Targeting of Deciduous Tooth Pulp Stem Cell–Derived Extracellular Vesicles on Telomerase-Mediated Stem Cell Niche and Immune Regulation in Systemic Lupus Erythematosus

Deciduous tooth pulp stem cells, which were first obtained from remnant dental pulp tissue in exfoliated deciduous teeth, are commonly referred to as stem cells from human exfoliated deciduous teeth (SHED) (1). Recently, a standard operation procedure for manufacturing clinical-grade SHED was established and secured the safety and quality associated with characteristics of SHED (2). SHED is composed of a unique population of postnatal mesenchymal stem/stromal cells (MSCs), which are able to undergo self-renewal, colony formation, proliferation, and multipotency into osteoblasts, odontoblasts, chondrocytes, adipocytes, and neural cells (1, 3). SHED exhibit a reduced tumorigenesis compared with bone marrow–derived MSCs (BMMSCs) (4). Therefore, the unique properties of SHED have been harnessed in a first-in-human clinical trial for the treatment of trauma-induced tooth injury (5). Recently, several studies have focused on the immunomodulatory function of SHED in immune cells, including dendritic cells, macrophages, and T lymphocytes (3, 6, 7). Single systemic SHED transplantation rescues immune conditions in various animal models of human autoimmune diseases, such as autoimmune encephalomyelitis, rheumatoid arthritis, and systemic lupus erythematosus (SLE) (3, 8, 9). However, the underlying mechanism of SHED-based therapy in autoimmune diseases remains unclear.

Telomerase reverse transcriptase (TERT), which consists of a catalytic subunit of telomerase, is responsible for maintaining telomere length to promote the chromosomal stabilization of telomeres (10). The majority of normal human somatic cells express undetectable levels of telomerase activity, whereas stem/progenitor cells exhibit varied telomerase activity (11). Both telomere length and telomerase levels play important roles in regulating stem cell behavior (12). BMMSCs are known to be a key element of the bone marrow (BM) microenvironment to support hematopoiesis and modulate the immune system (13, 14). Telomerase activity plays a critical role in the functions of stem cell niche formation and immune regulation of BMMSCs (15–17). Meanwhile, human and mouse BMMSCs (mBMMSCs) derived from SLE patients and SLE-model MRL/lpr mice exhibit the impaired functions of hematopoietic niche formation and immune regulation as well as cell proliferation and multipotency (18), suggesting that the telomerase activity-associated functions of recipient BMMSCs may participate in the pathogenesis and progression of SLE. However, little is known about the mechanism(s) of how donor SHED target recipient BMMSCs in terms of the functions for hematopoietic niche formation and immune regulation in MRL/lpr mice.

Single SHED transplantation employs multiple mechanisms, including direct and indirect cell‒cell contact (3, 6–8, 19). Recent studies have demonstrated that the singly transplanted donor SHED are engrafted in the targeted tissues and organs of recipients, at very low frequencies (3, 6), indicating that cell‒cell communication via trophic factors released from the parental SHED profoundly contributes to the underlying mechanism of SHED-based therapy. Most cells are known to release lipid bilayer membrane-bound particles, extracellular vesicles (EVs), which include exosomes (40–100 nm in diameter) and plasma membrane-derived microvesicles (100–1000 nm in diameter). Stem cell–derived EVs particularly contain enriched small RNAs, mostly microRNAs (miRNAs), compared with the parental cells, and facilitate cell‒cell communication under physiological and pathological conditions (20). SHED-releasing EVs (SHED-EVs) participate in intercellular communication to exert therapeutic effects via bioactive molecules, such as proteins and miRNAs, in acute inflammation, brain injury, and Parkinson disease (21–23). Our study showed that SHED-EV–containing RNA rescued the telomerase activity of recipient BMMSCs in estrogen-deficient mice, leading to the suppression of bone resorption through an improved osteoblastic function in recipient BMMSCs (24). However, whether SHED-EVs participate in the treatment of SLE and their potential mechanism of action have not yet to be elucidated. In this study, we investigated whether SHED-EVs contribute to the recovery of SLE-like disorders in MRL/lpr mice and whether they act as an intercellular communicator to recipient BMMSCs by rescuing the functions of hematopoietic niche formation and immune regulation via the activation of telomerase activity. Thus, this study aims to improve our understanding of the underlying mechanism of SHED-based therapy for SLE.

Human deciduous teeth were collected from discarded clinical samples from healthy pediatric donors (n = 3, 5–7 y) in the Department of Pediatric Dentistry, Kyushu University Hospital. The handling of human samples was approved by the Kyushu University Institutional Review Board for Human Genome/Gene Research (protocol no. 393-01, 678-01). We obtained written informed consent from the parents of each patient on behalf of the child donors. All experimental procedures in this study were performed in accordance with the relevant guidelines and regulations.

Immunocompromised NOD-SCID mice (female, 8 wk old) were purchased from CLEA Japan. (Tokyo, Japan). C57BL/6J and C57BL/6J-lpr/lpr (MRL/lpr) mice (female, 8 wk old) were purchased from Japan SLC (Shizuoka, Japan). All animal experiments were approved by the Institutional Animal Care and Use Committee of Kyushu University (approval no. A21-044-1 and A20-041-0).

SHED were cultured according to previous studies (1, 3). The dental pulp tissues of human deciduous teeth were treated with 0.3% collagenase type I (Worthington Biochemicals, Lakewood, NJ) and 0.4% dispase II (Sanko Junyaku, Tokyo, Japan) for 60 min at 37°C. The dissociated cells were passed through a 70-μm cell strainer and seeded in T-75 culture flasks with a growth medium. The growth medium consisted of 15% FBS (Equitech-Bio, Kerrville, TX), 100 µM l-ascorbic acid 2-phosphate (FUJIFILM Wako Pure Chemicals, Osaka, Japan), 2 mM l-glutamine (Nacalai Tesque, Kyoto, Japan), and 100 U/ml penicillin and 100 µg/ml streptomycin (Nacalai Tesque) in MEM Eagle α Modification (αMEM; Thermo Fisher Scientific, Waltham, MA). After overnight, nonadherent cells were removed, and the attached colony-forming cells (CFCs) were passaged to expand. Passage 3 SHED was determined by colony formation, cell surface Ag, and multipotency according to the previously established protocol (2) and used for further experiments.

Cells were pretreated for 3 d with small interfering RNA for Tert (siRNATert) and RAB27A (siRNARAB27A) and its control scrambled small interfering RNA (siRNACont) (20 nM each; Santa Cruz Biotechnology, Santa Cruz, CA) in αMEM without antibiotics using Lipofectamine RNAiMax (Thermo Fisher Scientific).

SHED were subconfluently grown in an EV-depleted conditioned medium (CM) containing 15% EV-depleted FBS, which were prepared at 4°C for 16 h at 10,000 × g by using an ultra-high speed centrifugal machine (himac CP80a; Hitachi, Tokyo, Japan) equipped with a swing rotor (P40ST; Hitachi) in 13PA tubes (Hitachi). Subsequently, CM was collected from 3-d SHED cultures (SHED-CM) with DMEM (Thermo Fisher Scientific). SHED-EVs were purified from the SHED-CM using an exoEasy Maxi kit (Qiagen, Valencia, CA) according to the manufacturer’s protocol. A fraction of the SHED-EVs were treated with RNase A (5 U/ml; Thermo Fisher Scientific) at 37°C for 3 h and incubated with RNase inhibitor (40 U/ml; Thermo Fisher Scientific) at room temperature for 10 min. As MOCK treatment, PBS was used instead of RNase. SHED-EVs were characterized by a previously reported method (24); the particle size was measured with an analyzer (NanoSight N3000; Malvern Panalytical, Malvern, U.K.). The surface Ag expression was analyzed using ExoAB Ab kit (ExoAB-KIT-1; System Bioscience, Palo Alto, CA) and R-PE–conjugated anti-rabbit IgG Ab (Cell Signaling Technology, Danvers, MA) by flow cytometric (FCM) analysis. Total RNA and total protein were extracted from SHED-EVs and SHED using miRNeasy Mini kit (Qiagen) and M‐PER mammalian protein extraction reagent (Thermo Fisher Scientific) mixed with a proteinase inhibitor mixture (Nacalai Tesque), respectively. RNA and protein were qualified using small RNAchips (Agilent, Santa Clara, CA) on a bioanalyzer (Agilent 2100; Agilent) and Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA) on a microplate spectrometer (Multiskan GO; Thermo Fisher Scientific).

MRL/lpr mice (16 wk old) were i.v. administered passage 3 SHED (1 × 10⁵ in 100 μl of PBS per 10 g of body weight) and SHED-EVs (100 μg in 100 μl of PBS) pretreated with or without RNase and sacrificed 4 wk posttransplantation. The age-matched C57BL/6J and MRL/lpr mice were infused with PBS (100 µl per 10 g of body weight) and used as experimental controls. Four weeks after the administration, peripheral blood, urine, and tissue samples (kidney, spleen, and long bones) were collected from the mice.

Serum anti-nuclear Ag (ANA) and anti-dsDNA IgG/IgM Abs were measured using mouse ANA (ANA/extractable nuclear Ag [ENA]) Ig's (Total A+G+M) ELISA Kit (Alpha Diagnostic, San Antonio, TX) and mouse anti-dsDNA IgG/anti-dsDNA IgM Specific ELISA Kit (Alpha Diagnostic), respectively. Serum creatinine and urine protein were measured using creatinine parameter assay kit (R&D Systems, Minneapolis, MN) and Bio-Rad protein assay (Bio-Rad Laboratories), respectively. Whole peripheral blood was stained with specific Abs, followed by treating with BD FACS Lysing Solution (BD Bioscience, Franklin Lake, NJ) to obtain PBMCs.

mBMMSCs were cultured according to a previous study (25). The mouse growth medium consisted of 20% FBS (Equitech‐Bio), 2 mM l‐glutamine (Nacalai Tesque), 55 μM 2‐mercaptoethanol (Thermo Fisher Scientific), and 100 U/ml penicillin and 100 µg/ml streptomycin (Nacalai Tesque) in αMEM (Thermo Fisher Scientific). Attached CFCs were passaged once, and passage 1 mBMMSCs were used for further experiments.

SHED-EVs (100 μg in 100 μl of PBS) were labeled with CFSE (Thermo Fisher Scientific) or PBS and i.v. infused into MRL/lpr mice (16 wk old). After 3 d of the infusion, bone samples were harvested. Cryosections of the bone samples were stained with DAPI (Nacalai Tasque). mBMMSCs (1 × 10⁶) were cultured with or without CFSE-labeled or nonlabelled SHED-EVs (20 μg/ml) for 24 h under the EV-depleted condition.

Total RNA was obtained from mBMMSCs and SHED using an RNeasy Mini Kit (Qiagen), and cDNAs were prepared by reverse transcription using a Revertra Ace qPCR kit (Toyobo, Osaka, Japan). Reverse transcription–quantitative PCR (RT-qPCR) was performed using an EagleTaq Universal Master Mix (Roche, Basel, Switzerland) with target-specific TaqMan probes (Thermo Fisher Scientific) for mouse Tert (Mm00436931_m1), mouse 18S rRNA (Mm03928990_g1), human RAB27A (Hs00608302_m1), and human 18S rRNA (Hs99999901_s1). Mouse and human 18S rRNA were used for normalization.

The activity of telomerase in mBMMSCs was evaluated by quantitative PCR–based telomeric repeat amplification protocol (RQ-TRAP) using a quantitative telomerase detection kit (Allied Biotech, Ijamsville, MD) according to the manufacturer’s instruction. HEK293T cells (Thermo Fisher Scientific) were used as positive controls. Heat-inactivated samples were used as negative controls.

mBMMSCs (4.0 × 10⁶) were mixed with hydroxyapatite tricalcium phosphate particles (HA/TCP) (40 mg; Zimmer, Warsaw, IN) and s.c. implanted under the dorsal skin of immunodeficient NOD-SCID mice (female, 10 wk old). Eight weeks after the implantation, the implant tissues were harvested and used for further histological, immunological, and cytological niche formation assays.

Paraffin-embedded sections of decalcified implant tissues were treated with H&E. The area of newly formed BM-like components was measured on seven randomly selected representative images of H&E-stained sections using Image J software (National Institutes of Health, Bethesda, MD). The results were shown as the ratios of the newly formed BM area to total tissue area.

Cells were obtained from mBMMSC implants after being treated with 0.4% dispase II (Sanko Junyaku) in HBSS (Nacalai Tesque) according to previous studies (16, 26) and passed through a 40-μm cell strainer. The single-cell suspension was stained with R-PE–conjugated primary Abs to hematopoietic cell markers, including Sca-1 (clone D7; BioLegend, San Diego, CA), c-Kit (clone 2B8; BioLegend), and CD45 (clone I3/2.3; BioLegend) and assessed by FCM analysis. As controls, R-PE–conjugated rat IgG2a (clone G013C12; BioLegend) and rat IgG2b (clone RTK4530; BioLegend) Abs were used.

For hematopoietic CFC assay, mouse BM cells (1 × 10⁴) were obtained from femur of wild-type C57BL/6 mice (B6-BMCs). They were resuspended with 2% FBS (Equitech-Bio) in IMDM (Thermo Fisher Scientific) and cultured in a methylcellulose-based medium (MethoCult GF M3444; Stem Cell Technologies, Vancouver, Canada) with or without the CM of mBMMSCs. The number of colonies was counted after 12–14 d as described previously (26, 27). The level of stem cell factor (SCF) was measured in the CM of mBMMSCs using a mouse SCF Quantikine ELISA kit (R&D Systems)

To obtain activated mouse CD4⁺ T cells, naive CD4⁺ T cells (1 × 10⁶ per well) were magnetically sorted from mouse splenic cells using Naive CD4⁺ T Cell Isolation Kit, Mouse (Miltenyi Biotec, Aubun, CA), and stimulated with plate-bounded anti-CD3ε (5 μg/ml; clone 145-2C11; BD Bioscience) and soluble anti-CD28 (1 μg/ml; clone 37.51; BD Biosciences) Abs for 3 d on 24-well multiplates with a complete medium. The complete medium consisted of 10% heat-inactivated FBS (Equitech-Bio), 2 mM l-glutamine (Nacalai Tasque), 50 mM 2-ME (Thermo Fisher Scientific), and antibiotic mixture (Nacalai Tesque) in DMEM (Thermo Fisher Scientific). mBMMSCs (2 × 10⁵ per well) were incubated overnight on wells of 24-well multiplates. The activated CD4⁺ T cells (2 × 10⁵ per well) were cocultured with mBMMSCs for 6 d in the complete medium supplemented with specific mixtures for inducing IL-17–secreting Th17 cells (mouse TGF-β1 [10 U/ml; Peprotech, Rocky Hill, NJ], IL-6 [10,000 U/ml; Peprotech], anti-mouse IL-2 Ab [10 μg/ml; clone JES6-1A12; BioLegend], anti-mouse IL-4 Ab [10 μg/ml; clone 11B11; BioLegend], and anti-mouse IFN-γ Ab [10 μg/ml; clone XMG1.2; BioLegend]) and for inducing regulatory T cells (Tregs) (TGF-β1 [10 U/ml; Peprotech] and IL-2 [2000 U/ml; Peprotech]). For apoptosis assay, the activated T cells (2 × 10⁵ per well) were cocultured with mBMMSCs in complete medium for 3 d.

Mouse PanT cells were isolated from mouse spleen using a PanT Isolation Kit, Mouse (Miltenyi Biotec), and activated with plate-bounded anti-CD3ε (1 μg/ml; clone 145-2C11; BD Biosciences) and soluble anti-CD28 (1 μg/ml; clone 37.51; BD Biosciences) Abs for 3 d in the complement medium. The activated T cells (1 × 10⁶ per mouse) or PBS were i.v. injected into NOD-SCID mice (10–12 wk old). Two days after the T cell adoption, mBMMSCs (1 × 10⁶ cells per 10 g of body weight) were i.v. transplanted into the T cell–adopted mice. Survival of the mice was inspected daily. MRL/lpr mice (16 wk old) were i.v. transplanted mBMMSCs (1.0 × 10⁵ in 100 µl PBS per 10 g of body weight) and sacrificed 4 wk posttransplantation. As experimental controls, the age-matched NOD-SCID, C57BL/6J, and MRL/lpr mice were infused with PBS (100 μl per 10 g of body weight).

All samples were measured with a flow cytometer (FACSVerse; BD Biosciences) and analyzed using FACSuite software (BD Biosciences). The positive cell number (percentage) of single-labeled cells was determined by comparing with the corresponding control cells stained with the corresponding isotype-matched Ab with a false-positive rate of less than 1%. For detecting Tregs, cells were stained with PerCP-conjugated anti-CD4 (clone GK1.5; BioLegend), FITC-conjugated anti-CD8a (clone 53-6.7; BioLegend), and allophycocyanin-conjugated anti-CD25 (clone 3C7; BioLegend) Abs (1 μg each) for 30 min on ice, followed by incubating with R-PE–conjugated anti-Foxp3 Ab (clone 150D; BioLegend) using a Foxp3 Fix/Perm Buffer set (BioLegend). For detecting Th17 cells, cells were stained with PerCP-conjugated anti-CD4 (clone GK1.5; BioLegend) and allophycocyanin-conjugated anti–IFN-γ (clone XMG1.2; BioLegend) Abs, followed by incubating with R-PE–conjugated anti–IL-17 Ab (clone TC11-18H10.1; BioLegend) using a Foxp3 Fix/Perm Buffer set (BioLegend). The apoptotic T cells were detected by staining with PerCP-conjugated anti-CD4 (clone GK1.5; BioLegend) Ab, followed by allophycocyanin Annexin-V Apoptosis Detection Kit with 7-AAD (BioLegend).

The data are shown as mean ± SEM. Comparisons between two groups were performed using an independent two-tailed Student t test. Multiple group comparisons were performed by one-way ANOVA followed by Tukey post hoc test. Kaplan–Meier and Kruskal–Wallis tests were used for the survival assays. A p value < 0.05 was considered statistically significant. All statistical analyses were performed using PRISM 6 software (GraphPad, La Jolla, CA).

SHED were isolated from deciduous tooth of multiple healthy donors according to the previous reports (1–3) and i.v. transplanted SHED into MRL/lpr mice at the age of 16 wk. As reported previously (3), systemic SHED transplantation rescued SLE-like disorders compared with the age-matched nontransplanted MRL/lpr and wild-type C57BL/6 mice 4 wk posttransplantation (Supplemental Fig. 1).

Recently, we demonstrated that SHED exhibit significant immunomodulatory properties compared with human BMMSC, as indicated by inhibiting Th17 cells in vitro and elevating the ratio of Tregs via Th17 cells in MRL/lpr mice in vivo (3). In BMMSC-based therapy for MRL/lpr mice, Fas expressing on donor BMMSC-releasing exosomes improved the functions of recipient BMMSCs (28). However, it remains unclear whether donor SHED target the functions of recipient BMMSCs in MRL/lpr mice. To investigate the hematopoietic niche-forming and immunomodulatory functions of recipient BMMSCs, mBMMSCs were isolated from SHED-transplanted MRL/lpr, nontransplanted control MRL/lpr, and control wild-type C57/BL6 mice, referred to as SHED-BMMSCs, lpr-BMMSCs, and B6-BMMSCs, respectively, and implanted with HA/TCP powders as carriers under the back skin of immunocompromised SCID mice. lpr-BMMSCs showed an impaired hematopoietic niche formation compared with B6-BMMSCs 8 wk postimplantation, as indicated by the decreased formation of de novo BM-like components and reduced number of Sca-1⁺, c-Kit⁺, and CD45⁺ cells in the implants by histological and FCM assays (Fig. 1A–C). lpr-BMMSCs secreted lower amounts of SCF than B6-BMMSCs by ELISA (Fig. 1D). B6-BMCs were cocultured with mBMMSCs. lpr-BMMSCs suppressed the hematopoietic colony formation compared with B6-BMMSCs by CFC assay (Fig. 1E). SHED-BMMSCs recovered the hematopoietic niche-forming function, as indicated by the increased formation of de novo BM-like components; induced number of Sca-1⁺, c-Kit⁺, and CD45⁺ cells; and enhanced functions of SCF secretion and hematopoietic colony formation by histological, FCM, immunological, and CFC assays (Fig. 1A–E).

We then found that lpr-BMMSCs showed an impaired in vitro immunomodulatory function compared with B6-BMMSCs, as indicated by the increased CD4⁺IL-17⁺IFN-γ^- cells, decreased CD4⁺CD25⁺Foxp3⁺ Tregs, and enhanced CD4⁺AV⁺7AAD⁺ Th17 cells by FCM analysis (Fig. 1F, 1G). lpr-BMMSC transplantation had no effect on the lifespan of T cell–adopted SCID mice (Fig. 1H). Secondary transplantation of lpr-BMMSCs did not improve not only the increased levels of serum autoantibodies, urine protein, and serum creatinine in MRL/lpr mice 4 wk posttransplantation by ELISA and biochemical assays (Supplemental Fig. 2A, 2B) but also the increased CD4⁺IL17⁺IFN-γ⁻ PBMNCs, decreased CD4⁺CD25⁺Foxp3⁺ PBMNCs, and suppressed ratio of Tregs to Th17 cells by FCM assays (Supplemental Fig. 2C, 2D). SHED-BMMSCs recovered the in vitro immunomodulatory function, as indicated by the suppression of CD4⁺IL17⁺IFN-γ cells and enhancement of CD4⁺CD25⁺Foxp3⁺ and CD4⁺AV⁺7AAD⁺ cells (Fig. 1F, 1G). SHED-BMMSC transplantation prolonged the lifespan of the T cell–adopted SCID mice (Fig. 1H) and alleviated SLE-like disorders in MRL/lpr mice (Supplemental Fig. 2). Thus, systemic SHED transplantation rescued the hematopoietic niche-forming and immunomodulatory functions of BMMSCs in MRL/lpr mice.

Previous studies have indicated that low levels of telomerase activity play important roles in the stem cell niche-forming and immunomodulatory functions of BMMSCs (16, 17). RQ-TRAP and RT-qPCR assays revealed that telomerase activity and Tert expression were lower in lpr-BMMSCs than in B6-BMMSCs but rescued in SHED-BMMSCs (Figure 2A, 2B). The functional knockdown by siRNATert reduced the telomerase activity and Tert expression in SHED-BMMSCs (Fig. 2A, 2B).

We then found that siRNATert-treated SHED-BMMSCs showed the reduced de novo formation of BM-like components and decreased number of Sca-1⁺, c-kit⁺, and CD45⁺ cells in the implants compared with siRNACont-treated SHED-BMMSCs by histological and FCM assays (Fig. 2C–E). By ELISA and CFC assay, siRNATert-treated SHED-BMMSCs exhibited the suppressed SCF secretion and hematopoietic colony formation (Fig. 2F, 2G) and inhibited the in vitro immunoregulatory function, as indicated by the enhanced induction of CD4⁺IL-17⁺IFN-γ⁻ cells and suppressed induction of CD4⁺CD25⁺Foxp3⁺ and CD4⁺AV⁺7AAD⁺ cells (Fig. 2H, 2I). Furthermore, siRNATert pretreatment attenuated the in vivo immunoregulatory function of SHED-BMMSCs, as seen by the prolonged lifespan in T cell–adopted SCID mice, and improved SLE-like disorders in recipient MRL/lpr mice (Fig. 2J, Supplemental Fig. 3). Thus, these results suggest that telomerase activity participated in the hematopoietic niche-forming and immunomodulatory functions of BMMSCs in MRL/lpr mice.

Given that the frequency of engrafted SHED is not significant in the BM of MRL/lpr mice (3, 24), we hypothesized that indirect communication of donor SHED is involved in the rescue of BMMSC functions in MRL/lpr mice. We focused on EVs because diverse trophic factors within MSC-releasing EVs are known to contribute to cell‒cell communication in MSC-based therapy (29, 30). Because RAB27A plays important roles in both secreting EVs from parent cells and maintaining homeostasis within hematopoietic system under inflammation (31, 32), we tested whether RAB27A-dependent SHED-EVs are involved in the therapeutic efficacy of BMMSCs in MRL/lpr mice by functional knockdown assay using siRNARAB27A. siRNARAB27A-treated SHED expressed a lower level of RAB27A expression in siRNACont-treated SHED (Fig. 3A). We systemically infused siRNARAB27A-treated SHED into MRL/lpr mice. siRNARAB27A knocked-down SHED attenuated the therapeutic efficacy of SHED transplantation in MRL/lpr mice 4 wk posttransplantation, but siRNACont-treated SHED did not (Fig. 3B–E). These findings suggest that SHED-EVs may participate in cell‒cell communication between donor SHED and recipient BMMSCs in MRL/lpr mice.

SHED-EVs were purified from SHED-CM and characterized by multiple assays as reported previously (24). The particle size of SHED-EVs ranged from 69 to 478 nm in diameter, and the average diameter and concentration were 228 ± 3.7 nm (mean ± SEM) and 1.3 × 10⁹ ± 2.3 × 10⁸ particles per milliliter, respectively, by particle tracking assay (Supplemental Fig. 4A). SHED-EVs highly expressed CD9 (81.2 ± 2.7%), CD63 (44.3 ± 4.2%), and CD81 (85.7 ± 5.6%), but not CD90 (4.2 ± 0.9%), by FCM assay (Supplemental Fig. 4B). The concentrations of total protein, small RNAs, and miRNAs in SHED-EVs were 798.3 ± 54.2 μg/ml, 4.8 ± 0.74 ng/µl, and 2.6 ± 0.63 ng/µl, respectively (Supplemental Fig. 4C). RNase treatment depleted the small RNAs and miRNAs within SHED-EVs to 2.5 ± 0.61 ng/µl and 1.3 ± 0.19 ng/µl, respectively (Supplemental Fig. 4C), whereas MOCK treatment with PBS did not. Both RNase and MOCK treatments retained the particle size, surface Ag expression, and protein content within the EVs (data not shown). These findings suggest that the RNase treatment can be used to evaluate the efficacy of the RNA content within SHED-EVs.

CFSE-labeled SHED-EVs pretreated with MOCK and RNase were loaded in lpr-BMMSC cultures. Both CFSE-labeled EVs were detected in lpr-BMMSCs, but nonlabelled SHED-EVs pretreated with MOCK were not (Fig. 4A). The SHED-EV loading recovered the Tert expression and telomerase activity in lpr-BMMSCs by RT-qPCR and RQ-TRAP, whereas RNase-pretreated SHED-EV loading did not (Fig. 4B, 4C). We then i.v. administrated SHED-EVs (100 μg per mouse) into 16-wk-old MRL/lpr mice. The dose of SHED-EVs was determined by previous studies (24, 28, 33). Systemic administration of SHED-EVs improved the peripheral autoantibody levels, renal functions, and levels of CD4⁺IL-17⁺IFN-γ^– and CD4⁺CD25⁺Foxp3⁺ PBMNCs of MRL/lpr mice 4 wk postadministration (Fig. 4D–H). Systemic administration of RNase-pretreated SHED-EVs attenuated the improved efficacy of SHED-EV administration (Fig. 4D–H). These findings suggest that diverse RNAs within SHED-EVs contribute to rescue SLE-like disorders in MRL/lpr mice.

By cell tracking assay, we found that CFSE-labeled SHED-EVs were taken up into the BMCs of recipient BM 7 d postadministration (Fig. 5A). We then isolated BMMSCs from MRL/lpr mice infused with MOCK (PBS), MOCK-pretreated SHED-EVs, and RNase-pretreated SHED-EVs, referred to as lpr-BMMSCs, EV-BMMSCs, and RNase-EV-BMMSCs. EV-BMMSCs recovered the Tert expression and telomerase activity compared with lpr-BMMSCs, whereas RNase-EV- BMMSCs did not, by RT-qPCR and RQ-TRAP (Fig. 5B, 5C). In s.c. implant assay, EV-BMMSCs exhibited the expanded de novo BM-like structure and enhanced Sca-1⁺, c-kit⁺, and CD45⁺ cells compared with lpr-BMMSCs (Fig. 5D–F). EV-BMMSCs enhanced the SCF production and hematopoietic colony formation by ELISA and CFC assay (Fig. 5G, 5H). RNase-EV-BMMSCs completely attenuated the in vitro and in vivo advantage of EV-BMMSCs (Fig. 5D–H).

We further investigated the effects of systemic SHED-EV administration on the in vitro and in vivo immunoregulatory functions of lpr-BMMSCs. SHED-EV administration rescued the in vitro immunoregulatory functions of EV-BMMSCs, as indicated by the suppressed induction of CD4⁺IL-17⁺IFN-γ⁻ cells and enhanced induction of CD4⁺CD25⁺Foxp3⁺ and Annexin-V⁺7AAD⁺ cells (Fig. 6A, 6B). EV-BMMSC transplantation prolonged the lifespan of T cell–adopted SCID mice (Fig. 6C) and improved the SLE-like disorders of MRL/lpr mice, as evidenced by the reduction of peripheral autoantibody levels, renal nephritis, and peripheral immune status by ELISA, biochemical assays, and FCM analysis (Fig. 6D–G). RNase-EV-BMMSCs completely attenuated the in vivo impact of EV-BMMSCs (Fig. 6). These findings suggest that RNA contents within SHED-EVs rescue the hematopoietic niche-forming and immunoregulatory functions of recipient BMMSCs via telomerase activity in MRL/lpr mice.

Single SHED transplantation has multiple underlying mechanisms, including direct cell‒cell contact and indirect cell‒cell communication with trophic factors, such as cytokines, within exosomes (3, 6–8, 19, 22). However, it is unclear whether SHED-EV–packaged factors contribute to cell‒cell communication between parental donor SHED and recipient damaged tissue/microenvironment to cure SLE. Recent studies have shown that s.c. transplanted BMMSCs participate in hematopoietic niche formation by recruiting recipient hematopoietic stem/progenitor cells (13, 27). Donor BMMSC-organized ectopic hematopoiesis can rejuvenate age-dependent multiple organ dysfunctions, such as BM hematopoiesis and renal function, prolonging the lifespan of the recipients (26), suggesting that the donor BMMSC-mediated hematopoietic niche-forming function plays an important role in curing diseases. The present RNA-depletion study showed the attenuation of the therapeutic recovery of SHED-EV administration to the immune tolerance and renal function in MRL/lpr mice and recipient BMMSC-mediated functions of hematopoietic niche formation and immune regulation via telomerase activity. These findings suggest a novel underlying mechanism of SHED-EV–based SLE therapy, in which the diverse RNA contents within SHED-EVs target the recipient BMMSC functions for hematopoietic niche formation and immune regulation via telomerase activity.

Telomerase activity contributes to the diverse cellular functions of MSCs, including self-renewal, cell proliferation, osteogenic differentiation, and tissue regeneration (15, 26, 34). Interestingly, telomerase in MSCs regulates the immunomodulatory function by recruiting Fas ligand expression (17). A mechanism of BMMSC-based therapy is evaluated: the Fas expression on donor BMMSC-releasing exosomes is transferred to improve recipient BMMSC functions via the epigenetic regulation of the miR-29b–Dnmt1–Notch cascade in MRL/lpr mice (28). Recently, we found that SHED-EVs contain MIR346, which binds to a region in the 3′-UTR of TERT mRNA to upregulate TERT expression (35), and participate in rescuing bone reduction in estrogen-deficient mice via epigenetically regulating Tert mRNA expression in recipient BMMSCs (24). The present RNA-depletion study showed the attenuation of advantage of SHED-EV administration to the hematopoietic niche-forming and immunomodulatory functions of recipient BMMSCs via the rescuing of Tert mRNA expression and its associated telomerase activity in MRL/lpr mice. These findings suggest that the RNAs within SHED-EVs are responsible for recovering the functions of hematopoietic niche formation and immune regulation in recipient BMMSCs through the epigenetically regulated TERT–telomerase activity pathway. Because the precise miRNA(s) within SHED-EVs responsible for rescuing Tert gene expression in recipient BMMSCs remains to be identified, additional experiments may be necessary to identify the critical RNA factor(s) in SHED-EVs to understand the mechanism of single SHED transplantation in MRL/lpr mice.

Which RNA content within SHED-EVs acts integrally on rescuing recipient BMMSCs to cure SLE disorders in MRL/lpr mice remains to be fully elucidated. In a recent study, SHED-CM is determined to be composed of a range of cytokines and chemokines with therapeutic effects, including antifibrotic, anti-inflammatory, and immunosuppressive (7). Multiplex-factor SHED-CM shows multifaceted benefits in the treatment of acute and chronic inflammation and autoimmune diseases (8, 36), implying that different factors in SHED-CM cooperatively correct the etiological factor-caused pleiotropic phenotypes in tissues and organs. Recent studies have suggested that SHED-EVs contain multiple small RNAs, including miRNAs, which exert immunosuppressive effects as an immunomodulator (8, 19). The present RNA-depletion study showed the attenuation of benefits of SHED-EVs in recipient BMMSCs in MRL/lpr mice, as reported in estrogen-deficient osteoporosis model mice (24). Thus, the interactive regulation of multiple RNAs within SHED-EVs may be implied as a mechanism of single SHED transplantation in MRL/lpr mice.

Our findings suggest that diverse RNA(s) within SHED-EVs may participate in the therapeutic mechanism of cell‒cell communications in SHED-based therapy for SLE by targeting the Tert gene of recipient BMMSCs. However, we do not exclude other cell‒cell communication mechanisms in SHED-based therapy. TERT activity in MSCs participates in a range of therapeutic properties, including proliferation, differentiation, tissue regeneration, and immunomodulation (15, 17, 34). Some of these mechanisms may function via TERT‒mitochondria interactions, such as the mitochondrial TERT matrix, which protects mitochondria against oxidative stress-induced damage (19), suggesting that the activity of mitochondria contributes to MSC functions. Mitochondrial transfer has been reported as a novel cell‒cell communication of MSCs in the rescue of tissue damage (37, 38). Donor-provided-MSC mitochondria improve recipient cells to exert therapeutic benefits in acute lung injury and diabetic nephropathy (39, 40). The cellular function of impaired recipient cells is rescued by mitochondria transferred from donor cells in vitro (41), and MSC-mediated mitochondrial transfer via Miro1 enhances the therapeutic potential of MSCs (42), suggesting that the cell‒cell communication via mitochondrial transfer can also be considered a mechanism for SHED-based therapy for SLE. Additional experiments are needed to fully elucidate the role of mitochondrial transfer in improving recipient BMMSCs for use in SHED-based immune therapies.

Taken together, the findings presented in this study demonstrate that systemic SHED-EV administration exerts therapeutic efficacy for SLE-like disorders in MRL/lpr mice by targeting recipient hematopoietic niche function to rescue the immune microenvironment regulated by recipient BMMSCs via revitalizing Tert-associated telomerase activity. Thus, to our knowledge, our findings provide novel insights into the mechanism of SHED-based therapy, revealing an important relationship between telomerase and recipient BMMSCs. Further research should focus on the molecular and cellular mechanism(s) of how transplanted donor MSCs rejuvenate impaired recipient BMMSCs in diseased recipients.

We thank Editage (https://www.editage.com/) for English language editing.

The authors have no financial conflicts of interest.