0882

1H-NMR of carnosine combined with 31P-NMRS to better characterize skeletal muscle pH dysregulation in Duchenne muscular dystrophy
Harmen Reyngoudt1,2, Suna Turk1,2, and Pierre G. Carlier1,2

1NMR Laboratory, Institute of Myology, Paris, France, 2CEA, DRF, I²BM, MIRCen, Paris, France

Synopsis

The finding of an alkaline Pi pool has been established with 31P-NMRS in healthy resting skeletal muscle and the dystrophic muscle of GMRD dogs and DMD patients. The pH values corresponding to this Pi pool corresponds better with extracellular pH. Intracellular pH, however, can also be measured with 1H-NMRS, using carnosine. In a group of DMD patients, we observed that pH determined with 31P-NMRS were systematically increased, whereas this was not always the case for pH based on the measurement of carnosine, revealing two groups in dystrophic muscle: (1) pH elevated with 31P and 1H and (2) pH only elevated with 31P.

Purpose

Quantitative nuclear magnetic resonance imaging (NMRI) and phosphorus NMR spectroscopy (31P-NMRS) have been used to investigate the mechanisms of Duchenne muscular dystrophy (DMD), the most common muscle-wasting disorder. Splitting of the inorganic phosphate (Pi) peak, showing a second Pi peak (Pi,b) upfield from Pi (Pi,a) was shown in healthy resting skeletal muscle [1] as well as in skeletal muscle of GRMD (Golden Retriever muscular dystrophy) dogs, the canine model for DMD [2], and in upper limb skeletal muscle of DMD patients [3], reflecting a more alkaline Pi pool. The pH values observed for Pi,b were too alkaline to correspond to healthy tissue (pH around 7.0), and corresponded better to pH values found in interstitial or extracellular space. Intracellular pH, however, can also be measured with proton NMRS (1H-NMRS), exploiting the two imidazole protons of carnosine (Fig. 1) [4]. In resting muscle, carnosine is exclusively found in the cytosol, suggesting pH calculated with 1H-NMRS might be different from pH calculated with 31P-NMRS in suffering muscle cells. The aim of this work was to perform both 1H-NMRS and 31P-NMRS at rest in DMD patients and healthy controls.

Methods

Data was acquired in 18 DMD patients (all male, 10 ± 2 yrs), as well as in 7 healthy controls (4 male, 31 ± 4 yrs) on a 3T Siemens Prisma system using a body matrix coil for quantitative NMRI, a dual-tuned 31P-1H surface coil for 31P-NMRS or a 15-channel volume knee coil for 1H-NMRS. 1H-NMRS data was obtained using a PRESS sequence with a TR of 3000 ms and a TE of 30 ms. Voxels were positioned in the gastrocnemius medialis muscle and voxel size depended on muscle size. Two 1H-NMR spectra were acquired: one water-suppressed (central frequency at 8.0 ppm, 64 averages) and one unsuppressed water spectrum (central frequency at 4.7 ppm, 16 averages). 31P-NMRS data was obtained from a non-localized excitation (TR = 4000 ms, 64 averages) with the coil wrapped around the calf (Fig. 1). NMRS data was processed with the AMARES [5] algorithm from jMRUI using optimized prior knowledge (Fig. 2,3). Additionally, an MSME sequence with 17 echoes was acquired which was processed using in-house python code, extracting water T2 and fat fraction values based on the tri-exponential model of Azzabou et al. [6]. Kruskal-Wallis tests for comparisons of groups and Spearman correlation analysis was performed (with P = 0.05 as significance level).

Results

Fig. 4a shows that pH values derived from 31P-NMRS were systematically increased in DMD patients. While there is good agreement between 1H and 31P pH determination in normal subjects (group A), discrepancies can be observed in DMD patients. Whereas all have alkaline pH when determined by 31P-NMRS, there are DMD subjects with normal intracellular pH as calculated from the carnosine resonance (group B), the others have an increased (intracellular) pH measured by the two methods (group C), as illustrated in Fig. 4a. Interestingly, the carnosine intracellular pH was never found alkaline in absence of concurrent Pi pH elevation. The same observation can be made for the relation between 1H derived pH and water T2. Abnormal intracellular pH is hardly ever associated with normal water T2 values (Fig. 4b). Elevated 31P-derived pH values are observed in both the presence and absence of elevated water T2 (Fig. 4c). Data of the three groups are summarized in Table 1.

Discussion

While pH alterations in the dystrophic muscle had been described long ago, the mechanism for the prominence of an alkaline Pi resonance in dystrophic muscle was still being debated. It may reflect cell pH dysregulation in leaky damaged myocytes or an increased volume fraction of the interstitium, impossible to distinguish using 31P-NMRS. Magnetization transfer experiments have indicated that, at least partly, the alkaline Pi pool was in exchange with ATP, and consequently originating from cells. The combined 1H and 31P approach used here identified that the two proposed mechanisms can exist. In muscles where 1H and 31P pH estimates are in agreement, the lesional mechanisms predominate. When 1H and 31P measures are discordant, it indicates that the alkaline Pi pool is extracellular, with an increased interstitial volume fraction being possibly related to fibrotic changes. Consequently, it will be worth investigating whether carnosine-based intracellular pH may provide an early indicator of dystrophic myocytes to therapy, as opposed to the 31P-NMRS-based pH, whose origin cannot be ascertained and may reflect predominantly the extent of chronic degenerative changes in dystrophic muscles.

Conclusion

Contrary to 31P-NMRS, carnosine 1H-NMRS can be used as a biomarker to assess specifically the intramyocytic pH changes in Duchenne patients.

Acknowledgements

No acknowledgement found.

References

[1] Kan HE., Klomp DWJ, Wong CS, et al. In vivo 31P MRS detection of an alkaline inorganic phosphate pool with short T1 in human resting skeletal muscle. NMR Biomed. 2010;23(8): 995–1000.

[2] Wary C, Naulet T, Thibaud JL, et al. Splitting of Pi and other 31P NMR anomalies of skeletal muscle metabolites in canine muscular dystrophy. NMR Biomed. 2012;25(10):1160–1169.

[3] Wary C, Azzabou N, Giraudeau C, et al. Quantitative NMRI and NMRS identify augmented disease progression after loss of ambulation in forearms of boys with Duchenne muscular dystrophy. NMR Biomed. 2015;28(9):1150–1162.

[4] Pan JW, Hamm JR, Rothman DL, et al. Intracellular pH in human skeletal muscle by 1H NMR. Proc. Natl. Acad. Sci. USA. 1988;85(21):7836–7839.

[5] Vanhamme L, van den Boogaart A, Van Huffel S. Improved method for accurate and efficient quantification of MRS data with use of prior knowledge. J. Magn. Reson. 1997;129(1):35–43.

[6] Azzabou N, de Sousa PL, Caldas E, et al. Validation of a generic approach to muscle water T2 determination at 3T in fat-infiltrated skeletal muscle. J. Magn. Reson. Imaging. 2015;41(3):645–653.

Figures

Fig. 1: Axial slice of T2 map with (red) voxel placed in gastrocnemius medialis of DMD patient, where the 1H-NMR spectra were acquired. In purple is indicated how the 31P-coil was positioned to acquire the 31P-NMR spectrum. For water T2 and fat fraction analysis, the same ROI as in the red voxel was drawn in ITK-Snap (open-source segmentation software).

Fig. 2: The two carnosine protons, C2-H and C4-H, resonate upfield of the water resonance at around 8.0 and 7.0 ppm, respectively. The water-suppressed 1H-NMR spectrum was frequency-corrected based on the water spectrum (4.7 ppm). The spectrum between 0 and 4 ppm was removed using the HLSVD algorithm. After processing with AMARES, the pH values were calculated with the Henderson-Hasselbalch equation using carnosine pKa, the alkalic and acidic limiting chemical shifts and the chemical shift difference between C2-H and water.

Fig. 3: 31P-NMR spectrum in the same patient with visible splitting of Pi. The sum of Pi,a and Pi,b is referred to as Pi,tot. 31P-NMRS ratio indices such as Pi,tot/PCr, Pi,b/Pi,tot, Pi,tot/γATP, and PCr/γATP were also evaluated with AMARES. The pH values were calculated with the Henderson-Hasselbalch equation using pKa of Pi, the alkalic and acidic limiting chemical shifts and the chemical shift difference between Pi and PCr. PME = phosphomonoesters; PDE = phosphodiesters; γATP, αATP and βATP = three resonances from adenosine triphosphate.

Fig. 4: (a) Strong correlation between pH 31P-NMRS and pH 1H-NMRS. Three different groups can be observed based on combinations of both pH-values. A: both pH values normal; B: pH 1H-NMRS normal and pH 31P-NMRS elevated; C: both pH values elevated; (b) Good correlation between pH 1H-NMRS and water T2 values; (c) Strong correlation between pH 31P-NMRS and water T2 values.

Table 1: Differences in pH values, 31P-NMRS indices, water T2 values and fat fraction values between the three groups.(1) significant difference between group A and group B; (2) significant difference between group A and group C; (3) significant difference between group B and group C (significant P-values all < 0.03).

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)
0882