Comparative Analysis of Dual‐miRNA Mediated Stimulation of Nucleus Pulposus and Bone Marrow Derived Stem Cells for Intervertebral Disc Repair Authors: Tara Ní Néill, Niamh Wilson, Jake McDonnell, Fergal J O'Brien, Caroline M Curtin, James E Dixon, Pieter A J Brama, Conor T Buckley Published: 2026 ABSTRACT ABSTRACTBackgroundIntervertebral disc (IVD) degeneration is the primary cause of low back pain (LBP). Disease progression is characterized by a shift in extracellular matrix homeostasis towards degradation over regeneration, a process that is a key target of emerging therapeutic strategies. Our laboratory has previously identified the dual delivery of FLR‐miRNA‐149‐5p mimic and 221‐3p inhibitor as being beneficial in generating an anti‐catabolic regenerative niche. Cell therapies have likewise demonstrated potential, though the most suitable cell source has been debated.Materials and MethodsIn this study, we explored the stimulation of nucleus pulposus and bone marrow derived stem cells with our therapeutic miRNA pair, with the aim of identifying the most suitable cell population for delivery to the IVD. Culture systems of increasing complexity ranging from monolayer to ex vivo organ culture were employed using caprine cells and IVD organs due to their physiological relevance to humans.ResultsWe demonstrated the maintenance of our previously reported anti‐catabolic niche, with significant downregulation of matrix degrading factors, ADAMTS5 and MMP13 and pro‐inflammatory cytokines, TNF and IL‐1β. Furthermore, delivery of dual‐miRNA stimulated nucleus pulposus cells in an organ culture model of mild IVD degeneration significantly upregulated the matrix markers, SOX9 and aggrecan. However, this matrix response was not identified in dual‐miRNA‐bone marrow derived stem cells delivered to goat IVD organs.ConclusionTaken together, these findings highlight the potential of nucleus pulposus cells as a regenerative cell source when combined with FLR‐miRNA‐149‐5p mimic and 221‐3p inhibitor for the treatment of LBP.Keywords: cell delivery, ex vivo organ culture, injectable therapeutics, intervertebral disc, microRNA Introduction Discogenic back pain represents a growing epidemiological concern worldwide, leading to neurological pain, impeded motor function, and diminished quality of life [1, 2]. Within the nucleus pulposus (NP) region of the intervertebral disc (IVD), breakdown of key extracellular matrix (ECM) proteins, such as the proteoglycan aggrecan, and collagen type II result in compromised tissue integrity. This dysregulation disrupts homeostatic balance, resulting in increased expression of catabolic enzymes such as disintegrin and metalloproteinase with thrombospondin motifs 5 (ADAMTS5) and matrix metalloproteinase 13 (MMP13) [3, 4, 5]. In parallel, a rise in pro‐inflammatory cytokines, specifically, interleukin‐1 beta (IL‐1β) and tumor necrosis factor (TNF) has been observed, amplifying the degenerative cascade [6, 7]. Treatments such as biomaterial delivery [8, 9, 10], cell therapies [11, 12, 13], and growth factors [14, 15, 16] have been trialed with varying degrees of success. More recently microRNAs (miRNAs), small non‐coding nucleic acids, have been explored as a promising therapeutic strategy, due to their ease of transfection, ability to regulate multiple genes, and lack of immunogenic response [17, 18, 19]. Of particular interest are miRNA‐149‐5p mimic, commonly identified as an anti‐inflammatory miRNA, and miRNA‐221‐3p inhibitor, associated with pro‐chondrogenic activity. Qin et al. reported that overexpression of miRNA‐149‐5p exerted a protective effect on NP cells under inflammatory conditions, restoring aggrecan and collagen type II expression following lipopolysaccharide (LPS) exposure, while concurrently downregulating MMP3 and ADAMTS4 [20]. Silencing of miRNA‐221‐3p has been primarily associated with cartilage‐related pathologies [21, 22], however in recent years has gained increasing attention in the IVD research field. Previous work has shown that treatment of disc derived cells with miRNA‐221 inhibitor significantly upregulated expression of collagen type II, SRY‐box transcription factor 9 (SOX9), and aggrecan (ACAN) at a gene level in monolayer culture [23]. Our laboratory have further demonstrated effective combinatorial delivery of these miRNAs with the cell penetrating peptide glycosaminoglycan (GAG)‐binding enhanced transduction (GET) peptide FLR [24, 25]. FLR specifically homes to GAG‐rich regions due to its fibroblastic growth factor 2 (FGF2) heparin binding domain, making it of particular interest to proteoglycan‐rich NP tissue [26, 27]. We have also demonstrated marked reductions in inflammatory and catabolic markers, alongside positive trends of pro‐ECM discogenic protein upregulation in rodent in vitro and ex vivo models [28, 29]. While miRNA delivery and transfection has shown promise as a therapeutic strategy for injured or damaged tissues, the synergistic potential of combining it with cell‐based approaches remains to be fully explored. Supplementing the injury site with viable cells could not only restore the tissue cell density but also promote paracrine interactions that support tissue homeostasis. Research to date has focused primarily on two principle cell populations as part of cell‐based regeneration strategies for disc repair, resident‐NP cells, and bone marrow‐derived mesenchymal stem cells (BMSCs), with both cell types offering distinct benefits and drawbacks. NP cells are particularly well suited to the disc milieu due to their inherent tolerance of acidic pH, nutrient poor environment, and altered osmotic conditions [30, 31, 32]. NP progenitor cells are currently being explored as a therapeutic source in human clinical trials conducted by Discgenics (NCT03955315 Japan, NCT03347708, United States), following preclinical studies in rabbits that demonstrated marked improvements in disc height index (DHI) after 8 weeks of cell treatment [33]. However, other studies have found that while NP cell transplantation does improve disc height, it remains below that of the healthy controls at the study end point, suggesting partial rather than full regeneration [34, 35, 36]. From 1994 to 2020 BMSCs were the most frequently transplanted cell type [37], valued due to their ability to differentiate towards a discogenic phenotype in vitro [38]. While their plasticity provides a potential advantage, the amount of matrix produced may be insufficient to elicit a regenerative response or restore lost disc height [39]. In an uninjured rabbit model, delivery of BMSCs significantly increased aggrecan and collagen type II gene expression, and sulphated (s)GAGs at the 6 month endpoint [40]. Delivery of BMSCs to a goat model of injured IVDs saw significant increases in proteoglycans compared to their hydrogel control, however no changes in collagen content or MRI score were identified [41]. A recent meta‐analysis investigating the effects of stem cell delivery reported positive outcomes in disc height restoration in rat models. However, this effect was not observed in ovine models, where no significant differences were found between control and stem cell groups, suggesting potential species‐specific responses [42] or effects related to the larger anatomical scale of ovine discs. Nonetheless, the therapeutic potential of BMSCs should not be overlooked given their multipotent capacity. To date, no studies have investigated the effects of miRNA or dual‐miRNA transfected cells delivered to the IVD, nor their combined potential to replenish the depleted cell population, drive regenerative processes, and modulate catabolic activity and inflammation. Therefore, the objective of this study was to investigate the effects of our previously identified dual‐miRNA stimulation approach on caprine NP cells and BMSCs in vitro and ex vivo, assessing the regenerative and anti‐catabolic potential of these transfected cell populations to determine their suitability for cell transplantation into degenerated IVDs. Materials and Methods Primary Cell Culture In accordance with the principles of the 3R's and to minimize animal use, goat NP and BMSCs were obtained from discarded material from skeletally mature Saanen goats used in a separate study approved by the University College Dublin Animal Research Ethics Committee (AREC‐P‐23‐17) and the Irish Healthy Products Regulatory Authority (HPRA AW18982/P142). Briefly, motion segments of the lumbar spine were isolated to allow for NP tissue dissection. The tissue was enzymatically digested in serum free low glucose DMEM (LG DMEM, Sigma‐Aldrich) containing pronase (100 U/mL, Millipore) and collagenase Type II (300 U/mL, Gibco) supplemented with 2% penicillin/streptomycin (pen/strep, Gibco) and 2.5 mg/mL Amphotericin B (Sigma‐Aldrich). Digestion was carried out at 37°C with gentle agitation (10 r.p.m.) until single cells were released. To isolate BMSCs, bone marrow was aspirated from the sternum with a 16G needle and subjected to sequential centrifugation and filtration steps. Cells were seeded at a density of 23 × 106/T175 flask to promote colony formation and cultured for 72 h. Red blood cells were then removed by washing with phosphate buffered saline (PBS, Sigma). Both cell types were cultured in LG DMEM as before supplemented with 10% fetal bovine serum (FBS, Gibco), referred to as expansion (XPAN) media, under physioxic conditions (5% O2 and 5% CO2) until passage 1, after which they were either used immediately for experiments or cryopreserved in liquid nitrogen (LN2). NP cells were used up to passage 3, and BMSCs up to passage 4 for all experiments. All experiments were performed under physioxic conditions unless otherwise stated. Tripotentiality Determination of BMSCs To assess the multilineage potential of isolated BMSCs, at passage 1 cells were treated with chondrogenic, adipogenic, and osteogenic medias as previously described [43, 44, 45]. Briefly, for chondrogenic differentiation, following expansion in XPAN, cell aggregates comprising of 2.5 × 105 cells were formed by centrifugation. The aggregates were then cultured for 21 days in a chemically defined medium (CDM), consisting of high glucose (HG) DMEM + GlutaMAX, 2% Pen/Strep (both Gibco), 2.5 mg/mL Amphotericin B (Sigma‐Aldrich), 5 mM sodium pyruvate, 40 μg/mL L‐Proline and 1.5 mg/mL bovine serum albumin (BSA, all Sigma‐Aldrich). For chondrogenic induction, the CDM was further supplemented with 1× insulin transferrin selenium (ITS, Gibco), 10 ng/mL TGF‐β3 (PeproTech), 100 nM dexamethasone, 50 μg/mL L‐ascorbic acid‐2‐phosphate and 4.7 mg/mL linoleic acid (all Sigma‐Aldrich). Adipogenic differentiation was induced in monolayer by seeding cells at a density of 1 × 103 cells/cm2 and supplementing XPAN media with 100 nM dexamethasone, 0.5 mM IBMX, and 50 μM indomethacin (all Sigma‐Aldrich). Osteogenic differentiation was performed similarly, with media instead supplemented with 100 nM dexamethasone, 10 mM β‐glycerol phosphate, and 56.6 μM L‐ascorbic acid‐2‐phosphate (All Sigma‐Aldrich). Differentiation outcomes were assessed with Alcian blue staining for chondrogenesis, Oil‐Red‐O for adipogenesis, and Alizarin red for osteogenesis. Results showing positive staining following chondrogenic, adipogenic, and osteogenic induction are presented in Figure S1. Determination of Transfection Efficiency in Monolayer NP and BMSC Cultures To achieve efficient miRNA transfection, a fluorescently labeled miRNA, miRDy547 (Dharmacon) was delivered to both cell types in monolayer culture using the FLR peptide, a glycosaminoglycan‐binding enhanced transduction (GET) peptide (Dixon Lab, Nottingham) [46, 47]. Briefly, cells were seeded at a density of 5 × 103 in μ‐Slide 18 well chamber slides (ibidi, GmbH) and allowed to attach overnight. Subsequently, to simulate future experimental conditions, cells were serum starved for 24 h and subjected to a cytokine challenge with TNF (50 ng/mL) and IL‐1β (10 ng/mL, both Fischer Scientific) for 24 h as previously reported [28, 29, 48]. miRNA complexes were prepared at increasing doses from 0 to 25 ng/μL in OptiMEM reduced serum media (Gibco), maintaining a charge ratio of N:P of 5. Complexes were allowed to form via electrostatic interactions for 30 min at room temperature before being added to the cells. Transfection was allowed to proceed for 6 h at 5% O2 and 37°C, upon which the complexes were removed and cultured in XPAN for 3 days. Cells were either fixed in 4% paraformaldehyde (PFA) for 12 min for immunofluorescence or trypsinized for flow cytometric analysis using a LSRFortessa Flow cytometer, with nuclear counterstaining performed using Hoechst 33324 (Merck Life Sciences Ltd.) in both assays. A non‐transfected (NT) control was included for all transfection experiments. Assessment of Protein Marker Expression Following miRNA‐Pair Delivery in Monolayer Culture Protein expression of specific markers in monolayer culture following miRNAs transfection was performed. Cells were seeded as previously described, with cytokine treatment applied as described in Section 2.3 for analysis of catabolic markers, while cultures without a cytokine challenge were used to assess anabolic protein expression. FLR complexes were formed with either a scrambled control miRNA (Dharmacon), or miRNA149‐5p mimic and miRNA‐221‐3p inhibitor combined at a concentration of 10 ng/μL. Protein expression was analyzed 3 days post‐transfection via immunofluorescence following antibody staining. To further examine the effects of transfection, NP cells and BMSCs were seeded in 6‐well plates at a density of 1 × 105 cells per well. Following overnight attachment, cells were exposed to serum starvation and a cytokine challenge as before. FLR‐scrambled miRNA complexes (10 ng/μL) were delivered and incubated for 6 h. Cells were maintained in XPAN media for 14 days with biweekly media changes, and biochemical analyses were performed at the end of the culture period. Microtissue Culture of miRNA Transfected NP and BMSCs To assess the effects of miRNA‐pair transfection in a more physiologically relevant 3D system, microtissues were generated from NP cells and BMSCs. Cells were seeded at a density of 1 × 106 in 6‐well plates. After overnight attachment, cells underwent serum starvation followed by cytokine treatment, both for 24 h as described above. Transfection proceeded for 6 h using 10 ng/μL of complexes containing either FLR‐scrambled miRNA or FLR‐miRNA‐149‐5p mimic combined with miRNA‐221‐3p inhibitor. Cells were allowed to recover overnight and the following day microtissues containing 5 × 104 cells each were formed following centrifugation at 800 g for 10 min in sterile agarose coated 48‐well plates. Media exchanges were performed twice weekly, with collected media pooled to allow for quantification of sGAG release during the culture period. Microtissues were maintained for either 3 or 14 days. At the relevant timepoints, 2–3 microtissues were pooled for biochemical analysis, and additional microtissues were fixed for 30 min in 4% PFA at room temperature for histological evaluation. Goat Ex Vivo Organ Culture and Induction of Mild Degeneration All goat tissue used for ex vivo organ culture was obtained with approval from the Animal Research Ethics Committee (AREC), University College Dublin (UCD) (AREC‐23‐17). Spines were dissected and excess fascia was removed to expose the entire disc region, whereupon motion segments were excised containing vertebra‐disc‐vertebra. The segments were incubated overnight in LG DMEM supplemented with 2% pen/strep, 0.1 μL/mL Amphotericin B (Sigma), and 100 μg/mL Kanamycin Sulfate (Gibco). To simulate disc biomechanics each intact motion segment was then cultured in XPAN media in a custom‐made bioreactor with an applied pressure of 0.4 MPa, a value within the human and goat physiological range [49, 50]. Mild degeneration was induced by a single injection of 200 μL of 2 units (U) chondroitinase ABC (cABC, Sigma‐Aldrich) via a 30G needle, and segments were maintained under these conditions for 14 days prior to treatment administration. miRNA Dosage for Ex Vivo Organ Culture Delivery Uptake of miRNA into mildly degenerated goat disc organs was evaluated using the fluorescently labeled miRDy547. Complexes of 10 and 25 ng/μL (200 μL per disc) were delivered via a 30G needle, with a sham OptiMEM (vehicle) NT control. Discs were cultured for 72 h, after which the intact IVDs were excised, snap frozen in liquid nitrogen and embedded in optimal cutting temperature (OCT) compound (Fischer Scientific). Upon freezing, sections of 10 μm were cut in the sagittal orientation using a Leica CM 3050S Cryostat. miRNA uptake by the cells was assessed using a Leica SP8 scanning confocal microscope (Leica, Germany), with nuclei counterstained using Hoechst 33342. Ex Vivo Organ Culture Treatment Administration Organ culture experiments were divided into five groups, as outlined in Table 1. A healthy control disc without cABC insult was included, while all other discs received cABC to induce mild degeneration, as determined in a pilot study by decreased staining for sGAGs and an increased MMP and ADAMTS response (Figure S2). Treatment groups were administered as follows: The NT group received a sham injection of OptiMEM (vehicle). The miRNA‐only group was injected with 200 μL of FLR‐miRNA‐149‐5p mimic + miRNA‐221‐3p inhibitor complexes at a concentration of 25 ng/μL. This higher dose was found to be the most efficient for organ culture delivery while maintaining strong viability in comparison to the NT control (Figure S3). For cell‐based treatments, NP cells and BMSC were transfected with FLR‐miRNA‐149‐5p mimic and miRNA‐221‐3p inhibitor as described in Section 2.3. Cells were transfected for 6 h; complexes were replaced with XPAN media and the cells allowed to recover overnight prior to trypsinization for ex vivo organ delivery. A total of 5.5 × 106 cells were delivered to each disc, a number predicted to restore goat matrix synthesis to a healthy level in vivo [51]. Prior to injection, both cell types were labeled with PKH‐26 dye (Sigma) as per the manufacturer's instructions to allow visualization. After termination of culture, NP tissue was dissected and removed from each disc, weighed, and stored at −70°C. Tissues were freeze dried using a standard procedure (500 mTorr, −10°C, 16–18 h, Harvest Right, USA), re‐weighed to determine hydration levels, and processed for biochemical evaluation. For histological evaluation, excess tissue and bone were removed, and discs were fixed in 4% PFA overnight at 4°C. TABLE 1Treatment groups, description and details for ex vivo organ culture.Group nameDegenerationTreatment descriptionTreatment detailsHealthyNoneNon‐degenerated healthy controlNo treatmentNon‐transfected (NT)200 μL 2 U chondroitinase ABC (cABC)Degenerated control200 μL Opti‐MEM (vehicle) treatedmiRNA‐only200 μL 2 U chondroitinase ABC (cABC)FLR‐miRNA‐pair treated200 μL FLR miRNA‐149‐5p + 221‐3p inhibitor (25 ng/μL)miRNA‐NP200 μL 2 U chondroitinase ABC (cABC)FLR‐miRNA‐pair transfected nucleus pulposus (NP) cells200 μL 5.5 × 106 NP cells transfected with FLR miRNA‐149‐5p + 221‐3p inhibitor (10 ng/μL)miRNA‐BMSC200 μL 2 U chondroitinase ABC (cABC)FLR‐miRNA‐pair transfected bone marrow derived stem cells (BMSCs)200 μL 5.5 × 106 BMSCs transfected with FLR miRNA‐149‐5p + 221‐3p inhibitor (10 ng/μL) Treatment groups, description and details for ex vivo organ culture. Evaluation of Cell Viability in Microtissues and Ex Vivo Disc Organ Cultures At days 3 and 14 of culture, microtissues were washed with PBS and incubated for 1 h at 37°C in a staining solution consisting of 2 μM Calcein (Sigma‐Aldrich) and 4 μM ethidium homodimer (EthD‐1, SLS Scientific Laboratory Supplies). For evaluation of cell viability in ex vivo disc organ cultures, the intact disc was dissected from the vertebral bodies, washed with PBS, and incubated for 1 h at 37°C in the live/dead staining solution. Following aspiration of the staining solution and three washes of PBS, the tissue was embedded in OCT and snap frozen in liquid nitrogen. Sections of 10 μm were acquired on a Leica 3050S Cryostat. Sections were imaged on a Leica SP8 confocal microscope to allow visualization of live and dead cells (485 nm excitation, 530 nm emission for Calcein, 530 nm excitation, 645 nm emission for EthD‐1). Images were analyzed using ImageJ software (version 1.54p) to generate maximum‐intensity projection z‐stacks for quantification of live and dead cell populations. Biochemical Quantification of in Vitro and Ex Vivo Organ Culture Experiments For in vitro experiments pooled media was freeze‐dried as previously described, and both cells and media were digested for 18 h at 60°C in 100 mM sodium phosphate/5 mM Na2EDTA buffer with 3.88 U/mL papain enzyme (all Merck Life Science Ltd.). Tissue samples from ex vivo organ cultures were weighed before and after freeze drying to assess hydration and were similarly digested with papain. DNA content was quantified using the Hoechst Bisbenzimide 33258 dye assay (Merck Life Science Ltd.) with a calf thymus DNA standard. sGAGs were determined using the dimethylmethylene blue binding assay (DMMB) (Blyscan, Biocolor Ltd.), with a chondroitin sulphate (CS) standard. Total collagen content was determined by hydroxyproline measurement. Samples were hydrolyzed in 38% concentrated hydrochloric acid for 18 h at 110°C, followed by a chloramine‐T assay, and collagen content calculated using a hydroxyproline‐to‐collagen ratio of 1:7.69 [52]. Histology and Immunocytochemistry of in Vitro and Ex Vivo Organ Culture Following fixation, microtissues were embedded in 4% (w/v) agarose and sequentially dehydrated through graded ethanol solutions and embedding in paraffin wax. Sections of 7 μm thickness were prepared using a microtome (Leica Microsystems). Goat discs were decalcified in Decalcifying Solution‐Lite (Merck Life Sciences Ltd.) for approximately 4 weeks, until softened, followed by incubation in 30% sucrose solution and OCT embedding. Sections of 10 μm were prepared using a Leica CM3050 S Cryostat. Histological staining included hematoxylin and eosin (H&E) to visualize cellular morphology, picrosirius red (PR) for collagen deposition, and alcian blue (AB) and safranin‐O for GAG detection. Imaging was performed on an Aperio AT2 Slide Scanner (Leica). For immunocytochemistry, antigen retrieval was performed on paraffin embedded microtissue sections using sodium citrate buffer (pH 6) heated to 100°C in a pressure cooker for 1 min. Both microtissue and organ culture membranes were permeabilized with 0.1% Triton X‐100 and blocked for 1 h in 5% BSA in PBS‐Tween. All antibodies used are detailed in Table 2. Primary antibodies were applied overnight at 4°C followed by incubation with relevant species‐matched Alexa Fluor secondaries for 2 h at room temperature. Nuclei were counterstained with Hoechst Bisbenzimide 33342. Imaging was performed on a Leica Scanning SP8 confocal microscope, where the corrected total cell fluorescence was quantified for regions of interest, (i.e., the microtissues or NP tissue) in ImageJ software (version 1.54p), and normalized to cell number. TABLE 2Primary and secondary antibodies utilized for immunocytochemistry.AntigenSpeciesConcentrationSupplier and codeAggrecanRabbit1:250Invitrogen, MA532695Collagen type IIRabbit1:250Invitrogen, PA599159Collagen type IAIRabbit1:150Invitrogen, PA5‐29569SOX9Rabbit1:250Invitrogen, PA586301ADAMTS5Rabbit1:250Invitrogen, PA532142MMP13Mouse1:125Invitrogen, MA514247TNFRabbit1:100Invitrogen, PA519810IL‐1βRabbit1:200Invitrogen, P420BAnti‐Rabbit Alexa Fluor 488Goat1:500Invitrogen, A27034Anti‐Mouse Alexa Fluor 488Goat1:2000Invitrogen, A11001 Primary and secondary antibodies utilized for immunocytochemistry. Statistical Analysis All statistical analyses were performed using GraphPad Prism version 10.4.0. Comparisons between protein expression levels in monolayer were assessed using a one‐way ANOVA followed by Tukey's post hoc test. Differences between groups at different time points were determined using a two‐way ANOVA with Tukey's post hoc test. Statistical significance was designated as p ≤ 0.05. A minimum of three independent biological replicates were performed for all experiments, and each data point presented represents an individual biological donor. Graphical data is presented as mean ± standard deviation (SD). Results Efficient microRNA Internalization Was Achieved, and the Scramble Control Showed no Functional Effects in NP Cell or BMSC Cultures The success of miRNA therapy hinges upon effective nanoparticle delivery, while the vector employed should not adversely alter cell function. Efficient internalization was achieved following FLR‐miRDy547 transfection at concentrations above 5 ng/μL (Figure 1A), with a plateau reached at 10 ng/μL (mean NP 79.4% ± 6.0, mean BMSC 85.0% ± 10.5, p ≤ 0.0001 for both NP and BMSC compared to 0 ng/μL, Figure 1B). At the highest concentration of 25 ng/μL, NP cell number decreased significantly compared to the 10 ng/μL dose (p = 0.016), with BMSCs following a similar trend (5 ng/μL vs. 25 ng/μL, p = 0.085), indicating a detrimental effect at elevated concentrations (Figure 1C). FLR‐scrambled miRNA transfection did not induce functional changes, with no significant differences in DNA, sulphated glycosaminoglycan (sGAG), or collagen levels in either two‐ or three‐dimensional cultures (Figure S3). Dosage study of FLR delivered fluorescently tagged miRDy547 in goat nucleus pulposus (NP) and bone marrow‐derived stem cells (BMSC) 3 days post‐transfection, showing a plateau in internalization efficiency at 10 ng/μL for both cell types. (A) Representative images of miRDy547 internalization. (B) Degree of miRDy547 internalization efficiency (%) as assessed by flow cytometry. (C) Fold change in cell count normalized to non‐transfected (NT) control. Statistical analysis: one‐way ANOVA with Tukey's post hoc analysis. N = 3 independent biological donors. * p < 0.05, *** p < 0.001, **** p < 0.0001 indicates statistical significance. Scale bar = 200 μm. miRNA‐Paired Delivery Induced Cell Type‐Specific Differences in Matrix Degrading Enzyme Expression in Monolayer Cultures Driving ECM protein production while reducing catabolic activity is critical when assessing potential treatments targeting IVD degeneration. In monolayer culture, a modest upward trend in aggrecan protein expression was observed in comparison to the NT control (Figure 2A,B BMSC p = 0.0737), while collagen type II was unchanged 3 days post‐transfection for both cell types (Figure 2C,D). However, following cytokine stimulation to promote a catabolic response, the cell types exhibited distinct post‐transfection responses. The key ECM degrading protein ADAMTS5 was significantly upregulated in transfected BMSCs compared to NP cells (Figure 2E,F, p < 0.001), while transfected NP cells showed a significant decrease in expression relative to NT controls (p = 0.015). Similarly, MMP13 expression was increased in BMSCs and significantly higher than in NP cells (Figure 2G,H, p = 0.019). Comparison of extracellular matrix (ECM) markers, and catabolic enzymes 3 days post miRNA‐pair transfection in goat nucleus pulposus (NP) and bone marrow‐derived stem cells (BMSCs). (A) Fluorescence intensity was quantified for aggrecan (A and B, ACAN), collagen type II (C and D, COL2), ADAMTS5 (E and F), and MMP13 (G and H) and normalized to the non‐transfected (NT) control, with representative images presented for each condition. Statistical test is one‐way ANOVA with Tukey's post hoc analysis. N = 6 independent biological donors, * p < 0.05, *** p < 0.001 indicates statistical significance. Scale bars = 50 μm. Microtissues Maintained Strong Viability Throughout the Culture Period To investigate the effect of miRNA‐pair stimulation in a more relevant 3D microtissue model, we first confirmed that viability was maintained throughout the culture period. DNA content increased significantly for both NT‐NP and miRNA‐BMSC treated groups from day 3 to 14 (Figure 3A,B p = 0.0142 and p = 0.0081, respectively). However, no differences were identified between any of the groups at the culture endpoint. Live/dead staining at both early and late timepoints confirmed these findings. Although a significant reduction in viability was observed in the NT‐BMSC group compared to both NP groups (Figure 3C,D, p = 0.0067 compared to NT‐NP and p = 0.0024 compared to miRNA‐NP), viability remained above 88% across all groups at both timepoints. DNA quantification, representative hematoxylin and eosin (H&E) staining, and viability of non‐transfected (NT) and FLR‐miRNA‐149 mimic + miRNA‐221‐inhibitor (miRNA‐pair) 3D microtissues formed from nucleus pulposus (NP) and bone marrow‐derived stem cells (BMSCs) at timepoints 3 (T3) and 14 (T14) days of culture. (A) DNA content (μg DNA/sample). (B) Representative H&E images of NP and BMSC microtissues. (C) Live cell viability (%) in microtissues was assessed using live/dead staining. (D) Representative images of live/dead staining. Statistical test is a two‐way ANOVA with Tukey's post hoc analysis. N = 6 independent biological donors. * p < 0.05, ** p < 0.01 indicates statistical significance. Scale bars = 100 μm. miRNA Transfection Promoted Divergent Upregulation of Early and Late‐Stage Extracellular Matrix Markers in 3D Cultures of NP and BMSCs Given the crucial role of GAGs in the formation of ECM, we assessed total sGAG production in microtissue cultures, alongside proteins of interest at early and late‐stages (Figure 4). sGAG deposition was significantly upregulated across all groups following 2 weeks of microtissue culture (Figure 4A–C). The total sGAGs deposited by miRNA‐BMSCs were significantly higher than in NP groups (Figure 4A p = 0.0044 vs. NT‐NP, and p = 0.0009 vs. miRNA‐NP). However, this effect between groups was abrogated after normalization to DNA content, with significant differences remaining only between timepoints (Figure 4B, p < 0.0001). Qualitatively this was reflected in increased GAG production as visualized by Alcian blue staining, with deeper staining observed at later timepoints across all groups (Figure 4C). Immunofluorescence analysis revealed a significant increase in the early chondrogenic marker SOX9 in miRNA‐pair transfected NP cells at the end of culture compared to NT controls (Figure 4D,E, p = 0.0322). Conversely, the later‐stage proteoglycan ACAN was substantially upregulated in transfected BMSCs relative to transfected NP cells, and both NT controls (Figure 4F,G, p = 0.0282, p = 0.0088 respectively). Biochemical quantification, histological, and immunofluorescence analysis of matrix‐specific markers of non‐transfected (NT) and FLR‐miRNA‐149 mimic + miRNA‐221‐inhibitor (miRNA‐pair) 3D microtissues formed from nucleus pulposus (NP) and bone marrow‐derived stem cells (BMSCs) at timepoints 3 (T3) and 14 (T14) of culture. (A) Sulphated glycosaminoglycans (sGAGs) per sample (μg) and (B) normalized to DNA content (μg/μg) (C) Representative images of Alcian blue (AB) staining for sGAGs. (D) SOX9 protein expression was quantified using immunofluorescence, NP‐transfected cells exhibited significant increases by day 14. (E) Representative images of SOX9 protein staining. (F) Aggrecan protein expression as quantified by immunofluorescence. By day 14, BMSCs exhibited significantly higher expression compared to all other groups (G) Representative images of aggrecan staining. Statistical test is a two‐way ANOVA with Tukey's post hoc analysis. N = 6 independent biological donors, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 indicates statistical significance. Scale bars = 20 μm for main image and 100 μm for inset of entire microtissue. miRNA Stimulation Increased Total Collagen Content and Specific Collagen Markers in Microtissue Cultures Alongside GAG formation, collagens, particularly collagen type II, are crucial for maintaining NP ECM integrity. Total collagen content increased as a function of time across all treatment groups except NT‐NP (Figure 5A,B p ≤ 0.05). However, following normalization to DNA content this effect was abolished, with no differences observed between any of the groups (Figure 5C). Likewise, minor increases in Picrosirius red (PSR) staining intensity was observed in the transfected groups. miRNA‐BMSCs exhibited increased expression of the more fibrotic collagen type I compared with NT controls (Figure 5D,E, p = 0.033), while collagen type II showed a trend towards higher expression in the transfected BMSC group (Figure 5F,G, p = 0.1025). Biochemical quantification, histological, and immunofluorescence analysis of collagen‐specific markers of non‐transfected (NT) and FLR‐miRNA‐149 mimic + miRNA‐221‐inhibitor (miRNA‐pair) 3D microtissues formed from nucleus pulposus (NP) and bone marrow‐derived stem cells (BMSCs) at timepoints 3 (T3) and 14 (T14) of culture. (A) Collagen content per sample (μg) and (B) normalized to DNA content (μg/μg). (C) Representative histological images of microtissues following Picrosirius Red staining for collagen. (D) Protein expression of COL1A1 and (E) representative immunofluorescent images for DAPI/COL1, (F) COL2A1 protein expression and (G) representative immunofluorescent images for DAPI/COL2. Statistical test is a two‐way ANOVA with Tukey's post hoc analysis. N = 6 independent biological donors, * p < 0.05, ** p < 0.01 indicates statistical significance. Scale bars = 20 μm for main image and 100 μm for inset of entire microtissue. miRNA Transfection Induced Differential Catabolic and Inflammatory Responses Between Cell Types Restoring homeostatic balance in the degenerated IVD requires the suppression of catabolic and inflammatory mediators. Consistent with our monolayer studies, ADAMTS5 expression was substantially elevated in transfected BMSC cells compared to both NT controls and miRNA‐NP cells (Figure 6A,B p = 0.0236 and p = 0.0038, respectively). In contrast to the 2D experiments, MMP13 expression in miRNA‐BMSCs was strongly attenuated after 14 days in culture compared with controls (Figure 6C,D p < 0.0001). As observed in our monolayer studies, transfected NP cells demonstrated marked reductions in MMP13 expression (p = 0.0158). To gain deeper insight into inflammatory activity, IL‐1β and TNF were assessed. IL‐1β showed an upward trend in miRNA‐BMSCs compared with miRNA‐NP cells (Figure 6E,F p = 0.0746). TNF levels demonstrated modest decreases overall, however transfected NP cells recovered from an initial spike at day 3 and showed a significant reduction by day 14 (Figure 6G,H p = 0.028). Immunofluorescence protein quantification and representative images of catabolic enzymes and pro‐inflammatory cytokines in non‐transfected (NT) and FLR‐miRNA‐149 mimic + miRNA‐221‐inhibitor (miRNA‐pair) 3D microtissues formed from nucleus pulposus (NP) and bone marrow‐derived stem cells (BMSCs) at timepoints 3 (T3) and 14 (T14) of culture. (A and B) ADAMTS5 expression was significantly increased in the BMSC‐transfected group. (C and D) MMP13 protein expression was reduced for both cell types. (E and F) IL‐1β showed no significant changes, however there was evidence of an upward trend for the BMSC transfected groups. (G and H) TNF levels in miRNA‐pair transfected NP cells significantly decreased by the end of the study. Statistical test is a two‐way ANOVA with Tukey's post hoc analysis. N = 6 independent biological donors, * p < 0.05, ** p < 0.01, *** p < 0.001, ****p < 0.0001 indicates statistical significance. Scale bars = 20 μm for main image and 100 μm for inset of entire microtissue. Dual‐miRNA and Transfected‐Cell Delivery Promoted Discogenic Protein Expression While Suppressing Catabolic and Inflammatory Factors in Ex Vivo Organ Culture Ex vivo organ culture represents one of the most physiologically relevant models available in the disc field outside of in vivo models. The dose of 2 U cABC was established through a pilot study to determine a mild rather than severe degree of degeneration (Figure S2). We demonstrated a reduction in sGAGs through Alcian Blue staining (Figure S2A) and a significant increase in ADAMTS5 protein expression with 2 U compared to the healthy control following 14 days of degeneration (Figure S2C). It is important to note, that while the higher dose of 4 U elicited a stronger response in relation to sGAG degeneration alongside MMP13 and ADAMTS5 expression, this magnitude of response is consistent with severe degeneration, beyond the scope of our proposed treatment strategy. Furthermore, it has been shown in vivo that degeneration will proceed throughout the study period [53], therefore, a lower insult of 2 U cABC is a more appropriate model of mild degeneration. To ensure the suitability of the model, FLR‐miRDy547 uptake was assessed in mildly degenerated goat IVDs, with a dose of 25 ng/μL selected for further studies due to its significant overexpression and retention within cells (Figure S4A). While in monolayer culture this higher dose was detrimental to cell health, viability was unaffected by miRNA delivery in the organ, perhaps due to the protective ECM niche within the tissue (Figure S4B). Following delivery, PKH26‐labeled NP cells and BMSCs were confirmed to be retained within the NP region at the end of the culture period (Figure S4C). Following treatment with dual‐miRNA only or with stimulated cells, histological and biochemical assessment was carried out to assess cell and matrix composition. H&E staining demonstrated maintenance of the cell population despite degeneration and confirmed retention of transplanted NP cells and BMSCs as indicated by the black arrows in Figure 7A. These observations were supported by DNA quantification (Figure 7H). Proteoglycan content was evaluated using AB staining (Figure 7C). Although modest increases in AB staining were observed in treatment groups, quantitative sGAG measurements revealed no significant differences (Figure 7C,I). Specific markers were assessed to examine potential differences in early (SOX9) and late‐stage (aggrecan) matrix protein production. Delivery of miRNA‐pair transfected NP cells elevated SOX9 protein expression compared to the healthy, NT, and BMSC miRNA groups (Figure 7D,E, p = 0.042, 0.02, and 0.106, respectively). This enhancement of matrix markers was reflected in aggrecan synthesis, which was strongly upregulated in the NP‐miRNA group compared to controls (Figure 7F,G, p = 0.038). Collagen content assessed by PSR showed no substantial changes consistent with biochemical analysis (Figure 7B,J). Finally, no differences in the sGAG to collagen ratio was observed for any of the groups investigated (Figure 7K). All ex vivo organ culture biochemical data are presented in Figure S5. No changes to overall tissue hydration were observed (Figure S5A). DNA content also remained unchanged, irrespective of normalization to hydration, wet weight, or dry weight (Figure S5B). Similarly, no significant differences were observed for sGAGs, whether expressed per sample, or normalized to DNA content, hydration, wet weight, or dry weight (Figure S5C). Collagen content was unchanged across all normalization methods (Figure S5D) and the sGAG to collagen ratio remained consistent whether expressed per sample or per weight (Figure S5E). Overall, we identified the upregulation of anabolic markers following delivery of miRNA‐pair transfected NP cells, with modest increases to sGAG staining. Histological, biochemical, and immunofluorescence evaluation of goat ex vivo organ culture following dual delivery of miRNA‐149‐5p mimic and 221‐3p inhibitor, or miRNA‐pair stimulated goat nucleus pulposus (NP) and bone marrow‐derived stem cells (BMSC) following 28 days of culture. (A) Representative images of histological staining for hematoxylin and eosin (H&E), (B) Picrosirius red (PSR), (C) Alcian blue (AB). (D and E) SOX9 expression was elevated in NP miRNA cells compared to controls (healthy and non‐transfected, NT). (F and G) ACAN expression was also higher in the NP‐miRNA treated group compared to NT control. (H) Biochemical quantification of DNA (μg/sample), (I) sGAGs (μg/μg), (J) collagen (COL, μg/μg) production and (K) sGAG:COL (μg/μg) ratio during the 14‐day culture period. Statistical test is one‐way ANOVA with Tukey's post hoc analysis. N = 3 independent biological donors. * p < 0.05 indicates statistical significance. Scale bars for whole organ view = 2 mm, magnified images of the NP region = 400 μm, indicated by magnifying glass and white dashed lines. Following evaluation of discogenic markers, the effects of miRNA and miRNA transfected cells on the protein expression of catabolic and inflammatory targets was evaluated. cABC‐induced degeneration in the NT group elicited a substantial increase for all targets when compared to the healthy non‐degenerated controls, as indicated by the white arrow heads in Figure 8. ADAMTS5 expression was significantly suppressed by miRNA‐paired delivery (Figure 8A,B p = 0.0088). MMP13 levels were significantly attenuated across all treatments groups compared to the degenerated NT control (Figure 8C,D p = 0.0082, p = 0.007, and p = 0.0239, for miRNA‐pair, NP‐miRNA, and BMSC‐miRNA vs. NT, respectively). Similarly, IL‐1β activity was strongly diminished following treatment compared with NT controls (Figure 8E,F p = 0.0002 for miRNA‐pair, p < 0.0001 for both NP‐miRNA and BMSC‐miRNA). Finally, TNF production was suppressed across all treatment groups when compared to the degenerated NT controls (Figure 8G,H p = 0.001, p = 0.0004, p = 0.0007, for miRNA‐pair, NP‐miRNA, and BMSC‐miRNA, respectively). Collectively here we demonstrate the suppression of key ECM degrading proteins. Aside from the reduction of ADAMTS5, which was specific to miRNA‐pair delivery, no notable differences were observed between groups suggesting no clear advantage of NP cells versus BMSCs for suppressing these degradative factors in the NP region. Immunofluorescence analysis and representative images of catabolic enzyme and pro‐inflammatory cytokine protein levels in ex vivo organ culture following miRNA‐pair delivery alone, or transfected nucleus pulposus (NP) and bone marrow‐derived stem cell (BMSC), post‐28 days in culture. (A and B) ADAMTS5 protein expression was markedly reduced following miRNA‐paired delivery. (C and D) MMP13 was significantly downregulated across all treatment groups. (E and F) Protein production of IL‐1β was suppressed post‐treatment delivery. (G and H) TNF activity was strongly diminished by all treatments. Statistical test is one‐way ANOVA with Tukey's post hoc analysis. N = 3 independent biological donors. * p < 0.05, ** p < 0.01, *** p < 0.001, ****p < 0.0001 indicates statistical significance. Scale bars for whole organ view = 2 mm, magnified images of the NP region = 400 μm, indicated by magnifying glass and white dashed lines. White arrow heads indicate regions of increased protein expression. Discussion Dysregulation of the ECM of the IVD, which manifests clinically as LBP, currently lacks robust treatment options beyond conservative pain management and surgical intervention, representing a significant unmet clinical need to target and restore tissue homeostasis. The therapeutic potency of miRNA‐guided augmentation of cellular phenotypes has gained significant appreciation in recent years. Similarly, the efficacy of cell transplantation to promote IVD anabolism while curbing catabolism has been debated. While some studies report increased ECM production and improvements to disc height following cell transplantation, others have observed no changes in disc height, and minimal regeneration [33, 34, 35, 37, 39, 54]. In this study, we explored the delivery of two cell populations following dual‐miRNA stimulation, to determine whether the delivery of a transfected cell source offers advantages over administering the miRNA‐pair only. Central to our results was the sustained anti‐catabolic effect previously attributed to the combined delivery of FLR‐miRNA‐149‐5p mimic + miRNA‐221‐3p inhibitor [28, 29]. Furthermore, we demonstrated the upregulation of discogenic markers following the delivery of this miRNA‐pair transfected NP cells in a goat ex vivo organ culture model of mild IVD degeneration. Taken together, our results highlight the potency of miRNA‐149‐5p mimic +221‐3p inhibitor delivery while also identifying encouraging restorative trends following delivery of NP cells transfected with our miRNA pair in ex vivo organ culture. The onset of degeneration and decline of IVD health is characterized by the amplification of the catabolic and inflammatory milieu. In agreement with previous work from our laboratory, dual‐miRNA delivery strongly inhibited degradative protein markers. miRNA‐149‐5p in particular has been highlighted as an anti‐inflammatory miRNA, eliciting positive effects both in vitro and in vivo. Specifically, Lang et al. demonstrated that protein expression of IL‐1α, IL‐1β, and IL‐6 were significantly reduced in miRNA‐transfected versus non‐transfected wounds during healing and these effects persisted even under a TNF challenge [55]. Furthermore, in chondrocytes, cells that closely resemble NP cells in both morphology and behavior, miRNA‐149 delivery has been shown to markedly suppress the inflammatory cytokine IL‐6 and degradative enzyme MMP‐3 at both the mRNA and protein levels [56]. A recent study demonstrated that the inhibition of miRNA‐149‐5p in tert‐Butyl hydroperoxide (TBHP)‐challenged NP cells exacerbated the inflammatory response through Nod‐like receptor Pyrin domain containing 1 (NLRP1) activation [57], a critical inflammation mediator, further emphasizing the anti‐catabolic utility of this miRNA. Our findings support the anti‐degradative effects mediated by miRNA‐149 when delivered in tandem with miRNA‐221‐3p inhibitor. In both two‐ and three‐dimensional culture, NP cells exhibited greater resilience when exposed to cytokine challenges. Expression of ADAMTS5, MMP13, IL‐1β, and TNF in NP cells tended to be lower than in BMSCs, regardless of transfection. This may reflect an inherent adaptive advantage, as NP cells secrete low levels of TNF and IL‐1β, even in a healthy state [7, 58]. Serum starvation and cytokine challenges were applied in monolayer, which may have disproportionately affected BMSCs. Redondo‐Castro and colleagues identified a strong up‐regulation of anti‐inflammatory factors; granulocyte‐colony stimulating factors (GCSF), and IL‐1 receptor antagonist in cytokine‐stimulated spheroids in comparison to their monolayer counterparts, suggesting BMSCs are sensitive to challenges based on their culture dimensionality [59]. Conversely, cervical derived NP cells did not elicit an immunogenic response when co‐cultured with peripheral blood mononuclear cells (PBMCs) in monolayer, with no discernible increase in T cell markers or pro‐inflammatory cytokines [60]. These observations suggest that NP cells exhibit a comparatively lower inflammatory response to challenges, particularly in vitro. The suppression of degradative factors was most pronounced in the ex vivo model of mild degeneration, further supporting the implementation of physiologically relevant culture systems. We observed strong catabolic upregulation between our healthy and degenerative controls, validating the model, with both miRNA only and transfected cells strongly suppressing these elevated levels of catabolic expression. Encouragingly, this mirrors our previous work in rat ex vivo organ culture, demonstrating the utility of our dual‐miRNA across species [28, 29]. Interestingly, and contrary to our in vitro experiments, miRNA‐BMSCs exhibited an anti‐catabolic effect in the ex vivo model. BMSCs rely upon microenvironmental factors to exert immunomodulatory effects, which may have been further enhanced when delivered directly to an organ. In vitro, application of conditioned media from BMSCs has been shown to reduce cytokine expression in NP cells [61]. Moreover, MSC encapsulated hydrogels have demonstrated immunomodulatory potential, curbing inflammation and promoting an anti‐inflammatory state through macrophage polarization [62]. However, contrary results have similarly been reported, with Lyon's et al. demonstrating an increased M1 pro‐inflammatory macrophage response in MSC seeded scaffolds in a mouse calvarial defect [63]. Taken together, these findings highlight the importance of the culture model, with the immunomodulatory effect being suppressed in vitro, but enhanced ex vivo. In the context of IVD degeneration, catabolic suppression may represent the gateway to regeneration, but this must occur concurrently with matrix accumulation to restore lost tissue components. Minor increases in sGAGs were identified histologically and biochemically following treatment with miRNA‐NP cells. However, analysis of specific markers was particularly encouraging, with both SOX9 and aggrecan showing significant upregulation. Endogenous SOX9 staining is characteristically low in native disc tissue [64]. In our ex vivo organ culture, the most pronounced SOX9 expression was observed in the dual‐miRNA‐NP groups, indicating a treatment‐associated increase. Notably, dual‐miRNA treated microtissues, also demonstrated significant SOX9 upregulation. Feng et al. delivered NP cells and BMSCs to degenerated rabbit discs, which halted DHI decline, and resulted in improvements to histological scoring compared to the sham control [54]. Consistent with our study, albeit at the gene rather than protein level, aggrecan was significantly upregulated in both treatment groups compared to the degenerated control. Moreover, Chen and colleagues delivered NP cells and NP‐derived mesenchymal stem cells to rabbit IVDs, finding no regenerative differences between the treatment groups, although both showed improvement compared to the degenerated control [34]. In the case of SOX9 in particular, our findings suggest that the elevated SOX9 signal is primarily attributed to the delivery of transfected cells, rather than the reprogramming of resident cells, an important consideration when designing cell‐derived treatments. In our study, the detection of both early and late‐stage ECM production markers following delivery of transfected NP cells is encouraging, indicating a shift towards a regenerative response that may not have had sufficient time to fully mature. We induced organ degeneration by delivering the enzyme chondroitinase ABC, which degrades chondroitin sulphate GAG chains of proteoglycans [65], and is widely recognized for reducing GAG content whilst eliciting a catabolic response in both ex vivo and in vivo models of disc degeneration [53, 66, 67]. As the principal sulphated proteoglycan within the disc ECM, regulation of aggrecan catabolism is key to restoring homeostasis. In bovine cartilage explants, aggrecan has been shown to be rapidly cleaved by cABC into numerous smaller core peptides [68], suggesting these core fragments must be replenished and modified prior to the reformation of the total proteoglycan [69, 70]. The observed increase in SOX9, considered a discogenic master regulator [71, 72], and aggrecan in the NP‐miRNA group suggests the early establishment of a regenerative phenotype. Full glycosylation may not yet have been achieved, likely reflecting a limitation of our ex vivo organ culture model, in which maintaining cultures beyond 4 weeks is currently not feasible. Truncated aggrecan molecules lacking chondroitin sulphate chains are characteristic of aging and degenerative pathology [73, 74], and may potentially explain the disparity observed between sGAG and aggrecan levels. The upregulation of discogenic factors observed in this work is likely to be primarily influenced by the inhibition of miRNA‐221‐3p, the chondroinductive element of our dual‐miRNA therapy. In cartilage regeneration, silencing of miRNA‐221 has been shown to enhance the production of matrix markers. Microtissue cultures of MSCs demonstrated enhanced chondrogenesis and cell proliferation through degradation of Slug protein in cells treated with miRNA‐221 inhibitor [75]. More recently, the loading of collagen hydrogels with miRNA‐221 inhibitor similarly led to significant increases in collagen type II and aggrecan at the mRNA level and sGAGs at the protein level [76]. In the context of disc regeneration, Sheng and colleagues reported a protective effect of miR‐221 inhibition in cartilaginous endplate (CEP) cells, with significant upregulation of aggrecan and collagen type II protein expression [77]. We observed similar upward trends, particularly in more sophisticated culture models, reinforcing the potential of miRNA‐221 silencing as a strategy to counter IVD degeneration. The combined delivery of dual‐miRNAs produced a beneficial effect, enhancing pro‐anabolic and suppressing catabolic signals, reflecting the inherent biological complexity of simultaneously targeting multiple degenerative pathways. This combinatorial approach has been successful in other fields, including the treatment of various cancers [78, 79, 80], where it proved both therapeutically effective and non‐toxic. Similarly, in our study, paired‐miRNA delivery showed no adverse effects on cell viability in ex vivo organ culture. We have previously shown that dual delivery of miRNA‐149‐5p + miRNA‐221‐3p inhibitor outperformed single miRNA treatment, both in terms of ECM molecule accumulation and in the suppression of matrix degrading enzymes [28]. A similar effect was reported by Castaño et al., whereby combined delivery of miRNA‐210 + miRNA‐16 inhibitor loaded scaffolds significantly enhanced the osteogenic potential of MSCs when compared to individual‐miRNA transfection [81]. The synergistic effect of dual‐miRNA therapy likely arises from the modulation of complementary signaling pathways. For example, miRNA‐149 is known to target molecules involved in the NF‐κB pathway [56, 82], while miRNA‐221 has been shown to modulate members of the TIMP signaling family [83, 84]. These distinct targets may modulate both anabolic and catabolic factors, however further investigations are required to elucidate the multitude of pathways potentially involved. miRNAs represent an attractive therapeutic strategy due to their ease of complexation and delivery. In terms of translatability, miRNAs have been reported to have a half‐life of approximately 119 h (~5 days) [85], with downstream effects on protein expression typically emerging within approximately 3 days [86]. The intracellular concentration of miRNA is also influenced by cell proliferation rate. Consequently, in slowly dividing cell types such as NP cells [87], miRNA activity may be prolonged, particularly in compromised or degenerated tissues. This further supports their suitability for delivery to the NP region. However, the precise kinetics of miRNA activity within the IVD have not been characterized in vitro or in vivo, representing an important area for future investigation. A critical aspect of developing cell‐based therapies for IVD regeneration is the comprehensive evaluation of their effects on resident NP cells. In this study, suppression of catabolic enzymes and pro‐inflammatory cytokines was observed in the ex vivo organ culture model, supporting the role of dual‐miRNA treatment in establishing a protective niche within mildly degenerated IVDs. This effect was most pronounced in the dual‐miRNA group, suggesting that the miRNA combination modulates endogenous disc cells, consistent with our previous work [28, 29]. In contrast, both cell treatments exhibited less pronounced reductions in catabolic activity, potentially reflecting a weaker response from, and reduced reprogramming of resident NP cells. The most robust regenerative responses, indicated by increased SOX9 and aggrecan expression, were observed in the miRNA‐NP cell group. However, as staining was predominantly localized around the delivered cells and the ex vivo culture duration was relatively short, the broader impact on resident cells remains difficult to fully assess. This is particularly relevant given the longer timeframes typically required for substantial aggrecan deposition and sGAG accumulation. Cell and biomaterial delivery to the IVD is typically confined to the injection site, resulting in spatially localized effects [88, 89]. Coupled with the dense ECM of even mildly degenerated NP tissue [90], this likely limits full dispersion of delivered cells throughout the disc. In this context, physiological dynamic mechanical loading may play an important role in enhancing cell distribution over time, as cyclic loading has been shown to promote fluid flow and solute transport within the disc [49, 91]. However, such effects are not captured in the shorter culture times of ex vivo culture systems. Consequently, therapeutic efficacy may depend on sustained paracrine signaling over extended durations, which may not be adequately captured within the experimental timeframe. These limitations should be explicitly considered in the development and evaluation of cell‐based therapies, particularly when interpreting short‐term outcomes. Although a range of culture models were utilized here, certain limitations should be acknowledged. While static loading within the human physiological range [49] was applied in our bioreactor system, more sophisticated models could be utilized to better simulate the diurnal and complex loading pattern experienced through daily life. Recent studies using such dynamic systems have demonstrated modulation of the microenvironment, whereby anabolic genes were downregulated and catabolic genes upregulated following a regime of dynamic and static loading, perpetuating degeneration even in the absence of enzymatic insult [92]. Furthermore, dynamic loading may increase the diffusivity of treatments throughout the NP region [93]. However, our simplified model incorporating static loading and cABC enzyme delivery successfully induced degeneration while maintaining viability comparable to that observed in more complex dynamic systems over the same culture period [94]. Furthermore, PKH26 positive staining confirmed the presence of injected cells within the NP region, demonstrating their retention within the disc and supporting the utility of this approach as an accessible model. In this study we employed the CPP FLR to transfect our dual‐miRNA cargo to NP and BMSCs. However, alternative transfection methods, such as electroporation have also been utilized in the IVD field to deliver plasmids to NP cells [95]. These methods may be more suitable for the delivery of larger cargoes. While a direct injection approach was used here to deliver dual‐miRNA transfected cell populations in an ex vivo organ culture model, other strategies have leveraged the intrinsic homing capacity of BMSCs, enabling migration across the CEP into the NP with promising outcomes [96, 97]. Whilst aligning with the 3R principles [98], organ culture inherently limits the duration of experiments compared to in vivo studies, typically to a maximum of 4 weeks [99, 100], which restricts assessment of longer term effects. Regeneration of the in vivo GAG matrix to a healthy level has been predicted to require 12 months in a goat following delivery of 5.5 × 106 cells [51], which is not feasible when using organ culture models. Nonetheless, organ culture models provide valuable insights into the early stages of treatment, and the use of goat discs offers strong translational relevance for human IVD therapies due to their anatomical and matrix compositional similarity to human discs [67]. While direct comparative studies between goat and human NP cells remain limited, existing evidence suggests that goat NP cells exhibit higher metabolic activity and collagen synthesis rates, whereas GAG production appears comparable between species [51]. This difference may lead to an overestimation of regenerative outcomes in preclinical models. However, in the absence of direct head‐to‐head comparisons, such interpretations remain speculative. Furthermore, to our knowledge, no studies have directly assessed differences in the immunomodulatory capacity of goat and human NP cells. Nonetheless, given their anatomical, compositional and microenvironmental similarities [37], it could be hypothesized that human NP cells would exhibit broadly similar behavior, supporting the relevance of the goat model as a preclinical and potentially translatable platform. One aspect of degeneration not addressed in this study but warrants further investigation is nociceptive pain. Elevated levels of markers such as β‐nerve growth factor (β‐NGF) and brain derived neurotrophic factor (BDNF) are associated with increasing degeneration severity and substantially affect quality of life in patients with degenerated discs. In a clinical study of platelet lysate delivery to the IVD, no measurable improvement in disc height was observed, yet both the visual analogue scale (VAS) and Roland‐Morris disability questionnaire showed significant pain reduction [101]. This highlights that modulation of pain perception is a key factor in rehabilitation, and should be considered alongside regenerative strategies. TNF has been shown in vitro to upregulate BDNF and β‐NGF expression [102, 103, 104], suggesting that our TNF silencing approach may potentially impact pain perception, although this remains to be explored in future studies. In this study, we examined the effect of dual delivery of the anti‐inflammatory miRNA‐149‐5p mimic and the pro‐matrix miRNA‐221‐3p inhibitor with the cell penetrating peptide FLR to stimulate two candidate cell populations, NP and BMSCs to determine the most efficacious for delivery to a mildly degenerated IVD. Our findings highlight the potential of a miRNA‐pair transfected cell delivery approach to exert both anti‐catabolic and anti‐inflammatory effects in vitro and in ex vivo goat organ culture models, while demonstrating positive upregulation of matrix promoting markers. Stimulation of both NP and BMSCs with FLR‐miRNA‐149‐5p mimic + 221‐3p inhibitor elicited positive effects, primarily through the suppression of key catabolic factors, ADAMTS5, MMP13, and the pro‐inflammatory cytokines TNF and IL‐1β. Notably, delivery of dual‐miRNA treated NP cells to goat organs demonstrated particularly promising improvements in matrix‐related markers, SOX9 and aggrecan, suggesting the potential of this strategy as a cell therapy for IVD degeneration when combined with the miRNA‐pair. Furthermore, these results suggest that NP cells are more suited than BMSCs for delivery to the IVD post‐miRNA transfection, as they demonstrated combined anti‐catabolic and pro‐restorative activity. Collectively, these findings support further longer‐term investigations to assess the feasibility and efficacy of a dual‐miRNA NP cell‐based delivery strategy to target and attenuate the degradative cascade promoting IVD degeneration and the development of LBP. Conclusion The present study demonstrated that delivery of the therapeutic miRNA pair, miRNA‐149‐5p mimic and miRNA‐221‐3p inhibitor, suppressed key catabolic enzymes (ADAMTS5 and MMP13) alongside inflammatory mediators (TNF and IL‐1β) confirming the establishment of a robust anti‐catabolic microenvironment in a goat ex vivo organ culture model. Delivery of NP and BMSCs that were transfected with this miRNA pair produced similar effects, albeit to a slightly lesser extent. Importantly, only delivery of dual‐miRNA stimulated NP cells into organ culture elicited a significant anabolic response, as evidenced by upregulation of the matrix markers SOX9 and aggrecan while this effect was not replicated by miRNA‐BMSC delivery. Overall, this work suggests NP cells as a promising regenerative cell source when delivered in combination with FLR‐miRNA‐149‐5p and 221‐3p inhibitor for the treatment of IVD degeneration. Author Contributions Tara Ní Néill: conceptualisation, methodology, software, data curation, investigation, validation, formal analysis, visualisation, writing – original draft, writing – review and editing. Niamh Wilson: methodology, data curation, formal analysis, writing – review and editing. Jake McDonnell: methodology, data curation, formal analysis, writing – review and editing. Fergal J. O'Brien: writing – review and editing. Caroline M. Curtin: writing – review and editing. James E. Dixon: resources, writing – review and editing. Pieter A. J. Brama: investigation, resources, writing – review and editing. Conor T. Buckley: conceptualization, methodology, software, data curation, Investigation, validation, formal analysis, supervision, funding acquisition, visualisation, project administration, resources, writing – review and editing. Funding This work was supported by the European Research Council, ERC‐2019‐CoG‐864104; INTEGRATE. Conflicts of Interest Conor T. Buckley is an Editorial Board member of JOR Spine and co‐author of this article. He was excluded from editorial decision‐making related to the acceptance of this article for publication in the journal. The remaining authors declare no conflicts of interest. Supporting information Figure S1: Representative images of tri‐potentiality assessment of bone marrow‐derived stem cells. (A) Chondrogenic (chondro.) supplementation of pellet cultures demonstrated pellets with more structural integrity, alongside enhanced collagen and sulphated glycosaminoglycan (sGAG) production, as determined by Picrosirius red and Alcian Blue staining. (B) Following supplementation with adipogenic (adipo.) supplements, increased Oil‐Red O staining was achieved compared to untreated groups. (C) Supplementation with osteogenic (osteo.) media supplements resulted in increased Alizarin red staining representing increased calcium deposition. N = 3 biological replicates. All scale bars 200 μm. Figure S2: Goat ex vivo organ culture degeneration pilot to establish cABC dosage, following 14 days of cABC degeneration. (A) Representative Alcian blue staining for sulphated glycosaminoglycan (sGAG) deposition. Protein quantification normalized to healthy control and representative immunofluorescence images for catabolic proteins (B) MMP13, (C) ADAMTS5. Histology scale bar 3 mm, immunofluorescence scale bar. Statistical test is one‐way ANOVA with Tukey's post hoc analysis, n = 4 technical replicates. Figure S3: Functional assessment of scramble miRNA transfection on monolayer and 3D culture in goat nucleus pulposus (NP) and bone marrow‐derived stem cells (BMSCs). (A) No changes were identified biochemically following scramble miRNA transfection, with DNA, sulphated glycosaminoglycan (sGAG), and collagen production assessed. (B) Biochemical analysis of three‐dimensional culture likewise demonstrated no changes. (C) Representative images of histological staining for hematoxylin and eosin (H&E), Alcian blue (AB), and Picrosirius red (PR). Statistical test is one‐way ANOVA with Tukey's post hoc analysis. ns = non‐significant. N = 3 independent biological replicates. Scale bars 50 μm. Figure S4: Assessment of miRNA delivery efficiency, culture viability, and cell retention in ex vivo organ culture of goat intervertebral discs. (A) Delivery of a fluorescently tagged miRNA demonstrated efficient uptake 3 days post‐transfection. (B) Live/dead staining of sections of nucleus pulposus (NP) tissue following 28 days of culture. (C) PKH‐26 staining of cells delivered demonstrated retention during the 28‐day culture period. Statistical test is one‐way ANOVA with Tukey's post hoc analysis. * p < 0.05, ** p < 0.01, ns = non‐significant. Scale bars for A and B are 100 μm. Scale bars for whole organ view of C are 1 mm and 200 μm for inset. Figure S5: Supplementary biochemical analysis of goat intervertebral discs following ex vivo organ culture. (A) Tissue hydration, (B) DNA content, (C) sulphated glycosaminoglycans (sGAGs), (D) collagen content, and (E) sGAG to collagen ratio. N = 3 independent biological replicates. Acknowledgments This project has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement ERC‐2019‐CoG‐864104; INTEGRATE). Data Availability Statement The data that support the findings of this study are openly available in Zenodo at https://zenodo.org/uploads/18505003. References Associated Data Supplementary Materials Figure S1: Representative images of tri‐potentiality assessment of bone marrow‐derived stem cells. (A) Chondrogenic (chondro.) supplementation of pellet cultures demonstrated pellets with more structural integrity, alongside enhanced collagen and sulphated glycosaminoglycan (sGAG) production, as determined by Picrosirius red and Alcian Blue staining. (B) Following supplementation with adipogenic (adipo.) supplements, increased Oil‐Red O staining was achieved compared to untreated groups. (C) Supplementation with osteogenic (osteo.) media supplements resulted in increased Alizarin red staining representing increased calcium deposition. N = 3 biological replicates. All scale bars 200 μm. Figure S2: Goat ex vivo organ culture degeneration pilot to establish cABC dosage, following 14 days of cABC degeneration. (A) Representative Alcian blue staining for sulphated glycosaminoglycan (sGAG) deposition. Protein quantification normalized to healthy control and representative immunofluorescence images for catabolic proteins (B) MMP13, (C) ADAMTS5. Histology scale bar 3 mm, immunofluorescence scale bar. Statistical test is one‐way ANOVA with Tukey's post hoc analysis, n = 4 technical replicates. Figure S3: Functional assessment of scramble miRNA transfection on monolayer and 3D culture in goat nucleus pulposus (NP) and bone marrow‐derived stem cells (BMSCs). (A) No changes were identified biochemically following scramble miRNA transfection, with DNA, sulphated glycosaminoglycan (sGAG), and collagen production assessed. (B) Biochemical analysis of three‐dimensional culture likewise demonstrated no changes. (C) Representative images of histological staining for hematoxylin and eosin (H&E), Alcian blue (AB), and Picrosirius red (PR). Statistical test is one‐way ANOVA with Tukey's post hoc analysis. ns = non‐significant. N = 3 independent biological replicates. Scale bars 50 μm. Figure S4: Assessment of miRNA delivery efficiency, culture viability, and cell retention in ex vivo organ culture of goat intervertebral discs. (A) Delivery of a fluorescently tagged miRNA demonstrated efficient uptake 3 days post‐transfection. (B) Live/dead staining of sections of nucleus pulposus (NP) tissue following 28 days of culture. (C) PKH‐26 staining of cells delivered demonstrated retention during the 28‐day culture period. Statistical test is one‐way ANOVA with Tukey's post hoc analysis. * p < 0.05, ** p < 0.01, ns = non‐significant. Scale bars for A and B are 100 μm. Scale bars for whole organ view of C are 1 mm and 200 μm for inset. Figure S5: Supplementary biochemical analysis of goat intervertebral discs following ex vivo organ culture. (A) Tissue hydration, (B) DNA content, (C) sulphated glycosaminoglycans (sGAGs), (D) collagen content, and (E) sGAG to collagen ratio. N = 3 independent biological replicates. Data Availability Statement The data that support the findings of this study are openly available in Zenodo at https://zenodo.org/uploads/18505003.