Sex-specific physical activity thresholds influencing muscle function in older adults with XX genotype of ACTN3 R577X

Abstract

Physical activity is widely recommended to maintain muscle health in older adults, yet its effects may vary depending on genetic background. The ACTN3 R577X polymorphism leads to α-actinin-3 deficiency in XX homozygotes and may influence skeletal muscle performance. We examined the association between physical activity and muscle mass and function in 682 community-dwelling adults aged ≥ 65 years who participated in annual health examinations between 2021 and 2024. Skeletal muscle mass index was assessed using dual-energy X-ray absorptiometry, while grip strength, gait speed, and five-time chair stand performance were evaluated according to Asian Working Group for Sarcopenia criteria. Physical activity was quantified using metabolic equivalents derived from the International Physical Activity Questionnaire. No significant association was observed between ACTN3 R577X genotype and sarcopenia components. However, in women with ACTN3 XX, physical activity exceeding 441 MET-min/week was significantly associated with poorer chair stand performance (β: 4.19, 95% CI: 0.98 ~ 7.40), revealing a U-shaped relationship between physical activity and physical performance. Sex-specific physical activity thresholds may exist in older adults with the ACTN3 XX genotype, with physical activity above 441 MET-min/week being associated with poorer muscle function in women. This specific MET value is an exploratory finding, requiring more external validation.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-57340-5.

Introduction

Sarcopenia is characterized by progressive declines in muscle mass, strength, and physical performance¹, affecting about 10% of older adults and nearly half of those aged over 80 years^2,3. It markedly increases the risk of functional decline, disability, and mortality⁴. Its development is influenced by multiple factors, including aging^5,6, sex^7,8, ethnicity⁹, lifestyle^10–12, medication use^13,14, comorbidities¹⁵, and genetic factors¹⁶.

Physical activity is one of the most effective strategies to prevent sarcopenia^17,18. However, its effects vary among individuals. Older adults carrying the ACTN3 XX genotype tend to demonstrate smaller gains in muscle function after resistance or combined training than those carrying the RR or RX genotypes^19,20. Moreover, an interaction between ACTN3 R577X and physical activity has been reported, where XX carriers who exercised less than four days per week showed greater muscle loss²¹. These findings imply that genetic background plays a role in modifying the effects of physical activity.

The heritability of sarcopenia-related traits is estimated to be around 50%^22–25. Age-related muscle decline is mainly driven by the atrophy of type II fibers^26,27. The protein α-actinin-3 is specifically expressed only in type II fibers²⁸, which is encoded by ACTN3, forms part of the Z-line protein that stabilizes actin filaments and regulates force transmission during contraction²⁹. The ACTN3 R577X polymorphism (rs1815739) produces a premature stop codon and consequent deficiency of α-actinin-3 in XX homozygotes^30,31, increasing the risk of sarcopenia in the elderly^32,33.

Approximately 20% of Asians carry the XX genotype³⁴. Thus, examining the effects of physical activity on muscle function in older adults with this genotype is important. To our knowledge, no study has examined the relationship between metabolic equivalent–based physical activity (MET-min/week) and sarcopenia components among ACTN3 R577X genotypes in older adults. Therefore, this study aimed to explore the potential physical activity threshold that minimizes muscle-related risks in individuals carrying the ACTN3 XX using MET-min/week.

Materials and methods

Study participants

This study adopted a cross-sectional research design. We enrolled community-dwelling adults aged ≥ 65 years who participated annual health examinations from March 2021 to October 2024. A structured questionnaire was used to gather data on demographic characteristics (including age, BMI, marital status, and education level), medical history, medication use (including osteoporosis, hypertension, diabetes, corticosteroid use, and vitamin D supplementation), and lifestyle factors (including smoking status and physical activity). This study was approved by the Institutional Review Board of Tri-Service General Hospital (TSGHIRB Number: 2-102-05-028 and A202205093), was conducted in compliance with the ethical standards of national and institutional committees on human research and the Helsinki Declaration. Written informed consent was obtained from each participant.

Assessment of muscle mass and function

Sarcopenia was defined according to 2019 Asian Working Group for Sarcopenia (AWGS) criteria, including skeletal muscle mass index (SMMI), muscle strength and physical performance (including gait speed and five-times sit-to-stand time)¹. Appendicular skeletal muscle mass (ASM) was measured using Dual energy X-ray absorptiometry (DXA; Prodigy Series X-Ray Tube Housing Assembly, GE Medical Systems Lunar 3030 Ohmeda Dr Madison, Wisconsin, USA). The skeletal muscle mass index (SMMI) was calculated as ASM divided by height squared (ASM/height², kg/m²). The grip strength of the participants’ dominant hand was assessed three times via an analogue isometric dynamometer (Exacta™ Hydraulic Hand Dynamometer; North Coast Medical Inc., Gilroy, CA), with the highest value recorded. For gait speed, participants walked 10 m at their usual pace, and gait speed was calculated based on the time required to walk the middle 6 m. For chair stand time, the participant was instructed to sit with arms folded across the chest and their back against a chair. They were asked to stand up and sit down five consecutive times and avoid touching the back of the chair during the test. The time required to complete all five repetitions was recorded in seconds^35,36. All the assessments were performed on the same day.

Assessment of physical activity

Physical activity was quantified using metabolic equivalent of task (METs). MET values were derived from the Chinese version of the International Physical Activity Questionnaire (IPAQ)³⁷. The IPAQ consists of seven items that assess the frequency (days per week) and duration (minutes per day) of vigorous-intensity physical activity, moderate-intensity physical activity, walking (classified as moderate-intensity activity) and sitting time over the previous seven days. For each physical activity category, MET values were assigned according to IPAQ scoring protocol (vigorous = 8.0 METs, moderate = 4.0 METs, walking = 3.3 METs). Total physical activity was expressed as MET-minutes per week.

Genomic DNA extraction and genotyping

Genomic DNA was extracted from 5 mL of peripheral blood of participants using the MagPurix^® Kit (Zinexts Life Science Corp., New Taipei City, Taiwan) according to the manufacturer’s instructions. Genotyping of the ACTN3 R577X polymorphism was performed using the iPLEX^® HS panel for the MassARRAY^® System (Agena Bioscience, San Diego, CA, USA)³⁸. To ensure accuracy, 10% of the samples were randomly selected for repeat genotyping, yielding a concordance rate of 99%.

Covariates

Covariates in this study included demographic and clinical characteristics obtained through standardized self-administered questionnaires, including sex (male or female), age (≥ 65 years), marital status (married or unmarried), and educational level (≤ 12 years or > 13 years). Information on comorbidities (e.g., hypertension, diabetes) and medication use (e.g., corticosteroids, vitamin D supplements) was also collected. Height and weight were measured, and body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared (kg/m²). The presence of osteoporosis was determined based on T-scores derived from dual-energy X-ray absorptiometry (DXA) measurements.

Statistical analysis

Continuous variables were presented as mean ± standard deviation (SD) and compared using t-tests and ANOVA. Categorical variables were presented as counts (percentages) and compared using Chi-square test. The associations between physical activity and sarcopenia components were analysed using generalized linear models (GLMs) stratified by ACTN3 gene polymorphisms. Covariates were selected based on variables that differed significantly between participants with and without sarcopenia (Additional file 1: Table S1). Furthermore, to evaluate the potential nonlinear relationship between physical activity and sarcopenia components, restricted cubic splines (RCS) was conducted. The optimal number of knots in the RCS models was determined based on the Akaike Information Criterion (AIC)³⁹; three knots provided the best model fit. All statistical analyses were conducted using R version 4.3.2. A two-sided p-value < 0.05 was considered statistically significant.

Results

Participants characteristics

A total of 682 older adults were included with a mean age of 73.34 ± 7.02, and 41.8% were men. The prevalence of sarcopenia was 17.8%. Comorbidities included hypertension (38.0%), diabetes (13.7%), and osteoporosis (9.7%). Long-term corticosteroid use and vitamin D supplementation were reported by 4.5% and 26.9% of participants, respectively. The mean physical activity was 726.13 ± 476.73 MET-min/week. Mean SMMI was 6.64 ± 1.05 kg/m² in men and 5.83 ± 0.82 kg/m² in women. Mean grip strength was 28.27 ± 9.64 kg in men and 24.38 ± 8.86 kg in women. Average gait speed was 1.11 ± 0.32 m/s, and the mean for chair stand time was 12.38 ± 4.88 s (Table 1).

Table 1.

Participants characteristics.

Characteristica	Overall (N = 682)	Men (N = 277)	Women (N = 405)
Age, year	73.34 ± 7.02	74.50 ± 7.74	72.55 ± 6.38
BMI, kg/m2	24.38 ± 3.60	25.13 ± 3.10	23.87 ± 3.82
Married	116 (17.9%)	12 (4.5%)	104 (27.2%)
Education
≤ 12 years	350 (54.1%)	90 (34.1%)	260 (67.8%)
> 13 years	297 (45.9%)	174 (65.9%)	123 (32.1%)
Hypertension	252 (38.0%)	120 (44.0%)	132 (33.8%)
Diabetic	91 (13.7%)	42 (15.4%)	49 (12.6%)
Osteoporosis	64 (9.7%)	1 (0.4%)	63 (16.0%)
Corticosteroid usage	29 (4.5%)	8 (3.0%)	21 (5.5%)
Vitamin D usage	174 (26.9%)	65 (24.6%)	109 (28.5%)
Physical activityb	726.13 ± 476.73	844.08 ± 506.00	642.56 ± 436.63
ACTN3 R577X genotype
RR + RX	545 (82.6%)	219 (82.0%)	326 (83.0%)
XX	115 (17.4%)	48 (18.0%)	67 (17.0%)
Sarcopenia	104 (17.8%)	53 (21.7%)	51 (15.0%)
SMMI, kg/m2	6.50 ± 1.05	7.43 ± 0.76	5.84 ± 0.65
Grip strength, kg	27.39 ± 9.39	35.62 ± 8.03	21.74 ± 5.10
Gait Speed, meters/sec	1.11 ± 0.32	1.13 ± 0.34	1.09 ± 0.30
Five-time chair stand time, sec	12.38 ± 4.88	11.97 ± 4.94	12.66 ± 4.83

^a Values are means (SD) for continuous variables and percentages for categorical variables.

^b Physical activity was expressed as MET-minutes/week.

SMMI: Skeletal muscle mass index, MET: Metabolic Equivalent.

Association between ACTN3 R577X polymorphism and sarcopenia components

The genotype distribution of ACTN3 R577X was 36.5% (RR), 43.3% (RX) and 16.9% (XX), consistent with Hardy–Weinberg equilibrium (p = 0.061) (Table 2). The minor allele frequency (39.8%) was consistent with the 1000 Genomes Project (40%)³⁴. Among men, the prevalence of sarcopenia was not significantly different between RR + RX carriers (23.3%) and XX carriers (19.0%) (p = 0.549), with no significant differences observed for each sarcopenia component. Consistently, GLM analyses showed no significant associations between the ACTN3 R577X genotype and SMMI, grip strength, gait speed, or chair-stand time after adjustment for age and BMI. Similarly, among women, the prevalence of sarcopenia was not significantly different between RR + RX carriers (14.4%) and XX carriers (19.3%) (p = 0.349), with no significant differences observed for each sarcopenia component. GLM analyses confirmed that the ACTN3 R577X polymorphism was not significantly associated with any measure of muscle mass or function in women after adjustment (Table 3).

Table 2.

Comparison of sarcopenia components according to ACTN3 R577X genotype stratified by sex.

	Men			Women
Sarcopenia components	RR + RX (n = 193)	XX (n = 42)	p-value	RR + RX (n = 271)	XX (n = 57)	p-value
Sarcopenia			0.549			0.349
No	148 (76.7%)	34 (81.0%)		232 (85.6%)	46 (80.7%)
Yes	45 (23.3%)	8 (19.0%)		39 (14.4%)	11 (19.3%)
SMMI (Kg/m2)	7.44 ± 0.75	7.32 ± 0.75	0.322	5.84 ± 0.63	5.86 ± 0.71	0.865
Grip strength (Kg)	35.42 ± 8.15	35.26 ± 7.44	0.910	21.54 ± 5.11	22.77 ± 5.23	0.089
Gait speed (Meters/sec)	1.13 ± 0.35	1.14 ± 0.28	0.902	1.09 ± 0.31	1.11 ± 0.27	0.604
Five-time chair stand time (sec)	12.00 ± 5.02	11.57 ± 4.40	0.589	12.74 ± 5.56	12.83 ± 4.13	0.900

Hardy–Weinberg equilibrium (p = 0.061).

Table 3.

Association between ACTN3 R577X genotype and sarcopenia stratified by sex.

	Crude- β (95% CI)	p-value	Adj-βa (95% CI)	p-value	Crude-β (95% CI)	p-value	Adj-βa (95% CI)	p-value
SMMI (Kg/m2)
RR + RX	Ref		Ref		Ref		Ref
XX	− 0.12 (− 0.36 ~ 0.12)	0.322	− 0.02 (− 0.19 ~ 0.15)	0.835	0.02 (− 0.16 ~ 0.19)	0.865	0.02 (− 0.11 ~ 0.15)	0.799
Grip strength (Kg)
RR + RX	Ref		Ref		Ref		Ref
XX	− 0.15 (− 2.83 ~ 2.53)	0.910	− 0.17 (− 2.65 ~ 2.31)	0.895	1.24 (− 0.18 ~ 2.65)	0.089	1.24 (− 0.11 ~ 2.58)	0.072
Gait speed (Meters/sec)
RR + RX	Ref		Ref		Ref		Ref
XX	0.01 (− 0.10 ~ 0.11)	0.902	− 0.01 (− 0.12 ~ 0.09)	0.789	0.02 (− 0.06 ~ 0.10)	0.604	0.03 (− 0.05 ~ 0.10)	0.519
Five-time chair stand time (Sec)
RR + RX	Ref		Ref		Ref		Ref
XX	− 0.43 (− 2.00 ~ 1.13)	0.589	− 0.28 (− 1.78 ~ 1.22)	0.712	0.09 (− 1.34 ~ 1.52)	0.900	0.06 (− 1.30 − 1.43)	0.927

For binary outcomes (sarcopenia), odds ratios (OR) with 95% CI are reported, regression coefficients (β) with 95% confidence intervals (CI) are shown.

For continuous outcomes (SMMI, grip strength, gait speed, and five-time chair stand time).

^aAll result of Adj-beta were adjusted by age, BMI.

Association between physical activity and sarcopenia components

As shown in Table 4, higher physical activity was significantly associated with greater muscle mass and better function. In men, higher physical activity was associated with reduced risk of sarcopenia (adj-OR: 0.62; 95% CI: 0.42 ~ 0.91), greater SMMI (adj-β: 0.13; 95% CI: 0.07 ~ 0.20), faster gait speed (adj-β: 0.08; 95% CI: 0.04 ~ 0.12) and shorter chair stand time (adj-β: − 1.04; 95% CI: − 1.83 ~ − 0.26). Similar associations were observed in women for SMMI (adj-β: 0.09; 95% CI: 0.02 ~ − 0.15), grip strength (adj-β: 0.79; 95% CI: 0.16 ~ 1.41), gait speed (adj-β: 0.05; 95% CI: 0.02 ~ 0.08) and chair stand time (adj-β: − 1.03; 95% CI: − 1.57 ~ − .48).

Table 4.

Association between physical activity and sarcopenia stratified by sex.

Sarcopenia components	Men				Women
	Crude-OR (95% CI)	p-value	Adj-ORa (95% CI)	p-value	Crude-OR (95% CI)	p-value	Adj-ORa (95% CI)	p-value
Sarcopenia	0.60 (0.43 ~ 0.84)	0.003b	0.62 (0.42 ~ 0.91)	0.013b	0.83 (0.58 ~ 1.20)	0.334	0.86 (0.59 ~ 1.25)	0.421

	Crude-β (95% CI)	p-value	Adj-βa(95% CI)	p-value	Crude-β (95% CI)	p-value	Adj-βa (95% CI)	p-value
SMMI (Kg/m2)	0.13 (0.04 ~ 0.22)	0.007b	0.13 (0.07 ~ 0.20)	< 0.001b	− 0.01 (− 0.09 ~ 0.07)	0.840	0.09 (0.02 ~ 0.15)	0.006b
Grip strength (Kg)	0.97 (− 0.09 ~ 2.04)	0.075	0.44 (− 0.55 ~ 1.44)	0.385	0.79 (0.16 ~ 1.42)	0.014b	0.79 (0.16 ~ 1.41)	0.014b
Gait speed (Meters/sec)	0.10 (0.06 ~ 0.14)	< 0.001b	0.08 (0.04 ~ 0.12)	< 0.001b	0.06 (0.02 ~ 0.09)	0.001b	0.05 (0.02 ~ 0.08)	0.002b
Five-time chair stand time (s)	− 1.41 (− 2.18 ~ − 0.63)	< 0.001b	− 1.04 (− 1.83 ~ − 0.26)	0.010b	− 1.17 (− 1.71 ~ − 0.63)	< 0.001b	− 1.03 (− 1.57 ~ − 0.48)	< 0.001b

Physical activity was expressed as MET-minutes/week. Odds ratios (ORs) and 95% confidence intervals (CIs) were expressed per 500 MET-minutes/week. Beta coefficients and 95% CIs were also expressed per 500 MET-minutes/week.

^a All result of Adj-beta were adjusted by age, BMI.

^b p < 0.05.

Interaction of ACTN3 R577X polymorphism and physical activity on muscle mass and function

In men with RR + RX genotype, every 500 MET-min/week increase in physical activity was associated with significant improvements in gait speed (β: 0.08, 95% CI: 0.03 ~ 0.13) and chair stand time (β: − 1.31, 95% CI: − 1.94 ~ − 0.70). However, these effects were not observed in XX carriers. Similar patterns were found in women: every 500 MET-min/week increase in physical activity showed significant improvements in gait speed (β: 0.07, 95% CI: 0.03 ~ 0.10) and chair stand time (β: − 1.13, 95% CI: − 1.74 ~ − 0.54), whereas these associations were not observed in XX carriers. Notably, in older women, genotype–physical activity interactions on physical performance reached statistical significance (p-interaction = 0.021 for gait speed; p-interaction = 0.005 for chair stand time) (Table 5).

Table 5.

Interaction of ACTN3 R577X and physical activity on sarcopenia stratified by sex.

Sarcopenia components /Genotype	Men					Women
	Crude-OR (95% CI)	p-value	Adj- ORa (95% CI)	p-value	Interaction p-value	Crude- OR (95% CI)	p-value	Adj- ORa (95% CI)	p-value	Interaction p-value
Sarcopenia b					0.093					0.794
RR + RX	0.46 (0.26 ~ 0.81)	0.031b	0.42 (0.18 ~ 0.95)	0.092		1.09 (0.63 ~ 1.89)	0.758	1.08 (0.59 ~ 1.97)	0.800
XX	0.32 (0.08 ~ 1.29)	0.110	0.30 (0.06 ~ 1.27)	0.109		0.77 (0.21 ~ 2.78)	0.687	0.53 (0.03 ~ 9.19)	0.660

	Crude-β (95% CI)	p-value	Adj- βa (95% CI)	p-value	Interaction p-value	Crude- β (95% CI)	p-value	Adj- βa (95% CI)	p-value	Interaction p-value
SMMI (Kg/m2)					0.153					0.388
RR + RX	0.12 (0.01 ~ 0.23)	0.045b	0.06 ( − 0.05 ~ 0.17)	0.282		0.01 ( − 0.08 ~ 0.09)	0.939	0.00 ( − 0.09 ~ 0.09)	0.996
XX	0.20 ( − 0.05 ~ 0.45)	0.105	0.24 ( − 0.00 ~ 0.49)	0.053		− 0.12 ( − 0.38 ~ 0.13)	0.335	− 0.14 ( − 0.38 ~ 0.11)	0.276
Grip strength (Kg)					0.528					0.588
RR + RX	1.54 (0.34 ~ 2.74)	0.007b	0.50 ( − 0.65 ~ 1.66)	0.270		0.79 (0.09 ~ 1.49)	0.023b	0.70 ( − 0.01 ~ 1.40)	0.066
XX	− 0.69 ( − 3.14 ~ 1.27)	0.571	− 0.48 ( − 2.88 ~ 1.92)	0.648		1.04 ( − 0.86 ~ 2.94)	0.285	1.55 ( − 0.34 ~ 3.44)	0.103
Gait speed (Meters/sec)					0.412					0.021b
RR + RX	0.11 (0.06 ~ 0.16)	< 0.001b	0.08 (0.03 ~ 0.13)	0.001b		0.07 (0.03 ~ 0.11)	< 0.001b	0.07 (0.03 ~ 0.10)	0.002b
XX	0.02 ( − 0.06 ~ 0.10)	0.596	0.04 ( − 0.03 ~ 0.10)	0.300		− 0.05 ( − 0.15 ~ 0.05)	0.286	− 0.06 ( − 0.16 ~ 0.04)	0.181
Five-time chair stand time (sec)					0.113					0.005b
RR + RX	− 1.39 ( − 2.08 ~ − 0.70)	< 0.001b	− 1.31 ( − 1.94 ~ − 0.70)	0.008b		− 0.90 ( − 1.58 ~ − 0.21)	< 0.001b	− 1.13 ( − 1.74 ~ − 0.54)	0.001b
XX	0.44 ( − 0.94 ~ 1.81)	0.524	1.14 ( − 0.55 ~ 2.82)	0.413		0.43 ( − 0.77 ~ 1.63)	0.182	1.33 ( − 0.40 ~ 3.06)	0.133

Physical activity was expressed as MET-minutes/week. Odds ratios (ORs) and 95% confidence intervals (CIs) were expressed per 500 MET-minutes/week. Beta coefficients and 95% CIs were also expressed per 500 MET-minutes/week.

^a All result of Adj-beta were adjusted by age, BMI.

^b p < 0.05.

Nonlinear relationship between physical activity and physical performance stratified by sex and ACTN3 R577X polymorphism

RCS analyses were applied to examine the nonlinear association between physical activity and physical performance. In men with RR + RX genotype, PA showed a linear association with significant improvements in chair stand time (p overall = 0.003). In contrast, chair stand time tended to worsen with increasing physical activity but did not reach statistical significance in XX carriers in Fig. 1A. After adjustment for age and BMI, similar patterns were observed in Fig. 1C. In women with RR + RX genotype, physical activity was also linearly associated with significant improvements in chair stand time (p overall = 0.004). However, in women with the XX genotype, chair stand time (p nonlinear = 0.036) showed a significant U-shaped association with increasing physical activity in Fig. 1B. After adjustment for age and BMI, similar patterns were observed in Fig. 1D. Piecewise GLM further confirmed a threshold effect of physical activity on chair stand time in women with XX genotype, identifying an inflection point at 441 MET-min/week. Above this threshold, each 500 MET-min/week increase in physical activity was associated with a 4.19-s longer chair stand time (95% CI = 0.984 ~ 7.398), whereas no significant association was observed below the threshold (Additional file 1: Table S2). For gait speed, in men with RR + RX genotype, physical activity showed a linear association with significant improvements in gait speed (p overall = 0.003). In contrast, gait speed tended to worsen with increasing physical activity but did not reach statistical significance in XX carriers in Additional file 1: Fig. S1A. After adjustment for age and BMI, similar patterns were observed in Additional file 1: Fig. S1C. In women with RR + RX genotype, physical activity was also linearly associated with significant improvements in gait speed (p overall = 0.037). However, in women with the XX genotype, gait speed (p nonlinear = 0.182) showed a U-shaped trend with increasing physical activity in Additional file 1: Fig. S1B. After adjustment for age and BMI, similar patterns were observed in Additional file 1: Fig. S1D.

Fig. 1.

Restricted cubic spline (RCS) analyses of the association between physical activity (MET-min/week) and five-time chair stand time, stratified by sex and ACTN3 R577X genotype. Curves represent β coefficients with shaded areas indicating 95% confidence intervals (CIs). (A)–(B) Crude model; (C)–(D) models adjusted for age and BMI. The ACTN3 R577X RR + RX genotype group is shown in blue (dashed line), and the XX genotype group is shown in red (solid line). In men, no significant nonlinear association was observed between physical activity and chair stand time (p for nonlinear > 0.05). In women, a significant nonlinear (U-shaped) association was detected for XX carriers (Crude model: p for nonlinear = 0.029; adjusted model: p for nonlinear = 0.036) but not for RR + RX carriers. The inflection point was identified at approximately 441 MET-min/week in adjusted models; above this threshold, higher PA was associated with longer chair stand time.

Discussion

This study found no significant association between ACTN3 R577X polymorphism and sarcopenia components, whereas higher physical activity was linearly associated with greater SMMI and better muscle function in both sexes. However, in both men and women with the XX genotype, physical performance tended to decline as physical activity increased. Notably, physical activity exceeding 441 MET-min/week was associated with poorer chair stand performance in female XX carriers, revealing a U-shaped nonlinear association between physical activity and physical performance.

Consistent with previous studies, the ACTN3 R577X polymorphism was not significantly associated with sarcopenia components^30,40,41. Prior research has also demonstrated that higher physical activity was beneficial for muscle mass and function^17,18; however, these improvements were attenuated in XX carriers. For instance, Pereira et al. reported smallest gains in lower-limb functional performance in older women with XX genotype after high-speed power training²⁰ and Delmonico et al. found that older adults with XX genotype had smaller increase in knee extensor peak power following strength training⁴². Our findings support these observations, emphasizing that the effects of physical activity on muscle function may differ by genotype.

In women with XX genotype, increasing physical activity was insignificantly associated with poorer physical performance. However, a significant interaction between genotype and physical activity was observed. This may be due to a nonlinear relationship between physical activity and physical performance in women with the XX genotype. Therefore, we applied an RCS model to examine this nonlinear pattern. The analysis revealed a significant U-shaped association between physical activity and chair stand time in women XX carriers, but not in men. These results suggest a potential sex-specific threshold beyond which excessively high physical activity levels may be associated with poorer physical performance.

Human skeletal muscle mainly consists of Type I and Type II fibers. Type I fibers are involved during endurance exercise, while type II fibers are primarily involved in power and high-intensity exercise⁴³. The protein α-actinin-3, which is encoded by ACTN3, is expressed only in type II fibers²⁸; therefore, older adults carrying the XX genotype lack this protein, leading to reduced Type II fiber function and consequently poorer physical performance. Among sarcopenia components, gait speed showed a nonsignificant U-shaped relationship with physical activity, possibly because it involves both Type I and Type II fibers⁴⁴. Although the XX genotype affects Type II fibers, it does not affect Type I fibers; therefore, gait speed did not show a significant U-shaped association with physical activity. Grip strength primarily depends on Type II fibers⁴⁵, but the XX genotype did not significantly reduce grip strength. This may be explained by the fact that most participants mainly engaged in lower-limb physical activity such as walking, rather than upper-limb physical activity like tennis or badminton. Previous studies also showed that lower-limb physical activity provides limited benefits to upper-limb strength⁴⁶. Therefore, genotype–physical activity interaction was not observed for grip strength. The chair stand time primarily depends on Type II fibers^47,48, and was therefore more affected by XX genotype. This explains why this study observed significantly poorer chair stand performance among older adults with the XX genotype under excessive physical activity, especially in women.

The observed sex difference may be partly attributed to hormonal factors. Estrogen deficiency contributes to muscle weakness through multiple mechanisms, including apoptosis, myosin dysfunction and impaired regeneration, ultimately reducing force-generating capacity⁴⁹. Hence, postmenopausal women with the XX genotype tend to experience earlier Type II fiber atrophy⁵⁰. Meanwhile, older men with the XX genotype showed an insignificant trend toward longer chair stand time, probably due to their larger Type II fibers reserve. Furthermore, GTEx database demonstrates lower ACTN3 expression in XX carriers (p = 3.68 × 10⁻¹⁵⁸), supporting the link between α-actinin-3 deficiency and reduced type II fibers function⁵¹.

Previous studies mainly focused on the effects from physical activity frequency in older adults with XX genotype²¹. Instead of physical activity frequency, our study used METs to quantify physical activity and applying an RCS model. More importantly, we observed an exploratory physical activity threshold (441 MET-min/week) for older women with the XX genotype, filling the gap that previous studies had not addressed. Nevertheless, several limitations should be acknowledged. First, participants were recruited from a physical examination program, who may have higher health literacy. Second, physical activity was assessed from face-to-face interview questionnaire. However, self-reported assessments may be limited by recall bias, possible overestimation, and subjective comprehension^52,53. Third, the questionnaire did not distinguish between aerobic and resistance exercise. Because the ACTN3 XX genotype affects type II muscle fibers that are primarily recruited during resistance exercise, we could not further assess whether the observed associations differed by physical activity modality. Finally, given the cross-sectional design, the physical activity threshold was interpreted as an exploratory finding, and external validation in longitudinal studies is required.

While WHO guidelines recommend at least 150 to 300 min of moderate-intensity physical activity per week for older adults (equivalent to ≥ 600 MET-min/week)⁵⁴, the observed inflection point of 441 MET-min/week in female ACTN3 XX carriers suggests that standardized exercise prescriptions may not fully account for genetic and sex heterogeneity. This value should therefore not be interpreted as a rigid upper limit, but rather as supportive evidence for personalized exercise prescriptions. Given the practical barriers to widespread genetic screening and muscle mass assessment in primary care^55,56, clinicians can recommend moderate-intensity exercise while using simple screening methods, such as chair stand time^57,58, to monitor muscle-related risks and adjust physical activity intensity, duration, or modality accordingly. Furthermore, although muscle function was evaluated using the AWGS 2019 criteria, different definitions or measurement approaches for sarcopenia may influence clinical outcome interpretation^59,60.

Overall, our findings underscore the importance of integrating sex-specific considerations into exercise guidelines for older adults, as we observed that biological sex significantly affects how the ACTN3 XX genotype relates to muscle function. Specifically, we observed a physical activity threshold of 441 MET-min/week for female XX carriers. These observations support the hypothesis that biological sex influences the susceptibility of muscle function to physical activity in the absence of alpha-actinin-3. However, further longitudinal and interventional studies are required to validate this association and determine the optimal physical activity type and intensity for this subgroup.

Conclusion

This study suggests a potential physical activity threshold for older women carrying the ACTN3 XX genotype. Notably, this specific MET value is an exploratory finding derived from our cross-sectional dataset and does not define a clinical cut-off. While sex-specific physical activity thresholds may exist in older adults with the ACTN3 XX genotype, these results represent preliminary associations that require external validation.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1 ^{(115.6KB, docx)}

Author contributions

Conceptualization, H-C H, S-L S; Investigation, H-C H, H-T L; Resources, W-H F, W-T C, C-J C, C-C W, M-J C, S-B L; Data curation, H-C H; Formal analysis, H-C H; Methodology, M-C L, Y-H C, S-L S; Supervision, S-L S; Validation, M-C L, Y-H C, S-L S; Writing—original draft, H-C H; Writing—review & editing, M-C L, Y-H C, S-L S. All authors read and approved the final manuscript.

Funding

This study was funded by grant MOST111-2314-B016-011 from Ministry of Science and Technology, NSTC112-2314-B-016-037 and NSTC113-2314-B-016-031 from National Science and Technology Council, TSGH_D_113119 from Tri-Service General Hospital, TSGH-SS_E_113017 from Tri-Service General Hospital Songshan Branch, CHNDMC-113-11201 from Cheng Hsin General Hospital, RVHCY113012 from Taichung Veterans General Hospital in Taiwan.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.