Expanding and validating the biomarkers for mitochondrial diseases

Abstract

Mitochondrial diseases are highly heterogeneous metabolic disorders caused by genetic alterations in the mitochondrial DNA (mtDNA) or in the nuclear genome. In this study, we investigated a panel of blood biomarkers in a cohort of 123 mitochondrial patients, with prominent neurological and muscular manifestations. These biomarkers included creatine, fibroblast growth factor 21 (FGF21) and growth/differentiation factor 15 (GDF-15), and the novel cell free circulating-mtDNA (ccf-mtDNA). All biomarkers were significantly increased in the patient group. After stratification by the specific phenotypes, ccf-mtDNA was significantly increased in the Mitochondrial Encephalomyopathy Lactic Acidosis Stroke-like episodes syndrome (MELAS) group, and FGF21 and GDF-15 were significantly elevated in patients with MELAS and Myoclonic Epilepsy Ragged Red Fibers syndrome. On the contrary, in our cohort, creatine was not associated to a specific clinical phenotype. Longitudinal assessment in four MELAS patients showed increased levels of ccf-mtDNA in relation to acute events (stroke-like episodes/status epilepticus) or progression of neurodegeneration. Our results confirm the association of FGF21 and GDF-15 with mitochondrial translation defects due to tRNA mutations. Most notably, the novel ccf-mtDNA was strongly associated with MELAS and may be used for monitoring the disease course or to evaluate the efficacy of therapies, especially in the acute phase. KEY MESSAGES: • FGF21/GDF15 efficiently identifies mitochondrial diseases due to mutations in tRNA genes. • The novel ccf-mtDNA is associated with MELAS and increases during acute events. • Creatine only discriminates severe mitochondrial patients. • FGF21, GDF-15, and ccf-mtDNA are possibly useful for monitoring therapy efficacy.

Keywords: Biomarkers; Cell free circulating-mtDNA; Creatine; FGF21; GDF-15; Mitochondrial diseases.

Fig. 1

Evaluation of ccf-mtDNA in plasma from controls and mitochondrial patients. a Ccf-mtDNA (MT-ND2) in controls (CTRLS) and mitochondrial patients (MD). b Ccf-mtDNA (MT-ND2) in CTRLS, patients with mtDNA genetic defects (MDD), and nuclear DNA genetic defects (NDD). c Ccf-mtDNA (MT-ND2) in CTRLS and mitochondrial patients stratified by phenotypes. d ROC curves for ccf-mtDNA (MT-ND2) in MD and MELAS patients. AUC in MD was 0.61 (95% CI 0.50–0.72, p = 0.05), in MELAS 0.73 (95% CI 0.60–0.86, p < 0.01). *p value < 0.05, ***p value < 0.001

Fig. 2

Evaluation of ccf-mtDNA in longitudinal samples of MELAS patients. a-d Ccf-mtDNA (MT-ND2) evaluated in longitudinal samples in four different patients. SLE, stroke-like episode; SE, status epilepticus

Fig. 3

Evaluation of creatine in serum from controls and mitochondrial patients. a Creatine levels in controls (CTRLS) and mitochondrial patients (MD). b Creatine levels in CTRLS, patients with mtDNA genetic defects (MDD), and nuclear DNA genetic defects (NDD). c Creatine levels in CTRLS and mitochondrial patients stratified by phenotypes. d ROC curve for creatine in MD patients. AUC was 0.62 (95% CI 0.51–0.73, p = 0.06) for MD. **p value < 0.01

Fig. 4

Assessment of FGF21 and GDF-15 in serum from controls and mitochondrial patients. a–b FGF21 and GDF-15 in controls (CTRLS) and mitochondrial patients (MD). c–d FGF21 and GDF-15 levels in CTRLS, patients with mtDNA genetic defects (MDD), and nuclear DNA genetic defects (NDD). e–f FGF21 and GDF-15 in CTRLS and mitochondrial patients stratified by phenotypes. g ROC curves for FGF21 and GDF-15 in MD and MELAS/MERRF patients. AUC for FGF21 was 0.75 (95% CI 0.65–0.85, p < 0.01) in MD and 0.89 (95% CI 0.80–0.98, p < 0.0001) in MELAS/MERRF, whereas AUC for GDF-15 was 0.96 (95% CI 0.93–0.99, p < 0.0001) in MD and 0.99 (95% CI 0.98–1.00, p < 0.0001) in MELAS/MERRF. h Pearson’s correlation between FGF21 and GDF-15 in MD. One outlier belonging to the MELAS group was excluded for the correlation analysis. Pearson r and p value are indicated in the graph. #, Significance after Bonferroni’s correction (p < 0.005); $, significance after Benjamini–Hochberg’s correction (FDR 0.10). ****p value < 0.0001, ***p value < 0.001, *p value < 0.05

Fig. 5

Correlation analyses between biomarkers and alanine/lactic acid plasma concentrations in MELAS patients. a–d Pearson’s correlation of alanine plasma concentration and ccf-mtDNA (MT-ND2), creatine, FGF21, and GDF-15. e–h Pearson’s correlation of lactic acid plasma concentration and ccf-mtDNA (MT-ND2), creatine, FGF21, and GDF-15. Pearson r and p values are indicated in the graphs. #, Significance after Bonferroni’s correction (p < 0.006); $, significance after Benjamini–Hochberg’s correction (FDR 0.10)