In Vitro sup 31 Phosphorus-Magnetic Resonance Spectroscopy of Muscle Extracts in Malignant Hyperthermia-susceptible Patients

Muscle specimens from 17 patients referred to the Grenoble MH Diagnosis Center were studied. Muscle biopsy was performed because of a family history of MH or symptoms of MH during anesthesia. Open surgical biopsies of the vastus medialis muscles were performed under regional anesthesia. One or two muscle specimens 2-2.5 cm long and weighing 440 mg (270-1067 mg) were quickly frozen and stored in liquid nitrogen until the PCA extraction procedure. Other muscle specimens were processed for histologic analysis and for the contracture test, which was performed according to the European MH Group protocol. [10]The test was considered positive when the muscle specimens produced a sustained increase in baseline tension with an amplitude greater or equal to 0.2 g during application of both less or equal to 2% halothane and less or equal to 2 mM caffeine. The test was equivocal when a contracture was elicited by only one test substance. Other results were deemed MH-negative.

The PCA extracts from muscle specimens were prepared as described previously, [11]within a 3-month period after biopsy. Briefly, the frozen biopsy specimen was pulverized with a liquid nitrogen-chilled mortar and pestled to a fine powder. The tissue was transferred and homogenized in a Potter-Elvehjem homogenizer with 0.6 mol l sup -1 of PCA (10 ml g sup -1 wet weight). After centrifugation at 41,400g for 10 min at 4 degrees C, the pH of the supernatant was adjusted to 6.6-7.0 with potassium hydroxide. This procedure was repeated twice, and the resulting supernatant was lyophilized. The powder was then resuspended in D²O solution (Laboratories Merck-Clevenot, Dardilly, France), centrifuged at 12,500g for 10 min and the supernatant was lyophilized. All the samples obtained after this PCA extraction procedure were stored in liquid nitrogen. Before the MRS experiment, the extract was resuspended in 350 micro liter D²O solution at pH 8.4 and centrifuged at 12,500g for 10 min.

Magnetic resonance spectroscopic experiments were performed with a Bruker AM400 spectrometer (Bruker Spectrospin, Wissembourg, France) equipped with a 9.4-T, 89-mm diameter vertical bore magnet. The³¹Phosphorus spectra were acquired using a 5-mm diameter multi-nuclear probe tuned to³¹Phosphorus frequency (162.4 MHz) under the following conditions: single-pulse sequence of 15.5 micro second pulse length with a composite proton sequence decoupling; 4.05 s repetition time; 6 or 7 blocks of 1,600 scans; 16,384 data points per free induction decay; 6410-Hz spectral width; 3-Hz line broadening. The PCr was chosen as reference of chemical shift (0 ppm). The quantification was based on the integration of the peaks of phosphorylated compounds using UXNMR software (Bruker, Rheinstetten, Germany). Results were expressed as areas relative to the sum of PCr and inorganic phosphate (Pi) areas (PCr + Pi) or to the sum of the areas of all phosphorylated compounds visible by³¹P-MRS (Epsilon P).

The following criteria were used to identify the chemical compounds present in the PDE region: (1) the correspondence between the observed chemical shifts in the muscle PCA extracts and the known chemical shifts for pure compounds at different pH values (titration experiment), (2) the persistence of a single spectral line of enhanced intensity on addition of a standard at a final concentration of 5 mg ml sup -1 to the PCA extract at more than one pH value (coresonance experiment). The standard compounds were L-alpha-glycerophosphoryl-ethanolamine, -inositol, -serine, -choline (GPC; Sigma, Saint-Quentin Fallavier, France).

Sixteen muscle specimens were analyzed histologically by optical microscopy. Sections were cut on a freezing microtome and stained histochemically for myosin ATPase at pH 4.35 to perform the fiber type composition. The counting of fiber types and their mean diameters were determined using an image analyzer equipped with the Histo 200 software (Biocom, Lyon, France). The mean cross-sectional area of each muscle fiber type was then automatically calculated.

Data were expressed as medians and ranges. The³¹P-MRS and histologic parameters of MHS and MHN groups were compared using the nonparametric Mann-Whitney U test at the 0.05 level of significance.

The contracture tests revealed that seven patients (median 37 yr, range 16-63) were MHS and ten patients (median 23 yr, range 12-51) MHN. None were equivocal for responses to halothane or caffeine. Figure 1shows the PDE region in muscle PCA extract from an MHS patient, before and after sequential addition of PDE compounds. In the PDE region, a major peak was detected at 3.0 ppm. When standard compounds were added to the muscle PCA extract, there was an exact coincidence of the PDE peak only with the GPC standard (coresonance experiment). In addition, there were similar changes in the pH-dependent chemical shifts of the PDE peak of the muscle PCA extract spectrum, the pure GPC peak in D²O solution and the GPC peak when added to the muscle PCA extracts (Figure 2; titration experiment). The major peak in PDE region was thus assigned to GPC.

Typical³¹Phosphorus spectra of muscle PCA extract obtained from MHN and MHS patients are shown in Figure 3. In these spectra, the GPC/(Pi + PCr) ratio was measurably higher in the muscle PCA extract of the MHS patient. For the two groups, the phosphomonoesters (PME) to Epsilon P (PME/Epsilon P), GPC/Epsilon P and GPC/(Pi + PCr) ratios were significantly higher in MHS muscle extracts than in MHN extracts (Figure 4and Table 1). The mean of the sum of ranks was 6.3 versus 12.8 for the PME/Epsilon P ratio, 6.8 versus 12.1 for the GPC/Epsilon P ratio, and 6.7 versus 12.3 for the GPC/(Pi/PCr) ratio in the MHS and MHN groups, respectively. There was no difference between the muscle specimens of MHS and MHN patients concerning the presence of type I fibers (43% [27.9-52.8] vs. 41.4% [28.3-50.5], respectively) and the mean cross-sectional area of type I fibers (3,019 micro²[2206-4657] vs. 2,734 micro²[1,257-3,217], respectively). None of the patients had specific myopathy on histologic examination of muscle specimens.

A significantly higher relative concentration of GPC was found in muscle PCA extracts of the MHS patients as compared to the MHN patients. In the absence of histologic differences between muscle specimens of the two groups, these results suggest that GPC could be a marker of an impairment in the phospholipid metabolism in the skeletal muscle of MHS patients. However, there was an overlap between MHS and MHN values (Figure 4), limiting diagnostic potential of these in vitro results.

Each phosphorus-containing chemical group has a characteristic chemical shift position, which is determined by the molecular environment of the phosphorus atom contained within different phosphorylated compounds. However, ionic strength, pH, cationic, and protein concentration in cell compartments may complicate the assignment of a resonance peak recorded from intact tissue. Therefore, coresonance experiments at different pH levels are usually done to assign the resonances observed in³¹Phosphorus spectra of PCA extracts. [11,12]According to the PCA extract procedure, GPC was clearly identified as the primary component of the PDE peak in muscle PCA extract of MHS patients. The other PDE compounds appeared to be of minor contribution in muscle PCA extracts, as previously reported in normal muscle. [8]However, no absolute concentrations of GPC and other phosphorylated compounds were provided in the current study because of the absence of an external standard. In addition, the muscle specimens of different weights were resuspended in the same volume of D²O solution, leading to different signal-to-noise ratio between PCA extracts (Figure 3(A and B)). Internal references were thus chosen such as Epsilon Phosphorus and Pi + PCr, because there is a stoichiometric change between the decrease in the PCr peak intensity and the increase in the Pi peak intensity, so the sum Pi + PCr remains constant. When the PCA extract procedure is properly carried out, no water-soluble metabolites such as PCr, Pi, and GPC can be significantly lost by precipitation. [13]Finally, no significant difference in the sum of Pi and PCr has been found in quadriceps muscle between MHS and MHN patients. [14].

Relative concentration of GPC was significantly elevated in the vastus medialis muscle extracts of MHS patients. This appears to be in accordance with the elevated PDE/PCr ratio observed in vivo in³¹Phosphorus spectra of the gastrocnemius muscle. [3,4]It has been reported that GPC is absent from the human biceps muscle, [7]a finding that could explain the discrepancy in the PDE peak between forearm and leg muscles found in MHS patients by in vivo³¹P-MRS. [1-4]Conversely, the highest concentration of GPC was found in slow twitch muscle containing type I fibers, e.g., in soleus muscle, and to a lesser extent, in the quadriceps muscle. [7,8]No difference in the percentage of type I fibers between MHS and MHN muscles was found in the current study, as previously reported in vastus lateralis muscle. [15]Because none of the patients had specific myopathy, these results suggest that the elevated GPC could be related to the MH susceptibility.

Glycerophosphorylcholine is a water-soluble metabolite produced by the catabolism of the membrane phospholipids, from the pathway of the phosphatidylcholine deacylation: production of lysophosphatidylcholine via phospolipase A², followed by production of GPC via lysophospholipase. It has been suggested that GPC may regulate membrane phospholipid composition by acting on the enzyme lysophospholipase by a product inhibition. [16]An increase of GPC could thus result in an increase in lysophosphatidylcholine content, and consequently affect the membrane integrity. [9,17]This hypothesis could explain the general, although controversial, defect of membrane permeability in the MH syndrome. [18-20]However, similar levels of lysophosphatidylcholine and phosphatidylcholine have been found in vastus lateralis muscle between MHS and MHN patients. [21]In this way, an elevated GPC level could either imply an accelerated synthesis of phosphatidylcholine resulting in an accumulation of GPC, or a decreased hydrolysis of GPC. The determination of the enzymatic activities leading to the synthesis of phosphatidylcholine as well as to the degradation of GPC would thus be useful to elucidate the mechanisms that could account for the data.

The PME/Epsilon Phosphorus ratio was higher in the muscle PCA extracts of MHS patients than those of MHN patients. The phosphomonoesters include a number of compounds such as triose phosphates, hexose aliphatic phosphates, pentose phosphates, inosine and adenosine monophosphate, choline, and phosphoethanolamine. [12]However, no chemical identification of the PME region was performed in the current study and the significance of such results remains unknown.

These in vitro³¹P-MRS results demonstrate that the skeletal muscle of MHS patients exhibits a higher relative GPC content than that of MHN patients. Because this difference cannot be explained by histologic changes between MHS and MHN muscle specimens, these results suggest that GPC could be a marker of an impairment in the phospholipid metabolism in the skeletal muscle of MHS patients.

The authors thank J. Viret, the Department of Biophysics of The Military Health Research Center of Grenoble, for magnetic resonance spectroscopy facilities. They also thank Raphael Massarelli and Simon Doran, for their comments.