Vascular dysfunction as an additional pathomechanism in glutaric aciduria type I

The metabolic hallmark of glutaric aciduria type I (GA I) is the deficiency of glutaryl-CoA dehydrogenase (GCDH) with subsequent accumulation of glutaric acid, 3-hydroxglutaric acid (3-OH-GA) and glutaconic acid. Current concepts regarding pathomechanisms of GA I focus on investigations of excitotoxic effects of 3-OH-GA. To identify pathogenetically relevant genes, microarray analyses were performed using brain material from GCDH-deficient (GCDH^-/-) and control mice. These microarray data confirmed recent pathogenic models, but also revealed alterations in genes that had previously not been correlated to the disease, e.g. genes concerning vascular biology. Subsequent in vitro and in vivo experiments confirmed direct effects of 3-OH-GA on vascular permeability and endothelial integrity. Clinical observations underscore the involvement of vascular dysfunction. In MRI scans of GA I patients, subdural effusions as well as dilated transarachnoid vascular plexuses were detected independently of encephalopathic crises. In fact, some of these findings are already detectable shortly after birth. MRI scans of a GA I patient performed during an acute encephalopathic crisis detected a dilated intrastriatal vasculature with perivascular hyperintensity, indicating local extravasation. In conclusion, we hypothesize that 3-OH-GA affects prenatal development of vessels, thus leading to an increased vulnerability of endothelial structures and subsequent vascular dysfunction. These observations display an additional pathomechanism in GA I and might explain frontotemporal hypoplasia and chronic subdural effusions in this disease. Elucidation of the pathomechanisms of vascular dysfunction may give further insights into the pathogenesis of GA I. Glutaryl-CoA dehydrogenase (GCDH) deficiency (glutaric aciduria type I, GA I; McKusick 231670) is an autosomal recessively inherited neurometabolic disorder. The principal clinical manifestation is characterized by a distinct neuropathology, being complicated by acute encephalopathic crises between age 6 and 18 months (for recent review, see Goodman and Frerman 2001; Strauss et al 2003). GCDH deficiency leads to the accumulation of glutaric acid (GA), 3-hydroxyglutaric acid (3-OH-GA) and glutaconic acid in all body fluids of patients. Whereas elevated levels of GA were also found in GA II and GA III, changes in 3-OH-GA concentration appear to be specific for GA I. In several in vitro studies investigating the effects of GA and 3-OH-GA using primary cultures of rat brain neurons or mixed cortex cells, and corticostriatal slice cultures, 3-OH-GA was found to be more potent than GA with respect to cell toxicity (Das et al 2003; Kölker et al 1999, 2000; Ullrich et al 1999). These data support the concept that 3-OH-GA is the major pathogenic compound in the disease. Owing to the structural similarities of GA and 3-OH-GA to glutamate, in vitro studies have focused on ionotropic glutamate receptors, i.e. N-methyl-D-aspartate (NMDA) receptors. In addition, the effects of 3-OH-GA on cell survival could be prevented by NMDA receptor-specific antagonists, supporting the concept of NMDA receptor-mediated excitotoxic neuronal damage by 3-OH-GA (Kölker et al 2004; Ullrich et al 1999). However, no direct interaction between 3-OH-GA and NMDA receptors has been demonstrated so far. Moreover, the competitive inhibition of neuronal glutamate decarboxylase leading to reduced GABA and increased glutamate levels (Stokke et al 1976), and the 3-OH-GA-stimulated uptake of glutamate by rat brain astrocytes (Frizzo et al 2004) suggest indirect effects of 3-OH-GA in the brain involving different cell types and signal transduction systems (reactive oxygen species, mitochondrial energy depletion, Ca²⁺, NO synthetase) (Das et al 2003; Kölker et al 2001, 2002, 2004). In all in vitro studies, concentrations of 0.1-50 mmol/L are required to detect cellular effects of 3-OH-GA (Bjugstad et al 2001; Kölker et al 2000; Ullrich et al 1999); these are much higher than the levels measured in body fuids of GA I patients, which vary in the micromolar range (Baric et al 1999; Kölker et al 2003; Strauss et al 2003). However, the levels of 3-OH-GA in brain tissue and cerebrospinal fluid under basal and conditions of catabolic crisis are unknown. Therefore, additional factors contributing to the pathogenic alterations found in GA I patients which might modulate 3-OH-GA effects cannot be excluded.