Dermatomyositis (DM) and juvenile dermatomyositis (JDM) are similar yet distinct autoimmune diseases characterized by an erythematous rash and severe proximal muscle weakness. The classic diagnoses of DM and JDM are based on the findings of clinical evaluation, elevated serum levels of muscle enzymes, abnormal electromyographic results, and muscle inflammation on biopsy. More recently, magnetic resonance imaging (MRI) and P-31 magnetic resonance spectroscopy (MRS) have been employed as noninvasive techniques for the evaluation of patients with adult and juvenile DM (1–5).
MRI has proven to be a reliable indicator of the extent and severity of muscle inflammation. P-31 MRS monitors metabolic status by determining the levels of the high-energy phosphate compounds ATP and phosphocreatine (PCr), which are required for muscle contraction. These compounds are greatly reduced in the affected muscles of patients with DM and JDM (1, 4–6). During exercise and recovery, these patients demonstrate inefficient utilization and regeneration of ATP and PCr (1, 5, 6). Since MRI and MRS are noninvasive procedures with no known side effects, these techniques may be used to quantitatively evaluate the patients' responses to therapy on a longitudinal basis (7).
In the present study, we used P-31 MRS to investigate the abnormalities in magnesium levels in the diseased muscles of DM and JDM patients (8–10). Magnesium is important because it is required for all the enzymatic reactions involving ATP. Moreover, it is well established that magnesium deficiency leads to muscle weakness (11). Since magnesium levels in serum or red blood cells do not always correspond to those in tissue, it is essential to determine the levels of magnesium directly in the muscle itself (12). P-31 MRS measurements of magnesium levels in muscles are determined by the chemical shift of the β-phosphate peak of ATP when it binds to free Mg2+ and forms the enzymatically active Mg-ATP complex (13, 14). The determination of free Mg2+ levels is also important because the free cation represents the biologically active form of magnesium in vivo.
Magnesium is a required cofactor for more than 300 enzymatic reactions in muscle and is essential for mitochondrial membrane stability along with coupling in oxidative phosphorylation (15). Levels of intracellular free Mg2+ and Mg-ATP have been determined with P-31 MRS in a variety of human tissues, including red blood cells, brain, liver, and muscle (14, 16–18). Magnesium deficiency has been reported in involved tissues in a number of clinical conditions, including hypertension, arteriosclerosis, cardiac arrhythmias, diabetes mellitus, and eosinophilia-myalgia syndrome (19–21).
In this investigation, levels of free Mg2+ and Mg-ATP, which were determined in the quadriceps muscles, were markedly decreased in DM and JDM patients. The results show that treatment with prednisone and immunosuppressive drugs is associated with an increase in the levels of magnesium in muscle tissues. Furthermore, increased levels of free Mg2+ and Mg-ATP are concordant with improved strength and endurance in DM and JDM patients (8–10). Thus, free Mg2+ and Mg-ATP may play a significant role in the pathophysiology of DM and JDM.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
Patients and normal subjects. The 2 patient groups included 12 adults with DM and 10 children with JDM. The adult patients were referred by rheumatologists or dermatologists affiliated with the Vanderbilt University School of Medicine. The clinical features of the DM patients are shown in Table 1.
Table 1. Clinical and metabolic characteristics of the DM patients*
|Age/sex||Disease duration, months†||Rash||Inflammation‡||CK, IU/ml||Prednisone, mg/day§||MVC, pounds||PCr, mmoles/kg¶||ATP, mmoles/kg¶|
|All DM patients (n = 12)||47.3 ± 5.2; 10F/2M||–||–||–||672 ± 328||–||30 ± 3||15 ± 1||3.5 ± 0.2|
|Control subjects (n = 11)||41.7 ± 5.1; 10F/1M||–||0||0||30–210||–||43 ± 3||25 ± 1||5.5 ± 0.2|
The DM group included 11 Caucasian patients and 1 African American patient; 10 of the patients were women and 2 were men. At the time of their first MRS examination, the patients ranged in age from 22 years to 75 years; their mean ± SEM age was 47.3 ± 5.2 years, which was not significantly different from that of the adult control subjects (41.7 ± 5.1 years). Disease duration prior to the initial P-31 MRS examination was determined from the patients' clinical histories.
Each DM patient had a characteristic photosensitive, erythematous rash. Nine patients underwent muscle biopsy, the results of which were consistent with the diagnosis of DM in all 9 patients. Ten of the 12 patients had inflammation in the quadriceps muscles, as demonstrated by high signal intensities (brightness) on T2-weighted MR images. The presence of inflammation was further verified by calculation of T1 and T2 relaxation times and then scored on a scale of 0–3, where 0 = no inflammation, 1+ = mild, 2+ = moderate, and 3+ = severe (1, 6). Three patients had elevated serum creatine kinase (CK) levels. Evaluation of disease activity by serum CK activity has limited value, however, since some patients never show an elevation of this enzyme and low CK levels may be associated with a poor prognosis (7, 22).
Prednisone dosages at the time of the first P-31 MRS examination are also listed in Table 1. In patients 8–10, prednisone was tapered from high doses, resulting in the reappearance of symptoms when the drug was discontinued. Four patients received adjunct therapy: patient 3 received dapsone 200 mg/day, patient 4 received azathioprine 100 mg/day, patient 6 received azathioprine 50 mg/day and methotrexate 15 mg/week, and patient 7 received methotrexate 7.5 mg/week. The dosages and time courses of the therapeutic regimens for the DM patients were highly variable.
All DM patients reported fatigue and weakness, as demonstrated by the maximum voluntary contraction (MVC) of the quadriceps muscle, which was on average, 30% lower than the control value (P < 0.006) as reported previously (6). PCr and ATP, the 2 high-energy compounds required for muscle contraction, were reduced by ∼40% in the quadriceps muscles of DM patients at rest (P < 0.0001). The DM metabolic profile represents a substantially defective bioenergetic status, as detailed previously (6).
The juvenile dermatomyositis (JDM) group consisted of 10 Caucasian children (6 girls and 4 boys), most of whom were referred by pediatric rheumatologists at Vanderbilt University Medical School (Table 2). One patient was referred by Drs. Gloria C. Higgins and Linda K. Myers (Department of Pediatrics, University of Tennessee Medical School, Memphis) and was the subject of a previous report (23). The JDM patients ranged in age from 5 years to 16 years with a mean ± SEM age of 10.5 ± 1.3 years, which was not significantly different from that of the juvenile control subjects (11.3 ± 1.3 years). All patients were evaluated clinically by their referring pediatricians prior to the initial MRI and P-31 MRS examinations. Disease duration prior to the initial MRS examination was determined from the patients' clinical histories.
Table 2. Clinical and metabolic characteristics of the JDM patients*
| ||Age/sex||Disease duration, months†||Rash||Inflammation‡||CK, IU/ml||Prednisone§||MVC, pounds||PCr, mmoles/kg¶||ATP, mmoles/kg¶|
| 1||11/M||17||+||3+||145||40 mg qod||3 months||12||10||2.8|
| 2||15/F||1||+||3+||5,769||60 mg/day||14 days||20||12||3.6|
| 3||5/F||22||+||1+||68||15 mg qod||11 months||7||20||5.7|
| 4||16/F||6||+||2+||22||40 mg/day||6 months||30||21||4.5|
| 5||13/M||7||+||1+||46||60 mg/day||4 days||20||27||6.4|
| 7||5/F||3||+||2+||2,051||40 mg/day||6 days||8||18||4.0|
| 8||9/F||0.5||+||2+||5,545||60 mg/day||3 days||15||10||2.4|
| 9||11/F||3||+||1+||160||40 mg/day||10 days||15||18||4.3|
| 10||6/M||4||+||2+||79||30 mg/day||21 days||10||12||3.1|
|All JDM patients (n = 10)||10.5 ± 1.3; 6F/4M||–||–||–||1,858 ± 787||–||–||15 ± 2||16 ± 2||4.1 ± 0.4|
|Control subjects (n = 6)||11.3 ± 1.3; 2F/4M||–||0||0||30–210||–||–||32 ± 5||26 ± 1||6.6 ± 0.3|
All children in the JDM group had typical erythematous rash. Inflammation was observed in the quadriceps muscles of all 10 patients, as determined by the MRI criteria used for adult DM patients. The serum creatine kinase (CK) values were substantially elevated in 4 patients (patients 2, 6, 7, and 8), but the remaining 6 patients had CK values that remained in the normal range on repeated testing. Weakness and fatigue were observed in all JDM patients.
The prednisone dosages at the time of the first examination and the duration of treatment are also shown in Table 2. Patient 6 received no prednisone prior to the MR examinations but was started on 40 mg/day immediately thereafter. Six patients had been treated for ≤21 days, and 3 patients received prednisone for ≥3 months. In addition, patient 4 took 7.5 mg of methotrexate per week, patient 8 had 2 courses of methylprednisolone (1,000 mg/course), and patient 9 received 2 courses of intravenous immunoglobulin.
Muscle strength, measured as the MVC of the quadriceps muscle, was ∼53% below normal in the JDM patients (P < 0.003). The mean PCr level at rest was 38% below the value for normal children (P < 0.002), and ATP levels were reduced by 38% (P < 0.0004), indicating significant biochemical deficits (5).
The adult control group consisted of 11 subjects (10 women and 1 man), all of whom were healthy and nonobese. Their ages ranged from 21 years to 80 years (mean ± SEM 41.7 ± 5.1 years). The juvenile control group consisted of 6 healthy, nonobese children (2 girls and 4 boys), ranging in age from 7 years to 16 years (mean ± SEM 11.3 ± 1.3 years). MRI and MRS examinations were performed over the same time periods in the control subjects as in the patients.
These studies were approved by the Committee for the Protection of Human Subjects of the Vanderbilt Institutional Review Board.
Characterization of longitudinal improvements in DM and JDM patients during therapy. Six DM patients (patients 1–6 in Table 1) and 6 JDM patients (patients 1–6 in Table 2) were evaluated longitudinally. Tables 3 and 4 show the findings from 2 visits for these 2 groups of DM and JDM patients, respectively. The first visit was during a time when the patient's status was severe; the second visit was during a time when the clinical status was improved. As shown in the tables, each of the 12 patients demonstrated improvements in at least 4 of 6 clinical or metabolic parameters: prednisone dosage requirements, inflammation, MVC, PCr levels, ATP levels, and work:energy cost ratios. For the 6 DM or JDM patients with improved clinical status, the mean levels of PCr and ATP, as well as the work:energy cost ratios, showed statistically significant increases, as determined by Student's paired 2-tailed t-test.
Table 3. Characterization of improvement during prednisone treatment in DM patients
| ||MRS interval, months*||Prednisone (adjunct therapy)†||Inflammation‡||MVC, pounds||PCr, mmoles/kg§||ATP, mmoles/kg§||Work:energy cost ratio¶|
| 1||17||8 mg/day||None||3+||1+||15||30||15||17||2.7||3.6||11||27|
| 2||13||60 mg/day||None||1+||1+||40||40||12||20||3.2||4.3||27||31|
| 3||8||10 mg/day; (Dap. 200 mg/day)||2.5 mg/day; (Dap. 50 mg/day)||1+||0||25||60||19||24||3.9||5.0||29||31|
| 4||42||30 mg/day; (AZA 100 mg/day)||None; (AZA 50 mg/day)||3+||1+||30||60||13||23||4.0||5.3||25||64|
| 5||35||20 mg/day||None; (HCQ 400 mg/day)||2+||0||40||30||10||21||2.9||5.0||18||54|
| 6||3||15 mg/day; (MTX 15 mg/week; AZA 50 mg/day)||None; (MTX 10 mg/week)||1+||0||40||40||16||27||3.8||5.0||23||39|
|Mean ± SEM||–||–||–||–||–||32 ± 4||43 ± 6||14 ± 1||22 ± 1||3.4 ± 0.2||4.7 ± 0.3||22± 3||41 ± 6|
Table 4. Characterization of improvement during prednisone treatment in JDM patients
| ||MRS interval, months*||Prednisone (adjunct therapy)†||Inflammation‡||MVC, pounds||PCr, mmoles/kg§||ATP, mmoles/kg§||Work:energy cost ratio¶|
| 1||48||40 mg qod||30 mg 4 days/week; (MTX 12.5 mg/week)||3+||0||12||40||10||12||2.8||3.4||5||27|
| 2||25||60 mg/day||2.5/25 mg qod; (MTX 15 mg/week)||3+||0||20||40||12||19||3.6||3.9||24||40|
| 3||29||10 mg/day; (MTX 12.5 mg/week)||None; (IVIG)||0||0||10||20||14||17||4.0||4.4||11||39|
| 4||20||40 mg/day; (MTX 7.5 mg/week)||40 mg/day; (MTX 25 mg/week)||2+||1+||30||30||21||25||4.5||5.2||27||32|
| 5||9||30 mg qod||20 mg qod||0||0||25||30||21||29||5.1||6.5||43||46|
| 6||5||60 mg/day||35/30 qd||2+||0||15||30||14||27||3.7||6.2||12||47|
|Mean ± SEM||–||–||–||–||–||19 ± 3||32 ± 3||15 ± 2||22 ± 3||4.0 ± 0.3||4.9 ± 0.5||20 ± 6||38 ± 3|
Exercise protocol. Prior to the MRS examination, each patient was evaluated for muscle strength as previously described (1, 5, 6). The MVC of the quadriceps muscle was determined with the patient in the sitting position. Patients performed full leg extensions with application of increasingly heavy ankle weights. The patient was then placed in the supine position with the thigh flexed over a supporting wooden arch. The P-31 MRS surface coil was taped firmly to the thigh, and Velcro straps were used to stabilize the upper leg in the center of the magnet. The lower leg was free and unrestricted, allowing for full extension of the lower leg for exercise and a prompt return to the resting position.
The P-31 MRS spectra of the quadriceps muscle were acquired first during 6 minutes of rest. Immediately thereafter, a nonmagnetic weight equivalent to 25% of the MVC value was secured to the ankle, and the subject extended his or her leg by contracting the quadriceps muscle once every 5 seconds for 6 minutes (1, 5, 6). Subsequently, a weight equivalent to 50% of the MVC value was attached to the ankle, and the lifting exercise was repeated during the next 6 minutes. Finally, the subject rested for a recovery period of at least 6 minutes. The same protocol was employed for the adult and juvenile patients and control subjects.
P-31 MR spectroscopy. P-31 spectra were acquired and metabolite levels were calculated as described previously (1, 5–7) (Figure 1). The magnetic field was adjusted for maximum proton resolution (∼0.1 parts per million [ppm]). Spectra of the P-31 metabolites were acquired with a Siemens surface coil (8 cm diameter) at 25.7 MHz with a rectangular pulse of 500 μsec duration, a TR of 3 seconds, and a resolution of 1.95 Hz. Spectra were acquired during each minute of the 6-minute rest period, during both of 6-minute exercise intervals, and again at rest during the recovery period following exercise. In order to collect data during periods of steady-state kinetics, only the data from the last 4 minutes of each 6-minute exercise interval were summed for calculations of P-31 metabolite concentrations (Figure 1).
Figure 1. Comparison of P-31 magnetic resonance spectroscopy spectra of the quadriceps muscles of a normal juvenile control subject and a juvenile dermatomyositis (JDM) patient during rest. The positions of the inorganic phosphate (Pi) and phosphocreatine (PCr) peaks, along with the α-, β-, and γ-phosphate peaks of ATP are shown. The spectrum for the patient with JDM demonstrates significantly reduced levels of PCr and ATP.
Download figure to PowerPoint
The phasing of the spectra was performed with zero-order and first-order corrections of the Fourier-transformed free induction decay (FID), and baseline corrections were made using the parabolic method (1, 5–7). Resonance areas of inorganic phosphate (Pi), PCr, and ATP were corrected for saturation effects determined in the resting quadriceps muscles of healthy subjects. Relative concentrations of P-31 metabolites were calculated from the corrected resonance areas and a proportionality factor, which assumes that the area of the β-phosphorus peak of ATP for the control subjects at rest is equivalent to 5.5 mmoles of ATP per kg wet weight of muscle, as determined from biopsy (24). The same proportionality factor was then used for DM and JDM patients to calculate relative levels of P-31 metabolites from corrected resonance areas.
Further quality assurance was obtained by evaluation of a Siemens P-31 standard before each MRS examination. The variability in the resonance area of the standard over 1 year was ±5%. Two control subjects who were tested frequently over a 10-month period had stable metabolite levels while at rest, with a variation of ±6%. Intracellular pH was calculated from the chemical shift of the Pi peak relative to PCr, as measured in parts per million.
Determination of the levels of free Mg2+ and Mg-ATP complex. The method of Gupta et al (14) and the parameters of Mosher et al (25) were used to determine the concentrations of free Mg2+, free ATP, and Mg-ATP complex (Figure 2). Free Mg2+ was determined from the measured distance between the α-phosphate and β-phosphate peaks of ATP in the muscle spectrum. Free [Mg2+] was calculated directly, as follows:
KDMg-ATP, the dissociation constant for the Mg-ATP complex, can be corrected for pH (14). At pH 7.10, KDMg-ATP = 53 μM; δαβMg-ATP = 8.255 ppm, the distance between the α- and β-phosphate peaks for the Mg-ATP complex; and δαβATPfree = 10.832 ppm, the distance between the α- and β-phosphate peaks for free ATP (25, 26). δαβobs represents the experimentally observed distance between the α-ATP and β-ATP peaks in the muscles of the subjects. [ATPfree] and [ATPtotal] are the concentrations of free and total ATP, respectively. [ATPtotal], which was determined by P-31 MRS measurements, represents the sum of [ATPfree] and [Mg-ATP] (1, 5, 6). The derivation of the formulas for these determinations was previously detailed by Clauw and associates (18).
Figure 2. Chemical shift of the β-ATP peak in the spectrum for the quadriceps muscles of a patient with juvenile dermatomyositis (JDM). The β-ATP peak in the spectrum for the juvenile control subject exhibits a symmetrical triplicate shape, indicative of Mg-ATP complex. In the JDM patient's spectrum, the β-ATP peak demonstrates the effect of reduced magnesium binding, in that the peak is less symmetrical and shifted to the right. The distance (δ) from the half-height midpoint of the α-peak to the quarter-height midpoint of the β-peak was measured and utilized for the calculations described in Patients and Methods.
Download figure to PowerPoint
Statistical analysis. Statistical comparisons of data for patients and normal control subjects were performed using Student's 2-tailed t-test. In the longitudinal studies, individual patients were evaluated during periods of severe and improved disease status. Comparisons of patients during these 2 disease states were made using Student's paired 2-tailed t-test. P values less than 0.05 were considered statistically significant. The Microsoft Excel 97 program was used for regression analyses, which yielded r2 values, coefficients of determination, and P values. Statistical correlations of the data were considered significant for r2 values greater than 0.5 and for P values less than 0.05.
- Top of page
- PATIENTS AND METHODS
Abnormalities in magnesium and ATP levels in the muscles of DM patients during rest and exercise. In the resting period, levels of intracellular free Mg2+ in the quadriceps muscles of DM patients were not significantly different from those of the adult controls (Table 5). With exercise, free Mg2+ in the muscles of the controls showed a modest increase of 17%. In contrast, free Mg2+ in DM muscles decreased to less than half of the resting value. Thus, DM muscles had ∼40% lower levels of free Mg2+ than did control muscles during both phases of exercise (P < 0.029). In the recovery phase following exercise, free Mg2+ levels for patients remained significantly lower than control values (P < 0.012) (data not shown).
Table 5. Abnormalities in magnesium and ATP concentrations in the muscles of DM patients*
| Adult controls (n = 11)||0.83 ± 0.09||5.1 ± 0.1||5.9 ± 0.1||5.5 ± 0.2||0.37 ± 0.04||6.6 ± 0.6|
| DM patients (n = 12)||1.17 ± 0.23||3.2 ± 0.2||4.4 ± 0.4||3.5 ± 0.2||0.25 ± 0.06||7.0 ± 1.4|
|Exercise: 25% MVC|
| Adult controls (n = 11)||0.90 ± 0.13||5.1 ± 0.2||6.0 ± 0.2||5.5 ± 0.2||0.36 ± 0.05||6.4 ± 0.7|
| DM patients (n = 12)||0.54 ± 0.10||2.7 ± 0.2||3.3 ± 0.2||3.1 ± 0.2||0.36 ± 0.06||11.3 ± 1.5|
|Exercise: 50% MVC|
| Adult controls (n = 11)||0.96 ± 0.12||5.0 ± 0.2||5.9 ± 0.2||5.3 ± 0.2||0.31 ± 0.04||5.8 ± 0.6|
| DM patients (n = 12)||0.54 ± 0.13||2.9 ± 0.2||3.4 ± 0.2||3.3 ± 0.2||0.40 ± 0.07||11.9 ± 1.6|
The dissociation constant (KD) for Mg-ATP varies according to the intramuscular pH (14). However, each of the subjects exhibited only minimal variations in pH throughout rest and exercise. Moreover, the patient and control groups had similar average pH values (range 7.04–7.07). Therefore, a standardized KD of 53 μM (for a pH of 7.1) was employed for the Mg2+ calculations (26). Using this KD, mean levels of free Mg2+ in all subjects were essentially identical to the results obtained with pH corrections for each subject individually.
Levels of Mg-ATP, the enzymatically active form of ATP, were 37% lower in DM muscles than in control muscles during rest (P < 0.0001) (Table 5). During the first phase of exercise, DM patients lost 15% of their Mg-ATP (P < 0.0003), which was then 46% below the corresponding control value (P < 0.0001). The Mg-ATP level remained significantly lower (46%) during the recovery phase of the protocol.
Total magnesium, which is the sum of the free Mg2+ and the Mg-ATP levels, was 26% lower in DM muscles than in normal muscles during rest (P < 0.002) (Table 5). However, with the combined decreases in free Mg2+ and Mg-ATP during exercise, DM patients lost about one-fourth of their total magnesium (P < 0.001), and their mean total magnesium level was then 45% below normal values (P < 0.0001). These differences were also sustained through the final recovery period.
Total intramuscular ATP, as determined with P-31 MRS, was 37% less in the resting muscles of DM patients as compared with the controls (P < 0.0001) (6). DM patients exhibited an exercise-induced loss of total ATP (11%) (P < 0.002), resulting in an ATP level that was 46% below normal (P < 0.0001) and remained low throughout the MRS protocol.
The absolute levels of free ATP, determined as the total ATP minus the Mg-ATP values, were not significantly different between DM patients and controls during rest, exercise, or recovery. However, the percentage of free ATP, which represents the fraction of ATP that is enzymatically inactive, revealed important differences. The percentage of free ATP in DM muscles (7.0%) was similar to that of adult controls (6.6%). While this level remained constant in the controls, it rose to abnormally high levels (11–12%) in DM patients during exercise and was statistically significantly different from that of the controls (P < 0.009). The percentage of free ATP in DM muscles remained significantly higher during recovery (10%) (P < 0.016).
Using regression analyses, none of the metabolites, including Mgfree, Mg-ATP, Mgtotal, ATPfree, or % ATPfree, showed any statistically significant correlation with age, disease duration, prednisone dosage, or CK level in the DM patients.
Deficiencies in magnesium and ATP levels in the muscles of JDM patients during rest and exercise. Free Mg2+ in the muscles of the juvenile control subjects rose steadily during exercise, reaching a level that was 31% higher than at rest (P < 0.036) (Table 6). In contrast, free Mg2+ was normal in the resting muscles of JDM patients but decreased by 38% at the end of exercise. During the second phase of exercise, JDM muscles had 44% lower free Mg2+ than did the juvenile control muscles (P < 0.005). Free Mg2+ levels remained abnormally low during the recovery period (P < 0.006) (data not shown).
Table 6. Abnormalities in magnesium and ATP concentrations in the muscles of JDM patients*
| Juvenile controls (n = 6)||0.78 ± 0.06||6.2 ± 0.2||6.9 ± 0.3||6.6 ± 0.3||0.43 ± 0.04||6.5 ± 0.5|
| JDM patients (n = 10)||0.93 ± 0.17||3.8 ± 0.4||4.7 ± 0.4||4.1 ± 0.4||0.28 ± 0.05||6.9 ± 1.1|
|Exercise: 25% MVC|
| Juvenile controls (n = 6)||0.88 ± 0.03||5.9 ± 0.2||6.7 ± 0.2||6.2 ± 0.2||0.36 ± 0.01||5.7 ± 0.2|
| JDM patients (n = 10)||0.63 ± 0.05||3.6 ± 0.3||4.2 ± 0.3||3.9 ± 0.4||0.34 ± 0.06||8.3 ± 0.7|
|Exercise: 50% MVC|
| Juvenile controls (n = 6)||1.02 ± 0.08||5.9 ± 0.2||7.0 ± 0.2||6.3 ± 0.2||0.32 ± 0.03||5.0 ± 0.3|
| JDM patients (n = 10)||0.57 ± 0.09||3.7 ± 0.3||4.2 ± 0.3||4.1 ± 0.3||0.40 ± 0.06||9.9 ± 1.2|
During rest and exercise, the average pH was essentially unchanged, ranging from 7.05 to 7.08, with no significant differences between the JDM patients and the normal children. As with the adult subjects, a KD of 53 μM was used (26), and the resultant mean values were no different from those obtained with individual pH corrections for each subject.
In the diseased JDM muscles, Mg-ATP, total magnesium, and total ATP levels were between 32% and 39% below the juvenile control values during rest (P < 0.0005, P < 0.002, and P < 0.0004, respectively). In contrast to the adult patients, exercise did not result in significant changes in these compounds in JDM muscles, and the 3 metabolites remained lower than those of the juvenile controls during exercise and recovery (range of P values <0.0001 to <0.0008).
For JDM patients, the percentage of free ATP (6.9%) was similar to that in the juvenile controls (6.5%) during rest, but in JDM muscles, the % ATPfree rose to higher levels (8.3% and 9.9%) during exercise and became statistically significantly different from the corresponding control values (P < 0.013 and P < 0.008, respectively). The percentage of free ATP remained significantly higher (9.1%) during recovery (P < 0.011). These defects in magnesium and ATP levels are concordant with the muscle weakness observed in DM and JDM patients.
With regression analyses, none of these Mg or ATP metabolites (Mgfree, Mg-ATP, Mgtotal, ATPfree, or % ATPfree) showed any statistically significant correlation with age, disease duration, prednisone dosage, or CK level in the JDM patients.
Stratification of the adult and juvenile controls and the two patients groups. Each of the 4 groups of subjects exhibited levels of Mg-ATP that distinctly differentiated them from the 3 other groups (Figure 3). The muscles of juvenile controls demonstrated the highest levels of Mg-ATP during rest, exercise, and recovery. Adult controls showed Mg-ATP levels that were ∼13–17% less than those of the juvenile controls at all stages of the protocol (range of P values <0.001 to <0.016). The JDM patients had levels that were 26–35% lower than those of the adult controls (range of P values <0.0001 to <0.003). The DM patients had the lowest Mg-ATP concentrations, which were on average 14–24% lower than those of the JDM patients, and these differences were statistically significant only during exercise (P < 0.036 and P < 0.022 for the 2 exercise levels).
Figure 3. Differences in Mg-ATP levels in the quadriceps muscles of control subjects and dermatomyositis (DM) patients during rest, exercise, and recovery. Values are the mean ± SEM mmoles/kg wet weight of muscle. The muscles of the normal juvenile controls had higher levels of Mg-ATP than those of the normal adult controls throughout the exercise protocol (∗ = range of P values from <0.001 to <0.016). The muscles of juvenile DM (JDM) patients showed lower values of Mg-ATP than those of normal adult controls at all phases of the protocol (∗∗ = range of P values from <0.0001 to <0.003). Muscles of the DM patients exhibited the lowest concentrations of Mg-ATP, which were statistically significantly different from those of the JDM patients only during exercise († = P < 0.036 and P < 0.022 for 25% and 50% of the maximum voluntary contraction [MVC], respectively).
Download figure to PowerPoint
Free Mg2+ provided a somewhat different profile than that observed with Mg-ATP. The 4 groups of subjects were not statistically different at rest (Figure 4). During exercise, however, the 2 patient groups demonstrated significantly lower levels of free Mg2+ than their respective control groups (range of P values <0.005 to <0.037). These decreased levels persisted during recovery (range of P values <0.012 and <0.006 for DM and JDM patients, respectively). Of note, there were no significant differences in free Mg2+ levels between the DM and the JDM groups or between the adult and the juvenile controls.
Figure 4. Comparisons of free Mg2+ levels in the quadriceps muscles of control subjects and dermatomyositis (DM) patients during rest, exercise, and recovery. During rest, there were no statistically significant differences between any of the 4 groups of subjects. During exercise and recovery, juvenile DM (JDM) patients had lower free Mg2+ in the quadriceps muscles than did the juvenile control subjects (∗ = range of P values from <0.005 to <0.006). Adult DM patients also had less free Mg2+ compared with the adult control subjects (∗∗ = range of P values from <0.012 to <0.037). The adult and juvenile controls, as well as the adult and juvenile DM patients, showed no significant differences during any phase of the exercise protocol. MVC = maximum voluntary contraction.
Download figure to PowerPoint
Initially, the percentage of free ATP ranged between 6% and 7% in the 4 groups of subjects (Figure 5). With exercise, the percentage of free ATP rose for both the JDM and DM groups, so that the patients then had significantly greater percentages of free ATP than their respective control groups (P < 0.013). During the second exercise phase and recovery, the percentages for both patient groups remained significantly higher than the those for the control subjects (P < 0.016). There were no statistically significant differences between the 2 patient groups or between the control groups.
Figure 5. Differences in the percentage of free ATP relative to total ATP in the quadriceps muscles of control subjects and dermatomyositis (DM) patients during rest, exercise, and recovery. During rest, there were no statistically significant differences among the 4 groups of subjects. For the controls, the percentage of free ATP remained between 5% and 6% during the entire exercise protocol. However, for the juvenile DM (JDM) patients during exercise and recovery, a significantly higher percentage of the total ATP was unbound, as compared with that in juvenile controls (∗ = range of P values <0.008 to <0.013). Adult DM patients also had an abnormally high percentage of free ATP compared with that in adult controls (∗ = range of P values <0.002 to <0.016). MVC = maximum voluntary contraction.
Download figure to PowerPoint
Beneficial effects of prednisone and immunosuppressive therapy on Mg-ATP and free Mg2+ levels in the recovery period. Beneficial effects of therapy on Mg-ATP and free Mg2+ concentrations were observed during rest, exercise, and recovery. Presented here are the data for the effects during the recovery phase, which represents a period when oxidative phosphorylation via the mitochondria is predominant (27, 28).
The severely weak DM and JDM patients exhibited significant increases in Mg-ATP levels during the course of prednisone treatment (Figure 6). DM patient 1 showed only a modest increase in Mg-ATP (Figure 6A), but DM patients 2–6 showed greater increases, ranging from 44% to 171%. The mean Mg-ATP level in the DM patients' muscles increased substantially, from a mean ± SEM of 2.5 ± 0.2 to 4.3 ± 0.2 mmoles/kg wet weight of muscle (P < 0.003) (Figure 6B). The mean Mg-ATP level was initially 51% below the adult control values (P < 0.0001), but after treatment, it was only 17% below normal (P < 0.007).
Figure 6. Improvements in Mg-ATP levels during the course of prednisone and immunosuppressive therapy. Dermatomyositis (DM) and juvenile DM (JDM) patients were evaluated during periods of severe weakness and improvement. A, Over the course of time, all patients in both groups demonstrated substantial improvements in the intramuscular levels of Mg-ATP complex, as measured during the recovery period following exercise (∗ = P < 0.003 for the DM patients; ∗∗ = P < 0.035 for the JDM patients). B, Severely weak DM patients improved from a level of Mg-ATP that was 51% below that in the normal adult controls († = P < 0.0001) to a level that was 17% below normal (‡ = P < 0.007). Similarly, JDM patients improved from a level that was 44% below that in the normal juvenile controls (§ = P < 0.0001) to a level that was 23% below normal (¶ = P < 0.022).
Download figure to PowerPoint
JDM patients also showed a substantial individual variability in response to therapeutic regimens. Although JDM patient 3 showed a minimal change in Mg-ATP, the remaining 5 patients showed improvements ranging from 20% to 110%. During the course of treatment, mean Mg-ATP levels in JDM muscles rose from a mean ± SEM of 3.4 ± 0.3 to 4.6 ± 0.5 mmoles/kg wet weight of muscle (P < 0.035) (Figure 6B). Initially, the average Mg-ATP level for the patients was 44% below the normal mean of 6.0 mmoles/kg (P < 0.0001), and after therapy, it was only 23% below normal control levels (P < 0.022).
Free Mg2+ was not improved over the course of treatment in 2 DM patients (patients 1 and 6), while 2 patients (patients 2 and 3) improved modestly, and two others (patients 4 and 5) reached a level above or within 1 SEM of adult control values (Figures 7A and B). For the 6 DM patients, the mean free Mg2+ increased from 0.48 ± 0.07 to 0.73 ± 0.14 mmoles/kg during the transition from severe to improved disease status (Figure 7B). This represented a level that was initially 52% below normal in the severe disease state (P < 0.024) and only 27% below normal in the improved clinical state (P not significant).
Figure 7. Prednisone and immunosuppressive therapy result in elevation of free Mg2+ levels, as measured during the recovery period following exercise. A, Two patients (patients 1 and 6) in the dermatomyositis (DM) group did not show improvements in free Mg2+ levels; the remaining 4 patients demonstrated modest to substantial increases. Following treatment, all 6 juvenile DM (JDM) patients demonstrated increased free Mg2+ levels (∗ = P < 0.041). B, As a group, the weak DM patients initially showed free Mg2+ levels in muscle that were below those in normal adult controls (∗∗ = P < 0.024). With treatment, these patients improved to a level that was not significantly different from that of the adult controls. Severely weak JDM patients had free Mg2+ concentrations that were initially below those in normal juvenile controls († = P < 0.013). During therapy, these values increased to a mean level above that of the normal juvenile controls (P not significant).
Download figure to PowerPoint
The JDM patients demonstrated greater improvement in free Mg2+ during therapy than did the DM patients (Figure 7). Over the course of treatment, all 6 JDM patients showed increased levels of free Mg2+, with values as high or higher than the mean free Mg2+ level for the juvenile controls (Figures 7A and B). The average free Mg2+ level in JDM muscles increased from 0.58 ± 0.08 to 1.21 ± 0.17 mmoles/kg (P < 0.041) (Figure 7B). In the severe disease state, the mean JDM value was 36% below the normal value of 0.90 mmoles/kg (P < 0.013), and with improved disease status, the patients had a mean concentration that was higher than the level in the juvenile controls, although this difference was not statistically significant.
- Top of page
- PATIENTS AND METHODS
Intracellular magnesium is largely bound to nucleic acids, proteins, and phosphorylated metabolites (29). Therefore, free magnesium concentrations are low in most cell types. In skeletal muscle, free Mg2+ has been estimated to be between 0.5 and 1.5 mmoles/kg using various methods including P-31 MRS (16, 18, 30–32). In the present investigation, free Mg2+ levels in the muscles of normal control subjects (mean ± SEM 0.83 ± 0.09 mmoles/kg for adults and 0.78 ± 0.06 mmoles/kg for children) were within the reported range during rest (Tables 5 and 6). The levels of free Mg2+ increased modestly (17%) for healthy adult subjects during exercise and recovery. However, with juvenile controls, free Mg2+ showed a statistically significant increase of 31% during exercise and recovery. Roth and Weiner (33) predicted a 35% increase in free Mg2+ at the end of exercise based on calculated chemical equilibria and pH. In contrast, free Mg2+ levels in DM and JDM muscles were equivalent to those of the controls during rest, but showed a significant decrease (40%) in free Mg2+ to 0.55 mmoles/kg during exercise. These low levels of free Mg2+ may be related to weakness and fatigue in these patients, since this is the biologically active cationic form (15).
Adequate intracellular magnesium has several important roles in maintaining normal function and homeostasis in muscle. First, it is required for the optimization of enzymatic reactions, the maintenance of membrane integrity, and the stabilization of mitochondria (15, 26, 34). Second, free Mg2+ is important for the regulation of ATP-dependent ion channels for K+, Na+, and Ca+ transport (34). For example, free Mg2+ is responsible for maintaining a low level of calcium release from the sarcoplasmic reticulum in skeletal and smooth muscle. Resnick et al (35) demonstrated that low levels of Mg2+ resulted in the unchecked release of calcium in smooth muscle cells, causing fiber contraction. Hypercontractility of smooth muscle may, in turn, result in arteriolar vasoconstriction and, ultimately, hypoperfusion of oxygen and substrates required for muscle contraction. Thus, function and performance in diseased muscles can be seriously affected by magnesium deficiencies through numerous mechanisms.
High levels of total ATP and Mg-ATP are essential for optimal muscle function. In this study, levels of total ATP in normal adults remained constant during rest and exercise, with ∼93% of ATP complexed to magnesium (Mg-ATP). Resting DM patients had about 37% less total ATP or Mg-ATP than did controls, indicating serious abnormalities (Table 5). These deficits in ATP and Mg-ATP became more pronounced during exercise, due to significant hydrolysis of ATP in the muscles of some of the patients.
As previously shown, ATP is completely hydrolyzed under these conditions to adenine and adenosine, which subsequently exit from the muscle cells (1, 6, 36, 37). ATP loss during moderate exercise results in fatigue and prolonged recovery, since it may take as long as several hours to resynthesize and restore intracellular ATP levels. In contrast, the JDM group exhibited no significant exercise-induced losses in ATP (Table 6). JDM patients may have a higher rate of oxygen consumption and mitochondrial oxidative phosphorylation for more effective ATP maintenance as compared with DM patients.
When comparing the levels of total ATP or Mg-ATP in the muscles of normal children and adults, the young subjects had a better metabolic profile than did their adult counterparts (Figure 3). For instance, normal children exhibited steady-state total ATP or Mg-ATP levels that were ∼15% higher than those of healthy adults during rest and exercise. Similarly, JDM patients showed ∼14–24% higher concentrations of ATP or Mg-ATP as compared with the adult DM patients. It is noteworthy that both JDM and DM patients at rest exhibited similar ATP decreases of ∼35% compared with their respective control groups.
Since Mg-ATP is in equilibrium with intracellular free Mg2+, it has been suggested that a decrease in free Mg2+ could be a primary cause of low intracellular ATP (29). In the DM patients, when levels of free Mg2+ were decreased during exercise, ATP was concurrently lowered, but which of these was the primary loss remains uncertain. In the JDM patients, a similar decrease in free Mg2+ during exercise did not result in a further decrease in ATP, indicating that other controlling factors may independently affect ATP and free Mg2+ levels (38, 39).
At rest, 5–7% of the total ATP in all 4 groups of subjects was in a free and enzymatically inactive state (Tables 5 and 6). This was within the normal range, as previously reported (12, 32). During exercise and recovery, the decreased levels of free Mg2+ in JDM and DM muscles negatively correlated with statistically significant increases in relative levels of free ATP. Thus, the ever-present deficit of low Mg-ATP in the muscles of the patients was compounded by an increased percentage of inactive, free ATP.
The defects in muscle function in DM and JDM patients are thought to be a direct manifestation of the disease, and not a result of deconditioning. The reason is that the severe weakness in DM and JDM is primarily present in the proximal muscles, while distal muscles can remain essentially unaffected. It would be unusual for normal, deconditioned subjects to have difficulty combing their hair, shaving, getting in and out of the car, etc. It is possible that a person with proximal muscle weakness may, over the course of time, become deconditioned due to inactivity. However, it would be very difficult to determine what percentage of the defects in the thigh muscles (as measured in the exercise protocol with P-31 MRS) was due to disease or to deconditioning. Therefore, one assumes that the weakness and fatigue, along with the metabolic abnormalities and Mg deficits, are components of the overall disease process.
Therapy that improves clinical status is often associated with normalization of low magnesium levels in a variety of diseases. For example, patients with mitochondrial cytopathies demonstrate low levels of magnesium. After treatment with coenzyme Q, these patients exhibited significantly increased Mg2+ levels as well as improved mitochondrial function in muscle tissues (40). In diabetes, treatment simultaneously improves clinical status and raises magnesium levels (41, 42). Magnesium supplementation has been successful in reversing or moderating the symptoms of dietary deficiency, alcoholic withdrawal, hypertension, cardiac arrhythmias, coronary heart disease, and eosinophilia-myalgia syndrome (20, 21, 43–47). Magnesium supplementation has also been shown to improve physical performance in healthy athletes (48, 49).
Prednisone and immunosuppressive therapy often improve the clinical and metabolic status in DM and JDM patients (7). The current investigation shows that concentrations of Mg-ATP are also significantly increased with treatment (Figure 6). The adult DM patients showed considerable variation in the extent of improvement in free Mg2+ levels, but all JDM patients demonstrated substantial increases in free Mg2+ in accordance with the generally greater therapeutic benefits seen in children (Figure 7) (50). This correspondence between improved clinical status and increased magnesium levels suggests that magnesium may have a significant role in the pathophysiology of muscle abnormalities in dermatomyositis.
In conclusion, the decreases in biologically active free Mg2+ and Mg-ATP in DM and JDM muscles indicate a possible relationship to the well-known weakness associated with magnesium deficiency. Prednisone and immunosuppressive therapy increase the free Mg2+ and Mg-ATP levels, indicating correlations between magnesium and improved muscle function. Although magnesium supplements are sometimes used in the treatment of DM patients, no controlled clinical trials have been reported. Such studies would be of interest and could provide insights into the pathophysiology of these diseases.