Altered high-energy phosphate and membrane metabolism in Pelizaeus-Merzbacher disease using phosphorus magnetic resonance spectroscopy

Abstract

Pelizaeus-Merzbacher disease is an X-linked recessive leucodystrophy of the central nervous system caused by mutations affecting the major myelin protein, proteolipid protein 1. The extent of the altered in vivo neurochemistry of protein, proteolipid protein 1 duplications, the most common form of Pelizaeus-Merzbacher disease, is, however, poorly understood. Phosphorus magnetic resonance spectroscopy is the only in vivo technique that can assess the biochemistry associated with high-energy phosphate and membrane phospholipid metabolism across different cortical, subcortical and white matter areas. In this cross-sectional study, whole-brain, multi-voxel phosphorus magnetic resonance spectroscopy was acquired at 3 T on 14 patients with Pelizaeus-Merzbacher disease with protein, proteolipid protein 1 duplications and 23 healthy controls (all males). Anabolic and catabolic levels of membrane phospholipids (phosphocholine and phosphoethanolamine, and glycerophosphoethanolamine and glycerophosphocholine, respectively), as well as phosphocreatine, inorganic orthophosphate and adenosine triphosphate levels relative to the total phosphorus magnetic resonance spectroscopy signal from 12 different cortical and subcortical areas were compared between the two groups. Independent of brain area, phosphocholine, glycerophosphoethanolamine and inorganic orthophosphate levels were significantly lower (P = 0.0025, P < 0.0001 and P = 0.0002) and phosphocreatine levels were significantly higher (P < 0.0001) in Pelizaeus-Merzbacher disease patients compared with controls. Additionally, there was a significant group-by-brain area interaction for phosphocreatine with post-hoc analyses demonstrating significantly higher phosphocreatine levels in patients with Pelizaeus-Merzbacher disease compared with controls across multiple brain areas (anterior and posterior white matter, superior parietal lobe, posterior cingulate cortex, hippocampus, occipital cortex, striatum and thalamus; all P ≤ 0.0042). Phosphoethanolamine, glycerophosphoethanolamine and adenosine triphosphate levels were not significantly different between groups. For the first-time, widespread alterations in phosphorus magnetic resonance spectroscopy metabolite levels of Pelizaeus-Merzbacher disease patients are being reported. Specifically, increased high-energy phosphate storage levels of phosphocreatine concomitant with decreased inorganic orthophosphate across multiple areas suggest a widespread reduction in the high-energy phosphate utilization in Pelizaeus-Merzbacher disease, and the membrane phospholipid metabolite deficits suggest a widespread degradation in the neuropil content/maintenance of patients with Pelizaeus-Merzbacher disease which includes axons, dendrites and astrocytes within cortex and the myelin microstructure and oligodendrocytes within white matter. These results provide greater insight into the neuropathology of Pelizaeus-Merzbacher disease both in terms of energy expenditure and membrane phospholipid metabolites. Future longitudinal studies are warranted to investigate the utility of phosphorus magnetic resonance spectroscopy as surrogate biomarkers in monitoring treatment intervention for Pelizaeus-Merzbacher disease.

Keywords: Pelizaeus–Merzbacher disease; biochemistry; high-energy phosphate; membrane phospholipids; phosphorus magnetic resonance spectroscopy.

Graphical abstract

Figure 1

Pathological analysis of WM from a 60-year-old patient with PLP1 duplication. (A) Toluidine blue staining of a normal control subject. (B) Toluidine blue staining of patient with Pelizaeus–Merzbacher disease showing thinning of myelin and reduced axonal fibre density with an increase composition of myelin degradation products. (C) H&E staining also demonstrates very thin myelin and reduced numbers of oligodendrocytes. There are also irregular non-uniform aggregates which are presumably degradation products of myelin. (D) HE-LFB staining shows a reduction and thinning of myelinated fibres with diffuse myelin degradation products. Scale 50 µm.

Figure 2

Anatomical placement of the ³¹P MRS voxels. Five different in-plane axial views of the 12 right/left and medial regions of interest as indicated by the black square outlines that include the anterior and posterior WM (aWM and pWM), dorsolateral prefrontal cortex (dlPFC), inferior and superior parietal lobe (iPL and sPL), superior temporal gyrus (STG), anterior/body of the hippocampus (HIP), occipital cortex (OCC), striatum (STR), thalamus (THA) and the medial anterior and posterior cingulate (ACC and PCC).

Figure 3

Example of a quantified in vivo ³¹P MRS spectrum extracted from the left posterior WM area (pWM) of a participant. The top includes the Fourier transformation of the acquired (black line) and modelled (red line) signal, and the residual (difference between acquired and modelled signal) is shown below (black line). The individual modelled spectral peaks are shown at the bottom (blue line) and the assignment of the spectral peaks are as indicated.

Figure 4

³¹P MRS metabolite levels between groups and across regions. Mean PC (A), GPC (B), PCr (C) and Pi (D) levels (±standard error of the mean) of the 12 regions of interest for the patients with Pelizaeus–Merzbacher disease (red bars) and healthy participants (blue bars). Significant group differences after applying the Bonferroni correction (P < 0.0042) from post-hoc analyses are indicated by the * symbol.