Ginsenoside Rg1 enhances stem cell abundance in aging human skeletal muscle after resistance exercise

Abstract

Background

Ginsenoside Rg1 enhances senolytic effect of exercise in human skeletal muscle.

Purpose

To determine the effects of Rg1 on stem cell enrichment and differentiation in skeletal muscle following resistance exercise in women over 60.

Methods

In a randomized, double-blind, placebo-controlled crossover trial, 11 women (60-73 years) completed four sets of seated leg press (70% 1RM) 1 h after ingesting Placebo or Rg1 (10 mg). Muscle biopsies were assessed for activated mesenchymal stem cells (Stro-1⁺/Tom20^high), neural progenitors (Nestin⁺), and vascular progenitors (CD34⁺/CD31⁺) at 0 h and 24 h post-exercise.

Results

While exercise alone did not alter Stro-1⁺ cells and mitochondrial content (TOM20⁺) in muscle tissues, the high mitochondrial Stro-1⁺ cells was decreased by 45% (p < 0.05). Rg1 supplementation doubled high mitochondrial Stro-1⁺ cells and total mitochondrial content post-exercise (p < 0.05). Concentration gradient between the Stro-1⁺ cells and engaged myofibers is consistent with a potential mitochondrial delivery mechanism from bone marrow stem cells. Exercise reduced CD34⁺ cells (−68%, p < 0.05) and Nestin⁺ neural progenitors (−50%, p < 0.05), whereas Rg1 attenuated CD34⁺ cell depletion and increased Nestin⁺ neural progenitors by twofold (p < 0.05). Vascular endothelial progenitors (CD34⁺/CD31⁺) remained unchanged after exercise. Exercise decreased serum estradiol levels (−43%, p < 0.05), whereas Rg1 supplementation restored estradiol to pre-exercise levels and doubled progesterone concentrations.

Conclusion

Rg1 supplementation prevents exercise-induced sex steroid depletion and markedly enhances stem cell abundance in skeletal muscle, which contributes to mitochondrial gains following resistance exercise in women over 60.

Graphical abstract

Pre-exercise ginsenoside Rg1 supplementation enhances activation of bone marrow–derived stem cells in skeletal muscle following exercise in women over 60 years of age.

1. Introduction

Resistance exercise is a potent stimulus for increasing muscle mass and strength, yet its efficacy declines with age. Eccentric contractions during resistance exercise induce myofiber damage [1], triggering acute inflammation in skeletal muscle [2]. During the immune response, myeloid cells infiltrate damaged muscle tissue and promote the clearance of unhealthy senescent cells in a process resembling Darwinian natural selection, which is swiftly followed by muscle regeneration [3,4]. Over time, this injury-induced immune process contributes to muscle rejuvenation, hypertrophy, and strength gains [3,5]. However, the specific bone marrow-derived cell populations that enter exercised human skeletal muscle, particularly for older individuals, are rarely reported.

Stro-1⁺ cells and CD34⁺ cells are widely recognized as bone marrow-derived pluripotent stem cells commonly used in transplantation to regenerate damaged tissues [6,7]. Stro-1⁺ cells are well-characterized mesenchymal stem cells known to promote neural regeneration [8]. CD34⁺ cells are known for their role in regenerating vascular progenitors [7,9]. CD34⁺ cells can be rapidly mobilized into circulation following acute resistance exercise [10], with their abundance in human skeletal muscle increasing within 24 h to support muscle regeneration [4]. Given the short lifespan of vascular endothelial cells (∼2 weeks) [11], continuous replacement of these rapidly aging cells is essential for maintaining healthy microcirculation in muscle tissue.

Recovery from exercise-induced cellular catabolism depends on balanced anabolic signaling. In postmenopausal women, the loss of anabolic steroids is pronounced [12], contributing to muscle atrophy and bone loss [13]. Estrogen and progesterone, the key anabolic steroids, play a central role in the production, mobilization, and differentiation of bone marrow-derived stem cells [14,15]. It is also consumed by skeletal muscle during and after exercise [16] and is critical for maintaining active stem cell populations in skeletal muscle during early postmenopause [17]. In addition, progesterone has been shown to stimulate the proliferation of neural progenitor cells [18], suggesting a broader role for sex steroids in supporting regenerative processes.

Therefore, the first aim of this study was to determine whether exercise can increase the abundance of Stro-1⁺ and CD34⁺ cells in the skeletal muscle of postmenopausal women with low sex steroid conditions. The second aim of the study was to determine whether the phytosteroid Rg1 supplementation can alter stem cell abundance in human skeletal muscle after resistance exercise for the postmenopausal women. Rg1 is a plant-derived steroid with similar structure of sex hormones [19], known to promote bone marrow stem cell proliferation [20] and survival [21]. Recent transplantation studies suggest that bone marrow stem cells donate mitochondria to support tissue repair [22,23]. Thus, we also examine the cellular localization of mitochondria in skeletal muscle for older women after resistance exercise in this study.

2. Materials and methods

2.1. Ethical approval

The study protocol was conducted in compliance with standards set by the updated (2013) Declaration of Helsinki. Ethical approval was obtained from the Institutional Review Board of University of Taipei (IRB-2021-045). Participants were given full explanation of the purpose, experimental procedure, and the potential risks of participation. Written informed consent was received prior to the commencement of the study.

2.2. Participants

A total of 14 sedentary women (aged 60–73 years) with no habitual exercise (less than once per week) were initially recruited through experimental advertisements. Three participants dropped out due to schedule conflicts and were excluded from the study. Eleven participants (mean age 64.8 years, range 60–73 years) completed the study. The minimum sample size was determined using G∗Power software with an effect size of |ρ| = 0.5, α error probability = 0.2, and power (1–β error probability) = 0.8. An additional 10% of participants were recruited to account for potential dropouts.

Inclusion criteria were sedentary women aged over 60 years who engaged in exercise less than once per week. Exclusion criteria were the use of insulin, sex hormones, or anti-inflammatory medications. Most participants were under metformin, statins, and/or antihypertensive medications. After medical clearance, these medications were discontinued 2 days before the exercise session and withheld until 24 h after the biopsy, as antidiabetic medications may influence exercise training responses [24]. All participants were also instructed to discontinue nutritional supplements (vitamins) two days before their visit, continuing until 24 h post-biopsy. Upon arrival, each participant verbally confirmed adherence to these requirements during questioning by the researcher.

2.3. Experimental design

A randomized, double-blind, placebo-controlled, counter-balanced crossover study was conducted with a 3-week washout period between crossover trials. Participants were allocated using block randomization (block size = 2) to determine the order of Placebo and Rg1 conditions. The randomization sequence was generated by an investigator not involved in data collection, and allocation was concealed using sealed opaque envelopes. Blinding was maintained by ensuring that participants, investigators administering the intervention, and outcome assessors were unaware of group allocation. Randomization codes were generated and held by an independent party, interventions were provided in identical coded containers, and all samples and data were analyzed using participant identification numbers, with the allocation revealed only after completion of the primary analysis.

One hour before the exercise challenge, participants received either a Placebo or Rg1 in a counter-balanced order. Baseline sample collection (blood and muscle biopsy) was performed one month prior to the exercise challenge. Baseline blood and muscle biopsies were collected between 09:30 and 11:00 a.m. for all participants. Participants were instructed to refrain from any specific forms of exercise (e.g., mountain climbing or cycling) and to maintain their usual dietary habits for three days prior to the intervention visit.

2.4. Phytosteroid Rg1

Placebo and Rg1 capsules were visually identical and orally administered in a double-blind manner to both participants and experimenters. The Rg1 dosage of 10 mg was based on a previous study [25]. Following a 12-h overnight fast, participants arrived at the testing venue at 08:30 a.m. and were randomized to receive either Placebo or Rg1 in a counter-balanced crossover design. Capsules containing placebo (10 mg cornstarch) or Rg1 (10 mg; NuLiv Science, Inc., Brea, CA, USA) were orally consumed 1 h prior to resistance exercise. They were orally delivered with a beverage containing 8 g of complete protein and approximately 72 kcal, prepared by dissolving 20 g of Orgain Organic Protein (vanilla bean flavor) in 200 mL of warm water (Orgain Inc., Irvine, CA, USA). This vehicle drink provided a protein dose of approximately 0.15-0.2 g kg⁻¹ body weight, intended to supply adequate nitrogen and carbon for post-exercise recovery [26].

2.5. Resistance exercise

Maximal muscle strength (one repetition maximum or 1-RM) for each participant was assessed two weeks prior to the exercise challenge using a validated protocol [27]. Following dynamic lower-body stretching and knee bend exercises, participants performed one warm-up set of 10 repetitions at 50% of their estimated 1-RM (based on perceived capacity), a second set of 5 repetitions at 75%, and a third set of 1 repetition at 90–95%. After a 5-min rest, participants were given 3-5 attempts to reach their actual 1-RM, with weight increased incrementally by approximately 5-10% and 3-5 min rest intervals between attempts. The 1-RM test was administered in small groups by the lead researcher, a qualified personal trainer with 30 years of strength training experience. Proper technique, including controlled contraction speed, breathing, and safe back and knee positioning, was emphasized throughout the assessment. Participants were monitored for signs of pain or discomfort, and motivation was provided by both the instructor and fellow participants. All participants demonstrated maximal voluntary effort during testing. Each exercise session began with a standardized, instructor-led warm-up consisting of body weight squats, static stretching of the quadriceps, hamstrings, and calves, and gentle joint mobilization exercises targeting the ankles, knees, and hips. Participants then performed one set of seated leg press (10 repetitions at 50% of their 1-RM). During all warm-up and exercise sets, participants counted repetitions aloud and received verbal encouragement from both the researcher and a paired partner.

After the warm-up, all participants completed 4 sets of 10 repetitions of the seated leg press at 70% 1-RM, beginning 1 h after Placebo or Rg1 supplementation. Each repetition consisted of a controlled eccentric phase lasting 2-3 s and a maximal-speed concentric phase, performed continuously without pausing between repetitions. Knee flexion reached approximately 80-90°. A 90-s rest period was provided between sets, during which the paired partner performed her set. Rest intervals were recorded and consistently ranged from 90 to 120 s across participants and sets. Seat settings were standardized and maintained across all crossover trials. All individuals achieved the full 10 repetitions for each of the 4 sets on both exercise occasions (Placebo or Rg1 supplemented conditions).

2.6. Muscle biopsy

Muscle biopsies were performed by a certified physician using standard sterile procedures. Local anesthesia (2% lidocaine hydrochloride) was administered prior to the procedure. A 14-gauge Temno® disposable cutting needle (Cardinal Health, McGaw Park, IL, USA) was used to obtain samples from the vastus lateralis muscle at a depth of approximately 3 cm and ∼15 cm proximal to the patella, following the procedure previously described [25]. At each time point, three biopsy samples were collected through a single incision point to ensure sufficient tissue volume (∼15 mg). Samples were immediately transferred into 2 mL-Eppendorf tubes containing 10% neutral buffered formalin, then embedded in paraffin wax blocks by a trained histology technician. Formalin-fixed paraffin-embedded tissues were sectioned and mounted on glass slides for serial sectioning and subsequent immunofluorescent staining.

Baseline biopsies were collected between 09:30 and 11:00 a.m. from all participants, three weeks prior to the exercise challenge. Post-exercise biopsies were collected at two time points: immediately after exercise (0 h) from the right leg (09:30 to 11:00 a.m.), and 24 h post-exercise from the same position on the contralateral leg (left). Immediately following each biopsy, the incision site was compressed with sterile gauze for more than 5 min to minimize the risk of internal bleeding. A topical ice pack was then applied to the skin surface for an additional 10 min.

Following the procedure, participants consumed a standardized lunch and remained at the site to rest for at least 1 h before discharge. Consistency in post-biopsy care and meal content was maintained across all trials. Three days after the procedure, each participant was contacted for a status check. All reported no concerns.

2.7. Serum steroids

Blood samples were collected from the forearm at baseline and 24 h after resistance exercise (∼10 min before muscle biopsy). Samples were drawn into 10-mL glass tubes and allowed to clot on ice for 10 min. They were then centrifuged at 3000 rpm for 15 min at < 4 °C. The resulting serum was manually pipetted into Eppendorf tubes and stored at −76 °C until analysis. Serum samples were analyzed for total and free testosterone, estradiol (E2), progesterone (P4), luteinizing hormone (LH), sex hormone-binding globulin (SHBG), and inflammation marker interleukin-6 (IL-6).

2.8. Immunofluorescence staining

The staining procedures were performed by pathologists at Toson Technology Corporation (Zhubei City, Hsinchu, Taiwan). Formalin-fixed, paraffin-embedded muscle tissues were sectioned at 3 μm thickness, deparaffinized in xylene, and rehydrated through a graded ethanol series (99.9%, 95%, 85%, and 75%) for 2 min at each step.

Cell division in muscle tissue was assessed using 4′,6-diamidino-2-phenylindole (DAPI) staining, which binds strongly to adenine-thymine-rich regions of DNA to visualize nuclei located outside myofibers. Non-dividing nuclei exhibited strong, condensed blue fluorescence, while mitotic nuclei, typically seen in dividing stem cells (i.e. Stro-1+ cells and Nestin + cells), displayed expanding and fainter blue signals due to chromatin dispersion during cell division.

For immunofluorescence co-staining of muscle tissues, sequential staining was performed on each section to detect cell-specific antigens and determine their cellular localization. Primary antibodies were applied to bind target antigens, followed by fluorescence-labeled secondary antibodies targeting the Fc regions of the primary antibodies. Before detecting a second target, tissues were treated with BioTnA's immunoblock reagent (BioTnA, TATS01F, Kaohsiung, Taiwan), which contains multiple IgG affinity proteins, ions, and chelating agents. This reagent effectively masks IgG binding sites, rendering previously bound antibodies unrecognizable and enabling additional rounds of antibody staining. All staining and blocking procedures followed the manufacturer's instructions. After each immunoblock treatment, the staining process could be repeated for the detection of additional targets.

Muscle cross-sections for each participant from baseline and post-exercise samples during Placebo and Rg1 trials were placed on the same glass slide and subjected to immunofluorescence co-staining to detect markers of mesenchymal, vascular, and neurogenic stem cell lineages. Specifically, Stro-1 and TOM20 antigens were used to identify activated mesenchymal stromal cells; co-localization of CD31 with CD34 antigens was used to identify vascular lineage development; and Nestin antigen was used to identify neurogenic lineage development.

The primary and secondary antibodies to detect cell-specific antigens were as follows: CD34: Rabbit anti-human CD34 (ab81289, 1:500, Abcam, Cambridge, UK) with goat anti-rabbit IgG-iFluor 488 (green fluorescence; TAFB02-488, ready to use, BioTnA, Kaohsiung, Taiwan); CD31: Rabbit anti-human CD31 (TACD31-254, 1:200, BioTnA) with goat anti-rabbit IgG-iFluor 670 (pink fluorescence; TAFB02-670, BioTnA); Stro-1: Mouse anti-human Stro-1 (SC-47733-FITC, 1:200, Santa Cruz Biotechnology, CA, USA) with goat anti-mouse IgG-FAM 488 (green fluorescence; TAFB01-488, BioTnA); TOM20: Rabbit anti-human TOMM20 (GTX133756, 1:1000, Abcam/GeneTex, Hsinchu, Taiwan) with goat anti-rabbit IgG-Cy5 670 (pink fluorescence; TAFB02-670, BioTnA); Nestin: Mouse anti-human Nestin (ab18102, 1:200, Abcam) with goat anti-mouse IgG-FAM 488 (green fluorescence; TAFB01-488, BioTnA).

2.9. Image analysis

Stained images were analyzed using the Motic Digital Slide Assistant System Lite 1.0 (Motic Hong Kong Limited, Hong Kong, China) at 40 × magnification. Aggregation of seven or more nuclei within widened spaces surrounding disrupted myofibers was defined as cell infiltration (myofiber lesion). For immunofluorescence, representative images were captured at 20 × magnification using OLYMPUS OlyVIA 3.21 (OLYMPUS, Tokyo, Japan), with scale bars ranging from 200 to 2000 μm. To reduce background interference, slide brightness, contrast, and gamma were manually adjusted. Two independent investigators assessed each sample using predefined criteria and only results with an inter-rater correlation >0.8 were accepted.

Fluorescence images were analyzed using ImageJ (NIH, Bethesda, MD, USA) to quantify positively stained areas. The minimum detectable cell diameter was set at 0.8 μm, based on the average stem cell size (∼1 μm) reported by Ref. [28]. Mitochondria were labeled using the TOM20 antibody (pink fluorescence), showing substantially greater concentration in cells located outside the myofibers. The extramyofibrillar localization of mitochondria labeled by TOM20 antibody was further confirmed using the COX4 antibody in an additional muscle sample (image not reported). Activated mesenchymal stem cells were identified by co-localization of green (Stro-1⁺) and pink (TOM20⁺) fluorescence, along with mitotic features indicated by enlarged and faint DAPI-stained nuclei. Vascular stem cells were identified by the co-localization of green (CD34⁺) and pink (CD31⁺) fluorescence. Neural stem cells (Nestin⁺) appeared as green fluorescence. Nearly all Nestin⁺ cells in muscle tissue and displayed mitotic features similar to Stro-1⁺ cells.

2.10. Statistical analysis

Statistical analyses were performed using SPSS version 27 (IBM, Armonk, NY, USA). In this crossover RCT, each participant was compared with their own baseline values following exercise under both placebo and Rg1-supplemented conditions. Percentage changes from baseline after exercise between the placebo and Rg1 trials were analyzed using paired t-tests. Data are expressed as mean ± standard error (SE). A type I error probability of ≤5% was considered statistically significant, while values between 5% and 10% were regarded as moderately significant. Cohen's d was calculated to estimate effect size in post hoc analyses and interpreted as trivial (0–0.19), small (0.20–0.59), moderate (0.60–1.19), large (1.20–1.99), very large (2.0–4.0), and extremely large (>4.0). Data are presented as mean ± standard error (SE). Statistical significance was set at p ≤ 0.05.

3. Results

3.1. Consort diagram

Fig. 1 illustrates the flowchart of the double-blind, placebo-controlled crossover trial designed to assess the effect of Rg1 supplementation on stem cell abundance in skeletal muscle of women over 60 years following resistance exercise. Fourteen eligible postmenopausal women were initially enrolled; three withdrew due to personal scheduling conflicts. Eleven participants (aged 60–73 years) completed both arms of the crossover trial. Rg1 or Placebo was administered 1 h prior to the exercise session. Serum sex steroid concentrations were measured at baseline (pre-exercise) and 24 h post-exercise. Muscle biopsies were collected at baseline, immediately after exercise (0 h), and 24 h post-exercise.

Fig. 1.

CONSORT diagram of the randomized, placebo-controlled crossover trial.

3.2. Sex steroids

Fig. 2 illustrates the structural similarity between the phytosteroid Rg1 and key sex steroids. Resistance exercise significantly increased serum IL-6 levels by 45% compared to baseline (d = 0.6, p = 0.08), indicating physiological stress induced by the exercise protocol. Table 1 presents serum sex steroid profiles measured 24 h post-exercise. Exercise alone resulted in a 45% reduction in serum estradiol relative to pre-exercise baseline (d = 0.55, p = 0.06). This decline was completely reversed by Rg1 supplementation. While resistance exercise had no significant effect on circulating progesterone, Rg1 supplementation doubled progesterone levels above pre-exercise values (d = 0.68, p = 0.14). No significant changes in LH, testosterone, or SHBG were observed with either exercise or Rg1 supplementation.

Fig. 2.

Structural similarity between the phytosteroid Rg1 and endogenous sex hormones.

Table 1.

Effects of resistance exercise on sex steroids and IL-6. Serum was measured at baseline and 24 h post-exercise under placebo and Rg1-supplemented conditions. Data are presented as mean ± SE, along with the percent change from baseline. †p ≤ 0.1 and ∗ p ≤ 0.05 indicate significant differences from the pre-exercise baseline. Abbreviations: LH, luteinizing hormone; SHBG, sex hormone-binding globulin; IL-6, interleukin-6.

	Baseline	Placebo	Change	Rg1	Change
Testosterone (ng/mL)	0.24 ± 0.05	0.22 ± 0.04	-	0.24 ± 0.05	-
Free testosterone (pg/mL)	3.9 ± 0.8	3.7 ± 0.8	-	3.9 ± 0.8	-
Estrogen (pg/mL)	15.6 ± 3.2	8.8 ± 1.8	−43% †	17.8 ± 2.8	-
Progesterone (ng/mL)	0.14 ± 0.03	0.16 ± 0.03	-	0.27 ± 0.08	+92 % ∗
LH (mIU/mL)	22.7 ± 3.6	25.1 ± 3.3	-	22.3 ± 3.0	-
SHBG (nmol/L)	45 ± 8	43 ± 7	-	46 ± 8	-
IL-6 (pg/mL)	1.26 ± 0.24	1.83 ± 0.35	+45% †	1.60 ± 0.38	-

3.3. Stro-1⁺ mesenchymal stem cells in human skeletal muscle

Fig. 3 illustrates the morphological characteristics of Stro-1⁺ cells in skeletal muscle tissue. These cells are predominantly located around myofibers, particularly concentrated at sites of muscle damage, such as split or disrupted regions between fibers. Nearly all Stro-1⁺ cells exhibit features of mitotically active stem cells, as indicated by expanded and diffuse DAPI staining, in contrast to the condensed nuclear staining typical of non-dividing cells. This provides direct evidence of mitotically active stem cells in the sites of injured muscle tissue.

Fig. 3.

Stro-1⁺ mesenchymal stem cells in skeletal muscle of women aged over 60 years. (A) Representative images showing Stro-1⁺ stem cell infiltration in damaged skeletal muscle tissue immediately following resistance exercise. Dividing cells are identified by enlarged, faint nuclei stained with DAPI (blue) in contrast to solid nuclei, indicating active proliferation in areas of muscle regeneration. Stro-1⁺ cells (green fluorescence) also exhibit highly concentrated mitochondrial content in sharp contrast to myofibers, indicated by TOM20⁺ staining (pink fluorescence). The Rg1-supplemented condition displayed a substantially greater infiltration of Stro-1⁺ cells and higher mitochondrial density compared to the placebo. (B) Semi-quantitative analysis of cell marker expression during the 24-h recovery period. ∗p < 0.05 indicates a significant difference between Placebo and Rg1.

Most Stro-1⁺ cells demonstrate markedly higher mitochondrial content compared to the surrounding myofiber cytoplasm as revealed by TOM20 antibody labeling. This mitochondrial enrichment is a defining feature of activated, rather than quiescent, stem cells [29]. In some cases, the high mitochondrial Stro-1⁺ cells are fused into the subsarcolemmal region of myofibers showing much lower mitochondria concentration, with visual evidence of a mitochondrial transfer.

Fig. 3B illustrates the semi-quantitative analysis of immunofluorescence staining of skeletal muscle response of Stro-1⁺ cells to resistance exercise in postmenopausal women aged over 60. The total number of Stro-1⁺ cells in skeletal muscle remained unchanged at 0 h and 24 h post-exercise; however, the high-mitochondria subfraction of Stro-1⁺ cells decreased by 50% (d = 0.54, p = 0.1) immediately following resistance exercise and failed to return in 24 h recovery period. In contrast, Rg1 supplementation administered prior to exercise significantly increased Stro-1⁺ cell abundance by 114% (d = 0.57, p = 0.08) relative to pre-exercise baseline levels. This 93% increase from pre-exercise baseline was primarily driven by an elevation in the high-mitochondria subfraction of Stro-1⁺ cells (d = 1.0, p = 0.14) compared to the Placebo-supplemented condition. Resistance exercise alone did not significantly affect total mitochondrial content in skeletal muscle. However, Rg1 supplementation markedly enhanced mitochondrial content post-exercise, resulting in a 173% increase (d = 0.8, p = 0.03).

3.4. Nestin ⁺ neural stem cells in human skeletal muscle

Stro-1⁺ bone marrow-derived mesenchymal stem cells can differentiate into Nestin⁺ neural stem cells. Accordingly, we examined Nestin⁺ neural stem cells in human skeletal muscle following exercise, as shown in Fig. 4. Nearly all Nestin⁺ cells in muscle tissue were located near the extramyofibrillar border of damaged skeletal myofibers. Similar to Stro-1⁺ cells, almost all Nestin⁺ cells exhibited features of active mitosis, characterized by expanded and faint DAPI staining compared with the dense nuclei of non-dividing cells, visible in magnified immunofluorescence images (Fig. 4A). Notably, Nestin⁺ cells exhibiting active mitosis were more frequently observed in cell-infiltrated regions of exercised skeletal muscle following pre-exercise Rg1 supplementation compared with Placebo. Fig. 4B illustrates the immunofluorescence staining indicating responses of Nestin⁺ neural stem cells in exercised skeletal muscle under both Placebo- and Rg1-supplemented conditions.

Fig. 4.

Nestin ⁺ neural stem cells in skeletal muscle of women aged over 60 years. (A) Representative images showing Nestin⁺ stem cells in damaged skeletal muscle immediately following resistance exercise. Dividing Nestin⁺ cells are identified by enlarged, faint DAPI-stained nuclei (blue), contrasting with intact, solid nuclei, and indicating active neural regeneration. Rg1 supplementation markedly increased Nestin⁺ cell-mediated neural regeneration compared to the placebo condition, which showed a decreased amount of these cells immediately post-exercise. (B) Nestin⁺ cells change during the 24-h recovery period. ∗p < 0.05 indicates a significant difference between Placebo and Rg1.

3.5. Vascular progenitor cells (CD34⁺/CD31⁺) in human skeletal muscle

CD34⁺ cells are a well-characterized population of bone marrow-derived stem cells and contribute to the development of vascular progenitor cells (CD34⁺/CD31⁺) following exercise [7,9]. Therefore, we examined CD34⁺/CD31⁺ cells in the skeletal muscle of postmenopausal women, as shown in Fig. 5. CD34⁺ and CD31⁺ cells were more frequently colocalized in the cell-infiltrated areas of skeletal muscle (Fig. 5A). CD34⁺ cell numbers decreased by 68% 24 h after resistance exercise (d = 1.96, p < 0.01), whereas CD34⁺/CD31⁺ and CD31⁺ cell levels remained stable (Fig. 5B). Pre-exercise Rg1 supplementation completely reversed the depletion of CD34⁺ cells and modestly increased CD31⁺ cells above baseline levels by +33% (d = 0.53, p = 0.07).

Fig. 5.

CD34⁺/CD31⁺ cells in skeletal muscle of women aged over 60 years following resistance exercise. (A) Representative images showing CD31⁺ vascular endothelial cells (pink fluorescence) colocalized with CD34⁺ cells (green fluorescence) at 24 h post-exercise, forming a circular structure feature of capillary regeneration in damaged muscle tissue. Comparisons are shown for both Placebo- and Rg1-supplemented conditions. (B) Semi-quantitative analysis of CD34⁺ and CD31⁺ cell marker expression during the 24-h recovery period. †p < 0.05 indicates a significant difference compared to pre-exercise baseline.

4. Discussion

In women aged over 60 years, the loss of anabolic sex steroids may compromise recovery from exercise-induced catabolic stress in skeletal muscle [30]. In this study, we asked the question whether phytosteroid Rg1 supplementation can influence sex hormones, bone marrow stem cells, and mitochondrial content in the skeletal muscle of women aged 60-73 following an acute bout of resistance exercise. The key findings are as follows: 1) Rg1 supplementation reverses exercise-induced estrogen depletion and doubles progesterone in blood for the postmenopausal women; 2) Resistance exercise depletes both CD34⁺ cells and Stro-1⁺/Tom20^high cells within human skeletal muscle of the postmenopausal women. To the contrary, Rg1 supplementation increases Stro-1⁺ cell abundance (including Stro-1⁺/Tom20^high subfraction) and minimizes CD34⁺ cell depletion after exercise; 3) Resistance exercise acutely decreases Nestin⁺ neural stem cells in human skeletal muscle of the postmenopausal women, whereas Rg1 supplementation doubles the Nestin⁺ neural stem cells in human skeletal muscle after resistance exercise; 4) Resistance exercise alone failed to stimulate mitochondrial gain in skeletal muscle of the postmenopausal women. Rg1 substantially increased mitochondria content in skeletal muscle after resistance exercise.

Stem cell numbers in muscle tissue reflect the balance between demand and supply of regenerative sources from the bone marrow. Therefore, our new findings implicate that pre-exercise Rg1 supplementation increases stem cell supply from bone marrow to match an increased rate of stem cell differentiation in exercised skeletal muscle for postmenopausal women.

Exercised skeletal muscle is known to consume endogenous steroids and decrease circulating estrogen in postmenopausal women [16]. In this study, we provide the first human evidence showing the reversal of exercise-induced estrogen decline and elevated circulating progesterone following pre-exercise Rg1 supplementation in the postmenopausal women. It remains unknown as to how Rg1 prevents the exercise-induced decline in estrogen and increases progesterone. With similar chemical backbone to sex steroids, Rg1 has been shown to exert estrogen-like actions through receptor binding in vitro [31]. Furthermore, we could not preclude the possibility that the competitive binding of steroidal moiety of Rg1 metabolites to SHBG against the sex steroids allows them to be more available in circulation. The results of the study in increasing progesterone and preventing estrogen loss may be valuable for postmenopausal women with exercise habits to maintain stability of circulating sex steroids.

Sex steroids are required for normal production, mobilization, and differentiation of bone marrow-derived stem cells, including mesenchymal stromal cells and hematopoietic stem cells [14,15]. Stro-1⁺ cells are the stromal cells known to originate from bone marrow [32]. For postmenopausal women, the effects of resistance exercise on Stro-1⁺ cells in human skeletal muscle has not been previously reported. In this study, we observed a decreased high-mitochondria subfraction of Stro-1⁺ stem cells in exercised skeletal muscle during recovery. High-mitochondria stem cells are considered activated and mobilizable, unlike their quiescent state marked by high stemness and low mitochondrial activity [29]. Stem cells from bone marrow are now known as mitochondria donors to peripheral recipients, widely reported from in vitro and transplantation studies [33]. In this study, we further provide in vivo evidence supporting the potential role of Stro-1+ cells as mitochondrial donors for replenishing exercised human skeletal muscle.

CD34⁺ cells are primarily identified as hematopoietic stem cells originated from bone marrow [7]. For normal young individuals, exercise is known to cause a transient release of CD34⁺ cells from bone marrow into circulation which seeds and expands in human skeletal muscle in 24 h, suggesting an immediate demand to bone marrow for replenishing the local stem cell reserve in skeletal muscle [34]. In this study, we determined whether increased CD34⁺ cells in skeletal muscle following exercise can occur in sex steroid-compromised women, similar to young women [34]. Here, we have observed an acute decrease in CD34⁺ cells suggesting an increased stem cell demand of challenged muscle from bone marrow in the postmenopausal women. Unchanged levels of CD34⁺/CD31⁺ and CD31⁺ cells in both Placebo and Rg1 supplemented conditions indicates vascular progenitor cell demand for recovery is being met in cost of CD34⁺ cells following an acute bout of resistance exercise in the postmenopausal women.

The decreasing trend of CD34⁺ cells and Stro-1⁺ cells in exercised muscle implicate increased stem cell demands for neural network rewiring and blood vessel reconstruction on disrupted myofibers after an acute bout of resistance exercise. This acute response is essential to restore voluntary contractile function of newly regenerated myofibers while maintaining adequate delivery of oxygen, nutrients, and circulating humoral factors.

The skeletal muscle response to exercise in these postmenopausal women is markedly altered when Rg1 is supplemented 1 h prior to exercise. The most intriguing finding of the study is that Rg1 supplementation increased Stro-1⁺ cells and Nestin⁺ neural stem cells together with attenuating CD34⁺ cell depletion in skeletal muscle. This finding suggests that Rg1 enhances bone marrow stem cell mobilization to exercised skeletal muscle. For Stro-1⁺ cells, the Rg1-induced increase is mainly contributed by high mitochondria subfraction of Stro-1 bone marrow stem cells. The exercise response of Nestin⁺ neural stem cells was similar to Stro-1⁺ cells in Rg1-supplemented condition, suggesting that the Stro-1⁺ cells contribute to differentiation of the neural progenitors after exercise-induced muscle damage.

The underlying mechanism accounted for the Rg1 effect in observed stem cell abundance to exercised muscle requires more studies. We speculate that this response is associated with preventing estrogen depletion after exercise, implicated by previous findings of sex steroid effect on stem cell proliferation, mobilization, and differentiation [14,15]. Postmenopausal women aged over 60 are considered sex hormone compromised individuals.

Most Stro-1⁺ and Nestin⁺ cells observed in the muscle tissues of older women exhibited expanded nucleus morphologies with faint DNA staining, in contrast to most infiltrating cells, which showed condensed and intense DAPI staining, indicative of active regeneration. These findings suggest that Rg1 supplementation promotes the mobilization of bone marrow-derived stem cells into skeletal muscle and cell regeneration (mitosis) in response to exercise in older women. This finding may help explain the muscle-rejuvenating effects of Rg1 previously reported in young men following endurance exercise [25].

Most previous studies on exercise-induced mitochondrial gains in skeletal muscle relied on whole-muscle homogenates for analysis, which do not distinguish the cellular localization of mitochondria within parenchymal versus supporting cells. Recent in vitro studies have already demonstrated that stem cells can act as mitochondrial donors, aiding the restoration of function in injured tissues [22,23].

The present study provides the first in vivo evidence, using immunofluorescence co-staining, that Stro-1⁺ cells contain a markedly higher mitochondrial concentration compared with myofibers. The observed mitochondrial diffusion gradient suggests that the increased mitochondria in muscle tissues following Rg1 supplementation is primarily driven by the infiltration and differentiation of Stro-1⁺ cells into damaged myofibers. In contrast, in the exercise-only (placebo) condition, the absence of mitochondrial enhancement is explained by the unchanged abundance of Stro-1⁺ cells in skeletal muscle.

While mitochondria are well recognized for their role in aerobic ATP production, bone marrow stem cells predominantly rely on anaerobic glycolysis [35]. Therefore, the mitochondria-rich Stro-1⁺ cells are likely function as a mitochondrial donor to replenish contracting myofibers in muscle tissue.

Reduced sex hormone levels lead to the accumulation of dysfunctional mitochondria due to diminished mitophagy, as shown in animal studies [36]. In postmenopausal women, the increase in Stro-1⁺ stem cells (particularly in mitochondrial rich cells) in skeletal muscle following Rg1 supplementation is likely linked to the stabilization of sex hormone levels in response to exercise. Estrogen and progesterone play pivotal roles in the production, mobilization, and function of bone marrow-derived stem cells, and are essential for effective muscle repair after exercise [14,15]. Tissue repair depends on a sufficient supply of bone marrow-derived stem cells, which are better maintained when sex hormone levels are optimized during early life.

Contracting myofibers present a challenging environment for mitochondrial survival, as mitochondria have a relatively short half-life of approximately two weeks and are rapidly removed via mitophagy during exercise [37]. In contrast, mitochondrial biogenesis, requiring DNA replication and protein synthesis, occurs over a much longer timescale [38]. Therefore, the rapid increase in mitochondrial content observed in human skeletal muscle following an acute bout of exercise [39] cannot be fully explained by mitochondrial biogenesis within myofibers alone.

Findings from the current study suggest an additional, plausible mechanism: mitochondria may be, at least in part, produced in the bone marrow and promptly mobilized to contracting myofibers in response to damage-induced renewal signals. This finding is consistent with a recent clinical study in young men, demonstrating that human myofibers can acquire mitochondria via cell fusion [40]. This potential stem cell-mediated mitochondrial transfer mechanism may be critical for maintaining a young, functional mitochondrial pool, thereby ensuring efficient ATP production in myofibers, particularly under conditions of exercise-induced muscle damage or aging.

4.1. Limitation

All participants in this study were postmenopausal. It is important to note that these women were essentially self-selected for higher fitness, as they had already survived to 60–73 years of age, unlike those who experienced earlier mortality. They were screened according to specific inclusion criteria, resulting in a relatively healthier subset of their age group. Therefore, the findings of this study may not be fully generalizable to frailer women of similar age. Among the 11 participants, considerable variation in physical aging was observed despite similar chronological ages, indicating individual differences in the rate of aging. In addition, resting hemodynamic responses following Rg1 supplementation were not assessed, making it difficult to determine whether the observed changes reflect direct Rg1-mediated cellular effects or an amplification/modulation of normal exercise-induced responses. Incorporating these measurements in future studies would provide clearer mechanistic insight into Rg1 delivery to skeletal muscle.

Another limitation of this study is the small sample size, underscoring the need for larger studies to confirm these findings. Additionally, stem cells are not the only bone marrow–derived cells capable of carrying mitochondria. Recent studies have identified several immune cell types enriched with mitochondria [41,42], opening an exciting avenue for future research.

4.2. Conclusion

The results of this study suggest that the phytosteroid Rg1 effectively enhances the mobilization of stem cells (Stro-1⁺ and CD34⁺) to exercised skeletal muscle in postmenopausal women with low sex steroid levels. This acute response may account for the rapid mitochondrial gains observed following Rg1 supplementation, likely driven by stem cell trafficking into damaged myofibers. The adaptive benefits of exercise in this context appear to be linked to the sex steroid-stabilizing effects of Rg1. The outcomes are exploratory and require confirmation in larger mechanistic and clinical studies.

Ethics statement

Institutional ethical approval was obtained from the Institutional Review Board of University of Taipei (approved number: IRB-2021-045). The protocol for this study was written in compliance with standards set by the Declaration of Helsinki. Participants were given full explanation of the purpose, experimental procedure, and the potential risks of participation. Written informed consent was received prior to the commencement of the study. The study was registered with clinicaltrials.gov (NCT07093190). The study was carried out in accordance with current ethical standards of Declaration of Helsinki.

Declaration of competing interest

Professor Kuo received a joint grant from the National Science and Technology Council, Taiwan, and Nuliv Science, USA (NSTC 110-2622-H-845-004). Nuliv Science, USA, is a nutraceutical company involved in the development of the related product Senactiv®. All other authors declare no conflicts of interest.