Introduction

Low back pain (LBP) is the main cause of disability in the world [1] and is primarily triggered by intervertebral disc degeneration (IDD), a process characterized by catabolic events prompting extracellular matrix (ECM) degradation and cell loss, eventually altering disc morphology and biomechanics. To date, regenerative cell-based therapies are limited by poor post-transplantation viability, production scalability, and substantial donor variability [2]. Thus, cell-free therapeutics have been explored to overcome these hurdles. As anabolic effects exerted by transplanted cells are seemingly derived from their secretome, direct utilization of secreted factors may offer remarkable advantages over intradiscal cell delivery [3, 4]. In this context, extracellular vesicles (EVs) hold promise due to their role in intercellular communication and putative regenerative effects. With regard to IDD, mesenchymal stromal cell (MSC)-derived EVs have been demonstrated to reduce disc cell apoptosis, ECM degradation, tissue inflammation, and oxidative stress, both in vitro and in vivo [5].

Despite their notable chondrogenic potential, bone marrow-derived MSCs (BM-MSCs) exhibit significant differences from nucleus pulposus cell (NPC) progenitors [6], characterized by the expression of Tie2 receptors. Compared to Tie2−-NPCs, these cells have demonstrated higher self-renewal, proliferation, viability, and multidifferentiation potential, as well as increased ECM production [7, 8]. Recent studies have also demonstrated that NPCs expressing Tie2 present a superior chondrogenic differentiation capacity compared to BM-MSCs [9]. However, the percentage of Tie2+-NPCs is progressively exhausted with aging and advancing IDD, possibly explaining the limited ability of the IVD to perform self-repair [10].

Considering their regenerative capacity and recent advancements in culture methods to preserve Tie2 expression [11, 12], we sought to assess whether EVs secreted by this subpopulation might provide a promising cell-free therapy against IDD.

This study aimed to characterize and apply EVs derived from human NPCs cultured in standard and Tie2-enhancing conditions, to evaluate their biological effects in vitro, and to explore their regenerative potential in vivo.

Materials and methods

This study has been approved by the Institutional Review Board of Tokai University School of Medicine (Isehara, Kanagawa) under the approval number n. 221012. Informed consent for the collection and use of surgical waste products for research purposes was obtained from every patient. All animal experiments have been conducted according to the ARRIVE guidelines and in respect of the 3R principle. The detailed methodology is available as Supplementary Information.

Cell isolation and culture

Human IVD tissues were collected from surgical specimens at Tokai University Hospital (Isehara, Japan; Table 1). NPCs derived from young (mean age: 28.5 years) and healthy donors were cultured by standard non-Tie2-optimized (NPCSTD) or Tie2-enhancing culture strategies (NPCTie2; Fig. 1). Specimens were washed with phosphate-buffered saline (PBS), minced, and the NP tissue was carefully separated as previously described [7]. For NPCSTD, the tissue was digested in 0.25% trypsin–EDTA (Thermo Fisher, USA; 30 min, 37 °C), and then with 0.25 mg/mL collagenase P (Roche, Switzerland), Dulbecco’s Modified Eagle Medium-High Glucose (DMEM-HG; Wako, Japan), and 10% fetal bovine serum (FBS, Sigma-Aldrich, USA; 4 h, 37 °C). The yielded suspension was filtered through an 80-µm cell strainer and seeded on standard 100-mm dishes (~ 1 × 105 cells/dish). Following the guidelines of the Orthopedic Research Society (ORS) [13], NPCsSTD were cultured in DMEM-HG + 10% FBS + 1% penicillin/streptomycin (Gibco, USA) + 50 µg/mL L-ascorbic acid (Sigma-Aldrich) at 37 °C in 5% CO2 and 5% O2. The same protocol was adopted for NPCs from degenerative IVDs (dNPCs). After washing and mincing, the tissues undergoing the NPCsTie2+ culture approach were directly cultured in polystyrene six-well plates (IWAKI, Japan) with a commercially developed NPC-optimized, blended medium (MEMα:32%, DMEM:48%, FBS:20%; TUNZ Pharma Co., Ltd., Japan). Tissue fragments were cultured at 37 °C in 5% CO2 and 5% O2 for 14 days without media replenishment. Subsequently, cultured tissue fragments were digested as described above, and derived cells were seeded in monolayer at a density of 30,000 cells per 100-mm dish (545.5 cells/cm2) using the same medium [11]. A commercial cell line of human BM-MSCs (Poietics™, Lonza, USA) was cultured in DMEM-HG + 10% FBS + 1% penicillin/streptomycin at 37 °C in 5% CO2 and 21% O2 (Fig. 1).

Table 1 Summary of characteristics of the donor IVD samples
Full size table
Fig. 1
figure 1

Schematic design of the in vitro study. NPCs were isolated from young and healthy donors and cultured in monolayer according to either a standard (a) or a Tie2-enhancing (b) culture protocol. Human BM-MSCs were also cultured following standard methods. After culture expansion (c), conditioned media were collected (d) and processed for EV isolation (e) and subsequent characterization (f). Subsequently, degenerative NPCs were extracted from surgical specimens (g) and treated in vitro (h) with either EVs extracted from NPCs cultured in standard conditions (a) or in Tie2-enhancing conditions (b). Eventually, cell proliferation and senescence were assessed. Created with BioRender.com. Abbreviations: BM-MSCs bone marrow-derived stromal cells, EV extracellular vesicle, NPCs nucleus pulposus cells, TEM transmission electron microscopy, WB Western blot

Full size image

EV isolation

FBS was ultracentrifuged (110,000 g for 17 h) and filtered through a 0.22-μm membrane to obtain bovine EV-free FBS [14]. NPCsSTD, NPCsTie2+, and BM-MSCs at 50–60% confluency were cultured with EV-free media for two days after which conditioned media were collected to isolate EVs as previously described [14]. Briefly, media were centrifuged at 1500 g (15 min), 12,000 g (30 min), and filtered with a 0.22-μm membrane to remove cell debris and larger vesicles. Eventually, the media were ultracentrifuged at 110,000 g (70 min) at 4 °C. The EV pellets were resuspended in PBS and ultracentrifuged again at 110,000 g (70 min) at 4 °C. The resulting EV products were used directly or stored at − 80 °C.

EV characterization

EV characterization was performed according to the International Society for Extracellular Vesicles guidelines [15], through transmission electron microscopy (TEM), single-particle analysis (qNano, Izon Science, New Zealand) [16] and assessment of CD63, CD9, TSG101, and lamin A/C expression by Western blot, comparing NPCs, BM-MSCs, and their EVs. EVs were quantified using the micro-bicinchoninic acid assay (Thermo Fisher).

EV uptake

NPCs were stained with Hoechst 33342 (Lonza), and isolated EVs were incubated with PKH26 (Sigma‐Aldrich) for 5 min. Labeled EVs were applied onto NPCs for 180 min, and uptake was observed through confocal laser scanning microscopy (LSM700, Carl Zeiss, Germany).

EV effect in vitro

dNPCs were cultured using standard media ± 10 ng/mL IL-1β (PeproTech, USA) to further promote catabolism [17]. Both conditions were treated with NPCSTD-EVs or NPCTie2+-EVs at 25, 50, 75, and 100 μg/mL [18]. Dose–effect at days three and seven of culture was determined through the CCK-8 assay (Dojindo, USA). Cell proliferation was calculated as percent change compared to the DMEM-only group.

Cell senescence

dNPCs were seeded into six-well plates (2.5 × 105 cells/well) and treated for 72 h with or without IL-1β and/or 50 µg/mL EVs as explained above. Subsequently, cells were washed with PBS, fixed, and incubated overnight with β-galactosidase staining (Cell Signaling, USA). Cells were observed under a phase-contrast light microscope, capturing six random fields per well at 10 × magnification, to determine the proportion of positive cells to total cells in a blinded manner.

Western blot

EVs and intact cells were lysed in RIPA buffer (Thermo Fisher). Protein lysates were resolved by SDS-PAGE using NuPAGE 4–12% Bis–Tris Gel (Invitrogen), MES running buffer, and blotted onto PVDF membranes. Membranes were incubated with primary antibodies (1:1000) at 4 °C overnight and secondary antibodies (1:2000) at room temperature for 1 h. The primary antibodies used were: CD9 (sc13118, Santa Cruz Biotechnology, USA), CD63 (BD556019, BD Biosciences, USA), TSG101 (ab30871, Abcam, UK), lamin A/C (MAB3211, Sigma-Aldrich), and GAPDH (SAB2108668, Sigma-Aldrich). Immunoreactive bands were detected using an Amersham Healthcare ECL Prime Western Blotting Detection Reagent (Cytiva, Japan). Protein signals were quantified by scanning densitometry (ATTO, Japan).

Assessment of EVs on in vivo IDD

Sixteen Sprague–Dawley rats (10 to 12 weeks old, MIZUSETSU, Japan) were randomly assigned to sham (n = 4), NPCSTD-EV (n = 4), NPCTie2+-EV (n = 4), or BM-MSC-EV (n = 4) groups (Fig. 2). Males were chosen due to reduced behavioral variability [19]. Following acclimation and baseline data acquisition, IDD was induced by annular puncture from Co5/6 to Co7/8 as previously described [20]. Co4/5 and Co8/9 functioned as healthy controls. Discs were surgically exposed, and approximately 10 μL of NP tissue was aspirated with a 21G needle. Subsequently, 2 μL of either PBS-only (sham group), NPCSTD-EVs, NPCTie2+-EVs, or BM-MSC-EVs was injected through a 27G needle–Hamilton syringe combination. Approximately 1.5 × 106 EVs were injected per level [21].

Fig. 2
figure 2

Schematic design of the in vivo study. Sixteen Sprague–Dawley rats were randomly divided into four groups (n = 4 per group), underwent annular puncture of three contiguous tail intervertebral discs, and subsequent injection of either PBS, NPCSTD-EVs, NPCTie2+-EVs, or BM-MSC-EVs. Animals underwent pain behavioral assessments (Von Frey test and open field test) every other week, radiographic evaluation of the DHI, and analysis of serum biomarkers at baseline, weeks 4, and 12. Following sacrifice, gross anatomy and histological analyses were carried out to evaluate disc structural changes. Abbreviations: BM-MSC-EVs extracellular vesicles isolated from bone marrow-derived mesenchymal stromal cells, DHI disc height index, NPCSTD-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in standard conditions, NPCTie2+-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in Tie2-enhancing conditions, PBS phosphate-buffered saline. Created with BioRender.com

Full size image

Behavioral assessment

Behavioral tests were performed at baseline and biweekly post-transplantation as previously described [22, 23]. Animals were placed in a 60 × 60 cm open-field box and recorded for 10 min from a bird’s eye view. The day before the test, animals were acclimatized in the enclosure for 10 min. Room lighting and time of day were kept consistent. EthoVision XT (Noldus Information Technology, The Netherlands) was used to analyze 18 parameters correlated with LBP as per previous studies [23].

The Von Frey test was performed using an esthesiometer (IITC, 2391). Three animals at a time were allowed to acclimate for 15 min in clear acrylic chambers placed on a metal grid. A maximum force of 28 g was applied ramping up over 10 s and kept for a maximum of 40 s. Withdraw threshold and time were assessed sequentially and alternatively by applying the probe on hind paws, tail base, and surgical site. This process was repeated at least thrice, and the averages were calculated. Investigators were blind to group allocation during the behavioral tests.

Serum biomarkers

Peripheral venous blood samples were obtained at baseline, 4, and 12 weeks post-operatively. Serum was isolated by centrifugation, frozen at − 80 °C, and used for IL-6 quantification using a rat-specific ELISA kit (RayBiotech, USA) as per the manufacturer’s instructions.

Radiographic assessment

Radiographic assessment was conducted at baseline and 4, 8, and 12 weeks post-operatively. Images were obtained from the coccygeal region with the animals in the supine position using a fluoroscopic imaging intensifier (DHF-105CX, Hitachi, Japan) under 2.5% isoflurane inhalation. In a blinded manner, disc height index (DHI) was calculated and normalized to pre-transplantation DHI [24].

Macroscopic and histological evaluation

At 12 weeks, the animals were sacrificed, tails were dissected, and discs were explanted. Samples were fixed in 10% formalin and decalcified using Wako solution A (Wako). Functional spine units were sectioned and assessed through Thompson grading system [25] by two blinded investigators. Subsequently, tissue sections were prepared for histology with hematoxylin/eosin, Safranin-O/Fast-Green, and Picrosirius-red/Aclian blue staining. Sections were blindly scored based on the ORS Spine rat histological grading scheme [26] by two investigators.

Statistical analysis

All quantitative data are expressed as means ± standard deviation. Data normal distribution was determined with the Wilk–Shapiro test. Statistical analysis was performed using one-way or two-way ANOVA. The Kruskal–Wallis test was performed to analyze non-normal data. Statistical significance was set as p < 0.05. Formal analysis was performed using Prism 10 (GraphPad, USA).

Results

EV isolation and characterization

TEM showed the typical EV oval cup morphology (Fig. 3a). NPCSTD-EVs, NPCTie2+-EVs, and BM-MSC-EVs exhibited an average diameter of 106.1 ± 16.3, 106.0 ± 18.2, and 87.0 ± 12.8 nm, respectively (Fig. 3b). The expression of EV markers CD63, CD9, and TSG101 was higher in EV samples compared to cell lysates, and lower for lamin A/C (Fig. 3c). PKH26-labeling showed EV uptake within NPCs (Fig. 3d). Collectively, these results demonstrated the isolation of small EVs (< 200 nm) with the characteristics of exosomes.

Fig. 3
figure 3

Characterization of isolated EVs. a TEM images demonstrating the typical cup shape of purified EVs (scale bar = 200 nm). b EV size distribution showed that the highest abundance of nanoparticles was below 200 nm, thus designating them as exosomes. c Western blot of cell lysates and their EVs showing the expression of exosomal markers CD63, CD9, and TSG101 and the lack of the nuclear protein lamin A/C. d Observation of PKH26-labeled EVs (red) under confocal microscopy demonstrated the uptake of nanoparticles within dNPCs. dNPC nuclei were stained with Hoechst 33342 (blue). Scale bars = 50 μm (left and center); 20 μm (right). Representative images are shown. N = 6. Abbreviations: dNPCs degenerative nucleus pulposus cells, EVs extracellular vesicles, TEM transmission electron microscopy

Full size image

NPCSTD-EVs and NPCTie2+-EVs increased dNPC proliferation and reduced senescence

Dose–response evaluation of NPCSTD-EVs and NPCTie2+-EVs at day three showed that NPCTie2+-EVs significantly increased cell proliferation compared to NPCSTD-EVs at all concentrations (p < 0.05). Interestingly, the addition of IL-1β to NPCSTD-EVs (p < 0.05) and NPCTie2+-EVs (p < 0.01) further enhanced cell proliferation compared to NPCSTD-EVs alone (Fig. 4a). Similar results were observed at day seven (Fig. 4b). dNPCs treated with IL-1β showed an elongated, fibroblastic-like conformation, while NPCSTD-EV- and NPCTie2+-EV-treated cells exhibited a preserved morphology, even in the presence of IL-1β (Fig. 4c). Considering the lack of significant differences among tested EV concentrations, the intermediate value (50 μg/mL) was utilized in subsequent experiments. β-galactosidase staining (Fig. 5) demonstrated that NPCSTD-EVs and NPCTie2+-EVs significantly decreased senescent cell ratio compared to dNPCs cultured with IL-1β (p < 0.05). Furthermore, NPCTie2+-EVs prevented IL-1β-induced senescence (p < 0.05).

Fig. 4
figure 4

Cell proliferation after treatment with NPCSTD-EVs and NPCTie2+-EVs with and without IL-1β detected by the CCK-8 assay at three (a) and seven days (b). Cell proliferation was calculated as percent absorbance change compared to the control group. Horizontal dotted lines correspond to a 100% viability compared to the control. *p < 0.05; **p < 0.01; ***p < 0.001. N = 5. c Representative images of dNPCs at three days after culture. Despite a slight increase in proliferation, cells incubated with IL-1β showed an altered morphology as a possible consequence of a stress response. Conversely, dNPCs treated with NPCSTD-EVs and NPCTie2+-EVs displayed a normal cell shape, even in the presence of IL-1β. Scale bars = 100 μm. Abbreviations: dNPCs, degenerative nucleus pulposus cells; IL-1β interleukin-1β, NPCSTD-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in standard conditions, NPCTie2+-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in Tie2-enhancing conditions

Full size image
Fig. 5
figure 5

dNPC senescence detected by SA-β-gal staining. NPCSTD-EVs and NPCTie2+-EVs significantly decreased the percentage of senescent cells compared to dNPCs cultured with IL-1β. Moreover, NPCTie2+-EVs prevented the increase in cell senescence despite the presence of IL-1β. *p < 0.05. N = 3. b Representative images of SA-β-gal-stained cells showing a reduced ratio of senescent cells in NPC-EV-treated samples. Blue cells: senescent cells. 10 × magnification. Scale bars = 100 μm. Abbreviations: dNPCs degenerative nucleus pulposus cells, IL-1β interleukin-1β, NPCSTD-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in standard conditions, NPCTie2+-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in Tie2-enhancing conditions, SA-β-gal senescence-associated β-galactosidase

Full size image

NPC-EVs decreased mechanical hypersensitivity and modulated pain-like behaviors in vivo

IDD induction and transplantation procedures were successfully performed without complications. Animals displayed stable weight without discernible differences among cohorts (Supplementary Fig. 1). The Von Frey test demonstrated that NPCSTD-EVs and NPCTie2+-EVs significantly lowered the nocifensive response threshold compared to the sham and BM-MSC-EV groups at the surgical site (Fig. 6a) and tail base (Fig. 6B; p < 0.05). Interestingly, NPCTie2+-EV-treated rats consistently exhibited the highest pain threshold and time until withdrawal (Supplementary Fig. 2a). No statistically significant differences were found following sham or BM-MSC-EV injections. No mechanical hypersensitivity was demonstrated for hind paws in any tested groups (Supplementary Fig. 2b).

Fig. 6
figure 6

Results of the Von Frey test to evaluate mechanical pain hypersensitivity at the surgical site (a) and tail base (b). Treatment with NPCSTD-EVs and NPCTie2+-EVs resulted in an increased pain threshold corresponding to a higher average force needed to elicit a withdrawal response compared to animals injected with BM-MSC-EVs or PBS. Horizontal dotted line: force threshold (26 g) able to induce a significant nocifensive response in rats. *p < 0.05; **p < 0.01; ***p < 0.001. N = 4 animals per group. Abbreviations: BM-MSC-EVs extracellular vesicles isolated from bone marrow-derived mesenchymal stromal cells, NPCSTD-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in standard conditions, NPCTie2+-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in Tie2-enhancing conditions, PBS phosphate-buffered saline

Full size image

The open field test, aimed to assess changes in spontaneous pain-like behavior, showed no evident between-group differences. Rats in the NPCSTD-EV and NPCTie2+-EV groups showed higher mean velocity, total distance, movement duration, behavioral probability of walking and rearing unsupported, walking frequency, and rearing unsupported duration and frequency, albeit failing to reach statistical significance (Fig. 7a–r). Although not showing significant differences, IL-6 serum levels displayed a more substantial decrease in NPCTie2+-EV-injected animals compared to the other groups (Fig. 7S).

Fig. 7
figure 7

Rat pain behaviors assessed by the open field test at baseline, 2-, 4-, 6-, 8-, 10-, and 12-weeks post-transplantation. Representative behavioral parameters evaluated included mean velocity (a), total distance (b), duration of movement (c), duration of not moving (d), probability of walking (e), probability of unsupported rearing (f), probability of supported rearing (g), walking duration (h), walking frequency (i), twitching frequency (j), sniffing duration (k), sniffing frequency (l), duration of unsupported rearing (m), frequency of unsupported rearing (n), duration of supported rearing (o), frequency of supported rearing (p), grooming duration (q), grooming frequency (r). (s) Serum concentration of IL-6 in treated animals at baseline and 1- and 3-months post-transplantation. N = 4 animals per group. Abbreviations: IL interleukin

Full size image

NPCSTD-EVs and NPCTie2+-EVs maintained DHI

The sham group displayed a continuous DHI decrease, thus demonstrating successful IDD induction. BM-MSC-EV injection resulted in DHI reduction during the first eight weeks, with a slight non-significant increase at week 12. Interestingly, DHI in both NPC-EV groups did not display notable changes from baseline and was significantly higher than the sham and BM-MSC-EV groups (p < 0.05, Fig. 8).

Fig. 8
figure 8

Radiographic assessment of DHI changes at 4-, 8-, and 12- weeks post-transplantation compared to baseline. a DHI changes have been calculated as percent variation compared to baseline DHI. b Representative radiograms of rat tails at 12 weeks following treatment. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. N = 12 discs per group. Abbreviations: DHI disc height index

Full size image

NPCSTD-EVs and NPCTie2+-EVs preserved disc morphology

Gross IVD evaluation demonstrated significant degenerative alterations in the sham and BM-MSC-EV groups, while NPCSTD-EV and NPCTie2+-EV treatment preserved annulus fibrosus and nucleus pulposus morphology (Fig. 9a). Thompson scores for NPCSTD-EV- and NPCTie2+-EV-treated discs were significantly lower than in the sham and BM-MSC-EV groups (p < 0.01; Fig. 9b). Histological assessment confirmed these observations (Fig. 9c), with similarly lower scores for NPCSTD-EV and NPCTie2+-EV groups compared to both sham and BM-MSC-EV groups, although significant differences were obtained only compared to the former (p < 0.05; Fig. 9d).

Fig. 9
figure 9

Assessment of morphological changes following sham and EV injection. a Gross anatomical evaluation of dissected IVDs showed signs of advanced degenerative changes in the sham group and a notable reduction of disc height in the BM-MSC-EV group, with no clear difference between the AF and NP areas. Conversely, IVDs treated with NPCSTD-EVs and, especially, NPCTie2+-EVs displayed a preservation of the original IVD morphology with maintained disc height and demarcation of NP and AF. b The semiquantitative Thompson score of disc degenerative changes was significantly lower in IVDs injected with NPCSTD-EVs and NPCTie2+-EVs. Scale bars = 500 μm. c Histological sections of IVDs following hematoxylin–eosin (upper row), Safranin-O/Fast-Green (middle row), and Picrosirius-red/Alcian blue staining (lower row). NPC-EVs and, especially, NPCTie2+-EVs better preserved the NP morphology, proteoglycan content, and ECM structure compared to the other groups. Scale bars = 1 mm. d ORS Spine rat histological grading scores of sectioned IVDs. Representative images are shown. *p < 0.05; **p < 0.01; ***p < 0.001. N = 12 discs per group. Abbreviations: AF annulus fibrosus, BM-MSC-EVs extracellular vesicles isolated from bone marrow-derived mesenchymal stromal cells, ECM  extracellular matrix, EV extracellular vesicle, IVD intervertebral disc, NP nucleus pulposus, NPCSTD-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in standard conditions, NPCTie2+-EVs extracellular vesicles isolated from nucleus pulposus cells cultured in Tie2-enhancing conditions

Full size image

Discussion

In this study, we showed that NPC-EVs were able to promote dNPC proliferation and reduce senescence in vitro while ameliorating pain and attenuating IDD in vivo. Moreover, we demonstrated that implementing a Tie2-enhancing protocol could augment NPC-EV therapeutic potential, emphasizing the importance of optimizing culture conditions.

Tie2 expression is associated with a progenitor phenotype characterized by multipotency and self-renewal capacity. The number of Tie2+-NPCs decreases with age and advancing IDD, possibly explaining the progressive reduction of the IVD inherent self-repair capacity [7]. Therefore, Tie2+-NPCs may constitute a potent cell source for IVD regeneration [11]. Recent evidence has demonstrated that EVs function as vectors of anabolic and anticatabolic mediators secreted by cells exploited as transplantation products to treat IDD. Indeed, EVs carry several biomolecules involved in cell communication, fate, and metabolism [18]. Furthermore, the EV cargo directly reflects the pathophysiological state of the donor cell, hence modulating the target cell response accordingly [27]. We postulated that increasing Tie2 expression in NPCs before EV isolation would promote the acquisition of a cargo with enhanced regenerative properties. According to our results, NPCTie2+-EVs exhibited the highest reparative capacity, although slightly superior to NPCSTD-EVs. By being isolated from relatively healthy NPCs, the intrinsic Tie2 expression and/or metabolic state of donor cells might equally result in the acquisition of an anabolic EV cargo.

Furthermore, our study revealed intriguing insights into pain-related behaviors. While the Von Frey test successfully demonstrated a reduction in mechanical hypersensitivity by NPC-EVs, other behavioral outcomes and systemic levels of IL-6 did not yield significant differences, similar to previous studies [22, 23]. This could be attributed to the specific nature of tail IDD pain, which might not exert a substantial impact on overall behavior, especially considering the tail’s limited weight-bearing role. Additionally, the acute degeneration model employed might not fully replicate the complex environment necessary to mimic discogenic pain [28].

Differently from previous studies, no significant beneficial outcomes of BM-MSC-EVs were found in vivo. Indeed, it has been earlier shown that BM-MSC-EVs were able to blunt ECM degradation, decrease disc cell loss, and dampen tissue inflammation and oxidative stress [18]. These differences could stem from several variables, including variations in cell sources, donor characteristics, and culture conditions. Nonetheless, while previous studies reported the use of repeated EV injections to achieve their therapeutic outcomes [18], a single injection of our NPC-derived EV product yielded a significant regenerative effect. These discrepancies, coupled with the wide variability among EV doses, units applied (protein vs. nanoparticle concentration), and inaccurate reporting in previous studies [18], emphasize the urgent need to establish an optimal dosing regimen for EV-based products.

While the caudal IVDs offer accessibility, established efficacy for evaluating intradiscally delivered treatments, and reproducibility, it is worth noting that the tail puncture model employed in this study has inherent limitations. Indeed, this model cannot fully replicate the complexity of chronic human lumbar IDD, particularly in terms of variations in cell types (such as notochordal cells in rats vs. human NPCs [29]) and biomechanical factors. These disparities may restrict the comprehensive understanding of the regenerative potential of our EV product in a real-world scenario. Another crucial factor to address is the xenogeneic nature of our EV transplantation (i.e., human to rat), potentially triggering an immunogenic response. However, our study did not find a reaction to our transplant product. This is in line with previous research that convincingly showed that EVs, either administered locally or systemically, did not provoke significant toxic or immunogenic reactions among different species [18, 30]. Therefore, we do not view this as a substantial concern. Although showing promising regenerative outcomes, the cargo of our EV product has not been specifically characterized. Unraveling the specific EV content will be essential to describe and predict its eventual effects. Moreover, additional in vitro experiments are needed to characterize the interaction between EVs and target cells, involved pathways, and mechanisms of action.

In summary, our study demonstrated NPC-EV therapeutic potential, highlighting the regenerative prowess of EVs isolated from NPCTie2+. These EVs countered IL-1β-induced catabolic effects in vitro and preserved DHI and tissue morphology in vivo. Despite promising outcomes, clinical translation of EV products will demand extensive research, including cargo characterization and standardization.