SIRT1 deacetylase in aging-induced neuromuscular degeneration and amyotrophic lateral sclerosis

Abstract

SIRT1 is an NAD⁺ -dependent deacetylase that functions in a variety of cells and tissues to mitigate age-associated diseases. However, it remains unknown if SIRT1 also acts to prevent pathological changes that accrue in motor neurons during aging and amyotrophic lateral sclerosis (ALS). In this study, we show that SIRT1 expression decreases in the spinal cord of wild-type mice during normal aging. Using mouse models either overexpressing or lacking SIRT1 in motor neurons, we found that SIRT1 slows age-related degeneration of motor neurons' presynaptic sites at neuromuscular junctions (NMJs). Transcriptional analysis of spinal cord shows an overlap of greater than 90% when comparing alterations during normal aging with changes during ALS, revealing a substantial upregulation in immune and inflammatory response genes and a downregulation of synaptic transcripts. In addition, overexpressing SIRT1 in motor neurons delays progression to end-stage disease in high copy SOD1^G93A mice. Thus, our findings suggest that there are parallels between ALS and aging, and interventions to impede aging may also slow the progression of this devastating disease.

Keywords: NAD; SIRT1; aging; amyotrophic lateral sclerosis; neurodegenerative disease; neuromuscular junction.

Figure 1

SIRT1 levels in mouse spinal cord during normal aging and neuromuscular disease. (a) Western blot of spinal cord homogenate from postnatal C57Bl6 mice ranging in age from one week to over two years. One female mouse and one male mouse were assayed at each timepoint. (b) Densitometry was performed to compare SIRT1 levels after normalization to GAPDH at different ages. (c–f) Immunoblots of central nervous system and metabolic tissues from three male SOD1^G93A mice and three litter‐ and gender‐matched wild‐type mice were analyzed at 5.5 months of age. (g) Densitometry was performed to compare SIRT1 levels after normalization to GAPDH between wild‐type and SOD1^G93A mice. Unpaired t ‐tests were used to compare levels of SIRT1 normalized to GAPDH in wild‐type and SOD1^G93A animals for each tissue analyzed. Values reported as mean ± SD.

Figure 2

SIRT1 protects the neuromuscular junction during normal aging. (a) Representative images of motor axons labeled with synaptotagmin 2 (Syt2; Green) and nicotinic acetylcholine receptors (AChR; Red) marked with α‐bungarotoxin at NMJs from young mice analyzed at 3–4 months of age. (b) Full innervation, partial innervation, and denervation of neuromuscular junctions from four male motor neuron knockout mice and three litter‐ and gender‐matched control mice at 3–4 months of age. (c) Full innervation, partial innervation, and denervation of neuromuscular junctions from five male motor neuron transgenic mice and five litter‐ and gender‐matched control mice at 3–4 months of age. (d) Representative images of motor axons labeled with synaptotagmin 2 (Syt2; Green) and nicotinic acetylcholine receptors (AChR; Red) marked with α‐bungarotoxin at NMJs from aged mice analyzed at 18–24 months of age. White arrows indicate partially innervated NMJs. Scale bar = 20 μm. (e) Full innervation, partial innervation, and denervation of neuromuscular junctions from four male motor neuron knockout mice and four litter‐ and gender‐matched control mice at 18–24 months (average: 21.2 months, range: 19–24 months) of age. (f) Full innervation, partial innervation, and denervation of neuromuscular junctions from five male motor neuron transgenic mice and five litter‐ and gender‐matched control mice at 18–24 months (average: 21.9 months, range 21–24 months) of age.

Figure 3

Similar transcriptional alterations occur during aging and amyotrophic lateral sclerosis. (a) Heatmap comparing differentially expressed protein‐coding transcripts (absolute log fold change >1 and adjusted p‐value <0.05) in whole spinal cord from young wild‐type compared with old wild‐type and symptomatic SOD1^G93A mice compared with wild‐type controls. Dataset includes eight young male mice at 3–4 months and eight old male mice at 18–24 months to examine aging‐related changes. Whole spinal cords from three SOD1^G93A mice (two female and one male) and three wild‐type control mice (two female and one male) at 5 months were used to examine changes due to ALS. Red signal indicates high relative expression, and blue signal indicates low relative expression. (b) Venn diagram illustrating the overlap between transcripts upregulated (log fold change >1 and adjusted p‐value <0.05) in aged spinal cords compared with young controls and ALS spinal cords compared with healthy controls. (c) Major representative pathways upregulated in aging and disease cohorts. (d) Major representative pathways downregulated in aging and disease cohorts. (e) Upregulated genes in aging and ALS with highest fold change are compared with ImmGen transcriptional profiles for major classes of immune cells. The means‐normalized expression value of each gene in different immune cell populations is shown as a scatter plot.

Figure 4

SIRT1 overexpression in motor neurons is protective in the SOD G93A model of amyotrophic lateral sclerosis. (a) Representative images of NMJ clusters in symptomatic SOD1^G93A and SOD1^G93A mice that overexpress SIRT1 in motor neurons. White arrows point to denervated NMJs. Scale bar = 20 μm. (b) Quantification of innervated NMJs from wild‐type (four male), motor neuron transgenic (three male), SOD1^G93A (three male, one female), and SOD1^G93A mice that overexpress SIRT1 (three male, one female). Unpaired t tests were used to compare innervation in litter‐matched SOD1^G93A and SOD1^G93A mice that overexpress SIRT1. Values reported as mean ± SD. (c) Quantification of denervated NMJs from wild‐type (four male), motor neuron transgenic (three male), SOD1^G93A (three male, one female), and SOD1^G93A mice (three male, one female) that overexpress SIRT1. Unpaired t‐tests were used to compare denervation in litter‐matched SOD1^G93A and SOD1^G93A mice that overexpress SIRT1. Values reported as mean ± SD. (d) NMJs were characterized as innervated, partially innervated, or denervated depending on the degree of overlay between the motor axon and AChR clusters and quantification of NMJ innervation. (e) Kaplan–Meier survival curve comparing SOD1^G93A with litter‐ and gender‐matched motor neuron transgenic SOD1^G93A mice that overexpress SIRT1 in motor neurons (16 male SOD1^G93A mice; 15 male MNTg SOD1^G93A; 22 female SOD1^G93A mice; 16 female MNTg SOD1^G93A mice). *p < 0.05, log‐rank (Mantel–Cox) test. Comparison of synaptic signaling pathways in (f) motor neuron transgenic SOD1^G93A relative to SOD1^G93A mice (g) SOD1^G93A relative to wild‐type mice and (h) old relative to young mice.