Pathogenesis and pathophysiology
Article section 6 of 16.  Previous  Next

By Raphael Schiffmann MD

The pathophysiology has been recently reviewed (Berger et al 2014). There is no doubt that the principal biochemical abnormality is the accumulation of saturated very long-chain fatty acids. In the past there had been consensus that the accumulation of very long chain fatty acid was due the impaired capacity to form their Coenzyme A derivative, a reaction that takes place in the peroxisome and was found to be deficient in cultured skin fibroblasts of patients with X-ALD (Hashmi et al 1986; Lazo et al 1988; Wanders et al 1988). Very long-chain acyl-CoA esters are transported into peroxisomes by ABCD1 independently of additional synthetase activity (Wiesinger et al 2013). However, ABCD1 remains a susceptibility gene, necessary but not sufficient for inflammatory demyelination to occur (Berger et al 2014). Oezen and colleagues conducted a comprehensive study of mitochondrial function in muscle tissue of the X-ALD mouse model (Oezen et al 2005). They demonstrated increased VLCFA in that tissue, but found mitochondrial function to be normal in all respects. They concluded that mitochondrial dysfunction is not a cause of VLCFA excess in this tissue. Studies suggest that oxidative stress contributes to the pathogenesis to the X-ALD. Di Biase and colleagues demonstrated free oxygen radical disease in C6 glial cells enriched in hexacosanoic acid (Di Biase et al 2004). Vargas and colleagues reported increased levels of anti-oxidant activity in the plasma and increased levels of antioxidant enzymes such as catalase, superoxide dismutase and glutathione peroxidase in the erythrocytes and cultured skin fibroblasts of X-ALD patients (Vargas et al 2004). Oral administration of pioglitazone, an agonist of PPAR-gamma, to a mouse model of X-ALD restored mitochondrial content and expression of master regulators of biogenesis, neutralized oxidative damage to proteins and DNA, and reversed bioenergetic failure in terms of ATP levels, NAD+/NADH ratios, and pyruvate kinase and glutathione reductase activities (Morato et al 2013). The treatment halted locomotor disability and axonal damage in X-linked adrenoleukodystrophy mice (Morato et al 2013).This work also supports the hypothesis that mitochondrial biogenesis is affected in X-ALD.

At this time, the main uncertainty relates to the mechanism by which the very long-chain fatty acids excess leads to the demyelinating process that is the cause of the neurologic disability. Although the earlier studies of Powers and colleagues make it likely that the adrenocortical damage can be attributed directly to the storage of very long-chain fatty acids containing lipids, particularly cholesterol esters (Powers et al 1980), the nervous system damage cannot be traced to this alone. It has already been noted that the extent of nervous system involvement is highly variable. This variability is not at all related to the severity of the biochemical defect or the degree of very long-chain fatty acids accumulation (Boles et al 1991). Instead the variability is related to the presence and severity of a perivascular lymphocytic infiltration (Schaumburg et al 1975; Powers 1985). Plasmalogens play an important role in modulating the effect of VLCFA accumulation. Their low levels in the brain compared to other organs were shown to facilitate cerebral involvement (Brites et al 2009). It is this response and the associated breakdown of the blood-brain barrier that is present in the rapidly progressive forms of the disease in which myelin breakdown is extensive. It is mild or absent in patients with adrenomyeloneuropathy or the slowly progressive forms of the disease (Powers 1985; Powers and Moser 1998; Powers et al 2000). Feigenbaum and colleagues have reported that in regions that are actively demyelinating 47% of oligodendrocytes show evidence of apoptosis (Feigenbaum et al 2000). Inducible nitric oxide synthase expression is increased in the actively demyelinating zones (Gilg et al 2000). The principal neuropathological abnormality in adrenomyeloneuropathy is a distal axonopathy that involves most severely the gracile tract in the cervical region and the corticospinal tract in the lumbosacral region (Powers et al 2000). A study of the dorsal root ganglia in adrenomyeloneuropathy patients by Powers and colleagues indicates that although the total number of nerve cells is not diminished, there is a decrease in the proportion of large neurons, and unlike the inflammatory cerebral forms, there is no evidence of apoptosis (Powers et al 2001). They also made the intriguing observations that the mitochondria contain large lipid inclusions, suggesting that mitochondrial abnormalities contribute to the pathogenesis of the disease. Powers and colleagues have presented evidence that tumor necrosis factor alpha is involved in this response (Powers et al 1992). Griffin and colleagues typed the cells that make up the brain perivascular response and found that it was compatible with an immune-mediated response (Griffin et al 1985). These findings have led to the hypothesis that the alteration in brain lipid fatty acid composition that results from the primary genetic defect lead to a cytokine and immune mediated cascade, and that it is this response that causes the rapid breakdown of myelin. Further support for this hypothesis is provided by the report of Tagawa and colleagues that antigangliosides bind with enhanced affinity to gangliosides containing very long chain fatty acids (Tagawa et al 2002). Schmidt and colleagues have reported that 25% of adult X-ALD patients have antibodies to myelin oligodendrocyte glycoproteins compared to 10% in controls (Schmidt et al 2003). The inflammatory process is reflected in elevation of various cytokines and matrix metalloproteinases in CSF of X-ALD patients. The abnormalities and total CSF protein significantly correlate with disease severity determined by MRI (Lund et al 2012; Thibert et al 2012).

Ito and colleagues with the use of new cytochemical techniques demonstrated that most of the lymphocytes that accumulate in the cerebral forms of X-ALD were mostly CD8 cytotoxic T cells (Ito et al 2001). It is of great interest that these cells. CD1 molecules were present in the acute inflammatory lesion and were present in the astrocyte. The CD1 antigen is a lipid-containing, antigen-presenting molecule that contains a hydrophobic groove that can accommodate very long chain fatty acids up to a 32-carbon chain length. Its possible relevance to the very long chain fatty acids characteristic of X-ALD, thus, is apparent. Paintlia and colleagues have made a detailed correlation of the anatomical localization of inflammatory lesions in the postmortem tissues of X-ALD patients with the levels of very long-chain fatty acids in various lipid fractions and the expression of inflammatory cytokines and postulate that the very long-chain fatty acids excess in membrane domains associated with signal transduction pathways activates resident microglia and astrocytes and that this results in loss of oligodendroglia and myelin (Paintlia et al 2003). Eichler and colleagues showed that microglial apoptosis in perilesional white matter represents an early stage in lesion evolution and may be an appropriate target for intervention in X-ALD patients (Eichler et al 2008). Two publications demonstrate evidence of oxidative stress in X-ALD: one study demonstrates this in the plasma, white blood cells, and cultured skin fibroblasts of X-ALD patients (Vargas et al 2004), the other in C6 glial cells enriched in hexacosanoic acid (Di Biase et al 2004). Mayatepek and colleagues implicate leukotriene T4 in the pathogenesis of the inflammatory response (Mayatepek et al 2003). Studies in the animal model suggest that loss of ABCD1 gene function hampers oxidative stress homeostasis and may be reversible by alpha-tocopherol analogs (Fourcade et al 2008).

It was noted earlier that the phenotype of adrenoleukodystrophy varies widely. The sharpest contrast exists between the rapidly progressive childhood adrenoleukodystrophy and the milder more slowly progressive adrenomyeloneuropathy. Our studies of more than 600 families, and those of others, have shown that childhood cerebral adrenoleukodystrophy and adrenomyeloneuropathy co-occur frequently within the same kindred and even nuclear family, including brothers (Moser et al 1992). Segregation analysis suggests that the phenotypic variation in adrenoleukodystrophy is due to the action of a modifier gene (Smith et al 1999; Kemp et al 2012). Asheuer and colleagues conducted a study of the correlation between phenotype and biochemical changes and gene expression patterns in postmortem brain tissue of X-ALD patients (Asheuer et al 2005). They found that in patients with the severe childhood cerebral phenotype the levels of VLCFA in normal appearing white matter was significantly higher than in patients with milder phenotypes, supportive of the hypothesis that VLCFA excess contributes to pathogenesis. They also made the intriguing observation that the expression of ABCD4, which codes for 1 of the 4 human ATP binding cassette transporter proteins, is decreased in this form of ALD. The biological significance of this finding is not yet clear, but it is the first time that a genotype-phenotype correlation has been demonstrated in X-ALD. However, the co-occurrence of different phenotypes in 2 sets of identical twins suggests the presence of as yet unidentified environmental factors (Sobue et al 1994; Korenke et al 1996). The Tc2 c.776C>G polymorphism involved in methionine metabolism significantly increases the demyelinating phenotype in patients with X-ALD (Semmler et al 2009).

In This Article

Historical note and nomenclature
Clinical manifestations
Clinical vignette
Pathogenesis and pathophysiology
Differential diagnosis
Diagnostic workup
Prognosis and complications
References cited