Resistance-based training improves mitochondrial capacity and redox balance in aging adults, independent of polyphenol supplementation

Abstract

Aging is associated with declines in skeletal muscle function, mitochondrial capacity, and changes in redox balance, which collectively contribute to frailty and chronic disease risk. This study investigated the effects of a 12-week resistance training (RT) program combined with a small dose of high-intensity interval training (HIIT), with or without polyphenol supplementation, on mitochondrial respiratory capacity (MRC) and oxidative stress in middle-aged and older adults (55–70 years). Forty-one participants were randomized to receive either a polyphenol supplement or a placebo for 30 days before the training intervention. Following the training intervention, aerobic capacity, lean mass, and strength improved significantly in both groups. Training also increased MRC in the placebo group but not in the polyphenol group, which displayed higher MRC following the supplementation phase, possibly reflecting either a supplement effect or baseline variation. The training resulted in a 20 % decrease in skeletal muscle H₂O₂ emission across both groups, suggesting enhanced mitochondrial efficiency or antioxidant defenses. However, gene expression of selected antioxidants was unchanged, and plasma oxidative stress markers malondialdehyde (MDA) increased, and 3-nitrotyrosine (3-NT) remained unchanged. Circulating antioxidants showed distinct changes with training, as ascorbic acid increased with training in both groups, while α-tocopherol increased only in the placebo group and β-cryptoxanthin and retinol declined in the polyphenol group, suggesting potential supplement–nutrient interactions. Uric acid increased in both groups, likely reflecting exercise-induced purine turnover. In conclusion, combined RT and HIIT improved mitochondrial bioenergetics and muscle redox balance in middle-aged and older adults, whereas polyphenol supplementation did not augment these adaptations and may have blunted some vitamin-related responses. These findings underscore resistance-based exercise as a potent intervention for maintaining physical and mitochondrial health with age.

Graphical abstract

Highlights

• •Resistance training and a small dose of HIIT improved mitochondrial capacity in aging adults.
• •Training reduced muscle H₂O₂, indicating enhanced antioxidant or mitochondrial capacity.
• •Training increased α-tocopherol and ascorbic acid, but polyphenols may have blunted this effect.
• •Polyphenols did not boost training adaptations and may alter carotenoid and retinol metabolism.

1. Introduction

Aging entails a gradual loss of skeletal muscle mass and strength, accompanied by an increased risk of chronic and cardiometabolic diseases. A central hallmark of the aging process is mitochondrial dysfunction [1], an umbrella term encompassing impairment in essential mitochondrial processes, such as bioenergetics, ion homeostasis or intracellular signaling [2]. Of particular importance are disruptions in mitochondrial redox balance, leading to elevated production and/or release of reactive oxygen species (ROS), which have been implicated as an important driver of cellular aging and tissue dysfunction [3]. While ROS serve as essential signaling molecules in numerous cellular processes [4], excessive ROS can be detrimental, leading to lipid peroxidation, protein oxidation, and ultimately triggering apoptosis and tissue damage [5]. In the absence of cellular stress, ROS production is kept at a minimum and redox homeostasis is ensured by the cellular antioxidant system. The issue arises when the production of ROS exceeds the signaling level or the capacity of the antioxidant defense system, creating a state of oxidative stress [6], which has been implicated in the development and progression of age-related diseases, such as type 2 diabetes [7,8] and cardiovascular disease [9,10]. Mitochondria are the major source of ROS production, and at least 11 distinct sites of ROS production have been identified in the mitochondria [11]. Age-related mitochondrial dysfunction is closely linked to declines in physical function [12] and has been proposed as a contributing factor in the development of sarcopenia [13]. Consequently, interventions aimed at preserving or restoring mitochondrial capacity and enhancing redox regulation are of significant interest for supporting healthy aging.

Exercise training, particularly resistance training (RT), is well known to enhance muscle mass, strength and physical function in middle-aged and older adults. However, its effects on mitochondrial capacity and oxidative stress remain less well defined. Regarding mitochondrial adaptations, RT has been reported to increase mitochondrial respiratory capacity (MRC) [14], but the findings are inconsistent [15]. In fact, some studies suggest that RT reduces mitochondrial volume density, explained by a “dilution effect” as a consequence of increased muscle fiber cross-sectional area [16].

The influence of RT on oxidative stress in elderly individuals is similarly inconclusive. Some studies have demonstrated increased systemic [17] and muscle [18] levels of antioxidant proteins or mRNA [19], while others report no change [20] or even a decrease in muscle antioxidant protein levels [21]. Moreover, it remains uncertain whether these molecular adaptations translate into reduced levels of oxidative damage or reduced ROS emission directly within the skeletal muscle of elderly individuals, as few studies have directly assessed these outcomes. One study reported reduced muscle protein levels of 4-hydroxynonenal-modified proteins (4-HNE) [19], a marker of lipid peroxidation, in older adults following RT, whereas two other studies found no change in skeletal muscle H₂O₂ emission [22,23].

Similarly, the impact of high-intensity interval training (HIIT) on ROS regulation in aging skeletal muscle also remains unclear. HIIT is known to induce mitochondrial adaptations and improve oxidative capacity [24], yet whether these effects alter ROS production or oxidative damage in skeletal muscle of middle-aged and older adults remains unexplored. Studies in healthy young and middle-aged individuals have reported increased glutathione peroxidase (GPX) following HIIT [25], but no change in H₂O₂ emission rates [26]. In contrast, one study in obese adults demonstrated both increased antioxidant protein levels and decreased H₂O₂ emission rates in muscle following six weeks of HIIT [27], suggesting that this training modality may improve redox balance in populations particularly susceptible to oxidative stress, such as obese or aging individuals. Finally, although combined RT and HIIT are frequently applied in both research and clinical practice, no studies to date have examined how this approach influences redox balance in middle-aged and older adults. Filling this gap could provide important insights into the ways in which different training modalities modulate oxidative stress with age.

Nutritional strategies such as polyphenol supplementation have been proposed to modulate mitochondrial function and redox homeostasis and thus ameliorate age-related diseases [28]. This is largely based on animal and in-vitro studies with limited evidence from human clinical trials. However, different types of polyphenols have been reported to increase systemic antioxidant capacity [29,30] and upregulate muscle gene transcripts associated with mitochondrial function in elderly subjects [31]. Nevertheless, no studies have examined the effects of polyphenol supplementation on mitochondrial capacity and ROS emission directly in the skeletal muscle of elderly subjects. Furthermore, no studies to date have examined the combined effects of polyphenol supplementation and exercise on these outcomes.

Given the potential of both RT and HIIT to influence redox balance and mitochondrial function in aging individuals, the combination may produce additive or synergistic effects, warranting investigation through well-controlled clinical studies. As such, the aim of the current study was to investigate if a combination of RT and small dose of HIIT, with or without polyphenol supplementation could improve mitochondrial capacity and decrease oxidative stress and ROS capacity in muscle, in aging individuals.

2. Methods

2.1. Participants and study overview

Details of the study design, participant characteristics, eligibility criteria, testing procedures, and training protocol, have recently been published [32]. The study was conducted in accordance with the Declaration of Helsinki and received approval from the Ethics Committee of the Capital Region of Denmark. An overview of the study protocol is presented in Fig. 1. In short, 41 individuals (21 men and 20 women) aged 55–70 years were randomized into two groups: a polyphenol group (mean age 62 ± 4, age range 55–69) and a placebo group (mean age 61 ± 4, age range 55–70). Subjects reported to the lab in the fasted state for an initial screening visit, including measurements of body composition, blood samples, VO₂max and a battery of strength measurements. Subjects then ingested either a polyphenol or a placebo supplement daily for 30 days (loading phase) and reported back to the lab to repeat the measurements from the screening visit. The polyphenol supplement was derived from red- and blackcurrant and contained approximately 700 mg of polyphenols per dose. For full composition of the supplement, including a mass spectrometry analysis, see reference study [32]. Subjects then arrived in the lab for a third testing day in the morning in the fasted state, and a muscle biopsy was obtained from the right vastus lateralis muscle. Subsequently, participants performed 45 min of cycling at an intensity corresponding to 65 % of VO₂max, where gas exchange was measured throughout, and then rested in the supine position for 60 min. Blood samples were collected at rest and every 15 min during the exercise and 60 min of recovery. Following the initial three testing days, the participants then undertook a 12-week supervised resistance-based training program, consisting of two weekly resistance training (RT) sessions and one weekly RT + high intensity interval training (HIIT) session. RT sessions included leg press, leg extension, leg curl, lat pulldown, seated row, chest press, and shoulder press. Lower-body exercises progressed from 3 × 12 RM (weeks 1–3) to 4 × 6 RM (weeks 9–12), while upper-body exercises progressed from 15 RM to 10 RM over 12 weeks. The RT + HIIT session consisted of just lower body exercises, followed by 7–9 60-s high-intensity intervals, performed on a BikeErg cycle ergometer (Concept2, Morrisville, VT, USA). All sessions began with two 60-s warm-up intervals at 90 % of the resistance that elicited VO₂max in the PRE-test and subsequently 5–7 work intervals of 60 s with 60-s rest. From weeks 1–4, the protocol included three intervals at 100 % VO₂max and two all-out efforts. In weeks 5–8, this increased to four intervals at 105 % VO₂max and two all-out sets, progressing to four intervals at 110 % VO₂max and three all-out sets in weeks 9–12. Following the training intervention, subjects reported to the lab for two separate testing days, one day replicating the initial two testing days with body composition and performance tests, and one day replicating the testing day with biopsies and a moderate intensity cycling test at the same absolute workload as the first testing day.

Fig. 1.

Study protocol. Subjects reported fasted for a screening (A0), including blood samples, body composition measurements, fitness and strength tests. Subject were randomized to ingest either a polyphenol or placebo supplement daily for 30 days (loading phase). Subjects reported back to the lab after the loading phase to replicate the previous measurements and tests (A1). After 2–4 days, subjects arrived fasted to the lab, and a blood sample and muscle biopsy was obtained. Subjects then completed 45 min of moderate intensity cycling (65 % of VO₂max), with blood samples being taken at minute 15, 30 and 45. Subjects then rested in the supine position for 60 min, with blood samples being drawn every 15 min. Respirometry and fluorometry was performed using part of the obtained muscle biopsy on the same day. The subjects then undertook a 12-week training program, consisting of full-body resistance training (RT) 2 times/week, and lower body RT + 7–9 min if HIIT 1 time/week, while continuing supplement or placebo. Following the training program, testing day A1 was replicated (A2), followed by a final training session and then two days later testing day B1 was replicated (B2).

2.2. Muscle biopsies

Muscle biopsies were collected from the vastus lateralis using a Bergström needle and divided into three portions. One portion was embedded in OCT compound, frozen in isopentane cooled by liquid nitrogen, and stored at −80 °C. One portion was placed in BIOPS buffer (100 mM CaK₂EGTA, 100 mM K₂EGTA, 5.77 mM Na₂ATP, 6.56 mM MgCl₂·6H₂O, 20 mM Taurine, 15 mM Na₂Phospho-creatine, 20 mM Imidazole, 0.5 mM DTT, 50 mM MES) for respirometry and fluorometry analysis. Remaining tissue was snap-frozen in liquid nitrogen and stored at −80 °C for qPCR analysis.

2.3. Preparation and permeabilization of skeletal muscle fibers

Muscle fibers were cleaned of blood, adipose tissue, and connective tissue, then separated into smaller bundles using fine needles while kept on ice in BIOPS preservation buffer. Fibers were permeabilized for 30 min with saponin (50 μg/mL) under gentle agitation, then washed twice for 10 min in MiR05 buffer (EGTA 0.5 mM, MgCl₂.6H₂O 3 mM, K-lactobionate 60 mM, taurine 20 mM, KH₂PO₄ 10 mM, HEPES 20 mM, sucrose 110 mM, BSA 1 g/L at pH 7.1) with gentle shaking. Fiber bundles of approximately 2 mg were briefly dried on filter paper, weighed, and placed in oxygraph chambers to normalize oxygen flux rates to tissue wet weight.

2.4. High-resolution respirometry

All measurements were performed in duplicate using two Oxygraph-2k systems (Oroboros, Austria) in hyperoxygenated chambers (450 nmol O₂/mL). The following substrate titration protocol was used: Pyruvate (5 mM), malate (2 mM), and glutamate (10 mM) were added to assess leak respiration via complex I. ADP (5 mM), accompanied by MgCl2 (3 mM), was added to determine state 3 respiration with complex I substrates. Cytochrome c (0.01 mM) was then added to verify outer mitochondrial membrane integrity (no response observed). Succinate (10 mM) was added to assess complex I + II-supported respiration, followed by stepwise FCCP titration (0.5 μM) to reach maximal uncoupling. Oxygen flux was analyzed using the DatLab software, and expressed as pmol O₂/s/mg. Respiratory control ratio (RCR) was used as a measure of mitochondrial efficiency, by dividing maximal coupled respiration (complex I + II respiration, determined by the addition of succinate) with leak respiration (respiration with only pyruvate, malate and succinate).

2.5. High-resolution fluorometry

Mitochondrial H₂O₂ emission was determined fluorometrically, simultaneously with high-resolution respirometry by the O2k-Fluorometer (O2k-Fluo LED2-Module Fluorescence-Sensor Green), using the H₂O₂-sensitive probe Amplex UltraRed. The substrate titration for the protocol was as follows: Blebbistatin (5 μM) was first added to inhibit muscle fiber contraction. This was followed by superoxide dismutase (1.7 μM), Amplex Red (4 μM), and horseradish peroxidase (HRP, 4 μM) to catalyze the conversion of superoxide to H₂O₂ and generate the fluorescent signal via resorufin. Pyruvate (5 μM) and malate (10 μM) were added to assess state 2H₂O₂ emission with complex I substrates, followed by a stepwise titration of succinate (1, 2, and 3 mM) and finally ADP (5 mM) in combination with MgCl₂ (3 mM). To calibrate the fluorometer and account for changes in sensitivity, a 0.1 μM H₂O₂ injection was made after HRP addition and repeated after pyruvate/malate, after 3 mM succinate, and after ADP/MgCl₂. Data was analyzed using the DatLab software and normalized to muscle wet weight and oxygen flux rates (pmol H₂O₂/min/mg ww/JO₂).

2.6. Citrate synthase and β-hydroxyacyl-CoA activity

Citrate synthase (CS) and β-hydroxyacyl-CoA (β-HAD) activities were measured using spectrophotometry. Approximately 10 mg wet weight of muscle was homogenized in 600 μl 0.3 M K₂ HPO₄, 0.05 % BSA, pH 7.7 for 2 min on a Tissuelyzer (Qiagen, Venlo, Limburg, Netherlands). 6 μl of 10 % Triton was added, and the samples were left on ice for 15 min. For CS activity, the homogenate was diluted 50 times in a solution containing 0.4 mM acetyl-CoA, 0.6 mM oxaloacetate, 0.175 mM DTNB, 30 mM tris-HCI (pH 8.0). The change in DTNB to TNB at 37 °C was measured spectrophotometrically at 415 nm on an automatic analyzer, Cobas 6000 (Roche Diagnostics, Mannheim, Germany). For β-HAD activity, the homogenate was diluted 70 times in a solution containing 0.34 mM acetoacetyl-CoA, 6.96 mM NADH, 0.14 M EDTA, 27.83 mM imidazole (pH 7.0). The changes in NADH at 37 °C were measured spectrophotometrically at 340 nm on an automatic analyzer, Cobas 6000 (Roche Diagnostics, Mannheim, Germany). Enzyme activities are expressed as micromoles substrate per minute per gram of wet weight of muscle tissue.

2.7. Plasma redox markers and micronutrients

Plasma redox biomarkers 3-nitrotyrosine (3-NT) and malondialdehyde (MDA) were analyzed by in-house ELISA and HPLC with fluorescence detection after derivatization, respectively. Carotenoids, retinol and tocopherols were analyzed by HPLC, as were ascorbic acid and uric acid. All methods have been described previously [33].

2.8. Gene expression analysis

Gene expression was analyzed in PRE and POST muscle samples from both groups. Due to limited tissue availability, the analysis included 11 participants from the Polyphenol group and 15 from the placebo group. Total RNA was extracted using Trizol™ Reagent (ThermoFisher Scientific, Australia) with chloroform, per manufacturer's instructions. RNA concentration was measured with a Qubit fluorometer (Thermo Fisher Scientific, USA). cDNA was synthesized from 500 ng RNA using the Quantitect Reverse Transcription Kit (Qiagen, Germany), and qPCR was performed with the QuantiNova SYBR Green PCR Kit (Qiagen, Germany). Gene expression was quantified using the ΔΔCt method and normalized to Actin β as the housekeeping gene. Target genes included Mitochondrial transcription factor A (TFAM), Mitofusin-2 (MFN2), Copper-zinc-superoxide dismutase (CuZnSOD), manganese-dependent superoxide dismutase (MnSOD), and Catalase (CAT). Primer sequences are listed in Supplementary Data (Table S1).

2.9. Statistical analysis

A power calculation was performed in G∗Power (version 3.1.9.7, Universität Düsseldorf) for a repeated-measures ANOVA testing a group × time interaction (2 groups × 2 timepoints). We aimed to detect a moderate interaction effect (Cohen's f = 0.25) with α = 0.05 and 80 % power (1−β = 0.80). The chosen effect size (f = 0.25) reflects a conservative and realistic estimate of the expected training × supplement interaction based on similar human intervention studies. This yielded a required total sample size of 34 participants (17 per group). To account for dropouts, 51 participants were included, and ultimately 41 completed the study. See Ref. [32] for Consort diagram. All statistical analyses and figure generation were conducted using GraphPad Prism version 10.4.0 (Dotmatics, San Diego, CA, USA). To evaluate outcomes across the intervention periods, a two-way analysis of variance (ANOVA) was applied, with time specified as the within-subject factor and group as the between-subject factor. For outcomes where data were missing at certain time points, a mixed-effects model was employed to account for the unbalanced design, assuming data were missing at random. When a significant interaction between time and group was identified, post hoc comparisons were carried out with adjustment for multiple testing using the Šidák correction method. Comparisons of participant characteristics, before the intervention, between groups were performed using unpaired t-tests. Assumptions of normality were assessed using the Shapiro–Wilk test, quantile–quantile (QQ) plots, and inspection of residual distributions. Results are reported as mean ± standard deviation (SD), or as mean (bars) with individual values. A p-value of <0.05 was considered statistically significant.

3. Results

3.1. Participant characteristics

Descriptive participant characteristics before and after the intervention have been published previously [32]. The most important characteristics are listed in Table 1, but we refer to the previous publication for the full dataset, including hematological, fibertype-specific, metabolic and inflammatory adaptations. In brief, the training intervention increased VO₂max (p = 0.0001, main effect) and whole-body lean mass (p < 0.0001, main effect) significantly while lowering body fat percentage (p = 0.039, main effect), to the same extent in both groups (Table 1). Polyphenol supplementation did not affect body composition or physical performance and did not enhance training adaptations (Table 1).

Table 1.

Participant characteristics after one month of supplement intake, before the training intervention (PRE) and after 12 weeks of training with continuous supplementation (POST). Weight, BMI, whole-body fat mass, fat percentage, visceral adipose tissue, whole-body lean mass determined by DXA. VO₂max determined by ramp test on cycle ergometer. Data were analyzed using a two-way ANOVA. P-value <0.05 was considered significant and is indicated in bold. Data are presented as means ± SD.

Placebo (n = 21)			Polyphenol (n = 20)		P-value
	PRE	POST	PRE	POST	Time	Time XGroup
Sex (m/w)	11/10	–	10/10	–	–	–
Age (years)	62 ± 4	–	61 ± 4	–	–	–
HbA1C (mmol/mol)	38 ± 2	–	37 ± 3	–	–	–
Height (m)	1.74 ± 0.1	–	1.73 ± 0.1	–	–	–
Weight (kg)	79.1 ± 11.8	79.1 ± 11.8	78.6 ± 7.8	78.5 ± 7.7	0.561	0.476
BMI (kg/m 2)	26.1 ± 3.3	26.3 ± 3.4	26.2 ± 3.3	26.2 ± 1.8	0.646	0.527
Whole-body fat mass (kg)	26.4 ± 7.6	26.0 ± 8.0	26.7 ± 6.8	25.9 ± 6.2	0.077	0.463
Fat percentage (%)	33.4 ± 8.3	32.9 ± 8.7	34.2 ± 7.9	33.3 ± 7.8	0.039	0.639
Visceral adipose tissue (g)	1311 ± 732	1280 ± 792	1241 ± 758	1165 ± 658	0.143	0.530
Whole-body lean mass (kg)	52.6 ± 10.1	53.5 ± 10.4	51.9 ± 8.0	52.8 ± 8.7	<0.0001	0.946
VO 2 max (ml O2/min)	2279 ± 298	2361 ± 327	2362 ± 298	2552 ± 327	0.0001	0.195
VO 2 max (ml O2/min/kg)	29 ± 5	30 ± 4	30 ± 5	32 ± 5	0.002	0.179

3.2. Mitochondrial respiratory capacity

When comparing the two groups at baseline (after the supplement loading phase, before the training intervention), the polyphenol group displayed higher MRC than the placebo group (p = 0.002, main effect) (Fig. 2A). Following the training intervention, MRC increased significantly in the placebo group (p = 0.01, main effect) with significantly increased maximal respiration (CI + II, p = 0.01) and maximally uncoupled respiration (ETS, p = 0.0006) (Fig. 2B), while no significant changes were observed for the polyphenol group (p = 0.388) (Fig. 2C). No differences were observed in leak respiration or RCR between groups before or following the training intervention (data not shown).

Fig. 2.

Mitochondrial bioenergetics and enzyme activity. A) MRC at baseline (after 30 days of supplementation, before training) in the placebo group (beige bars) and the polyphenol group (green bars). B) Mitochondrial respiratory capacity (MRC) before (PRE) and after (POST) the training intervention in the placebo group. C) MRC before (PRE) and after (POST) the training intervention in the polyphenol group. D) Citrate synthase (CS) activity (μmol/g/min) before (PRE and after (POST) the training intervention. E) β-hydroxyacyl-CoA dehydrogenase (β-HAD) activity (μmol/g/min) before (PRE and after (POST) the training intervention. Data presented as individual values and mean (bars). P-value <0.05 is considered significant. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

3.3. Mitochondrial enzyme activity

Following the supplement loading phase, no significant differences were observed between groups in CS activity (p = 0.464) (Fig. 2D). CS activity did not change following the training intervention in either of the groups (p = 0.985 for the placebo group, p = 0.937 for the polyphenol group) (Fig. 2D). Likewise, no significant differences were observed in β-HAD activity between groups following the loading phase (p = 0.534) or following the training intervention (p = 0.743 for the placebo group, p = 0.614 for the polyphenol group) (Fig. 2E).

3.4. Mitochondrial H₂O₂ emission

There were no significant differences in H₂O₂ emission rates between groups at following the supplement loading phase (p = 0.362) (Fig. 3A). The training intervention resulted in significantly lower H₂O₂ emission in both the placebo group (Fig. 3B) and the polyphenol group (Fig. 3C) across all succinate concentrations (p < 0.0001 for all succinate concentrations in both groups), with no difference between groups. Upon inspection of the individual data points, it appeared that an outlier sample in each group may be driving the observed reductions. When these outliers were removed, the p-value increased slightly in the placebo group (from <0.0001 to < 0.01), while no significant change was observed in the polyphenol group. Outliers were kept in the dataset as we had no obvious reason to exclude them, and their exclusion did not alter the conclusions.

Fig. 3.

Mitochondrial H₂O₂ emission. Emission rates measured across three different succinate concentrations: Suc 1 (1 mM), Suc 2 (2 mM), Suc 3 (3 mM), in the presence of blebbistatin, HRP, amplex red, pyruvate and malate. A) Mitochondrial H₂O₂ emission at baseline (after 30 days of supplementation, before training) in the placebo group (beige bars) and the polyphenol group (green bars). B) Mitochondrial H₂O₂ emission rates before (PRE) and after (POST) the training intervention in the placebo group. C) Mitochondrial H₂O₂ emission rates before (PRE) and after (POST) the training intervention in the polyphenol group. Data presented as individual values and mean (bars). P-value <0.05 is considered significant. ∗∗∗∗p < 0.0001.

3.5. Micronutrients and redox markers

After the one-month supplement loading phase, plasma retinol levels decreased significantly in the polyphenol group (p < 0.05) (Table S2). No other plasma vitamins or oxidative stress markers were altered during this phase (Table S2).

Following the three-month training intervention, several changes were observed. In the polyphenol group, β-cryptoxanthin concentrations were significantly lower in the POST compared with the PRE condition, both at rest (p = 0.04) and at the end of exercise (p = 0.01), whereas no changes occurred in the placebo group (Table 2). Retinol was also reduced at the end of exercise in the POST condition compared with PRE, but only in the polyphenol group (p = 0.04) (Table 2).

Table 2.

Plasma vitamins and oxidative stress markers, measured at rest (Rest), at the end of 45 min of moderate intensity cycling (End-ex) and after 1 h of recovery (Recovery), before (PRE) and after (POST) 12 weeks of RT and HIIT training with or without polyphenol supplementation. Data were analyzed using a two-way ANOVA. P-value <0.05 was considered significant and is indicated in bold and with the following asterisks: ∗ = different from PRE. # = different from placebo. Data are presented as means ± SD.

Polyphenol (n = 20)					Placebo (n = 21)
Rest			End-ex	Recovery	Rest	End-ex	Recovery
β-Cryptoxanthin	PRE	0.16 ± 0.19	0.19 ± 0.21	0.15 ± 0.18	0.17 ± 0.11	0.22 ± 0.12	0.16 ± 0.10
(μmol/L)	POST	0.16 ± 0.20	0.18 ± 0.21	0.16 ± 0.19	0.13 ± 0.08∗	0.17 ± 0.08∗	0.14 ± 0.07
	Δ	−0.00 ± 0.13	−0.01 ± 0.16	0.01 ± 0.14	−0.04 ± 0.06	−0.05 ± 0.08	−0.02 ± 0.09
Lutein + Zeaxanthin	PRE	0.25 ± 0.26	0.30 ± 0.29	0.22 ± 0.24	0.30 ± 0.21	0.38 ± 0.26	0.26 ± 0.21
(μmol/L)	POST	0.25 ± 0.20	0.28 ± 0.21	0.23 ± 0.19	0.33 ± 0.29	0.40 ± 0.36	0.31 ± 0.29
	Δ	−0.01 ± 0.15	−0.02 ± 0.16	0.01 ± 0.13	0.03 ± 0.12	0.02 ± 0.10	0.05 ± 0.15
Lycopene	PRE	0.53 ± 0.23	0.56 ± 0.25	0.51 ± 0.22	0.60 ± 0.17	0.63 ± 0.21	0.58 ± 0.17
(μmol/L)	POST	0.58 ± 0.25	0.63 ± 0.31	0.57 ± 0.28	0.64 ± 0.27	0.70 ± 0.26	0.64 ± 0.25
	Δ	0.05 ± 0.22	0.07 ± 0.25	0.06 ± 0.24	0.05 ± 0.28	0.08 ± 0.35	0.06 ± 0.29
α-Carotene	PRE	0.22 ± 0.18	0.24 ± 0.20	0.22 ± 0.17	0.14 ± 0.14	0.17 ± 0.16	0.14 ± 0.25
(μmol/L	POST	0.26 ± 0.28	0.28 ± 0.33	0.26 ± 0.29	0.13 ± 0.10	0.15 ± 0.10	0.14 ± 0.10
	Δ	0.04 ± 0.19	0.16 ± 0.19	0.05 ± 0.19	−0.02 ± 0.10	−0.02 ± 0.10	−0.00 ± 0.11
β-Carotene	PRE	0.55 ± 0.28	0.61 ± 0.31	0.54 ± 0.26	0.44 ± 0.27	0.53 ± 0.30	0.43 ± 0.28
(μmol/L)	POST	0.56 ± 0.35	0.60 ± 0.41	0.58 ± 0.35	0.46 ± 0.31	0.53 ± 0.32	0.48 ± 0.29
	Δ	0.01 ± 0.24	−0.01 ± 0.26	0.04 ± 0.22	0.01 ± 0.19	0.00 ± 0.18	0.04 ± 0.23
Retinol	PRE	1.88 ± 0.31	2.12 ± 0.48	1.73 ± 0.43	2.04 ± 0.45	2.40 ± 0.55	1.89 ± 0.47
(μmol/L)	POST	1.80 ± 0.40	2.08 ± 0.43	1.71 ± 0.39	1.87 ± 0.52	2.19 ± 0.46∗	1.79 ± 0.51
	Δ	−0.07 ± 0.32	−0.04 ± 0.35	−0.01 ± 0.34	−0.17 ± 0.43	−0.21 ± 0.31	−0.10 ± 0.52
α –Tocopherol	PRE	29.2 ± 6.8	33.1 ± 7.4	28.2 ± 6.7	31.0 ± 6.6	37.0 ± 7.0	29.7 ± 6.0
(μmol/L)	POST	31.1 ± 7.1∗	35.2 ± 8.1∗	31.0 ± 7.2∗	30.7 ± 7.4	35.7 ± 8.3	30.1 ± 9.7
	Δ	1.9 ± 3.9	2.1 ± 4.0	2.8 ± 3.7	−0.3 ± 5.2	−1.3 ± 5.7	0.4 ± 8.9
γ-Tocopherol	PRE	1.88 ± 0.73	1.98 ± 0.71	1.74 ± 0.73	1.98 ± 0.75	2.29 ± 0.72	1.94 ± 0.83
(μmol/L)	POST	1.71 ± 0.71	1.79 ± 0.78	1.63 ± 0.68	2.02 ± 0.82	2.31 ± 0.89	1.94 ± 0.71
.	Δ	−0.17 ± 0.95	−0.19 ± 0.77	−0.10 ± 0.70	0.04 ± 0.64	0.02 ± 0.49	0.00 ± 0.63
Ascorbic acid	PRE	3.21 ± 2.25	3.99 ± 2.50	3.47 ± 2.23	2.80 ± 1.70	3.88 ± 1.93	2.82 ± 1.80
(mg/L)	POST	3.99 ± 3.12	5.63 ± 4.08∗	4.92 ± 4.30∗	5.11 ± 3.73∗	6.48 ± 3.92∗	5.05 ± 3.79∗
	Δ	0.78 ± 2.67	1.65 ± 2.64	1.45 ± 2.92	2.32 ± 3.10	2.60 ± 2.83	2.23 ± 3.02
Uric Acid	PRE	52.8 ± 10.8	55.0 ± 9.1	56.2 ± 9.6	54.9 ± 8.9	56.6 ± 10.1	58.0 ± 10.5
(mg/L)	POST	67.9 ± 15.3∗	71.6 ± 15.9∗	72.3 ± 15.8∗	66.9 ± 11.9∗	69.2 ± 12.8∗	67.7 ± 12.4∗
	Δ	15.2 ± 12.8	16.6 ± 14.3	16.1 ± 14.1	12.0 ± 13.1	12.6 ± 12.6	9.2 ± 6.8
MDA	PRE	1.40 ± 0.73	1.51 ± 0.42	1.31 ± 0.38	1.31 ± 0.34	1.42 ± 0.40	1.24 ± 0.30
(μmol/L)	POST	1.62 ± 0.45	1.83 ± 0.47∗	1.57 ± 0.38	1.74 ± 0.57∗	2.20 ± 0.91∗	1.62 ± 0.42∗
	Δ	0.23 ± 0.73	0.33 ± 0.53	0.26 ± 0.40	0.45 ± 0.49	0.78 ± 0.76#	0.38 ± 0.37
3-Nitrotyrosine	PRE	4.38 ± 1.82	4.72 ± 2.23	5.74 ± 3.70	5.67 ± 2.97	4.56 ± 2.07	4.47 ± 2.14
(pmol/mg)	POST	4.71 ± 1.86	4.59 ± 2.54	4.65 ± 2.02	4.62 ± 1.85	5.47 ± 2.92	5.76± 2.63
	Δ	0.34 ± 2.73	−0.13 ± 3.05	−1.10 ± 3.81	−1.05 ± 3.04	0.91 ± 3.44	1.29 ± 2.80#

In contrast, α-tocopherol increased in the placebo group during the POST condition, with higher concentrations at rest (p = 0.03), at the end of exercise (p = 0.04), and 1 h after exercise (p = 0.004), while no differences were detected in the polyphenol group (Table 2). Ascorbic acid also increased after training, in both groups. In the placebo group, levels were higher in the POST condition at the end of exercise (p = 0.01) and 1 h after exercise (p = 0.02). In the polyphenol group, significant increases were observed at rest (p = 0.002), at the end of exercise (p = 0.003), and 1 h after exercise (p = 0.003) (Table 2).

Uric acid concentrations increased markedly in both groups in the POST condition. Significant elevations were found at rest (placebo p < 0.0001; polyphenol p = 0.0003), at the end of exercise (placebo p < 0.0001; polyphenol p = 0.0002), and 1 h after exercise (placebo p < 0.0001; polyphenol p = 0.003). However, delta values (PRE to POST differences) did not differ between groups (Table 2).

Markers of oxidative damage were also altered by training. In the placebo group, MDA increased significantly at the end of exercise in the POST condition (p = 0.01), with non-significant trends toward higher levels at rest (p = 0.08) and 1 h post-exercise (p = 0.07) (Table 2). In the polyphenol group, MDA increased at all time points in the POST condition: at rest (p = 0.002), at the end of exercise (p < 0.0001), and 1 h after exercise (p = 0.005) (Table 2). MDA delta values were significantly higher in the polyphenol group compared with the placebo group at the end of exercise (p = 0.02) (Table 2). Similarly, 3-NT delta values were significantly higher 1 h after exercise in the polyphenol group compared with the placebo group (p = 0.04) (Table 2).

3.6. Gene expression

Analysis of the mitochondrial genes TFAM, MFN2, CuZnSOD, MnSOD, and Catalase revealed no statistically significant differences between the experimental groups (Fig. S3). Expression levels were stable across all participants, indicating that the intervention had no detectable effect on the transcription of these genes. Cycle threshold (Ct) values were normalized to the ACTB reference gene, and statistical testing confirmed the lack of significant variation (p > 0.05).

4. Discussion

This study investigated effects of resistance-based training and polyphenol supplementation on skeletal muscle mitochondrial capacity and redox balance in middle-aged and older adults. As expected from the well-documented benefits of exercise in aging populations, the training intervention improved aerobic capacity, lean mass and muscle strength. Importantly, these functional gains were accompanied by reductions in muscle H₂O₂ emission and enhanced mitochondrial bioenergetics. Together, these findings demonstrate that a combined program of resistance and interval-type training not only enhances physical performance but also improves mitochondrial function and redox regulation. Thus, exercise emerges as a powerful intervention to mitigate age-related declines in mitochondrial health and redox balance. Notably, polyphenol supplementation did not provide additional benefits beyond training alone, suggesting that the improvements observed were driven primarily by exercise.

4.1. Mitochondrial bioenergetics

The training intervention increased MRC significantly in the placebo group, while no difference was observed in the polyphenol group. Interestingly, the polyphenol group displayed higher MRC than the placebo group following 30 days of supplementation, indicating an enhancing effect of the supplement on the mitochondria. Polyphenols have previously been shown to stimulate mitochondrial biogenesis in mice and cell models [34], but this has yet to be observed in humans. However, since no biopsy was collected before the supplementation phase, it remains unclear whether the observed difference in MRC was a result of the polyphenol intervention or simply a pre-existing baseline variation. Nonetheless, this difference cannot be attributed to fitness levels, as VO₂max did not differ between groups before the intervention, nor were there significant differences in CS activity. The lack of increase in MRC following the training intervention could reflect a ceiling effect in the polyphenol group, where individuals with higher pre-training respiratory capacity had less scope for improvement. Alternatively, it is possible that the polyphenol supplement attenuated training-induced adaptations, as some previous studies have reported that antioxidant compounds can blunt training adaptations [[35], [36], [37], [38]]. This is supported also by a lack of increase in α-tocopherol following training in the polyphenol group, which was otherwise observed in the placebo group. Given that non-mitochondrial ROS sources such as NADPH oxidases play a central role in initiating exercise-induced signaling cascades [39], the possibility that polyphenol supplementation, which may modulate global and not just mitochondrial ROS, attenuated these ROS-dependent pathways provides a biologically plausible explanation for the lack of additive effect observed. However, this did not affect any of the physiological training adaptations (lean mass, strength, VO₂max etc.).

The increase in MRC following training is often attributed to increased mitochondrial content, or rather, mitochondrial volume density. In the present study, however, CS activity remained unchanged and therefore does not account for the observed increase in MRC. While CS activity is commonly used as a proxy for mitochondrial content [40], electron microscopy remains the gold standard for assessing mitochondrial volume density, and we cannot exclude the possibility of an undetected increase. Importantly, the observed increase in muscle mass, particularly the enlargement of type II fiber cross-sectional area, without a reduction in CS activity, suggests that total mitochondrial content may have increased proportionally to fiber growth, thereby maintaining overall density. Thus, our findings point to a distinction between mitochondrial content (the absolute amount of mitochondria) and mitochondrial volume density (the proportion of fiber volume occupied by mitochondria), with training-induced gains in respiratory capacity potentially reflecting improved mitochondrial quality, rather than changes in density per se. However, no change was observed in TFAM gene expression following the training intervention, underlining the lack of a response in mitochondrial biogenesis-related pathways.

Consistent with our results, some previous studies have shown an increase in MRC following RT [14,[41], [42], [43]], while others showed no changes [22,44], and mitochondrial content has been found to either decrease [45] or remain unchanged [14,22,41,43,46] following RT. The increase in MRC with no corresponding change in CS activity may therefore be related to adaptations within the mitochondria. Indeed, some previous studies have shown that resistance-trained individuals display higher mass-specific MRC [43], coupling efficiency [43], and cristae density [47] than untrained individuals, despite no differences in mitochondrial volume density. Though speculative, another explanation could be related to an increase in respiratory supercomplex formation, which has been observed following endurance training [48], but this remains to be investigated in relation to RT. It must be noted that the present study employed both RT and HIIT, and therefore, the effects cannot be credited exclusively to RT. HIIT is a very potent stimulator of mitochondrial biogenesis, and HIIT programs usually lead to increased mitochondrial content and volume density [49]. Since mitochondrial content did not increase in the present study, and HIIT comprised only a minor fraction of the total training volume, the observed adaptations are most likely attributable to RT or the combined training stimulus, rather than to HIIT alone. Moreover, HIIT may also increase intrinsic mitochondrial capacity [50], and it is possible that the combination with RT provided a stronger stimulus for intrinsic mitochondrial adaptations. Nonetheless, one cannot draw conclusions on the modality-specific contributions to the observed adaptations, and future studies wishing to disentangle differences between exercise modalities must employ groups of just one exercise modality.

Collectively, our findings indicate that RT in combination with HIIT increases mitochondrial capacity, and that the lack of changes in mitochondrial content may be related to intrinsic mitochondrial adaptations. Polyphenol supplementation alone may be associated with increased MRC, but further research is warranted to reach this conclusion.

4.2. Redox status

There was no difference in H₂O₂ emission following the supplement loading phase between groups, indicating that 30 days of supplement intake did not have any noticeable muscle antioxidant effects. To date, no other studies have examined the effect of polyphenol supplementation on ROS production directly in skeletal muscle. However, some studies have reported a reduced exercise-induced increase in protein carbonyls and thiobarbituric reactive acid substances (TBARS) following sweet potato- or cherry-derived polyphenol supplementation [51,52], while others show no difference in TBARS or 3-NT with cherry-derived polyphenols [53].

On the other hand, the training intervention led to a significant reduction in H₂O₂ emission of approximately 20 % in both groups. To our knowledge, only two previous studies have directly assessed the effects of RT on skeletal muscle H₂O₂ emission, both reporting no changes after 10–12 weeks of training [22,23], albeit in one study H₂O₂ was not normalized to oxygen flux. The discrepancies may also be attributable to methodological limitations in the previous studies, both of which included only 10–11 participants who undertook RT, compared with the 20 participants per group in the present study. Moreover, one study employed isolated mitochondria, which are known to show divergent results from permeabilized fibers [54].

The reduction in H₂O₂ emission after training observed in the present study is likely attributable to intramuscular adaptations, such as enhanced antioxidant defenses. Though we did not detect changes in catalase, CuZnSOD or MnSOD mRNA expression, previous studies have reported increases in antioxidant enzymes following RT at both mRNA [19] and protein [18,55] levels. A limitation to these findings is that the POST biopsy was obtained two days after the final training session, which may have prevented us from capturing transient changes in mRNA expression that typically occur within hours after exercise. Combined with the fact that we assessed only three endogenous antioxidants enzymes, this means we cannot exclude the possibility that the intervention enhanced endogenous antioxidant capacity in ways not detected by our measurements. Another possible explanation is changes within the mitochondria, such as increased respiratory efficiency. This interpretation is supported by our finding that MRC was increased with no concomitant increase in mitochondrial content, as discussed previously, perhaps because of increased cristae density or supercomplex formation, which limits ROS production [56,57]. Moreover, the reduction in H₂O₂ emission when normalized to oxygen flux could indicate a lower proportion of electrons leaking to form O₂⁻ and subsequently H₂O₂ per unit of oxygen consumed. As previously mentioned, the effects cannot be attributed solely to RT, as the protocol also included HIIT, which has been shown to increase antioxidant capacity in both plasma [58,59] and muscle [27,60].

Interestingly, the reduction in H₂O₂ emission did not translate into a reduction in plasma markers of oxidative damage, as MDA was increased and 3-NT was unchanged following the training intervention. The elevated MDA levels may reflect a residual effect of RT, as measurements were taken ∼48 h after the final training session. Previous work has shown that RT increases plasma MDA 24 h post-exercise in young men [61], and it is plausible that levels remain elevated longer in older adults due to slower recovery. Conversely, chronic RT has also been reported to reduce resting plasma MDA [62], although these findings were also in young men, who may differ from older adults in their oxidative stress responses to exercise. Nonetheless, RT has been shown to reduce other types of plasma oxidative damage markers in older adults, such as TBARS and lipid hydroperoxides [63,64], 4-hydroxynonenal (4HNE) [19], and myeloperoxidase (MPO) [65].

In general, only few exogenous antioxidants were affected by the supplement or training intervention. Training increased α-tocopherol (vitamin E) levels in the placebo group, consistent with reports habitual physical activity is positively associated with plasma α-tocopherol concentrations in elderly individuals [66]. As a major lipid-soluble antioxidant, α-tocopherol protects cell membranes against lipid peroxidation and oxidative damage, and higher vitamin E status has been associated with reduced risk of age-related diseases and greater longevity [67]. Elevated α-tocopherol levels have also been reported after high volumes of exercise, accompanied by increases in oxidative damage markers [68], suggesting that the rise may be mediated by exercise-induced oxidative stress. This interpretation is supported by the present observation of increased plasma MDA. Some previous studies also have shown that acute exercise increases plasma α-tocopherol levels [69,70], but these may be confounded by exercise-induced hemoconcentration; a reduction in plasma volume that occurs as fluid shifts from the vascular compartment to active tissues. This transient decrease in plasma volume concentrates circulating solutes, including lipoproteins and fat-soluble vitamins such as α-tocopherol, thereby artificially elevating their measured plasma concentrations without reflecting a true increase in total body content. Indeed, we also observed exercise-induced increases in most plasma vitamins, accompanied by elevations in hematocrit and total protein. Therefore, these changes are likely attributable to hemoconcentration rather than a true rise in circulating vitamin levels.

In the polyphenol group, α-tocopherol did not change after training. This could reflect a blunting effect of the supplement, as discussed previously, or alternatively, a supplement-induced rise during the loading phase that limited further adaptation. Although not statistically significant, α-tocopherol levels increased in the polyphenol group after supplementation, resulting in higher pre-training values compared to the placebo group, a difference that disappeared following the training intervention.

The training intervention also increased ascorbic acid (vitamin C) levels in both groups. This aligns with a previous study showing that six months of resistance training significantly elevated vitamin C levels in elderly individuals, even when combined with an antioxidant supplement [71]. These findings suggest that vitamin C is highly responsive to training, and its increases are not hindered by antioxidant intake. Vitamin C is another important antioxidant associated with lower mortality and CVD [72], and it acts in synergy with vitamin E to regulate oxidative stress [73].

Unexpectedly, β-cryptoxanthin and retinol levels declined after the training intervention, but only in the polyphenol group, suggesting that the supplement may have influenced the metabolism of these vitamins. Some evidence indicates that certain polyphenols can interfere with carotenoid absorption [74], which could partly explain the reduction in β-cryptoxanthin over time. Though the underlying mechanism for this decrease remains speculative, retinol and β-cryptoxanthin levels decreased in 14 of 20 participants, in the polyphenol group. It is important to keep in mind that we also observed some degree of variation in measurements of some micronutrients. It has previously been shown that interindividual variability can affect the response to antioxidant supplements and how these affect physiological measurements [75].

Finally, uric acid levels increased significantly at all time points in both groups following the training intervention. Although chronically elevated uric acid has been associated with renal- and cardiovascular disease [76,77], in this context the rise is more likely a transient adaptive response to the training stimulus. Since blood sampling was performed ∼48 h after the final training session, the increase likely reflects enhanced purine metabolism [78], as the participants were likely still recovering from the final training session. This interpretation is consistent with the evidence that uric acid can remain elevated for up to three days after intensive exercise [79], and that 12 months of RT has been shown to increase resting uric acid concentrations in older individuals [78]. Another possibility is that the rise in uric acid reflected a compensatory response to exercise-induced oxidative stress [78]. However, previous evidence shows that even a near-complete depletion of uric acid does not alter oxidative damage in gout patients, rendering this interpretation unlikely.

In summary, the training intervention reduced skeletal muscle H₂O₂ emission and increased plasma levels of some exogenous antioxidants, while polyphenol supplementation may have blunted some vitamin responses. These effects did not translate into lower plasma oxidative damage markers, in contrast to some previous findings, which may be related to the characteristics of the study population or the timing of sample collection.

4.3. Limitations of the study

As previously mentioned, a limitation of the present study is the lack of a muscle biopsy before the supplement loading phase. While the supplement loading-phase gives an opportunity to examine effects of the supplement alone on blood-based markers and physiological data, it makes it difficult to distinguish supplement-induced from training induced changes. However, training-related changes observed in the placebo group were not influenced by this, providing confidence that the overall training effects reported here are robust and not confounded by the supplementation protocol. Another consideration to the interpretation of mitochondrial bioenergetics is the use of permeabilized fibers, containing a mixture of type I and II muscle fibers, that may differ in mitochondrial characteristics. Though permeabilized fibers better reflect physiological conditions, isolated mitochondria offer a more targeted assessment of mitochondrial effects. However, isolated mitochondria may diverge from permeabilized fiber results and can exaggerate age-related mitochondrial changes [54]. The participants in the current study displayed a relatively mixed distribution of fiber types (51.3 ± 13.4 and 52.9 ± 12.4 % type I fibers in the placebo and polyphenol group respectively), something that did not change following the training intervention [32]. Finally, although participants were instructed to maintain their habitual diet, this was not tightly controlled. While the stable weight and BMI indicate no major changes in calorie-intake or energy expenditure, this does not account for potential changes in macro or micronutrient intake, which could potentially have influenced redox and vitamin levels.

5. Conclusion

This study demonstrates that a combined resistance- and interval-type training program improves aerobic capacity, muscle strength, lean mass, and mitochondrial capacity in middle-aged and older adults. These functional gains were accompanied by reduced skeletal muscle H₂O₂ emission and increases in certain circulating antioxidants, indicating enhanced mitochondrial efficiency and improved redox regulation. Together, these adaptations underscore resistance-based exercise as a potent intervention to counteract age-related declines in muscle health. Polyphenol supplementation was associated with higher mitochondrial capacity at baseline, but it did not augment the benefits of training and may have blunted some vitamin-related responses. Furthermore, improvements in muscle redox balance were not accompanied by reductions in plasma oxidative damage markers, which could reflect age-related differences in systemic recovery dynamics. Collectively, these findings highlight resistance-based training as a robust strategy to improve mitochondrial health and redox status in aging individuals, whereas the role of polyphenol supplementation remains uncertain and requires further investigation.

CRediT authorship contribution statement

Mathias Flensted-Jensen: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Validation, Visualization, Writing – original draft, Writing – review & editing. Cecilie Moe Weinreich: Investigation, Validation, Writing – review & editing. Ann-Sofie Kleis-Olsen: Investigation, Validation, Writing – review & editing. Filip Hansen: Investigation, Validation, Writing – review & editing. Nadia Stenner Skyggelund: Investigation, Validation, Writing – review & editing. Jeppe Rahbek Pii: Investigation, Validation, Writing – review & editing. Ryan Whitlock: Investigation, Validation, Writing – review & editing. Anders Karlsen: Investigation, Validation, Writing – review & editing. Arthur Ingersen: Investigation, Validation, Writing – review & editing. Dace Reihmane: Funding acquisition, Resources, Writing – review & editing. Daniela Weber: Investigation, Validation, Writing – review & editing. Tilman Grune: Investigation, Validation, Writing – review & editing. Olga Pivovarova-Ramich: Resources, Supervision, Writing – review & editing. Flemming Dela: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Flemming Dela reports financial support was provided by Asiros Nordic. Given his role as a member of the editorial board of redox Biology, co-author Tilman Grune had no involvement in the peer review of this article and had no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to another journal editor. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the supplementary data to this article:

Data availability

Data will be made available on request.