Earlier studies have shown that the progressive, unrelenting muscle loss associated with Duchenne muscular dystrophy (DMD) involves an imbalance between the rates of synthesis and degradation of muscle proteins. Although previous studies have suggested that oxandrolone may be beneficial in DMD, the mechanism of action of oxandrolone on muscle in DMD remains unclear. To address these issues, we combined stable isotope studies and gene expression analysis to measure the fractional synthesis rate of myosin heavy chain (MHC), the key muscle contractile protein, the transcript levels of the isoforms of MHC, and global gene expression profiles in four children with DMD before and after 3 mo of treatment with oxandrolone. Gastrocnemius muscle biopsies and blood samples were collected during the course of a primed 6-h continuous infusion of l-[U-13C]leucine on two separate occasions, before and after the 3-mo treatment with oxandrolone (0.1 mg·kg−1·day−1). Gene expression analysis was done with microarrays and RT-qPCR. In response to the treatment, MHC synthesis rate increased 42%, and this rise was accounted for, at least in part, by an upregulation of the transcript for MHC8 (perinatal MHC). Gene expression data suggested a decrease in muscle regeneration as a consequence of oxandrolone therapy, presumably because of a decrease in muscle degeneration. These findings suggest that 1) oxandrolone has a powerful protein anabolic effect on a key contractile protein and 2) larger and longer-term studies are warranted to determine whether these changes translate into meaningful therapy for these patients.
- stable isotope
- mass spectrometry
duchenne muscular dystrophy (DMD) is a severe and progressive form of human muscular dystrophy with a lethal outcome. Although the absence of the protein dystrophin, which is part of a glycoprotein complex located on the intracellular surface of the cellular membrane, was identified as the molecular basis of the disease (25, 28), and despite advances in our understanding of dystrophin function, the precise pathogenesis of the severe muscle damage remains unclear. The most striking manifestations of this disorder are loss of skeletal muscle mass and strength, resulting in severe disability. Previous studies in DMD have suggested that the loss of muscle is likely the result of an imbalance between muscle protein synthesis and degradation (18, 42). In DMD, dystrophin deficiency is present in muscle from fetal life onward. However, the loss of muscle is progressive. The progressive nature of the muscle loss implies that there are crucial secondary factors that compound the problems caused by dystrophin deficiency. The gradual loss of regenerating capacity of muscle, which depends on protein synthesis, may be an important factor in this pathogenetic process (18, 41, 42).
The absence of a cure for DMD underscores the importance of adjunctive therapeutic strategies. A recent study by Fenichel et al. (13) suggested an ameliorative effect of oxandrolone, an oral synthetic analog of testosterone, on muscle strength in DMD. Although a subsequent study by the same group did not find a significant change in the average manual muscle test (MMT) score, they found significant improvement in quantitative muscle testing (14), and no major adverse reactions were noted. The expected increase in linear growth was also seen in boys treated with oxandrolone. These results suggest that oxandrolone may slow down or stabilize the progression of weakness. Oxandrolone has also been used successfully to enhance lean body mass in other debilitating clinical situations such as human immunodeficiency syndrome infection and chronic alcoholic cirrhosis, lean body mass and upper- and lower-body maximal voluntary strength in the elderly, and linear growth in Turner syndrome and improves albumin levels, net protein balance, and lean body mass in severely burned children (20, 45, 48, 54). An acute anabolic effect of oxandrolone on the fractional synthesis rate (FSR) of mixed muscle protein has also been reported in healthy young men (46). In contrast to the rather less sensitive clinical measurements of muscle strength, changes in the FSR of proteins can easily be detected in the short term by their direct measurement using stable isotope tracer dilution techniques (3).
Because oxandrolone has proven protein anabolic effects in other diseases, and may have some beneficial effect in slowing the progress of weakness in DMD, we hypothesized that oxandrolone would provide an overall protein anabolic effect in DMD. To test this hypothesis, we measured the FSR of myosin heavy chain (MHC; a key muscle contractile protein) and albumin (the most abundant single protein in humans) before and after a 3-mo treatment with oxandrolone in DMD and determined mRNA expression levels of the isoforms of MHC by microarray analysis and RT-qPCR.
MATERIALS AND METHODS
Four boys (ages: 8.3, 10.4, 12.8, and 16.7 yr) with diagnosis of DMD were enrolled in the study. One of the subjects (age: 16.7 yr) was wheelchair dependent. The others were ambulatory. Informed written consent was obtained according to protocols approved by the local Institutional Review Board at Baptist Medical Center (Jacksonville, FL). Subjects were studied before and after treatment with oxandrolone (0.1 mg·kg−1·day−1 po) for 3 mo.
l-[U-13C]leucine was purchased from Cambridge Isotope Laboratories (Woburn, MA). Solutions were prepared under sterile conditions, filtered through a 0.2-μm filter, and stored at 4°C in sterile containers before infusion. Solutions were proven to be sterile and pyrogen free before use in human subjects. All electrophoresis reagents were purchased from Bio-Rad Laboratories (Richmond, CA). Trifluoroacetic acid was purchased from Pierce (Rockford, IL). TRIzol, SuperScript II RT, and Taq polymerase were purchased from Invitrogen (Carlsbad, CA); DNase and proteinase K were from Sigma (St. Louis, MO); SYBR Green I and RiboGreen were from Molecular Probes (Eugene, OR); and TaqStart Antibody was from BD Biosciences Clonetech (Palo Alto, CA).
For at least 3 days before each study, all subjects were asked not to change their dietary habits or pattern of activity, and they maintained a dietary record. The subjects were admitted to the Clinical Research Center on the evening of the 4th day.
Muscle strength was measured using MMT, using the Medical Research Council system, at baseline, after 1 mo and at the end of the 3-mo intervention period. On the night before the study day, an intravenous line was inserted, and the line was kept patent by a slow infusion of saline. The isotope infusion study was performed in the morning in the postabsorptive state, after an overnight fast as reported earlier (33, 38). At 7:00 AM, a second catheter was inserted into a contralateral hand vein for blood sampling; the saline infusion was discontinued, and isotope infusion was performed as a 6-h primed (7.6 μmol/kg) continuous infusion of l-[U-13C]leucine (7.6 μmol·kg−1·h−1), as previously described (3, 22, 38, 52). Blood and gastrocnemius muscle biopsies (125 mg) were obtained as described earlier in our own previous studies (3). The muscle sample was cut into two aliquots (∼100 mg for MHC FSR and ∼25 mg for mRNA and microarray measurements), immediately frozen in liquid nitrogen, and kept at −80°C until analysis. At each visit, patients and caregivers were queried as to the occurrence of adverse effects, general physical examinations were performed, and blood chemistries were obtained.
Purification of MHC and Albumin
MHC was purified by a preparative continuous-elution electrophoresis technique described previously (2, 3). In brief, the tissue samples were homogenized in an SDS-pyrophosphate buffer (pH 7.4) and centrifuged. The supernatant containing the proteins was electrophoresed on a polyacrylamide gel (4%T-2.5%C resolving gel, pH 8.8, and 4%T-2.5%C stacking gel, pH 6.8) using a preparative electrophoresis cell (model 491 Prep Cell; Bio-Rad Laboratories, Hercules, CA) to separate MHC from other proteins. The purity of the MHC fractions in each sample was established by analytical SDS-PAGE on a 5% gel followed by silver staining (2). MHC samples were overloaded on the analytic gel to detect any contamination from other proteins. The pure MHC fractions were pooled together, and the protein was precipitated using TCA. Albumin was extracted from plasma by differential solubility in absolute ethanol from TCA-precipitated plasma proteins as described previously (5, 16, 31). The purified muscle MHC and albumin were hydrolyzed using 1.5 ml 6 N hydrochloric acid.
Measurement of Isotopic Enrichment in MHC, Albumin, and α-[13C]Ketoisocaproate
The measurements of [13C]leucine enrichment in MHC and albumin were performed by a gas chromatography-combustion-isotope ratio mass spectrometer (GC-combustion-IRMS) method, as described in detail elsewhere (1). The constituent amino acids resulting from muscle MHC and albumin hydrolysates were derivatized as their N-heptafluorobutyryl methyl esters for analysis of isotopic enrichment by GC/combustion/IRMS. Plasma enrichment of α-[13C]ketoisocaproate (KIC; [13C]KIC) was determined by GC-mass spectrometry (MS) with selected ion monitoring and electron impact ionization of an oxime t-butyldimethylsilyl derivative, as previously described (44).
The FSR of MHC (expressed as %/h) was calculated using the equation (3, 38, 53) FSR = [(ΔEMHC/EKIC × t)] × 100, where ΔEMHC represents the difference in 13C enrichment in MHC between the biopsies. EKIC is the plateau value of 13C enrichment of KIC (the precursor pool), and t represents the time interval between the biopsies. Because of limitation in biopsy sample size, 13C enrichment in tissue free leucine was not determined, and [13C]KIC enrichment was used as a precursor pool in calculating FSR.
The FSR of albumin (expressed in %/24 h) was calculated using the precursor/product relationship: FSR = 24 × 100 × (EAlbumin)/(EKIC), where EAlbumin is the slope of the 13C enrichment in the bound leucine residues in albumin from 2 to 6 h of isotope infusion (atom %excess/h), and 100 and 24 convert the result to a percent and per day (5, 16, 53). The absolute synthesis rate (ASR) of albumin, which is the amount of albumin synthesized per kg body weight per day, was estimated as the product of FSR and intravascular mass of albumin (5, 16). The intravascular albumin pool size was calculated by multiplying plasma albumin concentration (in mg/l) by the plasma volume (in liters), and plasma volume was calculated by multiplying body weight by 0.045 (39). The ASR was then expressed as milligram per kilogram body weight.
Purification of RNA
RNA was isolated from the muscle biopsy samples with TRIzol, following the manufacturer's instructions. Purified RNA was further treated with DNase and proteinase K and was finally passed over an RNeasy minicolumn (Qiagen, Valencia, CA), following the suppliers' instructions. The purified RNA was quantified by fluorescence using RiboGreen dye.
mRNA was linearly amplified (2 rounds) and biotin labeled using the protocol of Baugh et al. (8). Labeled product was purified on a Qiagen RNeasy column, and the integrity of each sample was assessed by 1% formaldehyde-agarose gel electrophoresis. Labeled, amplified RNA was applied to Affymetrix Human Genome U133 GeneChip sets (U133A and U133B). Hybridization and scanning were performed at the Center for Applied Genomics (Newark, NJ). Image analysis was performed with Affymetrix Microarray Suite, version 5.0. Color images of the microarrays were generated from the data in Bioconductor and inspected for regional abnormalities in hybridization (17).
Probe-level data were background corrected and normalized, and summary expression for each feature was calculated using the Robust Multiarray Average algorithms described by Irizarry et al. (26), using the affy package of Bioconductor (17).
Differences between pre- and posttreatment samples were determined using Significance Analysis of Microarrays (SAM) (49). Gene expression data were filtered for those genes with expression levels greater than background (an expression level >100 in >25% of the samples) and for those genes demonstrating adequate variability between the samples (intraquartile range >0.5; see Ref. 51). The resulting data set was analyzed by SAM, where paired t-tests (corrected for the large number of comparisons) were used to estimate significance. The algorithm reports the false discovery rate (q) for each gene, which is analogous to a P value.
Gene expression patterns were analyzed by MAPPFinder and GenMAPP (GenMAPP 2.0; see Ref. 10). GenMAPP assigns the Gene Ontology (GO) classifications to each gene represented on the array and color codes them as to whether they are expressed and how the expression levels differ between samples. MAPPfinder analyzes the resulting data set to identify GO classifications, where the percentage of genes found to be changed by oxandrolone treatment was different from the percentage that would be expected by random chance. These GO classifications are particularly relevant in determining whether related groups of genes changed in concert with treatment.
RT reactions were conducted with 1.0 μg purified RNA, as previously described (40). The reaction for quantitative PCR was 20 mM Tris·HCl, pH 8.3, 50 mM KCl, 3 mM MgCl2, 0.1% Tween 20, 200 mM each dNTP, and 500 nM each primer and contained 50 mg/ml BSA, 0.4 × SYBR Green I, 0.4 units of Taq polymerase, 0.088 mg of Taq start antibody, and one-tenth of the RT reaction. Purified amplicon of the specific primer set was used for quantitative standards in parallel reactions. Cycling was conducted in a LightCycler (Roche Applied Science, Indianapolis, IN) with parameters of 95°C for 2 min and 40 cycles of 95°C for 0 s, 65–55°C for 5 s (stepdown protocol of 1°C per cycle for the first 10 cycles), and 72°C for 15 s (30 s for MHC2 and -6). Melting curves were determined at the end of each PCR run to confirm the identity of the sample and control amplicons.
The primers used were MHC1 (forward: GGAGGAACAATCCAACGTCAA, reverse: TGACCTGGGACTCAGCAATG), MHC2 (forward: AAGGATACCCAGATCCACC, reverse: CTCAGCATTACGCTTTTGC), MHC3 (forward: AAGCAAGCCTTTACCCAGC, reverse: CATCAACCATCAGATCCTCC), MHC4 (forward: GAGTGAACTACAGACTTCC, reverse: AGTGTCCTTCAGTATTCC), MHC6 (forward: AGAGAAACTACCACATCTTCTACC, reverse: AATGAATGTCCACTCAATGC), MHC7 (forward: AGACAACCTGGCAGATGC, reverse: TTTCTCCTTCTCCACTTTGG), MHC8 (forward: CAGAATACCAGTCTCATTAACACC, reverse: TCCTTCAAGCTCACGTACC), MHC11 (forward: TTCCAAATATCACAGACACC, reverse: CAATTATAGCTCATTGCAGC), MHC13 (forward: GAACACAAGCCTGATAAATACC, reverse: AAGCTCATTTTCCAGCTCC), and glyceraldehyde-3-phosphate dehydrogenase (forward: CTCCTGCACCACCAACTG, reverse: GCCTGCTTCACCACCTTC). Primers were designed using Vector NTI software and synthesized by Operon Biotechnologies.
Quantification was accomplished by determining the “threshold cycle” (the second derivative of the resulting fluorescence curve) at which the amplicon is detected during the PCR and comparing this with the standard curve calculated from the parallel quantification reactions. These calculations are done with the LightCycler software (version 3.5). For each sample, RT reactions were done in duplicate, and a single PCR reaction was done on each RT reaction. Results of the duplicates were averaged. Data were log transformed so as to approximate a normal distribution.
Results are expressed as means ± SE. For all assays, pre- and postoxandrolone treatment results were compared using the paired Student's t-test, with a P value of <0.05 considered significant.
One subject (age: 12.8 yr) with DMD participated only in the baseline of the study and did not complete the second study. He developed elevated liver transaminases (compared with baseline values). Other liver function tests were normal, there was no clinical evidence of hepatic dysfunction, and liver transaminases returned to baseline values when oxandrolone was discontinued. Oxandrolone was, therefore, not restarted, and muscle biopsy was not performed at 3 mo. The other three subjects tolerated oxandrolone treatment well and completed the study.
FSR of MHC and Albumin: Effect of Oxandrolone
We measured the FSR of MHC before and after treatment with oxandrolone. The FSR of MHC increased by 42 ± 6% (n = 3; P = 0.02) in response to the intervention. An increase was observed in each of the subjects (Fig. 1). Muscle strength measured by MMT did not deteriorate during the study period.
Figure 2 summarizes the results on albumin FSR, its concentration, and its ASR. We found that the FSR, concentration, and ASR increased 43 ± 5, 4.7 ± 0.6, and 49 ± 6%, respectively, in response to oxandrolone treatment. All these parameters increased in each of the subjects.
MHC Isoform Transcripts by RT-qPCR and Gene Expression Profile: Effect of Oxandrolone
The Affymetrix U133 arrays contain 44,916 “features” representing 20,882 unique UniGene clusters, 5,829 genes not yet assigned to a UniGene cluster, and 68 controls. Across all the samples, genes from 12,129 UniGene clusters and 1,818 genes not yet assigned to a Unigene cluster were found to be expressed. The correlation coefficients (Pearson's) of the features between the pretreatment samples were 0.9841, 0.9924, and 0.9835. The complete microarray data set can be found in the Gene Expression Omnibus database (GEO; www.ncbi.nlm.nih.gov/geo), accession no. GSE1764.
The 30 most highly expressed genes in DMD muscle are listed in Table 1. Haslett et al. (21) performed microarray analysis on muscle tissue from 12 boys with DMD and 11 normal controls (6 male, 5 female). The raw data from these experiments were downloaded from GEO (accession no. GSE1004) and processed as described in materials and methods. Table 1 includes the ranking of each gene from the Haslett data, showing good agreement between the data sets. Many of the most highly expressed genes are ubiquitously expressed (GAPD, translation factors, ribosomal proteins), but many are muscle specific, including sarcolipin, myoglobin, myosins, tropomyosin 1, troponin C, etc. These confirm that, despite the presence of fibrous tissue found in DMD muscle, the majority of RNA came from muscle cells.
SAM was done using paired analysis. Genes from 42 UniGene Clusters and 23 genes not associated with a UniGene cluster were found to be upregulated (q < 0.05) with oxandrolone treatment. Genes with at least twofold upregulation are shown in Table 2. MHC8 was significantly upregulated; baseline expression was 10.0 ± 0.7 (log2 signal, means ± SE) with a change of 0.6 ± 0.2 (log2 signal ratio, q < 0.05). Although MHC4 was also identified as having been upregulated, the data show that in one of the three sample pairs, the gene was upregulated sevenfold. In the other two samples, there was no major change (Fig. 3). Genes from 1,128 UniGene Clusters and 37 genes not associated with a UniGene cluster were found to be downregulated (q < 0.05). Genes with at least 2.3-fold downregulation are shown in Table 3.
The imbalance between the number of up- and downregulated genes raised concerns about a systemic difference between the pre- and posttreatment data sets. To determine whether there was global gene downregulation, the expression levels for ubiquitously expressed invariant “housekeeping” genes (11) were examined. Of 427 housekeeping genes, the expression of 89% of these genes was not significantly different between the pre- and posttreatment results. For comparison, 89% of these housekeeping genes were also unchanged between DMD and normal muscle, using the data of Haslett et al. (21).
MAPPFinder was used to identify those gene function classifications that were impacted by oxandrolone treatment. Table 4 shows a generalized pattern of downregulation of energy pathways, transcription, and translation in response to oxandrolone. MAPPFinder analysis was also done on the DMD vs. normal muscle data set of Haslett et al. (21). Table 5 lists selected genes from the muscle development GO category and their change in expression levels. The table shows that most of the genes upregulated in DMD muscle were downregulated or not altered by oxandrolone treatment. The muscle development genes that were downregulated by oxandolone included a number related to muscle differentiation from precursor cells. Four and a half LIM domains 1 (FHL1) is a transcription factor thought to play an important role during the early stages of skeletal muscle differentiation (34). Interferon-related developmental regulator 1 (IFRD1) is involved with myoblast fate determination (19). Vesicle-associated membrane protein 5 (myobrevin, VAMP5) is involved in vesicle trafficking events that are associated with myogenesis (55). MAP kinase 12 (MAPK12) is thought to be a signal transducer of the stress-sensing mechanism during differentiation between myoblasts and myotubes (24). Ankyrin repeat domain 2 (ANKRD2) is also part of the stress-sensing mechanism in muscle (29). Myogenic factor 6 (MYF6) is a myogenic transcription factor (43). Histone deacetylase 4 (HDAC4) is involved with transcription regulation mediated by the myogenic factors myocyte enhancer factors 2C and 2D (37). Insulin-like growth factor I (IGF-I) is a potent anabolic agent for skeletal muscle (6), promotes muscle regeneration (7), and acts in an autocrine manner in regenerating muscle (27). The data from Haslett et al. showed an upregulation of IGF-I in DMD muscle (Table 5). This gene was downregulated in response to oxandrolone.
To determine MHC mRNA expression levels with more precision, real-time RT-qPCR was done (Table 6). As in the microarray analysis, MHC8 (skeletal, perinatal) was upregulated 1.6-fold (log2 ratio of 0.68 ± 0.18; P = 0.048) by oxandrolone treatment (Fig. 3). No other MHC mRNAs were significantly altered.
A limitation of this study is the small sample size. Because this was a demanding, labor-intensive study that involved multiple muscle biopsies and stable isotope infusions in a population of children with a devastating disease, recruitment typically was difficult. However, despite differences in age and disease stage, the findings of this study clearly demonstrate that the anabolic effect of oxandrolone on muscle in DMD may be mediated by a stimulation of the synthesis rate of MHC, a key contractile protein in skeletal muscle. A significant augmentation of the FSR of MHC in response to oxandrolone in children with DMD, as well as changes in gene expression profile and transcript levels of isoforms of MHC, was observed. This increase in the synthetic rate of MHC appears to be, at least in part, through an upregulation of the transcript for the perinatal isoform of MHC. The pathophysiology of DMD involves the accelerated degeneration of muscle tissue because of membrane weakness caused by the mutation in the dystrophin gene. In the early stages of the disease, accelerated regeneration of muscle allows for near-normal function. It is felt that it is the steady loss of regenerative capacity that results in the long-term decline in these patients (18, 41, 42). Consistent with this, Haslett et al. (21) and Chen et al. (9) compared global transcript levels from DMD muscle with those from normal muscle and found an overall upregulation of genes associated with muscle regeneration. The data presented here using the same technology showed that oxandrolone had the effect of downregulating many of these genes in DMD muscle. We therefore conclude that oxandrolone has the effect of reducing muscle regeneration. None of the subjects in this small, short-term study showed a decline in muscle function, and others have shown modest improvement in DMD patients treated with oxandrolone (14). Hence, if regeneration has been decreased with no change or improvement in muscle function, there must have been a concomitant decrease in muscle degeneration.
The stable-isotope studies carried out in the current studies revealed higher baseline FSR for MHC than those reported in normal subjects (3, 23). These higher rates of MHC synthesis are consistent with the previously reported high turnover rate of myofibrillar proteins (assessed by 3-methylhistidine and creatinine excretion) in DMD (4, 35, 36), mixed-muscle proteins in the mdx mouse (32), and the FSR of mixed-muscle protein in DMD and Becker dystrophy (42). We furthermore showed a clear rise in MHC synthesis in these subjects in response to oxandrolone. There was a modest increase in the expression level of MHC8 (perinatal MHC), which likely played some part. However, MHC8 represented only a fraction of the MHCs being produced by this tissue, and the modest increase does not seem enough to explain the changes. An alternate explanation is that there was an increase in the efficiency of the translation process for MHCs in general. A lack of a noticeable change (improvement) in muscle strength in the current study may, however, be related to the limited sensitivity of the MMT used for strength measurement, the duration of intervention, and the small sample size. The study by Fennichel et al. (14), on the other hand, demonstrated an improvement in the total muscle strength score in response to oxandrolone treatment for 6 mo (14).
Additionally, the increase in the synthesis rate and concentration of albumin, a nutritionally important liver protein, is intriguing. These, along with the increase in the ASR of albumin, suggest an overall increase in the turnover of albumin. The implications of these outcomes are not clear from this study, but animal studies have suggested that the peripheral breakdown of circulating albumin contributes substantially to the total amino acid output available for protein synthesis in other tissues (30), and various human studies have showed that amino acid availability is an essential component in regulating muscle protein metabolism (50). In any case, the increase in the concentration of albumin along with increased synthesis rates of albumin and MHC suggest that oxandrolone has an overall protein anabolic effect in DMD, and its effect is not limited to muscle only.
The gene expression data showed a generalized downregulation of genes involved in energy pathways, translation, and transcription. Despite this, there is evidence of increased efficiency of MHC synthesis. This implies that there has been a reduction in demands for synthesis of other proteins that was greater than the increase in demand for MHC synthesis. One potential mechanism, as stated above, is a putative reduction in muscle regeneration. Because muscle regeneration involves a high-energy cost and the need for accelerated protein synthesis, even a small reduction in muscle regeneration would be expected to substantially reduce the energy and synthesis needs of muscle tissue. This suggests that the major anabolic effect of oxandrolone may be in slowing down the degenerative process rather than upregulating the regeneration process, which is already overstretched. There is also a possibility that posttranslational changes may contribute to increased protein accretion. Indeed these hypotheses need to be validated independently.
The mechanism of action of anabolic steroids (androgens) on muscle on the molecular level remains poorly understood. The classical androgen pathway entails the steroid entering the cell by diffusion and binding to the androgen receptor. The complex translocates to the nucleus and acts as a transcription factor. This pathway has been implicated in a direct effect on muscle protein synthesis for testosterone (15) and oxandrolone (46). In the study presented here, we describe an increase in MHC synthesis with only a modest increase in MHC mRNA expression levels and a decrease in the genes involved in translation, suggesting this anabolic effect was not a direct genomic action. Androgens have been shown to stimulate recruitment of precursor cells into myoblasts (47), and long-term testosterone treatment has been shown to upregulate IGF-I gene expression (46), presumably through recruitment of myoblasts, since IGF-I is not expressed in mature muscle fibers (27). Our results, however, demonstrate a decrease in muscle regeneration and decreased IGF-I expression, further suggesting that this was not the predominant effect of oxandrolone here. Nongenomic effects of the androgen receptor have also been described, including a role in calcium regulation in skeletal muscles via an inositol triphosphate-mediated pathway (12). The effects of this pathway are unknown. Determining whether and how this or other pathways could be involved in slowing the muscle degeneration in DMD would require further studies.
In summary, we conclude that oxandrolone has anabolic effects on DMD muscle and also has the effect of decreasing muscle degeneration, easing the demands for muscle regeneration. We speculate that, by conserving regenerative capacity, long-term therapy with oxandrolone may prolong muscle function in these patients. Treatment approaches to incorporate these responses into a multidrug paradigm or other therapies could prove to be beneficial to DMD patients. Larger studies are needed to determine whether these tissue-level effects of oxandrolone translate into functional improvement for boys with DMD.
Microarray data were generated by the Service Facility of the Center for Applied Genomics, Public Health Research Institute, which is supported by the New Jersey Commission on Science and Technology. We received a travel grant from BTG for this study. This study was supported by grants from Nemours Research Programs.
We thank the volunteers for participating in this demanding study. We are grateful to Burnese Rutledge, Shiela Smith, and the nursing staff of the Clinical Research Center at the Wolfson Children's Hospital for superb assistance; Brenda Sager, Shawn Sweeten, Lynda Everline, Astride Altomare, Rebecca Macken, and Jeff Bailey for skilled technical assistance; Karen Lanier for excellent secretarial support; Dr. Michael Pollack for providing one DMD patient for the study; Dr. Jim Sylvester for helpful comments; Dr. Ted Kramer and Bio-Technology General Corp (BTG) for providing oxandrolone free of cost for the treatment; and Dr. Vicky Funanage for helpful comments and constant support.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 by American Physiological Society