Applications of metabolomics and proteomics to the mdx mouse model of Duchenne muscular dystrophy: lessons from downstream of the transcriptome

Abstract

Functional genomic studies are dominated by transcriptomic approaches, in part reflecting the vast amount of information that can be obtained, the ability to amplify mRNA and the availability of commercially standardized functional genomic DNA microarrays and related techniques. This can be contrasted with proteomics, metabolomics and metabolic flux analysis (fluxomics), which have all been much slower in development, despite these techniques each providing a unique viewpoint of what is happening in the overall biological system. Here, we give an overview of developments in these fields 'downstream' of the transcriptome by considering the characterization of one particular, but widely used, mouse model of human disease. The mdx mouse is a model of Duchenne muscular dystrophy (DMD) and has been widely used to understand the progressive skeletal muscle wasting that accompanies DMD, and more recently the associated cardiomyopathy, as well as to unravel the roles of the other isoforms of dystrophin, such as those found in the brain. Studies using proteomics, metabolomics and fluxomics have characterized perturbations in calcium homeostasis in dystrophic skeletal muscle, provided an understanding of the role of dystrophin in skeletal muscle regeneration, and defined the changes in substrate energy metabolism in the working heart. More importantly, they all point to perturbations in proteins, metabolites and metabolic fluxes reflecting mitochondrial energetic alterations, even in the early stage of the dystrophic pathology. Philosophically, these studies also illustrate an important lesson relevant to both functional genomics and the mouse phenotyping in that the knowledge generated has advanced our understanding of cell biology and physiological organization as much as it has advanced our understanding of the disease.

Figure 1

Applications of metabolomics and proteomics to the mdx mouse. (a) A typical high-resolution ¹H NMR spectrum from an aqueous extract of cardiac tissue from the mdx mouse. The chemical shift and splitting pattern of a given resonance (peak) enables the identification of the metabolite it belongs to, and the area under the resonance determines the concentration of that metabolite. (b) An orthogonal signal corrected partial least squares discriminate analysis plot of various mouse models of cardiac disease using solution state NMR spectroscopy. Key: circles, control + mdx; diamonds, model of cardiac hypertrophy (MLPKO); squares, model of cardiac arrhythmia (Scn^-/+); triangles, model of cardiac arrhythmia (ScnΔ ^/+). (c) Correlation analysis between identified proteins in a proteomic study of heart tissue from mdx mice and the intracellular concentration of taurine. When detected by ¹H NMR spectroscopy (bottom graph), taurine can be identified by two triple peaks at δ 3.25-3.27 and δ 3.42-3.46. The correlation heat map between spectral intensity and protein expression was used to determine which proteomic changes were associated with the increase in taurine in dystrophic muscle. The x axis is the chemical shift region containing the resonances from taurine; the y axis consists of protein spots detected in the two-dimensional gel electrophoresis. The color scale displays the correlation coefficients between the two sets of data (concentration of taurine against concentration of protein).

Figure 2

Metabolic flux ratios assessed in working control (C) and mdx mouse heart perfused with ¹³C-labeled substrates. Data are given as means ± standard errors (indicated in parentheses); n = 4-8 in each group. Reactions are shown in the part of the cell in which they take place. (a) Flux ratios. (i) Flux ratios reflecting the contribution of exogenous fatty acids (oleate) and carbohydrates (CHOs: lactate, pyruvate and glucose) to acetyl-CoA formation (energy) and oxaloacetate (OAA; anaplerosis) via oleate β-oxidation (OLE), pyruvate decarboxylation (PDC) and carboxyation (PC), respectively, and expressed relative to citrate synthesis (CS). (ii) Flux ratios reflecting the contribution of individual CHOs - as indicated by the individual arrows - to pyruvate formation, expressed in percentage of total. (b) Glycolytic rate, which reflects the production of lactate and pyruvate, in μmol × min^-1. (c) Tissue concentration of Krebs cycle intermediates, in μmol × g wet weight^-1. (d) Tissue aconitase activity, in μmol × min^-1 × mg protein^-1. *p < 0.05, ^#p < 0.001 for mdx versus control mouse hearts. Adapted from Khairallah et al. [54].