The application of Raman spectroscopy to the diagnosis of mitochondrial muscle disease: A preliminary comparison between fibre optic probe and microscope formats

Abstract

Muscle biopsy remains an important component of the diagnostic repertoire for patients with suspected mitochondrial disease, underpinning specialist histopathological and biochemical analyses. Raman spectroscopy has not yet been applied to mitochondrial disease, and new fibre optic systems, with advantages in terms of cost and portability, could provide a rapid means to identify muscle pathology. In this study, we aimed to explore the potential of two different formats of Raman spectroscopy to identify mitochondrial disease: a miniaturised fibre optic Raman system and a standard commercial Raman microscope. Raman spectra were recorded from muscle samples from healthy volunteers (n = 10) and patients with genetically confirmed mitochondrial disease (n = 15). Multivariate classification algorithms demonstrated a high level of disease classification performance with both the fibre optic probe system and microscope (area under receiver operating characteristic curves 0.80–0.82). Key spectral changes associated with mitochondrial disease concerned the α‐helical configuration of proteins. The results suggest that Raman spectroscopy of muscle is worthy of further investigation as a technique for the rapid identification of mitochondrial disease.

Non‐negative matrix factorisation of Raman spectra finds common patterns relating to neuromuscular disease across differing equipment configurations, preclinical models and human tissue

“…α‐helices are the dominant secondary protein structure in muscle [20], and the relative prominence of these in healthy muscle suggests a transition to β‐sheet structures in myopathy [21]. In previous analyses of microscope and fibre optic probe data, we observed a relative loss of α‐helix‐related peaks in both formats [6] and so it is reassuring that this is evident in the combined factorisation. Mode 7 was also significantly more prominent in healthy volunteers across both equipment formats (microscope p = 0.008; probe p < 0.0001) and shared some similar features to mode 1.…”

“…The balance between α‐helix and β‐sheet structures appears to be a robust biomarker of muscle health, noted not only in the preceding microscope/probe analysis on human tissue but also in our previous work in mice [5], human samples using more standard PCA feature extraction [6, 7] and the work of Gautam et al in fly models of myopathy [26]. Biomarkers of disease that cross the preclinical/clinical divide and can provide an equivalent readout of disease state in both settings are a priority area in fatal neuromuscular conditions [27–29].…”

“…In addition, differences were also apparent, particularly between 1200 and 1350 cm À1 (including peaks relating to tyrosine, phenylalanine and the CH 2 CH 3 deformation, proteins/lipids), perhaps relating to differences in power density and collection optics [6]. See Table S4 for further tentative peak assignments and references relating to average spectra.…”

Rapid identification of human muscle disease with fibre optic Raman spectroscopy

“…17 Recently, Raman spectroscopy has shown promise in the identification of neurological disorders, 18,19 including myopathies. [20][21][22] In the present study we have used ex vivo human muscle samples from patients with either genetically confirmed muscle disease, or patients under investigation for suspected muscle disease. Our aim was to test the hypothesis that portable fibre optic Raman spectroscopy can identify abnormal muscle without any sample preparation.…”

“…17 Recently, Raman spectroscopy has shown promise in the identification of neurological disorders, 18,19 including myopathies. 20–22…”