Accessory microRNA byproducts expand RNA interference via microprocessor-mediated cleavage activation

Debora Mazzetti; Michal O Nowicki; Himanshu Soni; Joshua D Bernstock; Maya Groff; Luisa Esposito; Diego A Hernandez; Lucia Altucci; Anna Krichevsky; Geoffrey Fell; E Antonio Chiocca; Hiroshi Nakashima; Marco Mineo; Pierpaolo Peruzzi

doi:10.1126/sciadv.adz8331

. 2026 Feb 13;12(7):eadz8331. doi: 10.1126/sciadv.adz8331

Accessory microRNA byproducts expand RNA interference via microprocessor-mediated cleavage activation

Debora Mazzetti ^1,^2,³, Michal O Nowicki ^1,², Himanshu Soni ^1,², Joshua D Bernstock ^1,², Maya Groff ⁴, Luisa Esposito ⁵, Diego A Hernandez ¹, Lucia Altucci ³, Anna Krichevsky ^2,⁶, Geoffrey Fell ⁷, E Antonio Chiocca ^1,², Hiroshi Nakashima ^1,², Marco Mineo ^1,², Pierpaolo Peruzzi ^1,^2,^*

PMCID: PMC12904199 PMID: 41686893

Abstract

RNA medicine is a promisingly expanding field in modern health care, but its use in genetically complex diseases, like cancer, has been challenging, mainly due to their reliance on multiple abnormal pathways. Here, we describe a microRNA-based platform that exploits previously unrecognized features of microRNA processing. Leveraging a microprocessor-dependent, cleave-activation strategy, this design allows us to expand biological impact by simultaneously up- and down-regulating desired microRNAs, while using them as structural enablers for other short, noncoding RNAs, such as aptamers. We demonstrate its biological potential in a glioblastoma model, where the simultaneous bidirectional modulation of five among the most deregulated microRNAs results in critical mass interference against the tumor. In parallel, microRNA-mediated chaperoning of an anti-p50 aptamer within the platform allows us to selectively block the nuclear factor κB pathway, a difficult-to-drug target. This work highlights the potential of chimeric microRNA clusters as an emerging therapeutic concept for cancer and other similarly multifactorial diseases.

Insights into microRNA processing guide the design of an RNA medicine strategy that enables precise multitargeting against cancer.

INTRODUCTION

Cancer is the epitome of a genetically complex disease, characterized by the interplay of multiple, adaptable, and often redundant abnormal pathways (1, 2). This is particularly true for glioblastoma (GBM), the deadliest brain cancer in adults, where a plethora of concomitantly active oncogenes (3) determine its characteristic heterogeneity and resistance to treatments, negating the efficacy of any targeted therapies attempted so far (4). The discovery and characterization of noncoding RNAs (ncRNAs) have further expanded the number of relevant aberrancies present in GBM and cancer in general (5), increasing this complexity without contributing to promising therapeutic solutions, yet. However, as RNA medicine is starting to show its potential for the treatment of monogenic diseases such as many congenital metabolic disorders (6, 7), it also provides the framework for newer strategies to combine precision with multitargeting needs. On one hand, RNA guarantees stringent target recognition, as postulated by the Watson and Crick rules (8), allowing one to design sequences virtually specific to any desired targets. Combinatorial targeting, instead, can be provided by microRNA genes, owing to their small size and clustered nature (9), and facilitated by permissive structural requirements (10, 11). Accordingly, cells recognize engineered RNA sequences as microRNAs as long as they form stereotypically shaped stem-loops (11).

Here, starting from the uncovering of accessory byproducts derived from microRNA processing, we demonstrate the rationale and different declinations of an ncRNA gene therapy platform. This is capable of carrying multiple, functionally different elements, which become biologically active upon cleavage from the primary sequence where they are encoded. As a proof of principle, we validate its translational potential in various models of GBM, a poorly druggable brain cancer.

RESULTS

microRNA processing releases accessory, elusive small ncRNA sequences

We performed short [<200 nucleotides (nt)] ncRNA sequencing from GBM cells expressing a previously described three-microRNA cluster [miR-128, miR-124, and miR-137 (Cluster 3, CL3) engineered from a miR-17-92 backbone] (9). In addition to the three mature microRNAs, we detected consistent reads mapping to the 5′ and 3′ flanking regions of the transgene (Fig. 1, A and B). This was unexpected since it is generally accepted that the only relevant component of any microRNA gene is its ~70-nt hairpin (pre-miR), which is cleaved out from the longer primary sequence (pri-miR) by microprocessor (Drosha and DGCR8) and then further processed by Dicer into a mature microRNA (12). This initial observation was confirmed using a second cell line [human embryonic kidney (HEK) 293 cells], which was infected with a lentivector carrying either CL3 or a green fluorescent protein (GFP) control transgene (fig. S1A). To further prove that this signal was not the result of spurious detection of the unprocessed primary sequence, we also analyzed RNA recovered from exosomes (fig. S1B), taking advantage of their enrichment for microRNAs and other short ncRNA (13). Here, reverse transcription quantitative polymerase chain reaction (RT-qPCR) confirmed the presence of the cleaved flanking regions but not of the primary sequence (fig. S1C). Furthermore, impairment of microRNA processing by Drosha knockout (KO) abrogated their detection, without affecting other exosome characteristics (fig. S1, C to E). Northern blot using specific probes for each flank confirmed the presence of two cleaved bands of ~200 nt, in addition to a ~1.4-kb band corresponding to the uncleaved primary sequence (Fig. 1C). The total length of the detected primary sequence was characterized by PCR, using progressively divergent primers, and confirmed to have a length of 1.35 kb (fig. S2A). The specificity of the probes was demonstrated by the absence of any detectable signal in control cells expressing GFP only (Fig. 1C). As an additional control, we used a third transgene expressing a scrambled CL3 (SCR), with nucleotide changes within the hairpin stem of each of the three microRNAs, which prevent cleavage by the microprocessor (fig. S2B) (7). This only produced the 1.35-kb band and not the smaller bands, validating them as cleaved products of the pri-miR transgene and dependent on correct microRNA processing (fig. S2C). The precise size of both flanking regions was then characterized by a progressively divergent primer approach as described above, using RNA enriched for fragments <500 nt to eliminate false positive signals from the longer uncleaved primary sequence (fig. S2, D to F).

Fig. 1. — (A) RNA structure prediction of CL3 construct. Each component is color-coded and labeled. (B) Volcano plot showing changes in short ncRNA from G34 patient-derived GBM stem-like cells expressing GFP-CL3 transgene versus GFP-only (CTRL) transgene. Results are based on three biological replicates per group. Gray dots represent P values > 0.05. (C) Agarose gel (1%) Northern blot of whole cell RNA obtained from HEK293 cells expressing CL3 and CTRL, same as described in (B). Cartoon shows the target of each tested probe within the transgene sequence. Stars denote the primary cluster sequence. Arrowheads denote cleaved flanking regions of the primary transcript. (D) RT-qPCR from RNA of G34 cells expressing CL3 grown in normal medium or in the presence of actinomycin D (10 μg/ml) for 6 hours. Color-coded RNA structures represent the predicted folding of the 3′ and 5′ flanks, respectively. Bars represent mean values + SD; n = 3 biological replicates. Each comparison was analyzed with Student’s t test, two tails. FC, fold change; RQ, relative quantification. (E) RT-qPCR of G34 cells expressing a modified CL3 sequence where the 5′ flank is replaced with a Tough Decoy (TuD), whose predicted structure is shown above. Cells were grown in normal medium or in the presence of actinomycin D (10 μg/ml) for 6 hours. Bars represent mean values + SD; n = 3 biological replicates. Each comparison was analyzed with Student’s t test, two tails. (F) Rationale for the design of a new microRNA-based platform. ****P < 0.0001; ns, nonsignificant.

To further explore the biological relevance of this observation and dissect whether they are only a transient byproduct or a more stable component, HEK293 cells were treated at different time points with actinomycin D to block RNA transcription. As expected, the primary sequence disappeared within 1 hour of treatment, as did the mRNA for ubiquitin C (UBC), while the small flanking segments remained detectable for at least 6 hours, behaving similarly to mature miR-124, which was used as a positive control for its well-known stability to degradation (Fig. 1D and fig. S3A) (14).

We hypothesized that the observed superior stability of the cleaved 3′ flank sequence compared to the 5′ one could be due to a more structured RNA configuration (15). To test this, the 5′ flanking region was replaced with an artificial TuD (Tough Decoy) sequence, which is folded as a double-stranded RNA (16). This modification resulted in a significant increase in its stability after actinomycin D treatment (Fig. 1E), while not interfering with the normal processing of the pri-microRNA sequence (fig. S3B). Together, these data suggest an additional layer in microRNA biology and the resulting opportunity to take advantage of it (Fig. 1F).

Repurposing of flanking sequences to expand the biological impact of microRNA genes

We hypothesized that these flanking regions, properly modified, could provide the unique opportunity to modulate the expression of multiple microRNAs bidirectionally. This would be crucial in a disease, like GBM, which is characterized by simultaneous down-regulation of tumor suppressor microRNAs and overexpression of oncomiRs (Fig. 2A). Since TuDs have been previously used to sponge microRNAs (17), we designed two progressively complex versions of CL3. The first is CL3_a1, with an anti-miR-21 TuD (Fig. 2B and fig. S4A) replacing the 5′ flank, and the second is CL3_a2, where an additional TuD against miR-210 (Fig. 2B and fig. S4B) is used to replace the 3′ flank of the original CL3 transgene. Targeting of miR-21 was chosen as the most common and best characterized oncomiR in GBM and many other cancers (18) and because of its demonstrated role in chemotherapy resistance in GBM (19). In addition, targeting of miR-210, another validated oncomiR in GBM (20), offered the possibility to use its specific hypoxia-related phenotype (21) to functionally validate the effects of its inhibition by our constructs and, therapeutically, to disrupt a well-described tumor pathway, which allows GBM to thrive in its usual hypoxic environment. RNA folding prediction of the resulting transgenes shows that the fundamental structure imparted by the three microRNA hairpins is not distorted by either flank modifications (Fig. 2B), and expression of the transgene in HEK293 cells results in proper processing, yielding high amounts of mature miR-124 and of the cleaved TuD-21 and TuD-210 fragments (Fig. 2C). Conversely, abrogation of microRNA processing using a Drosha KO model results in loss of the cleaved TuDs and persistence of the primary sequence only (Fig. 2D). RT-qPCR, performed in multiple cell lines, shows comparable levels of the three overexpressed microRNAs across all transgenes, while miR-21 and miR-210 are successfully down-regulated when the specific TuD is expressed (Fig. 2E and fig. S4, C and D). The measured down-regulation is at the same order of magnitude observed with either anti-miR oligonucleotides or a short hairpin RNA (shRNA)–like approach (miRZip) (fig. S4E).

Fig. 2. — (A) Volcano plot showing the most up-regulated and deregulated microRNAs in GBMs (n = 675) versus normal brain (n = 55). MicroRNAs used in following experiments are marked and color-coded. Gray dots represent P values >0.01. (B) Schematics and corresponding predicted RNA folding of engineered microRNA sequences. Each component is labeled and color-coded accordingly. (C) Polyacrylamide gel (4 to 20%) Northern blot from whole cell RNA of HEK293 cells transfected with different constructs. (D) Agarose gel (1%) Northern blot of whole cell RNA of CL3_a2 transgene expressed either in HEK293 cells Drosha WT or HEK293 cells Drosha KO. (E) microRNA expression in G34 cells infected with different constructs. Bars represent mean values + SD; n = 3 biological replicates. Statistical analysis was performed by two-way analysis of variance (ANOVA) compared with SCR with Dunnett’s multiple comparisons test. (F) Luminescence detection in HEK293 cells cotransfected with clusters and specific microRNA-responsive luciferase reporter genes (PSY-21, luciferase responsive to miR-21; PSY-210, luciferase responsive to miR-210). Bars represent mean values ± SD; n = 3 biological replicates. Statistical analysis was performed by one-way ANOVA compared with SCR with Dunnett’s multiple comparisons test. Cartoon represents the approach design for the luciferase reporter gene generation. mi-TS, microRNA target sequence. (G) Representative Western blot from whole cell protein lysate of G34 cells stably expressing different constructs. *P < 0.05; **P < 0.01; ****P < 0.0001; ns, nonsignificant.

Functionally, successful sponging of each microRNA was selectively interrogated and confirmed using a luminescence reporter system, where the seed sequence for miR-21 or miR-210 was cloned in the 3′ untranslated region (3′UTR) of a Renilla luciferase gene (Fig. 2F). Both CL3_a1 and CL3_a2 achieved a level of interference comparable to that obtained with anti-miR oligonucleotide transfection (fig. S4F). The appropriate changes of known targets of each modulated microRNAs were confirmed by Western blot (Fig. 2G) and RT-qPCR (fig. S4G).

At the cellular level, the increase in RNA interference resulted in progressive suppression of GBM cell growth in soft agar (Fig. 3, A and B). In addition, we demonstrate the specific effect of different microRNA perturbations, as inhibition of miR-21 severely affects cell motility (Fig. 3, C and D) and migration (Fig. 3E and fig. S5A), while inhibition of miR-210 greatly impairs cell survival under hypoxic conditions (Fig. 3, F to H, and fig. S5B) but not at physiologic oxygen concentrations (Fig. 3, A and B, and fig. S5C). Similar results were observed with an additional patient-derived glioma stem-like cells (GSCs) (fig. S5, D to F).

Translational relevance of combinatorial microRNA interference

Patient-derived GBM stem-like cells stably expressing lentiviral cassettes to modulate either single target microRNAs or their combinations were implanted intracranially into athymic nude mice. Animals implanted with tumors expressing an SCR reached terminal endpoint by day 13, at which time one mouse per group was euthanized and the size of the tumor analyzed by fluorescence microscopy (Fig. 4A). Mice implanted with CL3 had a median survival of 25 days, similar to our prior results (9); the CL3_a1 mice lived for a median of 66 days, whereas none of the CL3_a2 mice reached endpoint by day 150, when they were euthanized, and microscopic examination of the brain revealed no evidence of tumor (Fig. 4B), despite initial proof of successful implantation (fig. S6A). The apparent tumor involution observed with CL3_a2 was further substantiated with cleaved caspase-3 (CC3) immunostaining (fig. S6, B and C). Comparatively, mice implanted with tumors expressing single microRNA–modifying cassettes showed significant, albeit marginal benefits compared to SCR (fig. S6D). Together, this proves the significant synergism achieved by this clustered modulation approach (Fig. 4C). Moreover, a parallel, independent survival study also including the administration of DNA alkylating drug temozolomide (TMZ) confirmed not only that CL3_a2 mice did not develop tumor, but that this was also the case for the CL3_a1 group (fig. S6E), demonstrating a notable chemosensitizing effect of this miR-cluster strategy (9). A second GBM stem-like cell was used to confirm the severe impairment of intracranial tumor growth by the CL3_a2 construct (fig. S7, A and B).

Fig. 4. — (A) Fluorescence microscopy imaging of representative brain slices from mice implanted with G34 cells expressing different microRNA-modulating transgenes, all euthanized at day 13. Tumor is visualized by GFP expression over a Hoechst staining. Magnifying insets in the brain of CL3_a2 show evidence of scant tumor cells. Scale bars, 1000 μm (full brain); 20 μm (inset). (B) Kaplan-Meier survival curve of female athymic nu/nu mice intracranially implanted with 10,000 G34 cells stably expressing the different microRNAs expressing constructs. Six mice per group. Log rank test, corrected by Bonferroni analysis. Multiple sections of a CL3_a2 brain are presented to show no tumor presence at day 150. This analysis did not include animals used in (A). (C) Analysis of synergism among different microRNA combinations. Each bar represents the fold change gain in median survival of tumors expressing the specified transgene compared to SCR control. The horizontal black line represents the median survival of SCR, while dotted colored lines represent the expected survival if each specific microRNA combinations resulted in an additive effect only. Vertical black arrowed lines represent the extent of synergism shown for CL3, CL3_a1, and CL3_a2. Since no mice died in the CL3_a2 group, the median survival was established at day 150. **P = < 0.01; ***P = < 0.001; ****P < 0.0001.

Neither expression of CL3_a2 in human astrocytes nor direct administration of the transgene into mouse brain resulted in any impairments of cell viability, nor adverse events in animals, excluding the possibility that its observed effect on tumor cells is due to nonspecific toxicity (fig. S8, A to E). We speculate that this is at least in part due to the fact that the CL3_a2 cassette is not able to significantly modify the baseline expression of microRNAs in the context of normal brain, where their abundance is already profoundly up-regulated (for mir-124, miR-128, and miR-137) or down-regulated (for miR-21 and miR-210) compared to GBM (fig. S8E).

microRNA genes are amenable to further repurposing in the stem-loop region

The evidence of unaffected processing into mature microRNAs despite profound changes in their primary sequence led us to hypothesize that this platform could support further modifications in other regions. We reasoned that as long as the pre-miR is maintained as a stem-loop structure (hairpin), allowing Drosha cleavage, its loop component could also be modified or exchanged with other small ncRNAs, such as aptamers.

Aptamers are short nucleic acid sequences that work by binding to their specific molecular targets, interfering with their function (22). As a proof of principle, we chose a previously characterized anti-p50 [a component of nuclear factor κB (NF-κB)] RNA aptamer of ~50 nt (23, 24), which we noticed to be structurally similar to a pre-miR loop (fig. S9A).

Accordingly, we designed a chimeric microRNA based on the miR-7 hairpin, where its original loop was replaced by the p50 aptamer sequence (fig. S9, B and C). The resulting microRNA-driven aptamer anti-p50 (Mi-dAp50) was then engineered into the design of our original CL3 transgene, following our previously described protocol (25), creating CL3_Ap50 (Fig. 5A). The addition of Mi-dAp50 did not inhibit processing of the cluster’s microRNAs and did not produce a functional miR-7 (Fig. 5B). Instead, Northern blot using a probe for Ap50 showed a band of ~60 nt, which confirmed successful cleavage and release of Mi-dAp50 from the primary sequence (Fig. 5, C and D). PCR experiments in a Drosha KO model further validated that the existence of Mi-dAp50 depends on an intact microRNA processing machinery (fig. S9D). Evidence that Mi-dAp50 is functionally intact was obtained with RNA immunoprecipitation (RIP), which confirmed its binding to p50 (Fig. 5E). The retained property of Mi-dAp50 as a p50 inhibitor was proven by multiple approaches, including measuring its interference with the expression of known NF-κB–driven genes (Fig. 5F) and with a luciferase reporting system driven by an NF-κB responsive element (NF-κB-RE) (Fig. 5G). While the inhibition was readily evident in a Drosha wild-type (WT) context, no changes were observed in Drosha KO cells (Fig. 5G and fig. S9E). Mechanistically, chromatin immunoprecipitation (ChIP)–PCR confirmed that, in the presence of Mi-dAp50, the binding of p50 to the promoter of a representative NF-κB responsive gene was impaired (Fig. 5H).

Fig. 5. — (A) Schematics and corresponding predicted RNA folding of engineered microRNA sequences. Each component is labeled and color-coded accordingly. The black arrow represents the site of Drosha cleaved along the miR-7 stem sequence. (B) microRNA expression in HEK293 cells infected with different constructs. Bars represent mean values + SD; n = 3 biological replicates. Statistical analysis was performed by two-way ANOVA compared with SCR with Dunnett’s multiple comparisons test. (C) Polyacrylamide gel (4 to 20%) Northern blot from whole cell RNA of HEK293 cells transfected with different constructs. (D) Rationale for the design of a new microRNA-based platform. Mi-dAp50, microRNA-driven aptamer anti-p50. (E) PCR analysis of Mi-dAp50 in samples after RIP by p50 versus immunoglobulin G bait. Bars represent mean values ± SD; n = 3 biological replicates. Statistical analyses were performed by Student’s t test, two tails. (F) RT-qPCR showing gene expression level of known NF-κB target genes. RNA was isolated from HEK293 cells expressing different constructs. Bars represent mean values + SD; n = 3 biological replicates. Each comparison was analyzed with Student’s t test, two tails. (G) Luminescence detection in HEK293 expressing NF-κB responsive luciferase reporter genes and transfected with different clusters and specific, in presence or absence of TNF-α (10 ng/ml). Parallel experiments were done in Drosha WT and Drosha KO models. Bars represent mean values ± SD; n = 3 biological replicates. Statistical analysis was performed by Student’s t test, two tails. RLU, relative luminescence unit. (H) ChIP-PCR of p50 precipitated DNA from HEK293 cells expressing different constructs. The drawing clarifies IKBα as a known NF-κB responsive gene, while RLP30 is used as a negative control. Bars represent mean values ± SD; n = 3 biological replicates. Statistical analysis was performed by one-way ANOVA. *P < 0.05; **P = < 0.01; ***P = < 0.001; ****P < 0.0001; ns, nonsignificant. TSS, transcription start site.

Expanding the applicability of RNA aptamers

NF-κB activation is known to play a role in cell response after virus infection, activating defense pathways that decrease virus propagation (26). We reasoned that we could use an oncolytic virus derived from herpes simplex virus 1 (oHSV-1) (27) as a suitable model to prove the value of our microRNA platform in synergy with other established therapeutic modalities. After confirming that NF-κB becomes activated in GBM cells upon infection with oHSV-1 (fig. S10A) and that activation of NF-κB by tumor necrosis factor–α (TNF-α) makes these cells more resistant to virus replication and propagation (fig. S10, B to D), we characterized the differential permissivity to infection in T98G and G34 GBM cells expressing SCR versus CL3 versus CL3_Ap50, with the latter showing an increase of about 10-fold, while the three overexpressed microRNAs, per se, did not display any relevant function in this context (Fig. 6, A and B). After infection, a significant increase in dead cells was also observed in the CL3_Ap50 group (Fig. 6C), which also showed increased transcription of the HSV-1 genome (Fig. 6D). The observed difference in cell viability was not due to the isolated effect of the Mi-dAp50, since, without virus infection, CL3 and CL3_Ap50 cells showed a comparable proliferation and growth rate (fig. S11, A to C). A plaque-forming assay using equal volumes of the supernatant of GBM cells previously infected with oHSV-1 demonstrated that Mi-dAp50 significantly augmented virus replication and propagation (Fig. 6E and fig. S11, D and E). These findings were replicated in vivo, where nude mice previously implanted with GBM cells expressing the different CL transgenes were intratumorally infected with oHSV-1. As expected, in all cases, oHSV-1 prolonged survival compared to noninfected animals, but mice harboring the CL3_Ap50 tumor showed an almost threefold survival gain after oHSV-1 infection, compared to CL3 (Fig. 6F). Without oHSV-1 infection, CL3_Ap50 showed no gain in survival compared to CL3, both resulting in 1.5-fold gain compared to SCR control (Fig. 6F and fig. S11, F and G). Tumors harvested 7 days after oHSV-1 injection demonstrated a remarkable increase in virus spread through the CL3_Ap50 tumor compared to the CL3 tumor, both visually (Fig. 6G) and by measurement of oHSV-1 transcriptional activity (Fig. 6H) and oHSV-1 genome copy number (Fig. 6I).

Fig. 6. — (A) Fluorescence microscopy of T98G GBM cells expressing different RNA cluster constructs and infected with oHSV-1 [multiplicity of infection (MOI) = 0.5], acquired 12 hours after infection. Scale bar, 100 μm. (B) Fluorescence-activated cell sorting analysis for detection of GFP (oHSV-1 marker) in G34 cells expressing different RNA cluster constructs. (C) Quantification of cell viability with alamarBlue, 24 hours after oHSV-1 infection of T98G cells (MOI = 0.5), expressing different RNA constructs. Bars represent mean ± SD; n = 4 biological replicates. Statistical analysis by one-way ANOVA. (D) RT-qPCR quantification of oHSV-1 gene expression at different time points in G34 cells infected at MOI = 0.5. Bars represent mean values ± SD; n = 3 biological replicates. Statistical analysis by one-way ANOVA. (E) Representative images of plaque assay in Vero 2.2 cells incubated with an equal volume of conditioned supernatant from previously oHSV-1–infected (MOI = 0.5) G34 cells expressing different constructs. Plaques are visualized by GFP. Scale bar, 500 μm. (F) Kaplan-Meier survival curve of female athymic nu/nu mice intracranially implanted with 10,000 G34 cells stably expressing the different microRNAs constructs; n = 6. Log rank test, corrected by Bonferroni analysis. Arrowed lines represent different survival gains by oHSV-1. Graphic created in BioRender. Mazzetti, D. (2025) https://BioRender.com/ferhtf3. (G) Fluorescence microscopy images of cryosectioned brain slices obtained 9 days after oHSV-1 infection. Scale bars, 1000 μm (full brain); 25 μm (inset). (H) RT-qPCR analysis of oHSV-1 gene expression from tumor tissues referred to (G). Bars represent mean values + SD; n = 3 biological replicates. Each comparison was analyzed with Student’s t test, two tails. (I) Copy number quantification oHSV-1 genome recovered from tumor tissue related to (G) and (H). Bars represent mean values + SD; n = 3 biological replicates. Each comparison was analyzed with Student’s t test, two tails. *P < 0.05; **P = < 0.01; ****P < 0.0001; ns, nonsignificant. FSC-A, forward scatter area.

DISCUSSION

Here, we provide evidence for a gene therapy platform that can be armed with multiple and different ncRNA species, allowing for a high degree of biological interference in eukaryotic cells. This translates into a highly synergistic antitumor effect in various GBM models.

The only requirement for its engineering is the preservation of Drosha cleaving sites, which allows the proper release of each active component. This offers a flexible, plug-and-play scaffold for disease-specific multitargeting.

First described in the early 2000s (12), microRNAs have a proven role in regulating cell biology, but the initial enthusiasm for their use as therapeutics has been dampened by lack of power: Clinical trials have been scarce and overall disappointing, particularly in the oncology field (28–30). With this work, we propose a multilayered strategy to expand the potential of RNA therapy. This is not the first evidence of microRNA repurposing, as they have been previously used to encode shRNAs (31), artificial microRNAs (32), or engineered clusters (9, 25). However, to the best of our knowledge, this is the first time that accessory sequences making up the scaffolding component of a microRNA gene are observed to persist following its processing. In this regard, it is important to note that the length of both amplified flanks corresponds almost completely to the sequence that we originally borrowed from the pri-miR-128 gene when we designed the CL3 sequence (9, 25). It is intriguing to speculate that these parts of the native pri-miR-128 gene retain a so far unrecognized function, possibly associated with that of their embedded microRNA. Aside from the purely mechanistic aspect, which will need further investigation to be generalized, from a translational angle, this allows us to revisit the microRNA genetic structure as a vessel for introducing a combination of diverse, polyfunctional, small ncRNAs into cells, resulting in a robust increase in its overall biological effect and applicability. There are several advantages offered by this approach. For example, in addition to the benefit of coexpressing microRNAs and TuDs at the same time, this RNA polymerase II–driven platform allows improvement in TuD performance, as previously demonstrated (33). It also offers the possibility to use promoter-specific strategies for tissue-restricted expression.

Similarly, the therapeutic role of aptamers has been limited to their use as targeting carriers for other active molecules such as oligonucleotides (34), or to inactivate receptors within the cell membrane (35). On the contrary, their dependency on RNA polymerase III for intracellular expression, required by their short sequence (36), has significantly limited their use as endogenously produced genes. The Mi-dA design bypasses this shortcoming by allowing the cleavage of a properly sized aptamer structure from a longer RNA polymerase II–driven sequence. We chose to use an aptamer against p50 to demonstrate that this strategy could prove useful to inhibit an otherwise difficult-to-drug protein (37), as is also true for many other transcription factors (38). Moreover, intracellular production allows selective target inhibition only in the subset of transduced cells. This compartmentalized inhibition is beneficial when it involves pathways with such a pleiotropic effect as NF-κB (39). For example, its systemic disruption could be counterproductive by broadly impairing immune response in the context of cancer (40), or following oncolytic virotherapy, whose antitumor effect strongly depends on the activation of the adaptive immune response against the virus (41).

We chose to use miR-7 as the Trojan horse for the p50 aptamer as we had previously verified its correct processing within our cluster design (25). Given their stereotypical structure, many other microRNAs could expectedly be used instead for this specific purpose.

There will probably be limits to the length and structure of aptamers that can be fitted within the Mi-dA design, but with the rapid expansion of artificial intelligence algorithms for aptamer selection (42), we anticipate that it will be possible to generate an abundance of properly fitting sequences for any desired targets.

Although the choice of the five microRNAs modulated with our CL3_a2 transgene was informed by tumor expression analysis, it is not necessarily the best combination, neither was it the scope of this study. It is possible that other microRNAs could have an even stronger phenotype of what we have observed with CL3_a2. In addition, it is expected that different cancers might require alternative, disease-specific microRNAs. However, because all known microRNAs share the same hairpin structure, it is reasonable to predict an almost generalizable interchangeability, as we have previously shown (25). A limitation of this study is that we do not know yet the upper limits of this technology (i.e., how many sponges or microRNAs or Mi-dAs can be fitted before their processing becomes inefficient). It is encouraging that we have already demonstrated successful expression of up to six different microRNA hairpins within the same scaffold (25), even though decreasing processing efficiency was noted, particularly as the number of hairpin reaches 10 (fig. S12).

We have demonstrated the biological potential of this platform envisioning it as a therapeutic in itself, although suitable delivery strategies and comprehensive safety studies will be needed before clinical translation. Preclinical studies using adeno-associated virus type 2 vectors are currently underway to inform an incoming phase 1 clinical trial in patients with GBM. Nevertheless, given its small size (800 to 1300 nt), we anticipate that it will also work as embedded into genes already used to engineer oncolytic viruses or chimeric antigen receptor T cells, to overcome specific barriers to their functions (26, 43). When used in these applications, it will be crucial to exclude unintended consequences on innate immunity and uncontrolled viral dissemination caused by this approach.

While this platform has been primarily validated in GBM, the relevance of microRNAs in virtually any other cancers (44, 45), as well as in non-oncologic diseases, like cardiovascular (46) and neurodegenerative pathologies (47), is well recognized. Thus, the strategy described here is potentially applicable, albeit likely with different microRNA combinations, to any conditions characterized by multiple pathway abnormalities necessitating multitargeting.

Last, it is intriguing to consider that the existence of junk RNA remnants persisting after transgenic microRNA processing could also be true for some, if not many, cellular microRNAs. This could have important implications for a deeper understanding of their overall functioning, deserving further investigation.

In conclusion, the rethinking of microRNA genes in light of observations related to their structural constraints and processing, allows the design of a broader, modular, and more potent platform for RNA medicine, which is easy to engineer and offers a versatile tool for precise multitargeting.

MATERIALS AND METHODS

RNA sequencing

RNA from G34 GSCs infected with CL3 and control plasmid was isolated in TRIzol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. The same approach was adopted to isolate RNA from HEK293 cells and their exosomes. The RNA was processed by RealSeq Biosciences (Santa Cruz, CA, USA), and it was assessed for quantity and quality using Qubit Broad Range (catalog no. Q10210, Thermo Fisher Scientific, Waltham, MA, USA) and RNA TapeStation (catalog no. 5067-5576, Agilent, Lexington, MA, USA), following the manufacturer’s instructions. For DNA library preparation, RealSeq-Biofluids libraries were created with 100 ng of RNA input and 16 PCR cycles. Libraries were pooled to equal nanomolarity, purified, and size-selected using Pippin Prep (Sage Science, Cambridge, MA, USA) and then profiled with DNA TapeStation and double-stranded DNA (dsDNA) High Sensitivity Qubit before sequencing on the Singular G4 with single 100–base pair (bp) reads. The sequence of CL3 was provided to annotate the possible presence of the flanking regions of interest. For the RNA sequencing analysis of microRNA expression in GBM versus normal brain, curated data were obtained from an external dataset (48). References sets are GSE15825, GSE109628, GSE90603, GSE65626, GSE25631, The Cancer Genome Atlas (TCGA), and GSE165937.

Prediction of secondary RNA structures

For prediction of RNA secondary structure, the open-source program RNAFOLD (http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi) was used.

Exosome isolation

Exosomes were isolated from 30 ml of conditioned medium [Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) exosomes depleted; catalog no. A2720803, Gibco/Life Technologies, MA, USA] following procedures previously described (9). Briefly, the media were cleared from cells and cell debris by centrifugation at 3000g for 10 min, followed by 10,000g for 20 min and then filtered through a 0.22-μm polyvinylidene difluoride filter. Exosomes were finally isolated by ultracentrifugation at 100,000g for 70 min and either used for RNA or protein extraction.

Nanosight analysis

The exosomes isolated as above were processed and diluted 1:10 with double-filtered phosphate-buffered saline (PBS), analyzed using NanoSight LM10 (Malvern Instruments, Salisbury, UK), and quantified by NTA 3.0 program at the Nanosight Nanoparticle Sizing and Quantification Facility at the Massachusetts General Hospital, Boston, USA.

MicroRNA overexpression

For each transgene, the DNA sequence encoding each cluster was designed in silico by the authors and obtained as a bacterial plasmid from GeneArt (Life Technologies, Carlsbad, CA, USA). All DNA sequences were cloned into the lentiviral vector pCDH-CMV-MCS-EF1α-copGFP vector (System Biosciences, Palo Alto, CA, USA). For negative controls, both the empty vector (i.e., copGFP-only) and SCR sequences were used as previously described (9). Complete transgene sequences are provided in table S1. After infection, cells were sorted by GFP expression and used for downstream experiments.

Virus particle production

The lentiviral vector pCDH-CMV-MCS-EF1α-copGFP (catalog no. CD511B-1, System Biosciences) was used to produce lentiviruses in HEK293 cells using the ViraPower packaging system (Thermo Fisher Scientific). Lipofectamine 2000 (catalog no. 11668027, Life Technologies) was used to transfect DNA plasmids into HEK293 cells. The supernatant was collected 60 hours after transfection, and viral particles were concentrated by ultracentrifugation at 100,000g for 90 min at 4°C and resuspended in PBS. The same protocol was used for the lentiviral production of all constructs. For experiments requiring a red fluorescent protein (RFP) reporter gene, the cluster cassettes were subcloned into a pCDH-MSCV-MCS-EF1α-RFP+Puro lentivector (catalog no. CD713B-1, System Biosciences), and virions were produced as described above.

Northern blot

Northern blot assays were performed using DIG Luminescent Detection Kit for Nucleic Acids (catalog no. 11363514910, Roche, Basel, Switzerland), according to the manufacturer’s instructions and following the protocol described by Kim et al. (49).

All experiments involving the characterization of the primary sequence were run on a 1% agarose gel, 10% 10× MOPS [3-(N-morpholino)propanesulfonic acid], and 6.25% of 37% formaldehyde. Otherwise, Novex TBE (Tris Borate EDTA) Gels, 4 to 20% polyacrylamide gel (catalog no. EC62252BOX, Invitrogen), or 5% Criterion TBE-Urea Polyacrylamide Gel (catalog no. 3450086, Bio-Rad, Hercules, CA, USA) were used to characterize shorter RNA sequences. The probe sequences were designed using QIAGEN GeneGlobe software as custom miRCURY LNA miRNA and long noncoding RNA detection probes, and these are listed in table S2.

Cell cultures

G34 and G81 GSCs were previously obtained from human GBM operative specimens in the HCNL (Harvey Cushing Neuro-oncology Laboratories) laboratory. For GSCs, neurobasal medium (Gibco/Life Technologies, MA, USA) supplemented with 2 μM l-glutamine (Gibco), epidermal growth factor (20 ng/ml; PeproTech, Rocky Hill, NJ, USA), basic fibroblast growth factor 2 (20 ng/ml; PeproTech), B27 supplement (Life Technologies), and 1% penicillin/streptomycin (P/S; Gibco) were used. The human GBM cell lines U87-MG and T98G (purchased from American Type Culture Collection), mouse GBM cell line GL261 (gift from the HCNL laboratory), and HEK (human embryonic kidney) 293 cells were grown in DMEM (Gibco) supplemented with 10% FBS (Gibco) and 1% P/S (Gibco). Astrocytes were purchased from Lonza (catalog no. CC-2565, Basilea, Switzerland) and cultured according to the manufacturer’s protocol. HEK293 Drosha KO cells were purchased from Abcam (catalog no. AB266217, Cambridge, UK).

Gene expression studies

Total RNA was extracted using TRIzol (Invitrogen) according to the manufacturer’s protocol and as previously described (9). For microRNA analysis, specific TaqMan microRNA probes were purchased from Life Technologies and used to detect miR-124, miR-128, miR-137, miR-7, miR-21, miR-210, and miR-7 using a TaqMan Universal PCR Master Mix (Applied Biosystems, Carlsbad, CA, USA) and analyzed using Applied Biosystems Step One Plus 7500 RT-PCR apparatus. U6 small nuclear RNA (TaqMan probe) was used as a control. For mRNA analysis, cDNA for RT-PCR was synthesized using iScript cDNA Synthesis Kit (Bio-Rad). Analysis of mRNA expression was carried out using Power SYBR Green PCR Master Mix (Applied Biosystems) with primers listed in table S3. For specified experiments, total RNA was fractionated through column purification (Monarch RNA purification columns, New England Biolabs, Ipswich, MA, USA) following the manufacturer’s protocol.

For TuD detection, a Custom TaqMan Small RNA Assay probe was purchased from Life Technologies and used as described above. To determine the length of the primary sequence, progressively divergent primers were used to amplify cDNA using a Phusion High-Fidelity PCR Kit (New England Biolabs, Ipswich, MA, USA). All primer sequences used for PCR are listed in table S3. The PCR products were run in a 1.2% agarose gel at 100 V for 1 hour using 1 kb Plus DNA Ladder (catalog no. 10787018, Invitrogen) and visualized by SYBR safe staining.

Luciferase reporter assay

Twenty-nucleotide sequences complementary to miR-21 and miR-210 were cloned into the multiple cloning site (MCS) of a psy-CHECK2 vector (Promega, Madison, WI, USA), in the 3′UTR region of the Renilla luciferase gene, creating PSY-21 and PSY-210, respectively. Each plasmid was then cotransfected with different cluster constructs or specific miR inhibitors (mirVana miRNA inhibitors, Thermo Fisher Scientific) into HEK293 cells using Lipofectamine 2000 (Thermo Fisher Scientific). After incubation for 48 hours, the signals of Renilla and Firefly luciferase were quantified using the Dual-Luciferase Reporter Assay System Kit (catalog no. E1910, Promega).

For NF-κB luciferase experiments, the pGL4.32[luc2P/NF-κB-RE/Hygro] vector (Promega), which contains five copies of an NF-κB-RE that drives expression of the luciferase reporter gene luc2P, was used. This was subcloned into a lentivector plasmid and used to generate stable HEK293 and T98G GBM cell lines, which were then used for downstream experiments as detailed in each figure legend. Luciferase activity was detected using the Bright-Glo Luciferase Assay System (catalog no. E2610, Promega). Induction of NF-κB was obtained by the addition of TNF-α (10 ng/ml) to the culture medium.

Western blot

Cell pellets were lysed into radioimmunoprecipitation assay buffer containing protease and phosphatase inhibitors (catalog no. 78425, Thermo Fisher Scientific). Lysis was augmented with sonication, and the supernatant was cleared by centrifugation at 15,000 rpm for 10 min at 4°C. Protein concentration was measured using Bradford reagent assay. Twenty-five micrograms of whole cell lysate were loaded for blotting on 4 to 20% Criterion TGX Precast Midi Protein Gel (catalog no. 5671093, Bio-Rad). Primary antibodies are listed in table S4.

Clonogenic studies

G34 and G81 cells were dissociated and resuspended as single cells in stem cell medium containing 0.4% low melting temperature agarose (IBI Scientific, Peosta, IA, USA), as previously described (9). The cells were seeded at a density of 1000 per well in a 12-well plate in a total volume of 2 ml. After 10 min at room temperature, to allow medium jellification, cells were grown under standard conditions (37°C, 20% O₂, and 5% CO₂) for 2 weeks to allow time for sphere formation. For experiments under hypoxic conditions, cells were initially incubated at 1% O₂ for 48 hours and then put back to standard conditions for 2 weeks. Images were obtained with a Nikon ECLIPSE Ti motorized fluorescence microscope system, Japan (NIS-Elements 4.2). The number and size of spheres were analyzed by ImageJ software.

Wound healing assay

G34 cells expressing specific clusters were grown as a monolayer in six-well plates in DMEM with 10% FBS and 1% P/S. After 24 hours, a straight scratch was made using a 200-μl pipette tip. Cell migration was monitored every hour for 48 hours with a time-lap Nikon ECLIPSE Ti motorized fluorescence microscope system, Japan (NIS-Elements 4.2), and quantified using ImageJ software.

Transwell assay

G34 cells expressing different constructs were seeded in HTS 96-well multiwell permeable support system (catalog no. 734-0950, Corning FluoroBlok) with an 8-μm pore size floor, in DMEM with 10% FBS and 1% P/S. Cell migration was monitored over time for 48 hours with a time-lap Nikon ECLIPSE Ti motorized fluorescence microscope system, Japan (NIS-Elements 4.2), and quantified using ImageJ software.

Flow cytometry analysis

Cells under specified growing conditions were harvested, washed twice with BioLegend Cell Staining Buffer (catalog no. 420201, BioLegend, San Diego, CA, USA), and then stained for markers of cell death (Annexin A5 Apoptosis Detection Kit, catalog no. 640930, BioLegend) following the manufacturer’s protocol. G34 GSCs stably expressing each specific construct were infected with oHSV-1 and, after 24 hours, were harvested and washed twice with cold PBS with 2 mM EDTA. After fixation in 2% paraformaldehyde (PFA) for 15 min, cells were immediately analyzed for GFP expression with a BD LSRFortessa flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA). Data analysis was performed using FlowJo software (Tree Star, Inc. Ashland, OR, USA).

Animal studies

All animal experiments were performed in female, 6- to 8-week-old athymic mice (FoxN1 nu/nu, Envigo, South Easton, MA, USA), in compliance with all relevant ethical regulations applied to the use of small rodents and with approval by the Animal Care and Use Committees at the Brigham and Women’s Hospital and Harvard Medical School. As previously described (9), for intracranial tumor implantation, a stereotactic frame was used to inoculate each animal in the right striatum with 10,000 G34 or G81 GSCs (resuspended in 3 μl of PBS), stably expressing different microRNA transgenes specifically detailed in each figure panel. For experiments including intracranial injection of oHSV-1, 1 × 10⁶ plaque-forming units of virus, in a volume of 2 μl of Hanks’ balanced salt solution buffer, was administrated at the same stereotactic coordinates of previous tumor implantation at days specified in the corresponding figures.

For toxicity experiments, lentivectors expressing CL3_a2 transgene were intracranially injected into the right basal ganglia using a stereotactic frame. A total of 1 × 10⁵ infectious particles in 5 μl was injected per each mouse.

Mice were euthanized and perfused when they reached their predetermined endpoints, depending on the different experiments, and tissues were recovered for biochemical or histochemical analysis. For DNA and RNA extraction, animals were anesthetized and perfused transcardially with PBS (pH 7.4), followed by tissue dissection under microscope to isolate tumor tissue, or contralateral normal brain. For immunofluorescence staining, brains were fixed in 4% PFA and cryosectioned at 20-μm thickness. Sections were imaged using Nikon ECLIPSE Ti motorized fluorescence microscope system, Japan (NIS-Elements 4.2).

Brain slices were incubated for 3 hours with blocking buffer and overnight with antibody against CC3 at 1:400 dilution and GFP at 1:200 dilution (table S4), followed by Hoechst at 1:2000 dilution (catalog no. H3570, Thermo Fisher Scientific), and secondary anti-rabbit conjugated with Alexa-594 (catalog no. 711-586-152, Jackson ImmunoResearch Inc., West Grove, PA, USA) and anti-goat conjugated with Alexa-488 (catalog no. 705-545-003), both at 1:500 dilution. TMZ was administered in vivo by intraperitoneal injection at a concentration of 20 mg/kg for 5 consecutive days starting from the seventh day after intracranial tumor implantation, as previously described (9).

Chromatin immunoprecipitation

ChIP assay was performed using the SimpleChIP Plus Enzymatic Chromatin IP Kit (catalog no. 9005, Cell Signaling) according to the manufacturer’s protocol. Equal volumes of chromatin were immunoprecipitated with either antibody against p50 (catalog no. 13586, Cell Signaling) or rabbit immunoglobulin G (catalog no. P120-101, Bethyl Laboratories, Montgomery, TX, USA) as negative control. Primers for IKBα binding site (catalog no. 5552) and RLP30 (catalog no. 7014) were purchased from Cell Signaling.

RNA immunoprecipitation

Cells were harvested by centrifugation, rinsed in PBS, and lysed in RIP lysis buffer. Lysates were immunoprecipitated using an anti-p50 primary antibody (catalog no. 13586, Cell Signaling). RIP was performed using Magna RIP RNA-Binding Protein Immunoprecipitation Kit (catalog no. 17-700, Millipore, Billerica, MA, USA) following the manufacturer’s protocol. Primers used for the Mi-dAp50 detection are listed in table S3.

Advanced TaqMan PCR

Detection of the processed Mi-dAp50 was carried out using the TaqMan Advanced miRNA cDNA Synthesis Kit (catalog no. A28007, Applied Biosystems). This system uses 3′ poly(A) tailing and 5′ ligation of an adaptor sequence to extend the mature small RNA at both ends before reverse transcription. For amplification of Mi-dAp50, primers (listed in table S3) were designed to hybridize with the poly(A) tail and the known 5′ end of the target sequence. Last, RT-qPCR was performed (primers are listed in table S3).

oHSV-1 infection

Virus particles were produced and purified as previously described (50). All samples were handled in a biosafety level 2 (BL2) environment.

PCR verification of oHSV-1 infection

DNA and RNA were extracted from cells and mouse brain tumors using the Quick-DNA/RNA Miniprep Plus (catalog no. D7003, Zymo, Irvine, CA, USA) following the manufacturer’s protocol. A total of 10 ng of DNA and RNA was added per PCR reaction and amplified using the Luna Universal Probe One-Step RT-qPCR Kit (catalog no. E3006L, New England Biolabs) following the manufacturer’s protocol. Primers used for the detection are listed in table S3.

Metabolic assay

Quantitative cell death was evaluated using alamarBlue assay (catalog no. DAL1025, Invitrogen), following the manufacturer’s instructions, and fluorescence was measured with a Polar Star Omega cell plate reader (BMG LabTech).

Plaque assay

Green monkey Vero 2.2 cells grown as monolayers in a 96-well plate in DMEM supplemented with 10% FBS were incubated with 10 μl of conditioned medium obtained from G34 cells infected with oHSV-1 24 hours prior. Analysis of the resulting plaques was performed by visualizing the number and size of GFP foci at different time points and quantified with ImageJ.

Statistical and bioinformatics analysis

Statistical analyses were conducted using the appropriate statistical tests based on the experimental design and described in each display item using GraphPad Prism software. Unless otherwise specified, all results are based on three biological replicates. Quantification of microscopy-based assays was done using ImageJ. The statistical significance of each result is shown in each corresponding figure panel.

The small ncRNA sequencing data were processed through a bioinformatics pipeline involving filtering, trimming, quality control, alignment, and differential expression analysis. Raw FASTQ files were first processed with Cutadapt to remove adapter sequences and filter out reads shorter than 15 bp. Trimmed reads were sequentially aligned to multiple databases using Bowtie. Multimapped reads were assigned to the best alignment. DESeq2 was used for differential expression analysis, requiring a raw count matrix and metadata for pairwise comparisons.

Illustrations

Some of the illustrations have been created using BioRender (www.biorender.com) through an academic subscription to Harvard Medical School.

Acknowledgments

We thank all members of the HCNL laboratories and F. Slack for helpful discussions.

Funding:

This manuscript was supported by NINDS grants K08NS101091, NINDS 1R01NS116144, and the Distinguished Scientist Award from the Sontag Foundation to P.P.

Author contributions:

Conceptualization: P.P., D.M., E.A.C., and L.A. Methodology: P.P., D.M., M.O.N., M.M., G.F., D.A.H., J.D.B., M.G., L.E., and H.N. Investigation: D.M., P.P., M.O.N., M.M., H.S., D.A.H., and J.D.B. Visualization: D.M., P.P., L.E., and M.G. Supervision: P.P., E.A.C., and L.A. Data curation: D.M. Validation: D.M., P.P., and J.D.B. Formal analysis: P.P. and D.M. Software: D.M. Project administration: P.P., D.M., and M.O.N. Funding acquisition: P.P. Resources: P.P., M.O.N., E.A.C., J.D.B., H.N., and A.K. Writing—original draft: P.P. and D.M. Writing—review and editing: P.P., D.M., M.M., H.S., E.A.C., J.D.B., A.K., H.N., and G.F.

Competing interests:

The authors declare the following competing interests: P.P. holds equities in Ternalys Therapeutics. E.A.C. is an advisor to Amacathera, Bionaut Labs, Genenta, Inc., Insightec, Inc., DNAtrix, Inc., Seneca Therapeutics, and Synthetic Biologics. He has equity options in Bionaut 5 Laboratories, DNAtrix, Immunomic Therapeutics, Seneca Therapeutics, Synthetic Biologics, and Ternalys Therapeutics. He is a co-founder and on the Board of Directors of Ternalys Therapeutics. He has received research support from NIH, US Department of Defense, American Brain Tumor Association, National Brain Tumor Society, Alliance for Cancer Gene Therapy, Neurosurgical Research Education Foundation, Advantagene, NewLink Genetics, and Amgen. He is also a named inventor on patents related to oncolytic HSV-1 and ncRNAs. J.D.B. has an equity position in Treovir, Inc., an oHSV clinical stage company, and UpFront Diagnostics. J.D.B. is also on the Centile Bioscience, QV Bioelectronics, and NeuroX1 boards of scientific advisors. The other authors declare that they have no competing interests. Part of the methodology described is covered under US Patent 12,385,067 issued to P.P. and E.A.C., and US patent application 19/270,955 to P.P. and D.M.

Data and materials availability:

Any constructs or reagent are available upon request to the corresponding author (pperuzzi@bwh.harvard.edu). All data and code needed to evaluate and reproduce the results in the paper are present in the paper and/or the Supplementary Materials.

Supplementary Materials

The PDF file includes:

Figs. S1 to S12

Tables S1 to S4

Legends for data files S1 to S3

sciadv.adz8331_sm.pdf^{(4.1MB, pdf)}

Other Supplementary Material for this manuscript includes the following:

Data files S1 to S3

sciadv.adz8331_data_files_s1_to_s3.zip^{(2.5MB, zip)}

REFERENCES

1.Sanchez-Vega F., Mina M., Armenia J., Chatila W. K., Luna A., La K. C., Dimitriadoy S., Liu D. L., Kantheti H. S., Saghafinia S., Chakravarty D., Daian F., Gao Q., Bailey M. H., Liang W. W., Foltz S. M., Shmulevich I., Ding L., Heins Z., Ochoa A., Gross B., Gao J., Zhang H., Kundra R., Kandoth C., Bahceci I., Dervishi L., Dogrusoz U., Zhou W., Shen H., Laird P. W., Way G. P., Greene C. S., Liang H., Xiao Y., Wang C., Iavarone A., Berger A. H., Bivona T. G., Lazar A. J., Hammer G. D., Giordano T., Kwong L. N., McArthur G., Huang C., Tward A. D., Frederick M. J., McCormick F., Meyerson M., Van Allen E. M., Cherniack A. D., Ciriello G., Sander C., Schultz N., Oncogenic signaling pathways in The Cancer Genome Atlas. Cell 173, 321–337 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Flavahan W. A., Gaskell E., Bernstein B. E., Epigenetic plasticity and the hallmarks of cancer. Science 357, eaal2380 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Brennan C. W., Verhaak R. G., McKenna A., Campos B., Noushmehr H., Salama S. R., Zheng S., Chakravarty D., Sanborn J. Z., Berman S. H., Beroukhim R., Bernard B., Wu C. J., Genovese G., Shmulevich I., Barnholtz-Sloan J., Zou L., Vegesna R., Shukla S. A., Ciriello G., Yung W. K., Zhang W., Sougnez C., Mikkelsen T., Aldape K., Bigner D. D., Van Meir E. G., Prados M., Sloan A., Black K. L., Eschbacher J., Finocchiaro G., Friedman W., Andrews D. W., Guha A., Iacocca M., O’Neill B. P., Foltz G., Myers J., Weisenberger D. J., Penny R., Kucherlapati R., Perou C. M., Hayes D. N., Gibbs R., Marra M., Mills G. B., Lander E., Spellman P., Wilson R., Sander C., Weinstein J., Meyerson M., Gabriel S., Laird P. W., Haussler D., Getz G., Chin L., The somatic genomic landscape of glioblastoma. Cell 155, 462–477 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Rahman R., Trippa L., Lee E. Q., Arrillaga-Romany I., Fell G., Touat M., McCluskey C., Wiley J., Gaffey S., Drappatz J., Welch M. R., Galanis E., Ahluwalia M. S., Colman H., Nabors L. B., Hepel J., Elinzano H., Schiff D., Chukwueke U. N., Beroukhim R., Nayak L., McFaline-Figueroa J. R., Batchelor T. T., Rinne M. L., Kaley T. J., Lu-Emerson C., Mellinghoff I. K., Bi W. L., Arnaout O., Peruzzi P. P., Haas-Kogan D., Tanguturi S., Cagney D., Aizer A., Doherty L., Lavallee M., Fisher-Longden B., Dowling S., Geduldig J., Watkinson F., Pisano W., Malinowski S., Ramkissoon S., Santagata S., Meredith D. M., Chiocca E. A., Reardon D. A., Alexander B. M., Ligon K. L., Wen P. Y., Inaugural results of the individualized screening trial of innovative glioblastoma therapy: A phase II platform trial for newly diagnosed glioblastoma using bayesian adaptive randomization. J. Clin. Oncol. 41, 5524–5535 (2023). [DOI] [PubMed] [Google Scholar]
5.Anastasiadou E., Jacob L., Slack F. J., Non-coding RNA networks in cancer. Nat. Rev. Cancer 18, 5–18 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Zhu Y., Zhu L., Wang X., Jin H., RNA-based therapeutics: An overview and prospectus. Cell Death Dis. 13, 644 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Zogg H., Singh R., Ro S., Current advances in RNA therapeutics for human diseases. Int. J. Mol. Sci. 23, 2736 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Crooke S. T., Witztum J. L., Bennett C. F., Baker B. F., RNA-targeted therapeutics. Cell Metab. 27, 714–739 (2018). [DOI] [PubMed] [Google Scholar]
9.Bhaskaran V., Nowicki M. O., Idriss M., Jimenez M. A., Lugli G., Hayes J. L., Mahmoud A. B., Zane R. E., Passaro C., Ligon K. L., Haas-Kogan D., Bronisz A., Godlewski J., Lawler S. E., Chiocca E. A., Peruzzi P., The functional synergism of microRNA clustering provides therapeutically relevant epigenetic interference in glioblastoma. Nat. Commun. 10, 442 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Zeng Y., Cullen B. R., Efficient processing of primary microRNA hairpins by Drosha requires flanking nonstructured RNA sequences. J. Biol. Chem. 280, 27595–27603 (2005). [DOI] [PubMed] [Google Scholar]
11.Han J., Lee Y., Yeom K., Kim Y., Jin H., Kim S., Kim J., Han V. J., Kim D. S., Kim V. N., Molecular basis for the recognition of primary microRNAs by the Drosha-DGCR8 complex. Cell 125, 887–901 (2006). [DOI] [PubMed] [Google Scholar]
12.Bartel D. P., MicroRNAs: Genomics, biogenesis, mechanism, and function. Cell 116, 281–297 (2004). [DOI] [PubMed] [Google Scholar]
13.O’Brien K., Breyne M., Ughetto S., Laurent D., Breakefield C. M., RNA delivery by extracellular vesicles in mammalian cells and its applications. Nat. Rev. Mol. Cell Biol. 21, 585–606 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Zangari J., Ilie M., Rouaud F., Signetti L., Ohanna M., Didier R., Roméo B., Goldoni D., Nottet N., Staedel C., Gal J., Mari B., Mograbi B., Hofman P., Brest P., Rapid decay of engulfed extracellular miRNA by XRN1 exonuclease promotes transient epithelial-mesenchymal transition. Nucleic Acids Res. 45, 4131–4141 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Zhang K., Hodge J., Chatterjee A., Moon T. S., Parker K. M., Duplex structure of double-stranded RNA provides stability against hydrolysis relative to single-stranded RNA. Environ. Sci. Technol. 55, 8045–8053 (2021). [DOI] [PubMed] [Google Scholar]
16.Haraguchi T., Ozaki Y., Iba H., Vectors expressing efficient RNA decoys achieve the long-term suppression of specific microRNA activity in mammalian cells. Nucleic Acids Res. 37, e43 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Hollensen A. K., Bak M., Hasler M., Nielsen M. M., Møller M. M., Suppression of microRNAs by dual-targeting and clustered Tough Decoy inhibitors. RNA Biol. 10, 406–414 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Krichevsky A. M., Gabriely G., miR-21: A small multi-faceted RNA. J. Cell. Mol. Med. 13, 39–53 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Wong S. T., Zhang X. Q., Zhuang J. T., Chan H. L., Li C. H., Leung G. K., MicroRNA-21 inhibition enhances in vitro chemosensitivity of temozolomide-resistant glioblastoma cells. Anticancer Res 32, 2835–2841 (2012). [PubMed] [Google Scholar]
20.Wang P., Yan Q., Liao B., Zhao L., Xiong S., Wang J., Zou D., Pan J., Wu L., Deng Y., Wu N., Gong S., The HIF1α/HIF2α-miR210-3p network regulates glioblastoma cell proliferation, dedifferentiation and chemoresistance through EGF under hypoxic conditions. Cell Death Dis. 11, 992 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Huang X., Ding L., Bennewith K. L., Tong R. T., Welford S. M., Ang K. K., Story M., Le Q. T., Giaccia A. J., Hypoxia-inducible miR-210 regulates normoxic gene expression involved in tumor initiation. Mol. Cell 35, 856–867 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Keefe A. D., Pai S., Ellington A., Aptamers as therapeutics. Nat. Rev. Drug Discov. 9, 537–550 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Wurster S. E., Bida J. P., Her Y. F., Maher L. J. III, Characterization of anti-NF-κB RNA aptamer-binding specificity in vitro and in the yeast three-hybrid system. Nucleic Acids Res. 37, 6214–6224 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Reiter N. J., Maher L. J. III, Butcher S. E., DNA mimicry by a high-affinity anti-NF-κB RNA aptamer. Nucleic Acids Res. 36, 1227–1236 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Bhaskaran V., Yao Y., Bei F., Peruzzi P., Engineering, delivery, and biological validation of artificial microRNA clusters for gene therapy applications. Nat. Protoc. 14, 3538–3553 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Selman M., Ou P., Rousso C., Bergeron A., Krishnan R., Pikor L., Chen A., Keller B. A., Ilkow C., Bell J. C., Diallo J. S., Dimethyl fumarate potentiates oncolytic virotherapy through NF-κB inhibition. Sci. Transl. Med. 10, 1613 (2018). [DOI] [PubMed] [Google Scholar]
27.Kambara H., Okano H., Chiocca E. A., Saeki Y., An oncolytic HSV-1 mutant expressing ICP34.5 under control of a nestin promoter increases survival of animals even when symptomatic from a brain tumor. Cancer Res. 65, 2832–2839 (2005). [DOI] [PubMed] [Google Scholar]
28.Beg M. S., Brenner A. J., Sachdev J., Borad M., Kang Y. K., Stoudemire J., Smith S., Bader A. G., Kim S., Hong D. S., Phase I study of MRX34, a liposomal miR-34a mimic, administered twice weekly in patients with advanced solid tumors. Invest. New Drugs 35, 180–188 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.van Zandwijk N., Pavlakis N., Kao S. C., Linton A., Boyer M. J., Clarke S., Huynh Y., Chrzanowska A., Fulham M. J., Bailey D. L., Cooper W. A., Kritharides L., Ridley L., Pattison S. T., MacDiarmid J., Brahmbhatt H., Reid G., Safety and activity of microRNA-loaded minicells in patients with recurrent malignant pleural mesothelioma: A first-in-man, phase 1, open-label, dose-escalation study. Lancet Oncol. 18, 1386–1396 (2017). [DOI] [PubMed] [Google Scholar]
30.Kim T., Croce C. M., MicroRNA: Trends in clinical trials of cancer diagnosis and therapy strategies. Exp. Mol. Med. 55, 1314–1321 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Chang K., Elledge S. J., Hannon G. J., Lessons from Nature: MicroRNA-based shRNA libraries. Nat. Methods 3, 707–714 (2006). [DOI] [PubMed] [Google Scholar]
32.Arroyo J. D., Gallichotte E. N., Tewari M., Systematic design and functional analysis of artificial microRNAs. Nucleic Acids Res. 42, 6064–6077 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Bak R. O., Hollensen A. K., Primo M. N., Sorensen C. D., Mikkelsen J. G., Potent microRNA suppression by RNA PolII-transcribed “Tough Decoy” inhibitors. RNA 19, 28–293 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Millozzi F., Milán-Rois P., Sett A., Delli Carpini G., De Bardi M., Gisbert-Garzarán M., Sandonà M., Rodríguez-Díaz C., Martínez-Mingo M., Pardo I., Esposito F., Viscomi M. T., Bouché M., Parolini O., Saccone V., Toulmé J. J., Somoza Á., Palacios D., Aptamer-conjugated gold nanoparticles enable oligonucleotide delivery into muscle stem cells to promote regeneration of dystrophic muscles. Nat. Commun. 16, 577 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Nimjee S. M., Dornbos D. III, Pitoc G. A., Wheeler D. G., Layzer J. M., Venetos N., Huttinger A., Talentino S. E., Musgrave N. J., Moody H., Rempel R. E., Jones C., Carlisle K., Wilson J., Bratton C., Joseph M. E., Khan S., Hoffman M. R., Sommerville L., Becker R. C., Zweier J. L., Sullenger B. A., Preclinical development of a vWF aptamer to limit thrombosis and engender arterial recanalization of occluded vessels. Mol. Ther. 27, 1228–1241 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Mi J., Zhang X., Rabbani Z. N., Liu Y., Su Z., Vujaskovic Z., Kontos C. D., Sullenger B. A., Clary B. M., H1 RNA polymerase III promoter-driven expression of an RNA aptamer leads to high-level inhibition of intracellular protein activity. Nucleic Acids Res. 34, 3577–3584 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Yu H., Lin L., Zhang Z., Zhang H., Hu H., Targeting NF-κB pathway for the therapy of diseases: Mechanism and clinical study. Signal Transduct. Target. Ther. 5, 209 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Bushweller J. H., Targeting transcription factors in cancer—From undruggable to reality. Nat. Rev. Cancer 19, 611–624 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Guo Q., Jin Y., Chen X., Ye X., Shen X., Lin M., Zeng C., Zhou T., Zhang J., NF-κB in biology and targeted therapy: New insights and translational implications. Signal Transduct. Target. Ther. 9, 53 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Taniguchi K., Karin M., NF-κB, inflammation, immunity and cancer: Coming of age. Nat. Rev. Immunol. 18, 309–324 (2018). [DOI] [PubMed] [Google Scholar]
41.Ling A. L., Solomon I. H., Landivar A. M., Nakashima H., Woods J. K., Santos A., Masud N., Fell G., Mo X., Yilmaz A. S., Grant J., Zhang A., Bernstock J. D., Torio E., Ito H., Liu J., Shono N., Nowicki M. O., Triggs D., Halloran P., Piranlioglu R., Soni H., Stopa B., Bi W. L., Peruzzi P., Chen E., Malinowski S. W., Prabhu M. C., Zeng Y., Carlisle A., Rodig S. J., Wen P. Y., Lee E. Q., Nayak L., Chukwueke U., Gonzalez Castro L. N., Dumont S. D., Batchelor T., Kittelberger K., Tikhonova E., Miheecheva N., Tabakov D., Shin N., Gorbacheva A., Shumskiy A., Frenkel F., Aguilar-Cordova E., Aguilar L. K., Krisky D., Wechuck J., Manzanera A., Matheny C., Tak P. P., Barone F., Kovarsky D., Tirosh I., Suvà M. L., Wucherpfennig K. W., Ligon K., Reardon D. A., Chiocca E. A., Clinical trial links oncolytic immunoactivation to survival in glioblastoma. Nature 623, 157–166 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Wong F., He D., Krishnan A., Hong L., Wang A. Z., Wang J., Hu Z., Omori S., Li A., Rao J., Yu Q., Jin W., Zhang T., Ilia I., Chen J. X., Zheng S., King I., Li Y., Collins J. J., Deep generative design of RNA aptamers using structural predictions. Nat. Comput. Sci. 4, 829–839 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Delgoffe G. M., Xu C., Mackall C. L., Green M. R., Gottschalk S., Speiser D. E., Zehn D., Beavis P. A., The role of exhaustion in CAR T cell therapy. Cancer Cell 39, 885–888 (2021). [DOI] [PubMed] [Google Scholar]
44.Lee Y. S., Dutta A., MicroRNAs in cancer. Annu. Rev. Pathol. 4, 199–227 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Hayes J., Peruzzi P., Lawler S. E., microRNAs in cancer: Biomarkers, functions and therapy. Trends Mol. Med. 20, 460–469 (2014). [DOI] [PubMed] [Google Scholar]
46.Laggerbauer B., Engelhardt S., MicroRNAs as therapeutic targets in cardiovascular disease. J.Clin.Invest. 132, e159179 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Dhuppar S., Poller W. C., Murugaiyan G., MicroRNAs in the biology and hallmarks of neurodegenerative diseases. Trends Mol. Med. 31, 955–969 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Grigore I. A., Rajagopal A., Chow J. T., Stone T. J., Salmena L., Discovery of miRNA–mRNA regulatory networks in glioblastoma reveals novel insights into tumor microenvironment remodeling. Sci. Rep. 14, 27493 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Kim S. W., Li Z., Moore P. S., Monaghan A. P., Chang Y., Nichols M., John B., A sensitive non-radioactive Northern blot method to detect small RNAs. Nucleic Acids Res. 38, e98 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Nakashima H., Kaufmann J. K., Wang P.-Y., Nguyen T., Speranza M.-C., Kasai K., Okemoto K., Otsuki A., Nakano I., Fernandez S., Goins W. F., Grandi P., Glorioso J. C., Lawler S., Cripe T. P., Chiocca E. A., Histone deacetylase 6 inhibition enhances oncolytic viral replication in glioma. J. Clin. Invest. 125, 4269–4280 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figs. S1 to S12

Tables S1 to S4

Legends for data files S1 to S3

sciadv.adz8331_sm.pdf^{(4.1MB, pdf)}

Data files S1 to S3

sciadv.adz8331_data_files_s1_to_s3.zip^{(2.5MB, zip)}

Data Availability Statement

[R1] 1.Sanchez-Vega F., Mina M., Armenia J., Chatila W. K., Luna A., La K. C., Dimitriadoy S., Liu D. L., Kantheti H. S., Saghafinia S., Chakravarty D., Daian F., Gao Q., Bailey M. H., Liang W. W., Foltz S. M., Shmulevich I., Ding L., Heins Z., Ochoa A., Gross B., Gao J., Zhang H., Kundra R., Kandoth C., Bahceci I., Dervishi L., Dogrusoz U., Zhou W., Shen H., Laird P. W., Way G. P., Greene C. S., Liang H., Xiao Y., Wang C., Iavarone A., Berger A. H., Bivona T. G., Lazar A. J., Hammer G. D., Giordano T., Kwong L. N., McArthur G., Huang C., Tward A. D., Frederick M. J., McCormick F., Meyerson M., Van Allen E. M., Cherniack A. D., Ciriello G., Sander C., Schultz N., Oncogenic signaling pathways in The Cancer Genome Atlas. Cell 173, 321–337 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R2] 2.Flavahan W. A., Gaskell E., Bernstein B. E., Epigenetic plasticity and the hallmarks of cancer. Science 357, eaal2380 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R3] 3.Brennan C. W., Verhaak R. G., McKenna A., Campos B., Noushmehr H., Salama S. R., Zheng S., Chakravarty D., Sanborn J. Z., Berman S. H., Beroukhim R., Bernard B., Wu C. J., Genovese G., Shmulevich I., Barnholtz-Sloan J., Zou L., Vegesna R., Shukla S. A., Ciriello G., Yung W. K., Zhang W., Sougnez C., Mikkelsen T., Aldape K., Bigner D. D., Van Meir E. G., Prados M., Sloan A., Black K. L., Eschbacher J., Finocchiaro G., Friedman W., Andrews D. W., Guha A., Iacocca M., O’Neill B. P., Foltz G., Myers J., Weisenberger D. J., Penny R., Kucherlapati R., Perou C. M., Hayes D. N., Gibbs R., Marra M., Mills G. B., Lander E., Spellman P., Wilson R., Sander C., Weinstein J., Meyerson M., Gabriel S., Laird P. W., Haussler D., Getz G., Chin L., The somatic genomic landscape of glioblastoma. Cell 155, 462–477 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Rahman R., Trippa L., Lee E. Q., Arrillaga-Romany I., Fell G., Touat M., McCluskey C., Wiley J., Gaffey S., Drappatz J., Welch M. R., Galanis E., Ahluwalia M. S., Colman H., Nabors L. B., Hepel J., Elinzano H., Schiff D., Chukwueke U. N., Beroukhim R., Nayak L., McFaline-Figueroa J. R., Batchelor T. T., Rinne M. L., Kaley T. J., Lu-Emerson C., Mellinghoff I. K., Bi W. L., Arnaout O., Peruzzi P. P., Haas-Kogan D., Tanguturi S., Cagney D., Aizer A., Doherty L., Lavallee M., Fisher-Longden B., Dowling S., Geduldig J., Watkinson F., Pisano W., Malinowski S., Ramkissoon S., Santagata S., Meredith D. M., Chiocca E. A., Reardon D. A., Alexander B. M., Ligon K. L., Wen P. Y., Inaugural results of the individualized screening trial of innovative glioblastoma therapy: A phase II platform trial for newly diagnosed glioblastoma using bayesian adaptive randomization. J. Clin. Oncol. 41, 5524–5535 (2023). [DOI] [PubMed] [Google Scholar]

[R5] 5.Anastasiadou E., Jacob L., Slack F. J., Non-coding RNA networks in cancer. Nat. Rev. Cancer 18, 5–18 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Zhu Y., Zhu L., Wang X., Jin H., RNA-based therapeutics: An overview and prospectus. Cell Death Dis. 13, 644 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] 7.Zogg H., Singh R., Ro S., Current advances in RNA therapeutics for human diseases. Int. J. Mol. Sci. 23, 2736 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] 8.Crooke S. T., Witztum J. L., Bennett C. F., Baker B. F., RNA-targeted therapeutics. Cell Metab. 27, 714–739 (2018). [DOI] [PubMed] [Google Scholar]

[R9] 9.Bhaskaran V., Nowicki M. O., Idriss M., Jimenez M. A., Lugli G., Hayes J. L., Mahmoud A. B., Zane R. E., Passaro C., Ligon K. L., Haas-Kogan D., Bronisz A., Godlewski J., Lawler S. E., Chiocca E. A., Peruzzi P., The functional synergism of microRNA clustering provides therapeutically relevant epigenetic interference in glioblastoma. Nat. Commun. 10, 442 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] 10.Zeng Y., Cullen B. R., Efficient processing of primary microRNA hairpins by Drosha requires flanking nonstructured RNA sequences. J. Biol. Chem. 280, 27595–27603 (2005). [DOI] [PubMed] [Google Scholar]

[R11] 11.Han J., Lee Y., Yeom K., Kim Y., Jin H., Kim S., Kim J., Han V. J., Kim D. S., Kim V. N., Molecular basis for the recognition of primary microRNAs by the Drosha-DGCR8 complex. Cell 125, 887–901 (2006). [DOI] [PubMed] [Google Scholar]

[R12] 12.Bartel D. P., MicroRNAs: Genomics, biogenesis, mechanism, and function. Cell 116, 281–297 (2004). [DOI] [PubMed] [Google Scholar]

[R13] 13.O’Brien K., Breyne M., Ughetto S., Laurent D., Breakefield C. M., RNA delivery by extracellular vesicles in mammalian cells and its applications. Nat. Rev. Mol. Cell Biol. 21, 585–606 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Zangari J., Ilie M., Rouaud F., Signetti L., Ohanna M., Didier R., Roméo B., Goldoni D., Nottet N., Staedel C., Gal J., Mari B., Mograbi B., Hofman P., Brest P., Rapid decay of engulfed extracellular miRNA by XRN1 exonuclease promotes transient epithelial-mesenchymal transition. Nucleic Acids Res. 45, 4131–4141 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Zhang K., Hodge J., Chatterjee A., Moon T. S., Parker K. M., Duplex structure of double-stranded RNA provides stability against hydrolysis relative to single-stranded RNA. Environ. Sci. Technol. 55, 8045–8053 (2021). [DOI] [PubMed] [Google Scholar]

[R16] 16.Haraguchi T., Ozaki Y., Iba H., Vectors expressing efficient RNA decoys achieve the long-term suppression of specific microRNA activity in mammalian cells. Nucleic Acids Res. 37, e43 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] 17.Hollensen A. K., Bak M., Hasler M., Nielsen M. M., Møller M. M., Suppression of microRNAs by dual-targeting and clustered Tough Decoy inhibitors. RNA Biol. 10, 406–414 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] 18.Krichevsky A. M., Gabriely G., miR-21: A small multi-faceted RNA. J. Cell. Mol. Med. 13, 39–53 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] 19.Wong S. T., Zhang X. Q., Zhuang J. T., Chan H. L., Li C. H., Leung G. K., MicroRNA-21 inhibition enhances in vitro chemosensitivity of temozolomide-resistant glioblastoma cells. Anticancer Res 32, 2835–2841 (2012). [PubMed] [Google Scholar]

[R20] 20.Wang P., Yan Q., Liao B., Zhao L., Xiong S., Wang J., Zou D., Pan J., Wu L., Deng Y., Wu N., Gong S., The HIF1α/HIF2α-miR210-3p network regulates glioblastoma cell proliferation, dedifferentiation and chemoresistance through EGF under hypoxic conditions. Cell Death Dis. 11, 992 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Huang X., Ding L., Bennewith K. L., Tong R. T., Welford S. M., Ang K. K., Story M., Le Q. T., Giaccia A. J., Hypoxia-inducible miR-210 regulates normoxic gene expression involved in tumor initiation. Mol. Cell 35, 856–867 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Keefe A. D., Pai S., Ellington A., Aptamers as therapeutics. Nat. Rev. Drug Discov. 9, 537–550 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R23] 23.Wurster S. E., Bida J. P., Her Y. F., Maher L. J. III, Characterization of anti-NF-κB RNA aptamer-binding specificity in vitro and in the yeast three-hybrid system. Nucleic Acids Res. 37, 6214–6224 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] 24.Reiter N. J., Maher L. J. III, Butcher S. E., DNA mimicry by a high-affinity anti-NF-κB RNA aptamer. Nucleic Acids Res. 36, 1227–1236 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] 25.Bhaskaran V., Yao Y., Bei F., Peruzzi P., Engineering, delivery, and biological validation of artificial microRNA clusters for gene therapy applications. Nat. Protoc. 14, 3538–3553 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R26] 26.Selman M., Ou P., Rousso C., Bergeron A., Krishnan R., Pikor L., Chen A., Keller B. A., Ilkow C., Bell J. C., Diallo J. S., Dimethyl fumarate potentiates oncolytic virotherapy through NF-κB inhibition. Sci. Transl. Med. 10, 1613 (2018). [DOI] [PubMed] [Google Scholar]

[R27] 27.Kambara H., Okano H., Chiocca E. A., Saeki Y., An oncolytic HSV-1 mutant expressing ICP34.5 under control of a nestin promoter increases survival of animals even when symptomatic from a brain tumor. Cancer Res. 65, 2832–2839 (2005). [DOI] [PubMed] [Google Scholar]

[R28] 28.Beg M. S., Brenner A. J., Sachdev J., Borad M., Kang Y. K., Stoudemire J., Smith S., Bader A. G., Kim S., Hong D. S., Phase I study of MRX34, a liposomal miR-34a mimic, administered twice weekly in patients with advanced solid tumors. Invest. New Drugs 35, 180–188 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R29] 29.van Zandwijk N., Pavlakis N., Kao S. C., Linton A., Boyer M. J., Clarke S., Huynh Y., Chrzanowska A., Fulham M. J., Bailey D. L., Cooper W. A., Kritharides L., Ridley L., Pattison S. T., MacDiarmid J., Brahmbhatt H., Reid G., Safety and activity of microRNA-loaded minicells in patients with recurrent malignant pleural mesothelioma: A first-in-man, phase 1, open-label, dose-escalation study. Lancet Oncol. 18, 1386–1396 (2017). [DOI] [PubMed] [Google Scholar]

[R30] 30.Kim T., Croce C. M., MicroRNA: Trends in clinical trials of cancer diagnosis and therapy strategies. Exp. Mol. Med. 55, 1314–1321 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] 31.Chang K., Elledge S. J., Hannon G. J., Lessons from Nature: MicroRNA-based shRNA libraries. Nat. Methods 3, 707–714 (2006). [DOI] [PubMed] [Google Scholar]

[R32] 32.Arroyo J. D., Gallichotte E. N., Tewari M., Systematic design and functional analysis of artificial microRNAs. Nucleic Acids Res. 42, 6064–6077 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] 33.Bak R. O., Hollensen A. K., Primo M. N., Sorensen C. D., Mikkelsen J. G., Potent microRNA suppression by RNA PolII-transcribed “Tough Decoy” inhibitors. RNA 19, 28–293 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Millozzi F., Milán-Rois P., Sett A., Delli Carpini G., De Bardi M., Gisbert-Garzarán M., Sandonà M., Rodríguez-Díaz C., Martínez-Mingo M., Pardo I., Esposito F., Viscomi M. T., Bouché M., Parolini O., Saccone V., Toulmé J. J., Somoza Á., Palacios D., Aptamer-conjugated gold nanoparticles enable oligonucleotide delivery into muscle stem cells to promote regeneration of dystrophic muscles. Nat. Commun. 16, 577 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R35] 35.Nimjee S. M., Dornbos D. III, Pitoc G. A., Wheeler D. G., Layzer J. M., Venetos N., Huttinger A., Talentino S. E., Musgrave N. J., Moody H., Rempel R. E., Jones C., Carlisle K., Wilson J., Bratton C., Joseph M. E., Khan S., Hoffman M. R., Sommerville L., Becker R. C., Zweier J. L., Sullenger B. A., Preclinical development of a vWF aptamer to limit thrombosis and engender arterial recanalization of occluded vessels. Mol. Ther. 27, 1228–1241 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R36] 36.Mi J., Zhang X., Rabbani Z. N., Liu Y., Su Z., Vujaskovic Z., Kontos C. D., Sullenger B. A., Clary B. M., H1 RNA polymerase III promoter-driven expression of an RNA aptamer leads to high-level inhibition of intracellular protein activity. Nucleic Acids Res. 34, 3577–3584 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R37] 37.Yu H., Lin L., Zhang Z., Zhang H., Hu H., Targeting NF-κB pathway for the therapy of diseases: Mechanism and clinical study. Signal Transduct. Target. Ther. 5, 209 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] 38.Bushweller J. H., Targeting transcription factors in cancer—From undruggable to reality. Nat. Rev. Cancer 19, 611–624 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R39] 39.Guo Q., Jin Y., Chen X., Ye X., Shen X., Lin M., Zeng C., Zhou T., Zhang J., NF-κB in biology and targeted therapy: New insights and translational implications. Signal Transduct. Target. Ther. 9, 53 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R40] 40.Taniguchi K., Karin M., NF-κB, inflammation, immunity and cancer: Coming of age. Nat. Rev. Immunol. 18, 309–324 (2018). [DOI] [PubMed] [Google Scholar]

[R41] 41.Ling A. L., Solomon I. H., Landivar A. M., Nakashima H., Woods J. K., Santos A., Masud N., Fell G., Mo X., Yilmaz A. S., Grant J., Zhang A., Bernstock J. D., Torio E., Ito H., Liu J., Shono N., Nowicki M. O., Triggs D., Halloran P., Piranlioglu R., Soni H., Stopa B., Bi W. L., Peruzzi P., Chen E., Malinowski S. W., Prabhu M. C., Zeng Y., Carlisle A., Rodig S. J., Wen P. Y., Lee E. Q., Nayak L., Chukwueke U., Gonzalez Castro L. N., Dumont S. D., Batchelor T., Kittelberger K., Tikhonova E., Miheecheva N., Tabakov D., Shin N., Gorbacheva A., Shumskiy A., Frenkel F., Aguilar-Cordova E., Aguilar L. K., Krisky D., Wechuck J., Manzanera A., Matheny C., Tak P. P., Barone F., Kovarsky D., Tirosh I., Suvà M. L., Wucherpfennig K. W., Ligon K., Reardon D. A., Chiocca E. A., Clinical trial links oncolytic immunoactivation to survival in glioblastoma. Nature 623, 157–166 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R42] 42.Wong F., He D., Krishnan A., Hong L., Wang A. Z., Wang J., Hu Z., Omori S., Li A., Rao J., Yu Q., Jin W., Zhang T., Ilia I., Chen J. X., Zheng S., King I., Li Y., Collins J. J., Deep generative design of RNA aptamers using structural predictions. Nat. Comput. Sci. 4, 829–839 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R43] 43.Delgoffe G. M., Xu C., Mackall C. L., Green M. R., Gottschalk S., Speiser D. E., Zehn D., Beavis P. A., The role of exhaustion in CAR T cell therapy. Cancer Cell 39, 885–888 (2021). [DOI] [PubMed] [Google Scholar]

[R44] 44.Lee Y. S., Dutta A., MicroRNAs in cancer. Annu. Rev. Pathol. 4, 199–227 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R45] 45.Hayes J., Peruzzi P., Lawler S. E., microRNAs in cancer: Biomarkers, functions and therapy. Trends Mol. Med. 20, 460–469 (2014). [DOI] [PubMed] [Google Scholar]

[R46] 46.Laggerbauer B., Engelhardt S., MicroRNAs as therapeutic targets in cardiovascular disease. J.Clin.Invest. 132, e159179 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R47] 47.Dhuppar S., Poller W. C., Murugaiyan G., MicroRNAs in the biology and hallmarks of neurodegenerative diseases. Trends Mol. Med. 31, 955–969 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R48] 48.Grigore I. A., Rajagopal A., Chow J. T., Stone T. J., Salmena L., Discovery of miRNA–mRNA regulatory networks in glioblastoma reveals novel insights into tumor microenvironment remodeling. Sci. Rep. 14, 27493 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R49] 49.Kim S. W., Li Z., Moore P. S., Monaghan A. P., Chang Y., Nichols M., John B., A sensitive non-radioactive Northern blot method to detect small RNAs. Nucleic Acids Res. 38, e98 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R50] 50.Nakashima H., Kaufmann J. K., Wang P.-Y., Nguyen T., Speranza M.-C., Kasai K., Okemoto K., Otsuki A., Nakano I., Fernandez S., Goins W. F., Grandi P., Glorioso J. C., Lawler S., Cripe T. P., Chiocca E. A., Histone deacetylase 6 inhibition enhances oncolytic viral replication in glioma. J. Clin. Invest. 125, 4269–4280 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Accessory microRNA byproducts expand RNA interference via microprocessor-mediated cleavage activation

Debora Mazzetti

Michal O Nowicki

Himanshu Soni

Joshua D Bernstock

Maya Groff

Luisa Esposito

Diego A Hernandez

Lucia Altucci

Anna Krichevsky

Geoffrey Fell

E Antonio Chiocca

Hiroshi Nakashima

Marco Mineo

Pierpaolo Peruzzi

Roles

Abstract

INTRODUCTION

RESULTS

microRNA processing releases accessory, elusive small ncRNA sequences

Fig. 1. Uncovering of persistent junk byproducts from microRNA transgenes.

Repurposing of flanking sequences to expand the biological impact of microRNA genes

Fig. 2. Functional repurposing of microRNA flanking sequences.

Fig. 3. Biological activity of chimeric microRNA transgenes.

Translational relevance of combinatorial microRNA interference

Fig. 4. In vivo relevance of bidirectional RNA interference.

microRNA genes are amenable to further repurposing in the stem-loop region

Fig. 5. Feasibility and validation of microRNA hairpin loop repurposing.

Expanding the applicability of RNA aptamers

Fig. 6. Biological impact of Mi-dAp50 interference.

DISCUSSION

MATERIALS AND METHODS

RNA sequencing

Prediction of secondary RNA structures

Exosome isolation

Nanosight analysis

MicroRNA overexpression

Virus particle production

Northern blot

Cell cultures

Gene expression studies

Luciferase reporter assay

Western blot

Clonogenic studies

Wound healing assay

Transwell assay

Flow cytometry analysis

Animal studies

Chromatin immunoprecipitation

RNA immunoprecipitation

Advanced TaqMan PCR

oHSV-1 infection

PCR verification of oHSV-1 infection

Metabolic assay

Plaque assay

Statistical and bioinformatics analysis

Illustrations

Acknowledgments

Funding:

Author contributions:

Competing interests:

Data and materials availability:

Supplementary Materials

The PDF file includes:

Other Supplementary Material for this manuscript includes the following:

REFERENCES

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases