The Smn-independent beneficial effects of trichostatin A on an intermediate mouse model of spinal muscular atrophy

Abstract

Spinal muscular atrophy is an autosomal recessive neuromuscular disease characterized by the progressive loss of alpha motor neurons in the spinal cord. Trichostatin A (TSA) is a histone deacetylase inhibitor with beneficial effects in spinal muscular atrophy mouse models that carry the human SMN2 transgene. It is currently unclear whether TSA specifically targets the SMN2 gene or whether other genes respond to TSA and in turn provide neuroprotection in SMA mice. We have taken advantage of the Smn2B/- mouse model that does not harbor the human SMN2 transgene, to test the hypothesis that TSA has its beneficial effects through a non-SMN mediated pathway. TSA increased the median lifespan of Smn2B/- mice from twenty days to eight weeks. As well, there was a significant attenuation of weight loss and improved motor behavior. Pen test and righting reflex both showed significant improvement, and motor neurons in the spinal cord of Smn2B/- mice were protected from degeneration. Both the size and maturity of neuromuscular junctions were significantly improved in TSA treated Smn2B/- mice. Of interest, TSA treatment did not increase the levels of Smn protein in mouse embryonic fibroblasts or myoblasts obtained from the Smn2B/- mice. In addition, no change in the level of Smn transcripts or protein in the brain or spinal cord of TSA-treated SMA model mice was observed. Furthermore, TSA did not increase Smn protein levels in the hind limb muscle, heart, or liver of Smn2B/- mice. We therefore conclude that TSA likely exerts its effects independent of the endogenous mouse Smn gene. As such, identification of the pathways regulated by TSA in the Smn2B/- mice could lead to the development of novel therapeutics for treating SMA.

Figure 1. TSA treatment promotes survival, attenuates weight loss, and enhances motor function in Smn^2B/- mice.

(A) Kaplan-Meier Survival curves of TSA or vehicle (DMSO) treated Smn^2B/- mice and littermate controls. Median lifespan is 8 weeks in TSA treated Smn^2B/- mice (black arrow) in comparison to 23 days in DMSO-treated (black arrow). n = 9 (Smn^2B/- DMSO), 7 (Smn^2B/- TSA), 5 (Smn^2B/+ DMSO), 4 (Smn^2B/+ TSA). (B) Growth curves of TSA or vehicle (DMSO) treated Smn^2B/- mice and littermate controls from P13 to P25. DMSO-treated Smn^2B/- mice gradually lost weight during this time period. n = 16 (Smn^2B/- DMSO), 15 (Smn^2B/- TSA), 11(Smn^2B/+ DMSO), 14 (Smn^2B/+ TSA). (C) Righting reflex test. Mild improvement observed in the righting ability of TSA-treated Smn^2B/- mice from P15 onwards. n = 16 (Smn^2B/- DMSO) and 15 (Smn^2B/- TSA). (D) Pen test assay on Smn^2B/- mice treated with TSA or DMSO. Thirty seconds was set as the maximum time period. Improvement in latency to fall in TSA treated SMA mice (red line) became more apparent after P21 in comparison to DMSO treated SMA mice (black line). n = 16 (Smn^2B/- DMSO) and 15 (Smn^2B/- TSA).

Figure 2. TSA treatment reduces motor neuron loss in the spinal cord and improves myofiber histology in Smn^2B/- mice.

(A) Cross-sections through the L1 lumbar spinal cord from Smn^2B/+ heterozygous control mice, and Smn^2B/- mice treated with vehicle (DMSO) or TSA were processed for haematoxylin/eosin staining. Scale bar = 200 µm. (B) Cross-sections through the tibialis anterior (TA) muscles from Smn^2B/+ heterozygous control mice, and Smn^2B/- mice treated with DMSO or TSA were processed for haematoxylin/eosin staining. Scale bar = 200 µm. (C) Quantification of motor neuron cell body number in the spinal cord from TSA or vehicle treated mice. TSA treatment results in a significant increase in the number of motor neurons in the spinal cord of Smn^2B/- mice compared with DMSO treated Smn^2B/- mice. n = 3 (Smn^2B/- DMSO) and 4 (Smn^2B/- TSA).

Figure 3. TSA administration decreases neuromuscular junction loss in Smn^2B/- mice.

(A) Confocal micrographs show neuromuscular junctions (NMJs) labelled with antibodies against neurofilament (NF, red), synaptic vesicle protein 2 (SV2, red) and alpha-bungarotoxin (BTX, green). Note that whilst there is widespread NMJ pathology in vehicle treated mice, as evidence by full/partial denervation (white arrowhead/arrow respectively), NF accumulation (yellow arrow) and small, immature post-synaptic endplates, NMJs appear much more healthy in TSA treated mice, as evidenced by a decrease in denervation and NF accumulation, and an increase in post-synaptic size and maturity (scale bar = 50 µm). (B, C) Bar charts show an increase in the percentage of fully occupied endplates in both the transversus abdominis (TVA; B) and rectus abdominis (RA; C) muscles following TSA treatment. * P<0.05 by Mann Whitney test.

Figure 4. No alteration in the splicing pattern or levels of Smn mRNA in the brains of TSA-treated mice.

(A) RT-PCR for Smn and a known target of TSA, follistatin, was conducted in total RNA extracted from brains of Smn^2B/- mice or littermate controls treated with TSA or DMSO vehicle. Ratio of full length Smn versus Δ7Smn transcripts remained similar after TSA treatment in Smn^2B/- or in control mice. Transcript level of follistatin is elevated in TSA-treated mice. Actin serves as a loading control. (B) Real-time quantitative RT-PCR analyses of RNA from spinal cord and brain. No significant change in the level of Smn transcripts (full length Smn plus Δ7Smn) was revealed after TSA treatment (n = 3). Fold increase normalized to reference gene Gapdh.

Figure 5. TSA treatment does not increase Smn protein levels.

(A) Western blot analyses were performed with specific antibodies against Smn in protein extracts from brains, spinal cords, hind limb muscle, hearts, or livers of Smn^2B/- mice treated with either TSA or DMSO vehicle. Each lane represents one animal. β-tubulin or actin serve as a loading controls. (B) Quantification revealed no significant increase of Smn protein levels in brains, spinal cords, hind limb muscle, hearts, or livers of TSA-treated SMA mice (n = 3).

Figure 6. TSA treatment does not increase Smn protein levels in mouse embryonic fibroblasts and myoblasts obtained from Smn^2B/- mice.

Western blot analyses were performed with specific antibodies against acetylated histone H3 (A) or Smn (B) in protein extracts from TSA (100 nM) or DMSO vehicle treated MEFs obtained from Smn^2B/- mice. Total histone H3 or β-tubulin serves as a loading control, respectively. (C and D) A similar analysis was performed in TSA (100 nM) or DMSO vehicle treated myoblasts obtained from Smn^2B/- mice.