EP4683644A1

EP4683644A1 - Fully modified mir-34a and related conjugates, compositions and methods of use

Info

Publication number: EP4683644A1
Application number: EP24775784.2A
Authority: EP
Inventors: Andrea KASINSKI; Ahmed Mansour Abdelbaky ABDELAAL; Ikjot Singh SOHAL
Original assignee: Purdue Research Foundation
Current assignee: Purdue Research Foundation
Priority date: 2023-03-23
Filing date: 2024-03-22
Publication date: 2026-01-28
Also published as: KR20260005404A; JP2026509931A; AU2024239300A1; WO2024197262A1

Abstract

A fully chemically modified microRNA (miRNA), such as miR-34a, a conjugate comprising the (i) miRNA, (ii) a targeting ligand, such as folate, DUPA, or the ligand present in PSMA-617, and (iii) optionally, a group that improves tumor uptake of the conjugate, such as a. group comprising an albumin-binding moiety; a composition comprising the miRNA/conjugate; and methods of treatin cancer using same.

Description

70078-02 FULLY MODIFIED miR-34a AND RELATED CONJUGATES, COMPOSITIONS AND METHODS OF USE CROSS-REFEFRENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Appl. No.63/454,177, filed March 23, 2023, which is incorporated by reference as if fully set forth herein. STATEMENT OF GOVERNMENT SUPPORT This invention was made with government support under CA226259 and CA205420 awarded by the National Institutes of Health and under W81XWH-21-1-0181 awarded by the Department of Defense. The government has certain rights in the invention. INCORPORATION BY REFERENCE OF SEQUENCE LISTING This application contains a Sequence Listing which has been submitted electronically in ST26 format and is hereby incorporated by reference in its entirety. Said ST26 file, created on March 22, 2024, is named “1165172WO1.xml” and is 18,581 bytes in size. TECHNICAL FIELD This disclosure relates to chemically modified microRNA (miRNA), conjugates comprising same, compositions comprising miRNA/conjugates, and methods of using same to treat cancer. BACKGROUND MicroRNA (miRNA)-based therapeutics have emerged as potential therapeutic tools for treating multiple diseases due to their unique ability to modulate potently the expression of multiple genes. Reduced miRNA stability, potential immunogenic effects related to unmodified RNAs, and the lack of safe delivery vehicles are major drawbacks associated with transitioning miRNAs into the clinic. Unmodified miRNAs are rapidly degraded by nucleases, which hinder their activity, thereby necessitating the use of high and repetitive dosing and making them incompatible for in vivo applications. Several chemical modifications have been used to stabilize WKH^51$^^LQFOXGLQJ^^ƍ-O-PHWK\O^^DQG^^ƍ-fluoro modifications to the ribose and phosphorothioate 70078-02 substitutions to the backbone. The ribose modifications improve binding affinity and provide protection against nucleases, while phosphorothioate bonds confer additional resistance to exonucleases. Nonetheless, extensive modification could impede the silencing activity of miRNAs by altering target-gene affinity and increasing the stability of the RNA duplex, making it difficult for the RNA Induced Silencing Complex (RISC) to unwind the duplex and load the active strand. Thus, it is important to design and select carefully modifications that enhance RNA stability, but at same time are suitable for target gene repression and RISC loading and processing. In case of siRNAs and antisense oligonucleotides (ASOs), chemical modifications induced effective and prolonged silencing of the targeted transcripts, ultimately reducing the therapeutic doses. However, the same benefit has yet to be harnessed for miRNA duplexes. Being short noncoding RNA, miRNAs have the unique ability to downregulate multiple genes at the same time. For example, in case of tumor suppressive miRNAs, several oncogenic pathways that control cell proliferation, migration and invasion, resistance to apoptosis, and immune evasion can all be regulated by miRNA-34a, which targets the androgen receptor (AR), C-MYC, AXL, MET, sirtuin 1 (SIRT1), CD44, programmed death ligand-1 (PDL-1), and others. Despite the great benefit achieved through targeting multiple genes simultaneously, predicting the effect of chemical modification on the pleotropic activity of miRNAs is difficult. Thus, studies that laboriously evaluate the impact of chemical modifications on miRNA stability and activity are needed to advance modified miRNA therapeutics into the clinic. Ligand-mediated delivery of siRNAs and miRNAs has been developed to achieve safe and specific targeting of cancer cells. This approach relies on the use of a targeting ligand that has high affinity and specificity to a receptor that is upregulated by the targeted cells. For example, we previously developed a folate-miRNA conjugate (FolamiR), for delivery of miR- 34a (FolamiR-34a). Systemic delivery of FolamiR-34a to tumor-bearing mice resulted in downregulation of miR-34a target genes leading to inhibition of tumor growth in vivo. Despite the observed response, the activity of FolamiR-34a is limited by entrapment within the endosomes and by degradation by various nucleases. Although inclusion of an endosomal escape moiety allowed for increased cytosolic accumulation of miR-34a, the presence of cellular nucleases reduced the half-life of miR-34a, concealing the full potential of FolamiR-34a. One way to enhance miRNA stability is through the use of fully modified nucleotides. However, the 70078-02 direct impact of full chemical modification on the activity of tumor suppressive miRNA duplexes and how these modifications affect targeting are not well understood. A few studies used fully modified miRNA; however in vivo efficacy is lacking. For example, a fully modified let-7b FRQWDLQLQJ^DOWHUQDWLQJ^^ƍ-O-PHWK\O^^DQG^^ƍ-fluoro ribose bases, and phosphorothioate linkages that was conjugated to various lipids resulted in silencing of HMGA2 mRNA. However, in vivo tumor suppressive effects attributed to let-7b were not revealed, perhaps due to lack of a specific delivery vehicle. In an additional study a single- stranded oligonucleotide that mimics the active (antisense) strand of the miR-34a duplex was generated. This chemically modified, single-stranded oligonucleotide induced similar silencing of miR-34a target genes relative to an unmodified miR-34a duplex following transfection. Whether the single-stranded oligonucleotide worked better than a corresponding duplex with the same modifications was not determined. Moreover, efficacy of the chemically modified single- stranded oligos in vivo was not assessed. Indeed, the single-stranded oligo would be subject to several barriers including degradation by nucleases before reaching the targeting site. To overcome the forementioned challenges and to expand the understanding of the effect of chemical modifications on miRNA duplexes activity, we developed a chemically modified miR- 34a duplex and compared its stability and activity to partially modified miR-34a, both in cells and in vivo, following transfection or using the clinically relevant FolamiR-conjugate. In view of the above, it is an object of the present disclosure to provide fully modified miRNA, in particular for the treatment of cancer. It is another object of the present disclosure to provide conjugates comprising the fully modified miRNA. These and other objects and advantages, as well as inventive features, will be apparent from the detailed description provided herein. SUMMARY A fully chemically modified microRNA (miRNA), wherein the miRNA is modified with ^ƍ-O-methyl, ^ƍ-fluoro ribose bases, and phosphorothioate linkages, is provided. The miRNA can be miR-34a. The sense strand of the miRNA, which can be miR-34a, can have 15 nucleotides, whereas the antisense strand of the miRNA, which can be miR-34a, can have 22 nucleotides. Each strand of the miRNA can contain an alternating pattern of 2ƍ-O-methyl- 70078-02 modified and 2ƍ-fluoro-modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. In an embodiment, the sense strand can have the sequence: /52FG/*mC*/i2FU/mA/i2FA/mG/i2FA/mC/i2FA/mC/i2FU/mG/i2FC/*mC*/ i2FA//3AzideN/ (SEQ ID NO: 4) and the antisense strand can have the sequence: 5Phos/mU*/i2FG/*mG/i2FC/mA/i2FG/mU/i2FG/mU/i2FC/mU/i2FU/mA/i2FG/mC/i2FU/mG/i2F G/mU/i2FU/mG*/32FU/ (SEQ ID NO: 5), wherein m is 2ƍ-O-methyl; F is 2ƍ-fluoro; r is ribonucleotide; i is internal; * is a phosphorothioate bond; and Phos: 5ƍ-phosphate. A conjugate comprising the miRNA and a folate is also provided. The conjugate can further comprise a group that improves tumor uptake of the conjugate, such as a group that comprises or is an albumin-binding moiety. A conjugate comprising the miRNA and DUPA or the ligand present in PSMA-617 is further provided. The conjugate can further comprise a group that improves tumor uptake of the conjugate, such as a group that comprises or is an albumin-binding moiety. Still further provided is a composition comprising the miRNA and a pharmaceutically acceptable carrier, diluent, or excipient. Even still further provided is a composition comprising the conjugate and a pharmaceutically acceptable carrier, diluent, or excipient. A method of treating cancer in a subject is also provided. The method comprises administering to the subject a cancer-treating effective amount of the miRNA, optionally as a composition comprising the miRNA and a pharmaceutically acceptable carrier, diluent, or excipient, whereupon the subject is treated for cancer. The miRNA can be miR-34a. The sense strand of the miRNA, which can be miR-34a, can have 15 nucleotides, whereas the antisense strand of the miRNA, which can be miR-34a, can have 22 nucleotides. Each strand of the miRNA can contain an alternating pattern of 2ƍ-O-methyl-modified and 2ƍ-fluoro-modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. In an embodiment, the sense strand can have the sequence: /52FG/*mC*/i2FU/mA/i2FA/mG/i2FA/mC/i2FA/mC/i2FU/mG/i2FC/*mC*/ i2FA//3AzideN/ (SEQ ID NO: 4) and the antisense strand can have the sequence: 70078-02 5Phos/mU*/i2FG/*mG/i2FC/mA/i2FG/mU/i2FG/mU/i2FC/mU/i2FU/mA/i2FG/mC/i2FU/ mG/i2FG/mU/i2FU/mG*/32FU/ (SEQ ID NO: 5), wherein m is 2ƍ-O-methyl; F is 2ƍ-fluoro; r is ribonucleotide; i is internal; * is a phosphorothioate bond; and Phos: 5ƍ-phosphate. The cancer can be lung, breast, ovarian, or prostate cancer. Further provided is another method of treating cancer in a subject. The method comprises administering to the subject a cancer-treating effective amount of the conjugate comprising miR-34a and a folate, optionally as a composition comprising the conjugate and a pharmaceutically acceptable carrier, diluent, or excipient, whereupon the subject is treated for cancer. The conjugate can further comprise a group that improves tumor uptake of the conjugate, such as a group that comprises or is an albumin-binding moiety. The cancer can be lung, breast, ovarian, or colorectal cancer, or medulloblastoma. Still further provided is another method of treating cancer in a subject. The method comprises administering to the subject a cancer-treating effective amount of the conjugate comprising miR-34a and either DUPA or the ligand present in PSMA-617, optionally as a composition comprising the conjugate and a pharmaceutically acceptable carrier, diluent, or excipient, whereupon the subject is treated for cancer. The conjugate can further comprise a group that improves tumor uptake of the conjugate, such as a group that comprises or is an albumin-binding moiety. The cancer can be prostate cancer. BRIEF DESCRIPTION OF THE FIGURES The disclosed embodiments and other features, advantages, and aspects contained herein, and the matter of attaining them, will become apparent in light of the following detailed description of various exemplary embodiments of the present disclosure. Such detailed description will be better understood when taken in conjunction with the accompanying drawings. FIGS.1A-1F. Chemical composition and stability of partially (PM) and fully modified (FM) miR-34a. A) Chemical modification pattern of PM-miR-34a and FM-miR-34a (SEQ ID NOs: 1-3). B) Structure of different chemical modifications used in (A). C) Representative gel-Red-stained poly-acrylamide gel of PM and FM miR-34a shows successfully annealing miRNA duplexes as indicated by a shift in the mobility on the gel (n=3). D) 70078-02 Representative gel-Red-stained poly-acrylamide gel of unmodified (UM), PM and FM miR-34a after 50% serum exposure for different periods of times (left panel). The band intensity was quantified using image J software and normalized to time 0 in (E). F) FM-miR- 34a was incubated with 50% serum for the indicated periods of times followed by loading into poly- acrylamide gel and staining using gel Red. FIGS. 2A-2F. Comparison of cellular activity of PM and FM-miR-34a. A) Targeted silencing of miR-34a Renilla sensor post-transfection of MB231-miR-34a sensor cells with PM and FM-miR- 34a duplexes using different doses (n=3). B) Normalized firefly luciferase signal in BEAS-2B cells following co-transfection with a pmiRGlo plasmid (Promega) and PM-miR-34a, FM-miR-34a or NC duplexes (n=3). C) Western blot image shows significant reduction of MET, and CD44 post-transfection of MB-231 cells with 50 nM FM-miR-34a and PM-miR-34a duplexes for the indicated time points. D) Western blot image shows significant reduction of androgen receptor (AR) expression post-transfection of LNCaP cells with 50 nM FM-miR-34a and PM- miR-34a duplexes for the indicated time points. Evaluation of the expression of endogenous miR-34a targets (E) or non-targets (F) by qRT-PCR in MB-231 cells following transfection with 50 nM PM-miR-34a or FM-miR-34a duplexes (means ± SD, (n=3). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, relative to siluc2 (one-way Anova). GAPDH was used as an endogenous control. NC: siluc2. FIGS. 3A-3E. FM-miR-34a inhibits cancer cells proliferation, migration, and invasion. Evaluation of the effect of PM-miR-34a or FM-miR-34a on proliferation of MB-231 cells A) or LNCaP cells B), measured by SRB assay (means ± SD, n=3 **P < 0.01, ***P < 0.001, ****P < 0.0001, one-way Anova. C) Representative images of MB-231 cells that migrated through the 6 ^P-pore size transwell over 6 hours following transfection with PM- miR-34a or FM-miR-34a for 72 hr (n=3). D) Representative images of LNCaP cells that invaded through the Matrigel matrix over 96 hr following transfection with PM-miR-34a or FM-miR-34a for 48 hr (n=3). E) Evaluation of the effect of PM-miR-34a or FM-miR-34a on the proliferation of MB-231 cells using clonogenic assay post-transfection with NC or miR-34a duplexes (n=3). FIGS.4A-4F. FM-miR-34a activity requires AGO loading. A) quantification of miR- 34a using qrt-PCR following Ago immunoprecipitation. MB-231 cells were transfected with NC, 70078-02 PM-miR-34a, FM-miR-34a or miR-34a mimic followed by Ago immunoprecipitation and quantification of miR-34a. miR-34a expression in Ago-IP samples was normalized to miR-34a in IgG-IP and input. B-D) Effect of Ago2 knockdown on miR-34a activity indicating the contribution of Ago2 in FM-miR-34a activity, B) Renilla luciferase expression in MB- 231 sensor cells following transfection with NC, PM-miR-34a, FM-miR-34a or miR-34a mimic in presence or absence of siRNA against Ago2 (means ± SD, n=3, ****P < 0.0001, two-tailed Student’s t test). C-D) Western blot image shows a reduction of FM-miR-34a effect on MET or AR expression following transfection of MB-231 or LNCaP cells, respectively, with 50 nM PM- miR-34a or FM-miR-34a in the presence of 50 nM siAgo2. E) Proliferation of MB-231 cells measured by SRB 120 hours after transfection with 50 nM PM-miR-34a or FM-miR-34a in presence of 50 nM siAgo2 duplexes (means ± SD, n=3, **P < 0.01, two- tailed Student’s t test). F) Migration of MB-231 cells measured by migration assay after transfection with 50 nM PM- miR-34a or FM-miR-34a in presence of 50 nM siAgo2 duplexes. FIGS. 5A-5I. in vivo efficacy of fully modified miR-34a. A-B) Subcutaneous tumor growth using MB-231 cells transfected with 50 nM PM-miR-34a, FM-miR-34a or NC duplexes (n = 6 for NC and n= 4 for PM-miR-34a and FM-miR-34a ; error bars: means ± SEM; #: corresponds to ****P < 0.0001); tumor volume is shown in (A) and in (B) is a representative image showing the tumors harvested from the mice at the end-point of the study. C) Representative image shows Renilla luciferase sensor signal in nude mice implanted with MB- 231 sensor cells following intravenous injection with a single dose of 1.5 nmol folate-NC (siluc2), folate-PM-miR-34a or folate-FM-miR- 34a. D) Effect of folate-FM-miR-34a delivery on miR-34a-Renilla sensor signal over time (data normalized to day 0; error bars: means ± SEM, n=3). E) Western blot images show the protein expression of miR-34a targets (MET, CD44 and AXL) in excised MB-231 tumors at 120 h after intravenous injection with a single dose (1.5 nmol) of folate-NC (siluc2), folate-PM-miR-34a or folate-FM-miR-34a duplexes. F) miR-34a levels from excised MB-231 tumors quantified by quantitative reverse transcription polymerase chain reaction (qRT-PCR) 120 hours post-injection with different folate conjugates (n = 3 with at least 3 different technical replicates; error bars: means ± SD; one-way ANOVA). G) Tumor volume after treatment with different folate-miRNA conjugates (folate-NC: n=6; folate-miR-34a: n=5; folate-FM-miR-34a: n=6). Arrows represent treatment time (1.5 nmol, intravenous 70078-02 injection, once every 6 days). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; two-way ANOVA. Body weight measurement throughout the treatment period shows no significant change (error bars: means ± SD) as shown in (H). As a preliminary evaluation of the potential immune response, FM-FolamiR-34a or PM-FolamiR-34a were injected into the tail veins of immunocompetent mice (FVB.129 background) followed by quantification of IL-6 (I, left panel) and TNF-Į^(I, right panel) cytokines levels in the serum 2 hours post-injection. In comparison to the positive control, mice injected with LPS, neither FM-FolamiR-34a or PM- FolamiR-34a resulted in a significant increase in cytokine levels above the negative control. FIG.6 shows the synthesis of folate-DBCO by solid phase peptide synthesis method. FIG.7 is a graph of time (hr) vs. OD value (normalized to 24 hr). FIG.8 shows the synthesis of OTL-38 (folate-NIR) synthesis. FIG.9 shows confirmation of FM-miR-34a detection using qRT-PCR (means + SD, n=3). FIGS. 10A-10G. FM-miR-34a gene targeting is more robust than PM-miR-34a. a) Overall number of statistically significant (p <0.05) differentially expressed genes in MB-231 cells transfected with either PM-miR-34a or FM-miR-34a in comparison to siluc2-transfected (NC) or untreated (UT) cells. b) Volcano plots of up-regulated (orange) and down-regulated (blue) genes compared between cells transfected with PM-miR-34a and NC, or between FM- miR-34a and NC. Gene labels represent the top 6 up- and down-regulated genes based on lowest p-value and top 2 up- and down-regulated genes based on fold-change from each comparison. Dashed line represents p-value cutoff (0.05). Grey dots indicate non-significant genes. c) Overlap of statistically significant downregulated genes in PM-miR-34a vs. NC, FM-miR-34a vs. NC and FM-miR-34a vs. PM-miR-34a comparisons. d) Heatmap of differentially expressed genes sorted based on the most differentially regulated between the FM-miR-34a and NC. e) miRNA target enrichment analysis of statistically significant downregulated genes comparing PM-miR-34a to NC and FM-miR-34a to NC based on mirTarBase database. Dot size represent the number of target genes identified. From each comparison, the top 3 predicted miRNAs are labelled and their p-value, and the number of target genes identified from the experiment are indicated in the table. f) Visualization of selected biological processes from the gene set enrichment analysis comparing all statistically significant genes specific to the biological process 70078-02 for PM-miR-34a to NC (yellow) and FM-miR-34a to NC (red). For each sector, the x-axis represents genes, and the y-axis indicates fold-change. Red represents genes only modulated in FM-miR-34a condition, yellow represent genes only modulated in PM-miR-34a condition, and overlapping area represents genes common to both. All analysis was conducted using significantly altered genes (p < 0.05). g) Line plot indicating cumulative number of downregulated genes in PM-miR-34a vs. NC (black) or FM-miR-34a vs. NC (red) that overlap with known/predicted miR-34a targets in miRDB database and their rank. Target rank (y-axis) starts with the most highly ranked miR-34a target genes to the less stringent targets. Inset shows a similar plot for the top 100-ranked miR-34a targets. FIG.11 Silencing of miR-34a Renilla sensor following transfection of partially and fully modified miR-34a duplexes. Renilla luciferase expression at 24 h and 48 h post- transfection of MB-231 miR-34a sensor cells with various doses of PM-miR-34a and FM-miR- 34a duplexes (means ± SD, n=3). Data normalized to NC, one-way Anova, ****P < 0.0001. FIG.12. Chemical modification pattern of a FM-miR-34a (SEQ ID NOs: 1 and 14). While the present disclosure is susceptible to various modifications and alternative forms, exemplary embodiments thereof are shown by way of example in the drawings and are herein described in detail. DETAILED DESCRIPTION For the purposes of promoting an understanding of the principles hereof, reference will now be made to the embodiments illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of scope is intended by the description of these embodiments. On the contrary, this disclosure is intended to cover alternatives, modifications, and equivalents as may be included within the spirit and scope of this application as defined by the appended claims. The present disclosure is predicated, at least in part, on the discovery that miRNA can be fully modified (FM) to enhance its stability over 400-fold relative to unmodified miRNA without compromising its activity. Accordingly, the term “fully modified miRNA” includes an miRNA I where the 2ƍ-OH group of all ribose bases has been changed to either 2ƍ-F or 2ƍ-OMe and a duplex contains at least 7 phosphorothioate linkages. In some embodiments, fully modified 70078-02 miRNAs can further comprise a ^ƍ-vinylphosphonate in place of the 5ƍ-phosphate. In particular, the miRNA known as miR-34a has been fully modified and, when fully modified, miR-34a (FM- miR-34a) more robustly downregulates targets of miR-34a, including greater than 90% of targets such as CD44, AXL, and MET, as compared to partially modified (PM-miR-34a) miR-34a. Downregulation occurs in an Argonaute 2 (Ago2) protein-dependent manner. Enforced expression of FM-miR-34a in breast cancer cells resulted in stronger inhibition of proliferation and invasion and delayed tumor growth in comparison to partially modified miR-34a (PM-miR- 34a). Systemic delivery of a single dose of FM-miR-34a conjugated to folate (FM-FolamiR) induced stronger and prolonged downregulation of the expression of the target genes at a dose three-fold lower than what was administered for a PM-miR-34a conjugated to folate (PM- FolamiR). Surprisingly, FM-FolamiR significantly inhibited tumor growth in mice, leading to complete cures in some mice. In view of the above, it is an object of the present disclosure to provide a fully chemically modified microRNA (miRNA), such as the one shown in FIG.1A. The miRNA 100 is double- stranded and comprises a sense strand (also referred to as a passenger strand) 102 and an antisense strand (also referred to as a guide strand) 104. One strand can be longer than the other. Or one strand can be substantially the same length (e.g., a difference of 3 nucleotide bases) as the other strand. Each strand of the miRNA can independently range in length from about 12 nucleotides to about 40 nucleotides, such as 14-40, 16-40, 18-40, 20-40, 22-40, 24-40, 26-40, 28- 40, 30-40, 14-38, 14-36, 14-34, 14-32, 14-30, 14-28, 14-26, 14-24, or 14-22. The sense 102 and antisense 104 strands can be equal in length or unequal in length. In some embodiments the antisense strand 104 is longer, e.g., by 1, 2, 3, 4, 5, 6, or 7 nucleotides (e.g., by 1 to 5, 2 to 6, 3 to 7, or 2 to 7 nucleotides), than the sense strand 102. In some embodiments, the sense strand 102 of the miRNA can have 15 nucleotides, whereas the antisense strand 104 of the miRNA can have 22 nucleotides. In some embodiments, the fully chemically modified miRNA can comprise an antisense sequence that comprises at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, at least 18, at least 19, or at least 20 contiguous nucleotides of SEQ ID NO: 1, such as from 7 to 10, 7 to 15, 8 to 16, 7 to 19 or 8 to 20 contiguous nucleotides of SEQ ID NO: 1. In some embodiments, the seven contiguous nucleotides are 70078-02 GGCAGUG, which is the same “seed” sequence shared by the miR-34 family of miRNA, including miR-34a (UGGCAGUGUCUUAGCUGGUUGU; SEQ ID NO: 1), miR-34b (UAGGCAGUGUCAUUAGCUGAUUG; SEQ ID NO: 15), miR-34c (AGGCAGUGUAGUUAGCUGAUUGC; SEQ ID NO: 16), miR-449a (UGGCAGUGUAUUGUUAGCUGGU; SEQ ID NO: 17), miR-449b (AGGCAGUGUAUUGUUAGCUGGC; SEQ ID NO: 18), and miR-449c (UAGGCAGUGUAUUGCUAGCGGCUGU; SEQ ID NO: 19). See, for example, FIG.2 of J Biol Chem.2019 Mar 22; 294(12): 4381–4400, which is incorporated by reference as if fully set forth herein. Accordingly, the disclosure contemplates fully chemically modified miRNA versions of miR-34b, miR-34c, and miR-449a-c. The miRNA has a double-stranded or duplex region 106. Generally, the double-stranded (duplex) region is 12-25 nucleotide base pairs in length, such as 12, 14, 16, 18, 20, 22, or 24 nucleotide base pairs in length. In some embodiments, the double-stranded region is 20, 21, 22, 23 or 24 nucleotide base pairs in length. Making reference to FIG. 1A, when the sense and antisense strands differ in length, the miRNA has a single-stranded or “overhang” region 108. Generally, the single-stranded (overhang) region 108 is at least about 4 nucleotides, at least about 5 nucleotides, at least about 6 nucleotides, or at least about 7 nucleotides in length, such as 7 to 20 nucleotides in length or 4 to 20 nucleotides. In some embodiments, the single-stranded region 108 is at the 3ƍ end of the antisense strand 104. In other embodiments, the single-stranded region 108 is at the 5ƍ end of the antisense strand. The antisense strand 104 can align with (or near) the 5ƍ end of the sense strand 102. When the antisense strand 104 aligns with (or near) the 5ƍ end of the sense strand 102, a targeting ligand, such as folate, DUPA, or the ligand present in PSMA-617, can be attached to the 5ƍ end of the antisense strand, wherein folate has the structure: 70078-02 The structure of DUPA is shown below, as is the structure of the ligand present in PSMA-617 and other targeting ligands that can be used with the miRNA.

70078-02 (ligand present in PSMA-617). PSMA-617, itself, has the structure: . Alternatively, the antisense strand 104 can align with (or near) the 3ƍ end of the sense strand 102 as shown in FIG. 1A. Alignment with (or near) the 3ƍ end of the sense strand 102 can increase stability. When the antisense strand 104 aligns with (or near) the 3ƍ end of the sense strand 102, a targeting ligand, such as folate, DUPA, or the ligand present in PSMA-617, can be attached to the 3ƍ end of the sense strand 102. As another alternative, the antisense strand 104 can align with the sense strand 102 at any position along its length, in which case a targeting ligand, such as folate, DUPA, or the ligand present in PSMA-617, can be attached to the 5ƍ end or the 3ƍend of the sense strand 102, e.g., whichever end of the sense strand 102 is closer to the end of the antisense strand 104. The miRNA can be, and desirably is, at least partially, and desirably fully, modified. Examples of modifications include, but are not limited to, ^ƍ-O-PHWK\O^^^ƍ-fluoro ribose bases, and phosphorothioate linkages, as shown in FIG.1B. Any configuration of modifications, which increases stability without compromising activity, can be used including the inclusion of an extended nucleic acid that can be incorporated at the 3ƍ-end of the antisense strand 104. One example of a suitable extended nucleic acid is one of the formula: 70078-02 , wherein “Base” means A, T, C, or G; T¹ is H or OH; and X¹ is alkyl, such as C₁-C₃-alkyl, such as, e.g., CH₂, CH₂CH₂, and CH₂CH₂CH₂. Such an extended nucleic acid can, e.g., replace the phosphorothioate linkage between the G and the U at the 3ƍ-end of the antisense strand 104 in FIG.1A. In some embodiments, the miRNA can comprise at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 or more 2-fluoro ribose bases. The 2ƍ-fluoro ribose bases can all be present in one strand, i.e., the antisense strand 104 or the sense strand 102. In some embodiments, the sense and antisense strands 102/104 comprise at least 1, 2, 3, 4, 5, 6 or more 2ƍ-fluoro ribose bases. In some embodiments, the sense strand 102 comprises at least 1, 2, 3, 4, 5, 6 or 72ƍ-fluoro ribose bases, and the antisense strand 104 comprises at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or 112ƍ-fluoro ribose bases. The 2ƍ-fluoro modifications can occur on adjacent nucleotides, on alternating nucleotides, or a pattern of alternating and adjacent nucleotides. The positions/pattern of the 2ƍ-fluoro modifications on one strand can differ from the positions/pattern of the 2ƍ-fluoro modifications on the other strand. In some embodiments, the miRNA can comprise at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 or more 2ƍ-O-methyl modified nucleotides. The 2ƍ-O-methyl modified nucleotides can all be present in one strand, i.e., the antisense strand 104 or the sense strand 102. In some embodiments, the sense and antisense strands 102/104 comprise at least 1, 2, 3, 4, 5, 6 or more 2ƍ-O-methyl modified nucleotides. In some embodiments, the sense strand 102 comprises at least 1, 2, 3, 4, 5, 6 or 72ƍ-O-methyl modified nucleotides, and the antisense strand 104 comprises at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or 112ƍ-O-methyl modified nucleotides. The 2ƍ-O- methyl modified nucleotides can occur on adjacent nucleotides, on alternating nucleotides, or a pattern of alternating and adjacent nucleotides. The positions/pattern of the 2ƍ-O-methyl 70078-02 modified nucleotides on one strand can differ from the positions/pattern of the 2ƍ-O-methyl modified nucleotides on the other strand. Nucleotides in the single-stranded (overhang) region 108 can each be independently modified, such as 2ƍ-sugar modified, e.g., 2ƍ-fluoro, 2ƍ-O-methyl, thymidine (T), 2ƍ-O- methoxyethyl-5-methyluridine, 2ƍ-O-methoxyethyladenosine, and 2ƍ-O-methoxyethyl-5- methylcytidine. The 5ƍ or 3ƍ single-stranded region(s) can be modified, such as phosphorylated, e.g., with phosphorothioate or methylphosphonate internucleotide linkages, wherein the nucleotides can be the same or different. Other modifications include, but are not limited to, 5ƍ phosphorylation, such as with a phosphoryl analog. Examples of modifications include, but are not limited to, 5ƍ- monophosphate, 5ƍ-diphosphate, 5ƍ-triphosphate, 5ƍ-guanosine, 5ƍ-adenosine, 5ƍ- monothiophosphate, 5ƍ-monodithiophosphate, 5ƍ-phosphorothiolate,and 5ƍ-vinylphosphonate: , wherein “Base” means A, T, C, or G. In some embodiments, each strand of the miRNA contains an alternating pattern of 2ƍ-O- methyl-modified and 2ƍ-fluoro-modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. In some embodiments, the phosphorothioate linkages can be present in a substantial portion of the single stranded overhang region (e.g., phosphorothioate linkages can be present in the entire single stranded overhang region) of a longer strand. Thus, for example, each strand of the miRNA can contain phosphorothioate linkages at the 5’ and 3’ends, including multiple linkages that expand into the single stranded overhang of a longer strand. The miRNA can be miR-34a or a miR-34a mimic (see, e.g., U.S. Pat. Appl. Pub. Nos.2012/0288933, 2013/0123329, and 2015/0087607, each of which is incorporated by reference as if fully set forth herein). In an embodiment, the sense strand 102 can have the sequence: /52FG/*mC*/i2FU/mA/i2FA/mG/i2FA/mC/i2FA/mC/i2FU/mG/i2FC/*mC*/ i2FA//3AzideN/ (SEQ ID NO: 4) 70078-02 and the antisense strand 104 can have the sequence: 5Phos/mU*/i2FG/*mG/i2FC/mA/i2FG/mU/i2FG/mU/i2FC/mU/i2FU/mA/i2FG/mC/i2FU/mG/i2F G/mU/i2FU/mG*/32FU/ (SEQ ID NO: 5), wherein m is 2ƍ-O-methyl; F is 2ƍ-fluoro; r is ribonucleotide; i is internal; * is a phosphorothioate bond; and Phos: 5ƍ-phosphate. See FIG.1A. A conjugate comprising the miRNA, e.g., miR-34a, and a ligand is also provided. Any suitable/desirable ligand can be coupled to the miRNA. In various embodiments, the ligand is coupled covalently, either directly or indirectly via a linker (L). L can be any suitable linker. In one example, L can be a "non-releasable linker" or "non- cleavable linker. " "Non-releasable linker" or "non-cleavable linker" refers to a linker that cannot be cleaved under extracellular physiological conditions (e.g., a pH-labile, an acid-labile, an oxidatively labile, or an enzyme-labile bond). However, such a linker may include bonds that can be cleaved after entry into a cell. In another example, L can be a "releasable linker." A "releasable linker" refers to a linker that includes at least one bond that can be broken under physiological conditions (e.g., a pH- labile, acid-labile, oxidatively labile, or enzyme-labile bond). Releasable groups also include photochemically cleavable groups. Examples of photochemically cleavable groups include 2-(2- nitrophenyl)-ethan-2-ol groups, linkers containing o-nitrobenzyl, desyl, trans-o-cinnamoyl, m- nitrophenyl or benzylsulfonyl groups (see, for example, Dorman and Prestwich, Trends Biotech. 18:64-77 (2000); Greene and Wuts, Protective Groups in Organic Synthesis, 2nd ed., John Wiley & Sons, New York (1991); and U.S. Pat. Nos. 5,143,854; 5,986,076; 5,917,016; 5,489,678; and 5,405,783, all of which are hereby specifically incorporated by reference for their teachings regarding same). L can comprise a chain of atoms from about 3 atoms to about 30 atoms (for example, about 3 atoms to 30 atoms, 3 atoms to about 30 atoms, 3 atoms to 30 atoms, about 3 atoms to about 7 atoms, about 5 atoms to about 15 atoms, about 5 atoms to about 25 atoms, about 5 atoms to about 12 atoms, about 7 atoms to about 15 atoms, about 7 atoms to about 12 atoms, about 7 atoms to about 15 atoms or about 10 atoms to about 30 atoms) in length. L can comprise a chain of atoms from about 5 Å to about 45 Å in length, such as about 5 Å to 45 Å, 5 Å to about 45 Å, or 5 Å to 45 Å. L can comprise a peptide. L can comprise one or more phenylalanine residues, each of which is independently optionally substituted. L can comprise at least one phenylalanyl- 70078-02 phenylalanyl, in which at least one phenyl is independently optionally substituted. L can comprise a polyoligoethylene glycoln (POEGn), a polyethylene glycoln (PEGn), or a mixture thereof, wherein n = 1-36. In some embodiments, L can comprise at least one linker group, each linker group selected from the group consisting of polyethylene glycol (PEG), alkyl, sugar, and peptide. In some embodiments, the linker is a PEG- (e.g., pegylated-), alkyl-, sugar-, and peptide-based dual linker. The linker can be any suitable linker. For example, in some embodiments, the linker is a hydrophilic linker, such as a linker that comprises one or more of an amino acid (which are the same or different), an alkyl chain, a PEG monomer, a PEG oligomer, a PEG polymer, or a combination of any of the foregoing. In some embodiments, the linker comprises an oligomer of peptidoglycans, glycans, or anions. For a linker that comprises one or more PEG units, all carbon and oxygen atoms of the PEG units are part of the backbone unless otherwise specified. The “backbone” of the linker L can be the shortest chain of contiguous atoms forming a covalently bonded connection between T and X and/or T and A. In some embodiments, a polyvalent linker has a branched backbone, with each branch serving as a section of backbone linker until reaching a terminus. The L groups described herein can have any suitable length and chemical composition. For example, L can have a chain length of at least about 7 atoms (e.g., 7 atoms) in length. In one variation, L is at least about 10 atoms (e.g., 10 atoms) in length. In one variation, L is at least about 14 atoms (e.g., 14 atoms) in length. In another variation, L is between about 7 and about 31 (e.g., about 7 and 31, 7 and about 31, or 7 and 31), between about 7 and about 24 (e.g., about 7 and 24, 7 and about 24, or 7 and 24), or between about 7 and about 20 atoms (e.g., about 7 and 20, 7 and about 20, or 7 and 20) in length. In another variation, L is between about 14 and about 31 (e.g., about 14 and 31, 14 and about 31, or 14 and 31), between about 14 and about 24 (e.g., abougt 14 and 24, 14 and about 24, and 14 and 24), or between about 14 and about 20 (e.g., about 14 and 20, 14 and about 20, or 14 and 20) atoms in length. In another variation, L can have a chain length of at least 7 atoms (e.g., 7 atoms), at least 14 atoms (e.g., 14 atoms), at least 20 atoms (e.g., 20 atoms), at least 25 atoms (e.g., 25 atoms), at least 30 atoms (e.g., 30 atoms), at least 40 atoms (e.g., 40 atoms), from 1 to 15 atoms, 1 to 5 atoms, 5 to 10 atoms, 5 to 20 atoms, 70078-02 10 to 40 atoms or 25 to 100 atoms. An example of an L group having a chain length of 1 to 5 atoms is a group of the formula: wherein R¹ can be H, alkyl, arylalkyl, -alkyl-S-alkyl or arylalkyl or the side chain of any naturally or non-naturally occurring amino acid, and the like; and the numbers represent the atoms that are counted as being part of the chain, which in this example is three atoms. Examples of R¹ include H (i.e., glycine), alkyl (e.g., alanine, valine, isoleucine, and leucine), -alkyl-S-alkyl (e.g., methionine), arylalkyl (e.g., phenylalanine, tyrosine, tryptophan, and napthylalanine), and the like. The atom to which R¹ is attached can be chiral and can have any suitable relative configuration, such as a D- or L-configuration. The atoms used in forming L can be combined in all chemically relevant ways, such as chains of carbon atoms forming alkylene groups, chains of carbon and oxygen atoms forming polyoxyalkylene groups, chains of carbon and nitrogen atoms forming polyamines, and others. In addition, it is to be understood that the bonds connecting atoms in the chain can be either saturated or unsaturated, such that, for example, alkanes, alkenes, alkynes, cycloalkanes, arylenes, imides, and the like can be divalent radicals that are included in L. In addition, it is to be understood that the atoms forming the linker may also be cyclized upon each other to form saturated or unsaturated divalent cyclic radicals in the linker, such as radicals of the formulae: wherein each X² is independently CH₂, N (when there is a bond attached to X²), NH or O and each X³ is independently N, C (when there is a bond attached to X³) or CH. In each of the foregoing and other L groups described herein the chain forming the linker can be substituted or unsubstituted. Alternatively, or in addition to chain length, L can have any suitable substituents that can affect the hydrophobicity or hydrophilicity of L. Thus, for example, L can have a hydrophobic side chain group, such as an alkyl, cycloalkyl, aryl, arylalkyl, or like group, each of which is optionally substituted. If L were to include one or more amino acids, L can contain a 70078-02 hydrophobic amino acid side chain, such as one or more amino acid side chains from phenylalanine (Phe) and tyrosine (Tyr), including substituted variants thereof, and analogs and derivatives of such side chains. L can comprise portions that are neutral under physiological conditions. But L can comprise portions that can be protonated or deprotonated to carry one or more positive or one or more negative charges, respectively. Or L can comprise neutral portions and portions that can be protonated to carry one or more positive charges. Examples of neutral portions include poly hydroxyl groups, such as sugars, carbohydrates, saccharides, inositols, and the like, and/or polyether groups, such as polyoxyalkylene groups including polyoxyethylene, polyoxypropylene, and the like. Examples of portions that can be protonated to carry one or more positive charges include amino groups, such as polyaminoalkylenes including ethylene diamines, propylene diamines, butylene diamines and the like, and/or heterocycles including pyrrolidines, piperidines, piperazines, and other amino groups, each of which can be optionally substituted. Examples of portions that can be deprotonated to carry one or more negative charges include carboxylic acids, such as aspartic acid, glutamic acid, and longer chain carboxylic acid groups, and sulfuric acid esters, such as alkyl esters of sulfuric acid. Illustrative polyoxyalkylene groups include those of a specific length range from about 4 to about 20 (e.g., about 4 to 20, 4 to about 20, or 4 to 20) polyoxyalkylene (e.g., polyethylene glycol) groups. Illustrative alkyl sulfuric acid esters may also be introduced with click chemistry directly into the backbone. Illustrative L groups comprising polyamines include L groups comprising EDTA and DTPA radicals:

70078-02 ȕ-amino acids, and the like: and combinations thereof, wherein each R² is independently H, alkyl, arylalkyl, heterocyclylalkyl, ureido, aminoalkyl, alkylthio or amidoalkyl, such as in the side chains of naturally-occurring amino acids like alanine, valine, leucine, isoleucine, phenylalanine, tyrosine, tryptophan, serine threonine, asparagine, methionine, lysine, arginine, and histidine. Non- naturally occurring amino acids are also contemplated herein. The L groups can have any suitable molecular weight, such as from about 30 g/mol to about 1,000 g/mol (e.g., about 30 to 1,000; 30 to about 1,000; or 30 to 1,000), from about 30 g/mol to about 300 g/mol (e.g., about 30 to 300; 30 to about 300; or 30 to 300), about 100 g/mol to about 500 g/mol (e.g., about 100 to 500; 100 to about 500; or 100 to 500) or about 150 g/mol to about 600 g/mol (e.g., about 150 to 600; 150 to about 600; or 150 to 600). The terms "non-releasable linker" or “non-cleavable linker” are used interchangeably. As used herein, they refer to a linker that cannot be cleaved under extracellular physiological conditions (e.g., a pH-labile, acid-labile, oxidatively labile, or enzyme-labile bond). However, such a linker may include bonds that can be cleaved after entry into a cell. L can comprise carbonyl, aminoalkyleneamino, aminoalkylenecarbonyl, carbonylalkylenecarbonyl, thionocarbonyl, alkylene, cycloalkylene, aminoalkylene, alkylenecycloalkyl, alkylenecycloalkylenecarbonyl, aminoalkylenecycloalkylenecarbonyl, alkylenecarbonyl, cycloalkylenecarbonyl, 1-alkylenesuccinimid-3-yl, 1- (carbonylalkyl)succinimid-3-yl, alkylenesulfoxyl, sulfonylalkyl, alkylenesulfoxylalkyl, alkylenesulfonylalkyl, carbonyltetrahydro-2H-pyranyl, carbonyltetrahydrofuranyl, 1- (carbonyltetrahydro-2H-pyranyl)succinimid-3-yl or 1-(carbonyltetrahydrofuranyl)succinimid-3- yl, each of which is optionally substituted, and combinations thereof. In this example, L can further comprise an additional nitrogen (e.g., -NR³-, wherein R³ can be H or alkyl) such that L comprises alkylenecarbonyl, cycloalkylenecarbonyl, carbonylalkylcarbonyl or 1- (carbonylalkyl)succinimid-3-yl groups, each of which can be optionally substituted, bonded to the nitrogen to form an amide. Alternatively, L can further comprise a sulfur atom and alkylene 70078-02 or cycloalkylene groups, each of which can be optionally substituted with carboxy, and can be bonded to the sulfur to form a thiol. In yet another example, L comprises a sulfur atom and 1- alkylenesuccinimid-3-yl and 1-(carbonylalkyl)succinimid-3-yl groups bonded to the sulfur to form a succinimid-3-ylthiol. L can include alkyleneaminoalkylenecarbonyl, alkylene-thio-(carbonylalkylsuccinimid-3- yl), alkylenecycloalkylenecarbonyl, aminoalkylenecycloalkylenecarbonyl, alkylenecarbonyl, cycloalkylenecarbonyl and the like and combinations thereof, as further illustrated by the following formulae: , wherein the asterisk denotes a point of attachment to a group present in L, in T, in A, or in X; and wherein x and y are each independently 1, 2, 3, 4, or 5. L can have any suitable assortment of atoms in the chain, including C (e.g., -CH₂-, C(O)), N (e.g., NH, NR⁴, wherein R⁴ is, e.g., H, alkyl, alkylaryl, and the like), O (e.g., -O-), P (e.g., -O- P(O)(OH)O-), and S (e.g., -S-). For example, the atoms used in forming L can be combined in all chemically relevant ways, such as chains of carbon atoms forming alkyl groups, chains of carbon and oxygen atoms forming polyoxyalkyl groups, chains of carbon and nitrogen atoms forming polyamines, and others, including rings, such as those that form aryl and heterocyclyl groups (e.g., triazoles, oxazoles, and the like). In addition, the bonds connecting atoms in the chain in L can be either saturated or unsaturated, such that, for example, alkanes, alkenes, alkynes, cycloalkanes, arylenes, imides, and the like can be divalent radicals that are included in L. Further, the chain forming L can be substituted, e.g., with an -N(R⁴)2 group, or unsubstituted. Additional examples of L include L groups that include the groups 1-alkylsuccinimid-3- yl, carbonyl, thionocarbonyl, alkyl, cycloalkyl, alkylcycloalkyl, alkylcarbonyl, cycloalkylcarbonyl, carbonylalkylcarbonyl, 1-alkylsuccinimid-3-yl, 1- 70078-02 (carbonylalkyl)succinimid-3-yl, alkylsulfoxyl, sulfonylalkyl, alkylsulfoxylalkyl, alkylsulfonylalkyl, carbonyltetrahydro-2H-pyranyl, carbonyltetrahydrofuranyl, 1- (carbonyltetrahydro-2H-pyranyl)succinimid-3-yl, and 1-(carbonyltetrahydrofuranyl) succinimid-3-yl, wherein each group can be substituted or unsubstituted. Any of the aforementioned groups can be L or can be included as a portion of L. In some instances, any of the aforementioned groups can be used in combination (or more than once) (e.g., -alkyl-C(O)- alkyl) and can further comprise an additional nitrogen (e.g., alkyl-C(O)-NH-, -NH-alkyl-C(O)- or -NH-alkyl-), oxygen (e.g., -alkyl-O-alkyl-) or sulfur (e.g., -alkyl-S-alkyl-). Examples of such L groups are alkylcarbonyl, cycloalkylcarbonyl, carbonylalkylcarbonyl, 1- (carbonylalkyl)succinimid-3-yl, and succinimid-3-ylthiol, wherein each group can be substituted or unsubstituted. Conjugates may comprise releasable linkers for L if, e.g., release of A in vivo is desired. Releasable linkers for L are well-known in the art. L can be a “releasable linker” that is cleavable by an enzyme. The enzyme can be cathepsin, metalloproteinase, esterase, phosphatase, DNAase or pyrophosphatase. L can be cleavable by a reactive oxygen species (ROS). L can be p-aminophenol ether. L can be cleavable by hypoxic activation. L can be a quinone, a nitroaromatic, an aliphatic N-oxide, or a hetero- aromatic N-oxide. Or L can comprise a xN-xN portion (e.g., deoxythymidine-deoxythymidine (dT-dT) portion) that can be cleaved, e.g., by a DNAase, wherein x is a ribonucleotide or a deoxyribonucleotide; and each N is, independently, A, T, C, G, U, and combinations thereof. An example of such a dT-dT linker is:

70078-02 . An example of an miRNA having a dT-dT linker shown in FIG. 12. The miRNA shown in FIG. 12 also comprises a vinylphosphonate VKRZQ^DERYH^DW^WKH^^ƍ-end of the antisense strand 104, DQ^H[WHQGHG^QXFOHLF^DFLG^WKDW^FDQ^EH^LQFRUSRUDWHG^DW^WKH^^ƍ-end of the antisense strand 104, and a dT-G7^OLQNHU^^FOHDYDEOH^^DW^WKH^^ƍ-end of sense strand 102. In an embodiment, the sense strand 102 can have the sequence: /5mC/*mC*/mA/i2FG/mC/i2FU/mA/i2FA/mG/i2FA/mC/ i2FA/mC/i2FU/mG/mC*/*mC*/mU/T/T/3AzideN/ (SEQ ID NO: 14) and the antisense strand 104 can have the sequence: 5VPPhos/mU*/i2FG/*mG/i2FC/mA/i2FG/mU/i2FG/mU/i2FC/mU/i2FU/mA/i2FG/mC/i2 wherein m is 2ƍ-O-methyl; F is 2ƍ-fluoro; 70078-02 r is ribonucleotide; i is internal; * is a phosphorothioate bond; VP: ^ƍ-vinylphosphonate; and y: extended nucleic acid, which in FIG.12 is: While not wishing to be bound by any theory, it is believed that the extended nucleic acid WKDW^FDQ^EH^LQFRUSRUDWHG^DW^WKH^^ƍ-end of the antisense strand 104 will provide exonuclease resistance. The cleavable bond or bonds can be present in the interior of a cleavable linker and/or at one or both ends of a cleavable linker. It should be appreciated that such physiological conditions resulting in bond breaking include standard chemical hydrolysis reactions that occur, for example, at physiological pH, or as a result of compartmentalization into a cellular organelle such as an endosome having a lower pH than cytosolic pH. Illustratively, the bivalent linkers can undergo cleavage under other physiological or metabolic conditions, such as by the action of a glutathione-mediated mechanism. It is appreciated that the lability of the cleavable bond can be adjusted by including functional groups or fragments within the bivalent linker L that are able to assist or facilitate such bond breakage, also termed anchimeric assistance. The lability of the cleavable bond can also be adjusted by, for example, substitutional changes at or near the cleavable bond, such as including alpha branching adjacent to a cleavable disulfide bond, increasing the hydrophobicity of substituents on silicon in a moiety having a silicon-oxygen bond that can be hydrolyzed, homologating alkoxy groups that form part of a ketal or acetal that can be hydrolyzed, and the like. In addition, it is appreciated that additional functional groups or fragments can be included within the bivalent linker L that are able to assist or facilitate additional fragmentation of the PSMA binding drug linker conjugates after bond breaking of the 70078-02 releasable linker, when present. In one example, L can comprise one or more releasable linkers that cleave under the conditions described herein by a chemical mechanism involving beta elimination. Such releasable linkers include beta-thio, beta-hydroxy, and beta-amino substituted carboxylic acids and derivatives thereof, such as esters, amides, carbonates, carbamates, and ureas. Such linkers also include 2- and 4-thioarylesters, carbamates, and carbonates. An example of a releasable linker includes a linker of the formula: wherein n is an integer selected from 0, 1, 2, and 3, R⁵ is H or alkyl, R⁶ is hydrogen, or a substituent, including a substituent that can stabilize a positive charge inductively or by resonance on the aryl ring, such as alkoxy, and the like. The releasable linker can be further substituted. Assisted cleavage of releasable portions of L can include mechanisms involving benzylium intermediates, benzyne intermediates, lactone cyclization, oxonium intermediates, beta-elimination, and the like. In addition to fragmentation subsequent to cleavage of a releasable portion of L, the initial cleavage of the releasable linker can be facilitated by an anchimerically assisted mechanism. Thus, in the example of a releasable portion of L given above, the hydroxyalkanoic acid, which may cyclize, facilitates cleavage of the methylene bridge, by for example an oxonium ion, and facilitates bond cleavage or subsequent fragmentation after bond cleavage of the releasable linker. Alternatively, acid-catalyzed, oxonium ion-assisted cleavage of the methylene bridge can begin a cascade of fragmentation of this illustrative bivalent linker, or fragment thereof. Alternatively, acid-catalyzed hydrolysis of the carbamate may facilitate the beta elimination of the hydroxyalkanoic acid, which may cyclize, and facilitate cleavage of the methylene bridge by, for example, an oxonium ion. It is appreciated that other chemical mechanisms of bond breakage or cleavage under the metabolic, physiological, or cellular conditions may initiate such a cascade of fragmentation. It is appreciated that other chemical mechanisms of bond breakage or cleavage under the metabolic, physiological, or cellular conditions can initiate such a cascade of fragmentation. 70078-02 Illustrative mechanisms for cleavage of the bivalent linkers include the following 1,4 and 1,6 fragmentation mechanisms for carbonates and carbamates: wherein Nuc- is an exogenous or endogenous nucleophile, glutathione, or bio-reducing agent, and the like, and one of R⁷ and Z is T (or X) connected through other portions of the bivalent linker, and the other is X (or T) connected through other portions of the bivalent linker. The location of R⁷ and Z can be switched such that, e.g., the resulting products are Z-S-Nuc and HO- R⁷ or H₂N-R⁷. Although the above fragmentation mechanisms are depicted as concerted mechanisms, any number of discrete steps can take place to effect the ultimate fragmentation of the bivalent linker to the final products shown. For example, the bond cleavage can also occur by acid- catalyzed elimination of the carbamate moiety, which can be anchimerically assisted by the stabilization provided by either the aryl group of the beta sulfur or disulfide illustrated in the above examples. In those variations of this embodiment, the releasable linker is the carbamate moiety. Alternatively, the fragmentation can be initiated by a nucleophilic attack on the disulfide group, causing cleavage to form a thiolate. The thiolate can intermolecularly displace a carbonic 70078-02 acid or carbamic acid moiety and form the corresponding thiocyclopropane. In the case of the benzyl-containing bivalent linkers, following an illustrative breaking of the disulfide bond, the resulting phenyl thiolate can further fragment to release a carbonic acid or carbamic acid moiety by forming a resonance-stabilized intermediate. In any of these cases, the releasable nature of the illustrative bivalent linkers can be realized by whatever mechanism can be relevant to the chemical, metabolic, physiological, or biological conditions present. As described above, therefore, releasable linkers can comprise a disulfide group. Further examples of releasable linkers comprised in L can include divalent radicals comprising alkyleneaziridin-1-yl, alkylenecarbonylaziridin-1-yl, carbonylalkylaziridin-1-yl, alkylenesulfoxylaziridin-1-yl, sulfoxylalkylaziridin-1-yl, sulfonylalkylaziridin-1-yl, or alkylenesulfonylaziridin-1-yl groups, wherein each of the releasable linkers is optionally substituted. Additional examples of releasable linkers comprised in L can include divalent radicals comprising methylene, 1-alkoxyalkylene, 1-alkoxycycloalkylene, 1- alkoxyalkylenecarbonyl, 1-alkoxycycloalkylenecarbonyl, carbonylarylcarbonyl, carbonyl(carboxyaryl)carbonyl, carbonyl(biscarboxyaryl)carbonyl, haloalkylenecarbonyl, alkylene(dialkylsilyl), alkylene(alkylarylsilyl), alkylene(diarylsilyl), (dialkylsilyl)aryl, (alkylarylsilyl)aryl, (diarylsilyl)aryl, oxycarbonyloxy, oxycarbonyloxyalkyl, sulfonyloxy, oxysulfonylalkyl, iminoalkylidenyl, carbonylalkylideniminyl, iminocycloalkylidenyl, carbonylcycloalkylideniminyl, alkylenethio, alkylenearylthio or carbonylalkylthio groups, wherein each of the releasable linkers can be optionally substituted. Additional examples of releasable linkers comprised in L can include an oxygen atom and methylene, 1-alkoxyalkylene, 1-alkoxycycloalkylene, 1-alkoxyalkylenecarbonyl or 1- alkoxycycloalkylenecarbonyl groups, wherein each of the releasable linkers can be optionally substituted. Alternatively, the releasable linker can include an oxygen atom and a methylene group, wherein the methylene group can be substituted with an optionally substituted aryl, and the releasable linker can be bonded to the oxygen to form an acetal or ketal. Further, the releasable linker can include an oxygen atom and a sulfonylalkyl group, and the releasable linker can be bonded to the oxygen to form an alkylsulfonate. Additional examples of releasable linkers comprised in L can include a nitrogen (e.g., -NR⁵-, wherein R⁵ is H or alkyl) and iminoalkylidenyl, carbonylalkylideniminyl, 70078-02 iminocycloalkylidenyl, and carbonylcycloalkylideniminyl groups, wherein each of the releasable linkers can be optionally substituted, and the releasable linker can be bonded to the nitrogen to form a hydrazone. In an alternate configuration, the hydrazone can be acylated with a carboxylic acid derivative, an orthoformate derivative, or a carbamoyl derivative to form various acylhydrazone releasable linkers. Additional examples of releasable linkers comprised in L can include an oxygen atom and alkylene(dialkylsilyl), alkylene(alkylarylsilyl), alkylene(diarylsilyl), (dialkylsilyl)aryl, (alkylarylsilyl)aryl or (diarylsilyl)aryl groups, wherein each of the releasable linkers can be optionally substituted, and the releasable linker can be bonded to the oxygen to form a silanol. Additional examples of releasable linkers comprised in L can include two independent nitrogens (e.g., -NR⁵-) and a carbonylarylcarbonyl, a carbonyl(carboxyaryl)carbonyl, or a carbonyl(biscarboxyaryl)carbonyl, and the releasable linker can be bonded to the heteroatom nitrogen to form an amide and also be bonded to Z or R⁷ via an amide bond. Additional examples of releasable linkers comprised in L can include an oxygen atom, a nitrogen (e.g., -NR⁵-), and a carbonylarylcarbonyl, a carbonyl(carboxyaryl)carbonyl, or a carbonyl(biscarboxyaryl)carbonyl, and the releasable linker can form an amide and also be bonded to Z or R⁷ via an amide bond. In some embodiments, the ligand is folate. In other embodiments, the ligand is 2-[3-(1,3- dicarboxypropyl)ureido]pentanedioic acid (DUPA) or the ligand present in PSMA-617. “Folate” can be folic acid, a folic acid analog, or another folate receptor-binding molecule, including for example, analogs and derivatives of folic acid such as, without limitation, folinic acid (e.g., leucovorin), pteroylpolyglutamic acid, pteroyl-D-glutamic acid, and folate receptor-binding pterdines such as tetrahydropterins, dihydrofolates, tetrahydrofolates (e.g., 5-methyltetrahydrofolate (5-MTHF)), and their deaza and dideaza analogs. An “analog” or “derivative” with reference to a peptide, polypeptide or protein refers to another or identical amino acid sequence or structure of the original peptide, polypeptide or protein. An analog preferably satisfies at least one of the following: (a) a proteinaceous agent having an amino acid sequence that is at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, or at least 99% identical to the original amino 70078-02 acid sequence; (b) a proteinaceous agent encoded by a nucleotide sequence that hybridizes under stringent conditions to a nucleotide sequence encoding the original amino acid sequence; or (c) a proteinaceous agent encoded by a nucleotide sequence that is at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, or at least 99% identical to the nucleotide sequence encoding the original amino acid sequence. The terms “deaza” and “dideaza” analogs refer to the art-recognized analogs having a carbon atom substituted for one or two nitrogen atoms in the naturally occurring folic acid structure, or analog or derivative thereof. For example, the deaza analogs can include the 1- deaza, 3-deaza, 5-deaza, 8-deaza, and 10-deaza analogs of folate, folinic acid, pteropolyglutamic acid, and folate receptor-binding pteridines such as tetrahydropterins, dihydrofolates, and tetrahydrofolates. The dideaza analogs include, for example, 1,5-dideaza, 5,10-dideaza, 8,10- dideaza, and 5,8-dideaza analogs. The foregoing folic acid analogs are conventionally termed “folates,” reflecting their capacity to bind to folate receptors. Other folate receptor-binding analogs include aminopterin, amethopterin (methotrexate), N10-methylfolate, 2-deamino- hydroxyfolate, deaza analogs such as 1-deazamethopterin or 3-deazamethopterin, and 3’,5’- dichloro-4-amino-4-deoxy-N¹⁰-methylpteroylglutamic acid (dichloromethotrexate). The foregoing analogs and/or derivatives are also termed “a folate,” “the folate,” or “folates” reflecting their ability to bind to folate-receptors. Such molecules, when conjugated with exogenous molecules, can be effective to enhance transmembrane transport, such as via folate-mediated endocytosis. The foregoing can be used in the folate receptor-binding ligands described herein. L can comprise a chain of atoms from about 3 atoms to about 30 atoms (e.g., about 3 to 30, 3 to about 30, or 3 to 30) LQ^OHQJWK^^^/^FDQ^FRPSULVH^D^FKDLQ^RI^DWRPV^IURP^DERXW^^^Ⴒ^WR^DERXW^ ^^^Ⴒ^^H^J^^^DERXW^^^Ⴒ^^WR^^^^Ⴒ^^^^^Ⴒ^^WR^DERXW^^^^Ⴒ^^^RU^^^Ⴒ^^WR^^^^Ⴒ^^in length. L can comprise a peptide. L can comprise one or more phenylalanine residues, each of which is independently optionally substituted. L can comprise at least one phenylalanyl-phenylalanyl, in which at least one phenyl is independently optionally substituted. L can comprise a polyoligoethylene glycoln (POEGn), a polyethylene glycoln (PEGn), or a mixture thereof, wherein n = 1-36. The conjugate can further comprise a pharmacokinetic modulator. Examples of 70078-02 pharmacokinetic modulators include, but are not limited to, lipophiles, bile acids, steroids, phospholipid analogues, peptides, protein binding agents, polyethylene glycol (PEG), vitamins (e.g., vitamin E or biotin), cholesterol, fatty acids, cholic acid, lithocholic acid, dialkylglyceride, diacylglyceride, phospholipids, and sphingolipids.. In various embodiments, the conjugate further comprises a group that improves tumor uptake of the conjugate, such as a group that comprises or is an albumin-binding moiety. Examples of albumin binders include, but are not limited to, . A ligand can be coupled to the miRNA at various positions, such as the 3ƍ-end, the 5ƍ- end, or an internal position in accordance with methods known in the art and exemplified herein. In various embodiments, the ligand is coupled to the miRNA by a linker. A monomer having a chemical group suitable for participating in a click chemistry reaction can be incorporated, such as an azide- or alkyne-terminated linker. An example of a ligand coupled to an alkyne- terminated linker is shown in FIG.6 and is labeled Folate-DBCO: The folate ligand comprised in “Folate-DABCO” can be coupled via the alkyne into an miRNA having an azide at a terminus of the sense strand or the antisense strand, an example of 70078-02 which is provided in FIG. 1A. The resulting ligand-coupled miRNA can have the formula: , wherein R⁸ comprises a group comprising the miRNA, such as the group: , wherein R⁹ comprises the miRNA, such as an FM-miRNA like FM-miR-34a. An example of a ligand-coupled miRNA can have the formulae:

70078-02 wherein R⁹ comprises or is FM-miR-34a, which is an example of a “FolamiR” (supra). Ligands can be attached one or both strands. On some embodiments, ligands can be conjugated to nucleobases, sugar moieties, or internucleotidic linkages. Still further provided is a composition comprising the miRNA, or a conjugate comprising same, and a pharmaceutically acceptable carrier, diluent, or excipient. “Pharmaceutically acceptable” refers to those carriers, diluents, and excipients, which are, within the scope of sound medical judgment, suitable for use in contact with the tissues of the subject, such as an animal, in particular a human, without excessive toxicity, irritation, allergic response, or other problem or complication, commensurate with a reasonable benefit/risk ratio. The composition can be formulated for administration in solid or liquid form, including those adapted for intravenous, subcutaneous, intratumoral, topical, rectal, vaginal, nasal, pulmonary, ocular, parenteral, oral, sublingual, and transdermal administration. In various embodiments, compositions formulated for subcutaneous or intravenous (e.g., bolus or diffusible infusion) administration are employed. If desired, liposomes, DOPC, gold nanoparticles, and lipid formulations can be employed. Likewise, various methods of encapsulation, as are known in the art, can be employed. A method of treating cancer in a subject is also provided. The method comprises administering to the subject a cancer-treating effective amount of the miRNA, optionally as a composition comprising the miRNA and a pharmaceutically acceptable carrier, diluent, or excipient, whereupon the subject is treated for cancer. The miRNA can be miR-34a (see, e.g., 70078-02 U.S. Pat. Appl. Pub. No.2009/0227533, which is hereby incorporated by reference for its teachings regarding genes affected by miR-34a in cancer cells). The sense strand of the miRNA, which can be miR-34a, can have 15 nucleotides, whereas the antisense strand of the miRNA, which can be miR-34a, can have 22 nucleotides. Each strand of the miRNA can contain an alternating pattern of 2ƍ-O-methyl-modified and 2ƍ-fluoro-modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. In an embodiment, the sense strand can have the sequence: /52FG/*mC*/i2FU/mA/i2FA/mG/i2FA/mC/i2FA/mC/i2FU/mG/i2FC/*mC*/ i2FA//3AzideN/ (SEQ ID NO: 4) and the antisense strand can have the sequence: 5Phos/mU*/i2FG/*mG/i2FC/mA/i2FG/mU/i2FG/mU/i2FC/mU/i2FU/mA/i2FG/mC/i2FU/ mG/i2FG/mU/i2FU/mG*/32FU/ (SEQ ID NO: 5), wherein m is 2ƍ-O-methyl; F is 2ƍ-fluoro; r is ribonucleotide; i is internal; * is a phosphorothioate bond; and Phos: 5ƍ-phosphate. The cancer can be lung, breast, ovarian, or prostate cancer, for example. Further provided is another method of treating cancer in a subject. The method comprises administering to the subject a cancer-treating effective amount of the conjugate comprising miR-34a and a folate, optionally as a composition comprising the conjugate and a pharmaceutically acceptable carrier, diluent, or excipient, whereupon the subject is treated for cancer. The conjugate can further comprise a group that improves tumor uptake of the conjugate, such as a group that comprises or is an albumin-binding moiety. The cancer can be any cancer that overexpresses the folate receptor, such as many epithelial cancers, including cancers of the breast, lung, ovary, kidney, and colon, and various hematological malignancies, such as acute myeloid leukemia. In some embodiments, the cancer can be lung, breast, ovarian, or colorectal cancer, or medulloblastoma. The method also has application in the treatment of diseases involving over-expression of folate receptors/transporters, which are amenable to folate- and 5-methyltetrahydrofolate (5-MTHF)-mediated delivery. Still further provided is another method of treating cancer in a subject. The method comprises administering to the subject a cancer-treating effective amount of the conjugate comprising miR-34a and a ligand that targets prostate cancer, such as DUPA or the ligand present in PSMA-617, optionally as a composition comprising the conjugate and a 70078-02 pharmaceutically acceptable carrier, diluent, or excipient, whereupon the subject is treated for cancer. The conjugate can further comprise a group that improves tumor uptake of the conjugate, such as a group that comprises or is an albumin-binding moiety. . “Cancer-treating effective amount” is an amount of the miRNA, or a conjugate comprising same, that has a therapeutic effect in at least a sub-population of cancerous cells in a subject at a reasonable benefit/risk ratio applicable to any medical treatment. Actual dosage levels of the miRNA, or conjugate comprising same, can be varied to obtain a therapeutic effect in a given subject, taking into consideration the composition, the route of administration, and other factors, such as the age, sex, weight, condition, general health and prior medical history of the subject being treated. The unit dose can be less than 10 mg/kg body weight, such as less than 5, 2, 1, 0.5, 0.1, 0.05, 0.01, 0.005, 0.001, 0.0005, 0.0001, 0.00005 or 0.00001 mg/kg of body weight, and less than 200 nmole of miRNA per kg of body weight, such as less than 150, 125, 100, 75, 50, 25, 15, 7.5, 5.0, 2.5, 1.5, 0.75, 0.15, 0.075, 0.015, 0.0075, 0.0015, 0.00075, or 0.00015 nmole of miRNA per kg of body weight. The unit dose can be administered less frequently than once/day, such as less than every 2, 4, 8, 16 or 30 days. In some embodiments, the unit dose can be administered only once. The unit dose can also be administered with other traditional therapeutic modalities. The conjugate, or the composition comprising the conjugate and a pharmaceutically acceptable carrier, diluent, or excipient, can be administered by any suitable route, such as any suitable route employed in the treatment of cancer. Examples of suitable routes include, but are not limited to, parenterally, e.g., intradermally, subcutaneously, intramuscularly, intraperitoneally, intravenously, or intrathecally. Other routes include bladder infusion, nasal administration, inhalation, buccal absorption, transdermal, rectal and vaginal. Examples of parenteral dosage forms include aqueous solutions of the conjugate in an isotonic saline solution, a glucose solution, or other well-known pharmaceutically acceptable liquid carrier, such as an alcohol, a glycol, an ester, or an amide, suspension, or liposomes. The parenteral dosage form can be in the form of a reconstitutable lyophilizate. Prolonged-release dosage forms, such as biodegradable carbohydrate matrices, can be used. The above methods can be used in combination with other therapies. Examples of such therapies include, but are not limited to, chemotherapy, radiotherapy, immunotherapy, gene 70078-02 therapy, surgery, and the administration of other agents, such as immunomodulatory agents, EGFR-TKI (U.S. Pat. Appl. Pub. No.2014/0309278), sorafenib (e.g., for liver cancer; see U.S. Pat. Appl. Pub. No.2015/0246070), and hormones. Full modification of miRNA enhances stability (e.g., resistance to serum nucleases and increased intracellular half-life). Enhanced stability, coupled with specific, targeted delivery (e.g., folate receptors, such as on folate receptor-overexpressing cancer cells), enable enhanced activity at reduced and less frequent doses. Targeted delivery also reduces, if not eliminates, delivery to non-tumorigenic tissues. The disclosure relates to, among other things, the following enumerated Embodiments, which listing does not represent an order of importance: Embodiment 1. A fully chemically modified microRNA (miRNA), wherein the miRNA is modified with ^ƍ-O-PHWK\O^^^ƍ-fluoro ribose bases, and phosphorothioate linkages. Embodiment 2. The fully chemically modified miRNA of Embodiment1, wherein the miRNA is of the miR-34a family of miRNAs. Embodiment 3. The fully chemically modified miRNA of Embodiment1, wherein at least a portion of the miRNA is double stranded. Embodiment 4. The fully chemically modified miRNA of Embodiment3, wherein each strand can independently range in length from about 12 nucleotides to about 40 nucleotides. Embodiment 5. The fully chemically modified miRNA of Embodiment3, wherein one strand is longer than the other. Embodiment 6. The fully chemically modified miRNA of Embodiment5, wherein one strand is an antisense strand the other strand is an antisense strand and the antisense strand is longer by 1 to 7 nucleotides than the sense strand. Embodiment 7. The fully chemically modified miRNA of Embodiment3, wherein the miRNA comprises a duplex region and a single stranded region. Embodiment 8. The fully chemically modified miRNA of Embodiment7, wherein the duplex region is 12-25 nucleotide base pairs in length and/or the single stranded region is at least about 7 nucleotides in length. 70078-02 Embodiment 9. The fully chemically modified miRNA of Embodiment3, wherein one strand is an antisense strand the other strand is an antisense strand and the sense strand has 15 nucleotides and the antisense strand has 22 nucleotides. Embodiment 10. The fully chemically modified miRNA of Embodiment2, wherein the miRNA comprises a minimum length of 6 nucleotides and a maximum length of 24 nucleotides. Embodiment 11. The fully chemically modified miRNA of Embodiment1, wherein the miRNA comprises at least 6 contiguous nucleotide base pairs present in SEQ ID NO: 1. Embodiment 12. The fully chemically modified miRNA of Embodiment1, wherein the miRNA has at least 80 % identity to SEQ ID NO: 1 or a portion thereof. Embodiment 13. The fully chemically modified miRNA of Embodiment2, wherein the miRNA is of SEQ ID NO: 3, 4, or 14. Embodiment 14. The fully chemically modified miRNA of any one of claims 1-13, wherein each strand contains an alternating pattern of 2ƍ-O-methyl-modified and 2ƍ-fluoro- modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. Embodiment 15. The full chemically modified miRNA of Embodiment1, wherein the sense strand has the sequence: or /5mC/*mC*/mA/i2FG/mC/i2FU/mA/i2FA/mG/i2FA/mC/ i2FA/mC/i2FU/mG/mC*/*mC*/mU/T/T/3AzideN/ (SEQ ID NO: 14) and the antisense strand has the sequence: 5Phos/mU*/i2FG/*mG/i2FC/mA/i2FG/mU/i2FG/mU/i2FC/mU/i2FU/mA/i2FG/mC/i2FU/mG/i2F G/mU/i2FU/mG*/32FU/ (SEQ ID NO: 5) or 5VPPhos/mU*/i2FG/*mG/i2FC/mA/i2FG/mU/i2FG/mU/i2FC/mU/i2FU/mA/i2FG/mC/i2FU/mG/ i2FG*/mU*/i2FU*/mG*/3yU/ (SEQ ID NO: 1) wherein m is 2ƍ-O-methyl; F is 2ƍ-fluoro; r is ribonucleotide; i is internal; * is a 70078-02 phosphorothioate bond; Phos: 5ƍ-phosphate; VP: ^ƍ-vinylphosphonate; and y: extended nucleic acid. Embodiment 16. The full chemically modified miRNA of Embodiment1 comprising a sense sequence and an antisense sequence, the antisense sequence comprises at least 7 contiguous nucleotides of SEQ ID NO: 1. Embodiment 17. A conjugate comprising the fully chemically modified miRNA of any one of claims 1-16 comprising a folate ligand. Embodiment 18. The conjugate of Embodiment17, further comprising a group that improves tumor uptake of the conjugate. Embodiment 19. The conjugate of Embodiment18, wherein the group that improves tumor uptake of the conjugate comprises an albumin-binding moiety. Embodiment 20. The conjugate of Embodiment18, wherein the conjugate comprises a linker. Embodiment 21. The conjugate of Embodiment20, wherein the linker comprises carbonyl, aminoalkyleneamino, carbonylalkylenecarbonyl, thionocarbonyl, alkylene, cycloalkylene, aminoalkylene, alkylenecycloalkyl, alkylenecycloalkylenecarbonyl, aminoalkylenecycloalkylenecarbonyl, alkylenecarbonyl, cycloalkylenecarbonyl, carbonylalkylcarbonyl, and combinations thereof. Embodiment 22. The conjugate of Embodiment20, wherein the linker further comprises a group that improves tumor uptake of the conjugate. Embodiment 23. The conjugate of Embodiment22, wherein the group that improves tumor uptake of the conjugate comprises an albumin-binding moiety. Embodiment 24. A conjugate comprising the fully chemically modified miRNA of any one of claims 1-16 comprising a DUPA or a ligand of the formula:

70078-02 Embodiment 25. The conjugate of Embodiment24, further comprising a group that improves tumor uptake of the conjugate. Embodiment 26. The conjugate of Embodiment25, wherein the group that improves tumor uptake of the conjugate comprises an albumin-binding moiety. Embodiment 27. The conjugate of Embodiment24, wherein the conjugate comprises a linker. Embodiment 28. The conjugate of Embodiment27, wherein the linker comprises carbonyl, aminoalkyleneamino, carbonylalkylenecarbonyl, thionocarbonyl, alkylene, cycloalkylene, aminoalkylene, alkylenecycloalkyl, alkylenecycloalkylenecarbonyl, aminoalkylenecycloalkylenecarbonyl, alkylenecarbonyl, cycloalkylenecarbonyl, carbonylalkylcarbonyl, and combinations thereof. Embodiment 29. The conjugate of Embodiment27, wherein the linker further comprises a group that improves tumor uptake of the conjugate. Embodiment 30. The conjugate of Embodiment29, wherein the group that improves tumor uptake of the conjugate comprises an albumin-binding moiety. Embodiment 31. A composition comprising the fully chemically modified miRNA of any one of claims 1-16 and a pharmaceutically acceptable carrier, diluent, or excipient. Embodiment 32. The composition of Embodiment 31, wherein each strand of the fully chemically modified miRNA contains an alternating pattern of 2ƍ-O-methyl-modified and 2ƍ- fluoro-modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. Embodiment 33. A composition comprising the conjugate of any one of claims 17-23 and a pharmaceutically acceptable carrier, diluent, or excipient. Embodiment 34. A composition comprising the conjugate of any one of claims 24-29 and a pharmaceutically acceptable carrier, diluent, or excipient. Embodiment 35. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the fully chemically modified miRNA of any one of claims 1-15. Embodiment 36. The method of Embodiment35, wherein each strand of the fully chemically modified miRNA contains an alternating pattern of 2ƍ-O-methyl-modified and 2ƍ- fluoro-modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. 70078-02 Embodiment 37. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the composition of Embodiment31. Embodiment 38. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the composition of Embodiment32. Embodiment 39. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the composition of Embodiment33. Embodiment 40. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the composition of Embodiment34. Embodiment 41. The method of Embodiment35, wherein the cancer is lung, breast, ovarian, or prostate cancer. Embodiment 42. The method of Embodiment41, wherein the cancer is prostate cancer. Embodiment 43. The method of Embodiment36, wherein the cancer is lung, breast, ovarian, or prostate cancer. Embodiment 44. The method of Embodiment37, wherein the cancer is lung, breast, ovarian, or prostate cancer. Embodiment 45. The method of Embodiment38, wherein the cancer is lung, breast, ovarian, or prostate cancer. Embodiment 46. The method of Embodiment39, wherein the cancer is lung, breast, ovarian, or prostate cancer. Embodiment 47. The method of Embodiment40, wherein the cancer is lung, breast, ovarian, or prostate cancer. Embodiment EXAMPLES The following examples serve to illustrate the present disclosure and are not intended to limit the scope of the claimed invention in any way. 70078-02 Materials and Methods Cell culture MDA-MB-231 (hereafter referred to as MB-231) and LNCaP cells were obtained from ATCC. MB-231 cells selected for high folate receptor expression were a kind gift from Dr. Philip Low (Purdue University). MB-231-miR-34a reporter cells were generated previously. All the MB-231 strains were cultured in RPMI 1640 medium (no folic acid, Life Technologies), while LNCaP cells (CRL-1740™, ATCC) were cultured in RPMI-1640 medium (30-2001™, ATCC). Both media were supplemented with 10% fetal bovine serum (FBS; Sigma), penicillin (100 U/mL), and streptomycin (100 mg/mL) (HyClone, GE Healthcare Life Sciences). Cells were monitored monthly for lack of Mycoplasma using the MycoAlert Mycoplasma Detection Kit (Lonza). MDA-MB-231 cells overexpressing the folate receptor and MB-231-miR-34a sensor cells were authenticated by ATCC using short tandem repeat profiling. Preparation of miRNA duplexes and serum stability assay Unmodified, partially modified, and fully modified miR-34a duplexes were prepared by annealing corresponding sense and antisense strands at an equal molar ratio in the presence of annealing buffer [10 mM Tris buffer, pH 7 (Sigma), 1 mM EDTA (Sigma), 50 mM NaCl (Sigma)] followed by incubation at 95 °C for 5 minutes, and slow cooling to room temperature. Annealed oligos were then used for cell transfection or otherwise stored at í80°C. To prepare folate-miRNA conjugates (FolamiRs), the azide-containing sense strand was mixed with folate- DBCO (see FIG. 6) at a 1:10 molar ratio (sense strand: folate-DBCO) and was incubated at 23 °C for 10 hours with shaking. The next day, folate-miRNA conjugates were purified using Oligo Clean & Concentrator (Zymo Research) followed by annealing of the antisense strand at equal molar ratio in the presence of annealing buffer as mentioned above. To assess the stability in serum, miR-34a duplexes (50 pmol) were incubated in 50 % FBS (Sigma) at 37 °C for the indicated time points. At each time point, RNA samples were mixed with RNA loading dye and stored at -20 °C. After the last time point, samples were analyzed on a 15% polyacrylamide gel in Glycerol Tolerant Gel Buffer (GTB buffer) followed by staining RNA using Gel Red Nucleic Acid Gel Stain (Thermo Fisher Scientific, Biotium 41003). The sequences of the oligos can be 70078-02 found in Table 1. Table 1: chemical modification patterns and sequences of miR-34a and negative controls PM: partially modified; FM: fully modified: siLuc2 and siLuc+: anti-luciferase siRNAs used as a negative control (NC); miR: miRNA; m: 2ƍ-O-methyl; F: 2ƍ-fluoro; r: ribonucleotide; i: internal; *: phosphorothioate bond; Phos: 5ƍ phosphate In vitro Renilla luciferase assay MB-231 reporter cells were transfected with a negative control (NC) RNA, PM-miR-34a, or FM- miR-34a at the indicated concentrations using Lipofectamine RNAiMAX (Life Technologies). At each time point, Renilla-Glo Luciferase assay (Promega) was performed as per manufacture instructions. In brief, Renilla-Glo Luciferase substrate was mixed with Renilla- Glo buffer at 1:1000 dilution followed by addition into each well. After shaking the plates at room temperature for 10 minutes, Renilla luciferase signal was measured using a GloMax plate reader (Promega). In the case of Renilla luciferase assay following Ago2 knockdown, MB-231 sensor cells were seeded in individual wells of a 96 well plate. The following day, cells were co- transfected with 50 nM siRNA against Ago2 (GeneSolution GS27161; QIAGEN) or a control siRNA (4390846; Thermo Fisher Scientific) along with 10 nM NC, PM-miR-34a, FM-miR-34a 70078-02 duplexes, or miR- 34a mimic (MC11030; Ambion) using Lipofectamine RNAiMAX (Life Technologies). Renilla luciferase assay was performed as described above 48 hours post- transfection. Protein analysis using Western blot MB-231 or LNCaP cells (1 x 10⁵) were seeded in individual wells of a 24 well plate (coated with poly-D-lysine in case of LNCaP) followed by transfection with 50 nM of PM-miR- 34a, FM-miR- 34a, or siLuc2 (negative control) using Lipofectamine RNAiMAX (Life Technologies). To quantify the protein expression of miR-34a targets following Ago2 knockdown, MB-231 or LNCaP cells were seeded in individual wells of a 24 well plate followed by co-transfection with 50 nM siRNA against Ago2 (GeneSolution GS27161; QIAGEN) or a control siRNA (4390846; Thermo Fisher Scientific), along with 50 nM NC, PM-miR-34a, FM- miR-34a duplexes, or miR-34a mimic (MC11030; Ambion) using Lipofectamine RNAiMAX (Life Technologies). At each indicated time point, cells were lysed using RIPA buffer [Tris-HCl (pH 8.0, 50mM), N P-40 (1 %), Sodium chloride (150 mM), Sodium deoxycholate (0.5 %), SDS (0.1 %), ddH2O (up to 100 mL)] in the presence of 1X protease inhibitor cocktail (PIA32955, Thermo Fisher Scientific). Protein concentration was measured using the Pierce BCA Protein Assay kit. Protein lysate (50 Pg) was resolved on 12% TGX gels (Bio-Rad) and transferred to polyvinylidene difluoride (PVDF) membranes. After membrane blocking in LI-COR buffer for 1 hour at room temperature, the membrane was incubated overnight in the indicated primary antibody at 4°C. Following incubation with the corresponding secondary antibody, blots were scanned using Li-Cor Odyssey CLX (Li-Cor). Antibodies used: rabbit Androgen receptor (D6F11) XP (5153, Cell Signaling), rabbit MET (D1C2) XP (8198, Cell Signaling), mouse CD44 (156-3C11) (3570, Cell Signaling^^^PRXVH^ȕ-ACTIN (3700, Cell Signaling), rabbit AXL (C89E7) (8661, Cell Signaling), rabbit GAPDH (14C10) (2118, Cell Signaling). All the antibodies were used at 1:1000 dilution except the following: anti-Ago, clone 2A8 (MABE56; Millipore) and rabbit AXL (C89E7) (8661, Cell Signaling) were used at 1:500 dilution. mRNA quantification using qRT-PCR MB-231 cells (1 x 10⁵) were seeded in individual wells of a 24 well plate. The next day, 70078-02 cells were transfected with 50 nM PM-miR-34a, FM-miR-34a, or siLuc2 (negative control) using Lipofectamine RNAiMAX (Life Technologies). Forty-eight hours later, total RNA was isolated using the miRneasy Kit (217004, Qiagen) according to the manufacturer’s instruction. After removal of genomic DNA using DNase I digestion (79254, Qiagen), RNA integrity was evaluated by resolving on a 1.5% agarose gel. RNA concentration was quantified using a nanodrop. Total RNA (500 ng) was used to generate cDNA using the miScript Reverse Transcriptase kit (218161, Qiagen) using HiFlex buffer per the manufacturer’s instructions. Real- time polymerase chain reaction (qPCR) was performed using the SYBR Green PCR Kit (QIAGEN) with the following primers: Hs_AXL_1_SG, Hs_SIRT1_1_SG, Hs_MET_1_SG, Hs_GAPDH_1_SG, Hs_GNB2L1_2_SG, Hs_TNS4_1_SG, and Hs_ACTB_1_SG (QuantiTect Primer Assay; QIAGEN). Data were then analyzed using the 2^íǻǻ&W method and expressed as fold change. Cell proliferation assays The Sulforhodamine B (SRB, Sigma) assay was used to measure cell proliferation as previously reported. In brief, MB-231 or LNCaP cells were seeded onto individual wells of a 96 well plate (coated with poly-D-lysine in case of LNCaP). The next day, cells were transfected with the various miRNA duplexes (50 nM in case of MB-231 and 10 nM in case of LNCaP) using Lipofectamine RNAiMAX (Life Technologies). At the indicated time points, cells were fixed using 10% tricholoroacetic acid in complete media for 1 hour at 4 °C. Afterward, cells were stained with 0.04% (wt/vol) SRB in 1% acetic acid for 1 hour at 37°C followed by washing unbound dye five times with 1% acetic acid. Unbuffered Tris base (10 mM) was used to extract protein-bound dye and absorbance at 510 nm, which is a proxy for cell mass, was measured using a GloMax Multi+ spectrophotometer (Promega). For clonogenic assays, transfected MB- 231 cells were counted and plated at the density of 250 cells/well in 6 well plate. At the indicated time points, cells were stained using the Differential Quik^® staining kit (Polysciences, cat no. 26419-16). Cell migration and invasion assay For migration assays, 2 x 10⁵ MB-231 cells were seeded in each well of a 6-well plate. 70078-02 After 24 hours, cells were transfected with 5nM PM-miR-34a, FM-miR-34a, or NC (siLuc2) in 50% complete media using Lipofectamine RNAiMAX (Life Technologies) as per manufacturer instructions. Following 72 hours of transfection, the cells were trypsinized, counted and 6 x 10⁴ cells from each treatment were transferred to the apical chamber of 5µm pore-size transwell plates (07-200-149; Fisher Scientific). Basal media was added to the apical chamber and media containing 20% FBS was added to the basolateral chamber. After 12 hours, cells were fixed and stained using Differential Quik^® staining kit (26419-16; Polysciences, Inc.) as per manufacturer’s instructions. For invasion assays, 2 x 10⁵ LNCaP cells were seeded and transfected as mentioned above. Following transfection, 5 x 10⁴ cells were transferred to the apical chamber, coated with 100µl of 200µg/ml of Matrigel matrix (08-774-122; Fisher Scientific) at 37°C for 1 hour, of 8µm pore-size transwell plates (07-200-150; Fisher Scientific). To image migration or invasion chambers, cells were removed from the apical side of the porous membrane using cotton tip applicators and the insert was placed on a glass slide. Four fields were randomly selected, imaged using Olympus IX73 microscope at 10X magnification. Number of cells were quantified using ImageJ v1.53t (NIH). Images were converted to RGB stack and for the highest contrast stack, threshold was adjusted to “0-90” and analyzing particles by setting size (pixel²) to “50-Infinity". The data was compiled and analyzed using GraphPad Prism v9.4.1 (GraphPad Software, LLC). RNA sequencing MB-231 cells (2 × 10⁵) were seeded in individual wells of a 6 well plate. The next day, cells were transfected with the various miRNA duplexes (50 nM) using Lipofectamine RNAiMAX (Life Technologies). RNA was extracted from the cells after 48 hours using mirVana™ RNA Isolation Kit (Thermo Fisher, AM1560) including removal of genomic DNA using DNase I digestion (79254, Qiagen. Samples quantification and purity were determined by nanodrop, and samples integrity was confirmed using Agilent bioanalyzer (Agilent Technology, California USA). RNA sequencing library was prepared using poly A enrichment method using NEBNext® Ultra™ II RNA Library Prep Kit for Illumina® to remove ribosomal RNA. The library was then checked with Qubit and real-time PCR for quantification and bioanalyzer for size distribution detection. RNA sequencing was performed using NovaSeq 6000 platform with a 70078-02 paired end 150 base pair strategy. Bioinformatics analysis The raw reads were trimmed and aligned to GRCh38 (Ensembl release 104). DESeq2 (v1.36.0) was used to normalize the read count and determine differentially expressed genes³⁸. The p-value cutoff for statistically significant genes was 0.05 and no cutoff was used for log₂FC. Volcano plots were plotted using EnhancedVolcano package (v1.14.0) in R. Area-proportional Venn diagrams were adapted from DeepVenn tool³⁹. Heatmaps were generated using pheatmap package (v1.0.12) in R with distance measure set to “Euclidean” and clustering method set to “ward.D2”. Gene set enrichment and miRNA target enrichment analysis was performed using gprofiler2 package (v0.2.1) in R with statistical significance computed using g:SCS algorithm and set to 0.05⁴⁰. Terms with p-adj (corrected p-value) < 0.05 were selected for further analysis. The data for miRNA target enrichment analysis was exported and visualized using ggplot2 package (v3.3.6) in R. Circle plot for selected biological processes and their gene set expression was generated using circlize package (v0.4.15) in R. For miR-34a known/predicted target analysis, targets were exported from miRDB database⁴¹ and overlapped with genes downregulated in PM-miR-34a vs NC and FM-miR-34a vs NC comparisons. GraphPad Prism v9.5.0 (GraphPad Software, LLC) was used to visualize the results. All R analysis was performed using statistically significant genes (p < 0.05) or gene ontology terms (p-adj < 0.05) and was conducted in RStudio environment (v2022.12.0+353). RNA immunoprecipitation assay MB-231 cells (3 x 10⁶) were seeded in 10 cm plates. The next day, two plates of cells were transfected with 10 nM of each of miR-34a mimic (Ambion), PM-miR-34a, FM-miR-34a, or siLuc2 (negative control) duplexes using Lipofectamine RNAiMAX (Life Technologies). The transfection media was replaced with complete media 4 hours post-transfection. Twenty-four hours later, the culture media was discarded, and cells were washed twice with ice-cold PBS before cross linking at 400 mJ/cm² and then again at 200 mJ/cm² using a UV cross-linker (XL- 1000; SpectroLinker). Cell lysis buffer (1 x PBS, 1% vol/vol NP40, 0.5% wt/vol sodium deoxycholate, and 0.1% wt/vol SDS) was added to each plate in the presence of 1x protease 70078-02 inhibitor cocktail (PIA32955, Thermo Fisher Scientific) and RNase inhibitor (AM2696; Invitrogen) for 30 with shaking at 4°C. After shaking the cells were scraped into 1.5 ml microcentrifuge tubes and DNase I (79254, Qiagen) was added to remove genomic DNA. The resulting cell lysates were centrifuged at 16,000 x g for 20 minutes at 4°C and the supernatants were pre- cleared by incubating with 20 ul Dynabead Protein A beads (Life Technologies). Pre- cleared cell lysates were incubated with 2A8 anti-Ago (MABE56; Millipore) or normal mouse IgG (12-371; Millipore) overnight at 4°C. Afterward, Dynabead Protein A beads linked to the bridging antibody, Rabbit anti-PRXVH^,J*^^)FȖ^^1&^^^^^^2; Fisher Scientific) were added to each sample. Samples were incubated for 2 hours at 4°C followed by washing and resuspending the beads as previously described (22). RNA was extracted using Qiazol reagent (Qiagen) followed by ethanol precipitation. The miRscript II RT kit (Qiagen) was used to generate cDNA using the HiSpec buffer followed by quantitative reverse transcription polymerase chain reaction (qRT- PCR) using the SYBR Green PCR Kit (Qiagen). The following primers were used: Mir- 34a-5p (miScript primer assay; Qiagen) and RNU6B (non-target RNA, miScript primer assay; Qiagen. Data were then analyzed using the 2^íǻǻ&W method and expressed as fold change. Tumor implantation and in vivo experiments To evaluate the effect of FM-miR-34a on tumor growth, MB-231 cells were transfected with 50 nM PM-miR-34a, FM-miR-34a, or NC (siLuc2) using Lipofectamine RNAiMAX (Life Technologies) in 10 cm plates. Twenty-four hours later, cells were trypsinized, washed with 1x PBS, and mixed with Matrigel (Corning) at a 1:1 dilution. Cells (5 × 10⁶) were subcutaneously injected into the flank of 8-10-week-old female (NU/J, Foxn1^nu, strain #: 002019, Jackson Lab) mice. A vernier caliper was used to measure tumor volume at the indicated time points which was calculated using the following formula: tumor volume: length × width²/2. For the single dose study using FolamiRs, MB-231 sensor cells (7 x 10⁶) were injected into the flank of 10-12-week-old female (NU/J, Foxn1^nu, strain #: 002019, Jackson Lab) mice, which were maintained on a folate-deficient diet (TD.95247, Envigo) for 1 week prior to treatment and during the course of the experiment. When the tumor volume reached ~200 mm³, mice were treated with a single dose of folate-NC (siLuc2), PM-FolamiR-34a or FM-FolamiR- 34a (1.5 nmol) via tail vein injection. Luminescent signals were captured prior to treatment and 70078-02 over the course of 120 hours using Coelenterazine h Bioluminescent Substrate (PerkinElmer), which was administered intraperitoneally per the manufacturer’s instructions. Whole animal imaging was performed using Spectral AMI (Spectral Instruments). For extraction of protein and RNA from the tumor samples, individual tumors were harvested and stored in RNA later (Life Technologies) at -80 °C until processing. Tumor tissues (50 mg) were disrupted by grinding using liquid nitrogen in a cold mortar. The powder from each tumor sample was transferred into an Eppendorf tube followed by addition of RIPA buffer [Tris-HCl (pH 8.0, 50mM), NP-40 (1 %), Sodium chloride (150 mM), Sodium deoxycholate (0.5 %), SDS (0.1 %), ddH2O (up to 100 mL)] in the presence of 1X protease inhibitor cocktail (PIA32955, Thermo Fisher Scientific). Following centrifugation, an equal amount of protein lysate (50 µg) was resolved on TGX gels (Bio-Rad) followed by analysis of protein by immuno-detection. Total RNA was extracted from the tumor samples using mirVana™ RNA Isolation Kit (Invitrogen™, AM1560). cDNA was prepared using miScript II RT Kit (Qiagen) wiWK^WKH^VXSSOLHG^+L6SHF^%XIIHU^XVLQJ^^^^J^RI^WRWDO^ RNA. A standard curve (1 x 10³ copies to 1 x 10⁸ copies) was generated using the miR-34a mimic (Life Technologies). qRT-PCR reaction (At least 3 technical repeats per biological replicate) was performed using the miScript SYBR Green PCR Kit (Qiagen) and miRNA primer assays (Qiagen) in a QuantStudio 6 Flex Real-time PCR machine (Life Technologies). To determine the efficacy of FM-FolamiR-34a conjugates on tumor growth, MB-231 cells (5 x 10⁶) were injected into the flank of 8-10-week-old female (NU/J, Foxn1^nu, strain #: 002019, Jackson Lab) mice, which were maintained on a folate-deficient diet (TD.95247, Envigo) as mentioned in the single dosing study method. When the tumor volume reached ~150 mm³, mice were treated with folate-NC (siLuc2, n=6), PM-FolamiR-34a (n=5) or FM-FolamiR- 34a (n=6) all at 1.5 nmol via tail vein once every 6 days. Body weight was recorded, and the tumor volume of each mouse was measured every 3 days using a vernier caliper and was calculated using the following formula: tumor volume: (length × width²)/2. All protocols were approved by the Purdue Animal Care and Use Committee and were following the National Institutes of Health (NIH) guidelines for animal use. Quantification of serum cytokines To evaluate potential immune response in vivo, immune competent mice (FVB.129 70078-02 background) were injected with Lipopolysaccharide (LPS, 0.63 mg/kg, intraperitoneally), PBS, FM-FolamiR- 34a or PM-FolamiR-34a (1.5 nmol, n=4, tail vein). Two hours after injections, mice were euthanized, and whole blood was collected. Whole blood was incubated at room temperature for 1 hour, at which time serum was collected by centrifugation at 2000 x g for 10 minutes in a refrigerated centrifuge followed by storage at -80 °C until cytokine analysis. Interleukin-6 (IL-6) and tumor necrosis factor alpha (TNF-Į^^OHYHOV^ZHUH^PHDVXUHG^LQ^WKH^VHUXP^ samples using ELISA Max Deluxe Kit (Biolegend), according to the manufacturer’s instructions. Statistical analysis Statistical analysis was performed using Prism statistical package (GraphPad Software, version 9). The two-tailed Student’s t test was used to determine the statistical difference between two groups. One-way or two-way ANOVA was used to compare the differences between multiple groups and multiple comparisons were corrected using Dunnett’s post hoc test or Tucky’s post hoc test. Data are presented as means ± SD or means ± SEM as specified in the figure legends. Statistically significant p-values are as indicated in the corresponding figure legends. Example 1 Design, synthesis, and in vitro serum stability of partially and fully modified miR-34a When designing a modified RNA oligonucleotide for modulating gene expression, it is necessary to ensure that the incorporated modifications do not interfere with gene silencing. Previously, we synthesized miR-^^D^FRQWDLQLQJ^D^PLQLPDO^QXPEHU^RI^^ƍ-O-methyl modifications to the ribose sugars, which we referred to as partially modified miR-34a (PM-miR-34a). To understand the impact of full chemical modification on miRNA stability and activity, we designed a fully modified miR-34a (FM-miR-34a) in an asymmetric pattern that contained a 22 nucleotide-guide strand annealed to a 15-nucleotide complementary strand, which lowers the thermokinetic stability between the two strands facilitating strand displacement by the RNA Induced Silencing Complex (RISC). This pattern has been previously used to stabilize siRNA ^^^^^(DFK^VWUDQG^FRQWDLQV^DQ^DOWHUQDWLQJ^SDWWHUQ^RI^^ƍ-O-mHWK\O^DQG^^ƍ-fluoro modified sugars and WZR^SKRVSKRURWKLRDWH^OLQNDJHV^DW^WKH^^ƍ^DQG^^ƍ^HQGV^RI^HDFK^VWUDQG^WR^UHGXFH^LPPXQRJHQLFLW\^DQG^ 70078-02 provide exonucleases resistance (FIGS. 1A-1B). PM-miR-34a and FM-miR-34a duplexes were generated and confirmed (FIG.1C). The stability of FM-miR-34a was compared to the stability of PM-miR-34a and unmodified miR-34a duplexes by incubating the duplexes in 50% serum over a time course. While unmodified and PM-miR-34a degraded rapidly following exposure to serum, FM-miR-34a was completely resistant up to 24 hours and remained intact even after 72 hours of incubation (FIGS.1D-1F). Example 2 Comparison of FM-miR-34a to PM-miR-34a on target gene silencing To evaluate the effect of chemical modifications on miR-34a function, we compared the silencing activity of FM-miR-34a to PM-miR-34a on a synthetic target (a 100% complementary sequence) as well as endogenous biological targets of miR-34a. MB-231 cells engineered to express stably a miR-34a complementary sequence downstream of the Renilla gene (MB-231- 34a sensor cells) were used to evaluate the effect on sequences with 100% complementarity. Transfection of MB-231-34a sensor cells with FM-miR-34a or PM-miR-34a significantly downregulated Renilla luciferase expression suggesting that the various chemical modifications do not interfere with miR-34a silencing activity (FIG.2A and FIG.11). Similarly, both miR-34a constructs also downregulated an additional reporter based on firefly luciferase following transient transfection (FIG.2B). We also compared the silencing activity of FM-miR-34a and PM-miR-34a on multiple biological targets in a breast cancer cell line (MB- 231) and a prostate cancer cell line (LNCaP). In MB-231 cells, FM-miR-34a transfection resulted in a more robust downregulation of MET and CD44 (FIG.2C), while in LNCaP cells, the androgen receptor (AR) was similarly downregulated by both FM-miR-34a and PM-miR-34a (FIG.2D). We also compared the effect of FM-miR-34a and PM-miR-34a on the mRNA level of target genes following transfection of MB-231 cells. Both FM-miR-34a and PM-miR-34a significantly downregulated the miR-34a targets AXL, MET, and SIRT1, while neither significantly affected the levels of transcripts not predicted to be miR-34a targets (FIG. 2F). Collectively, these results indicate that the proposed full chemical modifications, when applied to miR-34a, result in similar or enhanced silencing of miR-34a target genes propelling evaluation of the entire transcriptome following transfection with FM-mi-34a. 70078-02 Example 3 Targeting by FM-miR-34a is more efficient and broader relative to targeting by PM-miR-34a To compare the activity of FM-miR-34a and PM-miR-34a at the global level and to evaluate if FM-miR-34a has any unintended off-target effects, we transfected MB-231 cells with either PM-miR-34a or FM-miR-34a, performed RNAseq, and evaluated gene expression and associated biological processes and pathways. Gene expression was more significantly altered in cells transfected with FM-miR-34a relative to cells transfected with PM-miR-34a whether data was normalized to cells transfected with a negative control (Figs.10A-10B) or untransfected cell (FIG. 10A), although there was substantial overlap. Of all the genes altered by PM-miR-34a, 62.2% of the downregulated genes (FIG.10C) and 59.8% of the upregulated genes were also significantly altered by FM-miR-34a. Of the downregulated genes AXL had the lowest p-value following both PM- and FM-miR-34a transfection (see FIG. 10B). In addition to a greater number of genes altered following FM-miR-34a transfection, genes were often more strongly altered in comparison to PM-miR-34a (FIG.10D). To determine if FM-miR-34a was indeed mimicking the targeting activity of endogenous miR-34a, the genes downregulated by FM-miR-34a and PM-miR-34a were evaluated for enrichment of miR-34a targets. The target enrichment analysis was conducted using experimentally validated targets found in the microRNA-target interactions database (mirTarBase). With regard to RNAs downregulated in the PM-miR-34a dataset, the top miRNA predicted to generate the change in transcripts was miR-193b-3p (p-value = 1.5e-24), followed by miR-34a-5p and miR-215-5p (FIG.10E). However, for RNAs downregulated in the FM- miR-34a dataset, the top predicted miRNA was miR-34a-5p (p-value = 6.4e-28) followed by miR-449a and miR-34c-5p – all members of the miR-34 family (FIG.10E). Further analysis verified that FM-miR-34a was mimicking endogenous miR-34a as there was a greater number of miR-34a target genes significantly altered in the FM-miR-34a gene set in comparison to PM- miR-34a gene set (190 vs 137). Because miR-34a regulates multiple cellular processes including cell cycle arrest, cell proliferation, programmed cell death and others^20,21, the major biological processes and pathways regulated by FM-miR-34a and PM-miR-34a was performed on both downregulated and upregulated genes. While the overall enrichment for biological processes, 70078-02 KEGG, and REACTOME terms was similar, PM-miR-34a was better at downregulating cell cycle-related processes. FM-miR-34a, on the other hand, was better at downregulating genes involved in cell proliferation and cell migration and upregulating programmed cell death-related processes (FIG 10F). To further highlight the outstanding targeting ability of FM-miR-34a, known and predicted miR-34a targets were compared between the RNAseq data obtained from PM-miR- 34a and FM-miR-34a transfected cells using the miRDB database. Cells transfected with FM- miR-34a not only had a larger number of miR-34a targets repressed than PM-miR-34a transfected cells (277 vs 191), but transfection with FM-miR-34a also resulted in a more robust downregulation of the targets (FIG.10G). Among the top 100 miR-34a predicted targets, FM-miR-34a downregulated 43% whereas PM-miR-34a downregulated only 27% of them (FIG.10G, inset). These results indicate that FM-miR-34a downregulates more miR-34a targets and causes more robust downregulation in comparison to PM-miR-34a. Example 4 Fully modified miR-34a inhibits cancer cell proliferation, migration and invasion in vitro The impact of full chemical modification of miR-34a on cancer cell proliferation, migration, and invasion was determined. In comparison to PM-miR-34a, FM-miR-34a transfected into MB-231 cells resulted in a significant and stronger inhibition of cell proliferation (FIG.3A) and migration (FIG.3C). In LNCaP cells, both FM-miR-34a and PM- miR-34a significantly inhibited cell proliferation (FIG. 3B) and invasion (FIG.3D) in a similar manner. These results are consistent with the effect of FM-miR-34a and PM-miR-34a on target genes (see FIGS. 2C-2D). To further compare the effect between FM-miR-34a and PM-miR- 34a on MB-231 cell proliferation, a clonogenic assay following transfection was performed. This assay allowed us to determine the effect on cell proliferation over a longer time course, which we hypothesized would be greater for FM-miR-34a due to its stability. As shown in FIG.3E, FM-miR-34a induced a significant inhibition of MB-231 clonogenic capacity, as indicated by smaller colonies and overall reduced number of colonies. Cell proliferation of non- tumorigenic BEAS-2B cells following transfection with either version of miR-34a was not altered, suggesting that the effect on cancer cells might be a result of addicted oncogenic 70078-02 signaling that is suppressed by miR-34a (see FIG.7). Overall, these results indicate that FM- miR-34a induces a comparable, or improved, inhibition of cell proliferation, migration, and invasion of cancer cells relative to PM-miR-34a. Example 5 The activity of the FM-miR-34a is dependent on loading into Argonaute (Ago) There are multiple mechanisms by which synthetic oligonucleotides could reduce the expression of target genes independently of the miRNA-mediated pathway. For example, antisense oligonucleotides can downregulate target genes by triggering RNase H-mediated degradation or by steric hindrance. To validate that FM-miR-34a functions using the same machinery as endogenous miR-34a, the necessity for Argonaute (Ago), the major component of RISC that is essential for endogenous miRNA activity, was assessed. RNA immunoprecipitation was performed in MB-231 cells following transfection of FM-miR-34a, PM-miR-34a, the commercial miR-34a mimic, or NC. Ago-loaded RNA was immunoprecipitated with an anti-Ago antibody followed by miR-34a quantification. Subsequent analysis determined that FM-miR-34a was loaded into Ago as efficiently as PM- miR-34a and the miR-34a mimic, suggesting that the enhanced silencing of FM-miR-34a is not due to better loading into Ago but instead that FM-miR-34a might exhibit higher affinity binding to endogenous targets (FIG.4A). To further confirm the role of Ago in FM-miR activity, MB-231 Ago2 was knocked down and the effect of FM-miR-34a silencing of the Renilla reporter (FIG.4B) or endogenous miR-34a genes (Figs. 4C-4D) was evaluated. The synthetic reporter and endogenous targets were all de-repressed when FM-miR-34a was combined with Ago2 knockdown. Additionally, the inhibitory effect of FM-miR-34a on proliferation, migration, and invasion of both MDA-MB-231 or LNCaP cells was lost when FM-miR-34a was combined with Ago2 silencing (FIGS.4E-4F). Overall, these results indicate that FM-miR-34a is loaded into Ago to mediate target gene silencing of both exogenous as well as endogenous targets leading to phenotypic effects. Example 6 Evaluation of FM-miR-34a activity in vivo 70078-02 To determine the effect of FM-miR-34a on tumor growth and development, MB-231 cells were transfected with FM-miR-34a or PM-miR-34a followed by implanting into immunodeficient mice. As shown in FIG.5A, cells transfected with FM-miR-34a oligos had a significant delay in tumor growth in comparison to cells transfected with PM-miR-34a. Tumors harvested from the FM-miR-34a group were smaller than those harvested from the PM-miR-34a group (Mean tumor weight of 0.12 g versus 0.55 g for PM-miR-34a group, FIG. 5B). Next, a folate-miRNA delivery strategy, which not only provides specific delivery to the tumor, but also completely gets rid of the proposed toxic delivery vehicle, was employed. However, because an encapsulating delivery vehicle is not used, the naked miRNA is subjected to both serum and cellular nucleases, which would mask the full impact of the miRNA if the miRNA were not modified. In this case, using FM-miR-34a we hypothesized a stronger and prolonged anti-tumor effect. To test this hypothesis, we initially validated delivery of folate-near infrared conjugates (folate-NIR, see FIG.8 for synthesis scheme) to folate receptor (FR)- overexpressing breast, cervical, and ovarian cancer cell lines. Folate-NIR conjugates bound specifically to FR expressing-MDA-MB-231, Hela, KB, and IGROV-1 cells and the binding was competed away in the presence of excess folate-glucosamine. Afterward, conjugated PM- miR-34a and FM-miR-34a sense strands were conjugated to folate followed by annealing of the antisense strands to generate PM- FolamiR-34a and FM-FolamiR-34a duplexes. To compare the activity of the FolamiRs, a single dose of PM-FolamiR-34a, FM-FolamiR-34a or folate-NC was injected into the tail vein of nude mice bearing MDA-MB-231 sensor cells. In this case, animals were administered 1.5 nM, a dose 3-fold lower than what was previously used to downregulate Renilla luciferase expression by the first-generation PM-FolamiR-34a conjugates. As observed previously, Renilla was downregulated in animals administered PM- Folamir-34a ~48 hours after systemic injection; however, the signal returned to baseline 24 hours later. Conversely, the signal in FM-FolamiR-34a treated mice was reduced 24 hours post- systemic injection and remained down until at least 96 hours post-injection (FIGS.5C-5D). To assess the effect of FM- FolamiR-34a on biological targets, tumors were harvested 120 hours post-injection, and the expression of various miR-34a targets was evaluated. MET, CD44 and AXL protein levels were significantly reduced in the FM-FolamiR-34a treated group in comparison to tumors harvested from PM-FolamiR-34a or folate-NC groups, confirming the 70078-02 ability of FM-miR-34a to silence its biological targets strongly in vivo (FIG. 5E). Higher miR- 34a copy number was detected in tumors collected from FM-FolamiR-34a treated mice in comparison to mice treated with a similar dose of PM-FolamiR-34a or folate-NC, likely due to the enhanced stability of FM-miR- 34a rather that poor cDNA synthesis (FIG.5F, FIG.9). Next, the efficacy of FM-FolamiR-34a and PM-FolamiR-34a was evaluated in MDA- MB-231 tumor-bearing mice. Folate-conjugates were administered at 1.5 nmol once every 6 days based on the effect of FM-FolamiR-34a on its biological targets at five days post-injection following a single dose administration (see FIG.5E). While PM-FolamiR-34a administration resulted in a delay in tumor growth (average tumor volume ~ 1.5-fold vs 3-fold for folate-NC), FM-FolamiR-34a significantly inhibited tumor growth with an average tumor volume that was less than, or equal to, the first day of treatment until the end of the 21-day study. It is also worth mentioning that tumors in two of the mice administered FM-FolamiR-34a shrunk to ~25-50% of their initial volume, and one mouse was a complete cure with no tumor tissue remaining 54 days post-treatment. Importantly, no significant changes were observed in the body weight throughout the study, suggesting the safety of FM-FolamiR-34a (FIG.5H). As a preliminary evaluation of the potential immune response, FM-FolamiR-34a or PM-FolamiR-34a were injected into the tail vein of immunocompetent mice (FVB.129 background) followed by quantification of IL-6 and TNF-Į^F\WRNLQHV^OHYHOV^LQ^WKH^VHUXP^^^KRXUV^SRVW-injection. In comparison to the positive control, mice injected with LPS, neither FM-FolamiR-34a or PM-FolamiR-34a resulted in a significant increase in cytokine levels above the negative control (FIG.5I). Collectively, these results indicate that FM-FolamiR-34a induces stronger and prolonged silencing of both synthetic and biological targets of miR-34a resulting in a significant delay in tumor growth in comparison to PM-FolamiR-34a in vivo. Discussion The stability and the activity of partially and fully modified miR-34a were directly compared using lipid transfection and folate-mediated miRNA delivery approaches. Full chemical modification of miR-34a extended its stability in comparison to unmodified or partially modified miR-34a. FM-miR-34a induced a stronger silencing of MET and CD44 70078-02 protein expression in MDA-MB-231 breast cancer cells when compared to PM-miR-34a. Consistent with that, FM-miR-34a transfected cells showed significant reduction in their ability to migrate. In the prostate cancer cells (LNCaP), FM-miR-34a induced a similar downregulation of AR protein expression and a comparable inhibition of invasion. The difference in the magnitude of silencing of MET, CD44 and AR by FM-miR-34a suggests that the effect of FM-miR-34a on different genes might be sequence-dependent. This also suggests that the full chemical modification approach can be generalized to induce targeting of multiple genes and in different cell lines and provides a rationale for future optimization of the chemical modification to achieve a better effect in case of LNCaP cells. The necessity of endogenous Ago for the function of FM-miR-34a was validated using immunoprecipitation and cellular activity assays. Using the more clinically relevant approach of miRNA delivery (FM-FolamiR-34a), the effect of chemical modifications on miR-34a activity in vivo was evaluated. Previously, it had been demonstrated that a dose of 5 nmol was needed to downregulate Renilla luciferase expression (used as a proxy for miR-34a activity) in vivo, and the effect only lasted for a few hours. Here a lower dose (more than 3-fold less, 1.5 nmol) was used to compare the effect of folate-PM-miR-34a with folate-FM-miR-34a. The data showed that FM-FolamiR-34a showed a stronger downregulation of Renilla luciferase expression, which stayed repressed for a longer time than with folate-PM-miR-34a. Also, systemically administered FM-FolamiR-34a (single dose, 1.5 nmol) showed a striking silencing of the biological targets (MET, CD44 and AXL) of miR-34a after five days. In addition, more FM-miR-34a was present in the tumor compared to PM-miR-34a, suggesting the enhanced stability of FM-miR-34a. Enhancing miRNA stability and subsequently prolonging its silencing effect could potentially eliminate the need of endosomal escape agents to release the RNA from the endosomes before getting degraded. This was seen in case of fully modified siRNA conjugated to N- acetylgalactosamine (GalNAc) ligand (GalNAc-siRNA). GalNAc-siRNA has a durable activity in vivo due, in part, to its enhanced stability and slow release from the acidic intracellular compartments. Overall, mRGLILFDWLRQ^RI^PL51$^GXSOH[HV^XVLQJ^^ƍ-O-methyl, and ^ƍ-fluoro ribose bases and phosphorothioate linkages enhanced both stability and activity of the miRNA. The combination of folate ligand and full chemical modifications will be beneficial to 70078-02 reduce the effective dose, and to avoid toxic side effects resulting from non-specific uptake or higher miRNA doses. All patents, patent application publications, journal articles, textbooks, and other publications mentioned in the specification are indicative of the level of skill of those in the art to which the disclosure pertains. All such publications are incorporated herein by reference to the same extent as if each individual publication were specifically and individually indicated to be incorporated by reference. In the event of inconsistent usages between this document and those documents so incorporated by reference, the usage in the incorporated reference should be considered supplementary to that of this document; for irreconcilable inconsistencies, the usage in this document controls. The invention illustratively described herein may be suitably practiced in the absence of any element(s) or limitation(s), which is/are not specifically disclosed herein. Thus, for example, each instance herein of any of the terms "comprising," "consisting essentially of," and "consisting of" may be replaced with either of the other two terms. Likewise, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. Thus, for example, references to "the method" includes one or more methods and/or steps of the type, which are described herein and/or which will become apparent to those ordinarily skilled in the art upon reading the disclosure. The term "or" is used to refer to a nonexclusive "or" unless otherwise indicated. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood to one of ordinary skill in the art. The following terms and phrases shall have the meaning indicated. The term "about," when referring to a number or a numerical value or range (including, for example, whole numbers, fractions, and percentages), means that the number or numerical range referred to is an approximation within experimental variability (or within statistical experimental error) and thus the numerical value or range can vary between 1% and 15% of the stated number or numerical range (e.g., +/- 5 % to 15% of the recited value, such as within 10%, within 5%, or within 1% of a stated value or stated limit of a range) provided that one of ordinary skill in the art would consider equivalent to the recited value (e.g., having the same function or result). The term "substantially" can allow for a degree of variability in a value or range, for 70078-02 example, within 90%, within 95%, 99%, 99.5%, 99.9%, 99.99%, or at least about 99.999% or more of a stated value or of a stated limit of a range. In addition, it is to be understood that the phraseology or terminology employed herein, and not otherwise defined, is for the purpose of description only and not of limitation. Any use of section headings is intended to aid reading of the document and is not to be interpreted as limiting. Further, information that is relevant to a section heading may occur within or outside of that particular section.

Claims

70078-02 WHAT IS CLAIMED IS: 1. A fully chemically modified microRNA (miRNA), wherein the miRNA is modified with ^ƍ-O-PHWK\O^^^ƍ-fluoro ribose bases, and phosphorothioate linkages. 2. The fully chemically modified miRNA of claim 1, wherein the miRNA is of the miR- 34a family of miRNAs. 3. The fully chemically modified miRNA of claim 1, wherein at least a portion of the miRNA is double stranded. 4. The fully chemically modified miRNA of claim 3, wherein each strand can independently range in length from about 12 nucleotides to about 40 nucleotides. 5. The fully chemically modified miRNA of claim 3, wherein one strand is longer than the other. 6. The fully chemically modified miRNA of claim 5, wherein one strand is an antisense strand the other strand is an antisense strand and the antisense strand is longer by 1 to 7 nucleotides than the sense strand. 7. The fully chemically modified miRNA of claim 3, wherein the miRNA comprises a duplex region and a single stranded region. 8. The fully chemically modified miRNA of claim 7, wherein the duplex region is 12-25 nucleotide base pairs in length and/or the single stranded region is at least about 7 nucleotides in length. 70078-02 9. The fully chemically modified miRNA of claim 3, wherein one strand is an antisense strand the other strand is an antisense strand and the sense strand has 15 nucleotides and the antisense strand has 22 nucleotides. 10. The fully chemically modified miRNA of claim 2, wherein the miRNA comprises a minimum length of 6 nucleotides and a maximum length of 24 nucleotides. 11. The fully chemically modified miRNA of claim 1, wherein the miRNA comprises at least 6 contiguous nucleotide base pairs present in SEQ ID NO: 1. 12. The fully chemically modified miRNA of claim 1, wherein the miRNA has at least 80 % identity to SEQ ID NO: 1 or a portion thereof. 13. The fully chemically modified miRNA of claim 2, wherein the miRNA is of SEQ ID NO: 3, 4, or 14. 14. The fully chemically modified miRNA of any one of claims 1-13, wherein each strand contains an alternating pattern of 2ƍ-O-methyl-modified and 2ƍ-fluoro-modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. 15. The full chemically modified miRNA of claim 1, wherein the sense strand has the sequence: /52FG/*mC*/i2FU/mA/i2FA/mG/i2FA/mC/i2FA/mC/i2FU/mG/i2FC/*mC*/ i2FA//3AzideN/ (SEQ ID NO: 4) or /5mC/*mC*/mA/i2FG/mC/i2FU/mA/i2FA/mG/i2FA/mC/ i2FA/mC/i2FU/mG/mC*/*mC*/mU/T/T/3AzideN/ (SEQ ID NO: 14) and the antisense strand has the sequence: 5Phos/mU*/i2FG/*mG/i2FC/mA/i2FG/mU/i2FG/mU/i2FC/mU/i2FU/mA/i2FG/mC/i2FU/mG/i2F 70078-02 G/mU/i2FU/mG*/32FU/ (SEQ ID NO: 5) or 5VPPhos/mU*/i2FG/*mG/i2FC/mA/i2FG/mU/i2FG/mU/i2FC/mU/i2FU/mA/i2FG/mC/i2FU/mG/ i2FG*/mU*/i2FU*/mG*/3yU/ (SEQ ID NO: 1) wherein m is 2ƍ-O-methyl; F is 2ƍ-fluoro; r is ribonucleotide; i is internal; * is a phosphorothioate bond; Phos: 5ƍ-phosphate; VP: ^ƍ-vinylphosphonate; and y: extended nucleic acid. 16. The full chemically modified miRNA of claim 1 comprising a sense sequence and an antisense sequence, the antisense sequence comprises at least 7 contiguous nucleotides of SEQ ID NO: 1. 17. A conjugate comprising the fully chemically modified miRNA of any one of claims 1-16 comprising a folate ligand. 18. The conjugate of claim 17, further comprising a group that improves tumor uptake of the conjugate. 19. The conjugate of claim 18, wherein the group that improves tumor uptake of the conjugate comprises an albumin-binding moiety. 20. The conjugate of claim 18, wherein the conjugate comprises a linker. 21. The conjugate of claim 20, wherein the linker comprises carbonyl, aminoalkyleneamino, carbonylalkylenecarbonyl, thionocarbonyl, alkylene, cycloalkylene, aminoalkylene, alkylenecycloalkyl, alkylenecycloalkylenecarbonyl, aminoalkylenecycloalkylenecarbonyl, alkylenecarbonyl, cycloalkylenecarbonyl, carbonylalkylcarbonyl, and combinations thereof. 70078-02 22. The conjugate of claim 20, wherein the linker further comprises a group that improves tumor uptake of the conjugate. 23. The conjugate of claim 22, wherein the group that improves tumor uptake of the conjugate comprises an albumin-binding moiety. 24. A conjugate comprising the fully chemically modified miRNA of any one of claims 1-16 comprising a DUPA or a ligand of the formula: . 25. The conjugate of claim 24, further comprising a group that improves tumor uptake of the conjugate. 26. The conjugate of claim 25, wherein the group that improves tumor uptake of the conjugate comprises an albumin-binding moiety. 27. The conjugate of claim 24, wherein the conjugate comprises a linker. 28. The conjugate of claim 27, wherein the linker comprises carbonyl, aminoalkyleneamino, carbonylalkylenecarbonyl, thionocarbonyl, alkylene, cycloalkylene, aminoalkylene, alkylenecycloalkyl, alkylenecycloalkylenecarbonyl, aminoalkylenecycloalkylenecarbonyl, alkylenecarbonyl, cycloalkylenecarbonyl, carbonylalkylcarbonyl, and combinations thereof. 29. The conjugate of claim 27, wherein the linker further comprises a group that improves tumor uptake of the conjugate. 70078-02 30. The conjugate of claim 29, wherein the group that improves tumor uptake of the conjugate comprises an albumin-binding moiety. 31. A composition comprising the fully chemically modified miRNA of any one of claims 1-16 and a pharmaceutically acceptable carrier, diluent, or excipient. 32. The composition of claim 31, wherein each strand of the fully chemically modified miRNA contains an alternating pattern of 2ƍ-O-methyl-modified and 2ƍ-fluoro-modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. 33. A composition comprising the conjugate of any one of claims 17-23 and a pharmaceutically acceptable carrier, diluent, or excipient. 34. A composition comprising the conjugate of any one of claims 24-29 and a pharmaceutically acceptable carrier, diluent, or excipient. 35. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the fully chemically modified miRNA of any one of claims 1-15. 36. The method of claim 35, wherein each strand of the fully chemically modified miRNA contains an alternating pattern of 2ƍ-O-methyl-modified and 2ƍ-fluoro-modified sugars and a phosphorothioate linkage at the 5ƍ and 3ƍ ends of the strand. 37. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the composition of claim 31. 38. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the composition of claim 32. 70078-02 39. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the composition of claim 33. 40. A method of treating cancer in a subject, which method comprises administering to the subject a cancer-treating effective amount of the composition of claim 34. 41. The method of claim 35, wherein the cancer is lung, breast, ovarian, or prostate cancer. 42. The method of claim 41, wherein the cancer is prostate cancer. 43. The method of claim 36, wherein the cancer is lung, breast, ovarian, or prostate cancer. 44. The method of claim 37, wherein the cancer is lung, breast, ovarian, or prostate cancer. 45. The method of claim 38, wherein the cancer is lung, breast, ovarian, or prostate cancer. 46. The method of claim 39, wherein the cancer is lung, breast, ovarian, or prostate cancer. 47. The method of claim 40, wherein the cancer is lung, breast, ovarian, or prostate cancer.