Renin-angiotensin-aldosterone system inhibitors improve membrane stability and change gene-expression profiles in dystrophic skeletal muscles

Abstract

Angiotensin-converting enzyme inhibitors (ACEi) and mineralocorticoid receptor (MR) antagonists are FDA-approved drugs that inhibit the renin-angiotensin-aldosterone system (RAAS) and are used to treat heart failure. Combined treatment with the ACEi lisinopril and the nonspecific MR antagonist spironolactone surprisingly improves skeletal muscle, in addition to heart function and pathology in a Duchenne muscular dystrophy (DMD) mouse model. We recently demonstrated that MR is present in all limb and respiratory muscles and functions as a steroid hormone receptor in differentiated normal human skeletal muscle fibers. The goals of the current study were to begin to define cellular and molecular mechanisms mediating the skeletal muscle efficacy of RAAS inhibitor treatment. We also compared molecular changes resulting from RAAS inhibition with those resulting from the current DMD standard-of-care glucocorticoid treatment. Direct assessment of muscle membrane integrity demonstrated improvement in dystrophic mice treated with lisinopril and spironolactone compared with untreated mice. Short-term treatments of dystrophic mice with specific and nonspecific MR antagonists combined with lisinopril led to overlapping gene-expression profiles with beneficial regulation of metabolic processes and decreased inflammatory gene expression. Glucocorticoids increased apoptotic, proteolytic, and chemokine gene expression that was not changed by RAAS inhibitors in dystrophic mice. Microarray data identified potential genes that may underlie RAAS inhibitor treatment efficacy and the side effects of glucocorticoids. Direct effects of RAAS inhibitors on membrane integrity also contribute to improved pathology of dystrophic muscles. Together, these data will inform clinical development of MR antagonists for treating skeletal muscles in DMD.

duchenne muscular dystrophy (DMD) is a fatal genetic muscle disease that results in progressive degeneration of skeletal and cardiac muscles and affects 1:5,000 boys (20). DMD is caused by complete absence of the dystrophin protein, which stabilizes muscle membranes by linking the subsarcolemmal cytoskeleton to a transmembrane glycoprotein complex. Currently, the standard-of-care treatment for DMD is prednisone, a glucocorticoid that delays loss of ambulation by an average of 2 yr but has numerous serious side effects (6). Glucocorticoids are widely used to treat both acute and chronic inflammation because of their anti-inflammatory and immunosuppressive effects; however, the underlying mechanism by which these glucocorticoid receptor (GR) agonists benefit skeletal muscles is not fully understood. Several laboratories have demonstrated that prednisone even worsens damage to both skeletal and cardiac muscles in DMD mouse models (13, 14, 31).

The angiotensin-converting enzyme inhibitor (ACEi) lisinopril and the mineralocorticoid receptor (MR) antagonist spironolactone are FDA-approved drugs with a long history of safety and efficacy in treating heart diseases and have minimal side effects (9, 21). Lisinopril and spironolactone indirectly and directly, respectively, target the steroid hormone MR through the renin-angiotensin-aldosterone system (RAAS). Lisinopril is now recommended for use in DMD patients starting at the age of 10 yr (19), and the addition of an MR antagonist has recently been demonstrated to further slow the progression of cardiomyopathy in DMD patients (27).

Our group has previously shown that combined treatment with lisinopril and spironolactone improved skeletal muscle function and pathology, in addition to the heart, in the utrn^+/−;mdx (“het”) mouse model of DMD (26). Mice treated with a combination of these drugs showed 80% of normal muscle force generation in both respiratory and limb muscles compared with only 40% of normal force observed in untreated het mice (26). Treatment also led to a significant reduction in ongoing skeletal muscle damage, but a direct effect on membrane integrity has not been tested.

We recently showed that MRs, not previously investigated in skeletal muscles, were present in limb and respiratory muscles from wild-type and dystrophic mice (8). The endogenous MR agonist aldosterone was able to induce a large number of gene-expression changes in normal differentiated human myotubes, supporting MR functions as a steroid hormone receptor in skeletal muscles (8). Together, these data suggest that ACEi and MR antagonists can have a direct therapeutic effect on skeletal muscles.

Inflammation exacerbates muscle damage and contributes to myonecrosis in dystrophic muscles (7). Previous studies have shown that treatment with anti-inflammatory glucocorticoids can shift macrophages from a proinflammatory M1 phenotype to an anti-inflammatory M2 phenotype and improve muscle pathology (7). MR has also been shown to regulate macrophage polarization in other tissues but has not been investigated in skeletal muscles (3). MR activation shifts macrophages to a proinflammatory M1 phenotype, whereas treatment with the nonspecific MR antagonist spironolactone or selective MR antagonist eplerenone promotes an anti-inflammatory M2 phenotype (3). Together, these data support that MR antagonism can result in anti-inflammatory effects similar to glucocorticoids, which may contribute to the improved muscle pathology observed in lisinopril plus spironolactone (LS)-treated dystrophic mice. The goal of this study was to begin to define the cellular and molecular effects of RAAS inhibitors on skeletal muscles. RAAS inhibitors, used in combination clinically in cardiology, were also compared with molecular changes resulting from standard-of-care glucocorticoid treatment.

MATERIALS AND METHODS

Animals.

All protocols were approved by the Institutional Animal Care and Use Committee of The Ohio State University, are in compliance with the laws of the United States, and conform to the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals. Dystrophin-deficient, utrophin, haplo-insufficient het male mice (38) on a C57BL/10 background were bred in house and used for treated groups or untreated controls. Flexor digitorum brevis (FDB) muscles were used for membrane integrity assays, since they are small, relatively thin muscles that can be dissected cleanly at tendons, allowing for intact surgical removal from the animal, and are typically used for laser-injury experiments in dystrophic mice. Quadriceps muscles were used for the genomic studies, since they are larger muscles that provide sufficient tissue to conduct gene-expression measurements and can be compared with previously published data sets available from this muscle type from various studies on dystrophic mice.

Confocal microscopy membrane integrity assay.

Eight 4-wk-old het male mice, housed two per cage, were given water bottles containing 132 mg/l lisinopril (CAS no. 83915-83-7; SBH Medical, Worthington, OH) + 250 mg/l spironolactone (S3378; Sigma-Aldrich, St. Louis, MO; dissolved in 0.1% ethanol) in reverse-osmosis water (n = 4) or reverse-osmosis water only (n = 4), as described in our previous preclinical studies (17, 26), for 2–4 wk. Medicated water bottles were replaced three times/week, and the volume of water consumed was recorded. Mice were weighed once/week. Mice consumed approximately the predicted dosages of 20 mg·kg⁻¹·day⁻¹ lisinopril and 37.5 mg· kg⁻¹·day⁻¹ spironolactone. Membrane integrity assays were performed on isolated FDB muscle bundles to evaluate the effect of LS on treated versus untreated het mice. A group of four C57BL/6J (C57) male mice were used as wild-types for comparison with untreated and treated het mice. FDB muscles were surgically isolated and placed in minimal Ca²⁺ Tyrode solution (140 mM NaCl, 5 mM KCl, 2 mM MgCl₂, and 10 mM HEPES, pH 7.2). Muscle bundles were mechanically separated at the tendon and then adhered on MatTek glass-bottomed Petri dishes using liquid bandage on the exposed tendons (New-Skin). Membrane disruption was induced in the presence of Tyrode supplement with 2.5 µM FM 4-64 dye (Thermo Fisher Scientific, Waltham, MA) and 2 mM Ca²⁺, with a FluoView FV1000 multi-photon confocal laser-scanning microscope (Olympus, Melville, NY). A circular region of interest was selected along the edge of the sarcolemma and irradiated at 24% of maximum infrared laser power for 3 s. Pre- and postdamage images were captured every 3 s for a total of 60 s. Extent of membrane damage was analyzed using the ImageJ Fiji software (National Institutes of Health, Bethesda, MD) by measuring the fluorescence intensity encompassing the site of damage with results represented as change in fluorescence signal relative to its starting signal (ΔF/F0), as described previously (34). Fibers for untreated het mice (n = 20), for LS-treated het mice (n = 24), and for C57 mice (n = 16 fibers, bred in house; The Jackson Laboratory, Bar Harbor, ME) were measured from the four mice used for each group. GraphPad Prism (GraphPad Software, La Jolla, CA) was used to fit the ΔF/F0 curve for each animal, calculate the area under the curve, and perform statistical analysis. The area under the curve represents a quantification of total dye accumulation during the entire period of laser damage. This calculation provides an index of the overall extent of sarcolemmal membrane disruption in each muscle fiber tested. Technical replicates for each biological replicate were averaged. All laser-injury experiments and accompanying analyses were performed by an operator blinded to the treatment groups.

Microarray analysis.

Het mice (n = 3/group) were treated from 4 to 6 wk of age with Teklad Rodent Chow (no. 7912), containing 133 mg/kg lisinopril (CAS no. 83915-83-7; SBH Medical) and either 666.66 mg/kg LS (S3378; Sigma-Aldrich) or 2,000 mg/kg eplerenone plus lisinopril (EL; Compound Transfer Program; Pfizer, New York, NY, prepared by Research Diets, New Brunswick, NJ) or water bottles containing 10 mg/l prednisolone (P6004; Sigma-Aldrich; dissolved in 0.04% ethanol) in reverse-osmosis water, replaced three times/week, or left untreated. Chow was used to compare treatment between MR antagonists, since eplerenone is not soluble in water. Mice were weighed weekly, and volume of water or chow consumed was recorded to calculate an average drug dosage of lisinopril (20 mg·kg⁻¹·day⁻¹), spironolactone (100 mg·kg⁻¹·day⁻¹), eplerenone (200 mg·kg⁻¹·day⁻¹), and prednisolone (1.5 mg·kg⁻¹·day⁻¹). All mice were dissected during the middle of the light cycle between 9 AM and 3 PM. One quadriceps muscle from each mouse was removed for RNA processing, and the other quadriceps muscle was embedded in optimal-cutting temperature medium and frozen on liquid nitrogen-cooled isopentane for histological analyses.

RNA from mouse quadriceps muscles was isolated using TRIzol reagent (Thermo Fisher Scientific), according to the manufacturer’s instructions. Samples were DNAse treated using RQ1 DNAse (Promega, Madison, WI) and further purified using the RNeasy mini kit (Qiagen, Germantown, MD) cleanup protocol, and integrity was verified using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). A 100-ng aliquot of total RNA was linearly amplified, and 5.5 µg cDNA was labeled and fragmented using the GeneChip WT PLUS Reagent Kit (Affymetrix, Santa Clara, CA), following the manufacturer's instructions. Labeled cDNA targets were hybridized to Affymetrix GeneChip Mouse Transcriptome Array 1.0 for 16 h at 45°C, rotating at 60 rpm. The arrays were washed and stained using the GeneChip Fluidics Station 450 and scanned using the GeneChip Scanner 3000. Arrays were normalized using the gene-level SST RMA algorithm in Expression Console software and comparisons made in Transcriptome Analysis Console software (Affymetrix). The microarray data have been deposited in the National Center for Biotechnology Information Gene Expression Omnibus (accession no. GSE84876). Gene groups were assigned using the Functional Annotation clustering tool from the Database for Annotation, Visualization and Integrated Discovery (DAVID).

Immunofluorescence.

Quadriceps muscle cryosections (8 µm) were stained with an antibody against mouse IgG (Alexa 488 goat anti-mouse IgG, 1:200; Thermo Fisher Scientific) to visualize ongoing muscle damage. Samples were mounted using Vectashield mounting medium (Vector Laboratories, Burlingame, CA) with 1 µl/ml 4′,6-diamidino-2-phenylindole and viewed using an Eclipse E800 microscope (Nikon Instruments, Melville, NY). Images were taken under a ×20 objective using a SPOT RT Slider digital camera and SPOT software.

RESULTS

Direct assessment of membrane stability effects of LS treatment on dystrophic skeletal muscles.

The results from previous preclinical studies support that treatment with lisinopril plus an MR antagonist reduces ongoing muscle damage, but a direct effect on membrane stability has not been investigated. To confirm the loss of membrane integrity in the het model, FDB muscle bundles from het mice were subjected to laser-induced sarcolemmal membrane damage. An infrared multiphoton laser produced localized damage to the sarcolemmal membrane of intact muscle fibers in the presence of a lipophilic FM4-64 dye, which is nonfluorescent in aqueous solution but fluoresces intensely once it enters the cell. When the membrane re-seals, dye entry ceases, and the fluorescent signal stabilizes. The tracking of the fluorescence intensity allows resolution of the extent of membrane disruption. Laser-injury assays demonstrated the predicted compromised membrane stability in untreated het mice compared with wild-type controls (Fig. 1). FDB muscles isolated from het mice treated for 4 wk with LS show a significant improvement in membrane integrity compared with untreated het mice (P < 0.0001; Fig. 1).

Fig. 1.

Lisinopril plus spironolactone (LS) treatment improves membrane integrity in het mice. A: representative confocal microscopy images of isolated FDB muscle bundles from het mice left untreated (top) or treated with LS (bottom) at indicated time points following injury by an infrared multiphoton laser (arrows) in the presence of FM4-64 dye. Original scale bar, 10 µm. B: FM4-64 dye influx over time in muscle fibers is quantified as the change/increase in red fluorescent signal over baseline (ΔF/F0) over a time course following induction of injury (time = 0 s; arrow). Fibers for injury experiments were isolated from 4 mice from each group: untreated het, LS-treated het, and C57 wild-type mice. Data are presented as means ± SE. C: quantification of the area under the curve (AUC) among untreated het (U), LS-treated hets, or C57 wild-type mice. One-way ANOVA, followed by Dunnett’s post hoc test, was used to identify significance between untreated hets and the 2 other groups. ****P < 0.0001 compared with untreated hets.

Treatment with LS or EL results in overlapping gene-expression changes.

Whereas increased membrane integrity could underlie some of the improved structure and function observed in dystrophic skeletal muscle following LS treatment, the molecular basis for the efficacy of RAAS inhibition is not clear. We have previously shown gene-expression differences between LS-treated and untreated het mice at the conclusion of a 16-wk study after phenotypes had significantly diverged (8, 26). These gene-expression differences represent the secondary benefits of treatment, in addition to the potential underlying therapeutic gene-expression changes. We have also shown that the endogenous MR agonist aldosterone can change gene expression in normal human myotubes (8). Here, we sought to begin to define the molecular changes that may underlie the efficacy of targeting MR in skeletal muscle. We compared treatment with LS, which can also bind glucocorticoid and androgen receptors, with lisinopril, plus the specific MR antagonist EL. These two treatments were recently shown to have equivalent efficacy in het mice (16).

We performed gene-expression microarray on 6-wk-old male het mice that had been treated for 2 wk with LS or EL or left untreated. These studies also included a group of het mice treated with prednisolone, the active metabolite of prednisone, to begin to differentiate between gene-expression changes that may underlie efficacy versus side effects of standard-of-care glucocorticoids. There were 66 gene-expression changes in quadriceps muscles between LS-treated and untreated het mice (Fig. 2). Treatment increased expression of 20 genes (2- to 7-fold) and decreased expression of 46 genes (2- to 7-fold). Similarly, EL treatment led to 70 gene-expression changes compared with untreated—with 23 genes increased (2- to 5-fold) and 47 decreased—with 45 of these reduced <7-fold (Fig. 2). S100a8 was decreased 160-fold by EL treatment, which was the largest fold change observed out of any of our previous microarrays. We reanalyzed the raw gene-expression data to determine whether this change was also present with LS treatment. We observed that there was an outlier in the untreated group, and when removed from the analysis, the S100a8 gene was decreased 93-fold in the LS group. In addition, S100a9, which functions as a dimer with S100a8, was now decreased in both EL and LS compared with untreated het mice (64- and 50-fold, respectively).

Fig. 2.

Microarray analysis of 2 wk-treated dystrophic het mice with a nonspecific or specific MR antagonist plus lisinopril results in a comparable number of gene-expression changes. Treatment with lisinopril plus spironolactone (LS) results in decreased expression of 46 genes and increased expression of 20 genes (66 genes total) compared with untreated het. Eplerenone plus lisinopril (EL) treatment results in a comparable number of gene-expression changes, 23 increased and 47 decreased (70 genes total). Thirty-three (50%) of these gene changes are conserved in EL-treated mice compared with untreated het. Treatment with standard-of-care prednisolone (Pred) results in 374 gene-expression changes, 122 increased and 252 decreased compared with untreated het. Only 62 genes of these genes are conserved with LS- or EL-treated mice versus untreated het; 312 genes are specific to prednisolone treatment. Ten gene-expression changes are present in both the LS and EL compared with untreated het but not in the prednisolone group. *Genes that are changed in the opposite directions with MR antagonist and prednisolone treatment. Dbp is represented in both the group of 10 gene changes conserved between LS and EL and as the gene-expression change (*) conserved with prednisolone but in the opposite direction.

Thirty-three gene-expression changes were conserved between LS and EL compared with untreated (Table 1 and Fig. 3A). Only 10 gene-expression changes conserved between LS and EL compared with untreated were not present or changed in the same direction in the comparison of prednisolone and untreated. Scd1, Scd2, Dbp, Lrtm1, Prkag3, and Nr1d1 were all increased by LS or EL treatment (5.3-, 4.4-, 3.1-, 3.0-, 3.0-, and 2.3-fold, respectively, in EL). Mt1, Pim1, Mt2, and Clu were reduced by LS or EL treatment (2.3-, 2.4-, 2.8-, and 2.9-fold, respectively, in EL). These genes were functionally classified as either ion binding or transcription factors and may underlie MR-specific efficacy.

Table 1.

Fold changes of gene-expression differences conserved in microarray comparing quadriceps muscles from lisinopril plus spironolactone (LS) and eplerenone plus lisinopril (EL) mice versus untreated het controls (Un)

Gene Symbol	Full Gene Name	Fold Change (LS/Un)	Fold Change (EL/Un)
Immune response and homeostasis
Cd14	CD14 antigen	−2.8	−2.8
Cd80	CD80 antigen	−2.9	−3.1
Csf2rb	Colony-stimulating factor 2 receptor, beta, low-affinity (granulocyte-macrophage)	−3.8	−3.5
Fcgr3	Fc receptor, IgG, low-affinity III	−2.2	−2.2
Il6	Interleukin 6	−3.9	−2.9
Mt1	Metallothionein 1	−2.1	−2.3
Nfkbia	Nuclear factor of kappa light polypeptide gene enhancer in B cells inhibitor, alpha	−2.8	−2.1
Tfpi2	Tissue factor pathway inhibitor 2	−2.3	−2.4
Thbs1	Thrombospondin 1	−6.2	−6.8
Scd2	Stearoyl-coenzyme A desaturase 2	4.1	4.4
Wfdc17	Whey acidic protein (WAP) 4-disulfide core domain 17	−2.1	−2.6
Regulation of transcription
Dbp	D Site albumin promoter-binding protein	4.2	3.1
Litaf	LPS-induced TNF-α factor	−2.4	−2.3
Mxd1	Myc-associated factor X (MAX) dimerization protein 1	−2.3	−2.6
Nr1d1	Nuclear receptor subfamily 1, group D, member 1	2.4	2.3
Nr4a3	Nuclear receptor subfamily 4, group A, member 3	−5.6	−4.0
Ion binding and transport
Adamts1	A disintegrin-like and metallopeptidase (reprolysin-type) with thrombospondin type 1 motif, 1	−2.5	−2.2
Mt2	Metallothionein 2	−2.4	−2.8
Pim1	Proviral integration site 1	−2.0	−2.4
Scd1	Stearoyl-coenzyme A desaturase 1	7.1	5.3
Zfp36	Zinc finger protein 36	−4.7	−3.5
Apoptosis and proteolysis
Clu	Clusterin	−2.1	−2.9
Alternative splicing
Asb15	Ankyrin repeat and SOCS box-containing 15	2.8	3.4
Plaur	Plasminogen activator, urokinase receptor	−2.4	−2.7
Prkag3	Protein kinase, AMP-activated, gamma 3 noncatalytic subunit	3.1	3.0
Transmembrane
Atp1b4	ATPase, Na+/K+ transporting, beta 4 polypeptide	2.0	2.6
Ifitm1	Interferon-induced transmembrane protein 1	−2.2	−2.1
Lrtm1	Leucine-rich repeats and transmembrane domain 1	2.5	3.0
Ms4a4a	Membrane-spanning 4-domains, subfamily A, member 4A	−2.7	−2.8
Tmem252	Transmembrane protein 252	−2.9	−3.0
Unknown or specific functions
Dusp5	Dual-specificity phosphatase 5	−2.5	−2.3
Hdc	Histidine decarboxylase	−6.6	−6.0
Sec14l5	SEC14-like 5 (Saccharomyces cerevisiae)	2.7	2.1

Genes were functionally clustered using the Functional Annotation clustering tool from the Database for Annotation, Visualization and Integrated Discovery (DAVID). Positive fold changes indicate that gene expression was increased with treatment and vice versa. The 10 gene-expression changes conserved between LS and EL, but not changed by prednisolone, are shown in bold.

Fig. 3.

Genes involved in immune response and transcriptional regulation are conserved between MR antagonists and standard-of-care glucocorticoid agonists. A: thirty-three gene-expression changes were conserved between lisinopril plus spironolactone (LS) and eplerenone plus lisinopril (EL). Out of these 33, only 10 were also not conserved with prednisolone treatment. B: there were 62 genes conserved between prednisolone (Pred) and LS or EL treated versus untreated het mice. C: an additional 312 genes were detected only in the Pred versus untreated het group. These genes encompass similar functional groups to those in A and B, including immune response, transcriptional regulators, ion binding, transmembrane, and alternative (Alt.) splicing. However, several functional groups specific to prednisolone treatment were also represented, including 35 genes involved in apoptosis or proteolysis (11.1%), 26 cytokines with chemokine activity (8.3%), 10 cytoskeleton (3.2%), 11 cell migration/motility (3.5%), 4 cell adhesion (1.3%), 3 drug metabolism (Met.; 1%), 3 behavioral response (1%), and 14 genes with GTPase activity (4.5%). D: seventeen gene-expression changes were conserved between Pred and EL, and (E) 21 were conserved between Pred and LS, subsets of the genes represented in B.

Short-term prednisolone treatment increases apoptotic genes and damages skeletal muscles.

Sixty-two genes were conserved between prednisolone and either LS or EL compared with untreated (Figs. 2 and 3B and Table 2). Of these 62 genes, 18 were increased with prednisolone treatment, and 44 were decreased. The highest fold change for increased genes was 3.9-fold, and only two genes were increased >3-fold. The largest fold change for decreased genes was 42-fold, and 23 of the 44 genes were reduced >3-fold. Overall, all three drug treatments reduced expression of more genes than they increased. Of the 62 genes conserved between prednisolone and either LS or EL compared with untreated het mice, the majority of genes could be classified into the functional groups of cytokine production/immune response (29%) or regulation of transcription (22.6%; Fig. 3B). One of the genes conserved among all three groups (Dbp), and three of the genes conserved between LS and prednisolone were changed in the opposite direction with mineralocorticoid antagonists compared with prednisolone (Fig. 2 and Table 2).

Table 2.

Fold changes of gene-expression differences in microarray comparing quadriceps muscles from prednisolone (P)-treated and untreated het mice conserved with either lisinopril plus spironolactone (LS) or eplerenone plus lisinopril (EL) versus untreated het controls (Un)

		Fold Change
Gene Symbol	Full Gene Name	P/Un	LS/Un	EL/Un
Regulation of cytokine production/immune response
Adora2b	Adenosine A2b receptor	−2.4	X	−2.1
Ccl9	Chemokine (C-C motif) ligand 9	−4.3	X	−5.2
Cd14	CD14 antigen	−2.4	−2.8	−2.8
Cd80	CD80 antigen	−3.4	−2.9	−3.1
Csf2rb	Colony stimulating factor 2 receptor, β, low-affinity (granulocyte-macrophage)	−3.5	−3.8	−3.5
Csf2rb2	Colony stimulating factor 2 receptor, β2, low-affinity (granulocyte-macrophage)	−3.0	X	−2.6
Fcgr3	Fc receptor, IgG, low-affinity III	−2.6	−2.2	−2.2
Il1b	Interleukin 1β	−42.1	X	−40.3
Il6	Interleukin-6	−3.6	−3.9	−2.9
Klrb1b	Killer cell lectin-like receptor subfamily B member 1B	−2.3	−2.3	X
Nfkbia	Nuclear factor of kappa light polypeptide gene enhancer in B cells inhibitor, alpha	−4.7	−2.8	−2.1
Oas1a	2′-5′ oligoadenylate synthetase 1A	−3.2	−2	X
Oas1g	2′-5′ oligoadenylate synthetase 1G	−2.9	−2	X
Srgn	Serglycin	−2.6	X	−2.1
Thbs1	Thrombospondin 1	−10.5	−6.2	−6.8
Tlr8	Toll-like receptor 8	−2.3	−2.1	X
Wfdc17	WAP four-disulfide core domain 17	−3.6	−2.1	−2.6
Zfp36	Zinc finger protein 36	−5.2	−4.7	−3.5
Regulation of transcription
Arntl	Aryl hydrocarbon receptor nuclear translocator-like	2.8	−2.3	X
Cebpd	CCAAT/enhancer binding protein (C/EBP), delta	−7.4	−4.5	X
Cited2	Cbp/p300-interacting transactivator, with Glu/Asp-rich carboxy-terminal domain, 2	2.6	X	2.5*
Dbp	D site albumin promoter binding protein	−2.2	4.2	3.1
Dmrt2	Doublesex and mab-3 related transcription factor 2	2.1	−2	X
Fgfrl1	Fibroblast growth factor receptor-like 1	2.3	X	2.1
Id1	Inhibitor of DNA binding 1	−2.9	X	−2.6
Id2	Inhibitor of DNA binding 2	−3.4	−2.4	X
Litaf	LPS-induced TN factor	−3.0	−2.4	−2.3
Mxd1	MAX dimerization protein 1	−3.2	−2.3	−2.6
Npas2	Neuronal PAS domain protein 2	2.5	−2.2	X
Nr4a3	Nuclear receptor subfamily 4, group A, member 3	−10.7	−5.6	−4
Pyhin1	Pyrin and HIN domain family, member 1	−4.5	−2	X
Rcor3	REST corepressor 3	2.2	X	2
Ion-binding and transport
Adamts1	A disintegrin-like and metallopeptidase (reprolysin type) with thrombospondin type 1 motif, 1	−3.1	−2.5	−2.2
F5	Coagulation factor V	−3.2	X	−2.2
Apoptosis
Casp8	Caspase 8	−2.1	−3.4*	X
Alternative splicing
Asb14	Ankyrin repeat and SOCS box-containing 14	2.0	2.8*	3.37*
Fgfrl1	Fibroblast growth factor receptor-like 1	2.3	X	2.1
Plaur	Plasminogen activator, urokinase receptor	−2.5	−2.4	−2.7
Transmembrane
Atp1b4	ATPase, Na+/K+ transporting, beta 4 polypeptide	3.6	2	2.6
Fcgr4	Fc receptor, IgG, low affinity IV	−4.2	−2.3	X
Ifitm1	Interferon induced transmembrane protein 1	−2.3	−2.2	−2.1
Ifrd1	Interferon-related developmental regulator 1	−2.6	X	−2.3
Ms4a4a	Membrane-spanning 4-domains, subfamily A, member 4A	−3.3	−2.7	−2.8
Pnpla3	Patatin-like phospholipase domain containing 3	2.2	2.1	X
Slc15a3	Solute carrier family 15, member 3	−2.0	−2.2	X
Sntb1	Syntrophin, basic 1	2.3	2.5	X
Tmem252	Transmembrane protein 252	−4.6	−2.9	−3
Ugcg	UDP-glucose ceramide glucosyltransferase	−2.9	X	−2.2
Unknown or specific functions
Dhrs7c	Dehydrogenase/reductase (SDR family) member 7C	3.9	X	2.2
Dusp5	Dual-specificity phosphatase 5	−2.6	−2.5	−2.3
Gm4980	Predicted gene 4980	2.5	X	2.1
Gm6307	Predicted gene 6307	2.6	X	2.5
Hdc	Histidine decarboxylase	−7.7	−6.6	−6
Irs2	Insulin receptor substrate 2	−2.5	−2.4	X
Lrrc30	Leucine rich repeat containing 30	2.7	2.1	X
Lrrc39	Leucine rich repeat containing 39	2.3	2.2	X
Prr33	Proline rich 33	2.4	X	2
Ptpn1	Protein tyrosine phosphatase, nonreceptor type 1	−2.1	−2.4	X
Sec14l5	SEC14-like 5 (S. cerevisiae)	2.3	2.7	2.1
Synpo2l	Synaptopodin 2-like	2.0	2.1	X
Tfpi2	Tissue factor pathway inhibitor 2	−3.0	−2.3	−2.4
Tnfaip6	Tumor necrosis factor-α-induced protein 6	−5.5	−7	X

Genes were functionally clustered using the Functional Annotation clustering tool from the Database for Annotation, Visualization and Integrated Discovery (DAVID). Positive fold changes indicate gene expression was increased with treatment and vice versa. Bold X's represent a lack of a 2-fold gene-expression change in the LS or EL treatment groups. Values underlined and italicized highlight gene-expression changes going in the opposite direction with P versus LS or EL treatment.

^*Functionally conserved gene family members in LS or EL comparisons (Cited4, Casp4, and Asb15).

Gene-expression changes (312) were present only in the prednisolone versus untreated het comparison and may represent changes associated with either glucocorticoid-specific muscle benefits or side effects (Figs. 2 and 3C and Table 3). Staining for IgG in quadriceps muscles of het mice treated for 2 wk and used for this microarray experiment (Fig. 4) recapitulated the increased skeletal muscle damage previously observed after longer treatment with prednisolone (14, 31). Genes (103) were increased by prednisolone, and 209 were reduced. Increases ranged from 2- to 6.9-fold with 17 gene-expression changes >3-fold. Decreased genes were reduced between 2- and 57-fold, with 64 genes reduced >3-fold. Classification of these genes identified several functional groups that were not present in the LS or EL versus untreated het comparison. The functional groups specific to prednisolone treatment were apoptosis or proteolysis (11.1%), cytokine/chemokine activity (8.3%), GTPase activity (4.5%), cytoskeleton (3.2%), cell migration and motility (3.5%), and cell adhesion (1.3%; Fig. 3C).

Table 3.

Fold changes of gene-expression differences specific to microarray comparing quadriceps muscles from prednisolone (Pred)-treated het mice versus untreated het controls (Un)

Gene Symbol	Full Gene Name	Fold Change (Pred/Un)
Immune response/regulation of immune response
Cd274	CD274 antigen	−2.0
Cd300ld	CD300 molecule-like family, member d	−2.2
Clec4n	C-Type lectin domain family 4, member n	−4.1
Cmip	c-Maf-inducing protein	−2.1
Ddx58	Asp-Glu-Ala-Asp (DEAD) box polypeptide 58	−2.1
Fcgr2b	Fc receptor, IgG, low-affinity IIb	−2.6
H2-Eb1	Histocompatibility 2, class II antigen E β	−2.7
Ifih1	Interferon-induced with helicase C domain 1	−2.4
Ifit1	Interferon-induced protein with tetratricopeptide repeats 1	−2.7
Ifit2	Interferon-induced protein with tetratricopeptide repeats 2	−2.2
Ifit3	Interferon-induced protein with tetratricopeptide repeats 3	−2.7
Oas2	2′-5′ Oligoadenylate synthetase 2	−5.5
Oasl2	2′-5′ Oligoadenylate synthetase-like 2	−3.2
Rnf125	Ring finger protein 125	−3.1
Rsad2	Radical S-adenosyl methionine domain-containing 2	−3.4
Samhd1	SAM domain and HD domain 1	−2.0
Tpd52	Tumor protein D52	−2.2
Trim25	Tripartite motif-containing 25	−2.4
Cytokines with chemokine activity
Adamts9	A disintegrin-like and metallopeptidase (reprolysin-type) with thrombospondin type 1 motif, 9	−5.0
Ccl8	Chemokine (C–C motif) ligand 8	−4.1
Ccl11	Chemokine (C–C motif) ligand 11	−3.2
Cfb	Complement factor B	−2.1
Cxcl1	Chemokine (C–X–C motif) ligand 1	−2.7
Cxcl9	Chemokine (C–X–C motif) ligand 9	−8.0
Cxcl13	Chemokine (C–X–C motif) ligand 13	6.9
Cxcr4	Chemokine (C–X–C motif) receptor 4	−2.9
Egf	Epidermal growth factor	2.1
Entpd1	Ectonucleoside triphosphate diphosphohydrolase 1	−2.1
Hck	Hemopoietic cell kinase	−2.0
Hif1a	Hypoxia-inducible factor 1, alpha subunit	−3.7
Igfbp3	Insulin-like growth factor-binding protein 3	2.7
Igh-VJ558	Immunoglobulin heavy chain (J558 family)	2.1
Il1r1	Interleukin 1 receptor, type I	−3.6
Il13ra1	Interleukin 13 receptor, alpha 1	−2.5
Il33	Interleukin 33	−8.7
Irf7	Interferon regulatory factor 7	−2.6
Lep	Leptin	2.3
Lrrc32	Leucine-rich repeat containing 32	−2.2
Ly86	Lymphocyte antigen 86	−2.8
Lyn	Yamaguchi sarcoma viral (v-yes-1) oncogene homolog	−3.1
Rap1b	RAS-related protein 1b	−2.1
Stat1	Signal transducer and activator of transcription 1	−2.5
Stat2	Signal transducer and activator of transcription 2	−2.9
Thbd	Thrombomodulin	−2.3
Regulation of transcription
Ankrd2	Ankyrin repeat domain 2 (stretch-responsive muscle)	3.5
Bach1	BTB and CNC homology 1	−2.0
Baz1a	Bromodomain adjacent to zinc finger domain 1A	−2.6
Bhlhe40	Basic helix-loop-helix family, member e40	−2.1
C7	Complement component 7	3.0
Ccnl1	Cyclin L1	−2.0
Crem	cAMP-responsive element modulator	−2.1
Ell2	Elongation factor RNA polymerase II 2	−2.0
Erg	Avian erythroblastosis virus E-26 (v-ets) oncogene related	−2.3
Fos	FBJ osteosarcoma oncogene	−5.1
Klf2	Kruppel-like factor 2 (lung)	−3.0
Mafb	v-maf Musculoaponeurotic fibrosarcoma oncogene family, protein B (avian)	−2.8
Mamstr	MEF2-activating motif and SAP domain-containing transcriptional regulator	2.0
Mnda	Myeloid cell nuclear differentiation antigen	−4.5
Nrip1	Nuclear receptor-interacting protein 1	−2.8
Parp14	Poly (ADP-ribose) polymerase family, member 14	−2.4
Per1	Period circadian clock 1	−6.4
Perm1	PPARGC1 and ESRR-induced regulator, muscle 1	2.5
Rcor3	REST corepressor 3	2.2
Rcor3	REST corepressor 3	2.4
Rorc	RAR-related orphan receptor gamma	2.4
Rxrg	Retinoid X receptor gamma	3.2
Tigd4	Tigger transposable element-derived 4	2.3
Ion binding and transport
Ano5	Anoctamin 5	2.4
Apobec2	Apolipoprotein B mRNA editing enzyme, catalytic polypeptide 2	2.3
Cacna1s	Calcium channel, voltage-dependent, L type, alpha 1S subunit	2.0
Car14	Carbonic anhydrase 14	6.6
Cd93	CD93 antigen	−2.5
Cdh5	Cadherin 5	−3.7
Clcn1	Chloride channel 1	2.6
Csrp3	Cysteine and glycine-rich protein 3	2.4
Cyp27a1	Cytochrome P450, family 27, subfamily a, polypeptide 1	2.1
Dtna	Dystrobrevin alpha	2.0
Fmo2	Flavin containing monooxygenase 2	2.1
Gabrr2	Gamma-aminobutyric acid (GABA) C receptor, subunit rho 2	2.1
Kcng4	Potassium voltage-gated channel, subfamily G, member 4	2.7
Kcnj2	Potassium inwardly rectifying channel, subfamily J, member 2	3.5
Kcnj12	Potassium inwardly rectifying channel, subfamily J, member 12	2.7
Mpi	Mannose phosphate isomerase	2.3
Mrc1	Mannose receptor, C type 1	−3.2
Parp12	Poly (ADP-ribose) polymerase family, member 12	−2.2
Pgm1	Phosphoglucomutase 1	−2.1
Plod2	Procollagen lysine, 2-oxoglutarate 5-dioxygenase 2	−2.4
Ppp2r3a	Protein phosphatase 2, regulatory subunit B, alpha	2.1
Prkd2	Protein kinase D2	−2.4
Rnf213	Ring finger protein 213	−3.4
S100a11	S100 calcium-binding protein A11 (calgizzarin)	−2.5
Slc10a6	Solute carrier family 10 (sodium/bile acid cotransporter family), member 6	−3.3
Slc16a3	Solute carrier family 16 (monocarboxylic acid transporters), member 3	2.4
Slc25a25	Solute carrier family 25 (mitochondrial carrier, phosphate carrier), member 25	−3.5
Slc30a2	Solute carrier family 30 (zinc transporter), member 2	2.6
Slc38a3	Solute carrier family 38, member 3	2.0
Slc38a4	Solute carrier family 38, member 4	2.4
Sln	Sarcolipin	3.8
Trim54	Tripartite motif-containing 54	2.1
Zfp366	Zinc finger protein 366	−2.0
Zswim6	Zinc finger SWIM-type containing 6	−2.3
GTPase activity
Arhgap20	Rho GTPase-activating protein 20	4.1
Gbp2	Guanylate-binding protein 2	−2.2
Gbp3	Guanylate-binding protein 3	−2.3
Gbp7	Guanylate-binding protein 7	−2.3
Gbp8	Guanylate-binding protein 8	−2.3
Gbp9	Guanylate-binding protein 9	−2.7
Gbp10	Guanylate-binding protein 10	−2.9
Gbp11	Guanylate-binding protein 11	−2.1
Gm12185	Predicted gene 12185	−2.5
Gvin1	GTPase, very large interferon-inducible 1	−2.3
Rac2	RAS-related C3 botulinum substrate 2	−2.1
Rapgef5	Rap guanine nucleotide-exchange factor (GEF) 5	−2.9
Rgs5	Regulator of G-protein signaling 5	−2.7
Tgtp2	T Cell-specific GTPase 2	−3.2
Apoptosis and proteolysis/regulation of apoptosis
Agt	Angiotensinogen (serpin peptidase inhibitor, clade A, member 8)	2.1
Aldh1a1	Aldehyde dehydrogenase family 1, subfamily A1	5.9
Asb2	Ankyrin repeat and SOCS box-containing 2	2.3
Asb11	Ankyrin repeat and SOCS box-containing 11	2.2
Asb12	Ankyrin repeat and SOCS box-containing 12	2.1
Atp1a2	ATPase, Na+/K+ transporting, alpha 2 polypeptide	2.1
Bcl3	B Cell leukemia/lymphoma 3	3.2
Ctla2a	Cytotoxic T lymphocyte-associated protein 2 alpha	−2.4
Ctla2b	Cytotoxic T lymphocyte-associated protein 2β	−3.1
Dapk2	Death-associated protein kinase 2	2.2
Ddit4	DNA damage-inducible transcript 4	−5.2
Dpep1	Dipeptidase 1 (renal)	2.5
Egfbp2	Epidermal growth factor binding protein type B	4.0
Egln3	Egl-9 family hypoxia-inducible factor 3	2.1
Eif2ak2	Eukaryotic translation initiation factor 2-alpha kinase 2	−2.6
Errfi1	ERBB receptor feedback inhibitor 1	−11.5
Fbxo32	F-Box protein 32	2.7
Fcer1g	Fc receptor, IgE, high affinity I, gamma polypeptide	−2.0
Fcgr1	Fc receptor, IgG, high affinity I	−6.1
Gadd45a	Growth arrest and DNA-damage-inducible 45 alpha	−2.0
Gadd45g	Growth arrest and DNA-damage-inducible 45 gamma	−10.0
Herc3	Hect domain and RLD 3	2.0
Hmox1	Heme oxygenase (decycling) 1	−4.4
Ifi204	Interferon-activated gene 204	−3.8
Mmp3	Matrix metallopeptidase 3	−3.6
Myc	Myelocytomatosis oncogene	−2.5
Nr4a1	Nuclear receptor subfamily 4, group A, member 1	−5.3
Pdia3	Protein disulfide isomerase-associated 3	−2.3
Pnp	Purine-nucleoside phosphorylase	−2.7
Sh3rf2	SH3 domain-containing ring finger 2	3.5
Smtnl1	Smoothelin-like 1	2.7
Stk17b	Serine/threonine kinase 17b (apoptosis inducing)	−2.5
Tsc22d3	TSC22 domain family, member 3	−3.4
Usp18	Ubiquitin-specific peptidase 18	−2.8
Xaf1	X-Linked inhibitor of apoptosis (XIAP)-associated factor 1	−2.9
Drug metabolism
Gsta3	Glutathione S-transferase, alpha 3	2.6
Gstk1	Glutathione S-transferase kappa 1	2.1
Mgst3	Microsomal glutathione S-transferase 3	2.1
Cytoskeleton
Arpc3	Actin-related protein 2/3 complex, subunit 3	−2.1
Cnn2	Calponin 2	−2.5
Fhl3	Four and one-half LIM domains 3	2.6
Msn	Moesin	−2.2
Myo5c	Myosin VC	2.2
Myoz1	Myozenin 1	2.2
Ptpn4	Protein tyrosine phosphatase, nonreceptor type 4	2.2
Tagln2	Transgelin 2	−2.1
Tpm4	Tropomyosin 4	−2.2
Ttll7	Tubulin tyrosine ligase-like family, member 7	2.6
Cell adhesion
Col15a1	Collagen, type XV, alpha 1	−2.5
Cytip	Cytohesin 1-interacting protein	−2.6
Lgals3bp	Lectin, galactoside-binding, soluble, 3 binding protein	−2.1
Vcan	Versican	−2.6
Behavioral response to stimulus
Amot	Angiomotin	2.1
Amot	Angiomotin	2.3
Sncg	Synuclein, gamma	2.3
Cell migration and motility
Bin2	Bridging integrator 2	−2.1
Ctgf	Connective tissue growth factor	−2.3
Dnaja1	DnaJ (Hsp40) homolog, subfamily A, member 1	−3.0
Itga1	Integrin alpha 1	−2.7
Itga6	Integrin alpha 6	−2.5
Myh9	Myosin, heavy polypeptide 9, nonmuscle	−2.3
Nus1	Nuclear undecaprenyl pyrophosphate synthase 1 homolog (S. cerevisiae)	−2.3
Ret	Ret proto-oncogene	−2.7
Rhob	Ras homolog gene family, member B	−2.0
S1pr1	Sphingosine-1-phosphate receptor 1	−3.2
Sema6c	Sema domain, transmembrane domain (TM), and cytoplasmic domain (semaphorin), 6C	2.2
Alternative splicing
BC094916	cDNA sequence BC094916	−3.3
Coq10b	Coenzyme Q10 homolog B (S. cerevisiae)	−2.9
Far1	Fatty acyl CoA reductase 1	−2.1
Htra3	HtrA serine peptidase 3	2.2
Ifi203	Interferon-activated gene 203	−2.8
Lgals9	Lectin, galactose-binding, soluble 9	−3.0
Lpcat2	Lysophosphatidylcholine acyltransferase 2	−2.0
Midn	Midnolin	−2.3
Mgat4a	Mannoside acetylglucosaminyltransferase 4, isoenzyme A	−2.1
Nnat	Neuronatin	3.2
Picalm	Phosphatidylinositol-binding clathrin assembly protein	−2.3
Pla2g4e	Phospholipase A2, group IVE	2.2
Ptbp3	Polypyrimidine tract-binding protein 3	−2.2
Rtp4	Receptor transporter protein 4	−2.5
Sp100	Nuclear antigen Sp100	−2.5
Trib1	Tribbles homolog 1 (Drosophila)	−2.2
Trp53inp2	Transformation-related protein 53-inducible nuclear protein 2	2.2
Zbp1	Z-DNA-binding protein 1	−2.9
Transmembrane
Abcb4	ATP-binding cassette, subfamily B (MDR/TAP), member 4	2.3
Alox5ap	Arachidonate 5-lipoxygenase activating protein	−3.8
Aqp4	Aquaporin 4	2.5
B3galt2	UDP-Gal: βGlcNAc β 1,3-galactosyltransferase, polypeptide 2	2.6
B4galt5	UDP-Gal: βGlcNAc β 1,4-galactosyltransferase, polypeptide 5	−4.6
Bst2	Bone marrow stromal cell antigen 2	−5.3
Cd53	CD53 antigen	−2.7
Chsy1	Chondroitin sulfate synthase 1	−2.2
Cpt1b	Carnitine palmitoyltransferase 1b, muscle	2.1
Ctxn3	Cortexin 3	3.6
Emp1	Epithelial membrane protein 1	−2.7
Gpr65	G-Protein-coupled receptor 65	−2.4
Gprc5a	G-Protein-coupled receptor, family C, group 5, member A	−2.1
Has1	Hyaluronan synthase1	−2.2
Ier3	Immediate early response 3	−2.7
Ifitm3	Interferon-induced transmembrane protein 3	−2.3
Mc5r	Melanocortin 5 receptor	3.0
Ms4a4b	Membrane-spanning 4-domains, subfamily A, member 4B	−2.3
Ms4a4c	Membrane-spanning 4-domains, subfamily A, member 4C	−7.6
Ms4a6b	Membrane-spanning 4-domains, subfamily A, member 6B	−2.6
Ms4a6c	Membrane-spanning 4-domains, subfamily A, member 6C	−2.5
Ms4a6d	Membrane-spanning 4-domains, subfamily A, member 6D	−3.5
Slc2a4	Solute carrier family 2 (facilitated glucose transporter), member 4	3.0
Slc7a8	Solute carrier family 7 (cationic amino acid transporter, y+ system), member 8	−2.7
Sptlc2	Serine palmitoyltransferase, long-chain base subunit 2	−2.1
St3gal5	ST3 β-galactoside alpha-2,3-sialyltransferase 5	2.2
Tecr	Trans-2,3-enoyl-CoA reductase	2.1
Tyrobp	TYRO protein tyrosine kinase binding protein	−2.1
Unknown or specific functions
1810011O10Rik	RIKEN cDNA 1810011O10 gene	−2.2
2310002L09Rik	RIKEN cDNA 2310002L09 gene	2.3
8430408G22Rik	RIKEN cDNA 8430408G22 gene	−7.3
9430020K01Rik	RIKEN cDNA 9430020K01 gene	−2.1
A730049H05Rik	RIKEN cDNA A730049H05 gene	−2.1
AA467197	Expressed sequence AA467197	−2.1
AI607873	Expressed sequence AI607873	−3.5
Aldh1a7	Aldehyde dehydrogenase family 1, subfamily A7	2.9
Apold1	Apolipoprotein L domain-containing 1	−9.8
Arrdc2	Arrestin domain-containing 2	−2.5
Cox7b2	Cytochrome c oxidase subunit VIIb2	−2.2
Dcaf4	DDB1 and CUL4-associated factor 4	2.1
Dennd4a	DENN/MADD-containing 4A	−2.3
Dusp1	Dual-specificity phosphatase 1	−3.7
Dusp10	Dual-specificity phosphatase 10	2.9
Eif1a	Eukaryotic translation initiation factor 1A	−2.2
Erdr1	Erythroid differentiation regulator 1	−10.0
Exd2	Exonuclease 3–5 domain-containing 2	2.7
Fam46a	Family with sequence similarity 46, member A	−2.0
Fam101b	Family with sequence similarity 101, member B	−2.3
Fcrls	Fc receptor-like S, scavenger receptor	−3.0
Gbas	Glioblastoma-amplified sequence	2.1
Gm22	Predicted gene, 22	−2.2
Gm4899	Predicted gene, 4899	−2.4
Gm4955	Predicted gene, 4955	−3.4
Gm6377	Predicted gene, 6377	−2.5
Gm6904	Predicted gene, 6904	−2.1
Gm10110	Predicted gene, 10110	2.2
Gm10999	Predicted gene, 10999	2.0
Gm11037	Predicted gene, 11037	−2.4
Gm11710	Predicted gene, 11710	−2.0
Gm11711	Predicted gene, 11711	−2.0
Gm21742	Predicted gene, 21742	−9.7
Gm21748	Predicted gene, 21748	−56.6
Gm21750	Predicted gene, 21750	2.1
Gm21857	Predicted gene, 21857	−4.2
Gm21860	Predicted gene, 21860	−56.6
Gm21887	Predicted gene, 21887	−9.3
Gys1	Glycogen synthase 1, muscle	2.1
Herc6	Hect domain and RLD 6	−2.6
I830012O16Rik	RIKEN cDNA I830012O16 gene	−3.5
Ifi44	Interferon-induced protein 44	−6.0
Ifi202b	Interferon-activated gene 202B	−3.4
Inmt	Indolethylamine N-methyltransferase	2.9
Kbtbd13	Kelch repeat and BTB (POZ) domain containing 13	2.1
Klhdc3	Kelch domain-containing 3	2.2
Klhl38	Kelch-like 38	2.6
Lrrc14b	Leucine-rich repeat containing 14B	2.2
Lsmem2	Leucine-rich single-pass membrane protein 2	2.0
Mettl7a1	Methyltransferase-like 7A1	2.4
Mettl21e	Methyltransferase-like 21E	6.6
Mndal	Myeloid nuclear differentiation antigen like	−2.2
Mup3	Major urinary protein 3	2.4
Myom3	Myomesin family, member 3	2.3
Odf3l2	Outer dense fiber of sperm tails 3-like 2	5.3
Pde4b	Phosphodiesterase 4B, cAMP specific	−2.1
Phf11b	PHD finger protein 11B	−3.1
Phf11d	PHD finger protein 11D	−2.3
Plac8	Placenta-specific 8	−2.5
Pnp2	Purine-nucleoside phosphorylase 2	−2.7
Ppp1r3c	Protein phosphatase 1, regulatory (inhibitor) subunit 3C	3.0
Prob1	Proline-rich basic protein 1	2.1
Ptma	Prothymosin alpha	−2.4
Pydc3	Pyrin domain-containing 3	−3.0
Pydc4	Pyrin domain-containing 4	−3.3
Retnlg	Resistin-like gamma	−5.7
Rilpl1	Rab-interacting lysosomal protein-like 1	2.2
Rragd	Ras-related GTP-binding D	2.1
Sbk2	SH3-binding domain kinase family, member 2	2.6
Serpine1	Serine (or cysteine) peptidase inhibitor, clade E, member 1	−9.6
Sh3pxd2b	SH3 and PX domains 2B	−2.0
Slfn1	Schlafen 1	−3.2
Slfn2	Schlafen 2	−3.7
Slfn4	Schlafen 4	−9.7
Slfn5	Schlafen 5	−2.3
Slfn8	Schlafen 8	−2.7
Tpm3-rs7	Tropomyosin 3, related sequence 7	−2.6
Tra2a	Transformer 2 alpha homolog (Drosophila)	−2.5
Trim30a	Tripartite motif-containing 30A	−3.9
Trim30b	Tripartite motif-containing 30B	−3.6
Trim30c	Tripartite motif-containing 30C	−3.4
Trim30d	Tripartite motif-containing 30D	−3.8
Wee1	WEE 1 homolog 1 (Schizosaccharomyces pombe)	−2.1
Xlr3b	X-Linked lymphocyte-regulated 3B	2.4
Xlr3c	X-Linked lymphocyte-regulated 3C	2.2

This list excludes the gene-expression changes also conserved with either LS or EL treatment, listed in Table 2. Genes were functionally clustered using the Functional Annotation clustering tool from the Database for Annotation, Visualization and Integrated Discovery (DAVID). Positive fold changes indicate that gene expression was increased with treatment and vice versa.

Fig. 4.

Short-term prednisolone treatment worsens ongoing muscle damage in dystrophic het mice. Immunofluorescence staining for serum IgG on representative quadriceps (Quad) sections from het mice treated for 2 wk with LS, EL, or prednisolone (Pred) and used for microarray analysis. Mice treated with 1.5 mg·kg⁻¹·day⁻¹ prednisolone (n = 3) had more ongoing damage than LS-treated (n = 3), EL-treated (n = 3), and untreated het mice (n = 3).

DISCUSSION

To determine whether the improvement in ongoing damage previously observed in dystrophic mice treated with LS was a direct or indirect effect of the drugs on striated muscle fiber sarcolemmal membranes, we carried out a laser-injury membrane-integrity assay. We found that het mice have reduced membrane integrity, as predicted by the loss of dystrophin and reduction of utrophin, similar to that seen in dystrophin-deficient mice (34). LS treatment for <4 wk is able to restore membrane integrity of muscles isolated from treated het mice to levels not different from that of wild-type mice. These data indicate that this drug treatment can decrease sarcolemmal membrane fragility through direct effects on skeletal muscle fibers or through modification of the muscle ECM. Since muscle bundles were used in these experiments, the native cellular connections to the matrix remain intact, and matrix characteristics could also contribute to improved sarcolemmal membrane integrity.

This increase in membrane integrity could result from multiple mechanisms. Furthermore, there are additional pathways that could contribute to improved muscle structure and function following treatment with LS. We used an unbiased gene-expression analysis approach to establish the effects of LS on the molecular level. Treatment of dystrophic het mice with EL or LS resulted in a comparable number of gene-expression changes (70 and 66, respectively), and ~50% of these changes were conserved between the two treatments. There were only 10 gene-expression changes conserved between LS and EL compared with untreated, which were not changed in the same direction by prednisolone compared with untreated.

Not included in this group of 10 genes is S100a8, the highest fold decrease in the EL group (160-fold), which was also not changed by prednisolone but required removal of an outlier to reveal a 93-fold change in LS. S100a8 works as a dimer with S100a9 to form calprotectin, a regulator of immune reactions and inflammatory processes and a marker of neutrophils. Previous studies have shown that infusion of IL-6 cytokine increases S100A8 expression in human skeletal muscle (22). IL-6 expression is also decreased approximately threefold in both LS and EL compared with untreated. Over 50% of gene-expression changes resulting from LS or EL treatment were also conserved in prednisolone-treated mice, and the majority of these genes are functionally classified as either cytokine production/immune response or transcriptional regulators. These data support that MR antagonists likely exhibit anti-inflammatory effects similar to prednisolone and that MR and GR may interact with each other both pre- and post-transcriptionally, particularly in inflammatory cells (35, 37).

Two of the 10 gene-expression changes specific to the LS and EL groups have known associations with altered membrane integrity in dystrophic muscles. Metallothioneins (Mt) 1 and 2, which are decreased by LS and EL treatment, are known to be increased in mdx limb muscles and reduced in mdx mice expressing a therapeutic utrophin transgene (2, 25). However, a recent study showed that an Mt1 mimetic in a laser-injury assay is protective in a mouse model of limb-girdle muscular dystrophy (1). Since Mt1 is also regulated by IL-6, different levels of metallothioneins may be protective at different stages of disease.

The remainder of the gene changes conserved between LS and EL, but not prednisolone, have known roles in muscle metabolism. Scd1, which encodes an enzyme essential for skeletal muscle lipid metabolism, showed the highest fold increase with treatment. The increase of Scd1 expression in skeletal muscle can enhance muscle triglyceride synthesis, glucose uptake, and exercise capacity (28). Diminished Scd1 activity can lead to increased levels of saturated fatty acids that result in inflammation, lipotoxicity, and insulin resistance (28).

Nr1d1 encodes the rev-ERBα nuclear receptor expressed in oxidative muscle and is known to increase exercise capacity (36). Pim1 encodes a serine/threonine kinase that regulates metabolism and is associated with high intramuscular fat in beef, and knockout of Clu, encoding clusterin in mice, is protective against a high-fat diet (15, 30). Prkag3 encodes the regulatory subunit of AMP kinase, which is known to play a major role in muscle metabolism and also to regulate the muscle circadian clock (33). Dbp, decreased by prednisolone and in muscle atrophy but increased by both LS and EL treatment, and Per3, increased in LS only, are circadian clock-controlled genes in muscle (24, 33). Clock genes have recently been shown to play an important purpose in normal muscle function and metabolism (12). These metabolic gene-expression changes may underlie the improved recovery from fatigue observed in diaphragm and extensor digitorum longus muscles from LS- and EL-treated mice, although the contribution of each gene product will require further investigation.

Two gene-expression changes were conserved between the het mice treated with LS or EL for 2 wk and mice treated for 16 wk with LS from our previous study compared with untreated hets (8). Cited4, which was decreased by 2-wk EL and 16-wk LS treatment, encodes a cAMP response element-binding protein-binding protein/p300 transcriptional coactivator-binding protein that has not been directly investigated in skeletal muscle. However, Cited4 was increased in exercised-induced cardiac hypertrophy and controls myocyte elongation and proliferation (4, 29). Adamst1, decreased by 2- and 16-wk LS treatment, EL treatment, and prednisolone treatment, is an anti-angiogenic regulator transiently increased during myotube maturation. Adamst1 is known to be drastically reduced in skeletal muscle by exercise and predicted to be involved in ECM remodeling (18).

There were 312 gene-expression changes present only in the prednisolone versus untreated het comparison. Functional groups specific to prednisolone treatment include genes involved in apoptosis or proteolysis, cytokines with chemokine activity, GTPase activity, cell migration and motility, cell adhesion, and cytoskeleton. Almost all apoptotic and proteolytic genes are increased with prednisolone treatment, whereas genes that negatively regulate apoptosis are all decreased. GR-induced increases in proapoptotic gene expression are seen in other tissues, including skin, bone, and neurons (5, 10, 23). The reduction of genes involved in cell motility and migration by prednisolone could contribute to impaired regeneration and explain the increased muscle damage associated with prednisolone treatment in animal models of DMD (11). Treatment with prednisolone also decreased expression of GTPase genes, including several members of the Gbp family that are induced by IFN-γ and play a key role in protective immunity (32).

This study has demonstrated that short-term treatment with LS is able to restore membrane integrity of isolated dystrophic mouse muscles and decrease inflammatory genes. The combination of these effects could contribute to the cell and molecular mechanisms that lead to the improved muscle structure and function in dystrophic muscle following treatment with LS. Microarray analysis identified several potential gene targets that may specifically underlie the efficacy of RAAS inhibitors on dystrophic muscles, while also identifying gene-expression changes that may contribute to the side effects of prednisolone on skeletal muscles. This information may be used to develop biomarkers to monitor RAAS inhibitor or glucocorticoid agonist treatment, develop a temporally coordinated treatment with both drug classes, or identify novel therapeutics with these more-specific treatment targets.

GRANTS

Funding for this study was provided by the National Institute of Neurological Disorders and Stroke Grant R01 NS082868) and National Heart, Lung, and Blood Institute Grant R01 HL116533 (to J. A. Rafael-Fortney), National Institute of Arthritis and Musculoskeletal and Skin Diseases Grant R01 AR063084 (to N. Weisleder), and Department of Defense Grant MD120063 (to J. A. Rafael-Fortney). Additional support was provided through a postdoctoral fellowship from The Ohio State University/Nationwide Children’s Hospital Center for Muscle Health and Neuromuscular Disorders (to S. Bhattacharya). Eplerenone (no longer under patent) was provided by the Pfizer Compound Transfer Program.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

J.A.C., S.B., and J.L. performed experiments; J.A.C., S.B., and J.A.R-F. analyzed data; J.A.C., N.W., and J.A.R-F. interpreted results of experiments; J.A.C. and S.B. prepared figures; J.A.R-F. drafted manuscript; J.A.C., J.L., N.W., and J.A.R-F. edited and revised manuscript; J.A.C., S.B., J.L., N.W., and J.A.R-F. approved final version of manuscript.