Serum neuronal, glial and mitochondrial markers in autosomal dominant optic atrophy and Leber hereditary optic neuropathy

Abstract

Leber hereditary optic neuropathy (LHON) and autosomal-dominant optic atrophy (ADOA) are the two most prevailing primary mitochondrial optic neuropathies. Both diseases preferentially affect the smallest retinal ganglion cells (GCs) of the papillomacular bundle, causing central visual loss in young patients. Although ADOA and LHON show striking similarities, including the convergence of underlying pathologic mitochondrial mechanisms, they differ clinically. The major distinction lies in the timing and progression of axonal damage during neurodegeneration. The exact reasons for these differences remain unclear, but they may, in part, be due to distinct patterns of mitochondrial dysfunction. To identify differences that could point to distinct degenerative processes, we investigated clinical features, optical coherence tomography (OCT) findings, laboratory biomarkers [serum neurofilaments light chain (sNfL), serum glial fibrillary acidic protein (sGFAP) and serum growth differentiation factor-15 (sGDF15)] in a cohort of patients with these two heritable optic neuropathies in the chronic phase. Our OCT analysis reveals a more profound GC layer and papillomacular bundle loss in LHON, whereas ADOA shows a sparser damage of the retinal nerve fibre layer, including fibres originating from the nasal retina. We also observed increased plasma levels of sNfL and GFAP in both groups, supporting the presence of ongoing neurodegeneration in both optic neuropathies. Finally, our findings suggest the retinal astrocytes may play a contributive role in the neurodegenerative process at the level of the optic nerve head, particularly in ADOA.

Keywords: ADOA; Glia; LHON; mitochondria; neurodegeneration.

Graphical Abstract

Figure 1

Mitochondrial genome and OPA1 variants. (A) Mitochondrial genome (mtDNA). The three common primary variants are at mitochondrial nucleotide positions mt.3460, mt.11778, and mt.14484. The variants affect MT-ND1, MT-ND4, and MT-ND6 subunit genes of complex I, respectively. (B) Distribution of OPA1 variants observed in our cohort. Each distinct variant observed in our cohort is represented above the gene: four patients had the in-frame deletion p.Lys979del, two patients had the frameshift p.Val958Glyfs*3, and two patients with the missense p.Ile143Thr and frameshift p.Gly112Trpfs*9 variant, respectively; the last two mutations are novel. The in-frame deletions and the frameshift variants are OPA1 variants predicted to lead to haploinsufficiency; instead, missense variants are associated with a dominant negative mechanism. OPA1 contains a basic domain (BD), a predicted transmembrane region (TM), two coiled-coil regions (CC), and the GTPase domain (GTPase).

Figure 2

Data points. RNFL (retinal nerve fiber layer) of Nasal (A), Temporal (B), Superior Temporal (C), Inferior Temporal (D), Superior Nasal (E), Inferior Nasal (F) quadrants in ADOA (Autosomal Dominant Optic Atrophy) patients (n = 9), LHON (Leber Hereditary Optic Neuropathy) patients (n = 11), and Healty Controls subjects (HC) (n = 34); statistical differences (Games Howell test, P < 0.05) are also reported.

Figure 3

Data points. GCL (Ganglion Cells Layers) of Nasal (A), Superior (B), Temporal (C), and Inferior (D) quadrants in ADOA patients (n = 8), LHON (n = 11), and HC subjects (n = 34). Single data points representing the observations for each of the ADOA and LHON patients and HC participants; statistical differences (Games Howell test, P < 0.05) are also reported.

Figure 4

Data points. Serum concentrations of neurofilament light chain (sNfL) (A), glial fibrillary acidic protein (sGFAp) (B), and growth/differentiation factor-15 (GDF-15) (C) in autosomal dominant optic atrophy (ADOA) (n = 9), Leber hereditary optic neuropathy (LHON) (n = 12), and healthy controls (HCs) (n = 40). Single data points (grey dots) representing the observations for each of the ADOA and LHON patients and healthy control participants are also reported together with the statistical differences (Ptukey test, P < 0.05).

Figure 5

Correlation between CGs data and log10 sNfL. Partial correlation between Log10 levels of serum neurofilament light chain (sNfL) and retinal ganglion cell layer thickness measured with OCT, with age as control variable, in autosomal dominant optic atrophy (ADOA) (n = 9), and Leber hereditary optic neuropathy (LHON) (n = 12). P and rho values obtained by the Spearman correlation test are indicated in the legend.