage-targeting moiety. Also provided are methods of prognosing and treating an inflammatory disease.

Description

[0001] COMPOSITIONS FOR DOWNREGULATING ZEB2 IN MACROPHAGES AND USES

[0002] THEREOF

[0003] RELATED APPLICATION / S

[0004] This application claims the benefit of priority of U.S. Provisional Patent Application No. 63 / 667,732 filed on July 4, 2024, the contents of which are incorporated herein by reference in their entirety.

[0005] SEQUENCE LISTING STATEMENT

[0006] The XML file, entitled 103916. xml, created on July 3, 2025, comprising 98,437 bytes, submitted concurrently with the filing of this application is incorporated herein by reference.

[0007] FIELD AND BACKGROUND OF THE INVENTION

[0008] The present invention, in some embodiments thereof, relates to compositions for downregulating Zeb2 in macrophages and uses thereof.

[0009] Cancer remains a leading cause of mortality worldwide, posing significant challenges to healthcare systems and affecting millions of individuals annually. Traditional cancer treatments, including chemotherapy, radiation therapy, and surgical interventions, have achieved varying degrees of success but are often associated with substantial side effects and limitations in efficacy, particularly in advanced or metastatic stages.

[0010] It is known that cancer cells evade the immune system through various mechanisms, such as downregulating antigen presentation, secreting immunosuppressive factors, and inducing regulatory T cells (Tregs) that inhibit anti-tumor immune responses. Given these challenges, immunomodulation has been suggested as a promising approach to enhance the immune system's ability to recognize and destroy cancer cells. Recent advancements in immunotherapy, particularly immune checkpoint inhibitors and adoptive cell transfer, have demonstrated the potential of harnessing the immune system to achieve durable responses in a subset of patients with several cancer types.

[0011] Macrophages are ubiquitous throughout the human body and remarkably represent one of the most functionally versatile cell types1. By identifying and neutralizing foreign and endogenous threats and phagocytosing cellular material, macrophages play a crucial role in maintaining tissue homeostasis2. Beyond these core functions, macrophages are integral to multiple intricate physiological processes, including extracellular matrix remodeling, metabolic regulation, neuronal nurturing in the central nervous system, and electroconductive regulation of the heartbeat3. This plasticity, enabled through sensing of diverse tissue signals, allows them to adapt to tissue- specific cues while preserving their fundamental macrophage identity4. In the tumor microenvironment (TME), monocytes and macrophages sense the unique tumor signals, including diverse tumor-associated cytokines, cellular stress, hypoxia, and cell death, through a multitude of sensors and regulatory circuits. These intricate pathways orchestrate their differentiation into tumor-associated macrophages (TAMs), key components of the TME, and typically play a pivotal role in the promotion of cancer progression and metastasis5,6. By facilitating processes such as immunosuppression, tissue remodeling, angiogenesis, and tumor cell invasion, TAMs actively contribute to tumor growth and the establishment of metastasis7 9. Consequently, increased monocyte infiltration and TAM differentiation in solid tumors are correlated with unfavorable prognosis, as well as resistance to conventional chemotherapy and immunotherapies7, 10 l 2. In contrast, certain types of TAM populations in the TME are associated with better survival of patients and response to anti-cancer therapies13. This indicates that TAM reprogramming bears the potential for novel classes of druggable targets to enhance anti-tumor immunity14,15.

[0012] Thus far, clinical trials targeting TAMs as monotherapy or in combination have fallen short in demonstrating clinical efficacy in patients with solid tumors. This, in part, can be attributed to an inadequate understanding of TAM checkpoints, a lack of effective biomarkers, and insufficient knowledge of the expected outcome of the targeting agent on TAM reprogramming. The commonly used classification system for functional macrophage polarity, known as the M1 / M2 model, was introduced approximately two decades ago. It divides macrophages into two distinct categories: Ml, known as 'classically' activated, associated with anti-tumoral phenotype, and M2, termed 'alternatively' activated, associated with immune suppressive / pro-tumorigenic phenotype. This terminology, although effective, oversimplifies the functional diversity of TAMs and the diverse functional effects of targeting different checkpoints on their gene programs. Recent advances in single-cell technologies unveiled some of these complexities, revealing the modular expression of several functional TAM programs, such as antigen presentation, phagocytosis, inflammation, immune regulation, and hypoxia16 I S. In line with the modular regulation of gene programs, expression of Ml and M2 gene signatures has been demonstrated to co-exist even in the same individual cells and across different tissues18. Moreover, while the conventional Ml and M2 markers failed to accurately predict clinical outcomes in patients, more refined segregation of TAM subsets using scRNA-seq data from large clinical patient cohorts identified CXCL9 versus SPP1 TAM polarity as an improved classifier for the prediction of patient survival13. A significant unmet need in the field is the knowledge of the molecular circuitry controlling TAM functions and the capacity to identify TAM reprogramming factors, ultimately shifting their phenotype from pro- to anti-tumor immunity.

[0013] Additional background art includes:

[0014] US Patent No. US9511050;

[0015] US Patent Application Publication No. US20080214436; and

[0016] Kortylewski, M., Swiderski, P., Herrmann, A. et al. (2009) Nat Biotechnol 27: 925-932.

[0017] SUMMARY OF THE INVENTION

[0018] According to an aspect of some embodiments of the present invention there is provided a composition comprising an agent capable of downregulating expression and / or activity of Zeb2 attached to a macrophage-targeting moiety or encapsulated in a particle comprising a macrophage-targeting moiety.

[0019] According to some embodiments of the invention, the agent is a polynucleotide.

[0020] According to some embodiments of the invention, the polynucleotide is an RNA silencing agent.

[0021] According to some embodiments of the invention, the RNA silencing agent is an siRNA or shRNA.

[0022] According to some embodiments of the invention, the RNA silencing agent is a microRNA.

[0023] According to some embodiments of the invention, the microRNA is selected from the group consisting of MiR-30a-5p, MiR-101, MiR-124, MiR-129, MiR-132, MiR-138, MiR-139- 5p, MiR-141, MiR-144, MiR-145, MiR-145-5p, MiR-153, MiR-154, MiR-155, MiR-187, MiR- 200a, MiR-200b, MiR-200c, MiR-203, MiR-205, MiR-206, MiR-211-5p, MiR-215, MiR-335, MiR-338-3p, MiR-454-3p, MiR-374b-5p, MiR-448, MiR-506, MiR-545, MiR-598, MiR-622, MiR-590-3p, MiR-769-3p, MiR-940, MiR-1179 and MiR-3653.

[0024] According to some embodiments of the invention, the agent is a genome editing agent.

[0025] According to some embodiments of the invention, the agent is a small molecule.

[0026] According to some embodiments of the invention, the small molecule is Teniposide.

[0027] According to some embodiments of the invention, the agent is a PROTAC.

[0028] According to some embodiments of the invention, the PROTAC is SR-1114 or SPFl bxil4-\

[0029] According to some embodiments of the invention, the macrophage-targeting moiety is specific for a macrophage surface marker.

[0030] According to some embodiments of the invention, the marker is a mannose receptor or a scavenger receptor. According to some embodiments of the invention, the marker is selected from the group consisting of toll-like receptor, CDl lb, CD68, CD163, TREM2, MSR1 and MRC1.

[0031] According to some embodiments of the invention, the marker is a toll-like receptor.

[0032] According to some embodiments of the invention, the toll-like receptor is TLR-9.

[0033] According to some embodiments of the invention, the macrophage-targeting moiety comprises a ligand or receptor of the marker.

[0034] According to some embodiments of the invention, the macrophage-targeting moiety comprises mannose.

[0035] According to some embodiments of the invention, the macrophage-targeting moiety comprises a CpG ODN.

[0036] According to some embodiments of the invention, the macrophage-targeting moiety comprises an antibody.

[0037] According to some embodiments of the invention, the agent is attached to the macrophage-targeting moiety via a linker.

[0038] According to some embodiments of the invention, the linker comprises a C3 spacer.

[0039] According to some embodiments of the invention, the particle is a lipid-based nanoparticle.

[0040] According to an aspect of some embodiments of the present invention there is provided a method of treating an inflammatory disease in a subject in need thereof, the method comprising administering to the subject a therapeutically effective amount of the composition, thereby treating the inflammatory disease in the subject.

[0041] According to an aspect of some embodiments of the present invention there is provided the composition, for use in treating an inflammatory disease in a subject in need thereof.

[0042] According to some embodiments of the invention, the method further comprising administering to the subject an anti-inflammatory agent.

[0043] According to some embodiments of the invention, the composition for use further comprising an anti-inflammatory agent.

[0044] According to some embodiments of the invention, the anti-inflammatory agent is Bacillus Calmette-Guerin (BCG).

[0045] According to some embodiments of the invention, the subject is characterized by a level of macrophages above a predetermined threshold in a biological sample obtained from the subject.

[0046] According to some embodiments of the invention, the subject is characterized by a level of Zeb2 above a predetermined threshold in the biological sample. According to some embodiments of the invention, the method further comprising determining a level of macrophages in a biological sample of the subject.

[0047] According to some embodiments of the invention, the method further comprising determining a level of Zeb2 in the biological sample.

[0048] According to an aspect of some embodiments of the present invention there is provided a method of prognosing an inflammatory disease in a subject, the method comprising determining a level of macrophages and a level of Zeb2 a biological sample of a subject diagnosed with the inflammatory disease, wherein the level of the macrophages and the level of the Zeb2 above a predetermined threshold is indicative of poor prognosis, thereby prognosing the inflammatory disease in the subject.

[0049] According to some embodiments of the invention, wherein a level of macrophages positive for the Zeb2 above a predetermined threshold is indicative of poor prognosis.

[0050] According to some embodiments of the invention, wherein the biological sample comprises a biopsy.

[0051] According to some embodiments of the invention, the level of macrophages is determined by expression of CD68.

[0052] According to some embodiments of the invention, the inflammatory disease is cancer.

[0053] According to some embodiments of the invention, the cancer is a solid cancer.

[0054] According to some embodiments of the invention, the cancer is selected from the group consisting of bladder cancer, lung cancer, stomach cancer, acute-myeloid leukemia and, colorectal cancer.

[0055] According to some embodiments of the invention, the cancer is selected from the group consisting of bladder and colon cancer.

[0056] According to some embodiments of the invention, the cancer is bladder cancer, administration is via intravesical injection.

[0057] 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 SEVERAL VIEW OF THE DRAWINGS

[0058] 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.

[0059] In the drawings:

[0060] FIGs. 1A-H demonstrate the unified atlas of tumor-associated macrophages programs. Figure 1A is a UMAP visualization of scVI embedding on 830,107 single cells integrated across 123 patients from breast, colon, and lung tumors and adjacent normal samples. Major cell-type lineages were annotated by conventional markers. Figure IB is a UMAP visualization of re- integrated sc VI embedding on 108,951 single cells from the myeloid cluster (Monocytes and macrophages, referred to herein as “MonMac atlas”). The small UMAP to the right shows the density enrichment of whether the cells came from a tumor sample or an adjacent healthy sample. Figure 1C is a dot plot showing normalized gene expression profiles in human monocytes, macrophages, and TAM clusters. Dot size represents the percentage of cells expressing the gene, and color represents the scaled mean gene expression of the cluster. Figure ID is a heatmap showing gene-gene pairwise local correlation of the MonMAC atlas on genes with significant autocorrelation (1000 genes, FDR < 0.05), grouped into 9 different gene modules using hotspot. Figure IE is a circular plot showing the enrichment of modules defined in Figure ID in TAM and healthy tissue macrophages from lung (alveolar macrophages), breast, and colon. Figure IF is a scatterplot showing the conservation between the human TAM enrichment over human MonMac and mouse TAM enrichment over mouse MonMac in the conserved genes of the top 1000 autocorrelated genes from the hotspot module analysis. Each dot is a gene, and the highlighted dot in red indicates that the gene is within our perturbation screen candidate. Figure 1G is a graph showing enrichment score of each module defined in Figure ID in the different BMDM stimulations. Figure 1H is a graph showing enrichment score of combined selected TAM modules (modules 4, 6, and 8) in the different BMDM stimulations.

[0061] FIGs. 2A-G demonstrate assembly of a functional gene perturbation network in TAMs using deep generative modeling. Figure 2A is a heatmap displaying the MrVI local sample distances aggregated for each gene knockout. The colors on the left indicate hierarchical clusters of perturbation with similar effects on the transcriptome. Figure 2B shows functional gene perturbation network in TAMs, each knockout represented a node in the network and was connected to its five nearest neighbors in the distance matrix and visualized in a 2D graph using MDS. Figure 2C is a dot plot showing normalized gene expression of the different functional modules defined by MrVI. Figure 2D shows enrichment of the functional modules as shown in Figure 2C aggregated for all cells within one gene knockout in the perturbation network (Figure 2B). The 5 top enriched or depleted perturbations are indicated in the anti and pro-tumor modules, respectively. Figure 2E is a chord diagram showing the effect of selected perturbation in the defined modules. Red and blue lines indicate upregulation and downregulation of the modules, respectively. Figure 2F is a graph showing anti-tumor enrichment score of the perturbations with the highest or lowest score (top 14 are indicated in each category). Figure 2G shows enrichment of CXCL9 gene polarity of human TAMs over perturbation network. 5 top enriched perturbations are indicated.

[0062] FIGs. 3A-F demonstrate that Zeb2 loss reprograms the epigenetic state of TAM. Figure 3A is a heatmap showing 35,906 ATAC enhancer peaks clustered with k-means (n = 15). The values are row-normalized z-scores of the normalized read counts within a region. Each column represents the pooled average reads of the duplicates from each KO condition. Figure 3B is a graph showing differential enhancer (gained or lost) counts across all knockouts normalized to control (P-adjusted < 0.05 and log2FoldChange > 0.75). Figure 3C is a graph showing percentage of Zeb2 KO differential enhancers (open or closed) belonging to the genes defining the different modules with Zeb2 enrichment of log2FoldChange > 0.5 and baseMean > 50. Figure 3D shows IGV tracks showing normalized ATAC-seq signals at selected loci for Zeb2 KO vs NT control. Dashed boxes highlight ATAC-seq peaks with differential signals (light gray for enhancers and dark gray for promoters), and TFs are indicated by the different colored boxes if their motifs are enriched in the differential peak. Peaks are highlighted with the same threshold in (Figure 3C). Figure 3E is a motif activity heatmap derived from GimmeMotifs showing enrichment or depletion of selected motifs in genes from pro and anti-tumor modules in technical duplicates of the regions enriched in the open or closed regions of the Zeb2 KO compared to the control. Figure 3F is a bar-plot showing gene expression fold change of Zeb2 KO normalized to the NT. The bars indicate the mean fold change, and the error bars indicate the standard error of the mean (SEM).

[0063] FIGs. 4A-K demonstrate that Zeb2 is a TAM checkpoint inhibiting T cell activation. Figure 4A shows Kaplan-Meier curves for classification of ZEB2hlgh / lowpatients based on the 50thexpression percentile in the TCGA PK dataset across pan-cancer (left), CD68 high LUSC (middle), and CD68 high BLCA (right) cancer patients stratified based on their ZEB2 expression. P-values were obtained using a log-rank test. Figure 4B is a bar-plot showing IFN-y protein levels from cell-free supernatants of naive or anti-CD3 / anti-CD28 antibody-activated T cell mono- or co-cultures with control or Zeb2 KO BMDMs. Bars show mean and error bars show SD (n = 4 biological replicates). Statistical significance was determined using a two-sided, unpaired t-test. Figure 4C is a bar-plot showing the frequency of proliferating CD8+ or CD4+ T cells from naive or anti-CD3 / anti-CD28 antibody-activated T cell mono- or co-cultures with control or Zeb2 KO BMDMs. Bars show mean, and error bars show SD (n = 12; 3 biological replicates x 4 technical replicates). Statistical significance was determined using a two-sided Wilcoxon test. Figure 4D is a schematic illustration of a CpG-siRNA conjugate molecule with the sequence of mouse Zeb2 siRNA (left panel) and experimental workflow (right panel). Figure 4E is a graph of tumor growth measurements in MC38 (n = 10) and MB48 (n = 5) tumor-bearing mice since treatment onset. Results are presented as means ± SEM. Statistical significance was determined using a two-way ANOVA or mixed-effects model with Tukey’s multiple comparisons test. Figure 4F is a boxplot displaying the anti-tumor enrichment score of TAMs acquired in early endpoint analysis across the three different treatments. Median with 25thto 75thpercentiles and whiskers within 1.5 * interquartile range are shown. Statistical significance was determined using the Kruskal-Wallis test with Dunnett’s multiple comparisons test. Figure 4G is a Volcano plot showing differentially expressed TAM genes between CpG-siRNAZeb2(dark magenta) and CpG-siRNACtrl(dark cyan) treatment from the early endpoint analysis. Dashed lines depict thresholds for genes within p < 0.05 & log2FC > 0.25. Figure 4H shows intratumoral CD8:Treg ratio at late endpoint analysis depicted as crossbars with mean and confidence intervals (n = 5-6 mice per group). Statistical significance was determined using one-way analysis of variance (ANOVA) with Dunnett’s multiple comparisons test. Figure 41 is a Volcano plot showing differentially expressed CD8 T cell genes between CpG-siRNAZeb2(dark magenta) and CpG-siRNACtrl(dark cyan) treatment from the late endpoint analysis. Dashed lines depict thresholds for genes within p < 0.05 & log2FC > 0.25. Figure 4J is a heatmap showing scaled median fluorescence intensity (FI) of index- sort-derived functional protein markers (left panel) and scaled mean expression of selected effector / memory genes (right panel) across tdLN CD8 T cell subsets and treatment groups. Statistical significance was determined using the Kruskal- Wallis test with Dunnett’s multiple comparisons test. Figure 4K is a bar-plot depicting orthotopic bladder tumor weight of untreated or CpG-siRNAZeb2-treated mice two days following last treatment. Results are presented as means ± SEM (n=8 mice). Statistical significance was determined using a two-sided, unpaired t-test.

[0064] FIGs. 5A-G demonstrate identification of conservation of TAM programs genes by analyzing scRNA-seq data from human and murine solid tumors. Figure 5A is a schematic workflow showing data collection and integration analysis of scRNA-seq datasets of human solid tumors from breast, colon, and lung using scVI deep generative model. Figure 5B is a dot plot showing normalized gene expression profile across human tumor atlas subsets. Figure 5C is a re- integrated MonMac atlas showing the integration of different tumors with the LISI integration score indicated on the right panel, comparing sc VI to the non-integrated PCA approach. Figure 5D is a MonMac atlas of murine tumors (B 16, MC38, and MCA205) with peritoneal macrophages as a reference for tissue-resident myeloid cells. Figure 5E is a dot plot showing normalized gene expression profiles across mouse tumor MonMac atlas subsets. Figure 5F is a UMAP of the sample aware MrVI latent space z of scRNA-seq of BMDMs stimulated with the different indicated cytokines and the enrichment of genes from human TAM modules. The heatmap on the right indicates a similarity between cytokines' effects on the gene expression of BMDMs. Figure 5G shows normalized gene expression profiles of screen candidate genes across the different BMDM stimulated with different cytokines.

[0065] FIGs. 6A-F demonstrate single-cell arrayed CRISPR screen of TAM-like BMDMs. Figure 6A is a scheme of CRISPR screen workflow from target selection to output. Figure 6B shows light microscope images of 4 different BMDM conditions. Cell survival is quantified using ImageJ and indicated in the right panel. Figure 6C is a heatmap showing the normalized z- score of selected genes (rows) across triplicates of NT control, Bachl KO, and Stat6 KO. Figure 6D shows FlowSom clusters of the 12 different clusters in each 12-well plate and their scaled MFI. Figure 6E shows quantification of cell distribution in each pool of 4 different perturbations. Figure 6F is a heatmap showing the normalized z-score of the corresponding genes to each perturbation in the screen. The gene expression of each perturbation is normalized against the control.

[0066] FIGs. 7A-D demonstrate assembly a functional gene perturbation network in TAMs using deep generative modeling. Figure 7A is a scheme of the application of MrVI deep generative model on the CRISPR screen. Figure 7B shows empirical cumulative distribution function (upper panel) and overlay on the UMPA of the MrVI z latent space (lower panel) for the first two principal components obtained from PCA of the local sample distances. Figure 7C shows UMAP of the MrVI z latent space displaying the indicated knockouts. Figure 7D shows differential gene expression, TF activity, and pathway analysis across the different functional perturbation clusters.

[0067] FIGs. 8A-E demonstrate assembly of a functional gene perturbation network in TAMs using deep generative modeling. Figure 8A is a heatmap of hotspot gene modules defined by MrVI CRISPR screen BMDMs. Figure 8B is a heatmap showing the -logio(padj) of a hypergeometric test on the gene overlaps between human and mouse modules. Figure 8C is a dot plot showing the enrichment of each knockout in each one of the defined modules in Figure 8A. Figure 8D shows projection of the different functional modules over the perturbation network. Figure 8E shows anti-tumor enrichment score across all candidate genes perturbation.

[0068] FIGs. 9A-D demonstrate assembly of a functional gene perturbation network in TAMs using deep generative modeling. Figure 9A is a heatmap showing the normalized z-score of selected genes (rows) for each perturbation in our CRISPR screen (columns). Each perturbation is normalized against the control by bootstrapping analysis of 100 resamples of 100 cells from each replicate of each perturbation. Figure 9B-C show chord diagrams showing the effect of selected perturbation in the defined modules. Red and blue lines indicate upregulation and downregulation of the modules, respectively. Figure 9D is a Volcano plot of differential gene expression between Zeb2 KO and NT control from the CRISPR screen using pseudo bulking and DESeq2. Important genes are indicated.

[0069] FIGs. 10A-F demonstrate that Zeb2 loss reprograms the epigenetic state of TAM. Figure 10A shows functional ATAC enhancer network where each node represents a TF perturbation and is connected to the 5 nearest neighbors in the distance matrix and visualized in 2D using MDS. Figure 10B shows differential enhancer (gained and lost) counts across all knockouts normalized to control. Figure 10C shows IGV tracks showing normalized ATAC-seq signals at selected loci for Zeb2 KO vs NT control. Dashed boxes highlight ATAC-seq peaks with differential signals (light gray for enhancers and dark gray for promoters), and TFs are indicated if their motifs are enriched in the differential peak. Peaks are highlighted with the same threshold in Figure 3C. Figure 10D is a heatmap showing 2127 ATAC promoter peaks clustered with k- means (n = 10). The values are row-normalized z-scores of the normalized read counts within a region. Each column represents the pooled average reads of the duplicates from each KO condition. Figure 10E shows IGV tracks showing normalized ATAC-seq signals at Zeb2 locus for Stat6 KO vs NT control. The dashed box highlights ATAC-seq enhancer peak with a differential signal, and TFs are indicated by the different colored boxes if their motifs are enriched in the differential peak. Peaks are highlighted with the same threshold in Figure 3C. Figure 10F is a bar-plot showing Zeb2 gene expression fold change of Stat6 KO normalized to the NT. The bars indicate the mean fold change, and the error bars indicate the standard error of the mean (SEM).

[0070] FIGs. 11A-N demonstrate that Zeb2 is a TAM checkpoint inhibiting T cell activation. Figure 11A is a dot plot showing mean Tlr9 expression across intratumoral immune cell subsets from scRNA-seq data of MC38 tumors. Figure 1 IB is a bar-plot showing median fluorescence intensity of Zeb2 protein expression in myeloid cells from untreated or CpG-siRNA-treated MB49 tumors. Data show mean with dots (n = 5) representing biological replicates. Statistical significance was determined using the Kruskal-Wallis test with Dunnett’s multiple comparisons test. Figure 11C is a bar-plot displaying the frequency of Zeb2+ myeloid cells from untreated or CpG-siRNA-treated MB49 tumors. Data show mean with dots (n = 5) representing biological replicates. Statistical significance was determined using one-way analysis of variance (ANOVA) with Dunnett’s multiple comparisons test. Figure 1 ID is a 2D plot of late endpoint analyzed, tdLN-enriched T cells showing paired single-cell UMAP embedding of their protein surface marker expression with FlowSOM cluster annotation (left) and scRNA-seq profiles and clusters (right). Figure HE is a heatmap showing scaled mean expression of differentially expressed marker genes for tdLN scRNA-seq clusters. Figure 11F is a heatmap depicting index-sort- derived scaled median fluorescence intensity (FI) expression of protein surface marker over the FlowSOM clusters of tdLN-enriched T cells. Figure 11G is a heatmap showing scaled median fluorescence intensity (FI) of index- sort-derived functional protein markers (left panel) and scaled mean expression of selected effector / memory genes (right panel) across tdLN CD4 T cell subsets and treatment groups. Statistical significance was determined using the Kruskal-Wallis test with Dunnett’s multiple comparisons test. Figure 11H is a 2D plot of intratumoral T cells from untreated or CpG-siRNA-treated MC38 tumors showing scRNA-seq UMAP and cluster annotation (left) with cluster- specific scaled mean expression of differential marker genes (right). Figure 111 is a heatmap showing scaled mean expression of NK effector genes across treatment groups. Figure 11 J is a heatmap displaying a scaled mean expression of differentially expressed marker genes for tumor monocyte and macrophage- specific clusters aggregated from untreated and CpG-siRNA-treated MC38 tumors. Figure 1 IK is a density plot showing the average cTET for each treatment condition by bootstrap resampling of 100 cells from each mouse for 500 resamples. Figure 11L is a 2D plot showing diffusion map with Zman-seq trajectory of monocyte to TAM single-cell combined from CpG-siRNA-treated MC38 tumors (left). Nodes represent clusters with mean cTET for the Zman trajectory. The right panel depicts the enrichment density for each treatment condition. Figure 1 IM is a segment plot displaying Gene Ontology (GO) enrichment analysis of biological processes (BP) using differentially expressed genes specific to TAMs (top) or CD8 T cells (bottom) between the CpG-siRNAZeb2and CpG- siRNACtrltreatment conditions. Figure UN is a dot plot showing normalized Zeb2 gene expression profile in monocytes and macrophages in different tumors (B16, MB49, MC38, and MCA205). Dot size represents the percentage of cells expressing the gene, and color represents the scaled mean gene expression per tumor. FIGs. 12A-B demonstrate that Zeb2 is a TAM checkpoint inhibiting T cell activation. The figures show Kaplan-Meier curves for classification of ZEB2hlgh / lowpatients based on the 50thexpression percentile in the TCGA PK dataset across CD68 low LUSC (Figure 12A) and CD68 low BLCA (Figure 12B) cancer patients stratified based on their Zeb2 expression. P-values were obtained using a log-rank test.

[0071] FIGs. 13A-B are schematic representations of the CpG-siRNAZeb2passenger strand (Figure 13A) and the siRNAZeo2guide strand (Figure 13B).

[0072] FIG. 14 is a scatter plot showing the correlation between the anti-tumor score (y axis) and the CXCL9:SPP1 polarity score (x axis) on the knockout screen. Each dot represents a KO condition, and the average score of the cells from the respective KO is plotted. The Pearson’s correlation coefficient and its p-value are shown on the plot.

[0073] FIGs. 15A-H demonstrate that ZEB2 loss reprograms TAM functions. Figure 15A is a Volcano plot of pseudo-bulk differential gene expression analysis between Zeb2 KO and non- treated (NT) control BMDMs from the CRISPR screen. Relevant genes are highlighted and labeled as indicated. Differential gene expression has been tested using DESeq2. Dashed line depicts threshold for genes within p(adj) < 0.05. Figure 15B is a bar plot showing IFNy protein levels from cell-free supernatants (left) and frequency of proliferating CD8+ or CD4+ T cells of naive or anti-CD3 / anti-CD28 antibody-activated T cell mono- or co-cultures with control or Zeb2 KO BMDMs. Bars represent mean and error bars indicate SD [n = 4 biological replicates (left) and n = 12 biological replicates (right)]. Statistical significance was determined using a two- sided, unpaired t-test (left) and a two-sided Wilcoxon test (right). Figures 15C-D show bar plots depicting flow cytometry analysis-derived frequency of CD8+ OT-I T cell (Figure 15C) or CD4+ OT-II T cell subsets (Figure 15D) after co-culturing with ovalbumin (OVA) peptide-pulsed control or Zeb2 KO BMDMs. Data show means with dots representing biological replicates (n = 2-3). Statistical significance was determined using a two-way ANOVA with Sidak’s multiple comparisons test. Figure 15E Left panel: Flow cytometry histogram of pHrodo bioparticle beads fluorescence intensity of Zeb2 KO and control BMDMs and BMDMs without pHrodo bioparticle beads phagocytosis. Figure 15E Right panel: Bar plot with S.E.M as errors depicting median fluorescence intensity of pHrodo bioparticle beads as a measurement for phagocytosis activity. Statistical significance was determined using a two-sided, Wilcoxon test. Figure 15F shows UMAP of the scVI latent space of scRNA-seq of in vivo splenic macrophages from Fcgrl- CRExZeb2fl / fl (CRE+) and Zeb2fl / fl (CRE) mice (Left panel). The color indicates whether cells are from Zeb2 KO (CRE+) or Zeb2 WT (CRE_) mice. On the right panel, the enrichment score of differentially upregulated Zeb2 KO genes from the CRISPR screen (as shown in A) onto the splenic macrophages UMAP. Figure 15G is a Volcano plot showing differentially expressed genes in scRNA-seq of splenic macrophages from Fcgrl-CRExZeb2fl / fl (CRE+; green) and Zeb2fl / fl (CRE; yellow) mice. Dashed line depicts threshold for genes within p(adj) < 0.05 for a Mann-Whitney U test with a false-discovery correction according to the Benjamini-Hochberg approach. Figure 15H is a Chord diagram displaying the difference in Jaccard similarity for each module. The Jaccard similarity between up- and downregulated genes in splenic macrophages from Fcgrl-CRExZeb2fl / fl (CRE+) verses Zeb2fl / fl (CRE_) was computed separately and the difference in Jaccard similarity between up- and downregulated genes and the respective modules is visualized (red links indicate positive association between Zeb2 KO signature and the respective gene module while blue links depict negative association).

[0074] FIG. 16 shows on the left panel a Volcano plot demonstrating differentially expressed genes in scRNA-seq of human TAM model of IL-4-cultured ZEB2 KO macrophages (green) and control macrophages (yellow). Dashed lines depict threshold for genes within p(adj) < 0.05 for a Mann-Whitney U test with a false-discovery correction according to the Benjamini-Hochberg approach. Pathway overrepresentation analysis for differentially expressed genes of human and mouse IL-4-cultured ZEB2 KO and control macrophages is shown on the right panel. Overrepresentation was assessed for each species separately by performing a one-tailed Fisher’s exact test and concomitant Benjamini-Hochberg correction of differentially expressed genes.

[0075] FIG. 17 shows tumor growth measurements of MC38 (n = 10) and MB48 (n = 5) tumors in tumor-bearing mice, each harboring two tumors inoculated at both flanks, while one tumor at one flank was intra-tumoral treated with the different treatments and the other tumor was left non-treated. Shown is the non-treated tumor. Results are presented as means ± SEM. Statistical significance was determined using a two-way ANOVA or a mixed-effects model with Tukey’s multiple comparisons test.

[0076] FIG. 18 is a bar plot depicting orthotopic bladder tumor weight of untreated, CpG- siRNACtrlor CpG-siRNAZeb2-treated mice two days after last treatment. Results are presented as means ± SEM (n = 12-15 mice). Statistical significance was determined using Kruskal-Wallis test with Dunnett’s multiple comparisons test between the untreated and the two treatment groups and a two-sided, Wilcoxon test between CpG-siRNACtrlor CpG-siRNAZeb2groups. Combined data from two independent experiments are shown.

[0077] FIGs. 19A-F demonstrates that ZEB2 loss reprograms macrophage functions. Figure 19A shows bar plots demonstrating median fluorescence intensity (MFI) of indicated protein surface markers from untreated, IL-4- or IFNy-treated NT control or Zeb2 KO BMDMs. Data represent means ± SD (n=3 biological replicates). Statistical significance was determined using a twoway- ANOVA with Sidak’s multiple comparisons. Figure 19B shows histograms demonstrating CellTrace™ Far Red cell proliferation dye fluorescence intensities for CD4+ and CD8+ T cells proliferation analysis of WT splenic T cells stimulated with anti-CD3 and anti-CD28 and co- cultured for 72 hours with either Zeb2 KO or NT control BMDMs or left alone or unstimulated. Figure 19C is a Heatmap depicting scaled MFI expression of protein surface markers over the FlowSOM defined clusters of CD4+ OT-II and CD8+ OT-I T cell subsets derived from the antigenpresentation co-culture assay with OVA peptides-pulsed NT control or Zeb2 KO BMDMs. Figure 19D is a UMAP visualization of flow cytometry-based surface marker expression with FlowSOM cluster-defined CD4+ OT-II and CD8+ OT-I T cell subsets derived from the antigen-presentation coculture assay with OVA peptides-pulsed NT control or Zeb2 KO BMDMs. Figure 19E shows flow cytometry dot plota depicting the FlowSOM-defined naive / effector / memory T cell subsets co-cultured with OVA peptide-pulsed NT control or Zeb2 KO BMDMs analyzed by flow cytometry. Figure 19F shows flow cytometry plots depicting the FlowSOM-defined T cell subsets (labels as in Figures 19D-E), which are positive for CD107a analyzed by flow cytometry. Dot size ranges indicate the scaled MFI of CD 107a per single-cell.

[0078] FIGs. 20A-B demonstrate that ZEB2 is a checkpoint for TAM reprogramming. Figure 20A shows mean Tlr9 expression across intratumoral immune cell subsets from scRNA-seq data of MC38 tumors (upper panel), histograms depicting flow cytometry-derived TLR9 expression (lower left panel) and bar plots presenting mean frequency of indicated MC38-derived intratumoral immune cell subsets out of CD45+ live cells (lower right panel). Dots represent individual mice (n=6) with errors indicating SD. Figure 20B upper panel: a bar plot showing mean Zeb2 gene expression measurement in IL-4 stimulated BMDMs treated with either CpG- siRNAZeb2, CpG-siRNACtrlor untreated (n=3; upper panel). Errors indicate SEM. Statistical significance was determined using a one-way ANOVA with Tukey’s multiple comparisons test. Figure 20B lower panel: box plot displaying the flow cytometry analysis-derived frequency of ZEB2+ MC38-derived intratumoral immune cell subsets (live CD45+ LY6G- ZEB2+) from mice that were either treated with CpG-siRNACtrlor CpG-siRNAZeb2intratumorally. Dots represent individual mice (n=6-7). Median with 25th to 75th percentiles and whiskers within 1.5 * interquartile range are shown. Statistical significance was determined using a two-way ANOVA test with Sidak’s multiple comparisons test.

[0079] FIGs. 21A-F demonstrate that ZEB2 is a checkpoint for TAM reprogramming in-vivo. Figure 21 A shows tumor growth measurements of LLC tumors in tumor-bearing mice (n = 9-10) injected intratumorally with 5 mg / kg CpG-siRNACtrl, CpG-siRNAZeb2or left untreated. Results are presented as means ± SEM. Statistical significance was determined using a mixedeffects model with Dunnett’s multiple comparisons test. Figure 2 IB is a dot plot showing normalized Zeb2 gene expression profile in monocytes and TAMs in different tumors (subcutaneous B16, MC38, and MCA205 and orthotopic MB49). Dot size represents the percentage of cells expressing the gene, and color represents the scaled mean gene expression per tumor. Figure 21C is a schematic illustration depicting the mannosylated lipid nanoparticle (mLNP) with mannose linked to DSPE-PEG encapsulating sgRNA. Figure 2 ID is a Volcano plot showing differentially expressed genes of index-sorted scRNA-seq of IL-4-treated murine Cas9 BMDMs between Zeb2 guide mLNPs (yellow) and the NT control guide mLNPs (green) using pseudo bulking and DESeq2. Dashed lines depict thresholds for genes within adjusted p-value < 0.05. Figure 2 IE is a Heatmap showing the - loglO(padj) of a hypergeometric test on the gene overlaps between differential genes of Zeb2 guide LNP versus LNP control guide with the geneset from the screen modules. Figure 21F shows tumor growth measurements of MC38 tumors (n = 8) in tumor- bearing mice intravenously injected (day 2, 4, 6) with either 1 mg / kg Zeb2-guide or non- targeting (NT) control guide carrying mLNPs. Data are presented as means ± SEM. Statistical significance was determined using a two-way ANOVA with Sidak’s multiple comparisons test.

[0080] DESCRIPTION OF SPECIFIC EMBODIMENTS OF THE INVENTION

[0081] The present invention, in some embodiments thereof, relates to relates to compositions for downregulating Zeb2 in macrophages and uses thereof.

[0082] Before explaining at least one embodiment of the invention in detail, it is to be understood that the invention is not necessarily limited in its application to the details set forth in the following description or exemplified by the Examples. The invention is capable of other embodiments or of being practiced or carried out in various ways.

[0083] Tumor-associated macrophages (TAMs) are major players in regulating the immunosuppressive tumor environment, and their abundance is highly correlated with poor clinical outcomes.

[0084] The present inventors set out to characterize the regulatory circuits controlling TAM functions using data-driven analysis of TAM programs across human solid tumors in conjunction with functional single-cell genomics. To this end, they integrated scRNA-seq data across human solid tumors and their corresponding healthy tissue and defined a set of candidate TAM regulators that are enriched across TAM-specific modules; and established an in vitro single-cell CRISPR screening approach on primary TAM-like cells to systematically assess the function of each candidate regulator (Examples 1-2 of the Examples section which follows). As shown in the Examples section which follows, using a deep generative model capable of learning a local representation for each candidate regulator, they constructed a gene perturbation network that linked individual target genes with prototypical functional modules in TAMs; and identified non- redundant pathways regulating distinct TAM functions, showing modular circuitry (Example 3 of the Examples section which follows). Importantly, the inventors identified Zeb2 as the master regulator of pro-tumor functions in TAMs, orchestrating the suppression of type I interferon response and antigen presentation alongside the activation of immune suppression programs. Following, the present inventors have demonstrated that genetic ablation of Zeb2 reprogrammed TAMs identity on chromatin, RNA, and protein levels (Examples 4-5 of the Examples section which follows). Interrogation of the Tumor Cancer Genome Atlas (TCGA) revealed a negative correlation between ZEB2 expression and patient survival in certain macrophage-rich solid tumors, including lung and bladder cancer. Functional macrophage coculturing assays defined Zeb2 as a critical regulator of TAM immunosuppression activity by inhibiting T cell proliferation and activation. Furthermore, selective in-vivo targeting of Zeb2 in macrophages using a CpG- siRNA2'*2DNA hybrid [a chimeric DNA molecule composed of the Toll-like receptor (TLR) 9 agonist unmethylated cytidine-phosphate guanosine (CpG) arm and a small interfering RNA (siRNA) sequence that specifically targets and silences Zeb2 in TAMs] reprogrammed TAMs and mobilized systematic anti-tumoral T cell responses, achieving complete tumor clearance as a monotherapy (Example 6 of the Examples section which follows). Moreover, an alternative Zeb2-targeting strategy using mannosylated lipid nanoparticles (mLNP-Zeb2) also showed efficacy, both in-vitro and in-vivo, further supporting the potential of ZEB 2 as a therapeutic target (Example 6 of the Examples section which follows).

[0085] Overall, the results indicate ZEB2 as a master switch of TAMs and suggest specific targeting of ZEB2 in macrophages as a potential therapeutic approach for the treatment of e.g., an inflammatory disease (e.g., cancer).

[0086] Thus, according to an aspect of the present invention, there is provided a composition comprising an agent specifically downregulating expression and / or activity of Zeb2 attached to a macrophage-targeting moiety or encapsulated in a particle comprising a macrophage-targeting moiety.

[0087] As used herein the term “Zeb2”, also known as zinc finger E-box-binding homeobox 2, refers to the ZEB2 gene (corresponding to Gene ID: 9839) and to its polynucleotide or polypeptide expression product. According to specific embodiments, the Zeb2 refers to the human Ze2, such as provided in the following Accession Numbers: NM_001171653, NM_014795, NP_001165124 and NP_055610 (SEQ ID NOs: 1-4). As used herein the phrase “downregulating expression and / or activity of Zeb2”, also interchangeably referred to herein as “inhibiting expression and / or activity of Zeb2”, refers to a decrease in expression and / or activity of the protein product of Zeb2 in the presence of the agent in comparison to same in the absence of the agent, as may be determined by e.g. PCR, Northern blot, RNA in situ hybridization stain Western-blot, ELISA, flow cytometry, immuno staining, activity assay such as reporter assay, immunoprecipitation assay, migration or invasion assay, and the like. Alternatively or additionally, as Zeb2 is a transcription factor, expression and / or activity may be determined indirectly by expression or activity of a gene Zeb2 is associated with its transcription. Such genes are known in the art and include e.g., E-Cadherin, N-Cadherin, Vimentin, MMP-2, SMAIL1, TWIST 1, or F0XC2. Alternatively or additionally, as the agent of some embodiments of the present invention has an effect on macrophages (i.e., upregulating macrophage anti-tumor functions and downregulating the pro-tumor functions), the decrease can also be determined by cytological assays e.g. assessment of morphological, phenotypic and transcriptional changes in macrophages, as further described in details in the Examples section which follows. According to specific embodiments, the decrease is a statistically significant decrease. According to specific embodiments, the change is in at least 5 %, 10 %, 20 %, 30 %, 40 % or even higher say, 50 %, 60 %, 70 %, 80 %, 90 % or more than 99 %. According to specific embodiments, the change is at least 1.5fold, at least 2 fold, at least 3 fold, at least 5 fold, at least 10 fold, or at least 20 fold as compared to same in the absence of the agent.

[0088] According to specific embodiments, the agent downregulates expression of Zeb2.

[0089] According to specific embodiments, the agent downregulates activity of Zeb2.

[0090] According to specific embodiments, the agent binds Zeb2.

[0091] Thus, according to specific embodiments, downregulating expression and / or activity of Zeb2 is specific for Zeb2. As used herein “specific for Zeb2” refers to the ability to specifically decrease the expression and / or biological function of Zeb2 and not of other genes or their expression products.

[0092] The selective inhibition can be manifested as higher affinity (e.g., Ka) of the agent to Zeb2 than to other genes or expression products thereof. Higher affinity can be of at least 5, 10, 100, 1000 or 10000 or 100,000 or 1,000,000 fold or more using the same affinity assay.

[0093] According to specific embodiments, the agent binds Zeb2 with no cross reactivity with other proteins or polynucleotides.

[0094] According to specific embodiments, the agent does not downregulate expression and / or activity of Zebl. As used herein the term “Zebl”, also known as zinc finger E-box-binding homeobox 1, refers to the ZEB1 gene (corresponding to Gene ID: 6935) and to its polynucleotide or polypeptide expression product. According to specific embodiments, the Zebl refers to the human Zel, such as provided in the following Accession Numbers: NM_001128128, NM_001174093, NM_001174094, NM_001174095, NM_001174096, NP_001121600, NP-001167564, NP_001167565, NP_001167566, NP_001167567 (SEQ ID NOs: 5-14).

[0095] According to other specific embodiments, the agent downregulates expression and / or activity of Zeb2 by affecting an upstream (also referred to herein as an activator) or downstream molecule (also referred to herein as an effector) which is specific to the Zeb2 pathway.

[0096] Downregulating expression and / or activity can be can be effected at the protein level (e.g., antibodies, small molecules, inhibitory peptides, enzymes that cleave the polypeptide, aptamers and the like) but may also be effected at the genomic (e.g. homologous recombination and site specific endonucleases) and / or the transcript level using a variety of molecules which interfere with transcription and / or translation (e.g., RNA silencing agents) of a target expression product described herein.

[0097] Downregulating expression and / or activity may be either transient or permanent.

[0098] The agent may be a reversible or an irreversible inhibitor.

[0099] The agent may be a competitive or non-competitive inhibitor.

[0100] Non-limiting examples of inhibitory agents are described in details hereinbelow.

[0101] Downregulating at the polypeptide level

[0102] According to specific embodiments, the downregulating agent is a polypeptide.

[0103] The term "polypeptide" or “protein” as used herein encompasses native peptides (either degradation products, synthetically synthesized peptides or recombinant peptides) and peptidomimetics (typically, synthetically synthesized peptides), as well as peptoids and semipeptoids which are peptide analogs, which may have, for example, modifications rendering the peptides more stable while in a body or more capable of penetrating into cells. Such modifications include, but are not limited to N terminus modification, C terminus modification, peptide bond modification, backbone modifications, and residue modification. Methods for preparing peptidomimetic compounds are well known in the art and are specified, for example, in Quantitative Drug Design, C.A. Ramsden Gd., Chapter 17.2, F. Choplin Pergamon Press (1992), which is incorporated by reference as if fully set forth herein.

[0104] Peptide bonds (-CO-NH-) within the peptide may be substituted, for example, by N- methylated amide bonds (-N(CH3)-CO-), ester bonds (-C(=O)-O-), ketomethylene bonds (-CO- CH2-), sulfinylmethylene bonds (-S(=O)-CH2-), a-aza bonds (-NH-N(R)-CO-), wherein R is any alkyl (e.g., methyl), amine bonds (-CH2-NH-), sulfide bonds (-CH2-S-), ethylene bonds (-CH2- CH2-), hydroxyethylene bonds (-CH(0H)-CH2-), thioamide bonds (-CS-NH-), olefinic double bonds (-CH=CH-), fluorinated olefinic double bonds (-CF=CH-), retro amide bonds (-NH-CO-), peptide derivatives (-N(R)-CH2-C0-), wherein R is the "normal" side chain, naturally present on the carbon atom.

[0105] These modifications can occur at any of the bonds along the polypeptide chain and even at several (2-3) bonds at the same time.

[0106] Natural aromatic amino acids, Trp, Tyr and Phe, may be substituted by non-natural aromatic amino acids such as l,2,3,4-tetrahydroisoquinoline-3-carboxylic acid (Tic), naphthylalanine, ring-methylated derivatives of Phe, halogenated derivatives of Phe or O-methyl- Tyr.

[0107] The polypeptides of some embodiments of the invention may also include one or more modified amino acids or one or more non-amino acid monomers (e.g., fatty acids, complex carbohydrates etc).

[0108] The term "amino acid" or "amino acids" is understood to include the 20 naturally occurring amino acids; those amino acids often modified post-translationally in vivo, including, for example, hydroxyproline, phosphoserine and phospho threonine; and other unusual amino acids including, but not limited to, 2-aminoadipic acid, hydroxylysine, isodesmosine, nor-valine, nor-leucine and ornithine. Furthermore, the term "amino acid" includes both D- and L- amino acids.

[0109] The polypeptides of some embodiments of the invention may be synthesized by any techniques known to those skilled in the art of peptide synthesis, for example but not limited to recombinant DNA techniques or solid phase peptide synthesis.

[0110] According to specific embodiments, the downregulating agent is an antibody.

[0111] According to specific embodiments, the antibody specifically binds at least one epitope of a target protein described herein.

[0112] As used herein, the term "epitope" refers to any antigenic determinant on an antigen to which the paratope of an antibody binds. Epitopic determinants usually consist of chemically active surface groupings of molecules such as amino acids or carbohydrate side chains and usually have specific three-dimensional structural characteristics, as well as specific charge characteristics.

[0113] The term "antibody" as used in this invention includes intact molecules as well as functional fragments thereof, such as Fab, F(ab')2, Fv, scFv, dsFv, or single domain molecules such as VH and VL that are capable of binding to an epitope of an antigen. The antibody may be mono-specific (capable of recognizing one epitope or protein), bi-specific (capable of binding two epitopes or proteins) or multi- specific (capable of recognizing multiple epitopes or proteins).

[0114] Suitable antibody fragments for practicing some embodiments of the invention include a complementarity-determining region (CDR) of an immunoglobulin light chain (referred to herein as “light chain”), a complementarity-determining region of an immunoglobulin heavy chain (referred to herein as “heavy chain”), a variable region of a light chain, a variable region of a heavy chain, a light chain, a heavy chain, an Fd fragment, and antibody fragments comprising essentially whole variable regions of both light and heavy chains such as an Fv, a single chain Fv (scFv), a disulfide- stabilized Fv (dsFv), an Fab, an Fab’, and an F(ab’)2.

[0115] As used herein, the terms "complementarity-determining region" or "CDR" are used interchangeably to refer to the antigen binding regions found within the variable region of the heavy and light chain polypeptides. Generally, antibodies comprise three CDRs in each of the VH (CDR HI or HI; CDR H2 or H2; and CDR H3 or H3) and three in each of the VL (CDR LI or LI; CDR L2 or L2; and CDR L3 or L3).

[0116] The identity of the amino acid residues in a particular antibody that make up a variable region or a CDR can be determined using methods well known in the art and include methods such as sequence variability as defined by Kabat et al. (See, e.g., Kabat et al., 1992, Sequences of Proteins of Immunological Interest, 5th ed., Public Health Service, NIH, Washington D.C.), location of the structural loop regions as defined by Chothia et al. (see, e.g., Chothia et al., Nature 342:877-883, 1989.), a compromise between Kabat and Chothia using Oxford Molecular's AbM antibody modeling software (now Accelrys®, see, Martin et al., 1989, Proc. Natl Acad Sci USA. 86:9268; and world wide web site www(dot)bioinf-org(dot)uk / abs), available complex crystal structures as defined by the contact definition (see MacCallum et al., J. Mol. Biol. 262:732-745, 1996), the "conformational definition" (see, e.g., Makabe et al., Journal of Biological Chemistry, 283: 1156-1166, 2008) and IMGT [Lefranc MP, et al. (2003) IMGT unique numbering for immunoglobulin and T cell receptor variable domains and Ig superfamily V-like domains. Dev Comp Immunol 27: 55-77].

[0117] As used herein, the “variable regions” and "CDRs" may refer to variable regions and CDRs defined by any approach known in the art, including combinations of approaches.

[0118] Functional antibody fragments comprising whole or essentially whole variable regions of both light and heavy chains are defined as follows:

[0119] (i) Fv, defined as a genetically engineered fragment consisting of the variable region of the light chain (VL) and the variable region of the heavy chain (VH) expressed as two chains; (ii) single chain Fv (“scFv”), a genetically engineered single chain molecule including the variable region of the light chain and the variable region of the heavy chain, linked by a suitable polypeptide linker as a genetically fused single chain molecule.

[0120] (iii) disulfide- stabilized Fv (“dsFv”), a genetically engineered antibody including the variable region of the light chain and the variable region of the heavy chain, linked by a genetically engineered disulfide bond.

[0121] (iv) Fab, a fragment of an antibody molecule containing a monovalent antigen-binding portion of an antibody molecule which can be obtained by treating whole antibody with the enzyme papain to yield the intact light chain and the Fd fragment of the heavy chain which consists of the variable and CHI domains thereof;

[0122] (v) Fab’, a fragment of an antibody molecule containing a monovalent antigen-binding portion of an antibody molecule which can be obtained by treating whole antibody with the enzyme pepsin, followed by reduction (two Fab’ fragments are obtained per antibody molecule);

[0123] (vi) F(ab’)2, a fragment of an antibody molecule containing a monovalent antigen-binding portion of an antibody molecule which can be obtained by treating whole antibody with the enzyme pepsin (i.e., a dimer of Fab’ fragments held together by two disulfide bonds); and

[0124] (vii) Single domain antibodies or nanobodies are composed of a single VH or VL domains which exhibit sufficient affinity to the antigen.

[0125] The antibody may be monoclonal or polyclonal.

[0126] Methods of producing polyclonal and monoclonal antibodies as well as fragments thereof are well known in the art (See for example, Harlow and Lane, Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, New York, 1988, incorporated herein by reference).

[0127] Antibody fragments according to some embodiments of the invention can be prepared by proteolytic hydrolysis of the antibody or by expression in E. coli or mammalian cells (e.g., Chinese hamster ovary cell culture or other protein expression systems) of DNA encoding the fragment. Antibody fragments can be obtained by pepsin or papain digestion of whole antibodies by conventional methods. For example, antibody fragments can be produced by enzymatic cleavage of antibodies with pepsin to provide a 5S fragment denoted F(ab')2. This fragment can be further cleaved using a thiol reducing agent, and optionally a blocking group for the sulfhydryl groups resulting from cleavage of disulfide linkages, to produce 3.5S Fab' monovalent fragments. Alternatively, an enzymatic cleavage using pepsin produces two monovalent Fab' fragments and an Fc fragment directly. These methods are described, for example, by Goldenberg, U.S. Pat. Nos. 4,036,945 and 4,331,647, and references contained therein, which patents are hereby incorporated by reference in their entirety. See also Porter, R. R. [Biochem. J. 73: 119-126 (1959)]. Other methods of cleaving antibodies, such as separation of heavy chains to form monovalent light-heavy chain fragments, further cleavage of fragments, or other enzymatic, chemical, or genetic techniques may also be used, so long as the fragments bind to the antigen that is recognized by the intact antibody.

[0128] Fv fragments comprise an association of VH and VL chains. This association may be noncovalent, as described in Inbar et al. [Proc. Nat'l Acad. Sci. USA 69:2659-62 (19720]. Alternatively, the variable chains can be linked by an intermolecular disulfide bond or cross- linked by chemicals such as glutaraldehyde. Preferably, the Fv fragments comprise VH and VL chains connected by a peptide linker. These single-chain antigen binding proteins (sFv) are prepared by constructing a structural gene comprising DNA sequences encoding the VH and VL domains connected by an oligonucleotide. The structural gene is inserted into an expression vector, which is subsequently introduced into a host cell such as E. coli. The recombinant host cells synthesize a single polypeptide chain with a linker peptide bridging the two V domains. Methods for producing sFvs are described, for example, by [Whitlow and Filpula, Methods 2: 97- 105 (1991); Bird et al., Science 242:423-426 (1988); Pack et al., Bio / Technology 11: 1271-77 (1993); and U.S. Pat. No. 4,946,778, which is hereby incorporated by reference in its entirety.

[0129] Another form of an antibody fragment is a peptide coding for a single complementarity- determining region (CDR). CDR peptides ("minimal recognition units") can be obtained by constructing genes encoding the CDR of an antibody of interest. Such genes are prepared, for example, by using the polymerase chain reaction to synthesize the variable region from RNA of antibody-producing cells. See, for example, Larrick and Fry [Methods, 2: 106-10 (1991)].

[0130] It will be appreciated that for human therapy or diagnostics, humanized antibodies are preferably used. Humanized forms of non-human (e.g., murine) antibodies are chimeric molecules of immunoglobulins, immunoglobulin chains or fragments thereof (such as Fv, Fab, Fab', F(ab').sub.2 or other antigen-binding subsequences of antibodies) which contain minimal sequence derived from non-human immunoglobulin. Humanized antibodies include human immunoglobulins (recipient antibody) in which residues form a complementary determining region (CDR) of the recipient are replaced by residues from a CDR of a non-human species (donor antibody) such as mouse, rat or rabbit having the desired specificity, affinity and capacity. In some instances, Fv framework residues of the human immunoglobulin are replaced by corresponding non-human residues. Humanized antibodies may also comprise residues which are found neither in the recipient antibody nor in the imported CDR or framework sequences. In general, the humanized antibody will comprise substantially all of at least one, and typically two, variable domains, in which all or substantially all of the CDR regions correspond to those of a non-human immunoglobulin and all or substantially all of the FR regions are those of a human immunoglobulin consensus sequence. The humanized antibody optimally also will comprise at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin [Jones et al., Nature, 321:522-525 (1986); Riechmann et al., Nature, 332:323- 329 (1988); and Presta, Curr. Op. Struct. Biol., 2:593-596 (1992)].

[0131] Methods for humanizing non-human antibodies are well known in the art. Generally, a humanized antibody has one or more amino acid residues introduced into it from a source which is non-human. These non-human amino acid residues are often referred to as import residues, which are typically taken from an import variable domain. Humanization can be essentially performed following the method of Winter and co-workers [Jones et al., Nature, 321:522-525 (1986); Riechmann et al., Nature 332:323-327 (1988); Verhoeyen et al., Science, 239: 1534-1536 (1988)], by substituting rodent CDRs or CDR sequences for the corresponding sequences of a human antibody. Accordingly, such humanized antibodies are chimeric antibodies (U.S. Pat. No. 4,816,567), wherein substantially less than an intact human variable domain has been substituted by the corresponding sequence from a non-human species. In practice, humanized antibodies are typically human antibodies in which some CDR residues and possibly some FR residues are substituted by residues from analogous sites in rodent antibodies.

[0132] Human antibodies can also be produced using various techniques known in the art, including phage display libraries [Hoogenboom and Winter, J. Mol. Biol., 227:381 (1991); Marks et al., J. Mol. Biol., 222:581 (1991)]. The techniques of Cole et al. and Boemer et al. are also available for the preparation of human monoclonal antibodies (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, p. 77 (1985) and Boerner et al., J. Immunol., 147(l):86-95 (1991)]. Similarly, human antibodies can be made by introduction of human immunoglobulin loci into transgenic animals, e.g., mice in which the endogenous immunoglobulin genes have been partially or completely inactivated. Upon challenge, human antibody production is observed, which closely resembles that seen in humans in all respects, including gene rearrangement, assembly, and antibody repertoire. This approach is described, for example, in U.S. Pat. Nos. 5,545,807; 5,545,806; 5,569,825; 5,625,126; 5,633,425; 5,661,016, and in the following scientific publications: Marks et al., Bio / Technology 10,: 779-783 (1992); Lonberg et al., Nature 368: 856-859 (1994); Morrison, Nature 368 812-13 (1994); Fishwild et al., Nature Biotechnology 14, 845-51 (1996); Neuberger, Nature Biotechnology 14: 826 (1996); and Lonberg and Huszar, Intern. Rev. Immunol. 13, 65-93 (1995).

[0133] As Zeb2 localized intracellularly, the antibody or antibody fragment capable can be an intracellular antibody (also known as “intrabodies”). Intracellular antibodies are essentially SCA to which intracellular localization signals have been added (e.g., ER, mitochondrial, nuclear, cytoplasmic). This technology has been successfully applied in the art (for review, see Richardson and Marasco, 1995, TIBTECH vol. 13). Intrabodies have been shown to virtually eliminate the expression of otherwise abundant cell surface receptors and to inhibit a protein function within a cell (See, for example, Richardson et al., 1995, Proc. Natl. Acad. Sci. USA 92: 3137-3141; Deshane et al., 1994, Gene Ther. 1: 332-337; Marasco et al., 1998 Human Gene Ther 9: 1627-42; Shaheen et al., 1996 J. Virol. 70: 3392-400; Werge, T. M. et al., 1990, FEBS Letters 274: 193-198; Carlson, J.R. 1993 Proc. Natl. Acad. Sci. USA 90:7427-7428; Biocca, S. et al., 1994, Bio / Technology 12: 396-399; Chen, S-Y. et al., 1994, Human Gene Therapy 5:595-601; Duan, L et al., 1994, Proc. Natl. Acad. Sci. USA 91:5075-5079; Chen, S-Y. et al., 1994, Proc. Natl. Acad. Sci. USA 91:5932-5936; Beerli, R.R. et al., 1994, J. Biol. Chem. 269:23931-23936; Mhashilkar, A.M. et al., 1995, EMBO J. 14: 1542-1551; PCT Publication No. WO 94 / 02610 by Marasco et al.; and PCT Publication No. WO 95 / 03832 by Duan et al.).

[0134] To prepare an intracellular antibody expression vector, the cDNA encoding the antibody light and heavy chains specific for the target protein of interest are isolated, typically from a hybridoma that secretes a monoclonal antibody specific for the marker. Hybridomas secreting anti-marker monoclonal antibodies, or recombinant monoclonal antibodies, can be prepared using methods known in the art. Once a monoclonal antibody specific for the marker protein is identified (e.g., either a hybridoma-derived monoclonal antibody or a recombinant antibody from a combinatorial library), DNAs encoding the light and heavy chains of the monoclonal antibody are isolated by standard molecular biology techniques. For hybridoma derived antibodies, light and heavy chain cDNAs can be obtained, for example, by PCR amplification or cDNA library screening. For recombinant antibodies, such as from a phage display library, cDNA encoding the light and heavy chains can be recovered from the display package (e.g., phage) isolated during the library screening process and the nucleotide sequences of antibody light and heavy chain genes are determined. For example, many such sequences are disclosed in Kabat, E. A., et al. (1991) Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242 and in the "Vbase" human germline sequence database. Once obtained, the antibody light and heavy chain sequences are cloned into a recombinant expression vector using standard methods.

[0135] For cytoplasmic expression of the light and heavy chains, the nucleotide sequences encoding the hydrophobic leaders of the light and heavy chains are removed. An intracellular antibody expression vector can encode an intracellular antibody in one of several different forms. For example, in one embodiment, the vector encodes full-length antibody light and heavy chains such that a full-length antibody is expressed intracellularly. In another embodiment, the vector encodes a full-length light chain but only the VH / CH1 region of the heavy chain such that a Fab fragment is expressed intracellularly. In another embodiment, the vector encodes a single chain antibody (scFv) wherein the variable regions of the light and heavy chains are linked by a flexible peptide linker [e.g., (Gly4Ser)3 and expressed as a single chain molecule. To inhibit marker activity in a cell, the expression vector encoding the intracellular antibody is introduced into the cell by standard transfection methods, as discussed hereinbefore.

[0136] Once antibodies are obtained, they may be tested for activity, for example via ELISA.

[0137] Another downregulating agent which can be used along with some embodiments of the invention is an aptamer. As used herein, the term “aptamer” refers to double stranded or single stranded RNA molecule that binds to specific molecular target. An aptamer can function at the polynucleotide or the polypeptide level, depending on its composition, and thus can e.g., inhibit translation of an mRNA or interfere with the activity of the polynucleotide or the protein. Various methods are known in the art which can be used to design target specific aptamers. The skilled artisan can employ SELEX (Systematic Evolution of Ligands by Exponential Enrichment) for efficient selection as described in Stoltenburg R, Reinemann C, and Strehlitz B (Biomolecular engineering (2007) 24(4):381-403).

[0138] According to specific embodiments, the downregulating agent is a PROTAC. PROTAC (proteolysis targeting chimera) is a bifunctional molecule composed to two active domains, one is a targeting domain that binds to a target protein and the other is capable of engaging an E3 ubiquitin ligase, thereby resulting in ubiquitination and subsequent degradation of the target protein via the proteasome. The targeting domain may be for example a polynucleotide, a polypeptide or a small molecule and the ligase recruiting domain is typically a small molecule or a polypeptide. PROTAC technology is known in the art and disclosed in e.g., in Nalawansha and Crews, Cell Chemical Biology (2020) 27: 998-1014, the contents of which are fully incorporated herein by reference. Non-limiting examples of PROTACs that can be used with specific embodiments of the invention include SR- 1114 and SPFl bxi4-\

[0139] Another examples of downregulating agent would be any molecule which interferes with Zeb2 activity (e.g., binding or interaction) by binding the target protein and / or cleaving the target protein. Such molecules can be a small molecule, antagonist, or inhibitory peptide.

[0140] It will be appreciated that a non-functional analogue of at least a binding or catalytic portion of the target can be also used as an inhibitory agent. Downregulating at the nucleic acid level

[0141] Inhibition at the nucleic acid level is typically effected using a nucleic acid agent, having a nucleic acid backbone, DNA, RNA, mimetics thereof or a combination of same. The nucleic acid agent may be encoded from a DNA molecule or provided to the cell per se.

[0142] Thus, according to specific embodiments, the downregulating agent is a polynucleotide.

[0143] As used herein the term “polynucleotide”, “oligonucleotide”, “nucleic acid sequence” or “nucleic acid molecule”, which are interchangeably used herein, refers to a single or double stranded nucleic acid sequence in the form of an RNA sequence (e.g., mRNA), a complementary polynucleotide sequence (cDNA), a genomic polynucleotide sequence (e.g., sequence isolated from a chromosome), a composite polynucleotide sequences (e.g., a combination of the above) or mimetic or analog thereof. This term includes polynucleotides derived from naturally occurring nucleic acids molecules (e.g., RNA or DNA), synthetic polynucleotide and / or oligonucleotide molecules composed of naturally occurring bases, sugars, and covalent intemucleoside linkages (e.g., backbone), as well as synthetic polynucleotides having non-naturally occurring portions, which function similarly to the respective naturally occurring portions.

[0144] According to specific embodiments, the polynucleotide is modified to render the polynucleotide to improve specificity, selectivity, affinity, and stability e.g., more resistant to degradation via e.g., exo- or endo- nucleases. Such modifications are well known in the art and disclosed e.g. in International Patent Application Publication No. WO2022 / 153322, U.S. Pat. No. 3,687,808; Kroschwitz, J. I., 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, the contents of which are fully incorporated herein by reference.

[0145] Non-limiting examples of modifications include a sugar modification, a nucleobase modification, and an internucleotide linkage modification.

[0146] Exemplary sugar modifications include, but are not limited to, 2'-modified nucleotide, e.g., a 2'-deoxy, 2'-fluoro (2'-F), 2'-deoxy-2'-fluoro, 2'-O-methyl (2'-0-Me), 2'-O-methoxyethyl (2'-0-M0E), 2'-O-aminopropyl (2'-O-AP), 2'-O-dimethylaminoethyl (2'-O-DMAOE), 2'-O- dimethylaminopropyl (2'-O-DMAP), 2'-O-dimethylaminoethyloxyethyl (2'-O-DMAEOE), 2'- Fluoroarabinooligonucleotides (2'-F-ANA), 2'-O— N-methylacetamido (2'-0-NMA), 2'-NH2 or a locked nucleic acid (LNA). Additional sugar modifications are described in Deleavey and Darnha, Chemistry and Biology (2012) 19: 937-954, incorporated herein by reference. Exemplary nucleobase modifications include, but not limited to, inclusion of locked nucleic acids (LNA), e.g., inclusion of nucleic acid analogues in which the ribose ring is “locked” by a methylene bridge connecting the 2’-0 atom and the 4’-C atom, ethylene nucleic acids (ENA), e.g., 2'-4'-ethylene-bridged nucleic acids, 2-amino-A, 2-thio (e.g., 2-thio-U), G-clamp modifications.

[0147] Exemplary intemucleotide linkage modifications include, but are not limited to, phosphorothioate, chiral phosphorothioate, phosphorodithioate, phosphotriester, aminoalkyl phosphotriester, methyl phosphonate, alkyl phosphonate (including 3'-alkylene phosphonates), chiral phosphonate, phosphinate, phosphoramidate (including 3'-amino phosphoramidate), aminoalkylphosphoramidate, thionophosphoramidate, thionoalkylphosphonate, thionoalkylphosphotriester, boranophosphate (such as that 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' to 5'-2'), boron phosphonate, phosphodiester, phosphonoacetate (PACE), morpholino, peptide nucleic acid (PNA) and threose nucleic acid (TNA). Various salts, mixed salts, and free acid forms of the above modifications can also be used. Additional internucleotide linkage modifications are described in Deleavey and Darnha, Chemistry and Biology (2012) 19: 937-954, incorporated herein by reference.

[0148] According to a specific embodiment, the modification comprises a phosphorothioate.

[0149] According to specific embodiments the polynucleotide has a phosphorothioate modified phosphate backbone (i.e., at least one of the phosphate oxygens is replaced by sulfur).

[0150] Other non-limiting examples of stabilized polynucleotides include nonionic DNA analogs, such as alkyl- and aryl-phosphonates (in which the charged phosphonate oxygen is replaced by an alkyl or aryl group), phosphodiester and alkylphosphotriesters, in which the charged oxygen moiety is alkylated. Polynucleotides which contain a diol, such as tetraethyleneglycol or hexaethyleneglycol, at either or both termini have also been shown to be substantially resistant to nuclease degradation.

[0151] It should be noted that any of the sequences provided herein may comprise deoxyribonucleic acids or ribonucleic acids; hence in any of these sequences the nucleotide “T” may be replaced with “U” and vice versa.

[0152] According to specific embodiments, the polynucleotide is a single stranded molecule.

[0153] According to specific embodiments, the polynucleotide is a double stranded molecule.

[0154] According to a specific embodiment, the agent comprises a first single stranded moiety (e.g., targeting moiety) and a second double stranded moiety (e.g., downregulating moiety), as further described hereinbelow. According to specific embodiments, the agent comprises a dsRNA in which one of the strands has an overhang and the first moiety has a complementary overhang. After annealing, both components are connected creating a desired construct. By controlling the length of the overhang and its makeup the strength and the specificity of the attachment can be controlled. Exemplary components of the overhang are 2'-O-methyl RNA (2'-0Me), 2'-Fluoro RNA (2'-F) or Locked Nucleic Acids (LNAs) or PNA. Extremely high melting temperatures of an LNA / LNA duplex allow for the use of much shorter overhangs. 2'-Fluoro RNA (2'-F) were reported to have lower toxicity then 2'-O-methyl RNA (2'-0Me). Use of all of the above increases the resistance of the oligonucleotide to cellular nucleases, see e.g., Kurreck et al. (2002, Braasch et al. (2002) and Braasch et al. (2003). Other exemplary sugar modifications include, for example, a 2'-O- methoxyethyl nucleotide, a 2'-0-NMA, a 2'-DMAE0E, a 2'-AP, 2'-hydroxy, or a 2'-ara-fluoro or extended nucleic acid (ENA), hexose nucleic acid (HNA), or cyclohexene nucleic acid (CeNA). The use of overhangs for the construction allows for: (i) use of smaller molecules, (ii) higher purity at lower cost, (iii) lower cost of final product and (iv) flexibility (construction of product on demand; possibility of matching of one component with multiple components). The use of a universal overhangs allows for the interchangeability of the components.

[0155] Polynucleotides designed according to the teachings of some embodiments of the invention can be generated according to any polynucleotide synthesis method known in the art such as enzymatic synthesis or solid phase synthesis. 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 polynucleotides is well within the capabilities of one skilled in the art and can be accomplished via established methodologies as detailed in, 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) and “Oligonucleotide Synthesis” Gait, M. J., ed. (1984) utilizing solid phase chemistry, e.g. cyanoethyl phosphoramidite followed by deprotection, desalting and purification by for example, an automated trityl-on method or HPLC.

[0156] According to specific embodiments, the polynucleotide is comprised in a nucleic acid construct suitable for expression in a cell e.g., a mammalian cell. 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. According to specific embodiments, inhibition can be achieved by RNA silencing. As used herein, the phrase "RNA silencing" refers to a group of regulatory mechanisms [e.g., RNA interference (RNAi), transcriptional gene silencing (TGS), post-transcriptional gene silencing (PTGS), quelling, co-suppression, and translational repression] mediated by RNA molecules which result in the inhibition or "silencing" of the expression of a corresponding protein-coding gene. RNA silencing has been observed in many types of organisms, including plants, animals, and fungi.

[0157] As used herein, the term "RNA silencing agent" refers to an RNA which is capable of specifically inhibiting or "silencing" the expression of a target gene. In certain embodiments, the RNA silencing agent is capable of preventing complete processing (e.g., the full translation and / or expression) of an mRNA molecule through a post-transcriptional silencing mechanism. RNA silencing agents include non-coding RNA molecules, for example RNA duplexes comprising paired strands, as well as precursor RNAs from which such small non-coding RNAs can be generated. Exemplary RNA silencing agents include dsRNAs such as siRNAs, miRNAs and shRNAs.

[0158] In one embodiment, the RNA silencing agent is capable of inducing RNA interference.

[0159] In another embodiment, the RNA silencing agent is capable of mediating translational repression.

[0160] According to an embodiment of the invention, the RNA silencing agent is specific to the target RNA (e.g., Zeb2) and does not cross inhibit or silence other targets or a splice variant which exhibits 99% or less global homology to the target gene, e.g., less than 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81% global homology to the target gene; as determined by PCR, Western blot, Immunohistochemistry and / or flow cytometry.

[0161] RNA interference refers to the process of sequence- specific post-transcriptional gene silencing in animals mediated by short interfering RNAs (siRNAs).

[0162] Following is a detailed description on RNA silencing agents that can be used according to specific embodiments of the present invention.

[0163] DsRNA, siRNA and shRNA - The presence of long dsRNAs in cells stimulates the activity of a ribonuclease III enzyme referred to as dicer. Dicer is involved in the processing of the dsRNA into short pieces of dsRNA known as short interfering RNAs (siRNAs). Short interfering RNAs derived from dicer activity are typically about 21 to about 23 nucleotides in length and comprise about 19 base pair duplexes. The RNAi response also features an endonuclease complex, commonly referred to as an RNA-induced silencing complex (RISC), which mediates cleavage of single- stranded RNA having sequence complementary to the antisense strand of the siRNA duplex. Cleavage of the target RNA takes place in the middle of the region complementary to the antisense strand of the siRNA duplex.

[0164] Accordingly, some embodiments of the invention contemplate use of dsRNA to inhibit protein expression from mRNA.

[0165] According to one embodiment dsRNA longer than 30 bp are used. Various studies demonstrate that long dsRNAs can be used to silence gene expression without inducing the stress response or causing significant off-target effects - see for example [Strat et al., Nucleic Acids Research, 2006, Vol. 34, No. 13 3803-3810; Bhargava A et al. Brain Res. Protoc. 2004;13: 115- 125; Diallo M., et al., Oligonucleotides. 2003;13:381-392; Paddison P.J., et al., Proc. Natl Acad. Sci. USA. 2002;99: 1443-1448; Tran N., et al., FEBS Lett. 2004;573: 127-134],

[0166] According to some embodiments of the invention, dsRNA is provided in cells where the interferon pathway is not activated, see for example Billy et al., PNAS 2001, Vol 98, pages 14428-14433; and Diallo et al, Oligonucleotides, October 1, 2003, 13(5): 381-392. doi: 10.1089 / 154545703322617069.

[0167] According to an embodiment of the invention, the long dsRNA is specifically designed not to induce the interferon and PKR pathways for down-regulating gene expression. For example, Shinagwa and Ishii [Genes & Dev. 17 (11): 1340-1345, 2003] have developed a vector, named pDECAP, to express long double-strand RNA from an RNA polymerase II (Pol II) promoter. Because the transcripts from pDECAP lack both the 5'-cap structure and the 3'-poly(A) tail that facilitate ds-RNA export to the cytoplasm, long ds-RNA from pDECAP does not induce the interferon response.

[0168] Another method of evading the interferon and PKR pathways in mammalian systems is by introduction of small inhibitory RNAs (siRNAs) either via transfection or endogenous expression.

[0169] The term "siRNA" refers to small inhibitory RNA duplexes (generally between 18-30 base pairs) that induce the RNA interference (RNAi) pathway. Typically, siRNAs are chemically synthesized as 21mers with a central 19 bp duplex region and symmetric 2-base 3'-overhangs on the termini, although it has been recently described that chemically synthesized RNA duplexes of 25-30 base length can have as much as a 100-fold increase in potency compared with 21mers at the same location. The observed increased potency obtained using longer RNAs in triggering RNAi is suggested to result from providing Dicer with a substrate (27mer) instead of a product (21mer) and that this improves the rate or efficiency of entry of the siRNA duplex into RISC. It has been found that position of the 3'-overhang influences potency of an siRNA and asymmetric duplexes having a 3 '-overhang on the antisense strand are generally more potent than those with the 3'-overhang on the sense strand (Rose et al., 2005). This can be attributed to asymmetrical strand loading into RISC, as the opposite efficacy patterns are observed when targeting the antisense transcript.

[0170] The strands of a double- stranded interfering RNA (e.g., an siRNA) may be connected to form a hairpin or stem-loop structure (e.g., an shRNA). Thus, as mentioned, the RNA silencing agent of some embodiments of the invention may also be a short hairpin RNA (shRNA).

[0171] The term "shRNA", as used herein, refers to an RNA agent having a stem-loop structure, comprising a first and second region of complementary sequence, the degree of complementarity and orientation of the regions being sufficient such that base pairing occurs between the regions, the first and second regions being joined by a loop region, the loop resulting from a lack of base pairing between nucleotides (or nucleotide analogs) within the loop region. The number of nucleotides in the loop is a number between and including 3 to 23, or 5 to 15, or 7 to 13, or 4 to 9, or 9 to 11. Some of the nucleotides in the loop can be involved in base-pair interactions with other nucleotides in the loop. Examples of oligonucleotide sequences that can be used to form the loop include 5'-CAAGAGA-3' and 5’-UUACAA-3’ (International Patent Application Nos. WO2013126963 and WO2014107763). It will be recognized by one of skill in the art that the resulting single chain oligonucleotide forms a stem-loop or hairpin structure comprising a double- stranded region capable of interacting with the RNAi machinery.

[0172] Synthesis of RNA silencing agents suitable for use with some embodiments of the invention can be effected as follows. First, the mRNA sequence is scanned downstream of the AUG start codon for AA dinucleotide sequences. Occurrence of each AA and the 3’ adjacent 19 nucleotides is recorded as potential siRNA target sites. Preferably, siRNA target sites are selected from the open reading frame, as untranslated regions (UTRs) are richer in regulatory protein binding sites. UTR-binding proteins and / or translation initiation complexes may interfere with binding of the siRNA endonuclease complex [Tuschl ChemBiochem. 2:239-245]. It will be appreciated though, that siRNAs directed at untranslated regions may also be effective, as demonstrated for GAPDH wherein siRNA directed at the 5’ UTR mediated about 90 % decrease in cellular GAPDH mRNA and completely abolished protein level (www(dot)ambion(dot)com / techlib / tn / 91 / 912.html).

[0173] Second, potential target sites are compared to an appropriate genomic database (e.g., human, mouse, rat etc.) using any sequence alignment software, such as the BLAST software available from the NCBI server (www(dot)ncbi.nlm.nih(dot)gov / BLAST / ). Putative target sites which exhibit significant homology to other coding sequences are filtered out.

[0174] Qualifying target sequences are selected as template for siRNA synthesis. Preferred sequences are those including low G / C content as these have proven to be more effective in mediating gene silencing as compared to those with G / C content higher than 55 %. Several target sites are preferably selected along the length of the target gene for evaluation. For better evaluation of the selected siRNAs, a negative control is preferably used in conjunction. Negative control siRNA preferably includes the same nucleotide composition as the siRNAs but lack significant homology to the genome. Thus, a scrambled nucleotide sequence of the siRNA is preferably used, provided it does not display any significant homology to any other gene.

[0175] It will be appreciated that, and as mentioned hereinabove, the RNA silencing agent of some embodiments of the invention need not be limited to those molecules containing only RNA, but further encompasses chemically-modified nucleotides and non-nucleotides as further described herein above and below.

[0176] Non-limiting examples of Zeb2 sequences that can be targeted by such RNA silencing (also known as the “passenger strand”) agent are provided in SEQ ID NOs: 15, 17, 19 and 21.

[0177] Non-limiting examples of guide sequences of such RNA silencing agents (e.g., siRNA) that can be used with specific embodiments of the invention are provided in SEQ ID NOs: 16, 18, 20 and 22.

[0178] According to specific embodiments, the silencing agent is provided as a single stranded nucleic acid sequence.

[0179] According to other specific embodiments, the silencing agent is provided as a double stranded nucleic acid sequence.

[0180] It should be noted that the first and second sequences are not required to be completely complementary, they only need to be substantially complementary to anneal under biological conditions and to provide a substrate for Dicer that produces a siRNA sufficiently complementary to the target sequence.

[0181] According to a specific embodiment, the guide strand comprises a sequence that is 100 % complementary, i.e., complete match, to the passenger strand.

[0182] According to specific embodiments, the nucleic acid sequence of the RNA silencing agent may be modified with e.g., phosphothioate. A non-limiting example of such sequence are provided in SEQ ID NO: 23 (passenger strand) and its complementary guide sequence SEQ ID NO: 24.

[0183] Thus, according to a specific embodiment, the agent comprises SEQ ID NOs: 23 and 24. miRNA and miRNA mimics - According to another embodiment the RNA silencing agent may be a miRNA.

[0184] The term "microRNA", "miRNA", and "miR" are synonymous and refer to a collection of non-coding single-stranded RNA molecules of about 19-28 nucleotides in length, which regulate gene expression. miRNAs are found in a wide range of organisms (viruses.fwdarw.humans) and have been shown to play a role in development, homeostasis, and disease etiology.

[0185] Below is a brief description of the mechanism of miRNA activity.

[0186] Genes coding for miRNAs are transcribed leading to production of a miRNA precursor known as the pri-miRNA. The pri-miRNA is typically part of a polycistronic RNA comprising multiple pri-miRNAs. The pri-miRNA may form a hairpin with a stem and loop. The stem may comprise mismatched bases.

[0187] The hairpin structure of the pri-miRNA is recognized by Drosha, which is an RNase III endonuclease. Drosha typically recognizes terminal loops in the pri-miRNA and cleaves approximately two helical turns into the stem to produce a 60-70 nucleotide precursor known as the pre-miRNA. Drosha cleaves the pri-miRNA with a staggered cut typical of RNase III endonucleases yielding a pre-miRNA stem loop with a 5' phosphate and ~2 nucleotide 3' overhang. It is estimated that approximately one helical turn of stem (~10 nucleotides) extending beyond the Drosha cleavage site is essential for efficient processing. The pre-miRNA is then actively transported from the nucleus to the cytoplasm by Ran-GTP and the export receptor Ex- portin-5.

[0188] The double-stranded stem of the pre-miRNA is then recognized by Dicer, which is also an RNase III endonuclease. Dicer may also recognize the 5' phosphate and 3' overhang at the base of the stem loop. Dicer then cleaves off the terminal loop two helical turns away from the base of the stem loop leaving an additional 5' phosphate and ~2 nucleotide 3' overhang. 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.

[0189] Although initially present as a double- stranded species with miRNA*, the miRNA eventually becomes incorporated as a single-stranded RNA into a ribonucleoprotein complex known as the RNA-induced silencing complex (RISC). Various proteins can form the RISC, which can lead to variability in specificity 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 removed and degraded. The strand of the miRNA:miRNA* duplex that is loaded into the RISC is 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.

[0190] The RISC identifies target nucleic acids based on high levels of complementarity between the miRNA and the mRNA, especially by nucleotides 2-7 of the miRNA.

[0191] 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 may 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).

[0192] The target sites in the mRNA may be in the 5' UTR, the 3' UTR or in the coding region. Interestingly, multiple miRNAs may regulate the same mRNA target by recognizing the same or multiple sites. The presence of multiple miRNA binding sites in most genetically identified targets may indicate that the cooperative action of multiple RISCs provides the most efficient translational inhibition. miRNAs may direct the RISC to down-regulate 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 is typically 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.

[0193] It should be noted that 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.

[0194] The term "microRNA mimic" or “miRNA mimic” refers to synthetic non-coding RNAs that are capable of entering the RNAi pathway and regulating gene expression. miRNA mimics imitate the function of endogenous miRNAs and can be designed as mature, double stranded molecules or mimic precursors (e.g., or pre-miRNAs). miRNA mimics can be comprised of modified or unmodified RNA, DNA, RNA-DNA hybrids, or alternative nucleic acid chemistries (e.g., LNAs or 2'-O,4'-C-ethylene-bridged nucleic acids (ENA)). For mature, double stranded miRNA mimics, the length of the duplex region can vary between 13-33, 18-24 or 21-23 nucleotides. The miRNA may also comprise a total of at least 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39 or 40 nucleotides. The sequence of the miRNA may be the first 13-33 nucleotides of the pre-miRNA. The sequence of the miRNA may also be the last 13-33 nucleotides of the pre-miRNA.

[0195] Preparation of miRNAs mimics can be effected by any method known in the art such as chemical synthesis or recombinant methods.

[0196] It will be appreciated from the description provided herein above that contacting cells with a miRNA may be effected by transfecting the cells with e.g. the mature double stranded miRNA, the pre-miRNA or the pri-miRNA.

[0197] The pre-miRNA sequence may comprise from 45-90, 60-80 or 60-70 nucleotides.

[0198] The pri-miRNA sequence may comprise from 45-30,000, 50-25,000, 100-20,000, 1,000- 1,500 or 80-100 nucleotides.

[0199] Non-limiting examples of microRNAs that can be used with specific embodiments of the invention are MiR-30a-5p, MiR-101, MiR-124, MiR-129, MiR-132, MiR-138, MiR-139-5p, MiR-141, MiR-144, MiR-145, MiR-145-5p, MiR-153, MiR-154, MiR-155, MiR-187, MiR-200a, MiR-200b, MiR-200c, MiR-203, MiR-205, MiR-206, MiR-211-5p, MiR-215, MiR-335, MiR- 338-3p, MiR-454-3p, MiR-374b-5p, MiR-448, MiR-506, MiR-545, MiR-598, MiR-622, MiR- 590-3p, MiR-769-3p, MiR-940, MiR-1179 and MiR-3653.

[0200] Antisense - Antisense is a single stranded RNA designed to prevent or inhibit expression of a gene by specifically hybridizing to its mRNA. Inhibition can be effected using an antisense polynucleotide capable of specifically hybridizing with an mRNA transcript encoding the target (e.g. MYC).

[0201] Design of antisense molecules which can be used to efficiently down-regulate a target must be effected while considering two aspects important to the antisense approach. The first aspect is delivery of the oligonucleotide into the cytoplasm of the appropriate cells, while the second aspect is design of an oligonucleotide which specifically binds the designated mRNA within cells in a way which inhibits translation thereof.

[0202] The prior art teaches of a number of delivery strategies which can be used to efficiently deliver oligonucleotides into a wide variety of cell types [see, for example, Jaaskelainen et al. Cell Mol Biol Lett. (2002) 7(2):236-7; Gait, Cell Mol Life Sci. (2003) 60(5):844-53; Martino et al. J Biomed Biotechnol. (2009) 2009:410260; Grijalvo et al. Expert Opin Ther Pat. (2014) 24(7):801-19; Falzarano et al, Nucleic Acid Ther. (2014) 24(l):87-100; Shilakari et al. Biomed Res Int. (2014) 2014: 526391; Prakash et al. Nucleic Acids Res. (2014) 42(13):8796-807 and Asseline et al. J Gene Med. (2014) 16(7-8): 157-65]

[0203] In addition, algorithms for identifying those sequences with the highest predicted binding affinity for their target mRNA based on a thermodynamic cycle that accounts for the energetics of structural alterations in both the target mRNA and the oligonucleotide are also available [see, for example, Walton et al. Biotechnol Bioeng 65: 1-9 (1999)]. Such algorithms have been successfully used to implement an antisense approach in cells.

[0204] In addition, several approaches for designing and predicting efficiency of specific oligonucleotides using an in vitro system were also published (Matveeva et al., Nature Biotechnology 16: 1374 - 1375 (1998)].

[0205] Thus, the generation of highly accurate antisense design algorithms and a wide variety of oligonucleotide delivery systems, enable an ordinarily skilled artisan to design and implement antisense approaches suitable for down-regulating expression of known sequences without having to resort to undue trial and error experimentation.

[0206] Nucleic acid agents can also operate at the DNA level as summarized infra.

[0207] Inhibition can also be achieved by inactivating the gene via introducing targeted mutations involving loss-of function alterations (e.g., point mutations, deletions and insertions) in the gene structure.

[0208] As used herein, the phrase “loss-of-function alterations” refers to any mutation in the DNA sequence of a gene which results in down-regulation of the expression level and / or activity of the expressed product, i.e., the mRNA transcript and / or the translated protein. Non-limiting examples of such loss-of-function alterations include a missense mutation, i.e., a mutation which changes an amino acid residue in the protein with another amino acid residue and thereby abolishes the enzymatic activity of the protein; a nonsense mutation, i.e., a mutation which introduces a stop codon in a protein, e.g., an early stop codon which results in a shorter protein devoid of the enzymatic activity; a frame-shift mutation, i.e., a mutation, usually, deletion or insertion of nucleic acid(s) which changes the reading frame of the protein, and may result in an early termination by introducing a stop codon into a reading frame (e.g., a truncated protein, devoid of the enzymatic activity), or in a longer amino acid sequence (e.g., a readthrough protein) which affects the secondary or tertiary structure of the protein and results in a non-functional protein, devoid of the enzymatic activity of the non-mutated polypeptide; a readthrough mutation due to a frame- shift mutation or a modified stop codon mutation (z.e., when the stop codon is mutated into an amino acid codon), with an abolished enzymatic activity; a promoter mutation, z.e., a mutation in a promoter sequence, usually 5' to the transcription start site of a gene, which results in down-regulation of a specific gene product; a regulatory mutation, i.e., a mutation in a region upstream or downstream, or within a gene, which affects the expression of the gene product; a deletion mutation, i.e., a mutation which deletes coding nucleic acids in a gene sequence and which may result in a frame- shift mutation or an in-frame mutation (within the coding sequence, deletion of one or more amino acid codons); an insertion mutation, i.e., a mutation which inserts coding or non-coding nucleic acids into a gene sequence, and which may result in a frame-shift mutation or an in-frame insertion of one or more amino acid codons; an inversion, i.e., a mutation which results in an inverted coding or non-coding sequence; a splice mutation i.e., a mutation which results in abnormal splicing or poor splicing; and a duplication mutation, i.e., a mutation which results in a duplicated coding or non-coding sequence, which can be in-frame or can cause a frame- shift.

[0209] According to specific embodiments loss-of-function alteration of a gene may comprise at least one allele of the gene.

[0210] The term "allele" as used herein, refers to any of one or more alternative forms of a gene locus, all of which alleles relate to a trait or characteristic. In a diploid cell or organism, the two alleles of a given gene occupy corresponding loci on a pair of homologous chromosomes.

[0211] According to other specific embodiments loss-of-function alteration of a gene comprises both alleles of the gene. In such instances the gene may be in a homozygous form or in a heterozygous form.

[0212] Methods of introducing nucleic acid alterations to a gene of interest are well known in the art [see for example Menke D. Genesis (2013) 51: - 618; Capecchi, Science (1989) 244: 1288- 1292; Santiago et al. Proc Natl Acad Sci USA (2008) 105:5809-5814; International Patent Application Nos. WO 2014085593, WO 2009071334 and WO 2011146121; US Patent Nos. 8771945, 8586526, 6774279 and UP Patent Application Publication Nos. 20030232410, 20050026157, US20060014264; the contents of which are incorporated by reference in their entireties] and include targeted homologous recombination, site specific recombinases, PB transposases and genome editing by engineered nucleases. Agents for introducing nucleic acid alterations to a gene of interest can be designed publically available sources or obtained commercially from Transposagen, Addgene and Sangamo Biosciences.

[0213] Following is a description of various exemplary methods used to introduce nucleic acid alterations to a gene of interest and agents for implementing same that can be used according to specific embodiments of the present invention.

[0214] According to specific embodiments, the downregulating agent is a genome editing agent.

[0215] Genome Editing using engineered endonucleases - this approach refers to a reverse genetics method using artificially engineered nucleases to cut and create specific double- stranded breaks at a desired location(s) in the genome, which are then repaired by cellular endogenous processes such as, homology directed repair (HDR) and non-homologous end-joining (NHEJ). NHEJ directly joins the DNA ends in a double-stranded break, while HDR utilizes a homologous sequence as a template for regenerating the missing DNA sequence at the break point. In order to introduce specific nucleotide modifications to the genomic DNA, a DNA repair template containing the desired sequence must be present during HDR. Genome editing cannot be performed using traditional restriction endonucleases since most restriction enzymes recognize a few base pairs on the DNA as their target and the probability is very high that the recognized base pair combination will be found in many locations across the genome resulting in multiple cuts not limited to a desired location. To overcome this challenge and create site-specific single- or double- stranded breaks, several distinct classes of nucleases have been discovered and bioengineered to date. These include the meganucleases, Zinc finger nucleases (ZFNs), transcription-activator like effector nucleases (TALENs) and CRISPR / Cas system.

[0216] Meganucleases - Meganucleases are commonly grouped into four families: the LAGLID ADG family, the GIY-YIG family, the His-Cys box family and the HNH family. These families are characterized by structural motifs, which affect catalytic activity and recognition sequence. For instance, members of the LAGLID ADG family are characterized by having either one or two copies of the conserved LAGLID ADG motif. The four families of meganucleases are widely separated from one another with respect to conserved structural elements and, consequently, DNA recognition sequence specificity and catalytic activity. Meganucleases are found commonly in microbial species and have the unique property of having very long recognition sequences (>14bp) thus making them naturally very specific for cutting at a desired location. This can be exploited to make site- specific double- stranded breaks in genome editing. One of skill in the art can use these naturally occurring meganucleases, however the number of such naturally occurring meganucleases is limited. To overcome this challenge, mutagenesis and high throughput screening methods have been used to create meganuclease variants that recognize unique sequences. For example, various meganucleases have been fused to create hybrid enzymes that recognize a new sequence. Alternatively, DNA interacting amino acids of the meganuclease can be altered to design sequence specific meganucleases (see e.g., US Patent 8,021,867). Meganucleases can be designed using the methods described in e.g., Certo, MT et al. Nature Methods (2012) 9:073-975; U.S. Patent Nos. 8,304,222; 8,021,867; 8, 119,381; 8, 124,369; 8, 129,134; 8,133,697; 8,143,015; 8,143,016; 8, 148,098; or 8, 163,514, the contents of each are incorporated herein by reference in their entirety. Alternatively, meganucleases with site specific cutting characteristics can be obtained using commercially available technologies e.g., Precision Biosciences' Directed Nuclease Editor™ genome editing technology.

[0217] ZFNs and TALENs - Two distinct classes of engineered nucleases, zinc-finger nucleases (ZFNs) and transcription activator-like effector nucleases (TALENs), have both proven to be effective at producing targeted double- stranded breaks (Christian et al., 2010; Kim et al., 1996; Li et al., 2011; Mahfouz et al., 2011; Miller et al., 2010).

[0218] Basically, ZFNs and TALENs restriction endonuclease technology utilizes a non-specific DNA cutting enzyme which is linked to a specific DNA binding domain (either a series of zinc finger domains or TALE repeats, respectively). Typically, a restriction enzyme whose DNA recognition site and cleaving site are separate from each other is selected. The cleaving portion is separated and then linked to a DNA binding domain, thereby yielding an endonuclease with very high specificity for a desired sequence. An exemplary restriction enzyme with such properties is Fokl. Additionally, Fokl has the advantage of requiring dimerization to have nuclease activity and this means the specificity increases dramatically as each nuclease partner recognizes a unique DNA sequence. To enhance this effect, Fokl nucleases have been engineered that can only function as heterodimers and have increased catalytic activity. The heterodimer functioning nucleases avoid the possibility of unwanted homodimer activity and thus increase specificity of the double-stranded break.

[0219] Thus, for example to target a specific site, ZFNs and TALENs are constructed as nuclease pairs, with each member of the pair designed to bind adjacent sequences at the targeted site. Upon transient expression in cells, the nucleases bind to their target sites and the Fokl domains heterodimerize to create a double-stranded break. Repair of these double-stranded breaks through the non-homologous end-joining (NHEJ) pathway most often results in small deletions or small sequence insertions. Since each repair made by NHEJ is unique, the use of a single nuclease pair can produce an allelic series with a range of different deletions at the target site. The deletions typically range anywhere from a few base pairs to a few hundred base pairs in length, but larger deletions have successfully been generated in cell culture by using two pairs of nucleases simultaneously (Carlson et al., 2012; Lee et al., 2010). In addition, when a fragment of DNA with homology to the targeted region is introduced in conjunction with the nuclease pair, the double-stranded break can be repaired via homology directed repair to generate specific modifications (Li et al., 2011; Miller et al., 2010; Umov et al., 2005).

[0220] Although the nuclease portions of both ZFNs and TALENs have similar properties, the difference between these engineered nucleases is in their DNA recognition peptide. ZFNs rely on Cys2- His2 zinc fingers and TALENs on TALEs. Both of these DNA recognizing peptide domains have the characteristic that they are naturally found in combinations in their proteins. Cys2-His2 Zinc fingers typically found in repeats that are 3 bp apart and are found in diverse combinations in a variety of nucleic acid interacting proteins. TALEs on the other hand are found in repeats with a one-to-one recognition ratio between the amino acids and the recognized nucleotide pairs. Because both zinc fingers and TALEs happen in repeated patterns, different combinations can be tried to create a wide variety of sequence specificities. Approaches for making site-specific zinc finger endonucleases include, e.g., modular assembly (where Zinc fingers correlated with a triplet sequence are attached in a row to cover the required sequence), OPEN (low- stringency selection of peptide domains vs. triplet nucleotides followed by high- stringency selections of peptide combination vs. the final target in bacterial systems), and bacterial one-hybrid screening of zinc finger libraries, among others. ZFNs can also be designed and obtained commercially from e.g., Sangamo Biosciences™ (Richmond, CA).

[0221] Method for designing and obtaining TALENs are described in e.g., Reyon et al. Nature Biotechnology 2012 May;30(5):460-5; Miller et al. Nat Biotechnol. (2011) 29: 143-148; Cermak et al. Nucleic Acids Research (2011) 39 (12): e82 and Zhang et al. Nature Biotechnology (2011) 29 (2): 149-53. A recently developed web-based program named Mojo Hand was introduced by Mayo Clinic for designing TAL and TALEN constructs for genome editing applications (can be accessed through www(dot)talendesign(dot)org). TALEN can also be designed and obtained commercially from e.g., Sangamo Biosciences™ (Richmond, CA).

[0222] According to specific embodiments, the genome editing agent comprises a CRISPR / Cas system.

[0223] CRISPR-Cas system - Many bacteria and archaea contain endogenous RNA-based adaptive immune systems that can degrade nucleic acids of invading phages and plasmids. These systems consist of clustered regularly interspaced short palindromic repeat (CRISPR) genes that produce RNA components and CRISPR associated (Cas) genes that encode protein components. The CRISPR RNAs (crRNAs) contain short stretches of homology to specific viruses and plasmids and act as guides to direct Cas nucleases to degrade the complementary nucleic acids of the corresponding pathogen. Studies of the type II CRISPR / Cas system of Streptococcus pyogenes have shown that three components form an RNA / protein complex and together are sufficient for sequence-specific nuclease activity: the Cas9 nuclease, a crRNA containing 20 base pairs of homology to the target sequence, and a trans-activating crRNA (tracrRNA) (Jinek et al. Science (2012) 337: 816-821.). It was further demonstrated that a synthetic chimeric guide RNA (gRNA) composed of a fusion between crRNA and tracrRNA could direct Cas9 to cleave DNA targets that are complementary to the crRNA in vitro. It was also demonstrated that transient expression of Cas9 in conjunction with synthetic gRNAs can be used to produce targeted double- stranded brakes in a variety of different species (Cho et al., 2013; Cong et al., 2013; DiCarlo et al., 2013; Hwang et al., 2013a, b; Jinek et al., 2013; Mali et al., 2013).

[0224] The CRIPSR / Cas system for genome editing contains two distinct components: a gRNA and an endonuclease e.g. Cas9.

[0225] The gRNA is typically a 20 nucleotides sequence encoding a combination of the target homologous sequence (crRNA) and the endogenous bacterial RNA that links the crRNA to the Cas9 nuclease (tracrRNA) in a single chimeric transcript. The gRNA / Cas9 complex is recruited to the target sequence by the base-pairing between the gRNA sequence and the complement genomic DNA. For successful binding of Cas9, the genomic target sequence must also contain the correct Protospacer Adjacent Motif (PAM) sequence immediately following the target sequence. The binding of the gRNA / Cas9 complex localizes the Cas9 to the genomic target sequence so that the Cas9 can cut both strands of the DNA causing a double-strand break. Just as with ZFNs and TALENs, the double-stranded brakes produced by CRISPR / Cas can undergo homologous recombination or NHEJ.

[0226] The Cas9 nuclease has two functional domains: RuvC and HNH, each cutting a different DNA strand. When both of these domains are active, the Cas9 causes double strand breaks in the genomic DNA.

[0227] A significant advantage of CRISPR / Cas is that the high efficiency of this system coupled with the ability to easily create synthetic gRNAs enables multiple genes to be targeted simultaneously. In addition, the majority of cells carrying the mutation present biallelic mutations in the targeted genes.

[0228] However, apparent flexibility in the base-pairing interactions between the gRNA sequence and the genomic DNA target sequence allows imperfect matches to the target sequence to be cut by Cas9. Modified versions of the Cas9 enzyme containing a single inactive catalytic domain, either RuvC- or HNH-, are called ‘nickases’. With only one active nuclease domain, the Cas9 nickase cuts only one strand of the target DNA, creating a single-strand break or 'nick'. A single- strand break, or nick, is normally quickly repaired through the HDR pathway, using the intact complementary DNA strand as the template. However, two proximal, opposite strand nicks introduced by a Cas9 nickase are treated as a double-strand break, in what is often referred to as a 'double nick' CRISPR system. A double-nick can be repaired by either NHEJ or HDR depending on the desired effect on the gene target. Thus, if specificity and reduced off-target effects are crucial, using the Cas9 nickase to create a double-nick by designing two gRNAs with target sequences in close proximity and on opposite strands of the genomic DNA would decrease off- target effect as either gRNA alone will result in nicks that will not change the genomic DNA.

[0229] Modified versions of the Cas9 enzyme containing two inactive catalytic domains (dead Cas9, or dCas9) have no nuclease activity while still able to bind to DNA based on gRNA specificity. The dCas9 can be utilized as a platform for DNA transcriptional regulators to activate or repress gene expression by fusing the inactive enzyme to known regulatory domains. For example, the binding of dCas9 alone to a target sequence in genomic DNA can interfere with gene transcription.

[0230] There are a number of publicly available tools available to help choose and / or design target sequences as well as lists of bioinformatically determined unique gRNAs for different genes in different species such as the Feng Zhang lab's Target Finder, the Michael Boutros lab's Target Finder (E-CRISP), the RGEN Tools: Cas-OFFinder, the CasFinder: Flexible algorithm for identifying specific Cas9 targets in genomes and the CRISPR Optimal Target Finder.

[0231] Non-limiting Examples of gRNA sequences that can be used with specific embodiments of the invention are provided in SEQ ID NOs: 27-31, 36-38 and 42.

[0232] In order to use the CRISPR system, both gRNA and Cas9 should be expressed in a target cell. The insertion vector can contain both cassettes on a single plasmid or the cassettes are expressed from two separate plasmids. CRISPR plasmids are commercially available such as the px33O plasmid from Addgene.

[0233] “Hit and run” or “in-out” - involves a two-step recombination procedure. In the first step, an insertion-type vector containing a dual positive / negative selectable marker cassette is used to introduce the desired sequence alteration. The insertion vector contains a single continuous region of homology to the targeted locus and is modified to carry the mutation of interest. This targeting construct is linearized with a restriction enzyme at a one site within the region of homology, electroporated into the cells, and positive selection is performed to isolate homologous recombinants. These homologous recombinants contain a local duplication that is separated by intervening vector sequence, including the selection cassette. In the second step, targeted clones are subjected to negative selection to identify cells that have lost the selection cassette via intrachromosomal recombination between the duplicated sequences. The local recombination event removes the duplication and, depending on the site of recombination, the allele either retains the introduced mutation or reverts to wild type. The end result is the introduction of the desired modification without the retention of any exogenous sequences.

[0234] The “double-replacement” or “tag and exchange” strategy - involves a two-step selection procedure similar to the hit and run approach, but requires the use of two different targeting constructs. In the first step, a standard targeting vector with 3' and 5' homology arms is used to insert a dual positive / negative selectable cassette near the location where the mutation is to be introduced. After electroporation and positive selection, homologously targeted clones are identified. Next, a second targeting vector that contains a region of homology with the desired mutation is electroporated into targeted clones, and negative selection is applied to remove the selection cassette and introduce the mutation. The final allele contains the desired mutation while eliminating unwanted exogenous sequences.

[0235] Site-Specific Recombinases - The Cre recombinase derived from the Pl bacteriophage and Flp recombinase derived from the yeast Saccharomyces cerevisiae are site-specific DNA recombinases each recognizing a unique 34 base pair DNA sequence (termed “Lox” and “FRT”, respectively) and sequences that are flanked with either Lox sites or FRT sites can be readily removed via site-specific recombination upon expression of Cre or Flp recombinase, respectively. For example, the Lox sequence is composed of an asymmetric eight base pair spacer region flanked by 13 base pair inverted repeats. Cre recombines the 34 base pair lox DNA sequence by binding to the 13 base pair inverted repeats and catalyzing strand cleavage and religation within the spacer region. The staggered DNA cuts made by Cre in the spacer region are separated by 6 base pairs to give an overlap region that acts as a homology sensor to ensure that only recombination sites having the same overlap region recombine.

[0236] Basically, the site specific recombinase system offers means for the removal of selection cassettes after homologous recombination. This system also allows for the generation of conditional altered alleles that can be inactivated or activated in a temporal or tissue-specific manner. Of note, the Cre and Flp recombinases leave behind a Lox or FRT “scar” of 34 base pairs. The Lox or FRT sites that remain are typically left behind in an intron or 3' UTR of the modified locus, and current evidence suggests that these sites usually do not interfere significantly with gene function. Thus, Cre / Lox and Flp / FRT recombination involves introduction of a targeting vector with 3' and 5' homology arms containing the mutation of interest, two Lox or FRT sequences and typically a selectable cassette placed between the two Lox or FRT sequences. Positive selection is applied and homologous recombinants that contain targeted mutation are identified. Transient expression of Cre or Flp in conjunction with negative selection results in the excision of the selection cassette and selects for cells where the cassette has been lost. The final targeted allele contains the Lox or FRT scar of exogenous sequences.

[0237] Transposases - As used herein, the term “transposase” refers to an enzyme that binds to the ends of a transposon and catalyzes the movement of the transposon to another part of the genome.

[0238] As used herein the term “transposon” refers to a mobile genetic element comprising a nucleotide sequence which can move around to different positions within the genome of a single cell. In the process the transposon can cause mutations and / or change the amount of a DNA in the genome of the cell.

[0239] A number of transposon systems that are able to also transpose in cells e.g., vertebrates have been isolated or designed, such as Sleeping Beauty [Izsvak and Ivies Molecular Therapy (2004) 9, 147-156], piggyBac [Wilson et al. Molecular Therapy (2007) 15, 139-145], Tol2 [Kawakami et al. PNAS (2000) 97 (21): 11403-11408] or Frog Prince [Miskey et al. Nucleic Acids Res. Dec 1, (2003) 31(23): 6873-6881]. Generally, DNA transposons translocate from one DNA site to another in a simple, cut-and-paste manner. Each of these elements has their own advantages, for example, Sleeping Beauty is particularly useful in region- specific mutagenesis, whereas Tol2 has the highest tendency to integrate into expressed genes. Hyperactive systems are available for Sleeping Beauty and piggyBac. Most importantly, these transposons have distinct target site preferences, and can therefore introduce sequence alterations in overlapping, but distinct sets of genes. Therefore, to achieve the best possible coverage of genes, the use of more than one element is particularly preferred. The basic mechanism is shared between the different transposases, therefore, we will describe piggyBac (PB) as an example.

[0240] PB is a 2.5 kb insect transposon originally isolated from the cabbage looper moth, Trichoplusia ni. The PB transposon consists of asymmetric terminal repeat sequences that flank a transposase, PBase. PBase recognizes the terminal repeats and induces transposition via a “cut- and-paste” based mechanism, and preferentially transposes into the host genome at the tetranucleotide sequence TTAA. Upon insertion, the TTAA target site is duplicated such that the PB transposon is flanked by this tetranucleotide sequence. When mobilized, PB typically excises itself precisely to reestablish a single TTAA site, thereby restoring the host sequence to its pretransposon state. After excision, PB can transpose into a new location or be permanently lost from the genome.

[0241] Typically, the transposase system offers an alternative means for the removal of selection cassettes after homologous recombination quite similar to the use Cre / Lox or Flp / FRT. Thus, for example, the PB transposase system involves introduction of a targeting vector with 3' and 5' homology arms containing the mutation of interest, two PB terminal repeat sequences at the site of an endogenous TTAA sequence and a selection cassette placed between PB terminal repeat sequences. Positive selection is applied and homologous recombinants that contain targeted mutation are identified. Transient expression of PBase removes in conjunction with negative selection results in the excision of the selection cassette and selects for cells where the cassette has been lost. The final targeted allele contains the introduced mutation with no exogenous sequences.

[0242] For PB to be useful for the introduction of sequence alterations, there must be a native TTAA site in relatively close proximity to the location where a particular mutation is to be inserted.

[0243] Genome editing using recombinant adeno-associated virus (rAAV) platform - this genome-editing platform is based on rAAV vectors which enable insertion, deletion or substitution of DNA sequences in the genomes of live mammalian cells. The rAAV genome is a single-stranded deoxyribonucleic acid (ssDNA) molecule, either positive- or negative-sensed, which is about 4.7 kb long. These single- stranded DNA viral vectors have high transduction rates and have a unique property of stimulating endogenous homologous recombination in the absence of double-strand DNA breaks in the genome. One of skill in the art can design a rAAV vector to target a desired genomic locus and perform both gross and / or subtle endogenous gene alterations in a cell. rAAV genome editing has the advantage in that it targets a single allele and does not result in any off-target genomic alterations. rAAV genome editing technology is commercially available, for example, the rAAV GENESIS™ system from Horizon™ (Cambridge, UK).

[0244] Methods for qualifying efficacy and detecting sequence alteration are well known in the art and include, but not limited to, DNA sequencing, electrophoresis, an enzyme-based mismatch detection assay and a hybridization assay such as PCR, RT-PCR, RNase protection, in-situ hybridization, primer extension, Southern blot, Northern Blot and dot blot analysis.

[0245] Sequence alterations in a specific gene can also be determined at the protein level using e.g., chromatography, electrophoretic methods, immunodetection assays such as ELISA and western blot analysis and immunohistochemistry. In addition, one ordinarily skilled in the art can readily design a knock-in / knock-out construct including positive and / or negative selection markers for efficiently selecting transformed cells that underwent a homologous recombination event with the construct. Positive selection provides a means to enrich the population of clones that have taken up foreign DNA. Non-limiting examples of such positive markers include glutamine synthetase, dihydrofolate reductase (DHFR), markers that confer antibiotic resistance, such as neomycin, hygromycin, puromycin, and blasticidin S resistance cassettes. Negative selection markers are necessary to select against random integrations and / or elimination of a marker sequence (e.g., positive marker). Non-limiting examples of such negative markers include the herpes simplex-thymidine kinase (HSV-TK) which converts ganciclovir (GCV) into a cytotoxic nucleoside analog, hypoxanthine phosphoribosyltransferase (HPRT) and adenine phosphoribosytransferase (ARPT).

[0246] Another examples of downregulating agent that work at the polynucleotide level would be any molecule which interferes with Zeb2 expression. Such molecules can be for example a small molecule. A non-limiting example of such a small molecule is Teniposide (VM-26, CAS No. 29767-20-2).

[0247] As mentioned, the composition disclosed herein comprises a macrophage-targeting moiety.

[0248] The term “macrophage-targeting moiety”, as used herein, relates to a functional group which serves to target or direct the downregulating agent or the composition comprising same described herein specifically to macrophages, facilitating attachment and / or internalization of the agent into a macrophage, and such that the agent may exert a therapeutic effect. According to specific embodiments, the targeting moiety induces internalization of the agent into the macrophage. Such targeting moieties include, but are not limited to, antibodies, cell surface receptor, ligands, hormones, lipids, sugars and dextrans.

[0249] According to specific embodiments, the macrophage-targeting moiety is specific for a macrophage surface marker.

[0250] As used herein, the phrase “macrophage surface marker” refers to a molecule that is preferentially expressed or presented on the cell surface of macrophages.

[0251] Such markers are typically proteins or glycoproteins, and are well known in the art. Exemplary markers are further provided infra. Methods of determining expression of a marker are well known in the art and include e.g., western blot, flow cytometry, ELISA, immuno staining followed by microscopy etc.

[0252] According to specific embodiments, the marker is overexpressed on macrophages. As used herein, the term “overexpressed” refers to an increased expression of a marker on macrophages as compared to other cell types e.g., other immune cells, non-immune cells, pathologic cells e.g., cancerous cells etc., as can be determined by e.g., flow cytometry, western blot, ELISA, immunostaining followed by microscopy.

[0253] According to specific embodiments, the increased expression is at least 1.5fold, at least 2 fold, at least 3 fold, at least 5 fold, at least 10 fold, or at least 20 fold as compared the level of the marker as measured using the same assay.

[0254] According to specific embodiments, the marker is exclusively expressed on macrophages (i.e., not on other cell types), as can be determined by e.g., flow cytometry, western blot, ELISA, immunostaining followed by microscopy.

[0255] As used herein, the term “macrophage” includes both circulating monocytes and to macrophages (also referred to as mononuclear phagocytes) present in a tissue.

[0256] According to specific embodiments, the macrophage comprises a tissue macrophage.

[0257] According to specific embodiments, the macrophage comprises a circulating monocytes.

[0258] According to specific embodiments, the macrophage comprises a tumor associated macrophage (TAM).

[0259] Non-limiting examples of macrophage surface markers that can be used with specific embodiments of the invention include toll-like receptor, CD14, CD40, CDl lb, CD64, CD68, F4 / 80 (mice) / EMRl (human), lysozyme M, MAC-l / MAC-3, CD163, TREM2, mannose receptor and scavenger receptor.

[0260] According to specific embodiments, the macrophage surface marker is a mannose receptor. Such receptors are known in the art and include MRC1 (macrophage mannose receptor 1, CD206) and MRC2 (macrophage mannose receptor 2, CD280).

[0261] According to specific embodiments, the macrophage surface marker is a scavenger receptor. Such receptors are known in the art and include e.g. MSR1 (SC ARA 1 or SR-A1), MARCO (SCARA2 or SR-A6), CD36 (SCARB2 or SR-B2), CD68, and LOX-1 (lectin-like oxidized LDL receptor- 1).

[0262] According to specific embodiments, the macrophage surface marker is selected from the group consisting of toll-like receptor, CDl lb, CD68, CD163, TREM2, MSR1 and MRC1.

[0263] According to specific embodiments, the macrophage surface marker is a toll-like receptor.

[0264] Toll-like receptors are a class of single-pass membrane-spanning receptors that bind to structurally conserved molecules derived from microbes. TLRs are a type of pattern recognition receptor (PRR) and their ligands (e.g., bacterial cell-surface lipopolysaccharides (LPS), lipoproteins, lipopeptides and lipoarabinomannan; proteins such as flagellin from bacterial flagella; double-stranded RNA of viruses or the unmethylated CpG motifs of bacterial and viral DNA; and certain other RNA and DNA), are known collectively as pathogen-associated molecular patterns (PAMPs). Endogenous ligands of TLRs have also been identified, including fibrinogen, heat shock proteins (HSPs), and DNA.

[0265] According to specific embodiments the TLR refers to the human TLR.

[0266] Ten TLRs have been identified in human so far, namely TLR-1 to TLR- 10.

[0267] According to specific embodiments, the TLR is TLR1 (corresponding to human Gene ID

[0268] 7096).

[0269] According to specific embodiments the TLR1 refers to the human TLR1, such as provided in the following GeneBank Number NP_003254.

[0270] According to specific embodiments, the TLR is TLR2 (corresponding to human Gene ID

[0271] 7097).

[0272] According to specific embodiments the TLR2 refers to the human TLR2, such as provided in the following GeneBank Numbers NP_001305716, NP_001305718, NP_001305719, NP_001305720 and NP_001305722.

[0273] According to specific embodiments, the TLR is TLR4 (corresponding to human Gene ID

[0274] 7099).

[0275] According to specific embodiments the TLR4 refers to the human TLR4, such as provided in the following GeneBank Numbers NP_003257, NP_612564 and NP_612567.

[0276] According to specific embodiments, the TLR is TLR5 (corresponding to human Gene ID

[0277] 7100).

[0278] According to specific embodiments the TLR5 refers to the human TLR5, such as provided in the following GeneBank Number NP_003259.

[0279] According to specific embodiments, the TLR is TLR6 (corresponding to human Gene ID 10333).

[0280] According to specific embodiments the TLR6 refers to the human TLR6, such as provided in the following GeneBank Number NP_006059.

[0281] According to specific embodiments, the TLR is TLR7 (corresponding to human Gene ID 51284).

[0282] According to specific embodiments the TLR7 refers to the human TLR7, such as provided in the following GeneBank Number NP_057646.

[0283] According to specific embodiments, the TLR is TLR8 (corresponding to human Gene ID

[0284] 51311). According to specific embodiments the TLR8 refers to the human TLR8, such as provided in the following GeneBank Numbers NP_057694 and NP_619542.

[0285] According to specific embodiments, the TLR is TLR9 (corresponding to human Gene ID 54106).

[0286] According to specific embodiments the TLR9 refers to the human TLR9, such as provided in the following GeneBank Number NP_059138.

[0287] According to specific embodiments, the TLR is TLR 10 (corresponding to human Gene ID 81793).

[0288] According to specific embodiments the TLR 10 refers to the human TLR 10, such as provided in the following GeneBank Numbers NP_001017388, NP_001182035, NP_001182036, NP_001182037 and NP_112218.

[0289] Thus, the macrophage-targeting moiety is capable of binding the macrophage surface marker.

[0290] According to specific embodiments, the targeting moiety specifically binds the macrophage specific marker.

[0291] As used herein, the term “specifically binds a macrophage specific marker” refers to the ability to bind the target marker at a higher affinity compared to other markers.

[0292] Higher affinity can be, for examples, of at least 2, 5, 10, 100, 1000, 10000, or 100,000 or 1,000,000 fold or more using the same affinity assay. Methods of determining binding are well known in the art and include e.g., flow cytometry, immunoprecipitation, BiaCore, HPLC, Surface Plasmon Resonance assay (SPR) and bio-layer interferometry Blitz® assay.

[0293] According to specific embodiments, the inhibitor binds the target with a Kd < 103M, 10 ’4M, 10 “5M, 10’6M, <10’7M, <10’8M, < 10’9M, 10’10M, 10’11M, 1012M, each possibility represents a separate embodiment of the present invention.

[0294] Hence, according to specific embodiments, the targeting moiety is an antibody. Detailed description on antibodies is further provided hereinabove.

[0295] According to specific embodiments, the antibody is not a depleting antibody.

[0296] Non-limiting examples of such antibodies can be commercially obtained e.g., the MACRO antibodies of ThermoFisher Scientific, the anti-CD68 antibodies of Abeam. Another non-limiting example is the anti-Clever-1 mAb FP-1305 described in Mantovani, A. et al. (2022) Nat Rev Drug Discov 21, 799-820.

[0297] According to specific embodiments, the targeting moiety comprises a ligand or receptor of the marker. The terms "ligand" and “receptor” as used herein in the context of a targeting moiety includes the intact native molecule as well as functional fragments, modified sequences and analogs thereof that maintains at least the ability of the intact native molecule to bind the target. According to specific embodiments, binding of the ligand / receptor to the marker induces internalization of the composition disclosed herein.

[0298] Such ligands / receptors are known in the art and depend on the marker selected. Non- limiting examples of macrophage targeting moieties that can be used with specific embodiments of the invention are disclosed in US Patent Application Publication No. US20080214436; and Kortylewski, M., Swiderski, P., Herrmann, A. et al. (2007) Nat Biotechnol 27: 925-932, the contents of which are fully incorporated herein by reference.

[0299] Thus, for example, according to specific embodiments, the macrophage targeting moiety comprises mannose, fucose and / or N- acetylgluco s amine residues, either as individual monosaccharides or presented at terminal positions on glycans of glycoproteins, glycolipids, polysaccharides etc., which is the ligand of a mannose receptor.

[0300] According to a specific embodiment, the macrophage targeting moiety comprises mannose.

[0301] According to other specific embodiments, the macrophage targeting moiety comprises a CpG ODN which is the ligand for a toll-like receptor, specifically TLR-9.

[0302] CpG ODN is a short oligonucleotide that contains a cytosine nucleotide ("C") followed by a guanine nucleotide ("G") motif. The "p" refers to the phosphodiester link between consecutive nucleotides, although some ODN have a modified phosphorothioate (PS) backbone instead. According to specific embodiments, the CpG ODN is single stranded. According to specific embodiments, the CpG ODN comprises deoxyribonucleic acids. According to specific embodiments, the CpG ODN is between 2 to 100 base pairs in size (e.g., between 8 and 40 base pairs in size). According to specific embodiments, the CpG are unmethylated. According to specific embodiments, the CpG ODNA contains a consensus mitogenic CpG motif represented by the formula:

[0303] 5' X1X2CGX3X4 3' wherein C and G are unmethylated, XI, X2, X3 and X4 are nucleotides and a GCG trinucleotide sequence is not present at or near the 5' and 3' ends. Examples of CpG ODNs are described in U.S. Pat. Nos. 6,194,388 and 6,207,646 and US Patent Application Publication No. US20080214436, each incorporated herein by reference.

[0304] According to specific embodiments, the CpG ODN comprises stabilized nucleotides (e.g., phosphorothioate stabilized nucleotides). Additional description on stabilization is provided hereinabove. Such CpG ODNs can be easily designed by the skilled in the art or commercially obtained from e.g., IDT, InvivoGen, Novus Biologicals, Creative Biogene and others.

[0305] Non-limiting examples of CpG ODNs that can be used with specific embodiments of the invention are provided in SEQ ID NOs: 32-35.

[0306] According to specific embodiments, the macrophage-targeting moiety is a CpG ODN and the downregulating agent is an RNA silencing agent.

[0307] According to specific embodiments, the macrophage-targeting moiety is a single stranded CpG ODN and the downregulating agent is a double stranded RNA silencing agent.

[0308] According to specific embodiments, the macrophage-targeting moiety is mannose and the downregulating agent is a genome editing agent (e.g., a CRISPR / Cas system).

[0309] The composition disclosed herein incorporates the downregulation agent and the macrophage-targeting moiety, thereby allowing specific downregulating of Zeb2 in macrophages.

[0310] Thus, the two functional moieties are in association with each other.

[0311] As used herein, the phrase “in association with” refers to direct or indirect binding (or attachment) of between downregulating agent and the targeting or via a particle comprising the targeting moiety.

[0312] Thus, according to specific embodiments, the downregulating agent is attached to the macrophage-targeting moiety.

[0313] A detailed description on conjugation methods is provided infra.

[0314] The attachment can be covalent or non-covalent.

[0315] The attachment may be direct or via a linker.

[0316] According to specific embodiments, the targeting moiety is attached (directly or via a linker) to a double stranded Zeb2 RNA silencing agent. Under this scenario, the targeting moiety (or the linker) can be attached to wither the passenger or guide strand.

[0317] According to specific embodiments, the targeting moiety is attached (directly or via a linker) to the passenger strand of the Zeb2 RNA silencing agent.

[0318] According to specific embodiments, the targeting moiety is attached (directly or via a linker) to the Zeb2 RNA silencing agent through a 3' terminal nucleotide and / or 5' terminal nucleotide.

[0319] According to specific embodiments, the targeting moiety is attached (directly or via a linker) to the Zeb2 RNA silencing agent through a 5' terminal nucleotide.

[0320] According to specific embodiments, the attachment is via a linker. As used herein, the term “linker”, or any “connecting group”, “spacer”, “intervening moiety”, “bridge” and grammatical equivalents thereof, describes a group (e.g., a substituent) that is attached to two or more moieties in the composition.

[0321] Any linker known in the art can be used with specific embodiments of the invention, including but not limited to a polynucleotide, a polypeptide, a synthetic linker, a chemical moiety, a polymer, a particle.

[0322] Non-limiting examples of linker groups include modified or unmodified nucleotides, nucleosides, polymers, sugars, carbohydrates, polyalkylenes such as polyethylene glycols and polypropylene glycols, polyalcohols, polypropylenes, mixtures of ethylene and propylene glycols, polyalkylamines, polyamines such as polylysin and spermidine, polyesters such as poly(ethyl acrylate), polyphosphodiesters, aliphatics, and alkylenes. Moreover, linkers / linker chemistries that are based on omega-amino-1,3- diols, omega-amino-l,2-diols, hydroxyprolinols, omega-amino-alkanols, diethanolamines, omega- hydroxy-1, 3-diols, omega-hydroxy-1, 2-diols, omega-thio-l,3-diols, omega-thio -1, 2-diols, omega- carboxy-1, 3-diols, omega-carboxy -1, 2-diols, co-hydroxy-alkanols, omega-thio-alkanols, omega- carboxy-alkanols, functionalized oligoethylene glycols, allyl amine, acrylic acid, allyl alcohol, propargyl amine, propargyl alcohol, and more, can be applied in this context to generate linkers of the appropriate length.

[0323] According to one embodiment, the linking compound is selected from a phosphodiester, a phosphorothioate, a carbamate, a methylphosphonate, a guanidinium, a sulfamate, a sulfamide, a formacetal, a thioformacetal, a sulfone, an amide and mixtures thereof.

[0324] According to one embodiment, the linker comprises a C3 spacer, a C6 linker or a triethylene glycol linker.

[0325] According to specific embodiments, the linker comprises a C3 spacer (CH2-CH2-CH2).

[0326] The linker of some embodiments of the invention may comprise one or multiple C3 spacers. Thus, according to specific embodiments, the linker comprises at least one, at least 2, at least 3, at least 4 or at least 5 C3 spacers.

[0327] According to specific embodiments, the linker comprises five units of a C3 carbon chain linker [(CH2-CH2-CH2)x5].

[0328] A non-limiting example of a composition comprising a GpG ODN attached via a linker to an siRNA downregulating agent is described in details in the Examples section which follows.

[0329] Thus, according to specific embodiments, the composition comprises SEQ ID NOs: 23, 24 and 33 and a C3 spacer linking SEQ ID NO: 23 to SEQ ID NO: 33. Such a composition is also shown in Figures 13A-B. According to specific embodiments, the composition comprises SEQ ID NOs: 23, 24 and 35 and a C3 spacer linking SEQ ID NO: 23 to SEQ ID NO: 35.

[0330] According to specific embodiments, the composition further comprises a cell penetrating moiety. Thus, according to specific embodiments, the downregulating agent may be attached to or encapsulated in a cell penetrating moiety.

[0331] As used herein the phrase "cell penetrating moiety" refers to a moiety which enhances translocation of the downregulating agent or composition comprising same across a cell membrane. Non-limiting examples of cell penetrating moieties include cell penetrating peptides, lipidic moieties and lipid particles.

[0332] As used herein, a "cell-penetrating peptide" is a peptide that comprises a short (about 12- 30 residues) amino acid sequence or functional motif that confers the energy-independent (i.e., non-endocytotic) translocation properties associated with transport of the membrane -permeable complex across the plasma and / or nuclear membranes of a cell. The cell-penetrating peptide used in the membrane-permeable complex of some embodiments of the invention comprises at least one non-functional cysteine residue, which is either free or derivatized to form a disulfide link with a double- stranded ribonucleic acid that has been modified for such linkage. Representative amino acid motifs conferring such properties are listed in U.S. Pat. No. 6,348,185, the contents of which are expressly incorporated herein by reference. The cell-penetrating peptides of some embodiments of the invention may include, but are not limited to, penetratin, transportan, plsl, TAT(48-60), pVEC, MTS, and MAP.

[0333] Non-limiting examples of lipidic moieties (i.e. naturally occurring or synthetically produced lipids) which may be used in accordance with specific embodiments the present invention include, fatty acids; fats; oils; waxes; cholesterol; cholic acid; a thioether; sterols; fat- soluble vitamins, such as vitamins A, D, E and K; monoglycerides; diglycerides, phospholipids, an aliphatic chain, a polyamine or a polyethylene glycol chain, or adamantane acetic acid, a palmityl moiety, or an octadecylamine or hexylamino-carbonyloxycholesterol moiety.

[0334] Additional description on lipid particles is provided hereinbelow.

[0335] Thus, according to specific embodiments, the downregulating agent is attached to the macrophage-targeting agent via a particle.

[0336] According to additional or alternative embodiments, the downregulating agent is encapsulated in a particle comprising the macrophage-targeting moiety.

[0337] As used herein, "particle" refers to a nano to micro structures which are not biological cells. The particle may be a synthetic carrier, gel or other object or material having an external surface which is capable of being loadable with (e.g., encapsulated in or attached to) the downregulating agent and the macrophage-targeting moiety. The particle may be either polymeric or non-polymeric preparations.

[0338] Exemplary particles that may be used according to specific embodiments of the invention include, but are not limited to, polymeric particles, microcapsules, liposomes, microspheres, microemulsions, nanoparticles, nanocapsules, nano-spheres, nano-liposomes, nano-emulsions, lipid nanoparticles and nanotubes.

[0339] According to specific embodiments, the particle if a lipid particle.

[0340] Such particles are well known to the skilled artisan and are also described in e.g., International Patent Application Publication Nos. WO2021 / 234699, the contents of which are fully incorporated herein by reference.

[0341] Suitable particles in accordance with some embodiments of the invention are preferably non-toxic.

[0342] The particles may have a charged surface (i.e., positively charged or negatively charged) or a neutral surface. Thus, agents which are used to fabricate the particles may be selected according to the desired charge required on the outer surface of the particles.

[0343] It will be appreciated that combinations of different lipids may be used to fabricate the particles of the present invention, including a mixture of more than one cationic lipid, a mixture of more than one anionic lipid, a mixture of more than one neutral lipid, a mixture of at least one cationic lipid and at least one anionic lipid, a mixture of at least one cationic lipid and at least one neutral lipid, a mixture of at least one anionic lipid and at least one neutral lipid and additional combinations of the above. In addition, polymer-lipid based formulations may be used.

[0344] According to some exemplary embodiments, the plurality of lipids of the lipid particles may be selected from, but not limited to: cationic lipids, phosphatidylethanolamines, ionizable lipids, membrane stabilizing lipids, phospholipids, and the like, or combinations thereof, each possibility represents a separate embodiment of the present invention.

[0345] In some embodiments, the membrane stabilizing lipids may be selected from, but not limited to: cholesterol, phospholipids (such as, for example, phosphatidylcholine, phosphatidylethanolamine, phosphatidylinositol, phosphatidylserine, phosphatidylglycerol, diphosphatidylglycerols), cephalins, sphingolipids (sphingomyelins and glycosphingolipids), glycoglycerolipids, and combinations thereof, each possibility represents a separate embodiment of the present invention. There are numerous polymers which may be attached to lipids. Polymers typically used as lipid modifiers include, without being limited thereto: polyethylene glycol (PEG), polysialic acid, polylactic (also termed polylactide), polyglycolic acid (also termed polyglycolide), apolylactie- polyglycolic acid' polyvinyl alcohol, polyvinylpyrrolidone, polymethoxazoline, polyethyloxazoline, polyllydroxyetlyloxazolille, solyhydroxypryloxazoline, polyaspartarllide, polyhydroxypropyl methacrylamide, polymethacrylamide, polydimethylacrylamide, polyvinylmethylether, polyhydroxyethyl acrylate, derivatized celluloses such as hydroxy methylcellulose or hydroxyethylcellulose.

[0346] The polymers may be employed as homopolymers or as block or random copolymers.

[0347] The particles may also include other components. Examples of such other components includes, without being limited thereto, fatty alcohols, fatty acids, and / or cholesterol esters or any other pharmaceutically acceptable excipients which may affect the surface charge, the membrane fluidity and assist in the incorporation of the biologically active lipid into the lipid assembly. Examples of sterols include cholesterol, cholesterol hemisuccinate, cholesterol sulfate, or any other derivatives of cholesterol. Preferred lipid assemblies according the invention include either those which form a micelle (typically when the assembly is absent from a lipid matrix) or those which form a liposome (typically, when a lipid matrix is present).

[0348] According to specific embodiments, the particle is a mannosylated particle.

[0349] In one embodiment, the particle is a lipid-based nanoparticle.

[0350] As used herein, the term "nanoparticle" refers to a particle or particles having an intermediate size between individual atoms and macroscopic bulk solids. Generally, nanoparticle has a characteristic size (e.g., diameter for generally spherical nanoparticles, or length for generally elongated nanoparticles) in the sub-micrometer range, e.g., from about 1 nm to about 500 nm, or from about 1 nm to about 200 nm, or of the order of 10 nm, e.g., from about 1 nm to about 100 nm.

[0351] The nanoparticles may be of any shape, including, without limitation, elongated particle shapes, such as nanowires, or irregular shapes, in addition to more regular shapes, such as generally spherical, hexagonal and cubic nanoparticles. According to one embodiment, the nanoparticles are generally spherical.

[0352] The core of the particle may be hydrophilic or hydrophobic. The core of the lipid-based nanoparticle may comprise some lipids, such that it is not fully hydrophilic.

[0353] Non-limiting examples of lipid nanoparticles that can be used with specific embodiments of the present invention and methods of producing same are described in e.g. Ramishetti et al. Adv Mater. 2020 Jan 30:el906128, International Patent Application Publication Nos. WO2016 / 189532, W02018 / 015881 and WO2018087753, WO2017194454 and US Patent

[0354] Application Publication no. US20130245107, the contents of which are fully incorporated herein by reference.

[0355] According to specific embodiments, the lipid-based nanoparticle comprises polyethylene glycol (PEG) and distearylphosphatidylethanolamine (DSPE).

[0356] According to specific embodiments, the lipid-based nanoparticle comprises polyethylene glycol (PEG), distearylphosphatidylethanolamine (DSPE) and a mannose group.

[0357] The lipid nanoparticle may be prepared by any of the methods known in the art, such as disclosed in e.g., Jayaraman et al. Angew chem. Jul 2012, Semple et al. Nat Biotech. 2010, Kauffman et al. Nano Lett, Oct 2015.

[0358] Lipid nanoparticles of some embodiments of the invention are also commercially available from e.g., GenScript

[0359] In a specific embodiment, the particle is a liposome. As used herein and as recognized in the art, liposomes include any synthetic (i.e., not naturally occurring) structure composed of lipid bilayers, which enclose a volume. Liposomes include emulsions, foams, micelles, insoluble monolayers, liquid crystals, phospholipid dispersions, lamellar layers and the like. The liposomes may be prepared by any of the known methods in the art [Monkkonen, J. et al., 1994, J. Drug Target, 2:299-308; Monkkonen, J. et al., 1993, Calcif. Tissue Int., 53: 139-145; Lasic D., Liposomes Technology Inc., Elsevier, 1993, 63-105. (chapter 3); Winterhalter M, Lasic D, Chem Phys Lipids, 1993 September;64(l-3):35-43].

[0360] The diameter of the liposomes used preferably ranges from 20-200 nm and more preferably from 20-100 nm. For sizing liposomes, extrusion, homogenization or exposure to ultrasound irradiation may be used. Homogenizers which may be conveniently used include microfluidizers (produced by Microfluidics of Boston, MA, USA) or microfluidic micro mixer (Precision NanoSystems, Vancouver, BC, Canada). In a typical homogenization procedure, liposomes are recirculated through a standard emulsion homogenizer until selected liposomes sizes are observed. The particle size distribution can be monitored by conventional laser beam particle size discrimination. Extrusion of liposomes through a small-pore polycarbonate membrane or an asymmetric ceramic membrane is an effective method for reducing liposome sizes to a relatively well-defined size distribution. Typically, the suspension is cycled through the membrane one or more times until the desired liposome size distribution is achieved. The liposomes may be extruded through successively smaller pore membranes to achieve a gradual reduction in liposome size. The liposomes may be unilamellar or may be multilamellar. Unilamellar liposomes may be preferred in some instances as they represent a larger surface area per lipid mass. The liposomes may be fabricated from a single phospholipid or mixtures of phospholipids. The liposomes may also comprise other lipid materials such as cholesterol. For fabricating liposomes with a negative electrical surface potential, acidic phospho- or sphingo- or other synthetic-lipids may be used. Preferably, the lipids have a high partition coefficient into lipid bilayers and a low desorption rate from the lipid assembly. Exemplary phospholipids that may be used for fabricating liposomes with a negative electrical surface potential include, but are not limited to phosphatidylserine, phosphatidic acid, phosphatidylcholine and phosphatidyl glycerol. Other negatively charged lipids which are not liposome forming lipids that may be used are sphingolipids such as cerebroside sulfate, and various gangliosides.

[0361] Other negatively charged lipids which are not liposome forming lipids that may be used are sphingolipids such as cerebroside sulfate, and various gangliosides.

[0362] The most commonly used and commercially available lipids derivatized into lipopolymers are those based on phosphatidyl ethanolamine (PE), usually distearylphosphatidylethanolamine (DSPE).

[0363] The lipid phase of the liposome may comprise a physiologically acceptable liposome forming lipid or a combination of physiologically acceptable liposome forming lipids for medical or veterinarian applications. Liposomes are spherical bilayer structures composed of different kind of amphiphatic lipids. These lipids have glycerol backbone with two hydrophobic acyl chains and a hydrophilic head group such as phosphate in combination or derivatives of the same may contain amine, imine, acid or alcohols. Typically, the acyl chain is between 12 to about 24 carbon atoms in length, and has varying degrees of saturation being fully, partially or non- hydrogenated lipids. Further, the lipid matrix may be of natural source, semi-synthetic or fully synthetic lipid, and neutral, negatively or positively charged.

[0364] Incorporating the downregulating agent and / or macrophage-targeting moiety to the particle can be effected concomitant with, or following particle assembly, by methods well known in the art such as disclosed in e.g. Ramishetti et al. Adv Mater. 2020 Jan 30:el906128, International Patent Application Publication Nos. W02018 / 015881 and WO2018087753, WO2017194454 and US Patent Application Publication no. US20130245107, the contents of which are fully incorporated herein by reference.

[0365] The desired amount of the downregulating agent and / or macrophage-targeting moiety attached to- or encapsulated in the particle varies depending on the type of the downregulating agent or macrophage-targeting moiety. However, it is preferable that these moieties be attached to- or encapsulated in the particle at a high loading efficiency.

[0366] Any of the moieties comprised in the composition may comprise a detectable (also referred to as “labeling”) moiety.

[0367] Examples of detectable moieties that can be used in the present invention include but are not limited to radioactive isotopes, phosphorescent chemicals, chemiluminescent chemicals, fluorescent chemicals, enzymes, fluorescent polypeptides and a radioactive isotope (such as

[0125] iodine). The detectable moiety can be a member of a binding pair, which is identifiable via its interaction with an additional member of the binding pair, and a label which is directly visualized. In one example, the label is a fluorescent protein or an enzyme producing a colorimetric reaction.

[0368] Examples of suitable fluorophores include, but are not limited to, phycoerythrin (PE), fluorescein isothiocyanate (FITC), Cy-chrome, rhodamine, green fluorescent protein (GFP), blue fluorescent protein (BFP), Texas red, PE-Cy5, and the like. For additional guidance regarding fluorophore selection, methods of linking fluorophores to various types of molecules see Richard P. Haugland, “Molecular Probes: Handbook of Fluorescent Probes and Research Chemicals 1992-1994”, 5th ed., Molecular Probes, Inc. (1994); U.S. Pat. No. 6,037,137 to Oncoimmunin Inc.; Hermanson, “Bioconjugate Techniques”, Academic Press New York, N.Y. (1995); Kay M. et al., 1995. Biochemistry 34:293; Stubbs et al., 1996. Biochemistry 35:937; Gakamsky D. et al., “Evaluating Receptor Stoichiometry by Fluorescence Resonance Energy Transfer,” in “Receptors: A Practical Approach,” 2nd ed., Stanford C. and Horton R. (eds.), Oxford University Press, UK. (2001); U.S. Pat. No. 6,350,466 to Targesome, Inc.]. Fluorescence detection methods which can be used to detect the fluorescent detectable moiety include, for example, fluorescence activated flow cytometry (FACS), immunofluorescence confocal microscopy, fluorescence in- situ hybridization (FISH) and fluorescence resonance energy transfer (FRET).

[0369] Further examples of detectable moieties, include those detectable by Positron Emission Tomagraphy (PET) and Magnetic Resonance Imaging (MRI), all of which are well known to those of skill in the art.

[0370] Methods of producing a molecule by attachment of any of the moieties comprised in the composition described herein are well known in the art and depend on the nature of the moieties, and include for example covalent, non-covalent, hydrogen, chemical and / or electrostatic bonds. The synthesis of the conjugate may involve the selective protection and deprotection of functional groups. Suitable protecting groups are well known for the skilled person in the art. For example, a general review of protecting groups in organic chemistry is provided by Wuts, P.G.M. and Greene T.W. in Protecting Groups in Organic Synthesis (4thEd. Wiley-Interscience), and by Kocienski P.J. in Protecting Groups (3rdEd. Georg Thieme Verlag).

[0371] For example, attachment of two polypeptides (e.g. covalent attachment e.g. via a peptide bond) can be effected by chemical cross-linking the two polypeptides by e.g. sulfo GMBS, glutaraldehyde (see e.g., G.T. Hermanson 1996, "Antibody Modification and Conjugation, in Bioconjugate Techniques, Academic Press, San Diego), 3-(2-pyridyldithio)propionic acid N- hydroxy succinimide ester (also called N-succinimidyl 3-(2pyridyldithio) propionate) ("SDPD") (Sigma, Cat. No. P-3415; see e.g., Cumber et al. 1985, Methods of Enzymology 112: 207-224), disulphide bond formation, click chemistry; by enzymatic methods such as sortase-mediated ligation, inteins; or by recombinant technology (i.e., translationally fusing the downregulating agent and the targeting moiety). Attachment of two polynucleotides can be effected for example by enzymatic methods, chemical or physical methods such as DNA ligase, use of restriction enzymes and DNA ligase, topoisomerase; DC (l-Ethyl-3-(3-dimethylaminopropyl)carbodiimide) mediated ligation [see e.g., J. March, Advanced Organic Chemistry: Reaction's, Mechanism, and Structure, pp. 349-50 & 372-74 (3d ed.), 1985; B. Neises et al. 1978, Angew Chem., Int. Ed. Engl. 17:522; A. Hassner et al. 1978, Tetrahedron Lett. 4475; E.P. Boden et al. 1986, J. Org. Chem. 50:2394 and L.J. Mathias 1979, Synthesis 561], Gibson assembly etc. Attachment of a polynucleotide to a polypeptide can be effected for example by cross-linking agents such as EDC (l-Ethyl-3-(3-dimethylaminopropyl)carbodiimide) or SMCC (Succinimidyl 4-(N- maleimidomethyl)cyclohexane-l -carboxylate), click chemistry, maleimide chemistry, enzymatic methods such as sortase-mediated ligation, T4 DNA ligase and fusion proteins, biotin- streptavidin linkage, UV light etc. Attachment of a small molecule to a polynucleotide or polypeptide can be effect for example by click chemistry, amide bond formation, thiol-ene reaction, biotin-streptavidin linkage, or genetic engineering (to allows incorporation of functional groups suitable for attaching small molecule during synthesis). In addition, methods of incorporating a polynucleotide or polypeptide to a particulated delivery vehicle, e.g., a liposome, or a nano- or microparticle, may be effects by any of the methods known in the art [e.g., Liposome Technology, Vol. II, Incorporation of Drugs, Proteins, and Genetic Material, CRC Press; Monkkonen, J. et al., 1994, J. Drug Target, 2:299-308; Monkkonen, J. et al., 1993, Calcif. Tissue Int., 53: 139-145; Lasic D D., Liposomes Technology Inc., Elsevier, 1993, 63-105. (Chapter 3); Winterhalter M, Lasic D D, Chem Phys Lipids, 1993 September;64(l-3):35-43; Ramishetti et al. Adv Mater. 2020 Jan 30:el906128, International Patent Application Publication Nos. WG2018 / 015881 and WO2018087753, WO2017194454 and US Patent Application Publication no. US20130245107, International Patent Application Publication NO. W02018 / 015881, US Patent Nos. 5,171,578, 5,204,096 and 5,258,499, the contents of which are fully incorporated herein by reference].

[0372] Thus, for Example, for synthesizing a composition comprising a CpG ODN attached via a five C3 spacer to a Zeb2 siRNA, the following steps may be performed:

[0373] (i) Synthesis of phosphorothioate polynucleotides: During synthesis of the CpG ODN-linker-siRNA passenger strand and of the siRNA guide strand, phosphorothioate linkages are introduced by using sulfur-containing reagents instead of the usual oxygen-containing ones. Thus, for example, as described in the Examples section which follows, the CpG-linker- siRNA passenger strand and the siRNA guide strand are synthesized seperately via a nucleoside phosphoramidite chemistry comprising a 4 steps synthesis cycle: (1) Deprotection of the 5’- hydroxyl group (Detritylation); (2) Coupling of nucleotide phosphoramidite to the 5’-hydroxyl group; (3) Capping of unreacted 5’-hydroxyl groups; and (4) Oxidation. Step (4) can be substituted with a sulfurization step for the synthesis of phosphorothioated oligonucleotides. These four steps are repeated in the above order until all nucleoside components are added. The five C3 spacer linker within the CpG-siRNA passenger strand is designed to provide an appropriate distance between the CpG ODN and the siRNA.

[0374] (ii) Following the complete synthesis and purification of the CpG-C3 linker-siRNA passenger strand and the siRNA guide strand, the two components are annealed to produce the CpG ODN-linker-siRNA complex.

[0375] (iii) The resulting CpG ODN-linker-siRNA complex is purified to remove any unreacted components by e.g., gel electrophoresis, HPLC, or affinity chromatography.

[0376] As shown in the Examples section which follows, the present inventors demonstrate that selective in-vivo targeting of Zeb2 in macrophages reprogrammed tumor associated macrophages (TAMs) and mobilized systematic anti-tumoral T cell responses, achieving complete tumor clearance as a monotherapy (Example 6 of the Examples section which follows). Hence, the present disclosure further suggests that the composition disclosed herein may be used for, but not limited to, treating an inflammatory disease (e.g., cancer).

[0377] Thus, according to an aspect of the present invention, there is provided a method of treating an inflammatory disease in a subject in need thereof, the method comprising administering to the subject a therapeutically effective amount of the composition, thereby treating the inflammatory disease in the subject.

[0378] According to an additional or an alternative aspect of the present invention, there is provided the composition for use in treating an inflammatory disease in a subject in need thereof. The term “treating” or “treatment” refers to inhibiting, preventing or arresting the development of a pathology (disease, disorder or medical condition) and / or causing the reduction, remission, or regression of a pathology or a symptom of a pathology. Those of skill in the art will understand that various methodologies and assays can be used to assess the development of a pathology, and similarly, various methodologies and assays may be used to assess the reduction, remission or regression of a pathology.

[0379] As used herein, the term “subject” includes mammals, e.g., human beings at any age and of any gender. According to specific embodiments, the term “subject” refers to a subject who suffers from the pathology (disease, disorder or medical condition, e.g., inflammatory disease e.g., cancer). Thus, according to specific embodiments, the subject is diagnosed with the pathology. According to specific embodiments, this term encompasses individuals who are at risk to develop the pathology.

[0380] According to specific embodiments, the subject is a human.

[0381] According to specific embodiments, the subject is characterized by a level of macrophages above a predetermined threshold in a biological sample obtained from the subject.

[0382] According to specific embodiments, the subject is characterized by a level of Zeb2 above a predetermined threshold in a biological sample obtained from the subject.

[0383] According to specific embodiments, the subject is characterized by a level of macrophages expressing Zeb2 above a predetermined threshold in a biological sample obtained from the subject.

[0384] Thus, according to specific embodiments, the methods and uses disclosed herein further comprise determining a level of macrophages in a biological sample of the subject.

[0385] According to specific embodiments, the methods and uses disclosed herein further comprise determining a level of Zeb2 in a biological sample of the subject.

[0386] Additional description on this characterization e.g., the determining procedure, the biological sample and the predetermined threshold is further provided in details hereinbelow.

[0387] Inflammatory diseases that can be treated according to some embodiments of the invention include, but are not limited to, chronic inflammatory diseases and acute inflammatory diseases.

[0388] Inflammatory diseases associated with hypersensitivity

[0389] Examples of hypersensitivity include, but are not limited to, Type I hypersensitivity, Type II hypersensitivity, Type III hypersensitivity, Type IV hypersensitivity, immediate hypersensitivity, antibody mediated hypersensitivity, immune complex mediated hypersensitivity, T lymphocyte mediated hypersensitivity and DTH. Type I or immediate hypersensitivity, such as asthma.

[0390] Type II hypersensitivity include, but are not limited to, rheumatoid diseases, rheumatoid autoimmune diseases, rheumatoid arthritis (Krenn V. et al., Histol Histopathol 2000 Jul; 15 (3):791), spondylitis, ankylosing spondylitis (Jan Voswinkel et al., Arthritis Res 2001; 3 (3): 189), systemic diseases, systemic autoimmune diseases, systemic lupus erythematosus (Erikson J. et al., Immunol Res 1998; 17 (l-2):49), sclerosis, systemic sclerosis (Renaudineau Y. et al., Clin Diagn Lab Immunol. 1999 Mar;6 (2): 156); Chan OT. et al., Immunol Rev 1999 Jun;169: 107), glandular diseases, glandular autoimmune diseases, pancreatic autoimmune diseases, diabetes, Type I diabetes (Zimmet P. Diabetes Res Clin Pract 1996 Oct;34 Suppl:S125), thyroid diseases, autoimmune thyroid diseases, Graves’ disease (Orgiazzi J. Endocrinol Metab Clin North Am 2000 Jun;29 (2):339), thyroiditis, spontaneous autoimmune thyroiditis (Braley-Mullen H. and Yu S, J Immunol 2000 Dec 15;165 (12):7262), Hashimoto’s thyroiditis (Toyoda N. et al., Nippon Rinsho 1999 Aug;57 (8): 1810), myxedema, idiopathic myxedema (Mitsuma T. Nippon Rinsho. 1999 Aug;57 (8): 1759); autoimmune reproductive diseases, ovarian diseases, ovarian autoimmunity (Garza KM. et al., J Reprod Immunol 1998 Feb;37 (2):87), autoimmune anti- sperm infertility (Diekman AB. et al., Am J Reprod Immunol. 2000 Mar;43 (3): 134), repeated fetal loss (Tincani A. et al., Lupus 1998;7 Suppl 2:S 107-9), neurodegenerative diseases, neurological diseases, neurological autoimmune diseases, multiple sclerosis (Cross AH. et al., J Neuroimmunol 2001 Jan 1;112 (1-2): 1), Alzheimer’s disease (Oron L. et al., J Neural Transm Suppl. 1997;49:77), myasthenia gravis (Infante AJ. And Kraig E, Int Rev Immunol 1999;18 (1- 2):83), motor neuropathies (Kornberg AJ. J Clin Neurosci. 2000 May;7 (3): 191), Guillain-Barre syndrome, neuropathies and autoimmune neuropathies (Kusunoki S. Am J Med Sci. 2000 Apr;319 (4):234), myasthenic diseases, Lambert-Eaton myasthenic syndrome (Takamori M. Am J Med Sci. 2000 Apr;319 (4):204), paraneoplastic neurological diseases, cerebellar atrophy, paraneoplastic cerebellar atrophy, non-paraneoplastic stiff man syndrome, cerebellar atrophies, progressive cerebellar atrophies, encephalitis, Rasmussen’s encephalitis, amyotrophic lateral sclerosis, Sydeham chorea, Gilles de la Tourette syndrome, polyendocrinopathies, autoimmune polyendocrinopathies (Antoine JC. and Honnorat J. Rev Neurol (Paris) 2000 Jan; 156 (1):23); neuropathies, dysimmune neuropathies (Nobile- Orazio E. et al., Electroencephalogr Clin Neurophysiol Suppl 1999;50:419); neuromyotonia, acquired neuromyotonia, arthrogryposis multiplex congenita (Vincent A. et al., Ann N Y Acad Sci. 1998 May 13;841:482), cardiovascular diseases, cardiovascular autoimmune diseases, atherosclerosis (Matsuura E. et al., Lupus. 1998;7 Suppl 2:S135), myocardial infarction (Vaarala O. Lupus. 1998;7 Suppl 2:S132), thrombosis (Tincani A. et al., Lupus 1998;7 Suppl 2:S 107-9), granulomatosis, Wegener’s granulomatosis, arteritis, Takayasu’s arteritis and Kawasaki syndrome (Praprotnik S. et al., Wien Klin Wochenschr 2000 Aug 25; 112 (15-16):660); anti-factor VIII autoimmune disease (Lacroix- Desmazes S. et al., Semin Thromb Hemost.2000;26 (2): 157); vasculitises, necrotizing small vessel vasculitises, microscopic polyangiitis, Churg and Strauss syndrome, glomerulonephritis, pauci-immune focal necrotizing glomerulonephritis, crescentic glomerulonephritis (Noel LH. Ann Med Interne (Paris). 2000 May; 151 (3): 178); antiphospholipid syndrome (Flamholz R. et al., J Clin Apheresis 1999; 14 (4): 171); heart failure, agonist-like P-adrenoceptor antibodies in heart failure (Wallukat G. et al., Am J Cardiol. 1999 Jun 17;83 (12A):75H), thrombocytopenic purpura (Moccia F. Ann Ital Med Int. 1999 Apr-Jun; 14 (2): 114); hemolytic anemia, autoimmune hemolytic anemia (Efremov DG. et al., Leuk Lymphoma 1998 Jan;28 (3-4):285), gastrointestinal diseases, autoimmune diseases of the gastrointestinal tract, intestinal diseases, chronic inflammatory intestinal disease (Garcia Herola A. et al., Gastroenterol Hepatol. 2000 Jan;23

[0391] (1): 16), celiac disease (Landau YE. and Shoenfeld Y. Harefuah 2000 Jan 16;138 (2): 122), autoimmune diseases of the musculature, myositis, autoimmune myositis, Sjogren’s syndrome (Feist E. et al., Int Arch Allergy Immunol 2000 Sep;123 (1):92); smooth muscle autoimmune disease (Zauli D. et al., Biomed Pharmacother 1999 Jun;53 (5-6):234), hepatic diseases, hepatic autoimmune diseases, autoimmune hepatitis (Manns MP. J Hepatol 2000 Aug;33 (2):326) and primary biliary cirrhosis (Strassburg CP. et al., Eur J Gastroenterol Hepatol. 1999 Jun; 11 (6):595).

[0392] Type IV or T cell mediated hypersensitivity, include, but are not limited to, rheumatoid diseases, rheumatoid arthritis (Tisch R, McDevitt HO. Proc Natl Acad Sci U S A 1994 Jan 18;91

[0393] (2):437), systemic diseases, systemic autoimmune diseases, systemic lupus erythematosus (Datta SK., Lupus 1998;7 (9):591), glandular diseases, glandular autoimmune diseases, pancreatic diseases, pancreatic autoimmune diseases, Type 1 diabetes (Castano L. and Eisenbarth GS. Ann. Rev. Immunol. 8:647); thyroid diseases, autoimmune thyroid diseases, Graves’ disease (Sakata S. et al., Mol Cell Endocrinol 1993 Mar;92 (1):77); ovarian diseases (Garza KM. et al., J Reprod Immunol 1998 Feb;37 (2):87), prostatitis, autoimmune prostatitis (Alexander RB. et al., Urology 1997 Dec;50 (6):893), polyglandular syndrome, autoimmune polyglandular syndrome, Type I autoimmune polyglandular syndrome (Hara T. et al., Blood. 1991 Mar 1;77 (5): 1127), neurological diseases, autoimmune neurological diseases, multiple sclerosis, neuritis, optic neuritis (Soderstrom M. et al., J Neurol Neurosurg Psychiatry 1994 May;57 (5):544), myasthenia gravis (Oshima M. et al., Eur J Immunol 1990 Dec;20 ( 12):2563), stiff-man syndrome (Hiemstra HS. et al., Proc Natl Acad Sci U S A 2001 Mar 27;98 (7):3988), cardiovascular diseases, cardiac autoimmunity in Chagas’ disease (Cunha-Neto E. et al., J Clin Invest 1996 Oct 15;98 (8): 1709), autoimmune thrombocytopenic purpura (Semple JW. et al., Blood 1996 May 15;87 (10):4245), anti-helper T lymphocyte autoimmunity (Caporossi AP. et al., Viral Immunol 1998;11 (1):9), hemolytic anemia (Sallah S. et al., Ann Hematol 1997 Mar;74 (3): 139), hepatic diseases, hepatic autoimmune diseases, hepatitis, chronic active hepatitis (Franco A. et al., Clin Immunol Immunopathol 1990 Mar;54 (3):382), biliary cirrhosis, primary biliary cirrhosis (Jones DE. Clin Sci (Colch) 1996 Nov;91 (5):551), nephric diseases, nephric autoimmune diseases, nephritis, interstitial nephritis (Kelly CJ. J Am Soc Nephrol 1990 Aug;l (2): 140), connective tissue diseases, ear diseases, autoimmune connective tissue diseases, autoimmune ear disease (Yoo TJ. et al., Cell Immunol 1994 Aug;157 (1):249), disease of the inner ear (Gloddek B. et al., Ann N Y Acad Sci 1997 Dec 29;830:266), skin diseases, cutaneous diseases, dermal diseases, bullous skin diseases, pemphigus vulgaris, bullous pemphigoid and pemphigus foliaceus.

[0394] Examples of delayed type hypersensitivity include, but are not limited to, contact dermatitis and drug eruption.

[0395] Examples of types of T lymphocyte mediating hypersensitivity include, but are not limited to, helper T lymphocytes and cytotoxic T lymphocytes.

[0396] Examples of helper T lymphocyte-mediated hypersensitivity include, but are not limited to, Thl lymphocyte mediated hypersensitivity and Th2 lymphocyte mediated hypersensitivity.

[0397] Autoimmune diseases

[0398] Include, but are not limited to, cardiovascular diseases, rheumatoid diseases, glandular diseases, gastrointestinal diseases, cutaneous diseases, hepatic diseases, neurological diseases, muscular diseases, nephric diseases, diseases related to reproduction, connective tissue diseases and systemic diseases.

[0399] Examples of autoimmune cardiovascular diseases include, but are not limited to atherosclerosis (Matsuura E. et al., Lupus. 1998;7 Suppl 2:S135), myocardial infarction (Vaarala O. Lupus. 1998;7 Suppl 2:S132), thrombosis (Tincani A. et al., Lupus 1998;7 Suppl 2:S 107-9), Wegener’s granulomatosis, Takayasu’s arteritis, Kawasaki syndrome (Praprotnik S. et al., Wien Klin Wochenschr 2000 Aug 25; 112 (15-16):660), anti-factor VIII autoimmune disease (Lacroix- Desmazes S. et al., Semin Thromb Hemost.2000;26 (2): 157), necrotizing small vessel vasculitis, microscopic polyangiitis, Churg and Strauss syndrome, pauci-immune focal necrotizing and crescentic glomerulonephritis (Noel LH. Ann Med Interne (Paris). 2000 May; 151 (3): 178), antiphospholipid syndrome (Flamholz R. et al., J Clin Apheresis 1999; 14 (4): 171), antibody- induced heart failure (Wallukat G. et al., Am J Cardiol. 1999 Jun 17;83 (12A):75H), thrombocytopenic purpura (Moccia F. Ann Ital Med Int. 1999 Apr-Jun;14 (2): 114; Semple JW. et al., Blood 1996 May 15;87 (10):4245), autoimmune hemolytic anemia (Efremov DG. et al., Leuk Lymphoma 1998 Jan;28 (3-4):285; Sallah S. et al., Ann Hematol 1997 Mar;74 (3): 139), cardiac autoimmunity in Chagas’ disease (Cunha-Neto E. et al., J Clin Invest 1996 Oct 15;98 (8): 1709) and anti-helper T lymphocyte autoimmunity (Caporossi AP. et al., Viral Immunol 1998;11 (1):9).

[0400] Examples of autoimmune rheumatoid diseases include, but are not limited to rheumatoid arthritis (Krenn V. et al., Histol Histopathol 2000 Jul;15 (3):791; Tisch R, McDevitt HO. Proc Natl Acad Sci units S A 1994 Jan 18;91 (2):437) and ankylosing spondylitis (Jan Voswinkel et al., Arthritis Res 2001; 3 (3): 189).

[0401] Examples of autoimmune glandular diseases include, but are not limited to, pancreatic disease, Type I diabetes, thyroid disease, Graves’ disease, thyroiditis, spontaneous autoimmune thyroiditis, Hashimoto’s thyroiditis, idiopathic myxedema, ovarian autoimmunity, autoimmune anti-sperm infertility, autoimmune prostatitis and Type I autoimmune polyglandular syndrome, diseases include, but are not limited to autoimmune diseases of the pancreas, Type 1 diabetes (Castano L. and Eisenbarth GS. Ann. Rev. Immunol. 8:647; Zimmet P. Diabetes Res Clin Pract 1996 Oct;34 Suppl:S125), autoimmune thyroid diseases, Graves’ disease (Orgiazzi J. Endocrinol Metab Clin North Am 2000 Jun;29 (2):339; Sakata S. et al., Mol Cell Endocrinol 1993 Mar;92 (1):77), spontaneous autoimmune thyroiditis (Braley-Mullen H. and Yu S, J Immunol 2000 Dec 15;165 (12):7262), Hashimoto’s thyroiditis (Toyoda N. et al., Nippon Rinsho 1999 Aug;57 (8): 1810), idiopathic myxedema (Mitsuma T. Nippon Rinsho. 1999 Aug;57 (8): 1759), ovarian autoimmunity (Garza KM. et al., J Reprod Immunol 1998 Feb;37 (2):87), autoimmune anti- sperm infertility (Diekman AB. et al., Am J Reprod Immunol. 2000 Mar;43 (3): 134), autoimmune prostatitis (Alexander RB. et al., Urology 1997 Dec;50 (6): 893) and Type I autoimmune polyglandular syndrome (Hara T. et al., Blood. 1991 Mar 1;77 (5): 1127).

[0402] Examples of autoimmune gastrointestinal diseases include, but are not limited to, chronic inflammatory intestinal diseases (Garcia Herola A. et al., Gastroenterol Hepatol. 2000 Jan;23 (1): 16), celiac disease (Landau YE. and Shoenfeld Y. Harefuah 2000 Jan 16;138 (2): 122), colitis, ileitis and Crohn’s disease.

[0403] Examples of autoimmune cutaneous diseases include, but are not limited to, autoimmune bullous skin diseases, such as, but are not limited to, pemphigus vulgaris, bullous pemphigoid and pemphigus foliaceus.

[0404] Examples of autoimmune hepatic diseases include, but are not limited to, hepatitis, autoimmune chronic active hepatitis (Franco A. et al., Clin Immunol Immunopathol 1990 Mar;54 (3):382), primary biliary cirrhosis (Jones DE. Clin Sci (Colch) 1996 Nov;91 (5):551; Strassburg CP. et al., Eur J Gastroenterol Hepatol. 1999 Jun; 11 (6):595) and autoimmune hepatitis (Manns MP. J Hepatol 2000 Aug;33 (2):326). Examples of autoimmune neurological diseases include, but are not limited to, multiple sclerosis (Cross AH. et al., J Neuroimmunol 2001 Jan 1 ; 112 (1-2): 1), Alzheimer’s disease (Oron L. et al., J Neural Transm Suppl. 1997;49:77), myasthenia gravis (Infante AJ. And Kraig E, Int Rev Immunol 1999;18 (l-2):83; Oshima M. et al., Eur J Immunol 1990 Dec;20 (12):2563), neuropathies, motor neuropathies (Kornberg AJ. J Clin Neurosci. 2000 May;7 (3): 191); Guillain- Barre syndrome and autoimmune neuropathies (Kusunoki S. Am J Med Sci. 2000 Apr;319 (4):234), myasthenia, Lambert-Eaton myasthenic syndrome (Takamori M. Am J Med Sci. 2000 Apr;319 (4):204); paraneoplastic neurological diseases, cerebellar atrophy, paraneoplastic cerebellar atrophy and stiff-man syndrome (Hiemstra HS. et al., Proc Natl Acad Sci units S A 2001 Mar 27;98 (7):3988); non-paraneoplastic stiff man syndrome, progressive cerebellar atrophies, encephalitis, Rasmussen’s encephalitis, amyotrophic lateral sclerosis, Sydeham chorea, Gilles de la Tourette syndrome and autoimmune polyendocrinopathies (Antoine JC. and Honnorat J. Rev Neurol (Paris) 2000 Jan;156 ( 1):23); dysimmune neuropathies (Nobile-Orazio E. et al., Electroencephalogr Clin Neurophysiol Suppl 1999;50:419); acquired neuromyotonia, arthrogryposis multiplex congenita (Vincent A. et al., Ann N Y Acad Sci. 1998 May 13;841 :482), neuritis, optic neuritis (Soderstrom M. et al., J Neurol Neurosurg Psychiatry 1994 May;57 (5):544) and neurodegenerative diseases.

[0405] Examples of autoimmune muscular diseases include, but are not limited to, myositis, autoimmune myositis and primary Sjogren’s syndrome (Feist E. et al., Int Arch Allergy Immunol 2000 Sep;123 (1):92) and smooth muscle autoimmune disease (Zauli D. et al., Biomed Pharmacother 1999 Jun;53 (5-6):234).

[0406] Examples of autoimmune nephric diseases include, but are not limited to, nephritis and autoimmune interstitial nephritis (Kelly CJ. J Am Soc Nephrol 1990 Aug;l (2): 140).

[0407] Examples of autoimmune diseases related to reproduction include, but are not limited to, repeated fetal loss (Tincani A. et al., Lupus 1998;7 Suppl 2:S 107-9).

[0408] Examples of autoimmune connective tissue diseases include, but are not limited to, ear diseases, autoimmune ear diseases (Yoo TJ. et al., Cell Immunol 1994 Aug;157 (1):249) and autoimmune diseases of the inner ear (Gloddek B. et al., Ann N Y Acad Sci 1997 Dec 29;830:266).

[0409] Examples of autoimmune systemic diseases include, but are not limited to, systemic lupus erythematosus (Erikson J. et al., Immunol Res 1998; 17 (l-2):49) and systemic sclerosis (Renaudineau Y. et al., Clin Diagn Lab Immunol. 1999 Mar;6 (2): 156); Chan OT. et al., Immunol Rev 1999 Jun; 169: 107). Infectious diseases

[0410] Examples of infectious diseases include, but are not limited to, chronic infectious diseases, subacute infectious diseases, acute infectious diseases, viral diseases, bacterial diseases, protozoan diseases, parasitic diseases, fungal diseases, mycoplasma diseases and prion diseases.

[0411] Graft rejection diseases

[0412] Examples of diseases associated with transplantation of a graft include, but are not limited to, graft rejection, chronic graft rejection, subacute graft rejection, hyperacute graft rejection, acute graft rejection and graft versus host disease.

[0413] Allergic diseases

[0414] Examples of allergic diseases include, but are not limited to, asthma, hives, urticaria, pollen allergy, dust mite allergy, venom allergy, cosmetics allergy, latex allergy, chemical allergy, drug allergy, insect bite allergy, animal dander allergy, stinging plant allergy, poison ivy allergy and food allergy.

[0415] Cancerous diseases

[0416] According to specific embodiments, the inflammatory disease is cancer.

[0417] Cancers which may be treated or prognosed by some embodiments of the invention can be any solid or non-solid tumor (including liquid cancer), cancer metastasis and / or a pre-cancer.

[0418] According to specific embodiments, the cancer is a malignant cancer.

[0419] Non-limiting examples of cancers include tumors of the gastrointestinal tract (colon carcinoma, rectal carcinoma, colorectal carcinoma, colorectal cancer, colorectal adenoma, hereditary nonpolyposis type 1, hereditary nonpolyposis type 2, hereditary nonpolyposis type 3, hereditary nonpolyposis type 6; colorectal cancer, hereditary nonpolyposis type 7, small and / or large bowel carcinoma, esophageal carcinoma, tylosis with esophageal cancer, stomach carcinoma, pancreatic carcinoma, pancreatic endocrine tumors), endometrial carcinoma, dermatofibrosarcoma protuberans, gallbladder carcinoma, Biliary tract tumors, prostate cancer, prostate adenocarcinoma, renal cancer (e.g., Wilms’ tumor type 2 or type 1), liver cancer (e.g., hepatoblastoma, hepatocellular carcinoma, hepatocellular cancer), bladder cancer, embryonal rhabdomyosarcoma, germ cell tumor, trophoblastic tumor, testicular germ cells tumor, immature teratoma of ovary, uterine, epithelial ovarian, sacrococcygeal tumor, choriocarcinoma, placental site trophoblastic tumor, epithelial adult tumor, ovarian carcinoma, serous ovarian cancer, ovarian sex cord tumors, cervical carcinoma, uterine cervix carcinoma, small-cell and non-small cell lung carcinoma, nasopharyngeal, breast carcinoma (e.g., ductal breast cancer, invasive intraductal breast cancer, sporadic ; breast cancer, susceptibility to breast cancer, type 4 breast cancer, breast cancer- 1, breast cancer-3; breast-ovarian cancer), squamous cell carcinoma (e.g., in head and neck), neurogenic tumor, astrocytoma, ganglioblastoma, neuroblastoma, lymphomas (e.g., Hodgkin's disease, non-Hodgkin's lymphoma, B cell, Burkitt, cutaneous T cell, histiocytic, lymphoblastic, T cell, thymic), gliomas, adenocarcinoma, adrenal tumor, hereditary adrenocortical carcinoma, brain malignancy (tumor), various other carcinomas (e.g., bronchogenic large cell, ductal, Ehrlich-Lettre ascites, epidermoid, large cell, Lewis lung, medullary, mucoepidermoid, oat cell, small cell, spindle cell, spinocellular, transitional cell, undifferentiated, carcinosarcoma, choriocarcinoma, cystadenocarcinoma), ependimoblastoma, epithelioma, erythroleukemia (e.g., Friend, lymphoblast), fibrosarcoma, giant cell tumor, glial tumor, glioblastoma (e.g., multiforme, astrocytoma), glioma hepatoma, heterohybridoma, heteromyeloma, histiocytoma, hybridoma (e.g., B cell), hypernephroma, insulinoma, islet tumor, keratoma, leiomyoblastoma, leiomyosarcoma, leukemia (e.g., acute lymphatic, acute lymphoblastic, acute lymphoblastic pre-B cell, acute lymphoblastic T cell leukemia, acute - megakaryoblastic, monocytic, acute myelogenous, acute myeloid, acute myeloid with eosinophilia, B cell, basophilic, chronic myeloid, chronic, B cell, eosinophilic, Friend, granulocytic or myelocytic, hairy cell, lymphocytic, megakaryoblastic, monocytic, monocytic- macrophage, myeloblastic, myeloid, myelomonocytic, plasma cell, pre-B cell, promyelocytic, subacute, T cell, lymphoid neoplasm, predisposition to myeloid malignancy, acute nonlymphocytic leukemia), lymphosarcoma, melanoma, mammary tumor, mastocytoma, medulloblastoma, mesothelioma, metastatic tumor, monocyte tumor, multiple myeloma, myelodysplastic syndrome, myeloma, nephroblastoma, nervous tissue glial tumor, nervous tissue neuronal tumor, neurinoma, neuroblastoma, oligodendroglioma, osteochondroma, osteomyeloma, osteosarcoma (e.g., Ewing's), papilloma, transitional cell, pheochromocytoma, pituitary tumor (invasive), plasmacytoma, retinoblastoma, rhabdomyosarcoma, sarcoma (e.g., Ewing's, histiocytic cell, Jensen, osteogenic, reticulum cell), schwannoma, subcutaneous tumor, teratocarcinoma (e.g., pluripotent), teratoma, testicular tumor, thymoma and trichoepithelioma, gastric cancer, fibrosarcoma, glioblastoma multiforme; multiple glomus tumors, Li-Fraumeni syndrome, liposarcoma, lynch cancer family syndrome II, male germ cell tumor, mast cell leukemia, medullary thyroid, multiple meningioma, endocrine neoplasia myxosarcoma, paraganglioma, familial nonchromaffin, pilomatricoma, papillary, familial and sporadic, rhabdoid predisposition syndrome, familial, rhabdoid tumors, soft tissue sarcoma, and Turcot syndrome with glioblastoma.

[0420] According to specific embodiments, the cancer is a pre-malignant cancer.

[0421] Precancers are well characterized and known in the art (refer, for example, to Berman JJ. and Henson DE., 2003. Classifying the precancers: a metadata approach. BMC Med Inform Decis Mak. 3:8). Classes of precancers amenable to treatment via the method of the invention include acquired small or microscopic precancers, acquired large lesions with nuclear atypia, precursor lesions occurring with inherited hyperplastic syndromes that progress to cancer, and acquired diffuse hyperplasias and diffuse metaplasias. Examples of small or microscopic precancers include HGSIL (High grade squamous intraepithelial lesion of uterine cervix), AIN (anal intraepithelial neoplasia), dysplasia of vocal cord, aberrant crypts (of colon), PIN (prostatic intraepithelial neoplasia). Examples of acquired large lesions with nuclear atypia include tubular adenoma, AILD (angioimmunoblastic lymphadenopathy with dysproteinemia), atypical meningioma, gastric polyp, large plaque parapsoriasis, myelodysplasia, papillary transitional cell carcinoma in-situ, refractory anemia with excess blasts, and Schneiderian papilloma. Examples of precursor lesions occurring with inherited hyperplastic syndromes that progress to cancer include atypical mole syndrome, C cell adenomatosis and MEA. Examples of acquired diffuse hyperplasias and diffuse metaplasias include AIDS, atypical lymphoid hyperplasia, Paget's disease of bone, post-transplant lymphoproliferative disease and ulcerative colitis.

[0422] According to specific embodiments, the cancer is a solid tumor.

[0423] According to specific embodiments, the cancer is selected from the group consisting of bladder cancer, lung cancer, stomach cancer, acute-myeloid leukemia and, colorectal cancer.

[0424] According to specific embodiments, the cancer is bladder cancer.

[0425] According to specific embodiments, the cancer is urothelial carcinoma.

[0426] According to specific embodiments, the cancer is colon cancer.

[0427] According to specific embodiments, the cancer is colon adenocarcinoma.

[0428] According to specific embodiments, the cancer is lung carcinoma.

[0429] According to specific embodiments, the cancer does not comprise myeloid cancerous cells. Non-limiting examples of such cancers include myeloid malignancy [e.g., acute myeloid leukemia (AML), chronic myeloid leukemia (CML), myelodysplastic syndrome (MDS)] and multiple myeloma.

[0430] According to specific embodiments, the cancer is not Histiocytic Sarcoma.

[0431] According to specific embodiments, the composition can be administered to a subject as a monotherapy.

[0432] According to other embodiments, the composition is administered to a subject in combination with other established or experimental therapeutic regimen to treat an inflammatory disease (e.g., anti-cancer agent) including, but not limited to analgesics, chemotherapeutic agents, radiotherapeutic agents, cytotoxic therapies (conditioning), hormonal therapy, antibodies, cell therapy and other treatment regimens (e.g., surgery) which are well known in the art. Non-limiting examples of anti-inflammatory drugs that can be administered in combination with the composition according to some embodiments of the invention include Alclofenac; Alclometasone Dipropionate; Algestone Acetonide; Alpha Amylase; Amcinafal; Amcinafide; Amfenac Sodium; Amiprilose Hydrochloride; Anakinra; Anirolac; Anitrazafen; Apazone; Balsalazide Disodium; Bendazac; Benoxaprofen; Benzydamine Hydrochloride; Bromelains; Broperamole; Budesonide; Carprofen; Cicloprofen; Cintazone; Cliprofen; Clobetasol Propionate; Clobetasone Butyrate; Clopirac; Cloticasone Propionate; Cormethasone Acetate; Cortodoxone; Deflazacort; Desonide; Desoximetasone; Dexamethasone Dipropionate; Diclofenac Potassium; Diclofenac Sodium; Diflorasone Diacetate; Diflumidone Sodium; Diflunisal; Difluprednate; Diftalone; Dimethyl Sulfoxide; Drocinonide; Endrysone; Enlimomab; Enolicam Sodium; Epirizole; Etodolac; Etofenamate; Felbinac; Fenamole; Fenbufen; Fenclofenac; Fenclorac; Fendosal; Fenpipalone; Fentiazac; Flazalone; Fluazacort; Flufenamic Acid; Flumizole; Flunisolide Acetate; Flunixin; Flunixin Meglumine; Fluocortin Butyl; FluoromethoIone Acetate; Fluquazone; Flurbiprofen; Fluretofen; Fluticasone Propionate; Furaprofen; Furobufen; Halcinonide; Halobetasol Propionate; Halopredone Acetate; Ibufenac; Ibuprofen; Ibuprofen Aluminum; Ibuprofen Piconol; Ilonidap; Indomethacin; Indomethacin Sodium; Indoprofen; Indoxole; Intrazole; Isoflupredone Acetate; Isoxepac; Isoxicam; Ketoprofen; Eofemizole Hydrochloride; Eomoxicam; Eoteprednol Etabonate; Meclofenamate Sodium; Meclofenamic Acid; Meclorisone Dibutyrate; Mefenamic Acid; Mesalamine; Meseclazone; Methylprednisolone Suleptanate; Momiflumate; Nabumetone; Naproxen; Naproxen Sodium; Naproxol; Nimazone; Olsalazine Sodium; Orgotein; Orpanoxin; Oxaprozin; Oxyphenbutazone; Paranyline Hydrochloride; Pentosan Polysulfate Sodium; Phenbutazone Sodium Glycerate; Pirfenidone; Piroxicam; Piroxicam Cinnamate; Piroxicam Olamine; Pirprofen; Prednazate; Prifelone; Prodolic Acid; Proquazone; Proxazole; Proxazole Citrate; Rimexolone; Romazarit; Salcolex; Salnacedin; Salsalate; Sanguinarium Chloride; Seclazone; Sermetacin; Sudoxicam; Sulindac; Suprofen; Talmetacin; Talniflumate; Talosalate; Tebufelone; Tenidap; Tenidap Sodium; Tenoxicam; Tesicam; Tesimide; Tetrydamine; Tiopinac; Tixocortol Pivalate; Tolmetin; Tolmetin Sodium; Triclonide; Triflumidate; Zidometacin; Zomepirac Sodium.

[0433] Non-limiting examples of anti-cancer drugs that can be administered in combination with the composition according to some embodiments of the invention include Acivicin; Aclarubicin; Acodazole Hydrochloride; Acronine; Adriamycin; Adozelesin; Aldesleukin; Altretamine; Ambomycin; Ametantrone Acetate; Aminoglutethimide; Amsacrine; Anastrozole; Anthramycin; Asparaginase; Asperlin; Azacitidine; Azetepa; Azotomycin; Batimastat; Benzodepa; Bicalutamide; Bisantrene Hydrochloride; Bisnafide Dimesylate; Bizelesin; Bleomycin Sulfate; Brequinar Sodium; Bropirimine; Busulfan; Cactinomycin; Calusterone; Caracemide; Carbetimer; Carboplatin; Carmustine; Carubicin Hydrochloride; Carzelesin; Cedefingol; Chlorambucil; Cirolemycin; Cisplatin; Cladribine; Crisnatol Mesylate; Cyclophosphamide; Cytarabine; Dacarbazine; Dactinomycin; Daunorubicin Hydrochloride; Decitabine; Dexormaplatin; Dezaguanine; Dezaguanine Mesylate; Diaziquone; Docetaxel; Doxorubicin; Doxorubicin Hydrochloride; Droloxifene; Droloxifene Citrate; Dromostanolone Propionate; Duazomycin; Edatrexate; Eflomithine Hydrochloride; Elsamitrucin; Enloplatin; Enpromate; Epipropidine; Epirubicin Hydrochloride; Erbulozole; Esorubicin Hydrochloride; Estramustine; Estramustine Phosphate Sodium; Etanidazole; Etoposide; Etoposide Phosphate; Etoprine; Fadrozole Hydrochloride; Fazarabine; Fenretinide; Floxuridine; Fludarabine Phosphate; Fluorouracil; Flurocitabine; Fosquidone; Fostriecin Sodium; Gemcitabine; Gemcitabine Hydrochloride; Hydroxyurea; Idarubicin Hydrochloride; Ifosfamide; Ilmofosine; Interferon Alfa-2a; Interferon Alfa-2b; Interferon Alfa-nl; Interferon Alfa-n3; Interferon Beta- I a; Interferon Gamma- I b; Iproplatin; Irinotecan Hydrochloride; Lanreotide Acetate; Letrozole; Leuprolide Acetate; Liarozole Hydrochloride; Lometrexol Sodium; Lomustine; Losoxantrone Hydrochloride; Masoprocol; Maytansine; Mechlorethamine Hydrochloride; Megestrol Acetate; Melengestrol Acetate; Melphalan; Menogaril; Mercaptopurine; Methotrexate; Methotrexate Sodium; Metoprine; Meturedepa; Mitindomide; Mitocarcin; Mitocromin; Mitogillin; Mitomalcin; Mitomycin; Mitosper; Mitotane; Mitoxantrone Hydrochloride; Mycophenolic Acid; Nocodazole; Nogalamycin; Ormaplatin; Oxisuran; Paclitaxel; Pegaspargase; Peliomycin; Pentamustine; Peplomycin Sulfate; Perfosfamide; Pipobroman; Piposulfan; Piroxantrone Hydrochloride; Plicamycin; Plomestane; Porfimer Sodium; Porfiromycin; Prednimustine; Procarbazine Hydrochloride; Puromycin; Puromycin Hydrochloride; Pyrazofurin; Riboprine; Rogletimide; Safingol; Safingol Hydrochloride; Semustine; Simtrazene; Sparfosate Sodium; Sparsomycin; Spirogermanium Hydrochloride; Spiromustine; Spiroplatin; Streptonigrin; Streptozocin; Sulofenur; Talisomycin; Taxol; Tecogalan Sodium; Tegafur; Teloxantrone Hydrochloride; Temoporfin; Teniposide; Teroxirone; Testolactone; Thiamiprine; Thioguanine; Thiotepa; Tiazofuirin; Tirapazamine; Topotecan Hydrochloride; Toremifene Citrate; Trestolone Acetate; Triciribine Phosphate; Trimetrexate; Trimetrexate Glucuronate; Triptorelin; Tubulozole Hydrochloride; Uracil Mustard; Uredepa; Vapreotide; Verteporfin; Vinblastine Sulfate; Vincristine Sulfate; Vindesine; Vindesine Sulfate; Vinepidine Sulfate; Vinglycinate Sulfate; Vinleurosine Sulfate; Vinorelbine Tartrate; Vinrosidine Sulfate; Vinzolidine Sulfate; Vorozole; Zeniplatin; Zinostatin; Zorubicin Hydrochloride. Additional antineoplastic agents include those disclosed in Chapter 52, Antineoplastic Agents (Paul Calabresi and Bruce A. Chabner), and the introduction thereto, 1202-1263, of Goodman and Gilman's "The Pharmacological Basis of Therapeutics", Eighth Edition, 1990, McGraw-Hill, Inc. (Health Professions Division).

[0434] Non-limiting examples of anti-cancer drugs that can be administered in combination with the composition according to some embodiments of the invention include abarelix, aldesleukin, aldesleukin, alemtuzumab, alitretinoin, allopurinol, altretamine, amifostine, anastrozole, arsenic trioxide, asparaginase, azacitidine, bevacuzimab, bexarotene, bleomycin, bortezomib, busulfan, calusterone, capecitabine, carboplatin, carmustine, celecoxib, cetuximab, cisplatin, cladribine, clofarabine, cyclophosphamide, cytarabine, dacarbazine, dactinomycin, actinomycin D, Darbepoetin alfa, Darbepoetin alfa, daunorubicin liposomal, daunorubicin, decitabine, Denileukin diftitox, dexrazoxane, dexrazoxane, docetaxel, doxorubicin, dromostanolone propionate, Elliott's B Solution, epirubicin, Epoetin alfa, erlotinib, estramustine, etoposide, exemestane, Filgrastim, floxuridine, fludarabine, fluorouracil 5-FU, fulvestrant, gefitinib, gemcitabine, gemtuzumab ozogamicin, goserelin acetate, histrelin acetate, hydroxyurea, Ibritumomab Tiuxetan, idarubicin, ifosfamide, imatinib mesylate , interferon alfa 2a, Interferon alfa- 2b, irinotecan, lenalidomide, letrozole, leucovorin, Leuprolide Acetate, levamisole, lomustine, CCNU, meclorethamine, nitrogen mustard, megestrol acetate, melphalan, L-PAM, mercaptopurine 6-MP, mesna, methotrexate, mitomycin C, mitotane, mitoxantrone, nandrolone phenpropionate, nelarabine, Nofetumomab, Oprelvekin, Oprelvekin, oxaliplatin, paclitaxel, palifermin, pamidronate, pegademase, pegaspargase, Pegfilgrastim, pemetrexed disodium, pentostatin, pipobroman, plicamycin mithramycin, porfimer sodium, procarbazine, quinacrine, Rasburicase, Rituximab, sargramostim, sorafenib, streptozocin, sunitinib maleate, tamoxifen, temozolomide, teniposide VM-26, testolactone, thioguanine 6-TG, thiotepa, thiotepa, topotecan, toremifene, Tositumomab, Trastuzumab, tretinoin ATRA, Uracil Mustard, valrubicin, vinblastine, vinorelbine, zoledronate and zoledronic acid.

[0435] According to specific embodiments, the composition is administered in combination with an immune-check point protein inhibitor. Non-limiting examples of such immune-checkpoint proteins include PD1, PDL-1 and CTLA4.

[0436] Examples of PD1 inhibitors include, without limitation, pembrolizumab, nivolumab, cemiplimab, spartalizumab, sintilimab, tislelizumab, toripalimab, dostarlimab, JTX-4014, INCMGA00012, AMP-224, and AMP-514.

[0437] Examples of PDL-1 inhibitors include, without limitation, atezolizumab, durvalumab, avelumab, KN035, CK-301, AUNP12, CA-170 and BMS-986189. Examples of CTLA4 inhibitors include, without limitation, ipilimumab, tremelimumab and nivolumab.

[0438] According to specific embodiments, the composition is administered in combination with Bacillus Calmette-Guerin (BCG).

[0439] According to specific embodiments, the disease is bladder cancer and the additional anti- cancer agent is Bacillus Calmette-Guerin (BCG).

[0440] The composition of some embodiments of the invention can be administered to an organism per se, or in a pharmaceutical composition where it is mixed with suitable carriers or excipients.

[0441] 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.

[0442] Herein the term "active ingredient" refers to the agent (e.g., agent downregulating Zeb2 in macrophages) accountable for the biological effect.

[0443] Hereinafter, the phrases "physiologically acceptable carrier" and "pharmaceutically acceptable carrier" which may be interchangeably used 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.

[0444] 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.

[0445] Techniques for formulation and administration of drugs may be found in “Remington’s Pharmaceutical Sciences,” Mack Publishing Co., Easton, PA, latest edition, which is incorporated herein by reference.

[0446] 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, intracardiac, e.g., into the right or left ventricular cavity, into the common coronary artery, intravenous, intraperitoneal, intranasal, or intraocular injections.

[0447] Conventional approaches for drug delivery to the central nervous system (CNS) include: neurosurgical strategies (e.g., intracerebral injection or intracerebro ventricular infusion); molecular manipulation of the agent (e.g., production of a chimeric fusion protein that comprises a transport peptide that has an affinity for an endothelial cell surface molecule in combination with an agent that is itself incapable of crossing the BBB) in an attempt to exploit one of the endogenous transport pathways of the BBB; pharmacological strategies designed to increase the lipid solubility of an agent (e.g., conjugation of water-soluble agents to lipid or cholesterol carriers); and the transitory disruption of the integrity of the BBB by hyperosmotic disruption (resulting from the infusion of a mannitol solution into the carotid artery or the use of a biologically active agent such as an angiotensin peptide). However, each of these strategies has limitations, such as the inherent risks associated with an invasive surgical procedure, a size limitation imposed by a limitation inherent in the endogenous transport systems, potentially undesirable biological side effects associated with the systemic administration of a chimeric molecule comprised of a carrier motif that could be active outside of the CNS, and the possible risk of brain damage within regions of the brain where the BBB is disrupted, which renders it a suboptimal delivery method.

[0448] 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.

[0449] According to specific embodiments, the composition is administered via intravesical injection. In intravesical the composition is directly administered into the bladder through the urethra using a catheter. By delivering medication directly into the bladder, intravesical injection maximizes the concentration of the drug at the target site (i.e. bladder e.g. when treating bladder cancer) while minimizing systemic exposure and potential side effects.

[0450] Pharmaceutical compositions of some embodiments of the 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.

[0451] Pharmaceutical compositions for use in accordance with some embodiments of the 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 which, can be used pharmaceutically. Proper formulation is dependent upon the route of administration chosen.

[0452] 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.

[0453] 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 if 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, sodium carbomethylcellulose; and / or physiologically acceptable polymers such as polyvinylpyrrolidone (PVP). If desired, disintegrating agents may be added, such as cross-linked polyvinyl pyrrolidone, agar, or alginic acid or a salt thereof such as sodium alginate.

[0454] 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.

[0455] Pharmaceutical compositions which 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. All formulations for oral administration should be in dosages suitable for the chosen route of administration.

[0456] For buccal administration, the compositions may take the form of tablets or lozenges formulated in conventional manner.

[0457] For administration by nasal inhalation, the active ingredients for use according to some embodiments of the 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 unit may be determined by providing a valve to deliver a metered amount. Capsules and cartridges of, e.g., 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.

[0458] The pharmaceutical composition described herein may be formulated for parenteral administration, e.g., by bolus injection or continues 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.

[0459] 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 acids esters such as ethyl oleate, triglycerides or liposomes. Aqueous injection suspensions may contain substances, which increase the viscosity of the suspension, such as sodium carboxymethyl cellulose, sorbitol or dextran. Optionally, the suspension may also contain suitable stabilizers or agents which increase the solubility of the active ingredients to allow for the preparation of highly concentrated solutions.

[0460] Alternatively, the active ingredient may be in powder form for constitution with a suitable vehicle, e.g., sterile, pyrogen-free water based solution, before use.

[0461] The pharmaceutical composition of some embodiments of the invention may also be formulated in rectal compositions such as suppositories or retention enemas, using, e.g., conventional suppository bases such as cocoa butter or other glycerides.

[0462] Pharmaceutical compositions suitable for use in context of some embodiments of the 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 effective to prevent, alleviate or ameliorate symptoms of a disorder (e.g., inflammatory disease e.g. cancer) or prolong the survival of the subject being treated.

[0463] 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.

[0464] For any preparation used in the methods of the invention, the therapeutically effective amount or dose 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.

[0465] 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, et al., 1975, in "The Pharmacological Basis of Therapeutics", Ch. 1 P-l).

[0466] Dosage amount and interval may be adjusted individually to provide levels of the active ingredient are sufficient to induce or suppress the biological effect (minimal 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.

[0467] 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.

[0468] 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.

[0469] Compositions of some embodiments of the 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 may also be accommodated by a notice associated with the container 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 or human or veterinary administration. Such notice, for example, may be of 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 compatible pharmaceutical carrier may also be prepared, placed in an appropriate container, and labeled for treatment of an indicated condition, as is further detailed above. As further shown in the Examples section which follows, the present inventors have uncovered that changes in Zeb2 expression are indicative of cancer prognosis. Specifically, expression of Zeb2 was correlated with poor patient survival in macrophage-rich solid tumors (Example 6 of the Examples section which follows). Consequently, specific embodiments, suggest increased levels of macrophages and Zeb2 can be used as a marker for e.g., prognosing, monitoring efficacy of treatment and treating of an inflammatory disease e.g., cancer.

[0470] Thus, according to an aspect of the present invention, there is provided a method of prognosing an inflammatory disease in a subject, the method comprising determining a level of macrophages and a level of Zeb2 a biological sample of a subject diagnosed with the inflammatory disease, wherein said level of said macrophages and said level of said Zeb2 above a predetermined threshold is indicative of poor prognosis, thereby prognosing the inflammatory disease in the subject.

[0471] As used herein the term “prognosing” refers to determining the outcome of the disease (inflammatory disease e.g., cancer).

[0472] As used herein “poor prognosis” refers to increased risk of death due to the disease, increased risk of progression of the disease (e.g., cancer grade), and / or increased risk of recurrence of the disease.

[0473] The phrase “biological sample” as used herein refers to a fluid or non-fluid biological sample which comprises cells and is obtained from the subject. Examples include but are not limited to, a blood sample, a urine sample, saliva, rinse fluid that may have been in contact with the tumor, a tissue biopsy, a tissue and an organ etc.

[0474] According to specific embodiments, the sample is an in-situ sample (e.g., of the cancer).

[0475] According to specific embodiments, the biological sample is a tissue or cell biopsy (e.g., cancer biopsy).

[0476] According to specific embodiments, the method of the present invention comprises obtaining the biological sample prior to the determining.

[0477] The biological sample can be obtained using methods known in the art such as using a syringe with a needle, a scalpel, fine needle aspiration (FNA), catheter and the like. According to specific embodiments, the biological sample is obtained by biopsy. Hence, according to specific embodiments, determining the level of macrophages and / or Zeb2 is effected ex-vivo or in-vitro.

[0478] Determining a level of macrophages and / or Zeb2 may be effected by any method known in the art.

[0479] Thus, for example, the level of macrophages can be determined by morphology or histochemistry. Additionally or alternatively, determining the level of macrophages may be effected using a macrophage surface marker. Such markers and methods of determining their expression level are known in the art and are further described in details hereinabove. According to specific embodiments, the macrophage marker is CD68.

[0480] In this context, “a level of Zeb2” refers to the degree of gene expression (e.g., mRNA or protein). Thus, determining the level of Zeb2 may be effected at the polynucleotide or polypeptide level and includes methods such as PCR, Northern blot, RNA in situ hybridization stain Western-blot, ELISA, flow cytometry, immunostaining and the like.

[0481] According to specific embodiments, the level of macrophages and the level Zeb2 is determined independently of each other, and increased levels of both above a predetermined threshold is indicative of poor prognosis.

[0482] According to other embodiments, a level of macrophages positive for Zeb2 above a predetermined threshold is indicative of poor prognosis.

[0483] As used herein the phrase “predetermined threshold” refers to a level that characterizes a sample of the same origin obtained from a subject not suffering from the pathology or a subject known to have good prognosis, assayed under the same conditions. Such a level can be experimentally determined by comparing samples from a subject not suffering from the pathology of interest or having a good prognosis to samples derived from subjects having poor prognosis. Alternatively, such a level can be obtained from the scientific literature and from databases.

[0484] According to specific embodiments, the predetermined threshold is derived from a control sample.

[0485] According to specific embodiments, the control sample is a healthy control sample.

[0486] According to specific embodiments, the pre-determined threshold can be determined in a subset of subjects with a known outcome.

[0487] According to specific embodiments, the control sample is obtained from the scientific literature or from a database.

[0488] According to specific embodiments, the increase above or below a predetermined threshold is statistically significant (e.g., P<0.05).

[0489] According to specific embodiments, the predetermined threshold is at least 1.5fold, at least 2 fold, at least 3 fold, at least 5 fold, at least 10 fold, or at least 20 fold as compared the level in a control sample as measured using the same assay, as further disclosed herein.

[0490] According to specific embodiments, the predetermined threshold is at least 2 %, at least 5 %, at least 10 %, at least 20 %, at least 30 %, at least 40 %, at least 50 %, at least 60 %, at least 70 %, at least 80 %, at least 90 %, e.g., 100 %, at least 200 %, at least 300 %, at least 400 %, at least 500 %, at least 600 % as compared the level in a control sample.

[0491] Specific embodiments of the present invention further contemplate making appropriate treatment decisions based on the prognosis. That is, a disease with poor prognosis is treated with a treatment regime suitable for poor prognosis according to e.g. established protocols; while a disease with good prognosis is treated with a treatment regime suitable for good prognosis according to other e.g. established protocols.

[0492] According to specific embodiments, the determined levels can indicate the likelihood that the inflammatory disease (e.g. cancer) will respond to the given therapy (personalized medicine). Hence, according to specific embodiments, the anti-inflammatory (e.g. anti-cancer) therapy is selected based on the levels of macrophages and Zeb. According to specific embodiments, when levels of macrophages and Zeb2 above a predetermined levels are indicated treatment comprises the composition disclosed herein.

[0493] Hence, according to an aspect of the present invention, there is provided a method of treating an inflammatory disease in a subject in need thereof, the method comprising:

[0494] (a) prognosing the inflammatory disease according to the method disclosed herein; and

[0495] (b) selecting a therapeutic agent based on the determined levels.

[0496] According to an additional or an alternative aspect of the present invention, there is provided a method of treating an inflammatory disease in a subject in need thereof, the method comprising:

[0497] (a) prognosing the inflammatory disease according to the method disclosed herein; and

[0498] (b) selecting a therapeutic agent based on the prognosis.

[0499] According to an additional or an alternative aspect of the present invention, there is provided a method of treating an inflammatory disease in a subject in need thereof, the method comprising:

[0500] (a) prognosing the inflammatory disease according to the method disclosed herein; and

[0501] (b) administering to the subject a therapeutically effective amount of an anti- inflammatory agent according to the determined levels.

[0502] According to an additional or an alternative aspect of the present invention, there is provided a method of treating an inflammatory disease in a subject in need thereof, the method comprising: (a) prognosing the inflammatory disease according to the method disclosed herein; and

[0503] (b) administering to the subject a therapeutically effective amount of an anti- inflammatory agent according to the prognosis.

[0504] As the levels of macrophages and Zeb2 can be used for determining disease (e.g., cancer) progression, the present invention also contemplates methods of monitoring treatment efficacy in subject in need thereof.

[0505] Thus, according to specific embodiments, the method of prognosing comprises monitoring disease state in a subject, comprising repeating the analysis following a predestined time interval.

[0506] According to an aspect of the present invention, there is provided a method of monitoring efficacy of treatment in a subject diagnosed with an inflammatory disease, the method comprising determining a level of macrophages and a level of Zeb2 in a biological sample of the subject following treatment, wherein a decrease in the expression level following treatment is indicative of efficaciousness of the treatment.

[0507] On the other hand, if there is no change in the levels, or in case there is an increase in the levels, then the therapy is not efficient in treating the subject and additional and / or alternative therapies (e.g., treatment regimens) may be used.

[0508] As used herein the term “about” refers to ± 10 %

[0509] The terms "comprises", "comprising", "includes", "including", “having” and their conjugates mean "including but not limited to".

[0510] The term “consisting of’ means “including and limited to”.

[0511] 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.

[0512] As used herein, the singular form "a", "an" and "the" include plural references unless the context clearly dictates otherwise. For example, the term "a compound" or "at least one compound" may include a plurality of compounds, including mixtures thereof.

[0513] 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.

[0514] 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.

[0515] 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.

[0516] When reference is made to particular sequence listings, such reference is to be understood to also encompass sequences that substantially correspond to its complementary sequence as including minor sequence variations, resulting from, e.g., sequencing errors, cloning errors, or other alterations resulting in base substitution, base deletion or base addition, provided that the frequency of such variations is less than 1 in 50 nucleotides, alternatively, less than 1 in 100 nucleotides, alternatively, less than 1 in 200 nucleotides, alternatively, less than 1 in 500 nucleotides, alternatively, less than 1 in 1000 nucleotides, alternatively, less than 1 in 5,000 nucleotides, alternatively, less than 1 in 10,000 nucleotides.

[0517] 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.

[0518] Various embodiments and aspects of the present invention as delineated hereinabove and as claimed in the claims section below find experimental support in the following examples. EXAMPLES

[0519] 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.

[0520] Generally, the nomenclature used herein and the laboratory procedures utilized in the present invention include molecular, biochemical, microbiological and recombinant DNA techniques.

[0521] MATERIALS AND METHODS

[0522] Mice - A founding breeding pair of Rosa26-Cas9 knock-in B6J mice was purchased from Jackson. Mice were then bred in the Weizmann Institute animal facility and backcrossed with wild-type mice (C57BL / 6); their progeny was crossed to produce Cas9-GFP homozygotes on a cleaner C57BL / 6 background. In all experiments, wild-type C57BL / 6 mice or Cas9-GFP young adult (7-11 weeks old) females were used. Mice were provided with food and water ad libitum and housed under a strict 12-hours light-dark cycle. All experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC).

[0523] Isolation of human PBMCs - Peripheral blood mononuclear cells (PBMCs) were isolated from fresh blood collected from a healthy donor (Weizmann IRB approval 2448-2) using sterile density gradient centrifugation with Ficoll-Paque (GE Healthcare Life Sciences).

[0524] Tumor cell lines - MC38 murine colon adenocarcinoma, MB 49 murine urothelial carcinoma and Lewis Lung Carcinoma (LLC) cell lines were cultured with RPML1640 medium (Gibco) supplemented with 10 mM HEPES buffer, 2 mM L-glutamine, 10 % heat- inactivated FBS, 1 mM sodium pyruvate, 1 % penicillin-streptomycin and lx non-essential amino acids. Cells were cultured in 100-mm tissue culture plates in an incubator with humidified air and 5 % CO2 at 37 °C. Cell lines were validated for lack of mycoplasma using an EZPCR mycoplasma kit (Biological Industries).

[0525] Cloning of guide RNAs into lentiGuide-puro plasmid - Guide RNA oligos were cloned into a lentiGuide-puro plasmid (addgene #52963). Briefly, gRNAs synthesized in pairs of oligonucleotides (IDT) with BsmBI-compatible overhangs were phosphorylated with T4 polynucleotide kinase (NEB) and annealed. The fragments were then inserted into lentiGuide- puro using Golden Gate cloning with BsmBI restriction enzyme (NEB) and T4 ligase (NEB). Ligated constructs were transformed into competent bacteria, and single clones were picked up and propagated. Finally, plasmids containing guide RNAs were purified and sequenced using Sanger sequencing with a primer for the U6 promoter to confirm the proper insertion of the guide. For each gene KO, two different guides obtained from the mouse CRISPR Brie KO library were cloned. Lentivirus production - Lentiviral particles were produced by transfecting HEK293T cells growing in tissue culture-treated six- well plates with a pool of two lentiGuide-puro plasmids (two different guides) per gene KO together with packaging plasmids, using the jetPEI transfection reagent (Polyplus-sartorius) according to the manufacturer’s protocol. Medium was replaced with DMEM+10XFBS 18 hours post-transfection, and media containing virus particles were collected 48- and 72-hours post- transfection. Virus particles from one well (-3.8 mL per well) were concentrated using Amicon Ultra 100 KDa 5ml centrifugal filter units (Millipore) spun down at 1100 x g at 4 °C to a final concentration of 100-200 pl per virus, aliquoted, and stored at ---80 °C until use.

[0526] Transduction of hone marrow-derived macrophages (BMDM) culture - Bone marrow was isolated from 8-weeks-old female Cas9-GFP mice by flushing the femora and tibiae with CIO medium (RPMI-1640 supplemented with 10 % heat-inactivated fetal bovine serum (FBS), lx MEM-eagle non-essential amino acid, 10 mM HEPES buffer, 1 mM sodium pyruvate, 2 mM L-glutamine, 1 % penicillin- streptomycin, and 50 pM p-mercaptoethanol) using a 21 -gauge needle. The suspension was then filtered through a 70 pm cell strainer. Cells were centrifuged at 350 x g for 5 minutes at 4 °C. The cell pellet was then resuspended in red blood cell lysis solution (Sigma-Aldrich) and incubated for 5 minutes at room temperature, followed by washing and resuspension in CIO medium. Cells were then cultured by plating 350k cells in 1 ml of CIO supplemented with 30 ng / ml human MCSF (Peprotech) per well in a 12-well non-tissue culture plate. On day 2, cells were transduced with lenti viruses by adding 10-20 pl concentrated virus per well in a triplicate per each knockout combined with 4 pg / ml Polybrene (Sigma Aldrich). On day 3, the medium was replaced with 1 ml CIO supplemented with 30 ng / ml human MCSF and 3.5 pg / ml puromycin. On day 6, the medium was replaced with 1 ml CIO supplemented with 30 ng / ml human MCSF and 20 ng / ml murine IL-4 (Peprotech). On day 7, cells were detached using Accutase solution (Sigma-Aldrich) and used in the different downstream analyses.

[0527] BMDM cytokine stimulation - To screen for the cytokine stimulation that generates T AM-programmed phenotype, BMDMs from WT C57BL / 6 were cultured as described above. On days 2 and 5, the medium was replenished with fresh CIO medium supplemented with 30 ng / ml human MCSF. On day 6, the medium was replenished with fresh CIO supplemented with 30 ng / ml human MCSF and 20 ng / ml of the different murine cytokines IL-4, IL-13, IL-10, IFN-y, and TGF-P (Peprotech). For unstimulated control, cells were left untreated. On day 7, cells were detached using Accutase solution (Sigma-Aldrich) and used for scRNA-seq. BMDM treatment with mLNP - 1 mg of mannosylated lipid-nanoparticles (mLNP) (purchased from GenScript. Cat# sc2337) loaded with Zeb2-targeting or non-targeting control (NT) guides (SEQ ID Nos: 42 and 43, respectively) were added to the CAS9+ BMDM culture (from Cas9-GFP mouse) on day 4 and the medium was replaced on day 6 with fresh CIO medium supplemented with 30 ng / ml human M-CSF and 20 ng / ml murine IE-4 (Peprotech). On day 7, cells were detached using Accutase solution (Sigma-Aldrich) and used for scRNA-seq.

[0528] Combinatorial indexing of the perturbed cells in 12-well plates for index sorting - To deal with the high number of wells used in the arrayed CRISPR screen (120 candidate genes x 3 replicates = 360 wells), a fluorophore -based combinatorial indexing approach was established, allowing minimizing the number of FACS tubes used to sort an equal number of cells (-384 cells) per one well in each 12-wells plate. For this, a staining protocol was calibrated with anti- CD45 antibodies. On day 7 of the BMDM: culture, cells were washed once with 2 ml pre- warmed PBS (- / -) and then stained with a different combination (Figure 6A) of fluorophore- conjugated anti-CD45 antibodies (APC, PE, APC / Cy7, PE-Dazzle 594, BV421, BV605 and BV711) in 1 ml of warm CIO medium for 15 minutes at 37 °C. Cells were then washed twice with 1 ml pre-warmed PBS (- / -) and detached using Accutase solution (Sigma- Aldrich) for 15 minutes at 37 °C. Afterwards, ice-cold 1 ml FACS buffer [PBS(- / -) supplemented with 0.5% BSA and 2 mM EDTA] was added to each well, and the 12 wells in each plate were combined and filtered through a 70 pm cell strainer. Cells were then centrifuged for 5 minutes at 350 x g at 4°C. Finally, the cell pellet was resuspended in 2 ml FACS buffer and moved to FACS tubes for a total of 32 different pools in 32 FACS tubes. From each FACS tube, single ceils were sorted into 12 different premade 384-well SPID-seq capture plates for scRNA-seq.

[0529] BMDM-T-cell coculture suppression assay - Spleens from 10-weeks-old female WT C57BL / 6 mice were isolated, mashed, and passed through a 70 pm strainer to prepare a single cell suspension and then subjected to a Pan T-cell isolation kit (Miltenyi Biotec, 130-095-130), according to the kit’s guidelines. T cells were then labeled with cell proliferation dye eFluorTM 450 (Thermo Fisher, 65-0842-85) according to the manufacturer’s guidelines. Perturbed IL-4- stimulated BMDMs were collected at day 7 from their wells using Accutase solution and kept in CIO medium until the coculture. For activating the T cells in the coculture wells, a 96-well U- shaped tissue-culture-treated plate was pre-coated with 1 pg Ultra-LEAF purified anti-mouse CD3 per well (BioLegend) for two hours at 37 °C and then washed twice with PBS (- / -). T cells and BMDMs were counted and cocultured in a 1 : 3 ratio (BMDM : T-cell) in 100 pL CIO medium per well, supplied with 2 pg / mL Ultra-LEAF purified anti-mouse CD28 (BioLegend) and kept in a 37 °C, 5 % CO2 incubator. Cells were harvested after 72 hours, stained with anti- CD4 and anti-CD8 antibodies and cell proliferation was analyzed using an LSRII FACS analyzer (BD). Murine IFN-y secretion was measured from the cocultures’ cell-free supernatant after 48 hours by an ELISA kit (BioLegend, 430801).

[0530] Antigen presentation assay - Zeb2 KO or NT control BMDMs stimulated with IL-4 were harvested on day 7 using Accutase, counted, and seeded at 20,000 cells per well in 100 mL of CIO medium in tissue culture-treated flat-bottom 96-well plates. Cells were incubated overnight to allow adherence. On day 8, BMDMs were treated with OVA257-264 or OVA332-339 peptides (0, 1.5, or 3 mg / ml) in 100 mL of CIO medium supplemented with 30 ng / ml human M-CSF (PeproTech) and 20 ng / ml murine IL-4 (PeproTech) for 8 hours at 37 °C in a 5 % CO2 incubator. Spleens from 12-week-old female OT-I and OT-II mice were collected, homogenized, and filtered through a 70 mmcell strainer to generate single-cell suspensions. T cells were purified using Pan T-cell and CD8 (Ly-2) isolation kits (Miltenyi Biotec) following the manufacturer’s protocol, yielding CD8+ OT-I and CD4+ OT-II cells separately. Peptide-loaded BMDMs were washed twice with 200 mL of CIO medium and co-cultured with 60,000 purified OT-I or OT-II T cells in 100 mL of CIO medium supplemented with 30 ng / ml human M-CSF and 20 ng / ml murine IL-4. After 48 hours, T cells were harvested, stained with activation markers, and analyzed using a Symphony S6 flow cytometer (BD Biosciences).

[0531] Phagocytosis assay - Zeb2 KO or NT control BMDMs stimulated with IL-4 were harvested on day 7 using Accutase, counted, and seeded at 50,000 cells per well in 100 mL of CIO medium in tissue culture-treated flat-bottom 96-well plates. Cells were incubated overnight to allow adherence. On day 8, BMDMs were washed and incubated with 0.3 mg / ml pre- sonicated pHrodo Red S. aureus Bioparticles Conjugate for Phagocytosis (ThermoFisher) for 10 minutes in the same culture medium. Cells were then washed twice with 200 mL MACS buffer and analyzed for their phagocytosis activity using a Symphony S6 flow cytometer (BD Biosciences) for PE fluorescence.

[0532] ZEB2 KO in human CD14-derived macrophages - PBMCs were isolated from whole blood of a healthy donor as indicated above. CD 14+ monocytes were further purified from the PBMCs using human CD 14 MicroBeads (Miltenyi Biotec). Cas9-ribonucleoprotein nucleofection into the primary monocytes was performed following the protocol by Hiatt et al. (Cell Rep. 2021, 35, 109105), using the DK-100 nucleofection program and P2 Lonza buffer with the Gene Knockout Kit v2 (Synthego) targeting ZEB2 (SEQ ID NOs: 36-38) and negative control (SEQ ID NO: 39). The edited CD 14+ monocytes were cultured in CIO medium supplemented with 50 ng / ml human M-CSF (Peprotech) for five days. On day 5, CD14+-derived macrophages were stimulated with human IL-4 (PeproTech) for two days. Cells were then harvested using Accutase and used for scRNA-seq.

[0533] CpG-siRNA conjugates design and synthesis - -Phosphothioated murine optimized CpG1668-ODN (TCCATGACGTTCCTGATGCT, SEQ ID NO: 33) and passenger strands (SS, SEQ ID NOs: 23 and 25, as delineated below) of siRNAs for Zeb and for Control (scrambled sequence that does not recognize any target), respectively, were linked using five units of a C3 carbon chain linker (iSpC3 linker (CH2-CH2-CH2)x5. The resulting constructs were hybridized to complementary siRNA guide (SEQ ID NOs: 24 and 26, as delineated below) strands to generate the CpG-siRNA conjugates. Sequences of the single-stranded constructs are as follows below (^indicates a phosphothioation site, r - indicates a ribonucleic acid): CpG-siRNAZeb2(pas senger strand) :

[0534] 5 ’ -CpG 1668-ODN / linker-rC*rC*rGrArArUrGrArGrArArArCrArArUrArUrCrArA- 3 ’ , its schematic representation is provided in Figure 13 A. siRNAZeb2(guide strand):

[0535] 5 ’ -rU* rU* rGr ArUr ArUrUrGrUrUrUrCrUrCrA rUrUrCrGrG* rU* rU-3 ’ , its schematic representation is provided in Figure 13B.

[0536] CpG-siRNACtrl(passenger strand ):

[0537] 5 ’ -CpG 1668-ODN / linker-rA*rU*rArArCrArCrGrArCrGrArArCrUrArGrArArUrA-3 ’ siRNACtrl(guide strand):

[0538] 5 ’ -rU*r A*rUrUrCrUr ArGrUrUrCrGrUrCrGrUrGrUrU*r A*rU-3 ’ .

[0539] To reiterate, the CpG-siRNA passenger strand and the siRNA guide strand were synthesized seperately. The starting point of each synthesis was a protected nucleoside linked via its 3’- oxygen to a controlled pore glass-based solid support. Nucleoside phosphoramidite chemistry was used for this synthesis; and the synthesis cycle comprised the following steps: (1) Deprotection of the 5 ’-hydroxyl group (Detritylation); (2) Coupling of nucleotide phosphoramidite to the 5’-hydroxyl group; (3) Capping of unreacted 5’-hydroxyl groups; and (4) Oxidation. Step (4) can be substituted with a sulfurization step for the synthesis of phosphorothioated oligonucleotides. These four steps were repeated in the above order until all nucleoside components were added. After the complete synthesis and purification of the CpG- siRNA passenger strand and the siRNA guide strand, the two components were annealed to produce the CpG-siRNA conjugate. qPCR for BMDM treated with CpG-siRNA conjugates - BMDMs from 8-weeks-old wild-type C57BL / 6 were cultured in a 96-well tissue culture plate and on day 5 cells were stimulated with 20 ng / ml murine IL-4 for two days. On day 7, cells were treated with 100 nM CpG-siRNAZeb2, CpG-siRNAControl conjugates or left untreated. Twenty-four hours after the treatment, cells in wells were washed with PBS and RNA was isolated using 12 mL of Dynabeads oligo(dT) (Invitrogen), washed, and eluted in 10 ml. mRNA was reverse transcribed using a High-Capacity cDNA Reverse Transcription kit (ThermoFisher) and cDNA was diluted 1:40 for qPCR measurement using mouse-specific Zeb2 primers (SEQ IS Nos: 40-41).

[0540] Subcutaneous mouse tumor models and treatment - MC38 (1.5 x 10°) or MB49 (1.0 x 106) cells were implanted subcutaneously on both flanks of adult female WT C57 / BL6 mice, while LLC (Lewis Lung Carcinoma) cells (1.0 x 106) were injected subcutaneously on a single flank. Tumor volumes were measured every 2 days with an electronic calliper by a researcher blinded to the treatment group. When tumors reached 80-100 mm3(typically 8-9 days post inoculation), mice were randomized into groups by tumor sizes. For MB49, MC38, and LLC models, one tumor per mouse was treated with intratumoral injections of CpG-siRNA conjugates (1, 2, or 5 mg / kg, respectively) or saline (untreated control) in a volume of 20-30 pl every' other day, while the contralateral tumor remained untreated. For experiments involving mannosylated lipid nanoparticles (mLNPs, GenScript) carrying either Zeb2 or non-targeting control CRISPR guides, female Cas9-GFP mice were used. Animals were implanted subcutaneously with MC38 tumor cells ( 1.5 x IO6) and tumor volumes were measured every 2 days as described. On day 6 after tumor cell implantation, mice intravenously received every' 2 days 1 mg / kg of either mLNPs carrying Zeb2-targeting guide (SEQ ID NO: 42) (mLNP-Zeb2) or control LNPs with non- targeting control guide (SEQ ID NO: 43) (mLNP-NT) in a volume of 100 ml PBS.

[0541] Orthotopic mouse bladder tumor model - The urinary bladder of 7- week-old female C57BL / 6 mice was catheterized using a 25-gauge Teflon catheter, followed by insertion of a 24- gauge soft-tipped guide wire through the urethra to the bladder wall. Monopolar coagulation current was then applied for 5 seconds. Subsequently, upon removal of the wire, a suspension containing (3 x 104) luciferase- stably transfected MB49 cells in 100 pl of saline was instilled into the bladder. On day 9, the mice were divided into three treatment groups with equal average luciferase signals and treated two times every other day with either 2 mg / kg of CpG-siRNAZeb2or CpG-siRNACtrlconjugates or PBS only (untreated group), administered via a 25-gauge Teflon catheter inserted into the bladder. To assess treatment efficacy, mice were sacrificed on day 15, and their bladders were extracted and weighed using an analytical scale.

[0542] Isolation of tumor-infiltrating leukocytes - Tumor-bearing mice were sacrificed at the described time points. To isolate tumor- infiltrating immune cells, tumors were digested using mechanical (gentleMACS C tube, Miltenyi Biotec) enzymatic digestion (RPMI-1640 supplemented with 0.1 mg / ml DNase type I (Roche) and 1 mg / ml collagenase IV (Worthington)) for 15 minutes at 37 °C. Ceils were then filtered through a 100 pm cell strainer, washed with ice-cold MACS buffer (PBS supplemented with 0.2 mM EDTA, pH 8, and 0.5 % BSA), and centrifuged for 5 minutes at 300 x g at 4 °C and then resuspended in ice-cold MACS buffer

[0543] Lymph node and spleen dissociation ~ Mice were sacrificed and spleen and tdLN tissues were isolated and mashed through a 100 pm cell strainer and washed with ice-cold MACS buffer. Cells were centrifuged at 400 x g for 5 minutes at 4 °C. The cell pellets were then resuspended in red blood cell lysis buffer (Sigma-Aldrich) and incubated for 5 minutes at room temperature, washed with ice-cold PBS, and centrifuged at 400 x g for 5 minutes at 4 °C and then resuspended in ice-cold MACS buffer for downstream applications.

[0544] Flow cytometry analysis and sorting of tumor and lymph node single-cell suspensions - Single-cell suspensions from tumors and tdLN were resuspended in ice-cold MACS buffer supplemented with Fc-blocking (1 : 200) (TruStain FcX, Clone 93, BioLegend) for 5 minutes to block Fc receptors before labeling with fluoroph ore-conjugated antibodies against cell surface epitopes followed by flow cytometry analysis and sorting. In the MC38 tumor model, cells were stained with CD45-BV711 (30-F11, BD Biosciences), CD68-PerCP-Cyanine5.5 (FA-11), CDl lb-BV605 (MI / 70), Ly-6G-PacificBlue (1A8), Ly-6C-FITC (HK1.4), CD90.2-Alexa Fluor®700 (53-2.1). CD4-PE-Cy7 (GK1.5). CD8a-PE / Dazzle™ 594 (53-6.7) and 0.1 ug / ml DAPI, all from Biolegend and sorting gates were DAPI- Ly6G- CD90.2+ / CD1 lb+. In the MC38 tdLN, cells w'ere stained with CD45-FITC (30-F11), TCRb-PerCP-Cyanine5.5 (H57-597), CD4-BV480 (RM4-5), CD8-PE-Cy7 (53-6.7), NKl.l-BV605(), CD1 Ic-PE-Dazzle™ 594 (N418), CD69-PE (H1.2F3), CD25-BV711 (PC61), CD44-APC (IM7), CD62L- Alexa Fluor®700 (MEL14), PD- 1 - APC / Cyanine7 (29F.1A12) and T cells were enriched by sorting DAPI- TCRb+ NKL1+Z- cells. In the MB49 tumor, cells were fixed using eBioscience™ intracellular fixation & permeabilization kit (Cat# 88-8824-00) and stained with Invitrogen™ LIVE / DEAD Fixable Aqua Dead Cell staining and Zeb2 antibody (ThermoFisher, Cat# PA5-78302) conjugated to Alexa Fluor 647 fluorophore using Zenon labeling kit (Cat# Z25308) and CDllb-PE. Myeloid cells were analyzed for the Zeb 2 MFI and Zeb2+ cells frequency to validate Zeb2 silencing following intratumoral injection with CpG-siRNAZeb2.

[0545] OMNI ATAC-sequencing method - Samples were processed in duplicates as described in63, with slight modifications. Briefly, each sample contained 25,000 cryo-preserved cells. Cells were thawed, washed once with PBS, and nuclei were isolated and permeabilized with OMNL ATAC buffer (10 mM Tris pH 7.5, 10 mM NaCl, 3 mM MgCl2, 0.1% NP-40, 0.1% Tween-20, 0.1% Digitonin). Tagmentation was carried out by incubating the nuclei in OMNI- AT AC transposition mix (IX TD buffer (Illumina 20034197), 0.01 % Digitonin, 0.1% Tween-20, 0.3X PBS) supplemented with 1.25 pl Tn5 enzyme (Illumina 20034197) for 60 minutes at 37 °C. The tagmentation reaction was stopped by adding transposition stop mix (900 mM NaCl, 30 mM EDTA, 1 % SDS, 1.6U Proteinase K (NEB P8107)) and incubating at 40 °C for 30 minutes. Transposed DNA was cleaned with Ampure XP beads (Beckman Coulter A63881). Libraries were completed and amplified with 8-9 cycles of PCR with barcoded adapters (IDT). The number of amplification cycles was determined by qPCR. Libraries were sequenced on a NovaSeq 6000 instrument (Illumina), allocating ~20 million paired-end reads per sample.

[0546] Single-cell sorting and RNA library preparation - Single cells were sorted using Symphony S6 (BD Biosciences) cell sorter into 384-well capture plates containing 100 nl of lysis solution, 3 pl of mineral oil, and 20 nM barcoded poly(T) reverse transcription primers. After sorting, plates were centrifuged and then snap-frozen on dry ice and stored at -80 °C. scRNA- seq libraries from sorted cells were prepared with SPID-seq scRNA-seq method as described in6. Briefly, polyadenylated mRNA from single cells sorted was barcoded during reverse transcription into cDNA and then amplified. Each plate was pooled, and cDNA was fragmented and amplified to generate Illumina sequencing-ready libraries. Each plate library was tested for library quality and DNA concentration.

[0547] Human single-cell RNA-seq data integration - In order to minimize batch effect and technical variance, the three largest publicly available studies with 10X sequencing technology from breast (GSE176078 - 130,246 single cells from 26 patients tumor samples with estimated normal-like cells with inferCNV), colon (GSE178341 - 371,223 single cells from 62 patients with adjacent normal and tumor samples), and lung (GSE154826 - 361,929 cells from 35 patients with adjacent normal and tumor samples) were collected. First, each dataset was preprocessed using scanpy with a minimum threshold of 500 UMI per cell before concatenating them to a unified gene space. After filtering, a total of 832,107 single cells for the pan-cancer integration atlas were obtained (Figure 5A). Following minimal filtering of cell-type unrelated genes such as mitochondrial genes (MT-), the top 10,000 highly variable genes were used for integration. The scVI was used to perform the integration on the patient level (using each patient as a batch key) and trained with 40 epochs. Then unsupervised graph-based clustering was applied to identify the major cell lineages with conventional markers (e.g., CD7 - T / NK, KRT18 - Epithelial cells, JCHAIN - B cells, C1QA, APOE - Myeloid cells). To construct the MonMac atlas, we further extracted the monocytes and macrophages from the myeloid cluster and filtered out dendritic cell populations. After filtering for cells with less than 5% mitochondrial reads and within 20000 UMI for total count, we have a total of 108,951 single cells for integration. We used the raw count of the MonMac atlas and re-integrated with scVI using the top 8,000 highly variable genes for integration and re-trained our model with 200 epochs using patients as the batch variable. We used the sc VI embedding to construct a neighborhood graph with the correlation metric and subsequently applied graph-based clustering to annotated different subsets of monocytes and TAMs with conventional markers (e.g., FCGR3A - CD16 Mon, S100A8 / 9 and VCAN - CD14 Mon, C1QC / APOE / CTSL / GPNMB / TREM2 - TAMs, MARCO and PPARG - Alveolar Macrophages).

[0548] Module analysis using hotspot - Although conventional clustering approaches can identify different subsets of cell populations, this is highly dependent on the clustering parameters as well as the latent representation of the data. In addition, highly homogeneous or plastic populations are extremely hard to separate based on clustering alone. Thus, it was proposed to define TAM gene programs with hotspot analysis26on the integrated scVI latent space. Conceptually, the top “local autocorrelated” genes for which the expression can be highly predictable by the neighborhood cells in the similarity graph constructed from the scVI latent space were select. Then, significantly autocorrelated genes were grouped into modules based on their co-expression in their neighborhood of cells. Since TAM has high plasticity, the approach is ideal for delineating the different TAM programs. The TAM modules were separated by assessing the heatmaps (Figures ID and 8B) and each program was annotated by the genes in each program.

[0549] Deriving gene module enrichment score - In order to calculate the anti-tumor enrichment score based on the defined gene programs, two different approaches were employed. The first approach used the AUCell package in R to score each gene program to derive an enrichment score for each single cell. In Figure IE, an additional dataset was included for the breast FOLR2+ TRM28, and thus an enrichment score was derived for the additional cells as well as cells from the MonMac atlas. Healthy colon TRM and healthy lung TRM were defined as macrophages originating from adjacent normal tissue and the conserved TAM population across tissues as TAM originating from tumor tissue. Following, the average enrichment in each of the populations was calculated and a z-score normalization per gene program was applied. For the mouse in-vitro culture data (Figure 1G), human genes were first converted to mouse genes using the human-mouse orthologs and the same z-score normalization per BMDM culture condition was applied. In order to summarize the gene programs to a more comprehensible score, directionality was added to the gene programs. Given a z-score enrichment per gene module denoted in group £ and module / , and an enrichment score was defined as follows: where the positive modules were defined to be Modules 4, 6, 8 and the negative modules to be Modules 1, 3, 5, and 7 to show the polarization of the culture conditions (Figure 1H). The second approach is simpler and was employed in Figures 2A-G. The enrichment z-score of each cell in each gene program was derived directly from the hotspot module analysis, and this was utilized to calculate the average enrichment in each of the knockout conditions and project the enrichment score onto the distance map (Figure 2D). Then, to summarize the effect, the positive modules were defined to be the good TAM effects (IFN -response, antigen presentation, and complement activation) and the negative modules to be the bad TAM effects (phagocytosis I+II and immune suppression I+II), and the enrichment score was derived for each of the knockout.

[0550] Read alignment - The scRNA-seq libraries were pooled at equimolar concentrations and sequenced on an Illumina NovaSeq 6000 sequencer with a sequencing depth ranging from 10k to 50k reads per cell. Reads were condensed into original molecules by counting the same unique molecular identifiers (UMI). It was ensured that the batches for analysis showed a low-level cross single-cell contamination (less than 3 %) by statistics on the detected spurious UMI in empty wells. Alignment of reads was done using the MARS-seq2.0 pipeline64,65. In short, reads were filtered for low-quality reads and subsequently mapped to the mouse reference genome mmlO using HIS AT (version 0.1.6), excluding reads with multiple mapping positions. The UCSC genome browser was used as a reference to assign exonic reads to genes. Cell UMI uniqueness was tested for 3kb aria. In cases where exons of different genes shared a genomic position on the same strand, reads were considered as a single gene with a concatenated gene symbol. scRNA-seq analysis of BMDM cultures across cytokine stimulation conditions - To investigate which polarizing cytokine conditions (IL-4, IL-10, IL-13, IFN-y, M-CSF, and TGF-P) best induce TAM-like phenotypes in BMDM, scRNA-seq data of BMDMs across conditions were preprocessed using scanpy66and analyzed with MrVI19. Raw count matrices of each plate were concatenated into a single anndata object. Cells were filtered using scanpy to contain between 150 - 4000 total counts per cell and less than 8 % mitochondrial counts. Counts were normalized to the number of total counts per cell, and log Ip transformed for visualization purposes. The 5000 most highly variable genes were determined based on the normalized variance of the raw counts using the seurat v3 implementation in scanpy. A MrVI model was trained with the cytokine condition as sample key and using the following model hyperparameters: n_latent = 30, n_latent_u = 5, qz_nn_flavor = ‘attention’, px_nn_flavor = ‘attention’, learn_z_u_prior_scale = False, z_u_prior = False, u_prior_mixture = True, u_prior_mixture_k = 20, the qz keyword arguments (use_map = True, stop_gradients = False, stop_gradients_mlp = True, dropout_rate = 0.03) and the px keyword arguments (stop_gradients = False, stop_gradients_mlp = True, dropout_rate = 0.03, h_activation = ‘nn.softmax’, low_dim_batch = True). The model has been trained using a batch size of 256, a learning rate of 3 x 10’3, a kl warmup of 50 epochs and an early stopping based on the elbo_validartion using a patience of 30. The trained model was used to compute a neighbor graph and a subsequent UMAP in scanpy using either the sample unaware u latent space or the sample aware z latent space. The aggregated local sample distances for each cytokine condition were computed and visualized in a heatmap using seaborn. The candidate genes obtained from the MonMac atlas were visualized in a heatmap aggregated by cytokine condition using scanpy.

[0551] Preprocessing of scRNA seq data from the CRISPR screen - Combinatorial index sort flow cytometry standard (FCS) files of the BMDM CRISPR screen were retrieved from FACSymphony™ S6 sorters (BD). Data transformation was performed by applying the hyperbolic arcsinh function. Scales were further transformed with the matrixStats package to normalize all marker index expressions between 0 and 1 with low (1 %) and high (99 %) percentile limits. Combined flow cytometry index sort data were visualized using UMAP embedding and clustered using FlowSOM with default settings and 100 self-organizing map nodes as described in67. Cells were meta-clustered with k=45 using the ConsensusClusterPlus package, merged, and annotated according to their well-index marker combination as described in68. Heatmap with a median expression of each index marker over the FlowSOM metaclusters were plotted using the pheatmap package (1.0.12). Proportions of each KO replicate per 12-well plate pool were depicted as bar plots using the ggplot2 package (3.4.2). Barcode and well-index sort annotations were added to the single-cell flow cytometry dataset to integrate with the corresponding scRNA-seq profiles for demultiplexing the arrayed CRISPR perturbation screen. First, specific mitochondrial genes, immunoglobulin genes, ribosomal genes, and genes linked with poorly supported transcriptional models (such as genes with the suffix “Rik” and so on) were removed, and the gene from downstream analysis if the total gene count was <10 across all cells were further removed. Then the cell was filtered if it had a total UMI of less than 300 and the function ‘isOutlier’ with the parameters (nmads = 2 and log=T) from the scatter R package was used to filter out outlier cells. After filtering and quality control, 85,333 single cells were obtained from a total of 120 perturbations, including non-targeting controls. scRNA seq analysis of draining lymph node and tumor in vivo model - Index-sorted, plate-based scRNA-seq analysis of intratumoral immune cells was performed using the Seurat R package 469. Cells were filtered with a 300 UMI cutoff and with less than 15 % mitochondrial genes, 8 % ribosomal genes, 5 % hemoglobin genes. After filtering based on UMI and gene percent expression, specific mitochondrial, ribosomal, hemoglobin, immunoglobulin, and genes, such as pseudogenes, were further removed. Genes with a total count < 3 across all cells were also removed. Afterwards, counts were log-normalized counts and highly variable genes were selected using the FindVariableFeatures function. Seurat-based linear-transformation scaling was applied prior to linear dimension reduction using PCA (RunPCA). KNN was computed based on the euclidean distance in PCA using FindNeighbors (Seurat), and graph-based clustering was performed using the Louvain algorithm of the FindClusters function (Seurat) with default settings. Single-cell clusters were visualized using UMAP or diffusion map, and differential gene markers were computed using the FindAllMarkers function with logFC=0.25 and min.pct=0.1 threshold. CD45+ sort-enriched tumor myeloid cells, including monocytes, MoMacs, and TAMs, were analyzed from the earlier endpoint analysis, while lymphocytes were investigated from CD45+ CD90+ enriched index-sort at the later endpoint analysis. Scaled mean expression of cluster-specific DEG markers were depicted in heatmaps using the pheatmap package. The enrichment z-score of the modules, as described above, was calculated for each cell within the TAM cluster to derive the anti-tumor score. TAM-specific or CD8 T cell-specific DEGs between CpG-siRNAZeb2and CpG-siRNACtrlwere depicted using Enhancedvolcano package (1.13.2) and used for downstream Gene Ontology (GO) enrichment analysis using the clusterProfiler package (3.18.1) with the Genome wide annotation for Mouse package org.Mm.eg.db (3.18). Proportions of CD8 T cell and Treg clusters within CD45+cells per individual tumor / mouse were used to show CD8 / Treg ratio across all conditions. Tumor monocyte and macrophage trajectory from CpG-siRNAZeb2and CpG-siRNACtrl-treated tumors were visualized with diffusion maps (destiny package, 3.4.0) using highly variable genes and a number of principal components n_pcs=50. Zman-seq geneset-based cTET values, derived from the AUC distribution, were calculated for each cell and the average for each monocyte and macrophage cluster (Figure 1 IK). Index-sorted, plate-based scRNA-seq analysis of tdLN-enriched T cells was performed using the Seurat pipeline (as described above). Flow cytometry sort index data from tdLN cells were transformed using the hyperbolic arcsinh function and scaled to normalize all marker index expressions between 0 and 1 as described above. Flow cytometry index data was visualized using UMAP embedding and clustered using FlowSOM. Cells were meta-clustered using the ConsensusClusterPlus package and annotated according to their lineage / subset marker expression. Heatmap with the median expression of each surface marker over the merged clusters was plotted using pheatmap package (Figure 1 IF). Flow cytometry sort index protein expression and FlowSOM cluster annotation for tdLN single-cells were matched and merged with their corresponding scRNA-seq profiles. Mulitomic paired tdLN single-cells were visualized using UMAP embedding of both index protein and RNA expression. CD8+ and CD4+ naive, effector, and central memory T cell subsets were defined by index protein expression without including tdLN Treg scRNA-seq cluster. Scaled average expression of selected naive, effector, and memory genes, as well as median fluorescence intensity for functional surface protein markers, were depicted in heatmaps for each tdLN CD8+and CD4+T cell subset per condition.

[0552] Assembly of a functional gene perturbation network in TAMs - To construct a network that represents the topology of the effect of each gene knockout in relation to all other perturbations, variational inference was utilized using MrVI (Figure 7A). Perturbations with less than 50 viable cells per knockout were excluded. Data from the remaining cells was normalized to the total counts per cell, and log Ip transformed using scanpy. The 5000 most highly variable genes were determined based on the normalized variance of the raw count layer using the scanpy implementation of seurat v3’s highly variable gene selection70. A MrVI model was trained with the respective gene knockout as sample key and using the following model hyperparameters: n_latent = 30, n_latent_u = 5, qz_nn_flavor = ‘attention’, px_nn_flavor = ‘attention’, learn_z_u_prior_scale = False, z_u_prior = False, u_prior_mixture = True, u_prior_mixture_k = 20, the qz keyword arguments (use_map = True, stop_gradients = False, stop_gradients_mlp = True, dropout_rate = 0.03) and the px keyword arguments (stop_gradients = False, stop_gradients_mlp = True, dropout_rate = 0.03, h_activation = ‘nn.softmax’, low_dim_batch = True). The model has been trained using a batch size of 1024, a learning rate of 3 x 10’3, a kl warmup of 50 epochs and an early stopping based on the elbo_validation using a patience of 30. The perturbation unaware latent space u and the perturbation aware latent space z were visualized by neighbor graph construction, followed by UMAP using scanpy. Local sample distances of the respective knockouts have been computed using MrVI, and the corresponding knockout- knockout distance matrix was subjected to hierarchical clustering using Ward's method in scipy. This yielded seven distinct clusters of perturbation effects. To embed the knockout-knockout distance matrix in a graphical network, a multidimensional scaling using igraph was employed. To this end, each knockout represented a node in the network and was connected to its five nearest neighbors in the distance matrix, which served as edges in the graph. To identify cell states that are the most variable across perturbation conditions, a PCA was computed on the MrVI single-cell knockout-knockout distance array and the first principal components (PCs) of sample variances were displayed on the embedding of the sample aware z latent space using scanpy. The empirical cumulative distribution function of the first PCs was computed for each knockout and visualized with line plots using seaborn to assess the extent to which single cells are affected by the respective perturbation.

[0553] Downstream analysis of the CRISPR screen - Differentially expressed genes between the functional gene knockout clusters of the gene perturbation network and non-targeting controls have been identified using a Mann- Whitney U test with a false-discovery correction according to the Benjamini-Hochberg approach. Key transcription factors for each cluster have been identified by converting murine genes to their human orthologs using mousipy. Transcription factor activities have been inferred using a univariate linear model in decoupleR and ranked in relation to the non-targeting controls using a t-test that overestimates the variance in each group71. Pathway activities have been inferred by performing an over-representation analysis of the MSigDB hallmark geneset collection using decoupleR. Gene expression, transcription factor activity, and pathway activity have been displayed using seaborn’s cluster-map. Gene modules were derived using hotspot as outlined above. The fold change in module activities for each knockout compared to the non-targeting controls has been visualized on the network graph using igraph. To overlay the CXCL9 signature13on the gene perturbation network, the 297 genes displaying a positive correlation coefficient with the CXCL9 polarization signature have been retrieved. The resulting geneset has been scored for geneset average activity across all knockouts and visualized on the gene perturbation network using igraph. Pseudo-bulk differential gene expression analysis between Zeb2 knockout samples and non-targeting control samples has been performed by aggregating the counts across cells for each biological replicate. Aggregated counts have been filtered to have at least ten counts across three biological samples. Resulting count matrices have been utilized to compute differentially expressed genes in DESeq247applying default parameters. To confirm the in-vitro findings of Zeb2 KO of the CRISPR screen in an in-vivo setting, a public scRNA-seq data from spleen macrophages obtained from Fcgrl- CRExZeb2fl / fl and Zeb2fl / fl littermate control mice (Scott, C.L. et al. Immunity, 2018, 49, 312— 325) was utilized. First scVI was used to derive the latent space and the data was visualized in UMAP, recapitulating the CRE+ (KO) and CRE- (WT) splenic macrophage clusters as shown in the original publication. Then, the upregulated differentially expressed geneset (p < 0.05) from the Zeb2 KO screen data was used to calculate an average geneset activity. Subsequently, the 70th percentile from the upregulation activity was taken as a threshold to derive an enrichment density using the scanpy function sc.tl.embedding_density for visualization. Differential gene expression of splenic macrophages from Fcgrl-CRExZeb2fl / fl and Zeb2fl / fl littermate control mice was assessed using a Mann-Whitney U test employing a false discovery correction according to the Benjamini-Hochberg approach. To compare the differentially expressed genes of conditional gene ablation to the gene modules derived from the knockout screen the Jaccard similarity between up- or downregulated genes and the genes present in the gene modules were computed. The Jaccard similarity between up- and downregulated genes in splenic macrophages from Fcgrl-CRExZeb2fl / fl (CRE+) verses Zeb2fl / fl (CRE-) was computed separately and the difference in Jaccard similarity between up- and downregulated genes and the respective modules was visualized in a Chord plot using the circlize package.

[0554] Survival analysis using TCGA - To investigate the relationship between Zeb2 gene expression and patient survival across human cancers, the TCGA PanCan atlas72was queried. Gene expression and survival data of the TCGA PanCan atlas have been accessed using the UCSC Xena platform73. Patients have been classified into Zeb2 and / or high / low based on the 50thpercent quantile of the indicated cancer entity. To assess survival differences associated with Zeb2 among macrophage-rich samples, patients were grouped into four categories according to their Boolean classification of CD68 and Zeb2 expression based on the 50thpercentile. Accordingly, the survival of the CD68hlghZeb2lowgroup was compared to that of the CD68hlghZeb2hlghgroup. Kaplan-Meier curves have been drawn using the survival and survminer packages. Differences in patient survival based on gene expression have been assessed using a log-rank test.

[0555] Quantification of cTET from Zman-seq - In order to quantify and derive the TAM trajectory from the in-vivo mouse tumors, an approach to estimate the continuous tumor exposure time (cTET) from the Zman-seq paper was developed. The gene sets from the TAM trajectory of monocytes (231 genes) to inflammatory TAMs (442 genes) to immunosuppressive TAMs (168 genes) were utilized. For each single cell, an enrichment score was first derived for each of the three genesets using AUCell. Then, using the timestamps (Mon - 12 to 24 hours, Inf TAMs - 24 to 36 hours, and Supp TAMs - 36 to 48 hours) from the Zman-seq trajectory as anchors, the cumulative distribution of each geneset was computed to construct an AUC curve per single cell. Then, the area under the curve was calculated and the reverse of the normalized area (as a larger area indicates an earlier timepoint and a smaller area indicates a later timepoint) was taken to derive the cTET value per cell as previously described53(Figure 11M). Tumor monocyte and macrophage trajectory from CpG-siRNAZeb2and CpG-siRNACtrl-treated tumors were then visualized with diffusion maps (destiny package, 3.4.0) using highly variable genes and the number of top principal components (n_pcs=50). The cTET values were visualized on the diffusion map and the average cTET per monocyte and macrophage cluster for the trajectory was shown (Figure UK). To quantitatively compare the TAM trajectory across the different conditions (PBS, CpG, and CpG_Zeb2), a bootstrapping approach was applied to ensure a fair comparison across the different mice. 100 cells per mouse were sampled repeatedly with 500 resamples and the average cTET in the myeloid population was calculated. Following, a histogram was then visualized accordingly to show the relative distribution of each treatment condition on the Zman-seq trajectory (Figure 1 IE).

[0556] AT AC processing and analysis - For low-level pre-processing, the Homer package74was used to call for peaks with the command (findPeaks -style factor) and subsequently peaks were merged with the command (mergePeaks -d 150) to curate a set of common peaks across perturbations and control for downstream analysis. Then, DESeq2 was used to normalize across the technical duplicates of the 15 knockouts and test for statistically significant differential peaks for each knockout against the non-targeting control. The differential peaks (with a log2 fold change > 0.5 and a base mean > 50 reads) in the Zeb2 KO were then tested for enriched TF- binding motifs using the algorithm findMotifGenome from the Homer package74.

[0557] Motif enrichment analysis of AT AC enhancers - For the GimmeMotifs TF motif enrichment analysis, the uniquely opened differential enhancer peaks from the Zeb2 KO that are within close proximity of the genes from the positive modules (210 peaks from IFN-response, antigen presentation, and complement activation) and closed differential enhancer peaks from the negative modules (219 peaks from phagocytosis and immune suppression modules) were used for the downstream motif enrichment analysis. Since flanking sequences of known motifs are heterogeneous across cell types, a consensus database of known TFs (gimme.vertebrate. v.5.0) was used as implemented in the GimmeMotifs package (v.0.17.2)75. The maelstrom function was applied to calculate the differential enrichment of motifs between Zeb2 and the control with default parameters, reducing or extending peaks to a uniform 200 bp around the peak summit. Motif activity was defined as positive or negative enrichment of -logio(p) for high or low ranking of the motif occurrence. For heatmap visualization, selected TFs expressed in the Zeb2 KO are shown.

[0558] 2D visualization on AT AC enhancer landscape - In order to visualize the enhancer landscape across the KO, the average reads were pooled across the duplicates in each KO and the log2 of the reads was taken for each peak. Then, 6636 peaks, which had a dispersion ratio of at least 0.75 (disp = mean(x) / var(x)), were selected for the construction of the 2D map. The spearman’s rho correlation metric was used to determine a n by n KO distance matrix and subsequently a principal coordinate analysis (PCoA) was applied on the distance matrix to generate the 2D map. The 5 nearest neighbor KO on the graph were connected with gray lines, and the same coloring scheme of the clusters of KO as defined in Figure 2C was used for visualization.

[0559] Differential gene expression and pathway conservation analysis of ZEB2 KO in human CD14-derived macrophages - Index-sorted, plate-based scRNA-seq data has been analyzed using scanpy.72 Raw count matrices of each plate were concatenated into a single anndata object. Cells with less than 150 unique genes or more than 14 % mitochondrial counts have been removed. Counts were normalized to the number of total counts per cell and log Ip transformed. Differential gene expression between ZEB2 KO macrophages and control macrophages has been tested using a Mann- Whitney U test employing a Benjamini-Hochberg correction. To evaluate whether pathways affected by Zeb2 KO were conserved between human and mouse macrophages, the differentially expressed genes of ZEB2 KO in human macrophages were compared to Zeb2 KO in murine macrophages. For this, murine gene names were converted to human orthologs using the mousipy library. Differentially expressed genes for both species were separately subjected to an overrepresentation analysis of pathways in the MSigDB hallmark geneset collection using decoupleR (Badia-i-Mompel, P., elez Santiago, J.V., Braunger, J., Geiss, C., Dimitrov, D., Mulle-Dott, S., Taus, P., Dugourd, A., Holland, C.H., Ramirez Flores, R.O., et al.. Gene regulation decoupleR: ensemble of computational methods to infer biological activities from omics data). Overrepresentation was tested using a one-tailed Fisher’s exact test and concomitant Benjamini-Hochberg correction for each species separately. The pathway overrepresentation results for each species were visualized in 2D scatter plots.

[0560] Quantification and statistical analysis - Data analyses were performed as indicated in the relevant Methods sections or by using GraphPad Prism (GraphPad Software). For in-vivo experiments, no statistical method was used to predetermine sample size; mouse numbers were determined based on the results of preliminary experiments. To detect outliers, the Robust Outliers Detection (ROUT) method (Q = 0.5%) in GraphPad was used. Two outliers were detected and removed from the experiment in Figure 4E (left) and one in Figure 18. For each dataset, normal distribution was confirmed using Anderson-Darling, Shapiro-Wilk, D’Agostino- Pearson, and Kolmogorov-Smirnov tests. For normal distributions, one-way analysis of variance (ANOVA) with Tukey’s post hoc test was used to compare three or more groups. When two groups were compared, an unpaired two-tailed student’s t-test (two-tailed, unequal variance) was used to determine statistical significance. When data was not normally distributed, a nonparametric test was used, either Kruskal-Wallis with Dunn’s post hoc test for multiple comparisons or the Wilcoxon test when two groups were compared. In repeated measurements of in-vivo tumor volume measurement, two-way ANOVA or mixed-effects models were used to detect statistical significance with Tukey’s post hoc test. Statistical significance indicated in all figures is as follows: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns: non-significant.

[0561] EXAMPLE 1

[0562] UNIFIED ATLAS OF TUMOR-ASSOCIATED MACROPHAGES (TAMs) PROGRAMS

[0563] In order to define the common programs of human TAMs, the present inventors first collected large publicly available patient cohorts of scRNA-seq datasets from human solid tumor tissues, including breast, lung, and colon cancers, as well as their adjacent healthy tissues (Figure 5A)2U23. After quality control and filtering, they used scVI24,25to integrate 832,107 single cells from 123 patients and annotated the major cell types using unsupervised clustering and well- established gene markers. (Figures 1A and 5B). To identify TAM-specific gene programs, the inventors focused on monocytes and macrophages and re-integrated 108,951 cells from the myeloid compartment to construct a comprehensive MonMac atlas with scVI, revealing diverse myeloid subsets by unsupervised graph-based clustering (Figures IB and 5C). This analysis identified five major clusters of human TAMs (I-V) alongside clusters of tissue macrophages and monocytes based on the enrichment density of the biopsies. Three monocyte clusters were defined characterized as CD14+classical and CD16+non-classical monocytes alongside intermediate monocytes, predominantly enriched in healthy tissues. While the monocyte subsets have distinct gene expression profiles, several TAM subsets overlap in their gene expression profiles among themselves and with healthy tissue-resident macrophages (TRMs), posing challenges in identifying TAM-specific gene programs (Figure 1C). To resolve this, the present inventors employed gene module analysis using hotspot26to identify distinct gene programs influenced by the TME signaling, enabling the identification of TAM-specific programs despite the overlap with healthy TRMs. This analysis revealed nine distinct gene modules across the MonMac atlas. Module 1 (Mod 1) was enriched with genes associated with the complement system (e.g., C1QA / B / C and C2) and antigen presentation (e.g., HLA-DMB, HIA-DQAF) together with canonical human alveolar macrophages markers (e.g., PPARG, MARCO, CD9, and MRC1), defining this module as an alveolar macrophage program27. Modules 5 and 7 were characterized by expression of genes associated with TRM (e.g., FOLR2, MAF, SLC40A1, LYVE1), annotated as TRM I and II28,29. Module 9 was characterized by monocyte markers (e.g., S100A8 / 9 / 12, FCN1). The rest of the modules include gene programs previously described in the context of TAMs, including IFN response (Mod3), phagocytosis (Mod4), immune suppression (Mod6), and angiogenesis (Mod8)16(Figure ID). In order to determine TAM-specific gene programs and distinguish them from healthy TRM programs, we used the gene programs to calculate the enrichment of healthy TRMs across breast, colon, and lung in comparison to TAMs in each of the modules (Figure IE). Modules 1, 2, 5, and 7 were more specific to healthy TRM, while modules 3, 4, 6, and 8 were more enriched in TAMs. To define the potential regulatory circuits underlying these TAM-specific programs functionally, the expression profiles and molecular functions were used to compile a list of candidate TAM regulators, prioritizing genes with transcription regulation and signaling functions.

[0564] To date, functional annotation of the regulatory factors controlling human TAMs has been limited due to their heterogeneity and the lack of common markers defining their activity. The present inventors assessed if the regulatory factors from their defined human TAM programs were conserved across different mouse tumor models. Therefore, multiple syngeneic mouse tumor models (MCA205, MC38, and B16) were collected and a mouse MonMac atlas with monocytes, TAMs, and TRM was constructed (Figures 5D-E). Using conserved human-mouse orthologs, the enrichment of the gene modules in the TAM population over monocytes and TRM was compared in both humans and mice. A high conservation of the genes was observed, with 83.9 % of overlapping candidate regulators from the human TAM programs being conserved in both human and mouse TAM (Figure IF). Overall, this analysis identified conserved TAM programs and their potential candidate regulators.

[0565] EXAMPLE 2

[0566] SINGLE-CELL ARRAYED CRISPR SCREEN OF TAM-PROGRAMMED BMDMs

[0567] The present inventors set out to establish a CRISPR knockout (KO) screening platform in a primary murine model of TAM -programmed cells ex-vivo. To this end, bone marrow cells isolated from CRISPR-Cas9 knock-in mice30were cultured in the presence of macrophage- colony stimulating factor (M-CSF) for seven days to yield bone-marrow-derived macrophages (BMDMs). To identify the relevant signals that activate the TAM programs, conditioning signals mimicking the TAM polarization in the TME were screened for. BMDMs were conditioned using different dominant TME signaling cytokines (IL-4, IL- 10, IL- 13, TGF-0, and IFN-y), known to induce macrophage polarization to T AM-like cells31, and these conditions were compared to previously defined TAM gene programs (Figure 5F). IL-4, the archetypical stimulation for the M2-phenotype, robustly activated the TAM gene programs, exhibiting the highest enrichment score for immunosuppressive TAM programs (Figures 1G-H), in agreement with the role of IL-4 in TAM differentiation both in-vitro and zn-vzvo32,33. Additionally, the candidate TAM regulatory genes were also highly enriched in the IL-4 TAM-programmed BMDMs (Figure 5G). Thus, IL-4 was selected to induce TAM-programmed BMDMs in the screening platform.

[0568] A lentiGuide-puro approach was used to transduce the CRISPR-Cas9 knock-in TAM- programmed BMDMs, followed by plate-based scRNA-seq profiling (Figures 6A-B). To evaluate the robustness of the screening procedure, two positive control perturbations were tested: targeting Stat6, the main transcription factor (TF) downstream to IL-4 signaling; and Bachl, a transcriptional repressor that negatively regulates macrophage tissue repair and TAM program through 7 / moxJ34,35. Statd-perturbed BMDMs failed to induce IL-4-dependent downstream genes such as Argl and Retnla (Figure 6C). Accordingly, Bachl perturbation resulted in enhanced expression of Hmoxl and other TAM suppressive genes, such as Sppl and Mmpl2, confirming previous findings regarding the Bachl-Hmoxl regulatory axis (Figure 6C). Thus, these findings rationalize a systematic scRNA-seq screening approach for characterizing the circuitry regulating TAM programs. To ensure consistency and efficiency in the experimental perturbations, a combined scRNA-seq and index- sorting-based cell barcoding strategy was developed (Figures 6A and 6D). This approach standardized all steps from virus preparation to cell seeding, transduction, and selection across all perturbations, aiming for uniform cell numbers in each knockout assessed. Among the 122 candidate genes, only KOs of Csfl r and Spil significantly reduced cell survival, confirming their essential roles in macrophage survival and differentiation, respectively36,37. This indicated that the rest of the evaluated gene KOs did not interfere with macrophage survival (Figure 6E). In addition, the corresponding gene of most gene KOs was downregulated compared to the non-targeting control, confirming the perturbation efficiency (Figure 6F). Collectively, the inventors established a CAS9-based arrayed CRISPR screen of primary murine TAM-programmed cells and systematically perturbed a comprehensive set of novel and established TAM regulators, resulting in QC-controlled and filtered 85,533 scRNA-seq profiles of specific KO of TAM-programmed BMDMs.

[0569] EXAMPLE 3

[0570] DEEP GENERATIVE MODELING DEFINES THE FUNCTIONAL TAM REGULATORY NETWORK

[0571] In order to delineate the effect of individual candidate regulators on functional macrophage programs, MrVI, a deep generative model capable of dissecting the effect of sample- level covariates, such as sex, age, or gene perturbations, on single-cell embeddings19was utilized. Using variational inference, Mr VI encodes gene expression of individual cells into a latent space u, which corresponds to the macrophage cell state and is designed to be decoupled from the respective gene perturbation (Figure 7A). Additionally, a second latent space z is learned, which corresponds to the perturbed macrophage cell state while accounting for the genetic perturbation. The mapping between the perturbation-agnostic and perturbation-dependent spaces also allowed estimating, for each cell, a distance measure between each pair of perturbations (comparing their respective effects on the cell; Figure 7A). Analyzing the first principal components among these pairwise local sample distances demonstrated that some candidate gene KO, such as Ythdfl, YthdjZ, Stat6, and ZebZ. stood out by high variation in a large fraction of cells within the respective KO, indicating that their perturbation effect was unique and pronounced in the majority of the cells analyzed (Figures 7B-C). By averaging these local pairwise distances across all cells and embedding the resulting distance matrix using multidimensional scaling (MDS), a prototypical network capturing the functional similarity for interrogated TAM regulators was constructed (Figures 2A-B). In this network, candidate gene KOs that evoke a similar effect on the single-cell states are located in close proximity to each other, while perturbation of putative regulators leading to distinct or even opposing macrophage state polarizations are embedded far from each other (Figure 2B). Hierarchical clustering of the distance matrix between the different KOs yielded seven clusters of prototypical responses to genetic perturbation in macrophages with distinct gene expression patterns and transcription factor activity (Figures 2A and 7D). The regulatory clusters obtained from this gene perturbation network could be utilized to reveal the functional circuits hardwiring macrophages in health and disease.

[0572] To decompose gene expression dynamics into functional TAM gene programs in a data- driven manner, the inventors used hotspot to group genes based on their autocorrelation utilizing the MrVI perturbation-aware latent space z. This yielded ten distinct functional gene programs that play important roles in TAM biology and share similarities with human TAM programs (Figures 2C and 8A-B). Notably, the distinct clusters of perturbation responses encompassed regulation across all major TAM biology modules ranging from antigen presentation to immunosuppression, thereby highlighting the functional modularity of macrophage programs (Figures 2D and 8C-D). For instance, the complement activation gene program was selectively induced in perturbations of regulatory clusters, which included Stat6 and ZebZ. suggesting that members of those clusters are potentially relevant checkpoints, blocking the cellular complement machinery in TAMs. On the contrary, perturbation of Tgfbrl-Smad family members downregulated the complement program (Figures 2C-E, 8C, and 9A-C). Furthermore, the type I IFN signaling response gene program was markedly induced upon perturbation of members from the m6A readers cluster (e.g., Ythdfl, Ylhdf2' p' Stat6 cluster (e.g., Stat6, Gpnmb, SppJ) and Zeb2 cluster, highlighting the potential of blocking these targets to accelerate a robust anti-tumor response in the TME. Interestingly, the antigen presentation program was co-regulated with the IFN-response program in perturbations of the Stat6 and Zeb2 clusters but not in the Ythdf cluster, indicating that these circuitries are intricately interwoven but also have distinct regulatory mechanisms, such as the post-transcriptional Ythdf circuitry inhibiting only the IFN pathway38,39(Figures 2E and 9C).

[0573] Recent studies highlighted the pro-tumor effects induced by TAM phagocytosis and efferocytosis and how interference with their activation enhances anti-tumor immunity40,41. In the above-described gene perturbation network, two programs of phagocytosis were found, I and II, the latter being enriched in lysosomal genes and cholesterol efflux transporters complexes associated with the efferocytosis processes (Figures 2C and 8A). Importantly, non-redundant functions of the different perturbation clusters were observed in those modules, highlighting the need for data-driven approaches to effectively reprogram pro-tumor TAM functions. Particularly, perturbation of the Tgfbrl-Smad and IL4-Stat6 pathways, which are known to drive pro-tumor functions, demonstrated non-redundant activity on these different pro-tumor TAM functions, indicating the need for dual targeting to reprogram the TAM pro-tumor phenotype (Figures 8C and 9B).

[0574] In conclusion, the functional TAM regulatory network provided a detailed circuitry regulating the different TAM programs (Figure 2E). This deep functional data can guide the rational design of TAM checkpoints or combinations of treatments to selectively interfere with specific gene programs of macrophage biology.

[0575] EXAMPLE 4

[0576] ZEB2 IS A NOVEL TAM REPROGRAMMING MASTER SWITCH

[0577] Using the comprehensive functional gene perturbation network, the present inventors next aimed to identify TAM reprograming hubs that upregulate macrophage anti-tumor functions and downregulate the pro-tumor functions. Three of the prototypical macrophage programs have been shown in multiple studies to be associated with anti-tumor TAM functions, including IFN response, antigen presentation, and complement activation42 46. Four of the remaining programs have been implicated in pro-tumor TAM functions, including different immune suppression modules and phagocytosis (Figures 2C and 8A). It was rationalized that effective TAM checkpoints would reprogram pro-tumor TAM into anti-tumor cells by increasing the anti-tumor programs and suppressing the pro-tumor programs. To assess this, the relative enrichment of each gene KO to the different gene programs was calculated by averaging the enrichment score of each single cell in each of the KO (Figures 2D, 8C-D). To estimate an anti-tumor score of the different gene KOs, the difference between the average enrichment of the anti-tumor modules (IFN-response, Antigen Presentation, and Complement Activation) and the pro-tumor modules (Immune Suppression I and II and Phagocytosis I and II) was calculated and ranked over all KOs (Figure 8E). Among all candidate regulators, including canonical TAM reprogramming factors such as Ythdf2, Stat6, Trem2, and TgfbrP, Zeb2 perturbation demonstrated the highest anti-tumor reprogramming score, while most of the candidates screened displayed only mild anti-tumor activity (Figures 2F and 8E). To validate this approach and the results in a human context, a recent study that reported a TAM gene expression polarity associated with improved patients’ survival, namely CXCL9.SPP1 polarity13was utilized. Projecting this human gene polarity score on the perturbation network revealed a strikingly similar anti-tumor ranking across the different gene KOs and revealed a significant correlation (R=0.67, p < 2.2e-16), with Zeb2 emerging as the top reprograming TAM checkpoint (Figures 2F, 2G, 9A, 9D).

[0578] In summary, the unbiased screen identified a complex TAM regulatory circuit controlling diverse pro- and anti-tumor functions, with Zeb2 demonstrating the most robust molecular switch for the reprogramming from pro-tumor macrophages into anti-tumor macrophages.

[0579] EXAMPLE 5

[0580] ZEB2 LOSS REPROGRAMS THE EPIGENETIC STATE OF TAMs

[0581] Since some of the most significant regulators of the TAM programs were TFs and chromatin modifiers, the inventors aimed to better understand whether they also reprogram the TAM epigenetic regulatory landscape. To this end, the top 15 relevant candidates that showed the most significant effects on gene expression levels we selected and their perturbation effects on the open chromatin regions were evaluated by ATAC-seq profiling of TAM-programmed BMDMs. After peak detection, raw peak counts were normalized using DESeq247across all samples and differential accessible peaks across the different KOs from the non-targeting control were computed. Limited changes in differential enhancer regions in most of the selected KOs were detected, except for Zeb2. Stat6, Mafb, and Cuxl, which displayed significant chromatin perturbation compared to control (Figures 3A-B and 10A). Stat6 and Mafb perturbation showed significant effects on the regulatory enhancer landscape, aligning with their pivotal roles in IL-4 signaling and macrophage differentiation, respectively35,48. Notably, similar to its major regulatory role on key TAM programs, Zeb2 KO demonstrated the most significant reprograming effects on the enhancer landscape, with 27,559 differential enhancers, including a dramatic shift in the promoter landscape compared to other regulatory factors KO and non-targeting controls (Figures 3B, 10B and 10D).

[0582] Subsequently, the inventors analyzed differential enhancer regions at loci corresponding to the defined TAM gene programs in Zeb2 KO. A highly enriched remodeling from closed to open enhancer regions in anti-tumor modules such as IFN response, antigen presentation, and complement activation was found (Figure 3C). To identify potential transcription factor cooperation in Zeb2 -regulated enhancers, the inventors searched for enrichment or depletion of TF motif co-occupancy. Significantly overrepresented motifs of inflammatory TFs like SlalE Nfkb, and Irfl / 9 in the differential regions of anti-tumor modules in Zeb2 KO were observed (Figures 3D). Conversely, motifs associated with IL-4 signaling and pro-tumor activity, such as Stat6, Bachl, and AP-1, were repressed49(Figures 3E and 10C). Interestingly, the results indicated a negative feedback motif between Zeb2 and Zebl previously described in hematopoietic stem and progenitor cells50,51. Upon Zeb2 KO, the DNA motif for Zebl was highly enriched in gained differential enhancer regions in anti-tumor gene modules (Figure 3D). Moreover, promoter activity and expression levels of Zebl and the regulated anti-tumor genes were significantly upregulated upon Zeb2 KO, indicating the repressive role of Zeb2 towards Zebl and the regulated genes (Figures 3F and 10D). Among the established pro-tumor genes regulated by Stat6 activity, such as Argl and Mrcl, a potential direct regulation of Zeb2 by Stat6 was observed. Specifically, a differential enhancer with a Stat6 binding motif near the Zeb2 locus was repressed in Stat6 KO (Figure 10E). This enhancer was previously reported as a Stat6 binding site driving transcription via RNAPII-pS5 binding in IL-4- stimulated BMDMs35. In line with this, a significant Zeb2 downregulation was observed in Stat6 KO (Figure 10F). This suggests that Zeb2 is one of the downstream targets of IL4 signaling through Stat6 activity, critical in macrophage anti-tumor activity32.

[0583] In summary, the chromatin perturbation analysis uncovered Zeb2 as a master reprograming factor of the TAM epigenetic landscape.

[0584] EXAMPLE 6 ZEB2 IS A TAM CHECKPOINT INHIBITING T CELL ACTIVATION AND ITS LOSS REPROGRAMS TAM FUNCIONALITY

[0585] In the human TAM gene module analysis of the MonMac atlas, ZEB2 was found to be enriched within the pro-tumor angiogenesis program defined by the expression of genes such as SPP1, VEGFA, ANGPLT4, and THBS1 (Figure ID). This gene signature highly resembles the SPP1+TAMs population previously reported to be correlated with poor prognosis of colorectal cancer patients43. In line with these findings, this TAM populatio...

Claims

WHAT IS CLAIMED IS:

1. A composition comprising an agent capable of downregulating expression and / or activity of Zeb2 attached to a macrophage-targeting moiety or encapsulated in a particle comprising a macrophage-targeting moiety.

2. The composition of claim 1, wherein said agent is a polynucleotide.

3. The composition of claim 2, wherein said polynucleotide is an RNA silencing agent.

4. The composition of claim 3, wherein said RNA silencing agent is an siRNA or shRNA.

5. The composition of claim 3, wherein said RNA silencing agent is a microRNA.

6. The composition of claim 5, wherein said microRNA is selected from the group consisting of MiR-30a-5p, MiR-101, MiR-124, MiR-129, MiR-132, MiR-138, MiR-139-5p, MiR-141, MiR-144, MiR-145, MiR-145-5p, MiR-153, MiR-154, MiR-155, MiR-187, MiR-200a, MiR-200b, MiR-200c, MiR-203, MiR-205, MiR-206, MiR-211-5p, MiR-215, MiR-335, MiR- 338-3p, MiR-454-3p, MiR-374b-5p, MiR-448, MiR-506, MiR-545, MiR-598, MiR-622, MiR- 590-3p, MiR-769-3p, MiR-940, MiR-1179 and MiR-3653.

7. The composition of any one of claims 1-2, wherein said agent is a genome editing agent.

8. The composition of claim 1, wherein said agent is a small molecule.

9. The composition of claim 8, wherein said small molecule is Teniposide.

10. The composition of claim 1, wherein said agent is a PROTAC.

11. The composition of claim 10, wherein said PROTAC is SR-1114 or SPFl bxi4-\12. The composition of any one of claims 1-11, wherein said macrophage-targeting moiety is specific for a macrophage surface marker.

13. The composition of claim 12, wherein said marker is a mannose receptor or a scavenger receptor.

14. The composition of claim 12, wherein said marker is selected from the group consisting of toll-like receptor, CDl lb, CD68, CD163, TREM2, MSR1 and MRC1.

15. The composition of claim 12, wherein said marker is a toll-like receptor.

16. The composition of any one of claims 14-15, wherein said toll-like receptor is TLR9.

17. The composition of any one of claims 12-16, wherein said macrophage-targeting moiety comprises a ligand or receptor of said marker.

18. The composition of claim 17, wherein said macrophage-targeting moiety comprises mannose.

19. The composition of any one of claims 15-17, wherein said macrophage-targeting moiety comprises a CpG ODN.

20. The composition of any one of claims 12-16, wherein said macrophage-targeting moiety comprises an antibody.

21. The composition of any one of claims 1-20, wherein said agent is attached to said macrophage-targeting moiety via a linker.

22. The composition of claim 21, wherein said linker comprises a C3 spacer.

23. The composition of any one of claims 1-22, wherein said particle is a lipid-based nanoparticle.

24. A method of treating an inflammatory disease in a subject in need thereof, the method comprising administering to the subject a therapeutically effective amount of the composition of any one of claims 1-22, thereby treating the inflammatory disease in the subject.

25. The composition of any one of claims 1-22, for use in treating an inflammatory disease in a subject in need thereof.

26. The method of claim 24, further comprising administering to the subject an anti- inflammatory agent.

27. The composition for use of claim 25, further comprising an anti-inflammatory agent.

28. The method of claim 26 or the composition for use of claim 27, wherein said anti- inflammatory agent is Bacillus Calmette-Guerin (BCG).

29. The method or the composition for use of any one of claims 24-28, wherein said subject is characterized by a level of macrophages above a predetermined threshold in a biological sample obtained from said subject.

30. The method or the composition for use of claim 29, wherein said subject is characterized by a level of Zeb2 above a predetermined threshold in said biological sample.

31. The method of any one of claims 24, 26 and 28-30, further comprising determining a level of macrophages in a biological sample of said subject.

32. The method of claim 31, further comprising determining a level of Zeb2 in said biological sample.

33. A method of prognosing an inflammatory disease in a subject, the method comprising determining a level of macrophages and a level of Zeb2 a biological sample of a subject diagnosed with the inflammatory disease, wherein said level of said macrophages and said level of said Zeb2 above a predetermined threshold is indicative of poor prognosis, thereby prognosing the inflammatory disease in the subject.

34. The method of claim 33, wherein a level of macrophages positive for said Zeb2 above a predetermined threshold is indicative of poor prognosis.

35. The method of any one of claims 29-34, wherein said biological sample comprises a biopsy.

36. The method or the composition for use of any one of claims 29-35, wherein said level of macrophages is determined by expression of CD68.

37. The method or the composition for use of any one of claims 24-36, wherein said inflammatory disease is cancer.

38. The method or the composition for use of claim 37, wherein said cancer is a solid cancer.

39. The method or the composition for use of claim 37, wherein aid cancer is selected from the group consisting of bladder cancer, lung cancer, stomach cancer, acute-myeloid leukemia and, colorectal cancer.

40. The method or the composition for use of claim 37, wherein said cancer is selected from the group consisting of bladder and colon cancer.

41. The method or the composition for use of any one of claims 39-40, wherein when said cancer is bladder cancer, administration is via intravesical injection.

Images