The Effect of Cutibacterium acnes Infection on Nerve Penetration in the Annulus Fibrosus of Lumbar Intervertebral Discs via Suppressing Oxidative Stress

Abstract

Modic changes (MCs) and low back pain are highly correlated and an economic burden to the society. Previous studies have shown that Cutibacterium acnes (C. acnes) infection can lead to MCs. The purpose of this study was to clarify whether and how C. acnes contributes to oxidative stress and nerve growth that potentially leads to low back pain. Neurons from the hippocampus or dorsal root ganglion (DRG) of Sprague-Dawley (SD) rats were cocultured with annulus fibrosus cells (AFCs) with or without the presence of the C. acnes supernatant in vitro. Cell viability, neurite length, oxidative stress, and neuro-related gene expression were examined. Furthermore, samples from the patients with MCs and SD rat model of MCs were used to validate the nerve growth results. Neurons from both the hippocampus and DRG showed neurites when cocultured with AFCs in the environment with/without the C. acnes supernatant. The average neurite length was significantly longer when exposed to the C. acnes supernatant in the hippocampal neuron (217.1 ± 90.0 μm versus 150.1 ± 68.1 μm in the control group) and in the DRG neuron (229.1 ± 91.3 μm versus 149.2 ± 64.8 μm in the control group). Hippocampal neurons showed upregulated expression levels of NeuN, Map2, and Psd95, while upregulation was only seen in Tuj-1 in DRG neurons. Suppressed oxidative stress could be observed using axon growth symbols. Degenerated disc structures and abnormal bone remodelling were found in animal models and clinical samples of MCs, with astrocytes, microglia, and neurons in the disc. Therefore, C. acnes infection was found to cause back pain in the presence of MCs by promoting nerve penetration into the annulus fibrosus by suppressing oxidative stress.

Figure 1

Representative MR image of the MC group (A1: T2-weighted; A2: T1-weighted) and control group (B1: T2-weighted; B2: T1-weighted). Representative preoperative (C1: T2-weighted; C2: T1-weighted) and postoperative (D1: T2-weighted; D2: T1-weighted) MR image of the MC animal model. Arrows indicate the segments with MCs or C. acnes injection. MCs: Modic changes.

Figure 2

Representative images of cocultured AFCs and hippocampal neuron in bright-field, Tuj-1/NeuN/DAPI/merged immunofluorescence, and neurite length measurement (20x, A–D, scale bar = 100 μm, the white arrows indicate where the neurite formed between neurons and AFCs). (e) shows the quantitative measurement of neurite length (means ± SD, ^∗P < 0.0.5). (f) Nerve-related gene expression in hippocampal neuron-AFC coculture environment (means ± SD, ∗ indicates P < 0.0.5). AFCs: annulus fibrosus cells; C. a: Cutibacterium acnes.

Figure 3

Representative images of cocultured annulus fibrosus cells and DRG neuron in (A) bright-field, Tuj-1/NeuN/DAPI/merged immunofluorescence, and neurite length measurement (20x, A–D, scale bar = 100 μm; the white arrows indicate where the neurite formed among neurons and neurogliocytes or AFCs). (e) shows the quantitative measurement of neurite length (means ± SD, ^∗P < 0.0.5). (f) Nerve-related gene expression in DRG neuron-AFC coculture environment (means ± SD, ∗ indicates P < 0.0.5). DRG: dorsal root ganglions; AFC: annulus fibrosus cells; C. a: Cutibacterium acnes.

Figure 4

(A, B) Representative images of annulus fibrosus cells treated with/without the C. acnes supernatant, stained with MitoSOX and DAPI (40x, scale bar = 50 μm), with MitoSOX density quantified in (c). (D–F) Representative images of DRG neurons cocultured with annulus fibrosus cells and C. acnes supernatant (40x, scale bar = 50 μm), with MitoSOX density quantified in (g); (h) shows the Ablim and Ryr2 gene expression in DRG neuron-AFC coculture environment (means ± SD, ∗ indicates P < 0.0.5). AFCs: annulus fibrosus cells; C. a: Cutibacterium acnes; DRG: dorsal root ganglions.

Figure 5

Representative images of the clinical MC group and control group with (A) HE staining (scale bar = 1.25 mm), (B) immunofluorescence of GFAP and Iba-1 (scale bar = 200 μm), and (C) immunofluorescence of Tuj-1 and NeuN (scale bar = 200 μm); images in the lower right corner are magnified from the boxed area of the corresponding images, with target protein indicated by coloured arrows. (d)–(f) show quantitative measurement of GFAP/Iba-1/NeuN-positive cells (means ± SD of a single slice, ∗ indicates P < 0.0.5). MCs: Modic changes.

Figure 6

Representative images of the SD rat MC group and control group with immunofluorescence of GFAP/iba-1 (A and C, scale bar = 100 μm) and Tuj-1/NeuN (B and D, scale bar = 100 μm); images in the lower right corner are magnified from the boxed area of the corresponding images, and white arrows indicate the GFAP/Iba-1/NeuN-positive neurons. (E) and (F) show the location of GFAP/Iba-1/NeuN-positive neurons in the IVD (cells with positive immunofluorescence were marked as bright yellow dots (scale bar = 1 mm)); (g)–(i) show quantitative measurement of GFAP/Iba-1/NeuN-positive cells (means ± SD of a single motion segment, ∗ indicates P < 0.0.5). MCs: Modic changes.