Human hair follicle-derived mesenchymal stem cells improve ovarian function in cyclophosphamide-induced POF mice

Abstract

Introduction

Mesenchymal stem cell (MSCs) of different tissue origins have become a new option for the treatment of premature ovarian failure (POF) as they can recovery the ovarian function. However, there were rarely researches about human hair follicle-derived mesenchymal stem cells (HF-MSCs) in POF.

Objectives

In this study, we compared the effects of HF-MSCs and human umbilical cord mesenchymal stem cells (HU-MSCs) on POF models to explore the underlying molecular mechanisms.

Methods

Female mice received intraperitoneal cyclophosphamid for 10 days to induce the POF mice model. One week after drug withdrawal, the mice were randomly divided into four groups according to the tail vein injection of drugs, which were: Control group (CON), Premature ovarian failure group (POF), HF-MSCs treatment group (P–H group) and HU-MSCs treatment group (P–U group). Which Treatment once a week for 4 consecutive times. Serum and ovarian tissues were collected 2 weeks after the last treatment, and fertility was performed by mating. ELISA, HE staining, transmission electron microscopy (TEM) were applied to evaluate the ovarian function, oocytes quantity and quality, and the mechanism was verified by qRT-PCR and western blot. In addition, the tumorigenic risk of organs was assessed by long-term observation.

Results

The POF mice model was successfully established by intraperitoneal injection of cyclophosphamide 100 mg/kg/d for 10 days. Compared with POF group, two weeks after transplantation, serum FSH decreased, AMH and E2 increased in the P–H and P–U groups of mice (p < 0.05), but there was no significant difference between the P–H and P–U groups (p > 0.05). In addition, the number of primary follicles, secondary follicles and antral follicles in both P–H and P–U groups were significantly increased (p < 0.05), while the atretic follicles was significantly decreased (p < 0.05). The pups in POF group was significantly lower than that in P–H group and P–U group (p < 0. 01). Furthermore, those effects was more significant in P–H group than in P–U group (p < 0.05). In addition, the mitochondrial ultramicrostructure of the ovaries in the four groups showed a significant difference in the mitochondrial morphologies and number. In the POF group, the mitochondria presented a spheroids structure with fewer numbers, serious vacuolation and a disordered mitochondrial cristae arrangement. Nevertheless, after MSCs transplantation into the P–H and P–U group, we could observe ameliorative mitochondrial cristae alignment and vacuolation, as well as a small number of long rod-like structures. Mechanism study showed that KEAP1 protein expression was decreased in the P–H group, which increased the nuclear translocation of NRF2 and upregulated the expression of downstream HO-1 protein. At last, the possibility of tumor development after transplantation of HF-MSCs was excluded by long-term observation and organ anatomical examination.

Conclusion

HF-MSCs can improve ovarian function in cyclophosphamide-induced POF mice, and the effects were superior to HU-MSCs. The underlying mechanism may by inhibiting ferroptosis of granulosa cells through KEAP1/NRF2/HO-1 pathway.

Introduction

Premature ovarian failure (POF) refers to ovarian failure before age 40 years, follicle stimulating hormone (FSH) > 40U/L, accompanied by decreased estrogen levels and menopausal symptoms, the fertility is nearly lost, and it's the end state of premature ovarian insufficiency (POI) [1]. It triggers bone, cardiovascular, and neurological pathologies, which seriously affects physical/psychological health and reduces quality of life [2,3,4]. Reports showed that the incidence of POF before 40 years old is 1%, and before 30 years old is 1‰ [5]. According to the WHO projections, the incidence of infertility is increasing worldwide and will become the third most prevalent disease after tumors and cardiovascular diseases [6]. Evidence shows that survivors treated after the onset of pubertal period with alkylating agents had a 29-fold higher rate of POF when compared with survivors treated before the onset of puberty without alkylating agents [7]. As the incidence of cancer becomes younger, the POF group caused by chemotherapy and radiotherapy is gradually drawn attention [8, 9]. Therefore, the preservation of fertility and the improvement of ovarian function in young unpregnant women with tumors will be the primary considerations.

At present, the main methods to protect fertility include oocyte cryopreservation, embryo cryopreservation, ovarian tissue cryopreservation and ovarian transplantation, and the application of gonadotropin-releasing hormone agonist (GnRHa) [10, 11]. The feasibility, safety and effectiveness of various measures are still challenging [9]. For example, there are still some technical difficulties and ethical issues in cryopreservation technology, and the routine clinical procedures of subsequent assisted reproductive technology (ART) are not widely beneficial to all female patients with POF who undergo in vitro fertilization (IVF) treatment, while the protective effect of GnRHa on ovarian function is still controversial [12]. Therefore, improved strategies to prevent chemotherapy-induced gonadotoxicity are still needed.

Stem cell therapy has lately been used as a potential alternative treatment option to help wounded tissues or organs heal and return to normal function. A popular topic in the field of reproductive medicine, mesenchymal stem cells (MSCs) are multipotent adult stem cells derived from a variety of sources. They are presently undergoing evaluation in numerous clinical trials across the globe [13]. It is still mostly unknown exactly how MSCs mediate the salvage and repair of damaged organs and tissues. Growing evidence suggests that the ability of MSCs to treat disease may arise from other than implantation and differentiation [13]. These mechanisms include the release of hormones and proteins through paracrine effects, the transfer of exosomes containing RNA and other molecules, and involvement in cell proliferation and anti-apoptotic processes [10, 14, 15].

To date, many animal models of POF have confirmed that administration of MSCs from various cell types can protect ovarian function, demonstrating the possibility of restoring ovarian function and structure [13]. In addition, in 2018, Professor Sun Haixiang transplanted umbilical cord mesenchymal stem cells (HU-MSCs) combined with collagen scaffolds, which can save the overall ovarian function of women with POF and successfully achieve clinical pregnancy [16]. In 2020, 61 POI patients participate clinical study showed that transplantation of HU-MSCs rescued the ovarian function, as indicated by increased follicular development and improved egg collection. Four of the participants had successful pregnancies and deliveries with healthy babies [17]. The clinical trial result sugggests a possible therapy for POI by HU-MSCs transplantation. Among many types of MSCs, human hair follicle-derived mesenchymal stem cells (HF-MSCs) have many advantages. Firstly, easy accessibility and abundant pool of autologous stem cells; Furthermore, no ethical restriction on homology and high differentiation ability [18,19,20]. However, the effects of HF-MSCs on POF has not been reported, which will bring a new direction for the regeneration and functional repair of ovarian tissue.

Distinguished from apoptosis by the generation of reactive oxygen species and lipid peroxidation, ferroptosis is a special kind of iron-dependent programmed death that was originally described by Dixon et al. in 2012 [21]. In recent years, the regulatory mechanisms of ferroptosis in the physiological and pathological processes of the female reproductive system, such as polycystic ovary syndrome, placenta praevia, and ectopic pregnancy, have received increasing attention [22]. However, studies on ferroptosis-related reproductive disorders are still limited and inconsistent, and few studies on iron death in POF have been reported so far [23].

In this study, we explored HF-MSCs for the first time and compared with HU-MSCs on the restoration effects of ovarian structure and function in cyclophosphamide-induced POF mice.

Methods

Hair follicles from healthy adults and umbilical cord tissues from healthy full-term fetuses were obtained with the approval of the Medical Ethics Committee of Qingyuan People's Hospital, China (Approval No. IRB-2023–017). Volunteers agreed to the isolation of their tissues, culture and expansion of MSCs for this experimental study and signed informed consent.

Isolation and culture of HF-MSCs

The HF-MSCs were isolated as described previously [24]. Briefly, 20 complete hair follicles were plucked from the occipital region of each of healthy volunteers. These hairs were washed three times in phosphate-buffered saline (PBS, 20012027, Thermo Fisher Scientific, USA) containing 5% penicillin/streptomycin solution (15070063, Thermo Fisher Scientific, USA). Remove excess tissue and inoculate hair follicles placed on the 24-well plate (801006, NEST, China), one piece of hair per well. Add MSCs culture medium (SC2013-G, TBD, China) and place in cell culture incubator at 37 °C and 5% CO₂. The medium was changed every 3 days. Seven days after, long shrinking like cells grew out of hair follicles. When the HF-MSCs had proliferated to approximately 80% confluence, the cells were digested with cell digest (TrypLE ™ Express, 12604021, ThermoFisher Scientific, USA). Passaging culture, identification of 5th generation cells, and animal experimental research on 4th to 6th generation cells.

Isolation and culture of HU-MSCs

The HU-MSCswere isolated as described previously [25]. Briefly, umbilical cords were washed with phosphate-buffered saline (PBS; Thermo Fisher Scientific, Waltham, MA, USA) containing 5% penicillin/streptomycin solution (P/S; 100 IU/mL penicillin, 100 IU/mL streptomycin; Thermo Fisher Scientific). Umbilical cord tissues were carefully dissected to remove vessels and then cut into pieces of approximately 1 mm3.Thereafter, fresh medium was gently added to the seeded plate, followed by incubation at 37 °C in a humidified 5% CO₂ incubator. The medium was changed every 3 days. After 10–14 days, the tissue explants were removed from the Petri dishes, leaving behind the attached cells. When the HU-MSCshad proliferated to approximately 80% confluence, the cells were digested with cell digest (TrypLE™ Express, 12604021, ThermoFisher Scientific, USA). passaging culture, identification of 5th generation cells, and animal experimental research on 4th to 6th generation cells.

Flow cytometry

To identify and isolate MSCs, cultured 5th generation cells were collected for flow assay. Cells were collected into centrifuge tubes, the supernatant was removed by 300 × g centrifugation, PBS was added to rinse once, and 4% paraformaldehyde was added to fix 15 min at room temperature. PBS was rinsed agin, 1%BSA was added to seal for 1 h at room temperature, and centrifuged to remove supernatant. Primary antibodies were added separately: CD44 (0.125 µg/test, 14–0441-82, Thermo Fisher Scientific), CD73 (0.25 µg/test, 11-0739-42, Thermo Fisher Scientific), CD90 (0.5 µg/test 14-0909-82, Thermo Fisher Scientific), CD105 (1:400, MA5- 17041, Thermo Fisher Scientific), and CD31 (0.125 µg/test, 48-0319-42, Thermo Fisher Scientific), CD45 (0.25 µg/test, 14-0459-82, Thermo Fisher Scientific), 4-degree incubation overnight, PBS rinsing three times. Then, cells were incubated with the anti-rabbit or anti-mouse secondary antibody (Alexa Fluor 488 and 555, 1:2,000, 4408S, 4409S, 4412S, 4413S, Cell,Signaling Technology) incubate at room temperature for 2 h, rinse with PBS for 3 times, and test on the machine.

Immunofluorescence staining

Cultured 5th generation cells were inoculated into 24-well plates for immunofluorescence detection. Cell fusion was about 80%, rinsed once with PBS, fixed with 4% paraformaldehyde at room temperature and protected from light for 15 min. Rinsed once with PBS, closed with 1% BSA at room temperature for 1 h. Primary antibodies were added separately to CD44 (0.125 µg/test,14-0441-82, Thermo Fisher Scientific), CD73 (0.25 µg/test,11-0739-42, Thermo Fisher Scientific), CD90(0.5 µg/test 14-0909-82, Thermo Fisher Scientific), CD105(1:400,MA5-17041, Thermo Fisher Scientific), CD31 (0.125 µg/test,48-0319-42, Thermo Fisher Scientific), CD45 (0.25 µg/test,14-0459-82, Thermo Fisher Scientific) were incubated overnight at 4 degrees Celsius and rinsed three times in PBS. Subsequently, cells were incubated with the corresponding goat antimouse secondary antibody or goat anti-rabbit secondary antibody (Alexa Fluor 488 and 555,1:1,000,4408S, 4409S, 4412S, 4413S, CST), incubated at room temperature for 2 h, rinsed 3 times with PBS. Cells were then counterstained with Hoechst 33,342 dye (10 μg/mL; Thermo Fisher Scientific) for 15 min, rinsed 3 times with PBS, and visualized under an inverted fluorescence microscope.

Identification of multi-differentiation of HF-MCSs/HU-MSCs

Osteogenic differentiation

Cells were inoculated in gelatin-treated six-well plates at a density of 4 × 10⁴ cells/cm², 2 mL of MSC medium per well, and place in cell culture incubator at 37 °C and 5% CO₂. When the cell fusion reached 60%, discarde supernatant, add osteogenic differentiation complete medium (TBD20190002, TBD, China) 2 mL/well, induce culture. The medium was changed every 3 days and cultured continuously for 4 weeks. At the end of the induction culture, discarde supernatant, PBS rinsed twice, and the cells were fixed with 4% neutral paraformaldehyde solution 2 mL/well at room temperature for 20 min. Then discarde fixed solution, PBS wash twice, and stained with alizarin red staining solution (the induction medium kit was included) 1 mL/well staining for 15 min. Discarde solution, PBS wash 3 times, observe under the microscope and take pictures.

Adipogenic differentiation

Cells were inoculated in gelatin-treated six-well plates at a density of 4 × 10⁴ cells/cm², 2 mL of MSC medium per well, and place in cell culture incubator at 37 °C and 5% CO₂. When the cell fusion reached 80%, discarde supernatant, add adipogenic differentiation medium A (TBD20190004, TBD, China) 2 mL/well, induce culture. After 72 h of incubation, discarde the original culture supernatant, add solution B (2 mL/well) to induce culture. After 24 h, discard the original culture supernatant, and add solution A (2 mL/well) to induce culture. Repeat the above steps about 3–5 times. When obvious lipid droplets appear in the cells, replace with B solution and continue culturing for 7 days. Replace with fresh B solution every 2 days and continue culturing until the lipid droplets are large enough. Then discarde supernatant, PBS rinsed twice, and the cells were fixed with 4% neutral paraformaldehyde solution 2 mL/well at room temperature for 20 min. Then discarde fixed solution, PBS wash twice, and stained with oil red O staining solution (the induction medium kit was included) 1 mL/well staining for 15 min. Discarde solution, PBS wash 3 times, observe under the microscope and take pictures.

Chondrogenic differentiation

The cell precipitate was added into chondrogenic induction complete medium (TBD20190003, TBD, China), centrifuged at 150 × g for 5 min, the precipitate was re-suspended in medium. When the cell density was adjusted to 5 × 10⁵ cells/mL, and 500 μL of the cell suspension was inoculated into 15 mL centrifuge tubes, centrifuged at 150 × g for 5 min. Loosen the lid of the tube, and gently put it into the incubator at 37 °C and 5% CO₂. After 24 h, the bottom of the centrifuge tube was gently ruffled to suspend the cell sediment clumps, and put it back into the incubator to continue culturing. Change the medium every 2 days, and induce culture for 4 weeks. Dewatering, paraffin embedding, and sectioning. Sections were dewaxed and rehydrated, stained with Alisin blue stain for 30 min, rinsed under running water for 5 min. Dewatering, transparenting and sealing with neutral gum. Observe under the microscope and take pictures.

Animal experiments

The C57BL/6J mice female (7–8 weeks, SPF) and male (9–10 weeks, SPF) were purchased from Guangdong Sja Biotechnology Co., Ltd [SCXK(GD) 2020-0052]. All mice were adapted for 1 week after their purchase and were housed in the Laboratory Animal Center of the Qingyuan People's Hospital [SCXK(GD) 2014- 0082] according to institutional guidelines for laboratory animals under temperature and light controlled conditions with a 12-h daily cycle and were fed ad libitum. All animal experiments were approved by the Institutional Animal Care and Use Committee (Approval No. IRB-2022-137). The work has been reported in line with the ARRIVE guidelines 2.0.

Establishment POF mice model and treatment with HF-MSCs/ HU-MSCs

To establish the POF model, mice were injected intraperitoneally with 100 mg/kg cyclophosphamide (CTX, C0768, Sigma-Aidrich, USA) daily for 10 days, and saline was injected into the control group. At the 1st week of CTX withdrawal, according to the different drugs injected into the tail vein, they were randomly divided into 4 groups, 20 animals in each group: CON, POF, P–H (HF-MSCs, 1.0 × 10⁶/200 μl/mouse, Passage 4–6), and P–U (HU-MSCs, 1.0 × 10⁶/200 μl/mouse, Passage 4–6), in which the CON and POF group were injected with PBS, once a week for 4 consecutive times. 2 weeks after the last injection, the mice were anesthetized by using inhalation anesthesia isoflurane (20231001, Hengfengqiang Biotechnology Co., Ltd, CHINA), weighed, the eyes were removed to collect peripheral blood, and sacrificed with cervical dislocation. Both ovaries were removed for subsequent experiments (Fig. 1). The remaining mice were mated with fertile male mice (2:1) for 60 days and checked for fertility.

Fig. 1

Schematic description of the experimental design. CTX (100 mg/kg/d) was administered by intraperitoneal injection for 10 days. The 1st week of CTX withdrawal, HF-MSCs (1.0 × 10⁶/mouse) or HU-MSCs (1.0 × 10⁶/mouse) or PBS were transplanted by tail vein injection, once a week for 4 consecutive times. Experimental analyses were performed 2 weeks after the last injection

In vivo imaging of mice

In order to track the transplanted cells in vivo, the cells were labeled overnight with DiR Iodide (fluorescent probe, 40757ES25, YEASEN, CHINA). Adding 5 μm DiR Iodide to serum-free medium and adjusting the cell suspension density to 1 × 10⁶/ml, incubating at 37 °C for 20 min, and centrifuging 5 min. Then, decante supernatant, slowly add PBS preheated at 37 °C to rinse twice, ready for subsequent animal experiments. After MSCs were tail vein injected 1 week, anesthesia by isoflurane. Observe under the AniView100 Multimode Live Animal Imaging System (Boluteng Biological Technology Co., Ltd, Guangzhou, CHINA) and take pictures.

ELISA analysis

Plasma from the four groups of mice was collected at 2 weeks after HF-MSCs treatment. An ELISA kit (JYM, CHINA) was employed to detect the levels of AMH, FSH, E2, malondialdehyde (MDA), SOD, according to the directions. In addition, mouse plasma harvested to evaluate the levels of ROS using an ELISA kit (MEIMIAN, CHINA). In brief, the test plate containing 10 μl serum sample and 40 ul sample diluent was added to per well. Then, 50 μl HRP-conjugate reagent was added to each well and incubated at 37 °C for 30 min. Wash buffer was used to wash the wells five times (10 s, per wash). Then, 50 μl of substrate A solution and substrate B solution were mixed together at 37 °C for 15 min. Then, 50 μl of stop solution was added to each well. Finally, a spectrophotometer (BioTek, USA) was used to test the light absorbance.

Serum ferrous ion (Fe²⁺) detection

Serum ferrous ion concentration was detected by ferrous ion colorimetric test kit (E-BC-K773-M, Elabscience, CHINA). Briefly, referring to the operating procedure of the kit, the mouse serum was diluted onefold and operated, and the OD value was detected at 593 nm, and the ferrous ion concentration in serum was calculated according to the standard curve.

Hematoxylin and eosin (HE) staining

In brief, tissues were fixed using 4% paraformaldehyde at 37 °C for 12 h first. An approximately 5 μm thick frozen section of tissue was prepared, fixed with 95% anhydrous ethanol for 2 min, hematoxylinstained for 5 min, and differentiated in a differentiation solution for 2 min. After soaking in weak ammonia for 3 min, the slices were washed with deionized water for 5 min, stained with eosin for 5 min, and washed with deionized water for 5 min. After soaking in 70%, 80%, and 90% ethanol solutions for 1 min, they were washed twice with anhydrous ethanol for 1 min and then in xylene for 1 min before being embedded in neutral sesame oil. After HE staining, the histologic changes of the ovary and the number of follicles at all levels were observed in the full field of view of each section, and the final follicle count was the sum of the counts of the three sections and divided by three.

Transmission electron microscopy (TEM)

Ovarian tissues from mice in four groups were collected and fixed in electron microscope fixative for 24 h (G1102, Servicebio, Wuhan, China). Following washing with 0.1 M cacodylate buffer, the tissues were fixed with 2% cacodylate buffered osmium tetroxide at room temperature for 1 h. Next, a series of ethanol concentrations (50% to 100%) was applied to dehydrate the samples. The embedded and sliced samples were subsequently stained with both methanolic uranyl acetate and lead citrate. TEM images were obtained with a JEM-1400 transmission electron microscope (Tokyo, Japan).

RNA extraction and quantitative real-time polymerase chain reaction

Total RNA was extracted from ovary using TaKaRa MiniBEST Universal RNA Extraction Kit (Takara, #9767, JAPAN), then reverse transcribed to generate cDNA (Takara, #RR036A, JAPAN). RT-PCR was performed using a 96-well optical plate with the CFX96 Real-time PCR System (BIO-RAD, USA) and TB Green Premix Ex Taq II FAST qPCR (Takara, #CN830A, JAPAN). Cycling conditions are as follows: 94 °C, 5 s; 60 °C, 34 s; 72 °C, 40 s, 40 cycles. The following primer sequences were used: KEAP1: forward, 5′-GGCAGGACCAGTTGAACAGT-3′; reverse, 5′-GGGTCACCTCACTCCAGGTA-3′; NRF2: forward, 5′-CTACAGTCCCAGCAGAGTG-3′; reverse, 5′-GCTCAGAAACCTCCTTCCA-3′; HO-1: forward, 5′-GAGCAGAACCAGCCTGAACTA-3′; reverse, 5′-GGTACAAGGAAGCCATCACCA-3′; GAPDH: forward, 5′-AGGTCGGTGTGAACGGATTTG-3′; reverse, 5′-TGTAGACCATGTAGTTGAGGTCA-3′. Mouse-Gapdh was used for internal normalization. The 2^–ΔΔCt method was employed to determine relative mRNA expression levels.

Western blotting

Ovaries were homogenized in lysis buffer (PRO-PREP™ Protein Extraction Solution, Intron, KOREA), centrifuged at 14,000 × g for 15 min; then, supernatants were diluted to 1 μg/μl with 4 × sample buffer (Bio-Rad, Hercules, CA, USA) and frozen at 20 °C. Proteins samples were boiled for 3 min. The extracted proteins were separated by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to PVDF membranes. Membranes were blocked with 5% non-fat dry milk in PBS containing 0.1% Tween 20, then incubated at 4 °C overnight with 1:1000 dilution of NRF2 antibody (CST, 12721S, USA), 1:1000 dilution of HO-1 antibody (CST, 86806S, USA), 1:2000 dilution of KEAP1 antibody (Proteintech, 10503-2-AP, CHINA), 1:100,000 dilution of the internal reference antibody GAPDH (Jackson Immuno Research, 163475, USA). Horseradish peroxidase-conjugated secondary antibodies were incubated for 1 h at room temperature, and immunoreactivity was detected using enhanced chemiluminescence reagent (Meilunbio, MA0186-1, CHINA) and recorded on ChemiDoc MP Imaging System (BIO-RAD, USA). Visualized bands were quantified by densitometry with NIH Image J software (https://imagej.nih.gov/ij/docs/faqs.html). Intensities of bands were expressed relative to that of the control.

Statistical analysis

Data are expressed as mean ± standard deviation (SD), with *: p < 0.05, **: p < 0.01, and ***: p < 0.001 indicating statistically significant differences. Student’s t-test and one-way analysis of variance (ANOVA) followed by Bonferroni’s post hoc test were used where appropriate. All statistical analyses were performed using SPSS 22.0 and GraphPad Prism v8.0 (USA).

Results

Characterization and differentiation of HF-MSCs and HU-MSCs

On the 3rd to 5th day of culture of hair follicles of healthy volunteers, long shrinkage-like cells grew from the dermal papilla or the outer hair root sheath (Fig. 2B). As the culture time increases, the number of cells gradually increased. When the cells confluence were about 80%, the cells were passaged. The fifth-generation cells were detected by flow cytometry and immunofluorescence, which expressed surface markers of MSCs. They were positive for CD44, CD73, CD90, and CD105, but negative for CD31 and CD45 (Fig. 2C, D). The long shrink-like cells grown from hair follicles were induced to differentiate by osteogenic induction solution. Which induced about 21 days with the production of postcalcium salt nodules, and the red color of the calcium salt nodules by alizarin red staining, indicating that the cells have the ability to differentiate osteoblastically. The long shrink-like cells grown from hair follicles are induced to differentiate by adipogenic induction solution. It induced for about 28 days later, lipid droplets were produced and stained with Oil Red O, the lipid droplets turned red, indicating that the cells had the ability to differentiate into adipocytes. The long shrink-like cells grown from hair follicles are induced to differentiate by chondrogenic induction solution. Which induced about 21 days later, the spheres were stained light blue by Alisin blue staining solution after sectioning, indicating that the cells had the ability to differentiate into chondrocytes (Fig. 2E).

Fig. 2

Generation and characterization of mesenchymal stem cells (MSCs) from human. A The chromosomes of HF-MSCs volunteer (46, XY) and HU-MSCs volunteer's umbilical cord (46, XY). B The morphology of MSCs. The image showed typical spindle-shaped morphology of adherent MSCs. C Flow cytometry analysis found MSCs markers CD44, CD73, CD90, CD105 are positive, and CD31, CD45 are negative. D Immunofluorescence staining analysis found MSCs markers CD44, CD73, CD90, CD105 are positive, and CD31, CD45 are negative. Scale bar: 150 μm. E, F The differentiation of MSCs. Adipogenesis: Triglyceride produced by differentiated adipocytes was stained with Oil Red O. Osteogenesis: Deposited calcium was stained with Alizarin Red. Chondrogenesis: Chondrogenic nodules were stained with Alcian Blue. Scale bar: 200 μm

On the 7th to 10th day of culture of UC pieces from healthy volunteers, fibroblast morphology cells grow out (Fig. 2B). As the culture time increases, the number of cells gradually increases. When the cell confluence is about 80%, the cells are passaged. The cells were collected, pooled, and subjected to phenotypic characterization and multipotency assays. The spindle-shaped cells that expressed the MSC differentiation markers CD44, CD73, CD90, and CD105 (Fig. 2C, D) and possessed multipotent differentiation potentials towards chondrocytes, osteoblasts, and adipocytes were classified as HU-MSCs (Fig. 2F).

Establishment of CTX-induced POF model and transplantation with HF-MSCs & HU-MSCs

The POF mice model was successfully established by intraperitoneal injection of cyclophosphamide 100 mg/kg/d for 10 days (Fig. 3A). Following CTX administration, mice were randomly divided into four experimental groups that were administered with PBS or HF-MSCs or HU-MSCs via the tail vein. Following the transplant procedure, mice that received HF-MSCs or HU-MSCs recovered both body weight and ovary index significantly more than POF group, and the improvement effect in the P-F group was better than in the P–U group (Fig. 3A, C, D). To track HF-MSCs and HU-MSCs in vivo after transplantation, DiR Iodide was co-cultured with MSCs before transplantation. The distribution of MSCs in vivo was detected by the AniView100 Multimode Live Animal Imaging System. One week after transplantation, luminescent signals were detected in the liver region of the P–H and P–U groups mice, and not in the CON and POF groups (Fig. 3B).

Fig. 3

Changes in ovary index of mice treated with CTX by HF-MSCs & HU-MSCs transplantation and tracking of MSCs distribution in vivo. A Body weights were measured during CTX exposure and HF-MSCs transplantation (n = 8). B Tracking of MSCs distribution in vivo by the AniView100 Multimode Live Animal Imaging System with DiR Iodide staining. C Ovary index = bilateral ovarian weight/body weight (mg/g), 2 weeks after the last transplantation of MSCs, mice were weighed and then executed, and bilateral ovaries were isolated (n = 8). D Ovaries from mice after 2 weeks of MSCs last transplantation. Scale bar = 1 mm, *: p < 0.05, **: p < 0.01, ***: p < 0.001, and ns: no significance

Restoration of ovarian structure and function

To further investigate the effects of HF-MSCs transplant on ovarian function, serum sex hormone and ovarian histology were analyzed. Compared with the POF group, two weeks after transplantation, serum FSH decreased, AMH and E2 increased in the P–H and P–U groups of mice (Fig. 4A, B, C, p < 0.05), but there was no significant difference between the P–H and P–U groups (Fig. 4A, B, C, p > 0.05). This indicates that both HF-MSCs and HU-MSCs treatments improved the serum sex hormone levels of the POF mice, and that the improvement effect was comparable between the two. Compared with the POF group, both the P–H and P–U groups showed significantly higher primary follicles, secondary follicles and antral follicles counts, while the atretic follicle count was significantly decreased (Fig. 4D, E, p < 0.05). Moreover, compared with P–U group, the antral follicles of P–H group was significantly higher and the atretic follicles was significantly lower (Fig. 4E, p < 0.05).It is indicating that HF-MSCs and HU-MSCs treatment can improve levels of follicles in POF mice, but HF-MSCs is better in increasing antral follicles and reducing atretic follicles.

Fig. 4

Improvement of ovarian function of mice treated with CTX by HF-MSCs & HU-MSCs transplantation. A, B, C The hormone levels of AMH, FSH, E2 were measured by ELISA on 2 weeks after the last transplantation of MSCs of each group. The error bars indicate SD (n = 6). D Each follicle type per ovary. AF, antral follicle; AtF, atretic follicle; SF, secondary follicle; PF, primary follicle; PdF, primordial follicle (n = 3). E Ovarian histology was analyzed 2 weeks after the last transplantation of MSCs using H&E staining (n = 3). *: p < 0.05, **: p < 0.01, ***: p < 0.001, and ns: no significance. Scale bar = 200 or 500 μm

HF-MSCs improve fertility in POF mice

Two weeks after the last transplantation of HF-MSCs, female mice in each group were mated with fertile males. Two mice in the POF group had no pregnancy within 2 months, and the infertility rate was 40%. The number of F1 pups respectively in CON group, POF group, P–H group and P–U group were 10.8 ± 1.6, 5.0 ± 1.0, 9.8 ± 1.5, 7.8 ± 0.8(Fig. 5). The pups in POF group was significantly lower than that in CON group, P–H group and P–U group (Fig. 5B, p < 0. 05). Moreover, the pups in P–H group was significantly more than P–U group, as 9.8 ± 1.5 vs. 7.8 ± 0.8 (Fig. 5B, p < 0. 05). The results demonstrated that HF-MSCs showed a stronger ability to improve fertility than hHU-MSCs. In addition, mice in the P–H group did not show any abnormal deaths within six months after the last transplantation, and organ autopsy examinations did not show any obvious organ developmental abnormalities (Fig. S1), suggesting that there is no obvious safety risk (e.g., tumorigenesis) associated with the transplantation of HF-MSCs.

Fig. 5

Fertility restoration by HF-MSCs transplantation in mice with CTX-induced POF. Reproductive outcomes in four mating experiments with fertile males. Mating occurred 2 weeks after the last HF-MSCs transplantation. A F1 pups from mice in the CON, POF, P–H and P–U groups. B Mean litter size per pregnant mouse for generation F1 (n = 5). Data represent mean ± SD, *: p < 0.05, **: p < 0.01, and ns: no significance

HF-MSCs changed oxidative stress and granulocyte mitochondrial distribution

To further confirm the underlying mechanisms of transplanted HF-MSCs in POF mice, Our results showed that compared with the POF group, the serum oxidative stress indexes were improved in HF-MSCs-transplanted mice, which significantly showed increased SOD levels, decreased ROS, MDA, and Fe²⁺ levels (Fig. 6A). It is suggested that HF-MSCs improve the ovarian function of POF mice, which may be related to oxidative stress and Ferroptosis. We further observed significant differences in the morphology and number of mitochondria in the granulosa cells of primary follicles by TEM. Specifically, the mitochondria in the POF group showed: reduced number, globular bodies, blurred membranes, severe vacuolization and reduced or even disappeared mitochondrial cristae (Fig. 6B). However, in the P–H group, which underwent HF-MSCs transplantation, we observed improved mitochondrial cristae arrangement and vacuolization, as well as a small number of long rod-like structures (Fig. 6B), and this phenotype was also present in the P–U group as well.

Fig. 6

The changes in serum oxidative stress indexes and mitochondrial structure after HF-MSCs transplantation. A The serum concentration of ROS, MDA, SOD were measured by ELISA, and Fe²⁺ were measured by chemiluminescence on 2 weeks after the last transplantation of MSCs of each group. The error bars indicate SD (n = 6). B Representative image of mitochondrial ultramicro structure from the granulosa cells of primary follicles detected by TEM (yellow arrow: long rod-like structures; red arrow: vacuolization; green arrow: mitochondrial cristae; blue arrow: globular bodies, blurred membranes) (n = 1). Data represent mean ± SD; *: p < 0.05, **: p < 0.01, and ns: no significance. Scale bar: 5 μm (original), 1 μm (enlarged)

HF-MSCs affects the expression of ferroptosis-related genes

We detected the expression of ferroptosis-related mRNA in ovarian tissues by qRT-PCR. Compared with the CON group, the mRNA expression level of KEAP1 was significantly increased in the POF group, while the mRNA expression levels of NRF2 and HO-1 were significantly decreased. Whereas the mRNA expression level of KEAP1 was significantly decreased in the P–H group transplanted with HF-MSCs, while the mRNA expression levels of NRF2 and HO-1 were significantly increased (Fig. 7A, p < 0.001). This situation was similarly in the P–U group after transplantation of HU-MSCs. Further protein level validation by western blotting revealed that the above genes were similarly expressed at the protein level (Fig. 7B, C). Therefore, we concluded that the potential mechanism of HF-MSCs improving ovarian function in POF mice by regulating Ferroptosis through KEAP1/NRF2/HO-1 pathway.

Fig. 7

The effect of transplanted HF-MSCs on cell ferroptosis of ovaries. A The mRNA expression of KEAP1, NRF2 and HO-1in ovarian of the four group mice (n = 3). B The most representative images (Cropped blots) of western blotting for KEAP1, NRF2 and HO-1 in four groups on 2 weeks after MSCs transplantation. Full-length blots/gels are presented in Supplementary Fig. S1. C Quantification of KEAP1, NRF2 and HO-1 expression levels (ratio to GAPDH) (n = 3). Data represent mean ± SD. *: p < 0.05, **: p < 0.01, ***: p < 0.001 and ns: no significance

Discussion

As a common complication of chemotherapy, POF seriously affects the quality of life of premenopausal women. And as tumor patients become younger, the damage to ovarian function by radiation and chemotherapy also affects the fertility of young infertile women [26]. Although several fertility preservation options are available today, it still urgently need improved treatment strategies to prevent chemotherapy-induced premature ovarian failure [9]. In this study, we evaluated for the first time of the restorative effects about HF-MSCs on ovarian function in cyclophosphamide-induced POF mice and compared the effects of HU-MSCs, which are currently more widely used. We found that treatment with HF-MSCs after chemotherapy restored ovarian function and improved fertility, which suggests that HF-MSCs are effective for ovarian protection and fertility preservation after chemotherapy in female cancer patients, and laterally argues the possibility of HF-MSCs to improve the life quality of the POF population in the clinic, providing a new direction for POF treatment.

Previous studies have shown that MSCs derived from umbilical cord, amniotic fluid, embryonic stem cells, bone marrow, adipose tissue, menstrual blood, and skin exhibit improved fertility in different animal models of chemotherapy-induced POF [12, 27, 28], as evidenced by the potential for restoring ovarian function and structure, reducing granulosa cell apoptosis, and restoring folliculogenesis [29, 30]. Furthermore, some of them had been demonstrated in clinical trials, especially HU-MSCs [13, 16, 17]. MSCs has become a hotspot of research in the field of reproductive endocrinology in recent years. Despite the many advantages of MSCs, most of them require invasive manipulation to harvest the source cells and are difficult to obtain a sufficient number of cells for clinical use with ethical risks, heterologous issues.

Adipose- and bone marrow-derived MSCs, for instance, need intrusive manipulation and a restricted number of donors. They also have a limited ability for proliferative senescence and are more likely to be pro-inflammatory when generated from older donors [31,32,33]. In addition, higher passage MSCs are more likely to trigger an innate blood-mediated inflammatory reaction, in vitro and in vivo, that affects the survival and function of MSCs [34]. The HF-MSCs derived from human hair follicles have a rich pool of autologous stem cells and homology, high differentiation capacity, low trauma, multiple access, and no ethical restrictions. Our studies had shown that the growth rate of HF-MSCs during early expansion is significantly higher than that of HU-MSCs (Fig. 2). Based on the above advantages, HF-MSCs were used as the MSCs for our study, while HU-MSCs were used as a positive control. At least 50% of the MSCs remains in the body 10 days after intravenous injection [35]. Our results showed that after injection of MSCs, the mice were imaged in vivo, but not in the ovary, indicating that MSCs did not homing to the ovary one week after injection. WANG et al. research compared the different transplantation methods of MSCs and found that intra-ovarian injection was the best transplantation method, tail-intravenously was the second, and within a month exert the most conspicuous to recovery of ovarian function [12]. Combined with the above factors, we chose tail-intravenously MSCs once a week for 4 consecutive times, and the curative effect was verified 2 weeks after the last injection. Moreover, there is no consensus about the dose of MSCs used in therapy, and dose selection is empirical [36]. Based on our study, 1.0 × 10⁶ MSCs were chose.

Follicular growth and development is a systematic process that requires both hormonal involvement and sophisticated regulation by various cytokines, growth factors and intracellular proteins, and that ovarian stromal cells and granulosa cells (GCs) play an important role in this process [9]. The study revealed that the P–H group had significantly more antral follicles and significantly fewer atretic follicles as compared to the P–U group (Fig. 4E). In addition, there were more litters in HF-MSCs than in HU-MSCs (Fig. 5). It has been shown that HF-MSCs transplantation into mice enhanced the capability for reproduction linked to ovarian reserve, and that HF-MSCs were more capable of restoring ovarian function than HU-MSCs.

However, the exact mechanism by which MSCs ameliorate chemotherapy-induced ovarian damage has not been elucidated. These mechanisms include the release of hormones and proteins through paracrine effects, the transfer of exosomes containing RNA and other molecules, and involvement in cell proliferation and anti-apoptotic processes [10, 14, 15]. CHEN et al. results showed that enhanced antioxidant function and activated Nrf2 signaling significantly improving premature ovarian failure [37]. Wang et al. research indicate that the increasing NRF2 in chemotherapy-injured cells was just compensatory and not enough to resist the accumulated stress. Upregulation of NRF2 could protect granulosa cells against cisplatin via elevating autophagic level [38]. The results of numerous experimental studies demonstrated that administering MSCs locally and systemically efficiently inhibited ferroptosis in wounded neurons, hepatocytes, cardiomyocytes, and nucleus pulposus cells while promoting the survival and regeneration of injured organs [39]. However, it is not clear what role ferroptosis plays in premature ovarian failure. Our results indicated that HF-MSC transplantation enhanced oxidative stress, iron overload, and granulocyte mitochondrial morphology and number in vivo, which in turn enhanced ovarian follicle counts. These findings raise the possibility that blocking ferroptosis may be a pertinent mechanism for using HF-MSCs to treat POF. Mitochondrial function is considered another potential factor affecting oocyte quality [12], and HF-MSCs treatment improved mitochondrial cristae alignment and vacuolization, as well as the appearance of a small number of long rod-like structures (Fig. 6B). Furthermore, the expression levels of NRF2 and HO-1 were elevated, and KEAP1 expression was reduced (Fig. 7). This further suggests that the potential mechanism by which HF-MSCs improve ovarian function in POF mice may be to improve ovarian function in POF mice by inhibiting ferroptosis through modulation of the KEAP1/NRF2/HO-1 pathway (Fig. 8).

Fig. 8

Pattern of improved ovarian function in POF mice by HF-MSCs. HF-MSCs improve the ovarian function of POF mice may by inhibiting ferroptosis of granulocyte through KEAP1/NRF2/HO-1 pathway

Although normal menstruation resumes later in reproductive-age POF patients after chemotherapy, patients are still at risk of early menopause and infertility due to cytotoxic damage. In the present study, our treatment with HF-MSCs improved body weight and ovarian mass in cyclophosphamide-induced POF mice (Fig. 3A, C, D) and significantly increased litter size (Fig. 5). Given the risk of complications use of MSCs, six months after the transplantation of MSCs, we assessed the health status of the mice without abnormal death and organ tumorigenesis, suggesting that HF-MSCs are safe for long-term effects in vivo. As a next step, further and more in-depth studies aiming to better understand the exact mechanisms underlying the therapeutic potential of MSCs are needed, as well as the need to explore the role of HF-MSCs on ovarian function in natural aging animal models. Our results suggest that HF-MSCs treatment can effectively improve ovarian function and fertility after chemotherapy, and the improvement is superior to that of HU-MSCs. our data may provide a clinical translational basis for MSCs in female fertility preservation and ovarian function repair.

Conclusion

HF-MSCs can improve ovarian function in cyclophosphamide-induced POF mice, and it’s effect superior to HU-MSCs. The underlying mechanism may by inhibiting ferroptosis of granulosa cells through KEAP1/NRF2/HO-1 pathway.