Measuring energy diffusion: phosphocreatine in human skeletal muscle

Introduction

The creatine kinase (CK) reaction plays a central role in energy transfer in the myocyte, shuttling high-energy phosphate (HEP) created in the mitochondria, to sites of utilization in the myofibrils [1,2]. At the mitochondria, the reverse reaction produces phosphocreatine (PCr) from adenosine triphosphate (ATP) and creatine (Cr). PCr diffuses to the myofibrils to regenerate ATP from adenosine diphosphate to fuel muscular contraction. The PCr energy transfer shuttle has been widely modeled. Knowledge of the diffusion coefficient of PCr, D_PCr, is key, but ill-defined in humans. Localized ³¹P MRS is uniquely able to access and quantify PCr and forward CK flux noninvasively [3]. The aim of this study is to develop and optimize a ³¹P MRS protocol for measuring D_PCr, and to characterize PCr energy transport in human muscle in relation to CK flux. We report D_PCr in 8 healthy subjects as a function of diffusion time (t_diff), and that PCr diffusion in muscle is anisotropic.

Methods

A 3-pulse, stimulated echo sequence was chosen to permit D_PCr measurements with long t_diff's (≫T₂) (t_diff =TM+32ms, TM is the mixing time). Diffusion encoding was provided via gradient lobes placed after the first and third RF pulses. Spectra were localized to a plane parallel to a 17-cm/8cm transmit/receive ³¹P coil set using depth-resolved surface coil spectroscopy (DRESS) [4], with the first pulse slice-selective (TR/TE= 8000/80ms; bandwidth =500Hz). D_PCr was measured at t_diff's corresponding to TM of 50,100,150,200,400,700, and 1000 ms, by acquiring one spectrum without diffusion gradients, and three spectra with diffusion gradients applied on the x, y, and z axes. The b-value was chosen to optimize precision of the D_PCr measurement (1000≤b≤2000 s/mm², b= [γ.G.δ]².t_diff, with γ is the gyromagnetic ratio, G is the diffusion gradient strength, and δ is the diffusion gradient duration). Averaging was adjusted to maintain adequate and comparable signal-to-noise ratio for each acquisition. Peak areas were fitted using CFIT [5]. D_PCr was obtained in the three directions (D_xx, D_yy, D_zz) and the mean diffusivity D_av was calculated as D_av= (D_xx+D_yy+D_zz)/3. The calf muscles of eight healthy subjects who gave informed consent were scanned on a Philips Achieva 3.0T system.

Results

Fig. 1 shows typical spectra from one volunteer at TM=200ms. Table 1 lists average D_xx, D_yy, D_zz, and D_av for PCr as a function of TM for all subjects. D_PCr is significantly higher along the fiber direction (∼z-axis) than transverse directions (P≤0.02 for D_zz vs D_yy and D_zz vs D_xx at TM≤150ms). The slow decrease in D_PCr with TM suggests that diffusion is ultimately restricted.

Fig. 1. PCr spectra from the calf muscle using stimulated echo DRESS at TM=200ms, without (S₀) and with diffusion weighting along the three physical axes (S_x,S_y,S_z).

Table 1. The diffusion coefficient of PCr in the calf muscle of eight healthy volunteers.

TM (ms)	D(×10-3 mm2/s)
	Dxx	Dyy	Dzz	Dav
50	0.44±.13	0.45±.10	0.81±.28	0.56±.12
100	0.41±.10	0.41±.10	0.58±.13	0.46±.08
150	0.43±.05	0.41±.06	0.60±.18	0.47±.07
200	0.46±.07	0.44±.07	0.57±.10	0.48±.04
400	0.39±.08	0.37±.09	0.51±.05	0.42±.06
700	0.38±.11	0.32±.11	0.40±.04	0.38±.06
1000	0.31±.12	0.30±.10	0.43±.08	0.35±.06

Discussion

We believe these are the first studies showing anisotropic and restricted PCr diffusion in human muscle. The values are comparable to data from animal limbs [6,7]. Prior ³¹P saturation transfer MRS in healthy human muscle shows that ∼27% of the PCr turns over through CK per sec [3]. This would yield a half-life for PCr of ∼2.6s at rest. Our results show that during this half-life, PCr will have diffused ∼65μm (extrapolated D_av ≈ 0.27 × 10^-3 mm²/s). Based on rat muscle [8], the expected separation between mitochondria and myofibril is up to 2μm. If this is similar in humans, PCr diffusion would certainly be fast enough to transfer HEP between mitochondria and myofibrils. This situation may be altered in diseases that affect energy transfer.