Respirometric Profiling of Muscle Mitochondria and Blood Cells Are Associated With Differences in Gait Speed Among Community-Dwelling Older Adults

Abstract

Background.

Gait speed provides an integrated measure of physical ability that is predictive of morbidity, disability, and mortality in older adults. Energy demands associated with walking suggest that mitochondrial bioenergetics may play a role in gait speed. Here, we examined the relationship between gait speed and skeletal muscle mitochondrial bioenergetics, and further evaluated whether blood-based bioenergetic profiling might have similar associations with gait speed.

Methods.

Participants in this study were comprised of two subsets (n = 17 per subset) and were overweight/obese (body mass index, 30.9 ± 2.37), well-functioning, community-dwelling older adults (69.1 ± 3.69 years) without major comorbidity. Gait speeds were calculated from a fast-paced 400 m walk test. Respiratory control ratios were measured from mitochondria isolated from leg skeletal muscle biopsies from one subset. Maximal respiration and spare respiratory capacity were measured from peripheral blood mononuclear cells from the other subset.

Results.

Individual differences in gait speed correlated directly with respiratory control ratio of mitochondria isolated from skeletal muscle (r = .536, p = .027) and with both maximal respiration and spare respiratory capacity of peripheral blood mononuclear cells (r = .585 and p = .014; r = .609 and p = .009, respectively).

Conclusions.

The bioenergetic profile of mitochondria isolated from skeletal muscle is associated with gait speed in older adults. Blood-based bioenergetic profiling is also associated with gait speed and may provide an alternative measure of mitochondrial function.

Gait speed integrates the effects of multiple subsystems including the central nervous system, peripheral nervous system, perception, muscle, bone/joint, and energy production/delivery.¹ It has been demonstrated to be predictive of future disability, cognitive deficits, institutionalization, falls, and mortality and is an important predictor of overall physical health and functional capacity and a key endpoint in clinical aging research.^2,3

Despite well-established health consequences associated with slow gait speed, little is known about its physiological determinants. Mitochondria are responsible for the majority of energy production in the body, specifically from oxidative phosphorylation. Aging is associated with declining mitochondrial bioenergetic capacity⁴; however, the role of mitochondrial function in overall health and physical ability, including gait speed, is unclear. Reliable assessments of mitochondrial function have the potential to inform on the role of bioenergetics in diminishing physical ability and may provide an objective way to monitor healthy aging.

There are numerous methods for measuring mitochondrial function in isolated organelles and whole cells.⁵ For example, measurements of adenosine triphosphate (ATP) provide direct assessment of bioenergetic output. Respirometric measures, on the other hand, can provide information about electron transport chain (ETC) activity. For isolated mitochondria, respiratory control is a commonly used index of ETC activity and reported by the increase in oxygen consumption after addition of ADP. Respiratory control is most often reported as a ratio, respiratory control ratio (RCR), of maximal capacity and respiration in the absence of substrate or upon inhibition of ATP synthesis. As a ratio, RCR can account for potential differences associated with mitochondrial isolation procedures. Analogous measures of respiratory control in intact cells require the use of chemical uncouplers because adenosine diphosphate (ADP) cannot readily cross the cell membrane. A preferred approach is to assess maximal oxygen consumption rate (Max-OCR) and spare respiratory capacity (SRC) using carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP). Recent studies show that measures of skeletal muscle mitochondrial bioenergetics correlate inversely with age, and directly with physical ability measured by gait speed and peak aerobic capacity.^6,7 Our objectives for this study were: (i) to determine whether the RCR of mitochondria isolated from leg skeletal muscle correlates with gait speed; and (ii) to investigate whether bioenergetic capacity (Max-OCR and SRC) measured from more easily accessible tissues (peripheral blood mononuclear cells [PBMCs]) could similarly correlate with differences in gait speed.

Methods

Participants

Two subsets of randomly chosen participants (n = 17 each) were included in this study of older (65–79 years), overweight and obese (body mass index [BMI] ≥ 28–35 kg/m²), sedentary men and women recruited to participate in a clinical trial of resistance training with or without dietary-induced weight loss. The assessments reported here were conducted at baseline, prior to intervention.

Eligible participants for both the parent clinical trial and substudy were in good health and had normal cognitive function, used no walking aids, and did not have uncontrolled diabetes or hypertension, abnormal liver or kidney function, or cancer requiring treatment in the past 2 years (see Supplementary Table 1 for detailed inclusion/exclusion criteria). The study was approved by the Wake Forest School of Medicine Institutional Review Board and all participants provided written, informed consent.

Gait Speed

Gait speed, calculated as m/s, was measured using a 400-m test. Participants were instructed to walk the 400 m distance (on a flat indoor surface 20 m in length) as quickly as possible, as prescribed by the parent study. Encouragement was given in a standardized fashion. The test has excellent reproducibility (intraclass correlation coefficient = 0.95) and directly correlates with measured maximal aerobic fitness.⁸

Muscle Biopsy and Blood Draw

Muscle biopsies and blood draws were performed in the morning after an overnight fast. A sample of vastus lateralis muscle (~100–120 mg) was extracted via percutaneous needle biopsy with local anesthesia. Tissues were processed immediately after collection. 8 mL of whole blood was collected into cell preparation tubes (Vacutainer) for PBMC separation.

Respirometry of Isolated Muscle Mitochondria

Vastus lateralis samples were visualized with a dissecting microscope and residual fat was manually teased apart and removed with fine forceps. About 50 mg of muscle tissue was used for mitochondrial isolation using differential centrifugation techniques based on the Chappell and Hansford method.⁹ Respirometric assays were performed using a Seahorse XF24-3 Analyzer (Billerica, MA), which measures the OCR (OCR) in a 24-well plate format. After isolation, 5 μg of mitochondrial protein, determined by bicinchoninic acid protein assay,¹⁰ was placed in each well of the Seahorse microplate. Microplates were centrifuged at 4ºC for 21 minutes at 2000G to promote attachment of mitochondria. During study protocol optimization, respiration driven by complex 1 using pyruvate/malate was compared with respiration driven by complex 2 using succinate/rotenone. Complex 2-driven respiration was consistently higher, similar to previous reports.¹¹ In order to ensure adequate sample size and standardization, all respirometric analyses of isolated mitochondria reported in this study were performed using succinate (10 mM) and rotenone (2 μM) to drive respiration via complex 2. ADP (2 mM) was added to induce state 3 respiration with saturating amount of substrate. This was followed by oligomycin (2 μM) addition which blocked complex 5 to provide a measure of state 4o respiration. The primary outcome for these measures was RCR, calculated as state 3/state 4o.¹¹ This measure can account for slight variations between mitochondrial sample preparations and experimental runs. In order to validate the presence of mitochondria in the final sample used for respirometric analysis, voltage-dependent anion channel 1 expression was verified by western blot at the end of each experiment. Voltage-dependent anion channel 1 enrichment was consistent between samples (see Supplementary Figure 1 for representative blot) and no samples were excluded from this study based on this verification. In a previous study, average isolated mitochondria RCR values for complex 2-driven respiration using the Seahorse XF system was 3.9 ± 0.6. We excluded two samples with RCR values <3 in order to ensure that samples included for analyses were within the expected RCR range based on previous work. As indicated in the results section, inclusion of these two samples in our analyses does not significantly alter our reported outcomes.

PBMC Respirometry

PBMCs were isolated from whole blood using Ficoll/isopaque. PBMCs were washed with phosphate-buffered saline and resuspended in XF assay buffer containing 1 mM Na⁺-pyruvate and 11 mM d-glucose (pH 7.4). A total of 500,000 cells were plated per well of the Seahorse microplate. Basal-OCR measures were followed by sequential addition of oligomycin (750 nM), FCCP (1 µM), and antimycin-A/rotenone (both 1 µM). Max-OCR was calculated after addition of FCCP, a potent mitochondrial uncoupler. The use of FCCP as a chemical uncoupler allows us to estimate maximal respiration. This is conceptually different from state 3 respiration which can only be measured when ADP can be exogenously introduced to mitochondria. SRC was calculated as the difference between Max-OCR and the Basal-OCR.

Statistical Analysis

Pearson correlation coefficients and partial correlations (individually adjusted for age, BMI, or gender) were calculated using SPSS v.21 (Armonk, NY) to examine relationships between the dependent variable (gait speed) and the independent variables (maximal respiration, SRC, and RCR). p Values of ≤0.05 were considered statistically significant.

Results

The demographics and bioenergetic data for each study subset are summarized in Table 1. Participants in both groups were of a similar age (p = .35), gender distribution (p = 1.00), and obesity status (p = .60) and had similar gait speeds (p = .32). Gait speed was plotted against skeletal muscle mitochondria RCR in Figure 1. Our results indicate a significant positive correlation indicating that higher RCR is associated with higher gait speed. Table 2 summarizes correlations between isolated mitochondria RCR and respiratory states 3 and 4o with gait speed. Partial correlations with independent adjustments for age, BMI, or gender are also presented. Two participants were excluded due to RCR values <3; when including these in the analysis, the correlations between RCR and gait speed remained significant (r = .507, p = .027, N = 19).

Table 1.

Demographic and Bioenergetic Characteristics of the Two Subsets of Participants Included in this Study

	Skeletal Muscle Subset (N = 17a)		PBMC Subset (N = 17)
% Female	41%		41%
	Mean ± SD	Range	Mean ± SD	Range
Age (years)	69.7±3.96	65–78	68.4±3.41	65–78
BMI (kg/m2)	31.1±2.41	27.2–35.0	30.7±2.33	27.0–34.9
Walk speed (m/s)	1.5±0.20	1.06–1.84	1.5±0.19	1.10–1.80
RCR	4.9±1.72	3.15–8.36
State 3 OCR (pmol/min/5 µg)	427.3±151.63	170.41–751.80
State 4o OCR (pmol/min/5 µg)	96.2±44.32	20.39–179.70
SRC (pmol/min/500,000 cells)			414.8±164.66	183.7–779.2
Max-OCR (pmol/min/500,000 cells)			589.3±213.44	312.8–1,006.7
Basal-OCR (pmol/min/500,000 cells)			174.5±57.47	91.0–298.0

Notes: BMI = body mass index; OCR = oxygen consumption rate; PBMC = peripheral blood mononuclear cell; RCR = respiratory control ratio; SRC = spare respiratory capacity.

^aTwo subjects were excluded from analysis based on the RCR cutoff <3.

Figure 1.

Gait speed (m/s) plotted against RCR (state 3/state 4o) of skeletal muscle mitochondria. RCR, respiratory control ratio.

Table 2.

Correlations Between Bioenergetics of Isolated Mitochondria (RCR, state 3, and state 4o) and Gait Speed Older Adults

Correlation	RCR		State 3		State 4o
	r	p Value	r	p Value	r	p Value
Pearson	.536	.027	−.13	.63	−.41	.09
Partial with age	.586	.017	−.12	.66	−.41	.12
Partial with BMI	.539	.031	−.10	.70	−.40	.12
Partial with gender	.514	.042	−.46	.08	−.56	.02

Notes: N = 17. Partial correlations were done independently of one another. BMI = body mass index; RCR = respiratory control ratio.

Gait speed was plotted against PBMC SRC (pmol/min/500,000 cells), Max-OCR (pmol/min/500,000 cells), and Basal-OCR (pmol/min/500,000 cells) (Figure 2A–C). Table 3 summarizes correlations between PBMC SRC, Max-OCR, and Basal-OCR and includes partial correlations with independent adjustments for age, BMI, or gender. Our results indicate that both SRC and Max-OCR are significantly positively associated with gait speed.

Figure 2.

Gait speed (m/s) plotted against (A) SRC (Max-OCR minus Basal-OCR; pmol/min/500,000 cells), (B) Max-OCR (pmol/min/500,000 cells), and (C) Basal-OCR (pmol/min/500,000 cells) measured in PBMCs. PBMC, peripheral blood mononuclear cell; SRC, spare respiratory capacity. OCR, oxygen consumption rate; FCCP, carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone.

Table 3.

Correlations Between PBMC Mitochondrial Bioenergetics (SRC, Max-OCR, and Basal-OCR) and Gait Speed in Older Adults

	SRC		Max-OCR		Basal-OCR
Correlation	r	p Value	r	p Value	r	p Value
Pearson	.609	.009	.585	.014	.425	.089
Partial with age	.598	.015	.573	.020	.416	.109
Partial with BMI	.612	.012	.585	.017	.427	.099
Partial with gender	.538	.032	.505	.046	.328	.216

Notes: N = 17. Partial correlations were done independently of one another. BMI = body mass index; OCR = oxygen consumption rate; PBMC = peripheral blood mononuclear cell; SRC = spare respiratory capacity.

Discussion

This study shows that mitochondrial function measured from myocyte mitochondria and circulating mononuclear cells is associated with gait speed in community-dwelling older adults. In previous studies, maximal skeletal muscle ATP production, measured by ³¹P imaging in 30 individuals⁶ and oxidative enzyme activity, measured by citrate synthase activity in 22 individuals¹² were correlated with gait speed (r = .40, p = .03; r = .45, p = .04, respectively) suggesting direct role for muscle mitochondrial metabolism in physical function. The same study reported that maximal respiration of permeabilized muscle fibers (measured by addition of ADP) and uncoupled respiration (measured by addition of FCCP), were not significantly correlated with gait speed (r = .25, p = .25; r = .31, p = .14, respectively). These results suggest that there are differences in the bioenergetic capacity of whole muscles and fibers. Our use of isolated organelles complements this work and indicates that intrinsic properties of mitochondrial ETC function are independently correlated to gait speed.

There are key differences between measuring mitochondrial function in isolated organelles and permeabilized cells. Use of isolated organelles eliminates differences in mitochondrial content and non-mitochondrial contributors to respiration that remain intact after permeabilization. While differences in the ratio of oxidative fibers exist at the muscular level, it remains unclear how such differences would affect the oxidative capacity of the isolated organelles we have studied. It has been reported that much of the loss of skeletal muscle cross-sectional area with age is due to decreased type II muscle fiber size,¹³ and in rats, typical type I skeletal muscles generally have lower oxidative capacity than type II skeletal muscles in both permeabilized fibers and isolated mitochondria.¹⁴ Future studies must account for muscle fiber type differences, which likely have implications for the bioenergetic capacity of our isolated mitochondrial prep. Various mitochondrial isolation techniques can result in altered bioenergetic function, such as significantly increased RCR compared with permeabilized myofibers accompanied by significantly reduced time to mitochondrial permeability transition pore opening and greater reactive oxygen species production.¹⁵ These factors are limitations to our study using only isolated mitochondria. Future studies should include respirometric profiling of both permeabilized fibers and isolated organelles in the same individuals. A further limitation of our study was that respiration of isolated mitochondria was only driven by complex 2 while blocking complex 1 with rotenone. Therefore, functional changes that may be associated with complex 1 are not addressed. While our study provides evidence that aspects of mitochondrial function driven by complex 2 activity are associated with physical ability in older adults, future studies will allow us to more comprehensively examine ETC function. For example, the use of alternate fuel sources such as malate and glutamate as well as N,N,N′,N′-tetramethyl-p-phenylenediamine (TMPD) and ascorbate may allow us to gain a more robust understanding of how the activity of specific ETC complexes may contribute physical function.

In contrast to our findings, a recent study did not find a relationship between PBMC respiration and vitality in middle-aged men, as measured by the Medical Outcomes Study Short Form vitality scale.¹⁶ The lack of association may stem from differences in the cohorts examined (middle-aged men vs older men and women) and the use of a self-assessment questionnaire versus a direct measure of physical ability. Future studies will examine various measures of physical function (eg, strength, power, endurance, balance) that impact physical ability in order to determine which are specifically influenced by mitochondrial bioenergetics.

In this study, two different cohorts were used for assessments of PBMC and muscle mitochondrial function. The conceptual premise of this work is that muscle bioenergetics is most relevant for understanding mobility function. Our finding that mitochondrial function in both blood and muscle were associated with gait speed suggests that either there may be a strong association of mitochondrial function between these tissues or that muscle and PBMC mitochondrial function tap into different processes that contribute to mobility. At the genetic level, there is increasing evidence that certain aspects of mitochondrial bioenergetics are systemic and shared across tissues. In two separate reports, a low-calorie diet decreased expression of genes involved in oxidative phosphorylation in PBMCs¹⁷ and in skeletal muscle biopsies.¹⁸ Similarly, decreased expression of oxidative phosphorylation genes is associated with type-2-diabetes in skeletal muscle¹⁹ and PBMCs.²⁰ A recent study assessed gene expression in response to a dietary intervention in PBMCs and skeletal muscle biopsies from the same individual.²¹ After an 8-week polyunsaturated fatty acid-supplemented diet, gene expression changes in PBMCs and muscle were strongly correlated suggesting that both tissues responded similarly to dietary intervention. Future studies should be undertaken to assess mitochondrial function in both blood and muscle in the same individual to determine if the contributions of both tissues to mobility are independent of one another. While direct comparisons of mitochondrial function in isolated muscle mitochondria and intact blood cells are not possible, respirometric analysis of muscle fibers may permit such a study.

The results presented in this brief report raise intriguing questions about the role of mitochondrial bioenergetics in the physical function of older adults. For example, the 400 m fast paced walk test incorporates various aspects of physical ability. In particular, fatigue plays an important role in this assessment. Future studies can be designed to examine the specific role that mitochondrial bioenergetics plays in fatigue experienced by older adults. Our data reveal that the RCR of skeletal muscle mitochondria is associated with gait speed independent of differences in BMI, age, and gender. Interestingly, our data also indicate that state 4o respiration tends to have a negative correlation with gait speed that reaches statistical significance when individually adjusting for gender. Lower state 4o respiration can underlie the increased RCR of mitochondrial from individuals with higher gait speeds, suggesting an important role for inner membrane leak in our reported findings. Further studies on the specific role of mitochondrial inner membrane leak on skeletal muscle bioenergetics and function are warranted. The role of gender should also be explored in greater detail since our data suggest that this can independently affect our readouts, particularly those that are reported as absolute values (ie, state 3 and state 4o). It is conceivable that ETC activity is intrinsically different between older men and women, or that mitochondrial isolation is affected by gender related differences in the physical properties of muscle tissue. It is likely that these gender effects are not evident in RCR because this measure of mitochondrial function is reported as a ratio, rather than an absolute value.

Potential confounders such as inflammatory status should also be analyzed in future studies as these have significant effects on both muscle and blood. It has been previously reported that inflammation, measured as high levels of interleukin-6, C-reactive protein, and interleukin-1RA, is significantly associated with poor physical performance and muscle strength in older persons.²² Two independent mechanisms can further contribute to the bioenergetic properties of PBMCs. Differences in mitochondrial ETC activity are the primary mediators of cellular respiration. However, PBMCs are a heterogeneous population of cells and that property can contribute to their overall bioenergetic profiles. Changes in monocyte and lymphocyte distribution, that may be related to inflammatory status, can affect OCR measurements performed on PBMCs. Future studies will tease apart the contributing effects of mitochondrial bioenergetics and cellular heterogeneity on the OCR of PBMCs.

Currently, there is no well-accepted blood biomarker that accurately predicts physical ability. Our findings indicate that respirometric analysis of PBMCs may provide such a measure. Compared with tissue biopsies, blood cells are easily accessible and suitable for multiple samplings over relatively short periods. This makes them appropriate for large-scale clinical trials and as a potential diagnostic tool. Further validation of blood-based bioenergetic profiling is required to determine its ability to estimate systemic bioenergetic capacity. In addition, future studies are needed in order to explore whether blood-based bioenergetic profiling has strong associations with morbidity and mortality and may provide a better prognostic indicator compared with other practical measures such as the short physical performance battery and gait speed.

Supplementary Material

Supplementary material can be found at: http://biomedgerontology.oxfordjournals.org/.

Funding

This work was supported by National Institutes of Health grants 5R01-AG020583 and 3R01-AG020583-09S1, and the Wake Forest Claude D. Pepper Older Americans Independence Center (P30-AG21332).