Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Apr 26;15(1):3563.
doi: 10.1038/s41467-024-47846-9.

LSD1 inhibition circumvents glucocorticoid-induced muscle wasting of male mice

Qingshuang Cai  1 , Rajesh Sahu  1 , Vanessa Ueberschlag-Pitiot  1 , Sirine Souali-Crespo  1 , Céline Charvet  1 , Ilyes Silem  1 , Félicie Cottard  1 , Tao Ye  1 , Fatima Taleb  1 , Eric Metzger  2 , Roland Schuele  2 , Isabelle M L Billas  1 , Gilles Laverny  1 , Daniel Metzger  1 , Delphine Duteil  3
Affiliations

LSD1 inhibition circumvents glucocorticoid-induced muscle wasting of male mice

Qingshuang Cai et al. Nat Commun. .

Abstract

Synthetic glucocorticoids (GC), such as dexamethasone, are extensively used to treat chronic inflammation and autoimmune disorders. However, long-term treatments are limited by various side effects, including muscle atrophy. GC activities are mediated by the glucocorticoid receptor (GR), that regulates target gene expression in various tissues in association with cell-specific co-regulators. Here we show that GR and the lysine-specific demethylase 1 (LSD1) interact in myofibers of male mice, and that LSD1 connects GR-bound enhancers with NRF1-associated promoters to stimulate target gene expression. In addition, we unravel that LSD1 demethylase activity is required for triggering starvation- and dexamethasone-induced skeletal muscle proteolysis in collaboration with GR. Importantly, inhibition of LSD1 circumvents muscle wasting induced by pharmacological levels of dexamethasone, without affecting their anti-inflammatory activities. Thus, our findings provide mechanistic insights into the muscle-specific GC activities, and highlight the therapeutic potential of targeting GR co-regulators to limit corticotherapy-induced side effects.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. LSD1 interacts with GR to control target gene expression in mouse myofibers at physiological glucocorticoid levels.
a Representative immunofluorescent detection of LSD1 (green) and GR (red) in gastrocnemius muscles of 9-week-old wild-type mice. Nuclei were stained with DAPI. A zoomed-in view of the confocal observation is shown on the top right panel. Scale bar, 50 µm. N = 3. b Representative western blot analysis of GR and LSD1 co-immunoprecipitation in gastrocnemius muscle nuclear extracts. Rabbit IgG served as a control for immunoprecipitation. N = 3 mice. Note that the observed discrepancies in the molecular weight on membranes decorated with anti-LSD1 antibody originate from the high sensitivity of LSD1 to salt and/or pH composition of the elution buffer. c, d Tag density map of LSD1 and GR in skeletal muscles, +/− 5 kb from the LSD1 (c) or the GR (d) peak center, and corresponding average tag density profiles. e Localization of GR and LSD1 at the Ddit4 locus. The four GR binding sites located at enhancer (GBSe1, GBSe2 and GBSe3) and promoter (GBSp1) regions are boxed in orange. f Two-step chromatin immunoprecipitation performed with indicated antibodies followed by qPCR analysis (ChIP-reChIP-qPCR) in gastrocnemius muscles of wild-type mice at GBSe1, GBSe2, GBSe3 and GBSp1 of Ddit4. N = 6 biological replicates. Mean ± SEM. Two-way ANOVA with Tukey correction. g Heatmap with hierarchical clustering depicting the mean centered normalized expression of genes differentially expressed in RNA-seq in gastrocnemius muscles of 9-week-old Ctrl, GR(i)skm-/- and LSD1skm-/- mice. h, i Pathway analysis of genes down- (h) and up-regulated (i) in gastrocnemius muscles of both GR(i)skm-/- and LSD1skm-/- mice at 9 weeks, with p values adjusted using the Benjamini-Hochberg. Source data are provided as a Source Data file.
Fig. 2
Fig. 2. LSD1 bridges GR at active enhancers and NRF1 at active promoters to control target gene expression.
a Tag density map of LSD1, GR, H3K4me1, H3K4me2 and H3K4me3 in skeletal muscles, +/− 5 kb from the LSD1 peak center, and corresponding average tag density profiles. b, c HOMER de novo motif analysis of LSD1 binding sites located at enhancer (b) or promoter (c) regions. p value: hypergeometric testing. d Representative western blot analysis of NRF1 and LSD1 co-immunoprecipitation in gastrocnemius muscle nuclear extracts. Rabbit IgG served as a control for immunoprecipitation. N = 3 mice. e Representative SDS-PAGE of recombinant LSD1 and NRF1 proteins immunoprecipitated with anti-LSD1 antibodies or rabbit IgG as a control. Input corresponds to 10% of the purified LSD1 and NRF1 proteins. f Representative western blot analysis of GR and NRF1 co-immunoprecipitation in gastrocnemius muscle nuclear extracts from control and LSD1skm-/- mice. Rabbit IgG served as a control for immunoprecipitation. N = 3 mice. g ChIP-qPCR analysis performed with anti-GR and anti-NRF1 antibodies, or rabbit IgG in gastrocnemius muscles of ctrl and LSD1skm-/- mice at the GBSe2 and GBSp1 of Ddit4. The promoter region of Pax7 was used as a negative control. N = 5 biological replicates. Mean ± SEM. One-way ANOVA with Tukey correction. h HOMER de novo motif analysis of NRF1 binding sites. p value: hypergeometric testing. i Tag density map of LSD1, GR, NRF1 and H3K4me3 in skeletal muscles, +/− 5 kb from the LSD1 peak center and corresponding average tag density profiles. j Localization of GR and LSD1 at the Fbxo31 locus. k Overlap among genes bound by LSD1, GR or NRF1 in skeletal muscles. l Pathway analysis on LSD1, GR and NRF1 common target genes in skeletal muscles, with p values adjusted using the Benjamini-Hochberg. m Relative Ddit4 and Fbxo31 transcript levels determined in C2C12 myotubes transfected with siRNA directed against Gr, Nrf1 and Lsd1 (siGR, siNRF1 or siLSD1, respectively), or with a scramble siRNA (siCtrl). N = 7 biologically independent samples. Mean ± SEM. One-way ANOVA with Tukey correction. Source data are provided as a Source Data file.
Fig. 3
Fig. 3. LSD1 is required for starvation-induced muscle atrophy.
a Representative western blot analysis of gastrocnemius muscle nuclear extracts from mice fed or starved for 12 or 24 h, immunoprecipitated with anti-GR or anti-LSD1 antibodies. Rabbit IgG served as a control for immunoprecipitation. N = 3 mice. b Body, gastrocnemius, tibialis, quadriceps, soleus, spleen and epWAT weights of 12-week-old Ctrl and LSD1(i)skm-/- mice fed or starved for 48 h. N = 9 mice. Mean ± SEM. Two-way ANOVA with Tukey correction. c, d Distribution (c) and average CSA (d) of gastrocnemius of 12-week-old Ctrl and LSD1(i)skm-/- mice fed or starved for 48 h. N = 3. Mean ± SEM. Two-way ANOVA with Tukey correction (c), one-way ANOVA with Tukey correction (d). e Maximal (Max) and average (Mean) grip strength of 12-week-old Ctrl and LSD1(i)skm-/- mice fed or starved for 48 h. N = 16 Ctrl fed, 12 Ctrl starved, 18 LSD1(i)skm-/- fed and 11 LSD1(i)skm-/- starved mice. Mean ± SEM. Two-way ANOVA with Tukey correction. f, g Representative western blot analysis (f) and corresponding quantification (g) of the indicated proteins in quadriceps of 12-week-old Ctrl and LSD1(i)skm−/− mice fed or starved for 48 h. α-TUBULIN was used as a loading control. N = 3 mice. Mean ± SEM. One-way ANOVA with Tukey correction. h Relative transcript levels of indicated genes determined in gastrocnemius of 12-week-old Ctrl and LSD1(i)skm-/- mice fed or starved for 48 h. N = 6 Ctrl Fed, 10 Ctrl Starved, 10 LSD1(i)skm-/- Fed and 10 LSD1(i)skm-/- Starved mice. Mean ± SEM. One-way ANOVA with Tukey correction. i Ultrastructure analysis of gastrocnemius muscles of 12-week-old Ctrl and LSD1(i)skm-/- mice fed or starved for 48 h. Mt mitochondria, S sarcoplasm, Z Z line. Black arrow indicates Z line disruption; white arrows indicate loss of myofilaments. This experiment has been performed on a minimum of 5 mice per group. Source data are provided as a Source Data file.
Fig. 4
Fig. 4. The GR/LSD1 complex is required to trigger the expression of genes involved in starvation-induced proteolysis upon food deprivation.
a Tag density map of LSD1, GR, H3K9me1 and H3K9me2 in skeletal muscles of mice fed or fasted for 24 h, +/− 5 kb from the LSD1 peak center and corresponding average tag density profiles. b Localization of LSD1, GR, H3K9me1 and H3K9me2 in skeletal muscles of mice fed or fasted for 24 h at indicated loci. c, d Representative western blot (c) and corresponding quantification (d) of LSD1 and GR protein levels in gastrocnemius muscle extracts from 12-week-old wild type mice fed or food deprived for 12, 24 or 48 h. GAPDH was used as a loading control. N = 2 representative mice (c) out of 7 mice (d). Mean ± SEM. One-way ANOVA with Tukey correction. e ChIP-qPCR analysis performed at indicated loci with anti-GR, anti-LSD1 or anti-H3K9me2 antibodies in skeletal muscle of wild-type mice at 0, 3, 6, 12 and 24 h of food deprivation. The promoter region of Pax7 and an unrelated region within the Ddit4 locus were used as negative controls. N = 3 mice. Mean ± SEM. One-way ANOVA with Tukey correction, with the following annotation on the figure. a: fed vs starved, p < 0.05. b: fed vs starved, p < 0.001. c: starved 3 h vs other time points, p < 0.05. d: starved 3 h vs other time points, p < 0.01. e: starved 3 h vs other time points, p < 0.001. Exact p values are detailed in “Source Data”. Source data are provided as a Source Data file.
Fig. 5
Fig. 5. LSD1 is required for dexamethasone-induced muscle wasting.
a, b Gastrocnemius, tibialis, quadriceps, soleus, spleen and white epWAT mass (a), and maximal (Max) and average (Mean) grip strength (b) of 12-week-old Ctrl and LSD1(i)skm-/- mice treated with dexamethasone (DEX) or a vehicle (Oil) for 72 h. N = 10 Ctrl Oil, 10 Ctrl DEX, 7 LSD1(i)skm-/- Oil and 7 LSD1(i)skm-/- DEX mice. Mean ± SEM. Two-way ANOVA (a) and one-way ANOVA (b) with Tukey correction. c Relative transcript levels of indicated genes in gastrocnemius of Ctrl and LSD1(i)skm-/- mice treated with DEX or Oil for 72 h. N = 4 Ctrl Oil, 5 Ctrl DEX, 6 LSD1(i)skm-/- Oil and 7 LSD1(i)skm-/- DEX mice biological replicates, the individual values of the technical replicates are presented on the graph. Mean ± SEM. One-way ANOVA with Tukey correction. d, e Representative western blot (d) and relative levels of the indicated proteins (e) in quadriceps of control and LSD1(i)skm−/− mice treated with DEX or Oil for 72 h. α-Tubulin was used as a loading control. N = 3 mice. Mean ± SEM. One-way ANOVA with Tukey correction. f Representative western blot analysis (left) and relative levels (right) of LSD1 and GR protein in gastrocnemius from three 12-week-old wild type mice treated with DEX or Oil for 24 or 72 h. GAPDH was used as a loading control. N = 3 mice. Mean ± SEM. Two-tailed t test. g Representative western blot of gastrocnemius from mice treated with DEX or Oil for 72 h immunoprecipitated with anti-GR or anti-LSD1 antibodies. Rabbit IgG served as a control for immunoprecipitation. N = 3 mice. h ChIP-qPCR analysis performed at Trim63 and Ddit4 loci with anti-GR and anti-LSD1 antibodies or rabbit IgG in skeletal muscles of Ctrl and LSD1(i)skm-/- mice treated with DEX or Oil for 72 h. N = 3 mice. Mean ± SEM. One-way ANOVA with Tukey correction. Source data are provided as a Source Data file.
Fig. 6
Fig. 6. Effect of LSD1 inhibition on dexamethasone-treated LHCN-M2 cells.
a Representative immunofluorescent detection of LSD1 (green) and GR (red) in LHCN-M2 cells. Nuclei were stained with DAPI. Scale bar, 50 µm. N = 3 biological replicates. b Representative western blot analysis of nuclear or cytoplasmic extracts of differentiated LHCN-M2 cells, immunoprecipitated with anti-GR, anti-LSD1 or anti-NRF1 antibodies. Rabbit IgG served as a control for immunoprecipitation. c Relative transcript levels of the indicated genes determined in LHCN-M2 myotubes treated with vehicle, DEX, DEX with CC-90011 (DEX + CC) or CC-90011 (CC) for 24 h. N = 6 biological replicates, the 12 individual values of the technical replicates are presented on the graph. Mean ± SEM. One-way ANOVA with Tukey correction. Source data are provided as a Source Data file.
Fig. 7
Fig. 7. LSD1 inhibition prevents dexamethasone-induced muscle wasting without affecting glucocorticoid anti-inflammatory activities.
a Gastrocnemius, tibialis, quadriceps, soleus, spleen and epididymal white adipose tissue (epWAT) mass of 12-week-old wild-type mice treated with a vehicle, DEX, DEX with CC-90011 (DEX + CC), or CC-90011 (CC) for 72 h. N = 6 vehicle, 5 DEX, 5 DEX + CC, 4 CC mice. Mean ± SEM. Two-way ANOVA with Tukey correction. b Average (Mean) and maximal (MAX) grip strength of 72 h vehicle, DEX, DEX + CC or CC treated 12-week-old wild-type mice. N = 6 vehicle, 5 DEX, 5 DEX + CC, 4 CC mice. Mean ± SEM. One-way ANOVA with Tukey correction. c ChIP-qPCR analysis performed at Trim63 and Ddit4 loci with anti-LSD1 and anti-GR antibodies or rabbit IgG in skeletal muscles of wild-type mice treated with vehicle, DEX or DEX + CC for 72 h. N = 3 mice. Mean ± SEM. One-way ANOVA with Tukey correction. d Relative transcript levels of indicated genes determined in gastrocnemius muscles of 72 h vehicle, DEX, DEX + CC or CC treated wild-type mice. N = 5 biological replicates, the 10 individual values of the technical replicates are presented on the graph. Mean ± SEM. One-way ANOVA with Tukey correction. e Blood count analysis of wild-type mice treated with vehicle, DEX, DEX + CC or CC for 72 h. RBC: red blood cells, WBC: white blood cells. N = 6 vehicle, 4 DEX, 4 DEX + CC, 4 CC mice. Mean ± SEM. One-way ANOVA with Tukey correction. f Representative contour plots (right panel) and corresponding quantification (left panel) of IL-17 expression in Th0- or Th17-induced wild-type CD4 + T-cells treated with vehicle, DEX, DEX with CC-90011 (DEX + CC) and CC-90011 (CC). IFNγ was used as a control of Th17 induction. N = 4 mice. Mean ± SEM. One-way ANOVA with Tukey correction. Source data are provided as a Source Data file.
Fig. 8
Fig. 8. LSD1 inhibition does not impair dexamethasone-induced improvement of experimental colitis symptoms.
a–c Water consumption (a), weight change (b) and clinical score (c) of 12-week-old DSS-exposed wild-type mice treated with vehicle (DSS+vehicle), DEX (DSS + DEX), DEX with CC-90011 (DSS + DEX + CC) or CC-90011 (DSS + CC). N = 5 mice per condition. Mean ± SEM. Two-way ANOVA with Tukey correction. d, e Colon representative images (d) and length measurement (e) of mice exposed to indicated treatments. N = 5 mice per condition. Basal values are represented by a dashed line. Mean ± SEM. One-way ANOVA, correction for multiple comparisons. f Relative transcript levels of IL-6 determined in colon. N = 5 mice per condition, all the individual values of the technical replicates are presented on the graph. Basal values are represented by a dashed line. Mean ± SEM. One-way ANOVA, correction for multiple comparisons. g Representative H&E staining of colons of mice from various groups. Red and black arrows denote inflammatory cells and exfoliated epithelial cells, respectively. GC goblet cell, LP lumina propria, SM muscularis mucosae composed of smooth muscle fibers. Scale bars, 100 µm. h Spleen weight of mice from various groups. N = 5 mice per condition. Basal values are represented by a dashed line. Mean ± SEM. Two-way ANOVA with Tukey correction. i Flow cytometry analysis of the lymphoid and the myeloid lineages from spleens of mice from various groups. N = 3 mice per condition. Basal values are represented by a dashed line. Mean ± SEM. Two-way ANOVA with Tukey correction. j Blood count analysis of mice from various groups. RBC: red blood cells. N = 5 mice per condition. Basal values are represented by a dashed line. Mean ± SEM. Two-way ANOVA with Tukey correction. k Gastrocnemius, tibialis, quadriceps and soleus mass relative to body weight (BW) of mice from various groups. N = 5 mice per condition. Basal values are represented by a dashed line. Mean ± SEM. Two-way ANOVA with Tukey correction. Source data are provided as a Source Data file.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Inaba H, Greaves M, Mullighan CG. Acute lymphoblastic leukaemia. Lancet. 2013;381:1943–1955. doi: 10.1016/S0140-6736(12)62187-4. - DOI - PMC - PubMed
    1. Clark AR, Belvisi MG. Maps and legends: the quest for dissociated ligands of the glucocorticoid receptor. Pharmacol. Therapeut. 2012;134:54–67. doi: 10.1016/j.pharmthera.2011.12.004. - DOI - PubMed
    1. Laugesen K, Jorgensen JOL, Petersen I, Sorensen HT. Fifteen-year nationwide trends in systemic glucocorticoid drug use in Denmark. Eur. J. Endocrinol. 2019;181:267–273. doi: 10.1530/EJE-19-0305. - DOI - PubMed
    1. Compston J. Glucocorticoid-induced osteoporosis: an update. Endocrine. 2018;61:7–16. doi: 10.1007/s12020-018-1588-2. - DOI - PMC - PubMed
    1. Nogue M, et al. Long-term corticosteroid use and dietary advice: a qualitative analysis of the difficulties encountered by patient. BMC Health Serv. Res. 2019;19:255. doi: 10.1186/s12913-019-4052-y. - DOI - PMC - PubMed

Publication types

MeSH terms