In vivo assessment of OXPHOS capacity using 3 T CrCEST MRI in Friedreich's ataxia

Abstract

Background: Friedreich's ataxia (FRDA) is a neurodegenerative disease caused by decreased expression of frataxin, a protein involved in many cellular metabolic processes, including mitochondrial oxidative phosphorylation (OXPHOS). Our objective was to assess skeletal muscle oxidative metabolism in vivo in adults with FRDA as compared to adults without FRDA using chemical exchange saturation transfer (CrCEST) MRI, which measures free creatine (Cr) over time following an in-magnet plantar flexion exercise.

Methods: Participants included adults with FRDA (n = 11) and healthy adults (n = 25). All underwent 3-Tesla CrCEST MRI of the calf before and after in-scanner plantar flexion exercise. Participants also underwent whole-body dual-energy X-ray absorptiometry (DXA) scans to measure body composition and completed questionnaires to assess physical activity.

Results: We found prolonged post-exercise exponential decline in CrCEST (τCr) in the lateral gastrocnemius (LG, 274 s vs. 138 s, p = 0.01) in adults with FRDA (vs. healthy adults), likely reflecting decreased OXPHOS capacity. Adults with FRDA (vs. healthy adults) also engaged different muscle groups during exercise, as indicated by muscle group-specific changes in creatine with exercise (∆CrCEST), possibly reflecting decreased coordination. Across all participants, increased adiposity and decreased usual physical activity were associated with smaller ∆CrCEST.

Conclusion: In FRDA, CrCEST MRI may be a useful biomarker of muscle-group-specific decline in OXPHOS capacity that can be leveraged to track within-participant changes over time. Appropriate participant selection and further optimization of the exercise stimulus will enhance the utility of this technique.

Keywords: Exercise; Friedreich’s ataxia; Magnetic resonance imaging; Mitochondrial disorders; OXPHOS; Oxidative metabolism; Skeletal muscle.

Fig. 1

CrCEST over time for a healthy 25yo male and a 30yo male with FRDA, respectively. (A, B) Maps of CrCEST (%asymmetry), rest–exercise–recovery protocol for (A) healthy 25yo male where τCr = 269 s, 263 s, 287 s and (B) 30yo male with FRDA where τCr = 741 s, 451 s, 367 s for the LG, MG, and soleus, respectively. The color bar indicates the intensity of the CrCEST signal, in proportion to the concentration of Cr in the muscle. (C, D) CrCEST over time for the same participants; prolonged recovery corresponds to decreased OXPHOS capacity. CrCEST in the LG at t = 60 s was omitted from the time series plot in Fig. 1D due to an error in image acquisition at that timepoint but was included in model fitting. LG: lateral gastrocnemius, MG: medial gastrocnemius, Sol: soleus

Fig. 2

Mean CrCEST timeseries for all adults in the (A) lateral gastrocnemius, (B) medial gastrocnemius, and (C) soleus. In the LG, median τCr = 274 s (IQI 221–309 s) in FRDA and τCr = 138 s (IQI 85–226 s) in controls (p = 0.01 for difference by Kruskal–Wallis test, Table 2)

Fig. 3

Post-exercise τCr in adults. Orange corresponds to controls, and blue corresponds to adults with FRDA. Median τCr in the LG was 138 s (IQI = 85–227 s) in controls and 274 s (IQI = 221–309 s) in FRDA (Table 2). In a linear regression analysis, FRDA disease status prolongs τCr by 131 s (95% CI = 28–234 s) in the LG, accounting for sex and age (p = 0.01, Table 3). This boxplot also illustrates the interaction between disease status and muscle group described, namely that disease-specific differences in post-exercise τCr in adults with FRDA are distinct in soleus as compared to LG. LG: lateral gastrocnemius, MG: medial gastrocnemius, Sol: soleus