Integrated single-cell functional-proteomic profiling reveals a shift in myofibre specificity in human nemaline myopathy: A proof-of-principle study

Abstract

Skeletal muscle is a complex syncytial arrangement of an array of cell types and, in the case of muscle-specific cells (myofibres), subtypes. There exists extensive heterogeneity in skeletal muscle functional behaviour and molecular landscape at the cell composition, myofibre subtype and intra-myofibre subtype level. This heterogeneity highlights limitations in currently applied methodological approaches, which has stagnated our understanding of fundamental skeletal muscle biology in both healthy and myopathic contexts. Here we developed a novel approach that combines a fluorescence-based assay for the biophysical examination of the sarcomeric protein, myosin, coupled with same-myofibre high-sensitivity proteome profiling, termed single myofibre protein function-omics (SMPFO). Applying this approach as proof-of-principle we identify the integrated relationship between myofibre functionality and the underlying proteomic landscape that guides divergent, but physiologically important, behaviour in myofibre subtypes in healthy human skeletal muscle. By applying SMPFO to two forms of human nemaline myopathy (ACTA1 and TNNT1 mutations), we reveal significant reduction in the divergence of myofibre subtypes across both biophysical and proteomic behaviour. Collectively we demonstrate preliminary findings of SMPFO to support its use to study skeletal muscle with greater specificity, accuracy and resolution than currently applied methods, facilitating that advancement in understanding of skeletal muscle tissue in both healthy and diseased states. KEY POINTS: Skeletal muscle is a complex tissue made up of an array of cell and sub-cell types, with the resident muscle cell - myofibre - critical for contractile function. Although single myofibre studies have advanced, existing methods lack the precision for simultaneous multidata analysis, hindering developments in our understanding of skeletal muscle. We introduce single myofibre protein function-omics (SMPFO), a method enabling functional analysis of sarcomeric myosin alongside global protein abundance within the same myofibre. In healthy myofibres SMyoMFO reveals extensive biochemical diversity in myosin heads, correlating with the abundance of metabolic and sarcomeric proteins, including subtype-specific patterns in sarcoglycan delta (SGCD). In contrast SMyoMFO uniquely reveals a reduction in diversity of myosin function and the myofibre proteome in two forms of nemaline myopathy, highlighting disease-associated alterations. This innovative approach provides a robust framework for investigating myofibre regulation and dysfunction in skeletal muscle biology.

Keywords: human muscle; myopathy; myosin; proteomics.

Figure 1. Single myofibre protein function‐omics (SMyoMFO) uniquely integrates single myofibre myosin state with same‐cell global proteome profiling

A, schematic representation of the SMyoMFO workflow that enables us to capture both datasets from N = 68 single myofibres. B, from healthy controls (N = 8). C, number of quantified protein hits and D, state of myosin in SRX% from MANT‐ATP chase assay. E, single myofibre proteome data hits ranked by expression (LFQ, Log2). F, gene ontology (GO) analyses of the 100 most enriched hits mapping to highly specific skeletal muscle terms. G, distribution of myofibre subtypes based on the relative expression of MYH7 (square), MYH2 (triangle), MYH1 (dot) within each single myofibre and ranked by the relative expression of MYH7. Inset histogram shows the distribution of type I (yellow), hybrid I/IIa (blue) and IIa (purple) subtypes. H and I, pseudo‐bulk differential abundance analysis of type I (yellow) ∼ type IIa (purple) myofibres with significant (unadjusted Xiao threshold < 0.05) hits relating to metabolic and muscle GO terms (N = 8). J, pseudo‐bulk analysis performed on single myofibres categorised as either high (N = 7, >65%, orange) or low (N = 7, <35%, blue) myosin SRX state (denoted in D) using unadjusted Xiao significance threshold. K, correlation coefficient volcano plot displaying the association between same myofibre SRX% and protein expression across all myofibres (unadjusted P‐values). On a myofibre subtype level (L) we show divergent patterns of association between SRX and protein expression with sarcoglycan delta (M) displaying significant divergence between type I and type IIa SRX and protein expression (LFQ log2). Unless otherwise stated N = 8 SkM biopsies were used for all analyses.

Figure 2. Single myofibre protein function‐omics (SMyoMFO) reveals a shrinking of diversity within both myosin conformational state and protein expression in single myofibres of ACTA1 and TNNT1 nemaline myopathy

A, immunofluorescence imaging of skeletal muscle (SkM) cross‐sections (upper) and single myofibres (lower) from control, ACTA1 and TNNT1 SkM. B–D, SMyoMFO captured MANT‐ATP (C) and global proteomic datasets (D) from N = 56 single myofibres from control, ACTA1 and TNNT1. E, single myofibre proteome data from control, ACTA1 and TNNT1 ranked by expression (LFQ, Log2) with gene ontology (F) analyses of the 100 most enriched hits suggesting a high specificity for SkM. G, as per Fig. 1G myofibre subtype distribution and frequency for control, ACTA1 and TNNT1 samples, highlighting enriched subtype diversity in patients with nemaline myopathy. H,I, differential protein expression analysis (unadjusted Xiao threshold < 0.05) of specific myofibre subtypes, comparing type I (ACTA1) and type IIa (TNNT1) with relevant controls. J, analysis of downregulated proteins from these comparisons reveals commonality in enriched biological terms. K, comparative analysis of all differentially expressed proteins highlights distinct overlap in dysregulation between control and ACTA1‐type I and TNNT1‐type IIa myofibres. L, of the 79 overlapping proteins 78 displayed the same direction of change compared to control. M, the exception being MYBC2. In control type I and type IIa myofibres, MYBPC2 expression is significantly different (analysis of variance (ANOVA)). However this differential abundance is reduced in type I and type IIa myofibres of ACTA1 and TNNT1, respectively. N = 3/condition, boxplot represents median, interquartile range with single myofibres highlighted by individual points. N, representation of differentially expressed proteins (unadjusted Xiao threshold < 0.05) identified in control type I ∼ type IIa and ACTA1‐type I ∼ TNNT‐type IIa data set. Highlighted are 35 differentially expressed proteins from control analysis that are identified in the entire comparative ACTA1‐type I ∼ TNNT1‐type IIa analysis, with only 14 of these proteins retaining significant differential abundance. O, unless otherwise stated, N = 3 SkM biopsies per condition (control, ACTA1, TNNT1) for all analyses depicted in the figure.