DNA methylation of SPARC and chronic low back pain

Abstract

Background: The extracellular matrix protein SPARC (Secreted Protein, Acidic, Rich in Cysteine) has been linked to degeneration of the intervertebral discs and chronic low back pain (LBP). In humans, SPARC protein expression is decreased as a function of age and disc degeneration. In mice, inactivation of the SPARC gene results in the development of accelerated age-dependent disc degeneration concurrent with age-dependent behavioral signs of chronic LBP.DNA methylation is the covalent modification of DNA by addition of methyl moieties to cytosines in DNA. DNA methylation plays an important role in programming of gene expression, including in the dynamic regulation of changes in gene expression in response to aging and environmental signals. We tested the hypothesis that DNA methylation down-regulates SPARC expression in chronic LBP in pre-clinical models and in patients with chronic LBP.

Results: Our data shows that aging mice develop anatomical and behavioral signs of disc degeneration and back pain, decreased SPARC expression and increased methylation of the SPARC promoter. In parallel, we show that human subjects with back pain exhibit signs of disc degeneration and increased methylation of the SPARC promoter. Methylation of either the human or mouse SPARC promoter silences its activity in transient transfection assays.

Conclusions: This study provides the first evidence that DNA methylation of a single gene plays a role in chronic pain in humans and animal models. This has important implications for understanding the mechanisms involved in chronic pain and for pain therapy.

Figure 1

Disc degeneration and behavioral signs of low back pain in aging and SPARC-null mice. Anatomy: The disc height index measured in 15-month old mice is smaller than that in 3-month old mice (a). 4-month old SPARC-null mice show smaller average disc height compared to age-matched WT controls (f). Pain Behavior: Aging mice exhibit signs of axial discomfort (b), cold sensitivity (c), but not mechanical sensitivity (d) in the hindpaw, in addition to overall motor impairment (e) when compared to 3-month old mice. 4-month old SPARC-null animals show a behavioral profile similar to older WT mice (g-j). * = p < 0.05, ** = p < 0.01, One-way ANOVA followed by Bonferroni's test (aging cohorts), two-tailed student t-test (SPARC-null vs WT). n = 8-15/group. Error bars = S.E.M.

Figure 2

Changes in expression and DNA methylation of SPARC in aging mice. SPARC mRNA expression (relative to gapdh) at different time points in life (b). Age-dependent changes in methylation of CG sites in the SPARC promoter (a) in IVDs as quantified by pyrosequencing (c). Treatment of 1-year old mice with the demethylating drug 5AC or vehicle (30 mg/kg, i.v. and 250 fmol i.t.) resulted in a 4-fold increase in SPARC mRNA expression and (d) decreased methylation of CG sites in the SPARC promoter as quantified by pyrosequencing in IVDs. Inset: Increased cold sensitivity following 5AC injection was significantly correlated with total SPARC methylation. * = p < 0.05, *** = p < 0.001. One-way ANOVA followed by Bonferroni's test (aging cohorts) and two-tailed student t-test (5AC vs. vehicle treated). Pearson's correlation (p = 0.03, r² = 0.46). n = 3-15/group. Error bars = S.E.M.

Figure 3

DNA methylation silences murine and human SPARC promoter activity. The mouse (a) and human promoter regions (b) were cloned into a pCPGL-basic plasmid (the CpG positions are indicated as balloons). There were no CpGs in the vector. The plasmids were either methylated or mock methylated in vitro and then transfected into HEK293 cells in the sense or antisense (1^st column) configuration for 48 h (c, d). Relative luciferase activity (percentage) in the extracts is shown as average per group (triplicate transfection). The decrease in expression in the methylated (3^rd column) vs. unmethylated (2^nd column) treatment groups indicates gene silencing. One-way ANOVA followed by Bonferroni's test. ** = p < 0.01 Error bars = S.E.M.

Figure 4

SPARC mRNA expression and DNA methylation in IVDs from chronic LBP patients with disc degeneration. Pain intensity measured with the numeric rating scale (a) and physical disability as determined by the Oswestry Disability Index (b) were increased in surgical patients, as was lumbar disc degeneration based on scoring of MRI images (c). The state of methylation of CG sites in the human SPARC promoter (d) in L4-L5 IVDs was increased as quantified by pyrosequencing (e) * = p < 0.05, ** = p < 0.01, *** = p < 0.001. Two-tailed student t-test. n = 5-8/group. Error bars = S.E.M.

Figure 5

Calculation of Disc Height Index from Mouse Lumbar Spinal X-ray. Lateral x-ray images of the intact lumbar spine were taken at 4× using a Faxitron ^® MX-20 (Faxitron X-Ray LLC, Lincolnshire, IL). Disc Height Index (DHI) was determined according to the following equation: Disc Height Index (DHI) = 2 × (DH1 + DH2 + DH3)/(A1 + A2 + A3 + B1 + B2 + B3) where A and B represent the length of the vertebral bone immediately rostral and caudal to the IVD, respectively; and DH represents the disc height between adjacent vertebrae.