Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2018 Mar 25;596(8):1467–1483. doi: 10.1113/JP275009

Oxidative capacity varies along the length of healthy human tibialis anterior

Andreas Boss 1, Linda Heskamp 1,✉, Vincent Breukels 1, Lauren J Bains 2, Mark J van Uden 1, Arend Heerschap 1
PMCID: PMC5899983  PMID: 29455454

Abstract

Key points

  • During exercise skeletal muscles use the energy buffer phosphocreatine.

  • The post‐exercise recovery of phosphocreatine is a measure of the oxidative capacity of muscles and is traditionally assessed by 31P magnetic resonance spectroscopy of a large tissue region, assuming homogeneous energy metabolism.

  • To test this assumption, we collected spatially resolved spectra along the length of human tibialis anterior using a home‐built array of 31P detection coils, and observed a striking gradient in the recovery rate of phosphocreatine, decreasing along the proximo‐distal axis of the muscle.

  • A similar gradient along this muscle was observed in signal changes recorded by 1H muscle functional MRI.

  • These findings identify intra‐muscular variation in the physiology of muscles in action and highlight the importance of localized sampling for any methodology investigating oxidative metabolism of this, and potentially other muscles.

Abstract

The rate of phosphocreatine (PCr) recovery (k PCr) after exercise, characterizing muscle oxidative capacity, is traditionally assessed with unlocalized 31P magnetic resonance spectroscopy (MRS) using a single surface coil. However, because of intramuscular variation in fibre type and oxygen supply, k PCr may be non‐uniform within muscles. We tested this along the length of the tibialis anterior (TA) muscle in 10 male volunteers. For this purpose, we employed a 3T MR system with a 31P/1H volume transmit coil combined with a home‐built 31P phased‐array receive probe, consisting of five coil elements covering the TA muscle length. Mono‐exponential k PCr was determined for all coil elements after 40 s of submaximal isometric dorsiflexion (SUBMAX) and incremental exercise to exhaustion (EXH). In addition, muscle functional MRI (1H mfMRI) was performed using the volume coil after another 40 s of SUBMAX. A strong gradient in k PCr was observed along the TA (P < 0.001), being two times higher proximally vs. distally during SUBMAX and EXH. Statistical analysis showed that this gradient cannot be explained by pH variations. A similar gradient was seen in the slope of the initial post‐exercise 1H mfMRI signal change, which was higher proximally than distally in both the TA and the extensor digitorum longus (P < 0.001) and strongly correlated with k PCr. The pronounced differences along the TA in functional oxidative capacity identify regional variation in the physiological demand of this muscle during everyday activities and have implications for the bio‐energetic assessment of interventions to modify its performance and of neuromuscular disorders involving the TA.

Keywords: skeletal muscle, 31P magnetic resonance spectroscopy, oxidative metabolism, phosphocreatine recovery, magnetic resonance imaging

Key points

  • During exercise skeletal muscles use the energy buffer phosphocreatine.

  • The post‐exercise recovery of phosphocreatine is a measure of the oxidative capacity of muscles and is traditionally assessed by 31P magnetic resonance spectroscopy of a large tissue region, assuming homogeneous energy metabolism.

  • To test this assumption, we collected spatially resolved spectra along the length of human tibialis anterior using a home‐built array of 31P detection coils, and observed a striking gradient in the recovery rate of phosphocreatine, decreasing along the proximo‐distal axis of the muscle.

  • A similar gradient along this muscle was observed in signal changes recorded by 1H muscle functional MRI.

  • These findings identify intra‐muscular variation in the physiology of muscles in action and highlight the importance of localized sampling for any methodology investigating oxidative metabolism of this, and potentially other muscles.

Introduction

Skeletal muscle disposes of efficient cellular buffer systems that enable sudden and large changes in energy expenditure during transitions from rest to exercise and recovery. Creatine kinase (CK) and phosphocreatine (PCr) provide such a buffering function: during conditions when the demand of ATP cannot be met by oxidative metabolism, typically at the onset of exercise, ATP is produced through hydrolysis of PCr by CK. More than thirty years ago it was demonstrated using 31P magnetic resonance spectroscopy (MRS) that the ATP required for the recovery of PCr is synthesized aerobically (Taylor et al. 1983). Thus, the mono‐exponential rate constant of PCr recovery (k PCr) after in‐bore exercise is a non‐invasive correlate of the muscle's oxidative capacity (Meyer, 1988; McCully et al. 1993; Kemp et al. 2015). It was shown to be higher in trained vs. untrained (Takahashi et al. 1995; Larsen et al. 2009), young vs. older (Fleischman et al. 2010), and in healthy subjects vs. patients with type II diabetes (Phielix et al. 2008; Bajpeyi et al. 2011). Traditionally, 31P MR spectra of muscles are recorded with a single surface coil and a simple pulse‐acquire method. The 31P signals originate from a region adjacent to the coil. The extent of this region is determined by the size and sensitivity profile of the coil, but otherwise the spectra are unlocalized. However, skeletal muscle is not a homogeneous entity. Most importantly, it is composed of metabolically different fibre types: oxidative type I, and glycolytic type II fibres, the spatial distribution of which is not uniform. In many human muscles, type I fibres are predominantly found in the deep parts, while type II fibres are more prominent in superficial areas (Lexell et al. 1983; Dahmane et al. 2005). In rodent lower hindlimb muscles, including the tibialis anterior, fibre distribution also varies along the length of the muscle, with the proportion of oxidative fibres being higher in the proximal part, and glycolytic fibres more prominent in the distal part of the muscle (Torrella et al. 2000; Wang & Kernell, 2001). As experiments using muscle biopsies point towards faster recovery of PCr in the oxidative type I compared to the glycolytic type II fibres (Söderlund & Hultman, 1991), k PCr might reflect differential muscle fibre distribution. In addition, intramuscular variations in k PCr may depend on spatial differences in oxygen supply for maximal oxidative phosphorylation. The presence of such variations along the length of muscles will have implications for the bio‐energetic assessment, in individual muscles, of interventions (e.g. exercise programmes) and of disease progression in neuromuscular disorders.

The general aim of this study was therefore to investigate whether variations in functional oxidative capacity occur along the length of healthy muscles. More specifically we selected the tibialis anterior (TA) for this study as this muscle is easy to approach and performs in a well‐defined and assessable way in foot dorsiflexion. To determine if proximal–distal differences in k PCr exist we employed a dedicated phased‐array probe with five coil elements to acquire localized 31P MR spectra with high signal‐to‐noise and time resolution along the length of the TA, during and after submaximal and exhaustive foot dorsiflexion.

To confirm that the dorsiflexion exercise properly activates the TA and to examine the possible involvement of other muscles to the exercise, specifically that of the extensor digitorum longus (EDL), we applied muscle functional 1H MRI (1H mfMRI) with a 1H volume coil, as an independent high resolution MR approach. Moreover, because the dynamics of the exercise‐induced signal alterations in 1H mfMRI are strongly correlated with k PCr (Schmid et al. 2014), it is relevant to examine intramuscular variations of these signal alterations as well. This is of additional interest as the signal of apparent transversal relaxation (T2*) weighted gradient‐echo (GE) images, used in 1H mfMRI, is sensitive to changes in deoxyhaemoglobin concentration and blood flow, although other factors may also be involved (Fleckenstein et al. 1988; Jacobi et al. 2012).

Methods

Ethical approval

This study was conducted according to the principles of the Declaration of Helsinki (version October 2013) and in accordance with the Medical Research Involving Human Subjects Act (WMO), except for registration in a database. The study was approved by the local medical ethics committee (registration number: NL47394.091.14) and all subjects gave written informed consent.

Subjects

Ten healthy young (18–35 years) normal weight (BMI 18–25 kg m−2) male volunteers were recruited for this study. Exclusion criteria were any form of metallic implant and claustrophobia. Subjects were instructed to abstain from exercise and alcohol for 24 h prior to the two experimental sessions. During the week prior to the second trial, the participants were instructed to fill in a validated questionnaire about their level of daily physical activity (Craig et al. 2003).

Ergometer

A home‐built MR compatible ergometer set‐up for dorsiflexion of the foot was used. The volunteer's right foot was tightly strapped to a shoe attached to a pedal. The pedal, in turn, was connected via a rope to a digital force gauge (Sauter FL 500, Balingen, Germany), located outside the magnet room. The force applied during the various exercise bouts was then projected onto the front panel of the magnet housing as online feedback to the volunteers (Fig. 1 A and B).

Figure 1. Schematic representation of the experimental set‐up.

Figure 1

Open in a new tab

A, the subject's right foot was placed in a shoe attached to a pedal which was connected with a rope to a force gauge located outside the scanner room. The leg was positioned in a 1H/31P birdcage coil with the five‐element 31P surface coil for receive strapped on the tibialis anterior. The force applied during the exercise was then projected onto the scanner as feedback to the volunteers. B, example of the projection of the beamer onto the front panel of the magnet housing. The subject was asked to keep his force level (green line) within the two red lines (+5% and −5% of the maximum voluntary contraction). C, 31P phased array surface coil strapped on the lower leg of a healthy volunteer with the five coil elements depicted.

Design and hardware

Each volunteer underwent two experimental sessions inside the clinical 3T MR system (Siemens Magnetom Trio, Erlangen, Germany). The first session served to determine the maximum voluntary contraction (MVC, maximum of three attempts) for foot dorsiflexion and was determined in each volunteer inside the MR scanner using the ergometer. Thereafter, the volunteers got acquainted with the set‐up and in‐bore exercise regime as they performed the same rest–exercise–recovery transitions as during the actual experiments of the second session which are described below.

To assess metabolic differences along the length of the TA, a home‐built 31P phased‐array probe (van Uden et al. 2012) was used for signal reception, consisting of five individual coil elements (size: 4 × 4.5 cm each, total size: 4 × 20 cm, overlap of elements for decoupling). The coil array was positioned on the TA of the right leg, with element 1 (E1) at the distal end of the muscle placed at a ±12 cm from the lateral malleolus, and fixed with Velcro straps (Fig. 1 C). Fish oil capsules in the centre of the two outer most elements allowed for verification of coil position in the MR images. The receive coil was combined with a commercially available 1H/31P birdcage coil (Rapid, Rimpar, Germany), detunable at the 31P‐frequency and 20 cm in length, which was carefully positioned in the head–foot direction to fully cover the 31P receive coils. This set‐up allowed for a homogeneous phosphorous excitation by the volume coil and recording of free induction decays (FIDs) of high signal‐to‐noise ratio (SNR) from the TA by the individual coil elements of the surface coil. The birdcage coil's proton part was further used for anatomical imaging, 1H mfMRI and 31P signal enhancement by employing the 1H–31P nuclear Overhauser effect (NOE).

MR acquisition during the second experimental session

Experiments in resting muscle

After a series of localizers, anatomical imaging was performed using a T1 weighted turbo spin echo (TSE) sequence with five transversal slices centred to the middle of each 31P coil element. If the fish oil capsules attached to the outermost coil elements did not appear directly above the TA muscle in the TSE images, the 31P receive coil array was repositioned. Thereafter, 2D 31P imaging was performed using a GE sequence (centre‐frequency on PCr, TR = 1500 ms, TE = 10 ms, NA = 6, FOV = 199 × 199 mm, matrix size = 16 × 16) with localization in the third dimension perpendicular to the coil by the position of the individual coil elements. For 31P MRS, the main magnetic field homogeneity was adjusted on a volume covering all of the muscles within the sensitive area of the transmit coil. Two fully relaxed 31P MR spectra (90° flip angle, 500 μs hard pulse, TR = 15 s, 6 averages) were performed either with or without NOE (1H decoupling with WALTZ 4, 1H frequency on the water frequency (Luyten et al. 1989)).

Experiments in contracting muscle (Fig. 2)

Figure 2. Succession of experiments applying isometric muscle contractions performed inside the clinical MR system.

Figure 2

Open in a new tab

First, 1H muscle functional magnetic resonance imaging (1H mfMRI) data was obtained, for which the volunteers performed 40 s of isometric dorsiflexions of the foot while 1H gradient echo echo‐planar imaging was applied for 16 min 40 s. Thereafter, 31P MRS was performed before, during and after exercise of the same intensity and duration (SUBMAX). Finally, an isometric incremental exercise was performed until exhaustion recorded with 31P MRS (EXH).

The post‐exercise signal intensity of 1H mfMRI takes at least 15 min (Schmid et al. 2014) to reach baseline levels which is considerably longer than the signals of 31P metabolites. Therefore, 1H mfMRI experiments were always performed before the dynamic 31P experiments. For the same reason, MVC was not repeated for the second experimental session to avoid any interference with 1H mfMRI. Furthermore, the submaximal 31P experiment always preceded the exhaustive one. For 1H mfMRI, T2* weighted GE‐EPI (TE = 29 ms, TR = 1 s) was applied during 1 min of rest, 40 s of isometric contractions performed at 60% of MVC, and during 15 min of recovery using five transversal slices (3 mm) corresponding to the centre of each 31P coil element. In separate experiments on three volunteers, 1H mfMRI was also performed without the 31P phased‐array probe tightly strapped on the lower leg to examine if the probe can cause blood flow obstruction along the length of the TA. To examine post‐exercise k PCr, 31P MR spectra (TR = 2 s, 2 averages per spectrum, 48° Ernst angle excitation, 1H–31P NOE enhanced) were acquired during submaximal exercise (SUBMAX) of 30 s rest, 40 s isometric contraction at 60% MVC and 5 min recovery. Finally, the volunteers performed exhaustive exercise (EXH) starting at 10% MVC, increasing by 10% every 30 s. The volunteers either stopped voluntarily or the test was interrupted when the volunteers could not maintain the requested force. 31P MRS (same parameters as for the previous experiment) was acquired during and after EXH for a total duration of 10 min 32 s.

Data post‐processing

Overlay of 31P imaging and anatomical imaging

We delineated the TA and EDL to determine the relative contribution of the two dorsiflexors to the total 31P signal by overlaying the 31P images and anatomical images in a subset of 8 volunteers using MIPAV (Medical Imaging Processing, Analysis and Visualization (MIPAV), http://mipav.cit.nih.gov). For the post‐exercise 1H mfMRI data, two ROIs surrounding the TA and EDL, and a third ROI surrounding the remaining muscles of the calf and peronei (C) were drawn using Fiji (Schindelin et al. 2012). In these three ROIs the signal intensity during recovery was normalized to baseline and a moving average filter with span of 11 data points was applied. The amount of signal change relative to the minimal value at the start of the recovery, as well as the slope at which the signal increased in the initial phase of the recovery were determined. The 31P spectra were fitted, after phase correction and frequency alignment, using the AMARES algorithm in JMRUI (Stefan et al. 2009) applying appropriate prior knowledge including Lorentzian line shape and multiplets for ATP. Since no Pi splitting was observed during exercise Pi was fitted as a singlet. PCr recovery was fitted to the following mono‐exponential function using Matlab (The Mathworks, Inc. Natick, MA, USA):

PCr (t)= PC r0+Δ PCr 1−e−kPCrt,

where k PCr is the recovery rate constant, PCr0 is PCr at the end of exercise, and ΔPCr is the recovery value of PCr (PCrrecovery) minus PCr0. Relative PCr depletion was obtained from the fitted values and defined as ΔPCr/(PCrrecovery). pH was determined from the chemical shift difference between Pi and PCr (Moon & Richards, 1973).

Statistical analysis

Linear mixed model procedures for repeated measures were used (SPSS version 22.0, IBM, Armonk, NY, USA) to investigate the linear dependence of the various 31P MRS and 1H mfMRI parameters on coil element or slice number, essentially as described by West (2009). For statistical modelling of the linear dependence of k PCr on coil element number, end‐exercise pH (pHendex) was included as an additional covariate. For the comparison of SUBMAX and EXH, averages of the 5 coil elements for k PCr, PCr depletion, and pH were computed in each volunteer and Student's paired t tests were performed. Standard linear regression was used to estimate associations between pHendex and k PCr, as well as between k PCr and parameters from 1H mfMRI. For this linear regression pooled data from all the volunteers per coil element or imaging slice were used. The level of significance α was set to 0.05. All results are means ± SEM, unless stated differently.

Results

Subject characteristics

The subjects’ anthropometric data along with their level of self‐reported physical activity are given in Table 1. All but one volunteers were engaged in sitting/sedentary professional activities. A relatively broad range of total physical activity time was reported, in total between 4 and 45 h per week at various intensities.

Table 1.

Anthropometric data and self‐reported physical activity

Subject characteristics
Anthropometric data
Age (years) 29 ± 1
Weight (kg) 79 ± 3
Height (m) 1.86 ± 0.03
BMI (kg m−2) 23.1 ± 0.8
Physical activity
Sitting ((hh:min) day−1) 07:40 ± 00:50 [01:40; 10:10]
Walking ((hh:min) week−1) 02:00 ± 00:30 [00:00; 05:10]
Cycling ((hh:min) week−1) 04:20 ± 01:00 [00:40; 09:00]
Heavy intensity activities ((hh:min) week−1) 03:20 ± 02:40 [00:00; 24:00]
Medium intensity activities ((hh:min) week−1) 02:30 ± 02:00 [00:00; 19:00]
Open in a new tab

Walking included work, displacement and leisure time activities, while cycling was commuting and displacement only. Heavy and medium intensity activities include for all but one volunteers exclusively leisure activities. Results are means ± SEM; minimum and maximum value in brackets.

MVC and isometric exercise

Average MVC determined during the practice session was 200 ± 7.5 N. Average force during 1H mfMRI and 31P MRS SUBMAX was 59 ± 0.5% and 59 ± 0.4% of MVC (P = 0.14), respectively, indicating a good compliance with the requested force output (i.e. 60% MVC). Average time to exhaustion during EXH was 165 ± 7 s.

General characteristics of 31P MR spectra

The proportion of 31P signal stemming from TA was determined using an overlay of 31P and 1H anatomical images and was found to be 64 ± 3%, 75 ± 3%, 84 ± 1%, 82 ± 1% and 79 ± 1% for E1 to E5, respectively. Thus, a dominant proportion of the signal originates from TA (Fig. 3 A). The cross‐sectional area of the TA on the 1H anatomical images was 371 ± 32 mm2, 586 ± 42 mm2, 749 ± 36 mm2, 663 ± 23 mm2 and 437 ± 26 mm2 for image slices 1 to 5, respectively. For the EDL, the cross‐sectional area was 294 ± 16 mm2, ± 305 ± 12 mm2, 279 ± 5 mm2, 283 ± 15 mm2 and 200 ± 13 mm2 for slices 1 to 5, respectively. 31P MR spectra obtained during exercise and recovery showed resonance for ATP, PCr and Pi with good SNR from all coil elements (see examples in Fig. 3 B and C). The variation in the signal integral of PCr during the experiments was stable enough for proper analysis (Fig. 3 D). Together, this demonstrates good data quality given the small volume of muscle tissue that the signals are arising from. The signal enhancement of PCr due to 1H–31P NOE was modest, but significant and very similar between the different coils elements (15% to 16% signal enhancement on average in a subset of 4 volunteers at rest).

Figure 3. Results of the 31P experiment in one volunteer from coil elements E1 (distal, left), E3 (middle) and E5 (proximal, right).

Figure 3

Open in a new tab

A, overlay of 31P maps on 1H images, indicating that the majority of the 31P signal originated from the tibialis anterior. B, 31P MR spectra at the end of submaximal exercise. C, 31P MR spectra at the end of recovery. D, the corresponding signal dynamics for phosphocreatine.

31P MRS after exercise

The post‐exercise recovery rate of PCr varied between the different coil elements, as indicated in Fig. 4 , which shows the average PCr intensity of the ten volunteers during exercise and recovery after SUBMAX exercise. Average k PCr varied significantly along the length of TA, being lower in distal compared to proximal regions after both SUBMAX and EXH (for both: P < 0.001, +0.27 min−1 with increasing coil number, Fig. 5 A and B). The statistical model used accounted for differences in pH at the end of exercise (pHendex) between coils and subjects, indicating that the pronounced difference in k PCr between coil elements cannot primarily be explained by alterations in pH.

Figure 4. Average normalized PCr signal intensity during recovery of submaximal (A and C) and exhaustive exercise (B and D).

Figure 4

Open in a new tab

PCr is normalized to pre‐exercise levels. A and B show results for all five coil elements for submaximal and exhaustive exercise, respectively (E1‐proximal to E5‐distal; without error bars). C and D show results for the two outermost coil elements E1 and E5 including error bars (SEM), for submaximal and exhaustive exercise, respectively. This demonstrated similar depletion of PCr, but a pronounced difference in the recovery rate. N = 10.

Figure 5. Post‐exercise PCr‐recovery rate constant (k PCr) for elements E1‐distal to E5‐proximal.

Figure 5

Open in a new tab

A, after submaximal (SUBMAX). B, after exhaustive exercise (EXH). A pronounced gradient in k PCr, increasing from distal to proximal regions of the tibialis anterior was found during both SUBMAX and EXH (P < 0.001 for both). C, correlation of k PCr of SUBMAX with k PCr of EXH for means of each coil element. D, plot of average pHendex vs. k PCr over all subjects of each coil element, indicating that pH cannot be the principal reason for the pronounced variation in k PCr between elements. Data are presented as means ± standard deviation. N = 10.

On average, k PCr was significantly lower after EXH than SUBMAX (P = 0.024, Fig. 5 C). A comparison of k PCr with pHendex for EXH and SUBMAX and all coil elements showed that a drastically lower pHendex was associated with an overall slightly lower k PCr in EXH compared to SUBMAX, but the difference in k PCr between coil elements was unchanged (Fig. 5 D). This further indicates that pH is not the main reason for the gradient in k PCr along TA. Pooling all subjects and coils, k PCr significantly correlated with pHendex in SUBMAX (r 2 = 0.17, P = 0.003) and in EXH (r 2 = 0.14, P = 0.008). k PCr was also fitted with a bi‐exponential model (Harris et al. 1976) and in 80% and 66% of the cases for SUBMAX and EXH, respectively, the fit converged to the same mono‐exponential solution. Importantly, there was no difference between coil elements, as a mono‐exponential solution was found in 14 (coil elements 1 and 3) or 15 (coil elements 2, 4, 5) out of 20 PCr recovery experiments per coil element.

31P MRS during exercise

Along with differences in oxidative capacity within TA measured by post‐exercise k PCr, parts of the muscle with a lower oxidative capacity might also show an earlier pH drop after the initial pH increase and a more pronounced utilization of PCr during exercise. While the relative maximum depletion of PCr was not correlated with any of the coil elements (P = 0.22) during SUBMAX, it varied slightly (estimated fixed effect: +0.02 with increasing coil number), but significantly (P < 0.001) with the coil elements during EXH (Fig. 6). pHendex and the lowest pH value during the experiment (pHmin; reached shortly after the end of the exercise due to protons released by the reversed creatine kinase reaction (Kemp et al. 2015)) showed no gradients along the length of the muscle, neither during SUBMAX, nor during EXH (Fig. 6). Time evolution of the PCr signal and the Pi‐to‐PCr ratio (not shown) during EXH was similar between coil elements, while pH started to drop earlier during exercise in E3 compared to the outer elements E5 and E1 (Fig. 7). In EXH, the depletion of PCr was more pronounced (0.53 ± 0.02 vs. 0.40 ± 0.02; P = 0.001), as was the decrease of end‐exercise pH (pHendex: 6.75 ± 0.03 vs. 7.07 ± 0.01; P < 0.001) compared to SUBMAX. The minimal pH was lower in EXH than SUBMAX (pHmin: 6.59 ± 0.03 vs. 6.93 ± 0.03; P < 0.001).

Figure 6. Relative PCr depletion (A and B) and pH (C and D) at the end of exercise (pHendex) and minimum (pHmin).

Figure 6

Open in a new tab

A, and C, submaximal exercise (SUBMAX). B and D, incremental exercise to exhaustion (EXH). During EXH, PCr depletion varied slightly, but significantly (P < 0.001) with coil number, i.e. displayed a gradient along the length of tibialis anterior. Data are presented as means ± standard deviation.

Figure 7. Average normalized PCr signal intensity (A and B) and pH (C and D) during the incremental exercise to exhaustion.

Figure 7

Open in a new tab

A and C, show results for the five coil elements (E1‐distal, to E5‐proximal; without error bars). B and D only show results for three coil elements (E1, E3, and E5) including error bars (SEM). pH values deviating more than ±2 times the standard deviation from the mean were removed. The average time to exhaustion in the ten volunteers was 165 ± 7 s. N = 10.

1H mfMRI response to exercise

The pronounced increase in the relative 1H mfMRI signal intensity in the early recovery phase after isometric exercise (+15.3 % in TA or +17.9 % in EDL on average in the most proximal image slice 5) indicates that both TA and EDL were recruited for the isometric dorsiflexion exercise, while the rest of the lower leg muscles were not, as reflected by the almost complete absence of post‐exercise response for the remaining muscles of the lower leg (Fig. 8). The slope of this signal increase varied with the slice number (TA and EDL: P < 0.001), increasing from distal to proximal (Fig. 9 A and B). The slope of the 1H mfMRI signal increase appeared to be lower for each slice when the 31P phased‐array probe is positioned on the TA (Fig. 9 C and D). However, a similar proximal–distal gradient of this slope along the TA was still present without this probe attached to the lower leg. Figure 10 A displays the association of the slope of TA and EDL, which were significantly correlated for the average over the volunteers per slice (slope: r 2 = 0.99; P < 0.001).

Figure 8. Average 1H mfMRI signal in the recovery phase after 40 s isometric exercise (60% MVC).

Figure 8

Open in a new tab

Data represent five slices (S1, distal to S5, proximal) corresponding to the centre of each of the five 31P‐coil elements (error bars omitted for clarity). For visual interpretation the lines were normalized to a minimum value at initial recovery. Both, tibialis anterior (A) and extensor digitorum longus (B) were recruited while the rest of the lower leg muscles (C) were not. The first 600 of 900 s of recovery are shown. Note that the high signal intensity in the first few seconds is a movement artefact. N = 10.

Figure 9. Post‐exercise dynamics of the 1H mfMRI signal.

Figure 9

Open in a new tab

A and B, slope during initial recovery from minimum to peak signal intensity in ten volunteers after 40 s isometric exercise (60% MVC) for the five imaging slices (S1‐distal to S5‐proximal) in tibialis anterior (A) and extensor digitorum longus (B). C, slope of the 1H mfMRI response in a subset of three volunteers without the placement of the 31P phased array probe on the tibialis anterior. D, slope of the 1H mfMRI response with placement of the 31P phased array probe on the tibialis anterior in the same three volunteers. Data are presented as means ± standard deviation.

Figure 10. Correlation between post‐exercise 1H mfMRI dynamics and recovery of phosphocreatine (k PCr).

Figure 10

Open in a new tab

A, significant correlation between the tibialis anterior (TA) and extensor digitorum longus (EDL) for the slope of the 1H mfMRI signal intensity. B, significant correlation of k PCr with the slope of the 1H mfMRI signal during the initial recovery after 40 s isometric exercise (60% MVC).

Correlation of k PCr (31P) with 1H mfMRI

The PCr recovery rate in TA was positively associated with the slope of the 1H mfMRI signal (Fig. 10 B). Highly significant correlations were observed for the average over the volunteers per slice or per coil element (r 2 = 1.0, P < 0.001).

Discussion

In this study we observed remarkable differences in oxidative capacity along the length of the tibialis anterior in healthy male volunteers by using phased‐array localised 31P MR spectroscopy. While the extent of PCr depletion and acidification during exercise was similar, the rate of PCr recovery after exercise was approximately twice as high in proximal compared to distal muscle parts, indicating a heterogeneous oxidative capacity within this muscle. It is important to note that the magnitude of this difference in k PCr within the same muscle is similar to the difference in k PCr between untrained and endurance trained (Larsen et al. 2009), or sprint and endurance trained TA muscles (Crowther et al. 2002) as assessed by 31P MRS using traditional surface coils. Thus proper placement of the coils may be critical in these assessments. We observed a similar proximal to distal gradient in the slope of the signal increase recorded by 1H mfMRI after exercise. These post‐exercise dynamics of the 1H mfMRI signal strongly correlated with k PCr derived from 31P MRS. It is noteworthy that these correlations stem from two consecutive exercise experiments, indicating a similar metabolic demand of both exercise bouts, which is also suggested by the very similar overall force during both experiments.

31P MRS studies repeatedly showed that end‐exercise low levels of pH and/or pronounced PCr depletion are associated with reduced PCr recovery rates (Lodi et al. 1997; van den Broek et al. 2007; Layec et al. 2013). In the present study, we indeed observed this correlation between pHendex and k PCr in an evaluation of the results from all subjects and coil elements. Furthermore, k PCr was suppressed after exhaustive exercise with lower pHendex and pHmin. This demonstrates that although the PCr signals arose from small muscle volumes, the signal‐to‐noise ratio was sufficient to measure the expected drop in k PCr and pH associated with intense exercise. The variation in mean k PCr along the coil elements, however, cannot be explained by pH, as the linear mixed model included pHendex as a covariate. This conclusion is supported by the observation that EXH exercise is associated with substantially lower pHendex compared to SUBMAX, while the difference in mean k PCr between coil elements remains the same.

Recovery of PCr requires an adequate supply of oxygen to the muscle cell (Taylor et al. 1983; Quistorff et al. 1993). Blood flow to the TA could be reduced due to compression of the anterior tibial artery by the surface coil, and thus potentially decrease k PCr. The additional 1H mfMRI experiments performed in a subset of three volunteers showed indeed that strapping the surface coil to the lower leg can result in a slower response compared to the same measurement without the 31P coil array. However, the gradient in the post‐exercise kinetics of the 1H mfMRI signal was not affected by the placement of the array.

The observed gradient in the rate of PCr recovery of the TA muscle thus has a biological origin which could be a proximo‐distal variation in fibre type composition and/or a variation in capillary density and local perfusion. The typical fibre length in TA is less than 7 cm, while this whole muscle is on average 26 cm long in male subjects (Ward et al. 2009). Even though TA muscle is only slightly pennated, the muscle fibres found distally are not the same as those proximally (Wang & Kernell, 2001). In rats and rabbits, the proportion of the area occupied by type I fibres gradually decreases from the proximal to the distal parts of TA (Wang & Kernell, 2001). Given the intrinsically higher capacity for PCr recovery in oxidative type I fibres (Söderlund & Hultman, 1991), a varying muscle fibre distribution along the TA muscle could explain the observed gradient in PCr recovery. There is evidence of non‐uniform muscle fibre distribution in cross‐sections of various human limb muscles, including the TA (Johnson et al. 1973; Elder et al. 1982; Dahmane et al. 2005) and the vastus lateralis (Elder et al. 1982), but we are not aware of studies on muscle fibre distribution along the length of human TA muscle.

Besides 31P MRS, muscle functional 1H MR imaging was performed before, during and after exercise of the same intensity and duration as for the submaximal exercise in the 31P experiments and with imaging slices placed at the middle of each 31P coil element. The slope of the signal increase in the TA after exercise increased from the distal to the proximal part of the muscle. Qualitatively, the signal response in 1H mfMRI after contraction is very similar to that of the brain: after a stimulus, an initial signal is observed increase, followed by a plateau phase, and subsequent decrease of the signal (Buxton, 2010). In the brain the response is in the order of 1–2%, lasting only a few seconds and is due to the so‐called blood oxygenation level dependent (BOLD) effect, caused by local susceptibility differences due to changing concentrations of blood deoxy‐ and oxyhaemoglobin. The post‐exercise effect in muscles on T2/T2* relaxation is higher, up to 30%, depending on exercise intensity, and lasts several minutes (Fisher et al. 1990). In skeletal muscle, the exact causes of these signal alterations after exercise are not entirely known; besides BOLD, changes in pH and osmotically driven fluid shifts may also affect T2/T2* relaxation (Fleckenstein et al. 1988; Jacobi et al. 2012).

If the observed gradient in the 1H mfMRI signal slope is mainly due to the BOLD effect, this could indicate decreasing capillary density and local perfusion in the proximal–distal direction and as a consequence a slower recovery of PCr in the distal parts. In rats, higher capillary density in proximal vs. distal regions of TA muscle was reported (Torrella et al. 2000), but we are unaware of studies investigating capillary density along human TA muscle. However, studies using other methods such as positron emission tomography (Mizuno et al. 2003) and near‐infrared spectroscopy (Koga et al. 2015) have demonstrated heterogeneity in perfusion and oxidative capacity of human skeletal muscle. Therefore, in future work it would be valuable to investigate perfusion along the length of the TA also, for example by MR techniques such as arterial spin labelling (Boss et al. 2006) or intravoxel incoherent motion imaging (Le Bihan et al. 1988; Filli et al. 2015).

Beyond the BOLD effect, it has been shown that tissue pH directly affects the muscle's intracellular transverse relaxation time (Louie et al. 2009). Furthermore, rapid osmotically driven alterations of intra‐ and extracellular water pool sizes due to changes of intracellular metabolite concentrations (Pi, Cr, PCr, lactate, acetylcarnitine etc.), may dynamically change the overall/apparent T2 and T2* of skeletal muscle (Meyer & Prior, 2000; Damon & Gore, 2005; Schmid et al. 2014). This requires intra‐ and extracellular (=interstitial and intravascular) water T2/T2* to be substantially different from each other. Intracellular and interstitial water T2 are indicated to be similar (around 30 ms), while T2 of vasculature is in the order of 140–180 ms (Belton et al. 1972; Hazlewood et al. 1974; Araujo et al. 2014). In the early phase of recovery, oxidatively produced ATP is utilized to re‐establish levels of PCr according to the net equation: Pi + Cr→PCr + H+. The protons produced by the CK reaction contribute to an initial lowering of intracellular pH, resulting in a (slightly) prolonged cellular T2 (Louie et al. 2009). Parts of the protons are, however, buffered, or together with lactate efficiently pumped out of the myocyte (Kemp et al. 2015). Therefore, intracellular metabolite concentration falls off quickly after the end of exercise, essentially as a function of the PCr recovery rate (Damon & Gore, 2005). Thus, the rapidly decreasing muscle metabolite concentration in the early phase of the recovery could lead to a fluid shift from intracellular to interstitial and via capillary exchange to vascular compartments, thereby resulting in a higher overall water T2 and T2*. Thus, increasing overall water T2/T2* as a consequence of changing PCr concentration could be an alternative explanation for the strong correlation of k PCr with the slope of the post‐exercise increase of the 1H mfMRI signal.

It is beyond the scope of this study to disentangle the complex nature of the 1H mfMRI response to exercise. However, our finding that k PCr is correlated with the 1H mfMRI response further supports the notion that metabolic events and oxidative capacity of the muscle are directly associated with the kinetics of the 1H mfMRI response (Schmid et al. 2014) or the extent (Vandenborne et al. 2000) and kinetics of muscle T2 after exercise (Reid et al. 2001). It was previously shown that the inverse of k PCr correlates with the time to peak of the post‐exercise 1H mfMRI signal of a single slice (Schmid et al. 2014). Here these findings are extended, demonstrating that this correlation holds true also for different regions of the same muscle.

In the present study, a pronounced post‐exercise 1H mfMRI signal increase was also observed for the EDL, while the remaining lower leg muscles do not show such an increase, which is in accordance with the biomechanics of dorsiflexion. Similar to TA, EDL displayed a proximal to distal gradient in the slope of the signal increase. Again, this functional gradient could reflect intramuscular variation in oxidative capacity or oxygenation. Most importantly, the fact that this gradient is also apparent in the EDL indicates that the gradient in k PCr in our 31P experiments of the TA muscle cannot be explained by a varying signal contamination by the EDL muscle along the different 31P coil elements. This is confirmed by the overlay of 31P and 1H anatomical images demonstrating that the vast majority of the 31P signal stems from the TA.

If oxidative capacity is higher in proximal parts of the TA muscle, we anticipated that the higher potential for aerobic energy provision should also be reflected in a slower utilization of PCr and a later drop of pH during exercise. An incremental test to exhaustion is expected to be the most sensitive for picking up differences between coil elements during exercise (Chilibeck et al. 1998). However, we observed little difference in PCr utilization, Pi to PCr ratio and pH in this phase, despite the pronounced gradient in the post‐exercise PCr recovery rate. We tested only for linear gradients along the muscle, and can therefore not exclude differences in certain parameters between one coil element and another. These differences must be, however, much smaller than those observed in the recovery phase. A limitation of this study is the use of continuous isometric exercise which is characterized by lower energy expenditure compared to concentric exercise (Ryschon et al. 1997) and higher blood flow occlusion compared to dynamic muscle contractions (Laaksonen et al. 2003). Thus, low oxidative energy provision in the entire TA during the muscle contractions could explain why the gradient in oxidative capacity is not or only slightly reflected in PCr utilization and pH. It may therefore well be that those parameters would vary to a greater extent if dynamic exercise was used. Alternatively, it cannot be ruled out that force production and therefore energy requirement is not uniform along the TA muscle. If the demand of ATP is higher in the proximal vs. distal regions of TA, the rate of PCr utilization may be similar despite a pronounced difference in oxidative capacity.

In our study we used five small separate coil elements for localization. Alternative technical approaches were recently presented for the localized assessment of k PCr (Greenman & Smithline, 2011; Parasoglou et al. 2012; Niess et al. 2016). For instance, using a phosphorous volume coil together with a fast 31P imaging method accelerated by compressed sensing at 7 T (Parasoglou et al. 2012). While this elegant method allowed PCr recovery rates to be determined in various muscles of a cross‐section of the lower leg, the simultaneous assessment of pH was not possible. Fast imaging of PCr and ATP in different muscles during exercise and recovery was demonstrated at 7 T using selective excitation and chemical shift displacement (Steinseifer et al. 2013). With single voxel localization (Meyerspeer et al. 2011) a more pronounced PCr depletion in the medial gastrocnemius during plantar flexions was observed compared to non‐localized experiments (Meyerspeer et al. 2012). Others reported different PCr recovery kinetics in different muscle groups in separate measurements (Larsen et al. 2009; Layec et al. 2013; Yoshida et al. 2013), as well as within a single experiment (Forbes et al. 2009; Valkovič et al. 2016). We are not aware of other studies investigating PCr recovery rates spatially resolved along the length of the same muscle.

In conclusion, pronounced differences in the functional oxidative capacity along the human tibialis anterior muscle were found by 31P MRS investigations with a specifically designed multi‐array coil, supported by 1H mfMRI examinations. These findings identify regional variation in the physiology of this muscle during everyday activities. An important practical implication of the proximal–distal gradient in k PCr is that reproducible positioning of a surface coil on the TA in conventional 31P MRS studies or voxel selection in localized 31P MRS are crucial, especially in longitudinal studies, such as to assess the effect of exercise training on k PCr. Moreover, for interventions it may be essential to assess multiple locations in muscles for a proper evaluation of their effects. The present and other studies (Schmid et al. 2014) also demonstrated that 1H mfMRI and k PCr assessed from 31P MRS can provide similar spatial information on in vivo oxidative capacity of skeletal muscles. Since 1H mfMRI can be done on any clinical MR system, and because of its much higher sensitivity, resulting in higher spatial and temporal resolution, this method may become a promising clinical tool to examine regionalization of metabolic capacity of skeletal muscle. The localized 31P MRS and 1H imaging approach reported here could also be used to investigate mechanisms of disease progression in neuromuscular diseases such as Duchenne muscular dystrophy and facioscapulohumeral muscular dystrophy, for both of which non‐uniform muscle fat infiltration along the proximo‐distal axis of lower limb muscles were recently observed (Janssen et al. 2014; Hooijmans et al. 2017).

Additional information

Competing interests

All authors have nothing to declare.

Author contributions

The experiments where performed at the Radiology and Nuclear Medicine department of the Radboud university medical centre, Nijmegen, The Netherlands. A.B. and A.H. were involved in the conception or design of the work. M.J.U. built the 31P‐coil and helped to develop the experimental set‐up. A.B. and L.H. were involved in the acquisition and analysis of the data and drafting the work. All authors were involved in revising the work critically for important intellectual content. All authors approved the final document and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

A.B. was supported by an Early Postdoc Mobility grant from the Swiss National Science Foundation and funding from the Centre for Systems Biology and Bioenergetics, Radboud University Nijmegen, The Netherlands. L.H. was supported by the European Community's Seventh Framework Programme (FP7/2007–2013) under grant agreement number 305697.

Acknowledgements

We would like to thank the volunteers for their dedication to the study. Miriam Lagemaat is thanked for providing the 31P‐MRS sequence, Tom Scheenen for providing the 31P MRI sequence and Anne Rijpma for her contribution to the statistical analysis.

Biographies

Andreas Boss trained in Human Movement Sciences at the ETH Zürich in Switzerland. He received his PhD at the University of Bern in the Magnetic Resonance Spectroscopy and Methods group led by Chris Boesch. Andreas’ research focus lies on the investigation of liver and muscle metabolism using MR spectroscopy.

graphic file with name TJP-596-1467-g001.gif

Linda Heskamp received her Master's degree at the University of Twente. She is currently a PhD candidate in the Biomedical Magnetic Resonance group at the Radboud University Medical Centre. Her research focuses on the use of Magnetic Resonance Imaging and Spectroscopy to further understand the muscle pathophysiology in muscular dystrophies, as well as understanding healthy muscle physiology and metabolism.

A. Boss and L. Heskamp contributed equally to this work.

Edited by: Michael Hogan & Bruno Grassi

References

  1. Araujo EC, Fromes Y & Carlier PG (2014). New Insights on human skeletal muscle tissue compartments revealed by in vivo T2 NMR relaxometry. Biophys J 106, 2267–2274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bajpeyi S, Pasarica M, Moro C, Conley K, Jubrias S, Sereda O, Burk DH, Zhang Z, Gupta A, Kjems L & Smith SR (2011). Skeletal muscle mitochondrial capacity and insulin resistance in type 2 diabetes. J Clin Endocrinol Metab 96, 1160–1168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Belton PS, Jackson RR & Packer KJ (1972). Pulsed NMR studies of water in striated muscle. I. Transverse nuclear spin relaxation times and freezing effects. Biochem Biophys Acta 286, 16–25. [DOI] [PubMed] [Google Scholar]
  4. Boss A, Martirosian P, Claussen CD & Schick F (2006). Quantitative ASL muscle perfusion imaging using a FAIR‐TrueFISP technique at 3.0 T. NMR Biomed 19, 125–132. [DOI] [PubMed] [Google Scholar]
  5. Buxton RB (2010). Interpreting oxygenation‐based neuroimaging signals: the importance and the challenge of understanding brain oxygen metabolism. Front Neuroenergetics 2, 8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chilibeck PD, McCreary CR, Marsh GD, Paterson DH, Noble EG, Taylor AW & Thompson RT (1998). Evaluation of muscle oxidative potential by 31P‐MRS during incremental exercise in old and young humans. Eur J Appl Physiol Occup Physiol 78, 460–465. [DOI] [PubMed] [Google Scholar]
  7. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, Pratt M, Ekelund U, Yngve A, Sallis JF & Oja P (2003). International physical activity questionnaire: 12‐Country reliability and validity. Med Sci Sports Exerc 35, 1381–1395. [DOI] [PubMed] [Google Scholar]
  8. Crowther GJ, Jubrias SA, Gronka RK & Conley KE (2002). A “functional biopsy” of muscle properties in sprinters and distance runners. Med Sci Sports Exerc 34, 1719–1724. [DOI] [PubMed] [Google Scholar]
  9. Dahmane R, Djordjevic S, Simunic B & Valencic V (2005). Spatial fiber type distribution in normal human muscle Histochemical and tensiomyographical evaluation. J Biomech 38, 2451–2459. [DOI] [PubMed] [Google Scholar]
  10. Damon BM & Gore JC (2005). Physiological basis of muscle functional MRI: predictions using a computer model. J Appl Physiol 98, 264–273. [DOI] [PubMed] [Google Scholar]
  11. Elder GC, Bradbury K & Roberts R (1982). Variability of fiber type distributions within human muscles. J Appl Physiol Respir Env Exerc Physiol 53, 1473–1480. [DOI] [PubMed] [Google Scholar]
  12. Filli L, Boss A, Wurnig MC, Kenkel D, Andreisek G & Guggenberger R (2015). Dynamic intravoxel incoherent motion imaging of skeletal muscle at rest and after exercise. NMR Biomed 28, 240–246. [DOI] [PubMed] [Google Scholar]
  13. Fisher MJ, Meyer RA, Adams GR, Foley JM & Potchen EJ (1990). Direct relationship between proton T2 and exercise intensity in skeletal muscle MR images. Invest Radiol 25, 480–485. [DOI] [PubMed] [Google Scholar]
  14. Fleckenstein J, Canby RC, Parkey RW & Peshock RM (1988). Acute effects of exercise on MR imaging of skeletal muscle in normal volunteers. AJR Am J Roentgenol 151, 231–237. [DOI] [PubMed] [Google Scholar]
  15. Fleischman A, Makimura H, Stanley TL, McCarthy MA, Kron M, Sun N, Chuzi S, Hrovat MI, Systrom DM & Grinspoon SK (2010). Skeletal muscle phosphocreatine recovery after submaximal exercise in children and young and middle‐aged adults. J Clin Endocrinol Metab 95, E69–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Forbes SC, Slade JM, Francis RM & Meyer RA (2009). Comparison of oxidative capacity among leg muscles in humans using gated 31P 2‐D chemical shift imaging. NMR Biomed 22, 1063–1071. [DOI] [PubMed] [Google Scholar]
  17. Greenman RL & Smithline HA (2011). The feasibility of measuring phosphocreatine recovery kinetics in muscle using a single‐shot 31P RARE MRI sequence. Acad Radiol 18, 917–923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Harris RC, Edwards RHT, Hultman E, Nordesjö LO, Nylind B & Sahlin K (1976). The time course of phosphorylcreatine resynthesis during recovery of the quadriceps muscle in man. Pflügers Arch 367, 137–142. [DOI] [PubMed] [Google Scholar]
  19. Hazlewood CF, Chang DC, Nichols BL & Woessner DE (1974). Nuclear magnetic resonance transverse relaxation times of water protons in skeletal muscle. Biophys J 14, 583–606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hooijmans MT, Niks EH, Burakiewicz J, Anastasopoulos C, van den Berg SI, van Zwet E, Webb AG, Verschuuren JJGM & Kan HE (2017). Non‐uniform muscle fat replacement along the proximodistal axis in Duchenne muscular dystrophy. Neuromuscul Disord 27, 458–464. [DOI] [PubMed] [Google Scholar]
  21. Jacobi B, Bongartz G, Partovi S, Schulte AC, Aschwanden M, Lumsden AB, Davies MG, Loebe M, Noon GP, Karimi S, Lyo JK, Staub D, Huegli RW & Bilecen D (2012). Skeletal muscle BOLD MRI: From underlying physiological concepts to its usefulness in clinical conditions. J Magn Reson Imaging 35, 1253–1265. [DOI] [PubMed] [Google Scholar]
  22. Janssen BH, Voet NBM, Nabuurs CI, Kan HE, De Rooy JWJ, Geurts AC, Padberg GW, Van Engelen BGM & Heerschap A (2014). Distinct disease phases in muscles of facioscapulohumeral dystrophy patients identified by MR detected fat infiltration. PLoS One 9, e85416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Johnson MA, Polgar J & Appleton D (1973). Data on the distribution of fibre types in thirty‐six human muscles an autopsy study. J Neurol Sci 18, 111–129. [DOI] [PubMed] [Google Scholar]
  24. Kemp GJ, Ahmad RE, Nicolay K & Prompers JJ (2015). Quantification of skeletal muscle mitochondrial function by 31P magnetic resonance spectroscopy techniques: a quantitative review. Acta Physiol (Oxf) 213, 107–144. [DOI] [PubMed] [Google Scholar]
  25. Koga S, Barstow TJ, Okushima D, Rossiter HB, Kondo N, Ohmae E & Poole DC (2015). Validation of a high‐power time resolved near‐infrared spectroscopy system for measurement of superficial and deep muscle deoxygenation during exercise. J Appl Physiol 118, 1435–1442. [DOI] [PubMed] [Google Scholar]
  26. Laaksonen MS, Kalliokoski KK, Kyrolainen H, Kemppainen J, Teras M, Sipila H, Nuutila P & Knuuti J (2003). Skeletal muscle blood flow and flow heterogeneity during dynamic and isometric exercise in humans. Am J Physiol Hear Circ Physiol 284, H979–H986. [DOI] [PubMed] [Google Scholar]
  27. Larsen RG, Callahan DM, Foulis SA & Kent‐Braun JA (2009). In vivo oxidative capacity varies with muscle and training status in young adults. J Appl Physiol 107, 873–879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Layec G, Malucelli E, Le Fur Y, Manners D, Yashiro K, Testa C, Cozzone PJ, Iotti S & Bendahan D (2013). Effects of exercise‐induced intracellular acidosis on the phosphocreatine recovery kinetics: a 31P MRS study in three muscle groups in humans. NMR Biomed 26, 1403–1411. [DOI] [PubMed] [Google Scholar]
  29. Le Bihan D, Breton E , Lallemand D, Aubin ML, Vignaud J & Laval‐Jeantet M (1988). Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging. Radiology 168, 497–505. [DOI] [PubMed] [Google Scholar]
  30. Lexell J, Henriksson‐Larsén K & Sjöström M (1983). Distribution of different fibre types in human skeletal muscles. 2. A study of cross‐sections of whole m. vastus lateralis. Acta Physiol Scand 117, 115–122. [DOI] [PubMed] [Google Scholar]
  31. Lodi R, Kemp GJ, Iotti S, Radda GK & Barbiroli B (1997). Influence of cytosolic pH on in vivo assessment of human muscle mitochondrial respiration by phosphorus magnetic resonance spectroscopy. MAGMA 5, 165–171. [DOI] [PubMed] [Google Scholar]
  32. Louie EA, Gochberg DF, Does MD & Damon BM (2009). Transverse relaxation and magnetization transfer in skeletal muscle: Effect of pH. Magn Reson Med 61, 560–569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Luyten PR, Bruntink G, Sloff FM, Vermeulen JWAH, Van Der Heijden JI, Den Hollander JA & Heerschap A (1989). Broadband proton decoupling in human 31P NMR spectroscopy. NMR Biomed 1, 177–183. [DOI] [PubMed] [Google Scholar]
  34. McCully KK, Fielding RA, Evans WJ, Leigh JS & Posner JD (1993). Relationships between in vivo and in vitro measurements of metabolism in young and old human calf muscles. J Appl Physiol 75, 813–819. [DOI] [PubMed] [Google Scholar]
  35. Meyer A (1988). A linear model of muscle respiration explains monoexponential phosphocreatine changes. Am J Physiol 254, C548–C553. [DOI] [PubMed] [Google Scholar]
  36. Meyer RA & Prior BM (2000). Functional magnetic resonance imaging of muscle. Exerc Sport Sci Rev 28, 89–92. [PubMed] [Google Scholar]
  37. Meyerspeer M, Robinson S, Nabuurs CI, Scheenen T, Schoisengeier A, Unger E, Kemp GJ & Moser E (2012). Comparing localized and nonlocalized dynamic 31P magnetic resonance spectroscopy in exercising muscle at 7T. Magn Reson Med 68, 1713–1723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Meyerspeer M, Scheenen T, Schmid AI, Mandl T, Unger E & Moser E (2011). Semi‐LASER localized dynamic 31P magnetic resonance spectroscopy in exercising muscle at ultra‐high magnetic field. Magn Reson Med 65, 1207–1215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mizuno M, Mimura Y, Iwakawa T, Oda K, Ishii K, Ishiwata K, Nakamura Y & Muraoka I (2003). Regional differences in blood flow and oxygen consumption in resting muscle and their relationship during recovery from exhaustive exercise. J Appl Physiol 95, 2204–2210. [DOI] [PubMed] [Google Scholar]
  40. Moon RB & Richards JH (1973). Determination of intracellular pH by 31P magnetic resonance. J Biol Chem 248, 7276–7278. [PubMed] [Google Scholar]
  41. Niess F, Fiedler GB, Schmid AI, Goluch S, Kriegl R, Wolzt M, Moser E & Meyerspeer M (2016). Interleaved multivoxel 31P MR spectroscopy. Magn Reson Med 77, 921–927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Parasoglou P, Feng L, Xia D, Otazo R & Regatte RR (2012). Rapid 3D‐imaging of phosphocreatine recovery kinetics in the human lower leg muscles with compressed sensing. Magn Reson Med 68, 1738–1746. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Phielix E, Schrauwen‐Hinderling VB, Mensink M, Lenaers E, Meex R, Hoeks J, Kooi ME, Moonen‐Kornips E, Sels JP, Hesselink MKC & Schrauwen P (2008). Lower intrinsic ADP‐stimulated mitochondrial respiration underlies in vivo mitochondrial dysfunction in muscle of male type 2 diabetic patients. Diabetes 57, 2943–2949. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Quistorff B, Johansen L & Sahlin K (1993). Absence of phosphocreatine resynthesis in human calf muscle during ischaemic recovery. Biochem J 291, 681–686. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Reid RW, Foley JM, Jayarama RC, Prior BM & Meyer RA (2001). Effect of aerobic capacity on the T2 increase in exercised skeletal muscle. J Appl Physiol 90, 897–902. [DOI] [PubMed] [Google Scholar]
  46. Ryschon TW, Fowler MD, Wysong RE, Anthony A & Balaban RS (1997). Efficiency of human skeletal muscle in vivo: comparison of isometric, concentric, and eccentric muscle action. J Appl Physiol 83, 867–874. [DOI] [PubMed] [Google Scholar]
  47. Schindelin J, Arganda‐Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez J‐Y, White DJ, Hartenstein V, Eliceiri K, Tomancak P & Cardona A (2012). Fiji: an open‐source platform for biological‐image analysis. Nat Methods 9, 676–682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Schmid AI, Schewzow K, Fiedler GB, Goluch S, Laistler E, Wolzt M, Moser E & Meyerspeer M (2014). Exercising calf muscle T2∗changes correlate with pH, PCr recovery and maximum oxidative phosphorylation. NMR Biomed 27, 553–560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Söderlund K & Hultman E (1991). ATP and phosphocreatine changes in single human muscle fibers after intense electrical stimulation. Am J Physiol 261, E737–E741. [DOI] [PubMed] [Google Scholar]
  50. Stefan D, Cesare FDi, Andrasescu A, Popa E, Lazariev A, Vescovo E, Strbak O, Williams S, Starcuk Z, Cabanas M, Ormondt D Van & Graveron‐Demilly D (2009). Quantitation of magnetic resonance spectroscopy signals: the jMRUI software package. Meas Sci Technol 20, 1–9.20463843 [Google Scholar]
  51. Steinseifer IK, Wijnen JP, Hamans BC, Heerschap A & Scheenen TWJ (2013). Metabolic imaging of multiple X‐nucleus resonances. Magn Reson Med 70, 169–175. [DOI] [PubMed] [Google Scholar]
  52. Takahashi H, Inaki M, Fujimoto K, Katsuta S, Anno I, Niitsu M & Itai Y (1995). Control of the rate of phosphocreatine resynthesis after exercise in trained and untrained human quadriceps muscles. Eur J Appl Physiol Occup Physiol 71, 396–404. [DOI] [PubMed] [Google Scholar]
  53. Taylor DJ, Bore PJ, Styles P, Gadian DG & Radda GK (1983). Bioenergetics of intact human muscle. A 31P nuclear magnetic resonance study. Mol Biol Med 1, 77–94. [PubMed] [Google Scholar]
  54. Torrella JR, Whitmore JM, Casas M, Fouces V & Viscor G (2000). Capillarity, fibre types and fibre morphometry in different sampling sites across and along the tibialis anterior muscle of the rat. Cells Tissues Organs 167, 153–162. [DOI] [PubMed] [Google Scholar]
  55. van Uden MJ, Veltien A, Janssen BH & Heerschap A (2012). 1H/31P birdcage coil combined with dedicated multi‐element 31P receive coil for optimal 31P MRSI of the tibialis anterior. Proc Intl Soc Mag Reson Med. [Google Scholar]
  56. Valkovič L, Chmelík M, Meyerspeer M, Gagoski B, Rodgers CT, Krššák M, Andronesi OC, Trattnig S & Bogner W (2016). Dynamic 31P‐MRSI using spiral spectroscopic imaging can map mitochondrial capacity in muscles of the human calf during plantar flexion exercise at 7 T. NMR Biomed 29, 1825–1834. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Vandenborne K, Walter G, Ploutz‐Snyder L, Dudley G, Elliott MA & De Meirleir K (2000). Relationship between muscle T2* relaxation properties and metabolic state: a combined localized 31P‐spectroscopy and 1H‐imaging study. Eur J Appl Physiol 82, 76–82. [DOI] [PubMed] [Google Scholar]
  58. van den Broek NMA, Graaf L De, Nicolay K & Prompers JJ (2007). Intersubject differences in the effect of acidosis on phosphocreatine recovery kinetics in muscle after exercise are due to differences in proton efflux rates. Am J Physiol Cell Physiol 293, C228–C237. [DOI] [PubMed] [Google Scholar]
  59. Wang LC & Kernell D (2001). Quantification of fibre type regionalisation: an analysis of lower hindlimb muscles in the rat. J Anat 198, 295–308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Ward SR, Eng CM, Smallwood LH & Lieber RL (2009). Are current measurements of lower extremity muscle architecture accurate? Clin Orthop Relat Res 467, 1074–1082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. West BT (2009). Analyzing longitudinal data with the linear mixed models procedure in SPSS. Eval Heal Prof 32, 207–228. [DOI] [PubMed] [Google Scholar]
  62. Yoshida T, Abe D & Fukuoka Y (2013). Phosphocreatine resynthesis during recovery in different muscles of the exercising leg by 31P‐MRS. Scand J Med Sci Sport 23, e313–319. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES