Coenzyme Q10 supplementation raises plasma levels without improving mitochondrial function in older adults

Abstract

Mitochondrial function is important to healthy aging, as it influences energy metabolism, oxidative stress, and physical performance. With age, mitochondrial function and biosynthesis of coenzyme Q10 (CoQ10) may change. CoQ10 serves as a key antioxidant and component of the electron transport system. Supplementation with CoQ10 may help preserve mitochondrial function and support healthy aging. Forty older community-dwelling adults (74 ± 4 years) received either daily oral CoQ10 supplementation (400 mg daily) or a placebo in a 12-week double-blinded, randomized, placebo-controlled design. Before and after the supplementation period, muscle biopsies were obtained. Subsequently, oral glucose tolerance tests (OGTT) and VO₂max tests were conducted. Mitochondrial respiratory capacity (MRC), mitochondrial H₂O₂ emission, and mitochondrial content were assessed in both isolated mitochondria and permeabilized muscle fibers. Levels and redox status of CoQ10 were measured in plasma, muscle tissue, and isolated skeletal muscle mitochondria. Additionally, resting metabolic rate, cognitive function, and body composition were investigated. Plasma levels of CoQ10 increased significantly without changes in redox status after the intervention. No changes between groups or time were observed in muscle and isolated mitochondria regarding MRC, H₂O₂ emission, mitochondrial content, and levels of CoQ10. Glucose homeostasis, VO₂max, and body composition were also unchanged. Twelve weeks of supplementation led to increased plasma levels of CoQ10, with unchanged levels in muscle tissue and isolated mitochondria. No differences in mitochondrial function, glucose homeostasis, and physical performance were found in a cohort of robust older adults.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s11357-025-02068-9.

Introduction

The global population is aging rapidly, with the number of individuals aged 65 years and older projected to more than double by 2050 (United Nations, Department of Economic and Social Affairs, 2023 [1]). While increased life expectancy is a significant public health achievement, it also raises an important challenge: ensuring that people not only live longer but also maintain health and functional independence as they age. This concept of “healthy aging” encompasses physical, cognitive, and metabolic resilience and has become a key goal in aging research [2].

At the cellular level, aging is closely linked to mitochondrial dysfunction, a hallmark of aging that may contribute to physical frailty and loss of physiological reserve [3–5]. Mitochondrial dysfunction, a broad term encompassing alterations in several key aspects of mitochondrial function, such as elevated oxidative stress and reduced capacity for energy production, is often observed with aging and may contribute to frailty [6–9]. Given the central role of mitochondria in energy metabolism, strategies aimed at preserving or improving mitochondrial function are considered critical for promoting healthy aging [3, 4]. One such strategy involves increasing antioxidant availability, which may offer a promising approach to mitigate oxidative stress and improve mitochondrial function [10]. Among them, coenzyme Q10 (CoQ10) has attracted particular interest due to its dual role as both an essential endogenous component of mitochondrial oxidative phosphorylation and a potent antioxidant in the lipid environment. Several studies have demonstrated the beneficial effects of dietary CoQ10 supplementation, particularly on cardiovascular health. Importantly, CoQ10 biosynthesis decreases with aging in rodents [11, 12], but in humans the evidence is not clear. Aside from the intrinsic biosynthetic capacity, aging is also associated with multiple factors (chronic diseases, drug interactions, oxidative stress), making individuals more prone to secondary deficiencies. Consequently, the beneficial effects of dietary supplementation in older individuals may be of greater significance because of the potential impact in supporting healthy aging and counteracting age-related degenerative diseases. On the other hand, CoQ10 bioavailability is limited due to its high lipophilicity, with less than 5% being absorbed through the gastrointestinal barrier. Nonetheless, the evidence suggests that supplementation significantly increases CoQ10 plasma levels [13, 14] where it is vehiculated in plasma lipoproteins, mainly in LDL. Remarkably, cellular uptake seems to display a tissue-specific uptake, with skeletal and cardiac muscle showing the lowest ability to incorporate exogenous CoQ10. Over the past decades, most studies investigating the bioenergetic functions of coenzyme Q10 have been performed in isolated mitochondria or cultured cells, where supplementation and functional assays can be tightly controlled. In contrast, in vivo studies remain more limited, particularly in humans. Clinical investigations directly assessing mitochondrial respiration or reactive oxygen species (ROS) production after CoQ₁₀ supplementation are scarce, and those conducted in humans represent only a very small fraction of the overall literature [14, 15], and these studies have primarily involved middle-aged individuals (range, 49–66 years). To date, no study has explored the impact of CoQ10 supplementation on mitochondrial function in a population above 70 years of age. This study aims to fill that gap by examining the effects of CoQ10 using high-resolution respirometry and fluorometry, including assessment of intramuscular CoQ10 levels in community-dwelling older adults.

Given the potential decline in CoQ10 levels and increased oxidative stress associated with the population of interest, we hypothesize that supplementation may have an effect on reducing mitochondrial ROS production and increasing plasma and muscle CoQ10 content.

Materials and methods

Approval

The project received approval from the Ethics Committee of Copenhagen (H-21071824) and was conducted at the Department of Biomedical Sciences, University of Copenhagen, in accordance with the guidelines of the Helsinki II Declaration.

Protocol and study design

The study was designed as a double-blinded, randomized, placebo-controlled human trial. Forty community-dwelling older (> 70 years) well-preserved participants with varying habitual activity levels (assessed by IPAQ) were included (20 males and 20 females) with a BMI of 25–30. Informed consent was obtained from all subjects involved in the study. To confirm eligibility, participants underwent a screening process that included completing questionnaires about health, lifestyle, and physical activity and undergoing health assessments, including anthropometric measures, blood pressure, and an ECG. Exclusion criteria included cardiovascular disease, smoking, treatment with statins, and pharmaceuticals known to interfere with the experimental results. However, mild to moderate hypertension and antihypertensive medication were accepted. Eligible participants were then randomly assigned to receive either 2 × 200 mg/day of oral Coenzyme Q10 supplementation or a placebo (soybean oil) for 12 weeks. Both of which were produced under Danish pharmaceutical control. Delivered in soft light-proof gelatin capsules in vegetable oil and the verum supplementation as ubiquinone. To ensure a stable and high Q10 plasma concentration during the intervention, participants were told to take two capsules in the morning and two more in the evening. Participants attended two identical experimental sessions: one prior to the supplementation period and the other the morning after completing the last dose of the 12 weeks of supplementation.

Participants reported to the laboratory after an overnight fast (10 h) and were instructed to refrain from strenuous exercise and alcohol for 24 h prior to the test. Participants were instructed to fill out a diet registration the day prior to the first test day and to follow the same diet before the second test day. The day began with a dual-energy x-ray absorptiometry (DXA) (Lunar iDXA, G&E Medical Systems Lunar, WI, USA, Encore Version 14.10.022) scan to measure body composition, followed by measurement of resting metabolic rate (RMR) by a ventilated hood system (COSMED, Q-NRG, Italy) after resting in a supine position. Next, a muscle biopsy was performed on the vastus lateralis muscle with a Bergström needle and manual suction [16]. Biopsies were separated into three portions and placed in different isolation buffers: sucrose (100 mM, KCl 100 mM, Tris–HCl 50 mM, KH2PO4 1 mM, EGTA 0.1 mM, BSA 0.2%, at pH 7.4) for analysis of isolated mitochondria, BIOPS (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 analysis of permeabilized fibers, or snap frozen in liquid nitrogen for later bulk analysis. Next, a venous catheter was inserted into the cubital vein to collect blood samples for an oral glucose tolerance test (OGTT), during which cognitive function was also assessed using the Montreal Cognitive Assessment (MoCA). Participants remained in bed throughout the extended 3-h duration of the OGTT. The experimental day concluded with a maximal oxygen consumption (VO₂max) test and a 30-s sit-to-stand test to assess cardiovascular fitness and muscle function. The exercise test was performed on a bicycle ergometer (COSMED, E100 ergoline, Germany). Pulmonary gas exchange was collected in a breath-by-breath analyzer system (Quark CPET, COSMED, Italy). The VO₂max test began with an initial workload of 35 W, which was increased by 20 W every minute until the participant reached voluntary exhaustion. To confirm that the participant had achieved their VO₂peak, at least two of the following criteria had to be met: a respiratory exchange ratio (RER) greater than 1.15, a plateau in VO₂ or heart rate despite increasing workload, or the inability to maintain a cadence above 60 rpm. At the end of the session, participants were provided with a light meal.

Measurements of Q10

Both CoQ10 levels and its oxidative status were quantified in all biological samples (muscle, plasma, and isolated mitochondria) by using an HPLC equipped with an electrochemical detector (ECD) (Shiseido Co. Ltd., Tokyo, Japan). The system allows the quantification of both reduced and oxidized forms of CoQ10 through a post-separation reducing column (Osaka Soda CQ-R; 20 mm × 2.0 mm ID) capable of fully reducing the ubiquinone peak and enabling discrimination between the two forms, as described by Andreani et al. [17]. The method used was adapted according to the biological matrix tested. In particular, plasma CoQ10 was extracted and quantified following the protocol described by Orlando et al., with minor modifications. Briefly, 50 µL of plasma was extracted with 250 µL of 1-propanol and vortexed for 30 s. After centrifugation at 20,900 × g for 2 min at 4 °C, 40 µL of the supernatant was injected. The mobile phase for elution consisted of 50 mM sodium perchlorate in methanol/isopropanol (95:5, v/v) at a flow rate of 0.32 mL/min. Differently, mitochondrial samples were eluted by using a phase composed of 50 mM sodium perchlorate in methanol/isopropanol (90:10, v/v) at a flow rate of 0.24 mL/min. Finally, intramuscular CoQ10 quantification was performed by adding 1 mL of 1-propanol and a stainless steel bead (7 mm diameter, Qiagen) for each sample extraction [18]. Biopsies were then lysed using a TissueLyser (Retsch MM 400, Haan, Germany) at 30 Hz for 2 min, and the extracted CoQ10 was quantified using the plasma protocol previously described. Plasma, mitochondrial, and intramuscular CoQ10 levels were normalized for cholesterol, CS activity, and dry weight, respectively. Oxidative status was expressed as percentage of ubiquinone/total CoQ10.

Mitochondrial analyses

Permeabilization of muscle fibers

The permeabilization of muscle fibers was carried out as previously described [19]. After permeabilization with saponin (Sigma-Aldrich, MO, USA; 5 mg saponin/1 mL BIOPS), the tissue (~ 2 mg of wet weight, exact weight was noted for normalization) was added to each oxygraph chamber (Oxygraph-2k, Oroboros, Austria). All chambers contained MIR05 (sucrose (110 mM), potassium lactobionate (60 mM), EGTA (0.5 mM), MgCl₂·6H₂O (3 mM), taurine (20 mM), KH₂PO₄ (10 mM), HEPES (20 mM), BSA (1 g/L), pH 7.1) and were held at 37 °C for high-resolution respirometry and fluorometry. Absolute measurements were normalized to mg wet weight of tissue within the chamber. All analyses were initiated immediately after permeabilization.

Isolation of mitochondria

The mitochondria were isolated from the biopsy via multiple centrifugation steps as previously described [20] and diluted in suspension buffer (sucrose 75 mM, mannitol 225 mM, Tris base 10 mM, EDTA 0.1 mM, at pH 7.4). Ten µL of isolated mitochondria suspension was added to each chamber containing MIR05 at 37 °C for high-resolution respirometry or fluorometry.

High-resolution respirometry

All measurements were performed as duplicates using oxygraphs (Oxygraph-2k, Orobos, Austria). Measures on permeabilized fibers were carried out in hyperoxygenated chambers containing ~ 450 nmol O₂/mL. The following protocol was used: the addition of pyruvate (5 mM), malate (2 mM), and glutamate (10 mM) was used to determine state 2 respiration with complex I–linked substrates. Cytochrome C (0.01 mM) was added to control the integrity of the outer mitochondrial membrane (data not shown). The Cytochrome C response was deemed acceptable with an increase in O₂ flux ≤ 10%. Determination of state 3 with complex I–linked substrates was assessed by the addition of ADP in stepwise increments (5 mM) with a single addition of MgCl₂ (3 mM). Succinate (10 mM) was added to determine state 3 respiration with complex I + II–linked substrates. Finally, FCCP was added with stepwise increments (0.5 µM) to reach maximal uncoupling.

Measurements on isolated mitochondria were performed at atmospheric oxygen levels corresponding to 200 nmol O₂/mL using the following protocol: Addition of isolated mitochondria (10 µL), followed by malate (2 mM) and pyruvate (5 mM) to determine state 2 leak respiration. A non-saturating amount of ADP (200 µM) was then added to cause an increase in respiratory flux for the determination of the P/O ratio. When the flux returned to baseline levels, the oxygen levels inside the chambers were restored by opening the chambers. ADP (5 mM) and MgCl₂ (3 mM) were then added, followed by the addition of glutamate (10 mM) and then succinate (10 mM) for the determination of complex I + II–linked state 3 respiration.

Mitochondrial efficiency was determined by P/O ratio and/or RCR as described previously [21]. P/O ratios are presented as the mean value of two chambers.

High-resolution fluorometry

In the protocol for permeabilized fibers, blebbistatin (5 µM) was added to prevent muscle fiber contraction [22]. The protocol proceeded with the addition of superoxide dismutase (1.7 μM), Amplex Red (4 μM), and horseradish peroxidase (HRP) (4 μM), which catalyzes the dismutation of O₂• − to H₂O₂ + O₂ and produces resorufin, a fluorescent molecule. To determine state 2 H₂O₂ emission with complex I–linked substrates, pyruvate (5 µM) and malate (10 µM) were added. Succinate was then added in three steps (1–2–3 mM). To calibrate the fluorometer, an addition of 0.1 μM H₂O₂ was performed after the addition of HRP. Additionally, three 0.1 µM H₂O₂ additions were performed, allowing for detection of changes in the assay sensitivity.

Protocols for isolated mitochondria were similar, but blebbistatin was not added. The protocols for isolated mitochondria were performed as a carbohydrate-supported (pyruvate (5 μM) + malate (10 μM)) protocol, and a fatty acid–supported (palmitoyl carnitine (2 μM) replaces pyruvate (data not presented)).

Enzyme activities

CS and HAD activity were measured in freeze-dried dissected muscle tissue as previously outlined [23], and the results were expressed per gram of dry weight. The measurement of CS and HAD activity in isolated mitochondria, however, involved slight modifications in the preparation process. Briefly, the isolated mitochondria were thawed on ice, thoroughly mixed, and centrifuged for 1 min at 1000 RPM at 4 °C. A 5-µL aliquot of the pellet was then resuspended in 245 µL of homogenization buffer and remixed. The resulting dilutions were immediately analyzed on a Cobas 6000, C 501 system (Roche Diagnostics, Mannheim, Germany). The results were expressed per mg of wet weight muscle used in the isolation protocol. The specific yield from the isolation process in the present study is undetermined, although it has previously been demonstrated to be approximately 40–50% [24]. Therefore, CS and HAD activity from muscle and isolated mitochondria are not directly comparable. In this study, CS activity serves as a marker of mitochondrial content [25].

Blood analysis

Blood samples were collected into tubes containing anticoagulants (dipotassium (K2) ethylene diamine tetraacetic acid (EDTA)/aprotinin; tripotassium (K3) EDTA/aprotinin; lithium heparin; sodium fluoride (NaF)/potassium oxalate (K-oxalate)) and centrifuged at 4000 g for 10 min at 4 °C using a Rotina 380R centrifuge (Axeb Lab Solutions) to separate the plasma. The plasma was then transferred into Eppendorf tubes and stored at −80 °C until further analysis. Plasma samples were analyzed using the COBAS e601 analyzer with an electrochemiluminescence immunoassay (ECLIA).

Statistics

Data from respirometry and fluorometry were analyzed in DatLab and further processed in Excel templates. Statistics were performed in GraphPad Prism (version 9.5.1). Continuous data were checked for normality and presented as mean ± standard deviation (SD). A two-way ANOVA was used when two categorical variables were analyzed. A mixed model was chosen in case of missing data or more than two categorical variables. The latter case had group (treatment or placebo), time (pre and post), and substrate as fixed factors, and the subject as a random factor. The model analyzed the interaction between the independent variables and the dependent variable of interest. To further elucidate the differences among specific groups or conditions, we conducted Šidák’s post hoc multiple comparisons. The significance level was set at p < 0.05.

Results

Participants

Forty older overweight adults were included in a well-matched cohort, as shown in Table 1. There was no significant difference in basic characteristics, body composition, and physical performance between groups and timepoints, except for a similar increase in MoCA (Montreal Cognitive Assessment) performance with time in both groups (p = 0.01). The majority of participants exhibited low fitness levels relative to their age group, with eight subjects being in the very low range, 16 in the low range, and 16 in the average range [26] in a Danish population [27]. Activity levels were not to be considered low but remained unchanged during the intervention, as assessed by the IPAQ. Representing a group of robust community-dwelling older individuals, compliance with the intake of capsules was 98% in both groups (assessed by collecting remaining capsules and comparing to expected number left). No adverse effects were reported.

Table 1.

Subject characteristics

	Active		Placebo
	Pre	Post	Pre	Post
N (F/M)	20 (10/10)	-	20 (10/10)	-
Age (years)	74 ± 4	-	75 ± 4	-
BMI (kg/m2)	27.2 ± 2.5	27.3 ± 2.5	27.8 ± 2.8	27.8 ± 3.0
Fat percentage (%)	35.1 ± 4.8	35.3 ± 4.4	35.5 ± 5.8	35.6 ± 5.9
Fat mass (kg)	27.2 ± 5.9	27.2 ± 6.3	28.1 ± 6.9	28.2 ± 7.3
Lean mass (kg)	51.5 ± 8.9	51.6 ± 8.7	52.7 ± 10.7	52.7 ± 10.9
Visceral fat (g)	1.42 ± 0.92	1.41 ± 0.86	1.59 ± 1.03	1.58 ± 0.99
Systolic BP (mmHg)	137 ± 17	135 ± 15	137 ± 14	137 ± 11
Diastolic BP (mmHg)	82 ± 10	81 ± 7	83 ± 7	83 ± 8
Resting heart rate (BPM)	62 ± 8	63 ± 8	62 ± 8	63 ± 8
VO2peak (L O2/min)	1.95 ± 0.45	1.95 ± 0.48	1.85 ± 0.57	1.95 ± 0.52
Fitness level (mL O2/min/kg)	25.2 ± 5.2	25.4 ± 5.0	22.4 ± 5.5	23.4 ± 4.4
Wattmax (watt)	140 ± 39	141 ± 38	144 ± 46	147 ± 47
RMR (kcal/day)	1444 ± 243	1445 ± 248	1546 ± 276	1570 ± 298
Sit-to-stand test (reps/30 s)	15 ± 3	16 ± 3	16 ± 4	17 ± 4
IPAQ (MET min/wk)	3096 ± 914	3243 ± 899	2621 ± 1123	2670 ± 1206
MoCA (points)#	25 ± 3	26 ± 3	25 ± 2	26 ± 2

Subject characteristics for the active and placebo groups. Data presented as mean ± SD

Abbreviations: N, numbers; BMI, body mass index, BP blood pressure, VO2peak peak oxygen uptake, Wattmax maximalload achieved during the maximal oxygen consumption test, RMR resting metabolic rate, MoCA Montreal Cognitive Assessment

#effect of time p<0.05

The plasma profile is shown in Table 2. Lipid profile markers were not different between groups and timepoints. The levels of total cholesterol and LDL were slightly elevated according to medical guidelines in Denmark (> 5 mmol/L and > 2.6 mmol/L, respectively) [28] None of the subjects received treatment for hypercholesterolemia.

Table 2.

Plasma markers

	Active		Placebo
	Pre	Post	Pre	Post
Total cholesterol (mmol/L)	5.2 ± 0.6	5.3 ± 0.9	5.3 ± 0.6	5.5 ± 0.6
HDL (mmol/L)	1.6 ± 0.5	1.5 ± 0.5	1.5 ± 0.3	1.4 ± 0.3
LDL (mmol/L)	3.3 ± 0.6	3.5 ± 0.8	3.6 ± 0.5	3.7 ± 0.5
Triglycerides (mmol/L)	1.1 ± 0.4	1.1 ± 0.5	1.2 ± 0.4	1.2 ± 0.4
FFA (µmol/L)	553 ± 141	505 ± 151	572 ± 184	561 ± 161
Glycerol (µmol/L)	55 ± 22	47 ± 15	55 ± 26	54 ± 22
HsCRP (mg/L)	1.6 (1.1–2.7)	1.4 (0.8–3.8)	1.1 (0.8–2.3)	1.2 (0.6–2.9)

Plasma markers for both groups: active and placebo. Data presented as mean ± SD for parametric data and median (IQR) for non-parametric data. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; FFA, free fatty acid; hsCRP, high-sensitivity C-reactive protein

Measurement of Q10

CoQ10 content in plasma significantly increased in the group given the active compound as expected; both total concentration (Table 3, p < 0.0001) and normalized to plasma cholesterol (Fig. 1 A, p < 0.0001). CoQ10 content in skeletal muscle normalized to the dry weight of muscle (Fig. 1B) and CoQ10 content normalized to citrate synthase activity (Fig. 1C), did not change across the intervention. Although not significant, total CoQ10 content in isolated mitochondria increased by 16% in the active group (p=0.4) (Table 3).

Table 3.

CoQ10 content and redox status

	Active		Placebo
	Pre	Post	Pre	Post
Total concentration
Plasma (nmol/L)	1065 ± 276*	5254 ± 2679	1047 ± 344*	1080 ± 378*
Skeletal muscle (nmol/L)	1216 ± 222	1219 ± 281	1306 ± 359	1312 ± 327
Mitochondria (µmol/mL)	25 ± 10	29 ± 12	23 ± 11	24 ± 10
Redox status (OxCoQ10/totalCoQ10(%))
Plasma (%)	9 ± 3	8 ± 2	10 ± 4	10 ± 3
Skeletal muscle (%)	93 ± 4	93 ± 5	92 ± 7	94 ± 4
Isolated mitochondria (%)	65 ± 8	62 ± 7	66 ± 8	63 ± 8

CoQ10 content and redox status for both groups: Active and placebo. Redox status calculated as: oxidized CoQ10/total CoQ10 × 100.  Data presented as mean ± SD. *p<0.0001 different from CoQ10 post

Fig. 1.

Normalized CoQ10 content in A plasma normalized to plasma concentration of cholesterol, B skeletal muscle normalized to amount of dry weight muscle, and C isolated mitochondria normalized to citrate synthase activity in the isolated mitochondria. ****Effect of time, p < 000.1

Reduced CoQ10, ubiquinol, is the active antioxidant, whereas ubiquinone is the oxidized form. The redox status of CoQ10 did not differ between groups and time points. The results were calculated as oxidized CoQ10 divided by total CoQ10 and reported in percentages.

Mitochondrial function

Mitochondrial content and enzymatic activity

Muscle citrate synthase (CS) and 3-hydroxyl-CoA dehydrogenase (HAD) activity were not significantly altered after the intervention (p = 0.7; p = 0.3, respectively) nor different between groups (p = 0.2; p = 0.4, respectively). Neither were CS and HAD activity altered in the isolated mitochondria (intervention: p = 0.9; p = 0.3, and group: p = 0.5; p = 0.8, respectively) (Table 4).

Table 4.

Enzymatic activity in skeletal muscle and isolated mitochondria

	Active		Placebo
	Pre	Post	Pre	Post
CS activity (µmol/g/min)
Isolated mitochondria	7.9 ± 3.1	8.1 ± 1.7	7.7 ± 3.6	7.3 ± 3.4
Muscle tissue	122 ± 38	125 ± 41	116 ± 26	109 ± 30
HAD activity (µmol/g/min)
Isolated mitochondria	11.3 ± 4.6	10.7 ± 2.0	11.4 ± 5.4	10.1 ± 4.9
Muscle tissue	148 ± 35	142 ± 40	139 ± 38	132 ± 40

CS and HAD activity in skeletal muscle tissue and isolated mitochondria. Data presented as mean ± SD

Mitochondrial H₂O₂ emission

Absolute H₂O₂ emission was unaffected by the intervention, and not different between the groups in both isolated mitochondria and permeabilized fibers. This was observed with both complex I–linked substrates (CI_leak) and complex I + II–linked substrates (CI + II_leak) at the highest concentration of succinate (3 mM) in both assays (Fig. 2 A, B). Additionally, no differences were found in isolated mitochondria when total H₂O₂ emission was normalized to oxygen flux (Fig. 2 C).

Fig. 2.

H₂O₂ emission in isolated mitochondria (left panel) and permeabilized fibers (right panel). Total mitochondrial ROS production in isolated mitochondria and permeabilized fibers (A and B, respectively) and mitochondrial H₂O₂ emission relative to oxygen consumption during oxidative phosphorylation (C and D, respectively)

Surprisingly, we observed a significant decrease in H₂O₂ emission in the placebo group with complex I–linked substrates when normalized to oxygen flux (Fig. 2D, placebo pre, 0.035 ± 0.028; placebo post, 0.017 ± 0.06 pmol s⁻¹ mg⁻¹/O₂ flux, p = 0.01). No differences were observed with complex I + II–linked substrates (Fig. 2D). The same results were found when normalizing H₂O₂ emission to citrate synthase activity (data not shown). Furthermore, in a fatty acid–supported ROS protocol in isolated mitochondria, we found no significant changes in H₂O₂ emission (data not shown).

Mitochondrial respiration using high-resolution respirometry

Maximal respiratory capacity (MRC) in absolute values was not significantly affected by the intervention. Specifically, maximal respiration with complex I and I + II–linked substrates for isolated mitochondria showed p > 0.05 for both groups and time (Fig. 3 A). Maximal respiration with complex I–linked substrates, complex I + II–linked substrates, and after the addition of FCCP (uncoupler to achieve electron transport system capacity, ETS) in permeabilized fibers showed p > 0.05 for both groups and time (Fig. 3B). After normalization of the flux to mitochondrial content (citrate synthase as proxy) to gain insight into intrinsic mitochondrial respiration, no differences were observed in respiratory capacity in the preparations (Fig. 3 C, D).

Fig. 3.

Maximal respiratory capacity (MRC) in isolated mitochondria (A) and permeabilized fibers (B). Maximal respiratory capacity normalized to citrate synthase activity (intrinsic mitochondrial function) in isolated mitochondria (C) and permeabilized fibers (D). CI, complex I (malate, pyruvate, glutamate, ADP); CI + II (CI + succinate); ETS, electron transport system capacity (FCCP titration)

Mitochondrial efficiency

Mitochondrial efficiency in isolated mitochondria was assessed by P/O ratio and respiratory control ratio (RCR) (Table 5). We observed no significant differences between groups and time. Neither did we observe changes in leak respiration (data not shown). Likewise, in permeabilized fibers, mitochondrial efficiency was assessed by RCR (Table 5), showing no significant changes. Neither did we observe changes in ADP sensitivity (data not shown).

Table 5.

Indices of mitochondrial function

	Active		Placebo
	Pre	Post	Pre	Post
Isolated mitochondria
P/O ratio	2.9 ± 0.4	2.8 ± 0.4	3.0 ± 0.5	3.0 ± 0.5
RCR	22.0 ± 4.7	20.7 ± 3.0	22.4 ± 3.3	21.1 ± 4.2
Permeabilized fibers
RCR	8.9 ± 2.7	9.3 ± 2.5	8.1 ± 1.6	9.9 ± 5.3

Indices of mitochondrial function. P/O ratio and RCR were calculated as described in the reference in the “Materials and methods” section. Data presented as mean ± SD

Glucose homeostasis

All subjects in both groups had normal fasting glucose and HbA1c levels before and after the intervention. There were no significant changes in all indices of glucose homeostasis after the intervention, nor were there any differences between groups (see Table 6).

Table 6.

Indices of glucose homeostasis

	Active		Placebo
	Pre	Post	Pre	Post
Fasting glucose (mmol/L)	5.5 ± 0.5	5.6 ± 0.6	5.7 ± 0.7	5.7 ± 0.8
HbA1c (mmol/mol)	37.2 ± 2.9	37.2 ± 2.1	37.7 ± 3.4	37.3 ± 2.9
Glucose AUC (mmol/L × 120 min)	1025 ± 145	1016 ± 187	1044 ± 196	1058 ± 209
Glucose AUC (mmol/L × 180 min)	1414 ± 224	1397 ± 273	1438 ± 286	1452 ± 310
Fasting insulin (pmol/L)	44 ± 19	47 ± 23	46 ± 25	45 ± 27
Insulin AUC (nmol/L × 120 min)	50 ± 21	51 ± 19	46 ± 21	49 ± 24
Insulin AUC (nmol/L × 180 min)	67 ± 25	69 ± 25	63 ± 33	68 ± 38
HOMA-IR	1.6 ± 0.8	1.7 ± 1.0	1.6 ± 0.9	1.6 ± 0.9
HOMA-β (%)	63 ± 23	66 ± 27	58 ± 31	59 ± 41
Matsuda Index	6.0 ± 3.5	5.8 ± 3.8	6.9 ± 4.5	6.7 ± 4.8

Indices of glucose homeostasis derived from OGTT. AUC calculated based on the trapezoidal method. HOMA-IR calculated as [glucose (mmol/L) × insulin (µU/L)]/22.5. HOMA-β calculated as 20 × fasting insulin (μIU/mL)/fasting glucose (mmol/mL) − 3.5. Matsuda Index calculated as 10,000/√[fasting glucose (mmol/L) × fasting insulin (pmol/L)] × [mean glucose (mmol/L) × mean insulin (pmol/L)]. Data presented as mean ± SD. Abbreviations: AUC, area under the curve; HOMA-IR, homeostatic model assessment for insulin resistance; HOMA-β, homeostatic model assessment for beta cell function; OGTT, oral glucose tolerance test

There were unequivocally no sex-based differences in the effect of the intervention on any of the parameters investigated.

Discussion

CoQ10 is widely recognized for its role in mitochondrial electron transport and as an antioxidant, making it a promising candidate for enhancing mitochondrial function and improving metabolic health in aging populations. Moreover, no study has employed such extensive analyses on both plasma, skeletal muscle, and isolated mitochondria from skeletal muscle, combined with a comprehensive phenotypic assessment that includes measures of glucose homeostasis, body composition, cognitive function, and aerobic fitness. In this study, we investigated whether 12 weeks of oral CoQ10 supplementation would result in a significant increase in muscle CoQ content paralleled by increased plasma levels, and whether supplementation could enhance mitochondrial bioenergetics, reduce oxidative stress, and improve glucose homeostasis and physical performance in older adults. Compliance with the intake was high, and supplementation resulted in a significant increase in plasma CoQ10 levels; however, no changes were observed in muscle or mitochondrial CoQ10 content, mitochondrial respiratory capacity, H₂O₂ emission, or mitochondrial content. We observed a statistically significant 1-point increase in MoCA scores in both groups, which likely indicates a learning curve, even though the test remains effective after retesting at 3 months. It is worthwhile to delve into these results, as the cutoff point for a normal result was reached after the intervention (from 25 to 26 points). Furthermore, glucose homeostasis, VO₂peak, and body composition remained unchanged. These findings suggest that CoQ10 exhibits limited uptake in skeletal muscle and functional impact under the conditions tested, despite increased systemic availability.

Tissue-specific content of CoQ10

The current finding of increased plasma CoQ10 after supplementation has been previously demonstrated in human studies (from acute up to 8 weeks of supplementation) [14, 18, 29–32]. The range of the baseline values (total CoQ10, range, 463–1858 nmol/L; CoQ10/chol, range, 86–296 nmol/mol) was similar to reference values, as suggested by others [33, 34]. In the present study, the plasma increase did not correlate with changes in CoQ10 concentrations in skeletal muscle or isolated mitochondria. Moreover, despite the rise in systemic CoQ10, redox status across the three different preparations remained unaltered.

Skeletal muscle CoQ10 levels remained unchanged, which is consistent with prior reports highlighting the difficulty of increasing intramuscular CoQ10 through supplementation [14, 35, 36], although some studies have shown a non-significant tendency [36]. Several factors may contribute to this, including the specific formulation of CoQ10. For instance, considerable effort has been invested in gathering evidence supporting improved bioavailability in various formulations, delivery systems, and the use of precursors [37, 38]. Despite variability in response, those who exhibited increases were not consistently those with the lowest baseline levels. Similarly, CoQ10 levels in isolated mitochondria were unaltered by supplementation. To our knowledge, this has only been previously measured in isolated mitochondria from human skeletal muscle in early work by Svensson et al. in healthy young participants [35]. In line with our results, Svensson and colleagues did not detect an increase in CoQ10 in muscle and isolated mitochondria.

While CoQ10 levels have been reported to decline with age in cardiac tissue [10], similar findings in skeletal muscle are sparse. There is a clear need for more data across a broader age spectrum and between sexes to understand physiological ranges and the relevance of age-related changes in muscle CoQ10 content. Furthermore, little is known about the variability in CoQ10 concentrations between different biopsies from the same muscle, making it challenging to interpret subtle within-subject changes.

To our knowledge, no studies have specifically examined the effects in older adults (> 70 years). Here, we contribute to the existing literature by providing additional evidence that oral ingestion of current CoQ10 formulations does not increase muscle or mitochondrial content in muscle.

Mitochondrial function

Mitochondrial respiratory capacity

We show that elevated plasma levels of CoQ10 do not necessarily translate into improved mitochondrial bioenergetics when muscle uptake is minimal. Despite a robust increase in plasma CoQ10, respiratory capacity in both permeabilized skeletal muscle fibers and isolated mitochondria from skeletal muscle remained unchanged following supplementation. This consistency was observed across multiple respiratory states, including complex I– and complex I + II–supported oxidative phosphorylation. These findings contradict our initial hypothesis that CoQ10 supplementation would enhance mitochondrial respiration in an older population. No previous studies have directly examined the effects of CoQ10 supplementation on mitochondrial respiratory capacity in older individuals; however, these results are consistent with previous findings from our laboratory, which involved CoQ10 supplementing statin-treated middle-aged men and women [14]. While CoQ10 supplementation has been associated with improvements in indices of physical performance or reductions in fatigue indices in some studies [30, 39], these outcomes do not necessarily reflect underlying changes in mitochondrial respiration. The lack of change in respiratory capacity may be a logical consequence of the unaltered CoQ10 content in skeletal muscle and isolated mitochondria, as observed in our study. Although CoQ10 is a central component of the electron transport chain, facilitating electron transfer between complexes I/II and III, its availability may not be a limiting factor under normal conditions. In fact, endogenous CoQ10 levels, if not impaired, may already be sufficient to support maximal or near-maximal mitochondrial flux, implying a possible saturation effect at the functional level. This raises the question of whether there exists a threshold or “critical level” of CoQ10 below which respiration becomes impaired — a level that our study participants may not have reached.

Furthermore, we observed no changes in citrate synthase activity, a standard proxy for mitochondrial content [25], suggesting that mitochondrial biogenesis was not upregulated in response to supplementation.

Overall, these findings underscore the complexity of CoQ10 biology in human tissues and suggest that simply increasing systemic CoQ10 availability is unlikely to enhance mitochondrial function unless intracellular delivery, particularly into muscle mitochondria, can be improved.

Mitochondrial hydrogen peroxide (H₂O₂) emission

We hypothesized that mitochondrial H₂O₂ emission would decrease in both permeabilized muscle fibers and isolated mitochondria from skeletal muscle following CoQ10 supplementation, driven by an expected increase in intramuscular CoQ10 content. However, the absence of changes in muscle or mitochondrial CoQ10 levels in our cohort may explain the unchanged H₂O₂ emission observed post-intervention. This aligns with our findings on respiratory capacity and suggests that mitochondrial function is unlikely to be enhanced in muscle tissue after CoQ10 supplementation without improved intracellular delivery. Our results are consistent with those of Dohlmann et al. [14], who also reported no change in H₂O₂ emission in skeletal muscle following CoQ10 supplementation in middle-aged individuals [14]. In contrast, Pham et al. observed a significant reduction in H₂O₂ emission normalized to oxygen flux during complex I + II leak respiration after 6 weeks of supplementation with CoQ10 or MitoQ in middle-aged men [15]. Differences in age, supplementation duration, formulations used, and protocol design (e.g., no placebo group) may account for these divergent findings. Interestingly, we observed a significant reduction in H₂O₂ emission following the placebo condition. This unexpected outcome can be partially attributed to methodological limitations in the H₂O₂ emission assay, particularly under conditions of high oxygen tension. As discussed by Gnaiger et al. [40], elevated oxygen concentrations can artificially inflate ROS emission measurements. Although we employed consistent protocols across all analyses, this remains an important consideration. To our knowledge, mitochondrial H₂O₂ emission has not previously been investigated in isolated mitochondria from human skeletal muscle in the context of CoQ10 supplementation. Our data, therefore, contribute novel insights that appear to align with the challenges known in skeletal muscle fibers. It is also possible that baseline ROS emission in our cohort was not elevated to a degree where CoQ10 could exert a measurable effect. Establishing normative data for mitochondrial ROS production across different age groups and physiological states/diseases would be valuable to contextualize intervention effects.

Glucose homeostasis

CoQ10 supplementation did not affect glucose metabolism in our cohort. There were no significant changes in insulin or glucose area under the curve (AUC) during OGTT, nor in derived indices such as HOMA-IR or HbA1c. Most prior studies examining the benefits of CoQ10 on glycemic control have focused on individuals with type 2 diabetes or metabolic syndrome, with unequivocal results across multiple meta-analyses [41, 42]. A recent meta-analysis confirmed modest improvements in fasting glucose, insulin, and HbA1c in such populations [43]. However, these effects are typically absent in metabolically healthy individuals. Our participants were older but generally healthy, with no evidence of impaired glucose regulation at baseline. The lack of effect is therefore likely due to limited room for improvement, underlining that CoQ10’s metabolic benefits may depend on underlying dysfunction.

Considerations and limitations

Firstly, this study was strengthened by its randomized, double-blinded, placebo-controlled design, high adherence, and relatively long intervention period compared to similar trials. However, while we used a high daily dose of CoQ10 (commonly 100–300 vs. 400 mg per day), newer speculations within the field suggest that splitting it into three daily doses may optimize absorption, though this is less practical in real-world settings. Our study was not powered for subgroup analyses by sex, so caution should be exercised when interpreting the lack of findings. We did not assess specific mechanistic markers such as mitochondrial protein complexes (e.g., via blue native gels) or a broad panel of antioxidant enzymes and genes. Interpreting physical activity with IPAQ has limitations. IPAQ is known to overestimate physical activity in older adults [44], but there is a lack of qualified PROMs for this purpose [45]. As CoQ10 functions within a wider antioxidant network, additional oxidative stress sources and compensatory pathways may have had masking effects. Another key limitation is the lack of measurable CoQ10 uptake in muscle and mitochondria, which may explain the absence of functional effects. It remains unclear whether this is due to biological limits in healthy older adults or insufficient tissue delivery. A deeper understanding of CoQ10 transport and bioavailability in aging is needed. Chronological age alone may not be a sufficient model to attain decreased CoQ10 and increased ROS production. In this regard, evaluating mitochondrial-targeted supplementation in older adults with comorbidities may better determine its role in promoting healthy aging. Therefore, these results might not apply to frail, multimorbid, or metabolically impaired individuals typically seen in a clinical geriatric population.

Conclusion

In this study, we investigated the effects of 12 weeks of CoQ10 supplementation in well-functioning older adults, with a focus on mitochondrial function, glucose homeostasis, and physical performance. While plasma levels of CoQ10 increased significantly, no corresponding changes were observed in skeletal muscle or isolated mitochondria. Mitochondrial respiratory capacity, ROS production, and mitochondrial content remained unchanged. Additionally, glucose homeostasis, VO₂peak, and body composition remained unaffected by the intervention. These findings suggest that despite improved systemic availability, CoQ10 supplementation did not translate into measurable physiological benefits in the target tissues or functional outcomes in this population of robust older individuals. It is relevant that future studies investigate a typical geriatric population encountered in clinical settings and improve formulations for skeletal muscle CoQ10 delivery.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contribution

Malte Schmücker, Cathrine Tranberg, Jacob Borch, Mette Killerup Kaae, Michelle Damgaard, Mathias Flensted-Jensen, Ida Blom, Marco Morosetti, Sara Barbarossa, and Patrick Orlando performed the experiments, analyzed the data, and discussed the results. Malte Schmücker, Luca Tiano, Flemming Dela, Jørn Wulff Helge, and Steen Larsen were involved in the study conception, data analysis, and interpretation, as well as discussing the results and writing the paper. All authors contributed to the article and approved the submitted version.

Funding

Internal funding and Nordea-Fonden. Pharma Nord donated Coenzyme Q10 and a placebo.

Data availability

The data supporting the findings in the study are available from the first author (M. Schmücker) or last author (S. Larsen) upon reasonable request.

Declarations

Institutional review board statement

The project received approval from the Ethics Committee of Copenhagen (H-21071824) and was conducted at the Department of Biomedical Sciences, University of Copenhagen, in accordance with the guidelines of the Helsinki II Declaration.

Informed consent statement

Informed consent was obtained from all subjects involved in the study.

Competing interests

S. Larsen is consulting for Pharma Nord. The other authors declare no conflict of interest. The funders had no role in the design of the study, in the collection, analysis, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.