Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jun 8.
Published in final edited form as: Can J Anaesth. 2021 Jan 6;68(6):761–772. doi: 10.1007/s12630-020-01895-y

Untargeted metabolomics profiling of skeletal muscle samples from malignant hyperthermia susceptible patients

Profilage métabolomique non ciblé d’échantillons de muscles squelettiques de patients susceptibles à l’hyperthermie maligne

Barbara Bojko 1,#, Tijana Vasiljevic 2,#, Ezel Boyaci 3, Anna Roszkowska 4, Natalia Kraeva 5, Carlos A Ibarra Moreno 6, Annabel Koivu 7, Marcin Wąsowicz 8, Amy Hanna 9, Susan Hamilton 10, Sheila Riazi 11, Janusz Pawliszyn 12
PMCID: PMC8185566  NIHMSID: NIHMS1706163  PMID: 33403543

Abstract

Purpose

Malignant hyperthermia (MH) is a potentially fatal hypermetabolic condition triggered by certain anesthetics and caused by defective calcium homeostasis in skeletal muscle cells. Recent evidence has revealed impairment of various biochemical pathways in MH-susceptible patients in the absence of anesthetics. We hypothesized that clinical differences between MH-susceptible and control individuals are reflected in measurable differences in myoplasmic metabolites.

Methods

We performed metabolomic profiling of skeletal muscle samples from MH-negative (control) individuals and MH-susceptible patients undergoing muscle biopsy for diagnosis of MH susceptibility. Cellular metabolites were extracted from 33 fresh and 87 frozen human muscle samples using solid phase microextraction and Metabolon® untargeted biochemical profiling platforms, respectively. Ultra-performance liquid chromatography-high resolution mass spectrometry was used for metabolite identification and validation, followed by analysis of differences in metabolites between the MH-susceptible and MH-negative groups.

Results

Significant fold-change differences between the MH-susceptible and control groups in metabolites from various pathways were found (P value range: 0.009 to < 0.001). These included accumulation of long chain acylcarnitines, diacylglycerols, phosphoenolpyruvate, histidine pathway metabolites, lysophosphatidylcholine, oxidative stress markers, and phosphoinositols, as well as decreased levels of monoacylglycerols. The results from both analytical platforms were in agreement.

Conclusion

This metabolomics study indicates a shift from utilization of carbohydrates towards lipids for energy production in MH-susceptible individuals. This shift may result in inefficiency of beta-oxidation, and increased muscle protein turnover, oxidative stress, and/or lysophosphatidylcholine levels.

Keywords: Metabolomics, Malignant hyperthermia

Malignant hyperthermia (MH) is a potentially fatal pharmacogenetic disorder of skeletal muscle, triggered by exposure to certain anesthetics.1 During an MH reaction, there is a sustained elevation in myoplasmic calcium (Ca2+) that leads to a hypermetabolic state.1,2 The muscle contracture test is a highly sensitive diagnostic test for MH susceptibility3 that requires a freshly-excised muscle to measure the contracture response to caffeine and halothane. The diagnosis of MH susceptibility may also be determined by identifying a pathogenic variant in one of the three MH-related genes—RYR1, CACNA1S, and STAC31,2—whose products play key roles in the process of excitation-contraction coupling and in maintenance of Ca2+ homeostasis in skeletal muscle.

In the absence of anesthetic triggers, about 50% of MH-susceptible (MHS) individuals4 present with musculoskeletal symptoms such as myalgia, cramps, heat sensitivity, and exertional rhabdomyolysis. Metabolic changes caused by defective excitation-contraction coupling1 in MHS individuals have long-lasting effects on skeletal muscle function that have been the subject of recent research. One study hinted at the increased oxidative stress as an underlying pathomechanism.5 Furthermore, an in vivo metabolic assessment using magnetic resonance spectroscopy in MHS individuals showed an impaired aerobic metabolic pathway with reduced adenosine triphosphate (ATP) production.6 Another study using permeabilized skeletal muscle cells from MHS individuals showed mitochondrial dysfunction at baseline.7 Some older studies showed a connection between Ca2+ dyshomeostasis and impaired fatty acid metabolism in MHS individuals.8-10

Through a novel holistic approach, we aimed to identify pre-existing impairments in the metabolic pathways underlying the MHS phenotype in the absence of anesthetic triggers. We hypothesized that clinically observed differences between MHS and MH-negative (MHN) individuals are reflected in measurable differences in myoplasmic metabolites. Furthermore, by identifying the constellation of biochemical markers peculiar to the MHS phenotype, we should be able to elucidate up-regulated or down-regulated metabolic pathways involved in MH susceptibility that could help identify potential drug targets for both MH prevention and treatment.

Methods

Patients

Following institutional research ethics board approval (13-7116-BE, approved on 7 February 2014), all individuals referred to the Malignant Hyperthermia Investigation Unit at University Health Network between January 2015 and December 2018 for a caffeine halothane contracture test (CHCT)3 were approached for this study. To avoid confounding factors, only individuals with self or family history of an MH reaction (without any other known medical conditions requiring medication) were enrolled in this study. Written informed consent was obtained from all participating individuals.

Muscle sample collection

The muscle biopsy and CHCT were done according to the North American Protocol.3 The excess of biopsied specimen was utilized for metabolomics analysis. We explored two approaches for high-throughput detection and quantification of cellular metabolites: 1) solid phase microextraction (SPME)—a solvent-free, single-step, minimally invasive sampling technique allowing concomitant sample extraction and quenching of the metabolic processes in fresh muscle tissues,11-14 and 2) an untargeted biochemical profiling procedure developed by Metabolon® (Durham, NC, USA).15 Both approaches use liquid chromatography-high resolution mass spectrometry that ensures data reproducibility and accurate metabolite identification.

Muscle sample preparation

Muscle biopsy samples were transported to the laboratory in carbogenated Krebs-Ringer buffer because it was technically difficult to dissect them in the operating room. Upon arrival to the laboratory, which on average was five minutes after harvest, muscle strips, dissected from the biopsied muscle tissue and measuring approximately 25 x 5 x 5 mm (~150 mg), were either immediately flash-frozen in liquid nitrogen and shipped on dry ice to Metabolon or used as fresh muscle for sampling with SPME fibres.

Solid phase microextraction sampling and processing

Three 7-mm mix-mode coating SPME fibres were preconditioned in 1:1 methanol:water and then rinsed with water. The SPME fibres were inserted for 15 min into a fresh muscle strip immersed in Krebs-Ringer solution. After sampling was completed, the fibres were removed, washed with water, packed individually in vials and stored at −80°C until analysis. The fibres were removed from −80°C and desorbed to 200 μL 1:1 acetonitrile:water for one hour on a vortex (1,500 rounds per minute agitation). A pooled matrix sample (generated by taking a small volume of each experimental sample), extracted water samples (process blanks), and a cocktail of different quality control standards were analyzed in parallel as controls. Instrument variability was determined by calculating the percent relative standard deviation (RSD) of the instrumental quality controls, injected after every tenth sample. The overall process variability was determined by calculating the percent RSD for all endogenous metabolites present in 100% of the pooled matrix samples. The extracted samples were analyzed in positive and negative ionization modes in random order on a liquid chromotography-mass spectrometry (LC-MS) system consisting of a binary pump, an Accela autosampler, and an Orbitrap mass spectrometer Q-Exactive equipped with a heated electrospray source (Thermo Fischer Scientific, Waltham, MA, USA). Chromatographic separation was carried out as described before.16

To confirm metabolite identity, significant compounds were analyzed under parallel reaction monitoring conditions via LC-MS using a Q-Exactive Quadrupole Orbitrap (Thermo Fischer Scientific™, Waltham, MA, USA) and a Dionex Ultimate 3000 UPLC system (Dionex Corporation, Bannockburn, IL, USA).

Metabolon® sampling and processing

The frozen muscle samples were prepared for analysis using the automated MicroLab STAR® system (Hamilton Company, Reno, NV, USA). The extract was divided into four fractions: two fractions for analysis by two separate reverse phase ultra-performance LC-tandem MS (RP/UPLC-MS/MS) methods with positive and negative ion mode electrospray ionization—one fraction for analysis by hydrophilic interaction chromatography/ultra-performance liquid chromatography-tandem mass spectrometry (HILIC/UPLC-MS/MS) with negative ion mode electrospray ionization, with the other fraction reserved for backup.17 All methods utilized the Waters ACQUITY ultra-performance liquid chromatography (Waters Corporation, Milford, MA, USA) and a Thermo Scientific Q-Exactive high resolution/accurate mass spectrometer interfaced with a heated electrospray ionization source and Orbitrap mass analyzer operated at 35,000 mass resolution. Full details concerning the LC gradient and MS conditions have previously been described.18 Quality control was similar to SPME processing as described above.

Data extraction, compound identification, and statistical analysis

The SPME data were obtained in raw format using Thermo Xcalibur software (Thermo Fischer Scientific™, Waltham, MA, USA) and converted using MSconvert.19 After conversion, the data were analyzed using R-statistical package (R Foundation for Statistical Computing, Vienna, Austria)20 to remove features whose RSD in the pooled quality controls was > 30% (acceptance criterion proposed by Want et al.)21 as well as features whose signal-to-noise (S/N) ratio was < 5 compared with pooled quality controls. Further data analysis and normalization were done using MetaboAnalyst software,22 and autoscaling.23 Data clustering and potential outliers were examined using principal component analysis and partial least squares discriminant analysis (PLS-DA). Details of validation of the PLS-DA model and identification of statistically significant features/metabolites using variable importance in projection (VIP) are provided in the Electronic Supplementary Material (ESM; eTables 1 and 2). Non-parametric tests (Wilcoxon’s rank t test, unequal group variance) based on a two-sided hypothesis were applied for the identification of significant metabolites with a significant P value set at < 0.05, and false discovery rate (FDR) based on Benjamini-Hochberg adjustment set at 5%.

Identification of metabolites was done by comparing their exact masses obtained from LC-MS with the Metlin24 or human metabolome database (HMDB),25 or from the Metabolon library of authenticated standards. Two-step confirmations were performed after determining the potential metabolites. The first step was to run both standards and samples under the previous chromatographic conditions, and the obtained retention time of the feature was compared with the standard’s retention time. The fragmentation patterns of the standard and the metabolite were compared and used as second confirmation. In absence of standards (for SPME samples), the fragmentation patterns were compared with patterns reported in the Metlin database or HMDB (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Steps from sampling to analysis for Metabolon® (upper panel) and for SPME (lower panel) techniques. Picture shows a sample of muscle inserted with SPME fibres. QC = quality control; SPME = solid phase micro-extraction; (RP)UPLC MS/MS = reverse phase ultra-performance liquid chromatography-tandem mass spectrometry

For Metabolon samples, standard statistical analyses were performed on log-transformed data. Welch’s two-sample independent t test was used for a two-sided hypothesis of whether the metabolite means were different between the two study groups. To correct for false discovery of statistically significant compounds due to multiple hypothesis testing, Benjamini-Hochberg adjustment method with a FDR (q value) of < 5% was used. Fold differences were determined by dividing relative abundance of a metabolite in the MHS samples by its relative abundance in the MHN samples. Metabolites with P < 0.05 and q < 0.05 were considered statistically significant in this study. Because of lack of previous data, a statistical power calculation was not conducted for either SPME or Metabolon approach. The sample size was based on the available study data. Both methods used post hoc assignment of biochemical pathways. Methodological details for both methods have been previously described.18,26,27

Results

Based on the inclusion criteria, muscle biopsy samples from 14 MHS and 19 MHN individuals obtained in 2015 and 2017 were analyzed by SPME technology; samples from 54 MHS and 33 MHN individuals obtained between 2015 and 2018 were analyzed by Metabolon technology. Nine MHS and five MHN samples were analyzed by both methods. Data from all recruited patients were included. The details of patients’ demographics are listed in Table 1.

TABLE 1.

Characteristics of the recruited patients

Patient group Number of patients (n) Males (n) Age
mean (SD)		 Caffeine contracture, g
Median (range)		 Halothane contracture, g
Median (range)		 Genetics
Negative
(n)		 Positive (n)
Pathogenic VUS*
SPME group
 MHS 14 8 35.5 (12) 0.3 (0–3.4) 1.6 (1–10.6) 4 0 4
 MHN 19 6 38.9 (8.8) 0.0 (0–0.2) 0.3 (0–0.6) 19 0 0
Metabolon® group
 MHS 54 28 34.5(13.9) 0.7 (0–8.6) 2.8 (1–12.5) 16 18 20
 MHN 33 15 35.5(13.8) 0.0 (0–0.2) 0.4 (0–0.6) 33 0 0
Open in a new tab

g = gram contracture; MHN = malignant hyperthermia negative; MHS = malignant hyperthermia susceptible; SD = standard deviation; SPME = solid phase microextraction

*

VUS = variant of uncertain significance. RYR1 and CACNA1S genes were fully sequenced in both MHS and MHN groups

Solid phase microextraction results

For the 2015 cohort, a total of 15,945 and 4,985 features (a peak or signal that represents a chemical compound) were obtained in positive and negative ionization modes, respectively. Results from the negative ionization mode, however, were excluded because the detected features were assessed as insignificant/non-informative using MetaboAnalyst. The 2017 cohort yielded 10,973 and 5,871 features for positive and negative ionization modes, respectively. Good clustering of data were obtained for both cohorts using partial least squares discriminant analysis (PLS-DA) (Fig. 2). The PLS-DA28 models were validated by examining the results of a ten-fold cross validation and using Q2 as a performance measure whereby the 2015 cohort yielded acceptable results (R2 = 0.84, Q2 = 0.56; see ESM, eTable 1), while 2017 cohort data were found to be over-fitted (ESM, eTable 1).

Fig. 2.

Fig. 2

Open in a new tab

Partial least squares discriminant analysis (PLS-DA) plots of “2015” cohort in positive ionization mode (A), “2017” cohort in positive ionization mode and negative ionization mode (B and C). The components were selected to show the optimum separation in two-dimensional view

Additionally, we performed a permutation test (n = 2,000) with separation distance (B/W), which revealed that the PLS-DA model of the 2015 cohort gave a test statistics of 0.06, while the 2017 cohort’s PLS-DA showed P values of 0.99 and 0.14 for positive and negative ionization mode, respectively (ESM, eTable 2). Further evaluation using non-parametric Wilcoxon’s rank t test with FDR < 0.05 yielded a total of 829 significant compounds with 108 being up-regulated in MHS cases for the 2015 cohort. Tentatively identified metabolites corresponded to acylcarnitines, vitamin D metabolites, long chain fatty alcohols, fatty acids, prostaglandins, amino acids, and selected lipid species. A single metabolite (16:0 lysophosphatidylcholine) was validated, further suggesting involvement of lipid species in MH susceptibility. Evaluation of the 2017 cohort revealed a limited number of significant features when using a non-parametric analysis of variance test, so heatmaps were used to determine features of potential interest according to their VIP scores. Data representing the involved pathways and statistically significant metabolites with highest or lowest fold changes for both cohorts can be seen in Table 2. (Detailed statistical information is included in the ESM eTable 3).

TABLE 2.

Solid phase microextraction results

Pathway Metabolite(s) MHS/
MHN
RSD (%)		 P value
1-Lipid metabolism
 Fatty acid metabolism (acyl carnitines) Alpha-linoeyl carnitine 2.50
(22.9)		 0.003
Pentadecanoylcarnitine 2.38
(11.2)		 0.004
Tiglylcarnitine 5.36
(10.5)		 <0.001
 Fatty acid metabolism Octadecanol 2.58
(11.5)		 0.002
(20R,22R)-20,22-hydroxycholesterol 2.72
(10.6)		 0.001
5-tetradecenoic acid 2.17**
(18.5)		 0.001
3-methylglutaconic acid 0.76**
(9.3)		 <0.001
(±)19,20-Dihydroxy-4Z,7Z,10Z,13Z,16Z-docosapentaenoic acid 2.94
(11.3)		 <0.001
 Fatty acid amides N-palmitoylethanolamide 1.91**
(21.1)		 0.003
 Polyunsaturated fatty acids Arachidonic acid 2.75**
(11.4)		 <0.001
Linoleic acid 1.18*
(8.2)		 <0.001
Gamma-linoleic acid 1.13**
(5.9)		 <0.001
 Phospholipid Phosphatidylcholines* 2.58–6.70
(25.3)		 <0.001
Phosphatidylethanolamines* 2.50–6.71
(19.8)		 <0.001
Phosphatidylserines* 6.71
(6.5)		 <0.001
Phosphatidic acids* 2.54–6.34
(15.9)		 <0.001
Lysophosphatidylethanolamines* 2.57
(10.4)		 0.002
16:0 LysoPC 2.65
(20.36)		 0.002
Phosphatidylinositols* 2.67 – 2.80
(24)		 0.01
Phosphatidylglycerols* 2.36
(15.8)		 0.004
LPA (0:0/18:0)
PA (18:0/0:0)		 3.64
(24.3)		 <0.001
PA(P20:0/22:6(4Z,7Z,10Z,13Z,16Z,19Z)) 3.31
(16.7)		 <0.001
 Triglycerides Triglycerides** 3.80
(18.7)		 <0.001
 Mitochondrial beta-oxidation of long chain saturated fatty acids 12-[[(9Z)-1-oxo-9-hexadecen-1-yl]oxy]-octadecanoic acid
9-[[(9Z)-1-oxo-9-hexadecen-1-yl]oxy]-octadecanoic acid		 2.90
(18.5)		 0.001
 Acylglycines Pristanoyl-glycine 0.86**
(6.5)		 0.012
Pentadecanoylglycine 0.53**
(9.8)		 0.017
4-aminohippuric acid 1.44**
(8.4)		 <0.001
Hydroxyvalerylglycine 1.64**
(21.2)		 0.002
 Eicosanoids Leukotrienes* 5.12
(11.1)		 <0.001
15-(R)-15-methylprostaglandin A2 Prostaglandin A2 methyl ester 3.22
(25.3)		 <0.001
16,16-dimethyl-Prostaglandin A2 2.94
(6.2)		 0.001
1a,1b-dihomo-15-deoxy-δ-12,14-Prostaglandin D2 2.51
(10.7)		 0.003
Prostaglandins* 2.85
(18.5)		 0.001
2-Amino acid metabolism
 Histidine metabolism L-histidine 1.28**
(15.2)		 <0.001
 Aminoacyl-t-RNA biosynthesis N-methyl-proline L-lysine 1.33**
(10.4)		 0.001
4-amino-3-hydroxybutyrate L-threonine L-homoserine 0.24**
(3.6)		 0.002
 Dipeptide Asparaginyl-proline
Prolyl-asparagine		 1.20**
(28.7)		 0.006
3-Vitamin D metabolism 3-Deoxyvitamin-d3 2.03
(9.5)		 0.009
Vitamin D and derivatives* 2.43–5.12
(19.1)		 0.003
18-acetoxy-1α,25-dihydroxyvitamin D3 3.59
(17.9)		 <0.001
3-Deoxy-3-azido-25-hydroxyvitamin D3 3.42
(15.5)		 0.02
24-hydroxycalcitriol 2.54
(19.6)		 0.002
Open in a new tab

Highlights of the involved pathways with fold difference of significant metabolites

*

A large number of different hits returned for one or more m/z (mass to charge ratio) values

**

VIP score (measure of variable’s importance) obtained from a partial least squares discriminant analysis (PLS-DA) plot. MHN = malignant hyperthermia negative; MHS = malignant hyperthermia susceptible; RSD = relative standard deviation

Metabolon results

The data set includes 498 compounds of known identity—i.e., metabolites detected in skeletal muscle samples. Following log transformation and imputation of missing values, with the minimum observed value for each compound, Welch’s two-sample t test identified metabolites that differed significantly between MHS and MHN experimental groups. Based on an FDR < 0.05, a total of 94 metabolites were significantly different between MHS and MHN, of which 67 metabolites were significantly increased and 27 significantly decreased in the MHS group compared with in the MHN group.

Table 3 shows four involved pathways—lipid, amino acid, carbohydrate, and glutathione—represented by some of the validated statistically significant metabolites with the highest and the lowest fold changes between the two groups. Consistent with SPME findings, the results indicate accumulation of medium and long chain fatty acids, and increased levels of phosphatidylcholines, protein degradation products, and oxidative stress indicators.

TABLE 3.

Metabolon results: highlights of the involved pathways with fold difference of validated significant metabolites

Pathway Metabolite MHS/MHN
Mean (SD)		 P value
1-Lipid metabolism
 Fatty acid metabolism (acyl carnitine) Laurylcarnitine (C12) 3.92 (0.65) <0.001
Myristoylcarnitine (C14) 4.29 (0.31) 0.008
Palmitoylcarnitine (C16) 4.61 (0.91) 0.002
Palmitoleoylcarnitine (C16:1)* 3.34 (0.28) <0.001
Stearoylcarnitine (C18) 4.23 (0.53) <0.001
Linoleoylcarnitine (C18:2)* 4.52 (0.32) <0.001
Oleoylcarnitine (C18:1)* 3.44 (0.74) <0.001
 Polyunsaturated fatty acid Arachidonate (20:4n6) 3.98 (0.34) <0.001
Dihomo-linolenate (20:3n3 or n6)* 3.17 (0.56) 0.003
Hexadecatrienoate (16:3n3) 2.87 (0.64) 0.003
 Fatty acid synthesis Malonylcarnitine 4.38 (0.41) <0.001
 Monoacylglycerol 1-oleoyl-GPG (18:1) 0.76 (0.08) 0.004
1-arachidonoyl-GPI (20:4) 0.85 (0.02) <0.001
1-stearoyl-GPC (18:0) 0.70 (0.13) 0.001
1-palmitoyl-GPC (16:0) 0.64 (0.05) <0.001
 Diacylglycerols Stearoyl-linoleoyl-glycerol (18:0/18:2) [2] 3.21 (0.37) <0.001
Oleoyl-linoleoyl-glycerol (18:1/18:2) 3.93 (0.63) 0.002
Oleoyl-linolenoyl-glycerol (18:1/18:3) 3.06 (0.58) 0.009
Oleoyl-arachidonoyl-glycerol (18:1/20:4) 3.23 (0.92) 0.006
Palmitoleoyl-oleoyl-glycerol (16:1/18:1) [2] 4.35 (0.45) 0.002
Palmitoyl-linoleoyl-glycerol (16:0/18:2) [2] 3.89 (0.84) <0.001
 Phospholipid 1-16:0-Lysophosphatidylcholine* 4.92 (0.38) <0.001
1-palmitoyl-2-docosahexaenoyl-GPC (16:0/22:6) 3.20 (0.95) 0.004
1-palmitoyl-2-palmitoleoyl-GPC (16:0/16:1)* 2.84 (0.27) <0.001
2-Amino acid metabolism
 Histidine metabolism Histidine* 3.39 (0.59) 0.008
1-methylhistidine 5.64 (0.37) <0.001
Carnosine* 4.49 (0.63) <0.001
3-Methylhistidine 5.81 (0.79) <0.001
 Dipeptide Glycylvaline 4.27 (0.77) 0.001
Isoleucylglycine 4.89 (0.51) 0.009
Leucylglycine 4.03 (0.18) 0.002
Phenylalanylglycine* 4.25 (0.85) 0.002
Prolylglycine 4.43 (0.49) <0.001
Valylglycine 4.01 (0.29) 0.003
 N-Acetylated amino acids N-acetyl glycine 0.85 (0.01) <0.001
N-acetyl aspartate 0.74 (0.04) 0.004
N-acetyl serine 0.78 (0.09) <0.001
N-acetyl glutamate 0.70 (0.12) 0.005
N-acetyl methionine 0.82 (0.10) 0.001
3-Carbohydrate metabolism Phosphoenolpyrovate 2.08 (0.37) 0.002
Glucurante 0.71 (0.14) 0.001
Alpha ketoglutarate 0.68 (0.06) 0.002
Isocitrate 0.72 (0.12) 0.007
4-Glutathione metabolism Glutathione, oxidized(GSSG) 5.02 (0.44) 0.003
Cysteinylglycine 4.71 (0.80) <0.001
5-oxoproline 3.85 (0.41) <0.001
Ophthalmate 3.54 (0.37) <0.001
Open in a new tab
*

Indicates metabolites that were significantly increased in nine MHS patients whose muscle samples were analyzed by both solid phase microextraction and Metabolon® approaches. MHS = malignant hyperthermia susceptible

Discussion

Comparative analysis of muscle samples from MHS and MHN patients revealed significant differences in levels of metabolites involved mainly in lipid, carbohydrate, and glutathione metabolism. Both SPME and Metabolon analysis revealed a similar pattern of metabolite differences between the two patient groups. Solid phase microextraction was performed on a fresh muscle sample, yielding a large number of metabolites. Nevertheless, the validation of metabolites identified by Metabolon was more comprehensive.

One of the major findings in the MHS group was increased levels of polyunsaturated fatty acids, and reduced levels of several monoacylglycerols. Free fatty acids can be derived from lipolysis of triacylglycerols, and the rate of fatty acid β-oxidation is regulated by the availability of coenzyme A and by the carnitine shuttle. Elevated levels of free fatty acids in the MHS group may indicate that they are “primed” for ATP generation from lipid oxidation. Lower levels of monoglycerols indicate increased lipolysis, resulting in the elevated levels of polyunsaturated fatty acids. These results point towards an increased lipid utilization for energy production. The shift of energy production to lipids from carbohydrate sources is also inferred from the accumulation of phosphoenolpyruvate along with reduced levels of other intermediates such as glucuronate, alpha ketoglutarate, and isocitrate, which are indicators of an impaired Krebs cycle. The excessive accumulation of acylcarnitines can result from inefficient importation of fatty acids by “overwhelmed mitochondria” (i.e., inefficiency of lipolysis pathway due to overuse). Our findings are in agreement with previous studies in animals and with our own in vivo metabolic study.6,29-34

Significant increases in oxidized glutathione, cysteinylglycine, 5-oxoproline, and ophtalmate found in MHS muscle, consistent with previous studies, indicate impaired antioxidant utilization and increased oxidative stress associated with MH susceptibility.32,33,35 Cysteinylglycine and 5-oxoproline are glutathione catabolites generated when glutathione is recycled in the cell. Ophtalmate is a compositional derivative of glutathione, and its synthesis increases with increased oxidative stress.36

Approximately six-fold increased levels of 3-methylhistidine (an index of the rate of muscle protein breakdown), was present in the MHS compared with the MHN group. Elevated levels of histidine pathway metabolites as observed in the MHS group are suggestive of greater muscle activity and turnover.

The increased levels of diacylglycerol found in the MHS group is consistent with a recent report37 of a higher prevalence of hyperglycemia in MHS patients, as elevated diacylglycerol levels are involved in the development of insulin resistance.38 Diacylglycerols can activate TRPC3 and TRPC6 channels. These same channels were recently shown to be responsible for the increased extracellular Ca2+ influx in an MH mouse model.39 Recent evidence also showed that wide-ranging metabolic alterations in MHS individuals stimulate glycogen breakdown and may hamper glucose uptake and utilization inside muscle, thus promoting hyperglycemia in MHS individuals.40

One of the validated findings on SPME was an increase in 16:0 lysophosphatidylcholine, a product of hydrolysis of phosphatidylcholine catalyzed by the enzyme phospholipase A2 (PLA2). Increased phospholipase A2 activity results in higher levels of lysophosphatidylcholines and lysophosphatidylethanolamines,41 possibly due to increased free fatty acid release from mitochondria in MHS patients.8-10,42 Lysophosphatidylcholine is also known to be an effector of fatty acid-induced insulin resistance,43 a condition that seem to be more prevalent in MHS patients.37,40

Phospholipase A2, regulated by phosphorylation and Ca2+,44 plays a role in formation of reactive oxygen species, influencing contractile properties and fatigue characteristics of skeletal muscles.41,42,45 In particular, the phospholipase A2-dependent increase in long chain fatty acid(s) increased the release of Ca2+ from skeletal muscle mitochondria in a porcine MH model.46 Thus, increased activity of phospholipase A2 resulting in increased lysophosphatidylcholine in MHS patients may contribute to increased levels of fatty acids and increased oxidative stress, according to our findings using both approaches.

Other features pointed to several different families of lipids, with strictly returned hits for phosphoinositols. This finding corroborates results of a previous study,47 which showed higher basal levels of inositol phosphate products (over 13-fold increase) in MHS swine compared with MHN swine, confirming a correlation between increased levels of phosphoinositols and MH susceptibility.

Our study has some limitations. First, we limited our recruitment to patients with no major morbidities to minimize potential effects of unrelated conditions and their medications on metabolite profiles; however, this decision affected our sample size. Second, MH is a rare disorder, so it is challenging to obtain a very large patient cohort to minimize the confounding effects. Third, our fresh sampling technique (SPME) did not go through full validation of compounds, but results from Metabolon have been fully validated. Fourth, although muscle samples were kept in oxygenated buffer during preparation, the possibility of ischemia and its differential effects between genotypes cannot be excluded. Despite these limitations, the results of our metabolomics profiling of MH susceptibility are encouraging, providing preliminary data on metabolic pathway alterations in MHS individuals in the absence of anesthesia.

The prominence given to MH crises triggered by anesthetics has obscured the fact that MH susceptibility is a pleomorphic condition, manifested with various abnormal phenotypes in the absence of anesthetic triggers. Our holistic approach using metabolomics technology shows up- and down-regulation of various biochemical pathways in MHS patients in the absence of anesthesia, showing their underlying abnormal metabolism. Overall, our findings of faster protein turnover and increased oxidative stress in MHS may indicate fragile muscles with lower threshold to rhabdomyolysis in response to various environmental triggers. The varied degrees of impairment of the affected pathways may also account for the variable clinical spectrum of MHS—from patients with no symptoms to those complaining of fatigue, heat intolerance, muscle cramps, myalgia, and rhabdomyolysis. These abnormalities have important clinical and pharmacologic implications for the anesthetic management of MHS patients beyond the risk of MH reaction.

In summary, our metabolomics study on MHS individuals showed a high muscle protein turnover as per increased levels of histidine pathway metabolites, increased oxidative stress, and a shift from carbohydrates towards lipids for energy production, which might reduce efficiency of the beta-oxidation pathway, corroborating previous evidence of mitochondrial involvement as a secondary effect of skeletal muscle Ca2+ dyshomeostasis. Taken together, these findings reflect metabolic impairment of MHS skeletal muscle in the absence of anesthetics.

Supplementary Material

supplementary materials

Acknowledgements

The authors would like to extend special thanks to Miao Yu, PhD, postdoctoral fellow at University of Waterloo, Ontario, Canada for aid with the data processing; Nikita Looby, MSc. and Sofia Lendor, MSc, research associates at University of Waterloo, Ontario, Canada for aid in data processing, interpretation and performing the experiments; and Thermo Scientific for the access to Q-Exactive Focus Orbitrap.

Funding statement

This work was supported by the Natural Sciences and Engineering Research Council of Canada, Millipore-Sigma to Janusz Pawliszyn, and Physicians’ Services Incorporated grant to Sheila Riazi.

Footnotes

Disclosures Sheila Riazi has received a consulting fee from Norgine Pharmaceuticals.

Editorial responsibility This submission was handled by Dr. Hilary P. Grocott, Editor-in-Chief, Canadian Journal of Anesthesia.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Prior presentations This work has been presented in part at the annual European MH group meeting on 22 May 2016 in Ulm, Germany; the annual meeting of Canadian Anesthesiologists’ Society on 27 June 2016 in Vancouver, Canada; the annual meeting of American Society of Anesthesiologists on 23 October 2016 in Chicago, US; the Pittcon Conference and Expo on 27 February 2018 in Orlando, US; and Metabolomics Circle 2018 on 27 October 2018 in Torun, Poland.

Electronic supplementary material The online version of this article (https://doi.org/10.1007/s12630-020-01895-y) contains supplementary material, which is available to authorized users.

Contributor Information

Barbara Bojko, Department of Chemistry, University of Waterloo, Waterloo, ON, Canada; Department of Pharmacodynamics and Molecular Pharmacology, Faculty of Pharmacy, Collegium Medicum in Bydgoszcz, Nicolaus Copernicus University in Toruń, Bydgoszcz, Poland.

Tijana Vasiljevic, Department of Chemistry, University of Waterloo, Waterloo, ON, Canada.

Ezel Boyaci, Department of Chemistry, University of Waterloo, Waterloo, ON, Canada; Department of Chemistry, Middle East Technical University, Ankara, Turkey.

Anna Roszkowska, Department of Chemistry, University of Waterloo, Waterloo, ON, Canada; Department of Pharmaceutical Chemistry, Medical University of Gdansk, Gdansk, Poland.

Natalia Kraeva, Malignant Hyperthermia Investigation Unit, Department of Anesthesia, University Health Network, University of Toronto, 323-200 Elizabeth Street, Toronto, ON M5G 2C4, Canada.

Carlos A. Ibarra Moreno, Malignant Hyperthermia Investigation Unit, Department of Anesthesia, University Health Network, University of Toronto, 323-200 Elizabeth Street, Toronto, ON M5G 2C4, Canada.

Annabel Koivu, Malignant Hyperthermia Investigation Unit, Department of Anesthesia, University Health Network, University of Toronto, 323-200 Elizabeth Street, Toronto, ON M5G 2C4, Canada.

Marcin Wąsowicz, Malignant Hyperthermia Investigation Unit, Department of Anesthesia, University Health Network, University of Toronto, 323-200 Elizabeth Street, Toronto, ON M5G 2C4, Canada.

Amy Hanna, Department of Molecular Physiology and Biophysics, Baylor College of Medicine, Houston, Texas, USA.

Susan Hamilton, Department of Molecular Physiology and Biophysics, Baylor College of Medicine, Houston, Texas, USA.

Sheila Riazi, Malignant Hyperthermia Investigation Unit, Department of Anesthesia, University Health Network, University of Toronto, 323-200 Elizabeth Street, Toronto, ON M5G 2C4, Canada.

Janusz Pawliszyn, Department of Chemistry, University of Waterloo, Waterloo, ON, Canada.

References

  • 1.Rosenberg H, Pollock N, Schiemann A, Bulger T, Stowell K. Malignant hyperthermia: a review. Orphanet J Rare Dis 2015. 10.1186/s13023-015-0310-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Riazi S, Kraeva N, Hopkins PM. Malignant hyperthermia in the post-genomics era: new perspectives on an old concept. Anesthesiology 2018; 128: 168–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Larach MG. Standardization of the caffeine halothane muscle contracture test. Anesth Analg 1989; 69: 511–5. [PubMed] [Google Scholar]
  • 4.van Ruitenbeek E, Custers JA, Verhaak C, et al. Functional impairments, fatigue and quality of life in RYR1-related myopathies: a questionnaire study. Neuromuscul Disord 2019; 29: 30–8. [DOI] [PubMed] [Google Scholar]
  • 5.Lanner JT, Georgiou DK, Dagnino-Acosta A, et al. AICAR prevents heat-induced sudden death in RyR1 mutant mice independent of AMPK activation. Nat Med 2012; 18: 244–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Thompson SJ, Riazi S, Kraeva N, et al. Skeletal muscle metabolic dysfunction in patients with malignant hyperthermia susceptibility. Anesth Analg 2017; 125: 434–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chang L, Daly C, Miller DM, et al. Permeabilised skeletal muscle reveals mitochondrial deficiency in malignant hyperthermia-susceptible individuals. Br J Anaesth 2019; 122: 613–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Cheah KS, Cheah AM, Fletcher JE, Rosenberg H. Skeletal muscle mitochondrial respiration of malignant hyperthermia-susceptible patients. Ca2+-induced uncoupling and free fatty acids. Int J Biochem 1989; 21: 913–20. [DOI] [PubMed] [Google Scholar]
  • 9.Fletcher E, Tripolitis L, Erwin K, et al. Fatty acids modulate calcium-induced calcium release from skeletal muscle heavy sarcoplasmic reticulum fractions : implications for malignant hyperthermia. Biochem Cell Biol 1990; 68: 1195–201. [DOI] [PubMed] [Google Scholar]
  • 10.Fletcher JE, Mayerberger S, Tripolitis L, Yudkowsky M, Rosenberg H. Fatty acids markedly lower the threshold for halothane-induced calcium release from the terminal cisternae in human and porcine normal and malignant hyperthermia susceptible skeletal muscle. Life Sci 1991; 49: 1651–7. [DOI] [PubMed] [Google Scholar]
  • 11.Collins F Has the revolution arrived? Nature 2010; 464: 674–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Bojko B, Cudjoe E, Gómez-Ríos GA, et al. SPME - quo vadis? Anal Chim Acta 2012; 750: 132–51. [DOI] [PubMed] [Google Scholar]
  • 13.Bojko B, Gorynski K, Gómez-Ríos GA, et al. Solid phase microextraction fills the gap in tissue sampling protocols. Anal Chim Acta 2013; 803: 75–81. [DOI] [PubMed] [Google Scholar]
  • 14.Goryński K, Bojko B, Nowaczyk A, Bucinski A, Pawliszyn J, Kaliszan R. Quantitative structure–retention relationships models for prediction of high performance liquid chromatography retention time of small molecules: endogenous metabolites and banned compounds. Anal Chim Acta 2013; 797: 13–9. [DOI] [PubMed] [Google Scholar]
  • 15.Metabolon®. Available from URL: www.metabolon.com (accessed September 2020).
  • 16.Vuckovic D, Pawliszyn J. Systematic evaluation of solid-phase microextraction coatings for untargeted metabolomic profiling of biological fluids by liquid chromatography - mass spectrometry. Anal Chem 2011; 83: 1944–54. [DOI] [PubMed] [Google Scholar]
  • 17.Evans AM, Bridgewater BR, Liu Q, et al. High resolution mass spectrometry improves data quantity and quality as compared to unit mass resolution mass spectrometry in high-throughput profiling metabolomics. Metabolomics 2014. 10.4172/2153-0769.100132. [DOI] [Google Scholar]
  • 18.Evans AM, DeHaven CD, Barrett T, Mitchell M, Milgram E. Integrated, nontargeted ultrahigh performance liquid chromatography/electrospray ionization tandem mass spectrometry platform for the identification and relative quantification of the small-molecule complement of biological systems. Anal Chem 2009; 81: 6656–67. [DOI] [PubMed] [Google Scholar]
  • 19.Chambers MC, Maclean B, Burke R, et al. A cross-platform toolkit for mass spectrometry and proteomics. Nat Biotechnol 2012; 30: 918–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.The R Foundation. The R Project for Statistical Computing. Available from URL: http://www.r-project.org (accessed September 2020).
  • 21.Want EJ, Wilson ID, Gika H, et al. Global metabolic profiling procedures for urine using UPLC – MS. Nat Protoc 2010; 5: 1005–18. [DOI] [PubMed] [Google Scholar]
  • 22.Chong J, Soufan O, Li C, et al. MetaboAnalyst 4.0: towards more transparent and integrative metabolomics analysis. Nucleic Acids Res 2018; 46: W486–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.van den Berg RA, Hoefsloot HCJ, Westerhuis JA, Smilde AK, van der Werf MJ. Centering, scaling, and transformations: improving the biological information content of metabolomics data. BMC Genomics 2006. 10.1186/1471-2164-7-142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Guijas C, Montenegro-Burke JR, Domingo-Almenara X, et al. METLIN: a technology platform for identifying knowns and unknowns. Anal Chem 2018; 90: 3156–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Wishart DS, Feunang YD, Marcu A, et al. HMDB 4.0: the human metabolome database for 2018. Nucleic Acids Res 2018; 4: D608–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Vasiljevic T, Singh V, Pawliszyn J. Miniaturized SPME tips directly coupled to mass spectrometry for targeted determination and untargeted profiling of small samples. Talanta 2019; 199: 689–97. [DOI] [PubMed] [Google Scholar]
  • 27.Reyes-Garceś N, Diwan M, Boyaci E, et al. In vivo brain sampling using a microextraction probe reveals metabolic changes in rodents after deep brain stimulation. Anal Chem 2019; 91: 9875–84. [DOI] [PubMed] [Google Scholar]
  • 28.Westerhuis JA, Hoefsloot HC, Smit S, et al. Assessment of PLSDA cross validation. Metabolomics 2008; 4: 81–9. [Google Scholar]
  • 29.Durham WJ, Aracena-Parks P, Long C, et al. RyR1 S-nitrosylation underlies environmental heat stroke and sudden death in Y522S RyR1 knock-in mice. Cell 2008; 133: 53–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Berman MC, Harrison GG, Bull AB, Kench JE. Changes underlying halothane-induced malignant hyperpyrexia in landrace pigs. Nature 1970; 225: 653–5. [DOI] [PubMed] [Google Scholar]
  • 31.Giulivi C, Ross-Inta C, Omanska-Klusek A, et al. Basal bioenergetics abnormalities in skeletal muscle from ryanodine receptor malignant hyperthermia-susceptible R163C knock-in mice. J Biol Chem 2011; 286: 99–113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Boncompagni S, Rossi AE, Micaroni M, et al. Characterization and temporal development of cores in a mouse model of malignant hyperthermia. Proc Natl Acad Sci USA 2009; 106: 21996–2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Wei L, Salahura G, Boncompagni S, et al. Mitochondrial superoxide flashes: metabolic biomarkers of skeletal muscle activity and disease. FASEB J 2011; 25: 3068–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Fletcher JE, Rosenberg H, Beech J. Interactions of fatty acids with the calcium release channel in malignant hyperthermia. Adv Exp Med Biol 1991; 301: 57–69. [DOI] [PubMed] [Google Scholar]
  • 35.Michelucci A, De Marco A, Guarnier FA, Protasi F, Boncompagni S. Antioxidant treatment reduces formation of structural cores and improves muscle function in RYR1Y522S/WT mice. Oxid Med Cell Longev 2017. 10.1155/2017/6792694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Soga T, Baran R, Suematsu M, et al. Differential metabolomics reveals ophthalmic acid as an oxidative stress biomarker indicating hepatic glutathione consumption. J Biol Chem 2006; 281: 16768–76. [DOI] [PubMed] [Google Scholar]
  • 37.Altamirano F, Riazi S, Ibarra Moreno CA, et al. Is malignant hyperthermia associated with hyperglycaemia? Br J Anaesth 2019; 122: e3–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Moro C, Galgani JE, Luu L, et al. Influence of gender, obesity, and muscle lipase activity on intramyocellular lipids in sedentary individuals. J Clin Endocrinol Metab 2009; 94: 3440–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Lopez JR, Kaura V, Hopkins P, et al. Transient receptor potential cation channels and calcium dyshomeostasis in a mouse model relevant to malignant hyperthermia. Anesthesiology 2020; 133: 364–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Tammineni ER, Kraeva N, Figueroa L, et al. Intracellular calcium leak lowers glucose storage in human muscle, promoting hyperglycemia and diabetes. Elife 2020. 10.7554/eLife.53999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Wieland SJ, Fletcher JR, Gong Q, Rosenberg H. Effects of lipid-soluble agents on sodium channel function in normal and MH-susceptible skeletal muscle cultures. In: Blanck TJ, Wheeler DM, editors. Mechanisms of Anesthetic Action in Skeletal, Cardiac, and Smooth Muscle. NY: Plenum Press; 1991. p. 9–19. [DOI] [PubMed] [Google Scholar]
  • 42.Fletcher J, Rosenberg H, Michaux K, Tripolitis L, Lizzo FH. Triglycerides, not phospholipids, are the source of elevated free fatty-acids in muscle from patients susceptible to malignant hyperthermia. Eur J Anaesthesiol 1989; 6: 355–62. [PubMed] [Google Scholar]
  • 43.Han MS, Lim YM, Quan W, et al. Lysophosphatidylcholine as an effector of fatty acid-induced insulin resistance. J Lipid Res 2011; 52: 1234–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Qiu ZH, Gijo MA, De Carvalho MS, Spencer DM, Leslie CC. The role of calcium and phosphorylation of cytosolic phospholipase A2 in regulating arachidonic acid release in macrophages. J Biol Chem 1998; 273: 8203–11. [DOI] [PubMed] [Google Scholar]
  • 45.Yoda E, Hachisu K, Taketomi Y, et al. Mitochondrial dysfunction and reduced prostaglandin synthesis in skeletal muscle of Group VIB Ca2+ -independent phospholipase A2y deficient mice. J Lipid Res 2010; 51: 3003–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Cheah KS, Cheah AM. Skeletal muscle mitochondrial phospholipase A2 and the interaction of mitochondria and sacroplasmic reticulum in porcine malignant hyperthermia. Biochim Biophys Acta 1981; 638: 40–9. [DOI] [PubMed] [Google Scholar]
  • 47.Scholz J, Roewer N, Troll U, et al. Malignant hyperthermia and inositol phosphate metabolism in the heart and skeletal musculature (German). Anasthesiol Intensivmed Notfallmed Schmerzther 1991; 26: 450–3. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplementary materials
NIHMS1706163-supplement-supplementary_materials.pdf (217KB, pdf)

RESOURCES