Diet-Induced Obesity Affects Muscle Regeneration After Murine Blunt Muscle Trauma-A Broad Spectrum Analysis

Abstract

Injury to skeletal muscle affects millions of people worldwide. The underlying regenerative process however, is a very complex mechanism, time-wise highly coordinated, and subdivided in an initial inflammatory, a regenerative and a remodeling phase. Muscle regeneration can be impaired by several factors, among them diet-induced obesity (DIO). In order to evaluate if obesity negatively affects healing processes after trauma, we utilized a blunt injury approach to damage the extensor iliotibialis anticus muscle on the left hind limb of obese and normal weight C57BL/6J without showing any significant differences in force input between normal weight and obese mice. Magnetic resonance imaging (MRI) of the injury and regeneration process revealed edema formation and hemorrhage exudate in muscle tissue of normal weight and obese mice. In addition, morphological analysis of physiological changes revealed tissue necrosis, immune cell infiltration, extracellular matrix (ECM) remodeling, and fibrosis formation in the damaged muscle tissue. Regeneration was delayed in muscles of obese mice, with a higher incidence of fibrosis formation due to hampered expression levels of genes involved in ECM organization. Furthermore, a detailed molecular fingerprint in different stages of muscle regeneration underlined a delay or even lack of a regenerative response to injury in obese mice. A time-lapse heatmap determined 81 differentially expressed genes (DEG) with at least three hits in our model at all-time points, suggesting key candidates with a high impact on muscle regeneration. Pathway analysis of the DEG revealed five pathways with a high confidence level: myeloid leukocyte migration, regulation of tumor necrosis factor production, CD4-positive, alpha-beta T cell differentiation, ECM organization, and toll-like receptor (TLR) signaling. Moreover, changes in complement-, Wnt-, and satellite cell-related genes were found to be impaired in obese animals after trauma. Furthermore, histological satellite cell evaluation showed lower satellite cell numbers in the obese model upon injury. Ankrd1, C3ar1, Ccl8, Mpeg1, and Myog expression levels were also verified by qPCR. In summary, increased fibrosis formation, the reduction of Pax7⁺ satellite cells as well as specific changes in gene expression and signaling pathways could explain the delay of tissue regeneration in obese mice post trauma.

Keywords: C57BL/6J; fibrosis; microarray; obesity; satellite cells; skeletal muscle; trauma.

Figure 1

General workflow for evaluation of present microarray data. After normalization of data and generation of heatmaps for each comparison, a general DEG analysis per time point and diet for the generation of a time-lapse heatmap was carried out. The total amount of genes for normal weight and obese mice at each time was compared to screen for unique and shared genes using Venn diagrams at the indicated time points. Final pathway analysis was carried out with ClueGO (Bindea et al., 2009), a plugin for Cytoscape (Lotia et al., 2013).

Figure 2

Body weight, force input, and soft tissue trauma induction of the left hind leg of C57BL/6J mice. Values are given with mean ± standard deviation; n = 9. Statistical analysis by two-sided homoscedastic t-test. (A) Weight distribution for normal weight and obese animals (^*indicates p ≤ 0.05); Force input into extensor iliotibialis anticus, values showed no significance. (B) Left hind limb of normal weight and obese C57BL/6J mice before and after trauma induction. Hematoma formation was observed within 24 h.

Figure 3

Monitoring of the regeneration process after induction of a blunt muscle injury in female normal weight and obese C57BL/6J mice. (A) FLASH and T2W-RARE recordings of two female C57BL/6J mice receiving either ND or HFD. Arrow = edema. Same animals were used for the whole time course. Prescan (Control) did not show any signs of injury, whereas an edema can be seen starting 1 h post-injury. Edema formation increases until 6 h post-injury, then recedes within 21 days, whereas the obese mouse shows signs of edema 8 days after injury. The normal weight mouse seems to recover within the observed time frame. (B) Hematoxylin-eosin (HE) staining of muscle tissue sections. The staining shows damaged muscle, inflammatory cells and subsequent myofiber regeneration within 21 days post-injury. Triangle = fat inclusion, arrow = newly regenerated myofiber. (C) Sirius Red staining of muscle tissue sections. Triangle = fat inclusion, arrow = fibrosis. Pictures were taken with Olympus IX81 using Xcellence v.1.2. Scale = 200 μm at 10x magnification; Scale = 50 μm at 40x magnification.

Figure 4

Differentially expressed genes in normal weight and obese mice respectively (p ≤ 0.05; fold change ≥ ± 1.5).

Figure 5

Differentially expressed genes, when comparing trauma vs. control (p ≤ 0.05; fold change ≥ ± 1.5) in normal weight and obese mice. Unique and shared genes with their total amount for each time point are shown. Blue = normal; Red = obese. Webtool from Ghent University Bioinformatics and Evolutionary Genomics.

Figure 6

Pathway analysis of differentially expressed genes showing the highest impact over the whole course of time (hits ≥ 3; 23 in total). Data was analyzed using ClueGO (Bindea et al., 2009), a Cytoscape (Lotia et al., 2013) plugin designed by Bindea et al. (2009). Each association to an ontology was counted as one hit, regardless of gene quantity.

Figure 7

Detection of C3ar and C5ar in muscle tissue of normal weight and obese mice by IHC. (A) Anti-C3ar-labeled muscle shows stronger signals for normal weight when compared to obese. (B) Anti-C5ar-labeled muscle shows a partially increased signal, and slightly increases over time, also indicating the negative role of obesity. Pictures were taken with Olympus IX81 using Xcellence v.1.2. Scale = 200 μm at 10x magnification; Scale = 50 μm at 40x magnification.

Figure 8

Determination of Pax7-positive satellite cells by IHC. (A) Tissue sections of female normal weight and obese C57BL/6J mice, labeled with anti-Pax7, and stained with DAB. Triangle = satellite cell. Pictures were taken with Olympus IX81 using Xcellence v.1.2. Scale = 50 μm at 40x magnification. (B) Calculated satellite cells per mm². Six animals per time point à three sections per animal were independently counted by three researchers, averaged and depicted. Black = normal weight; Gray = obese. Statistical analysis by homoscedastic two-sided t-test. ^*Indicates p ≤ 0.05.

Figure 9

Expression levels of Ankrd1, C3ar1, Ccl8, Mpeg1, and Myog determined by qPCR on Ppia as housekeeper. X-fold values ±SEM were determined by normalization on corresponding controls of each time point and diet. Statistical analysis by two-sided homoscedastic t-test. ^*Indicates p ≤ 0.05, n = 3.