Abstract

Childhood cerebral adrenoleukodystrophy (CCER), adrenomyeloneuropathy (AMN) and AMN with cerebral demyelination (AMN-C) are the main phenotypic variants of X-linked adrenoleukodystrophy (ALD). It is caused by mutations in the ABCD1 gene encoding a half-size peroxisomal transporter that has to dimerize to become functional. The biochemical hallmark of ALD is the accumulation of very-long-chain fatty acids (VLCFA) in plasma and tissues. However, there is no correlation between the ALD phenotype and the ABCD1 gene mutations or the accumulation of VLCFA in plasma and fibroblast from ALD patients. The absence of genotype–phenotype correlation suggests the existence of modifier genes. To elucidate the mechanisms underlying the phenotypic variability of ALD, we studied the expression of ABCD1, three other peroxisomal transporter genes of the same family (ABCD2, ABCD3 and ABCD4) and two VLCFA synthetase genes (VLCS and BG1) involved in VLCFA metabolism, as well as the VLCFA concentrations in the normal white matter (WM) from ALD patients with CCER, AMN-C and AMN phenotypes. This study shows that: (1) ABCD1 gene mutations leading to truncated ALD protein are unlikely to cause variation in the ALD phenotype; (2) accumulation of saturated VLCFA in normal-appearing WM correlates with ALD phenotype and (3) expression of the ABCD4 and BG1, but not of the ABCD2, ABCD3 and VLCS genes, tends to be correlated with the severity of the disease, acting early in the pathogenesis of ALD.

INTRODUCTION

X-linked adrenoleukodystrophy (ALD, MIM 300100) is a severe peroxisomal disorder with variable phenotypic expression (1,2). Childhood cerebral ALD (CCER) affects ∼40% of patients and is characterized by progressive cerebral demyelination with an inflammatory response in the white matter (WM) of the brain that leads to vegetative stage or death within 2–5 years. Adult-onset adrenomyeloneuropathy (AMN) is characterized by progressive paraparesis, probably as a result of axonal degeneration and secondary demyelination in the spinal cord. Nearly 60% of ALD patients develop this phenotype between 20 and 50 years. AMN is often portrayed as a relatively mild variant of ALD, but 30% of male AMN patients develop cerebral demyelination (AMN-C) with the same poor prognosis as in children (1).

ALD is caused by mutations in the ABCD1 gene that inactivate the peroxisomal ALD protein (ALDP) (3). ALDP belongs to a small family of ATP-binding cassette (ABC) peroxisomal transporters that includes three other members: ABCD2 or ALDR (ALD-related protein) (4), ABCD3 or PMP70 (5) and ABCD4 or P70R (6,7). The phenotypic presentation of identical mutations in the ABCD1 gene is highly variable, ranging from CCER to mild AMN even within the same family (8).

ALD patients accumulate very-long-chain fatty acids (VLCFA) in plasma and tissues, but the exact mechanism by which ABCD1 gene mutations affect the metabolism of VLCFA is not known. As for ABCD1 gene mutations, there is no correlation between VLCFA accumulation in the plasma and fibroblasts from ALD patients and the phenotypic expression of the disease (1). A mouse model of ALD accumulates VLCFA in the brain and spinal cord in a way similar to that found in ALD patients and develops after 15 months clinical and neuropathological features that resemble AMN (912).

ALDP, like the other three ABC proteins (PMP70, ALDRP and P70R) identified up to now in peroxisomes, is a half-size ABC transporter that must dimerize to be functional. The dimerization partner may play a role in determining the transporter substrate (13). Disruption of the PXA1 and PXA2 genes of Saccharomyces cerevisiae results in impaired growth in oleic acid medium, suggesting that the half-size peroxisomal ABC transporters Pat1p and Pat2p encoded by these genes act as heterodimers (1417). Homo- and hetero-dimerization occurs in vitro among ALDP, ALDRP and PMP70 by using the yeast two-hybrid system and co-immunoprecipitation (18,19). However, homodimerization of ALDP seems to prevail in vivo (20). From the 356 non-recurrent mutations identified in the ABCD1 gene (http://www.x-ald.nl), 48% are frame shift, nonsense and insertion/deletion mutations leading to potential truncated proteins. Because ALDP interacts with itself and with other peroxisomal ABC transporters, these truncated proteins could exert a dominant-negative effect and influence the disease phenotype.

The striking absence of phenotype–genotype correlation in ALD also suggests the existence of stochastic factors and/or modifier genes. The function of the ABCD2, ABCD3 and ABCD4 genes in higher eukaryotes is unknown, but their sequence homology with the ABCD1 gene indicates that they may have related or overlapping functions. ABCD2 and ABCD3 genes can complement the biochemical defect in ALD fibroblasts when over-expressed (19,21). The possible functional equivalence of the ABCD1 and ABCD2 genes has been repeatedly suggested (22,23). The ABCD2, ABCD3 and ABCD4 genes are therefore candidate genes that may play a role in the phenotypic expression of ALD.

Activation of VLCFA to CoA derivatives by synthetases plays a pivotal role in directing VLCFA toward a specific metabolic fate, including degradation by peroxisomal β-oxidation, or incorporation into complex lipids (24). Six full-length cDNAs that encode enzymes with very long-chain acyl-CoA synthetase (VLACS) activity have been cloned (25). These enzymes belong to the ‘ACSVL/FATP’ protein family, which includes a peroxisomal VLACS (VLCS or ACSVL1) with activity that is reduced in ALD fibroblasts (26,27). This peroxisomal synthetase is thought to be involved in the biochemical pathology of ALD (28). Min and Benzer (29) described a Drosophila melanogaster mutant, ‘bubblegum’, characterized by neurodegeneration and elevated levels of VLCFA. BG1, the mammalian homolog of the protein encoded by the defective gene in the Drosophila mutant ‘bubblegum’, activates VLCFA to VLFCA-CoA. This activity classifies BG1 as a VLACS enzyme. This enzyme is not found in peroxisomes, but the BG1 gene is expressed in brain, spinal cord, adrenal gland and testis that are affected in ALD. By modulating the fate of VLCFA, the VLCS and BG1 genes may play a role in the variable phenotypic expression of ALD.

To investigate the role conveyed by ABCD1-truncating mutations and ABCD2, ABCD3, ABCD4, VLCS and BG1 genes on the non-allelic phenotypic diversity of ALD, we studied the levels of expression of these genes in the normal cerebral WM from patients with CCER, AMN-C and AMN and from age-matched controls using quantitative real-time RT–PCR. For the ABCD2, ABCD3, ABCD4, VLCS and BG1 genes, we adopted the hypothesis that if any of these genes play a significant role as a modifier gene, differences in gene expression should be detectable in the normal WM of ALD patients who had different phenotypes before the onset of the demyelinating process in the studied WM. We also evaluated the VLCFA content of normal WM from ALD patients with the CCER, AMN-C and AMN phenotypes to find out whether there was any correlation between brain VLCFA concentration and phenotype, in contrast to what is observed in ALD fibroblasts.

RESULTS

ABCD1 gene mutations

The ABCD1 mutations identified in ALD fibroblast and brain samples are shown in Table 1. Fifty percent were missense mutations, 26.5% frame shift mutations, 20.5% nonsense mutations and 3% microdeletions. As expected from previous findings (19), no correlation was found between ABCD1 gene mutations and phenotype. The distribution of the ABCD1 gene mutations within each phenotype was as follows: CCER (27.3% frame shift, 36.4% missense, 27.3% nonsense and 9% microdeletion); AMN-C (28.5% frame shift, 57% missense and 14.5% nonsense) and AMN (28.5% frame shift, 43% missense and 28.5% nonsense).

Histopathological examination of ALD and control WM

WM sections of ALD patients and controls (Table 2) were stained with Luxol fast blue (LFB) to detect demyelination, demyelination edge and normal-looking area. Brain tissue sections were processed for RT–PCR amplification when two to three adjacent sections showed no sign of demyelination with LFB staining and no perivascular cuffs of lymphocytes using hematoxylin and eosin (H&E) staining (Fig. 1). However, WM of ALD patients did contain fewer ramified microglia and more non-ramified microglial cells than the control WM when labeling was performed with anti-Iba1 antibody (Fig. 1). This indicates that transition of resting microglia to activated microglia occurs early in the normal WM of ALD patients, including those with AMN.

Expression of the ABCD1, ABCD2, ABCD3, ABCD4, VLCS and BG1 genes in normal fibroblasts and brains

We first analyzed the expression of the ABCD1–4, VLCS and BG1 genes in normal fibroblast and brain samples using quantitative real-time RT–PCR assay (Fig. 2).

In fibroblasts, VLCS-, BG1- and ABCD2-gene mRNA levels were very low, hindering reliable quantification by real-time quantitative RT–PCR (threshold cycle, Ct≈40). The quantitative expression of these three genes was therefore not investigated in ALD fibroblasts. In comparison, the expression of these genes was at least 1000-fold higher in brain samples. ABCD1 and ABCD4 transcripts were detected at low levels and ABCD3 transcript at medium levels in fibroblasts.

There was no difference in the WM expression of the ABCD1, ABCD2, ABCD3, ABCD4, VLCS or BG1 genes in the child and adult controls. The median level of ABCD4-gene expression was 1.4-fold higher (P=0.2) in adult than that in child control WM, and this difference was taken into account for the comparison with ALD samples. In the gray matter (GM), the ABCD1, ABCD3, ABCD4 and BG1 genes were expressed at the same level in child and adult controls. The level of expression of the ABCD2 and VLCS genes was higher (1.4-fold) in the GM of the child controls (P<0.01 and P<0.05, respectively). There was no significant difference in the expression of these genes according to the localization within the brain (P>0.5).

However, each gene was not expressed at the same level in the WM and GM (Fig. 2). The expression of the ABCD1 and ABCD4 genes was moderate with WM/GM expression ratios of 1.6 and 2.5, respectively. The expression of the ABCD2 and VLCS genes was low in WM and moderate in GM. The WM/GM expression ratios of these genes were 0.06 and 0.02, respectively. The ABCD3 and BG1 genes were highly expressed in the brain, at very similar levels in the GM and WM (WM/GM ratio=1.2 and 1.3, respectively).

Expression of the ABCD1, ABCD2, ABCD3, ABCD4, VLCS and BG1 genes in ALD fibroblasts and brains

The ABCD1 gene.

The median level of ABCD1 gene expression was lower in ALD than in control fibroblasts (Table 3). No correlation was found between the level of ABCD1 expression and the phenotype. ABCD1 expression levels were strongly reduced in 4/6 CCER (∼7.3-fold), 2/5 AMN-C (∼9.8-fold) and 3/5 AMN (∼11-fold) fibroblasts. Except for one ALD fibroblast sample with an unknown mutation, all ALD fibroblasts exhibiting a marked decrease in the expression of the ABCD1 gene had frame shift or nonsense mutations leading to truncated proteins.

The median level of ABCD1 gene expression was also reduced in both WM and GM of ALD patients and showed no correlation with phenotype (Table 3). In the WM, expression of the ABCD1 gene was strongly reduced in 3/6 CCER (∼12-fold) and in 3/9 AMN-C (∼14-fold). Similar changes were observed in the GM. These six patients had frame shift or nonsense mutations leading to truncated proteins. The down-regulation of truncated ABCD1 transcripts is likely mediated through nonsense-mediated decay (NMD).

The ABCD2 gene.

There is evidence for functional overlap between ABCD1 and ABCD2 genes in vivo (23). Variation of ABCD2-gene expression in the brains of ALD patients could therefore contribute to modifying the phenotypic expression of the disease. The levels of expression of the ABCD2 gene in normal-appearing WM from patients with CCER, AMN-C and AMN were however similar and no different from the levels observed in age-matched controls (Table 3). The expression of the ABCD2 gene was the same in the GM from controls and ALD patients (Table 3).

The ABCD3 gene.

The levels of ABCD3 gene expression were the same in CCER, AMN-C, AMN and control fibroblasts (Table 3). In the WM, ABCD3-gene expression was slightly reduced 1.3-fold in AMN-C (P=0.03) and AMN samples (Table 3).The expression of the ABCD3 gene was the same in the GM from controls and ALD patients (Table 3).

The ABCD4 gene.

The ABCD4 gene encodes a peroxisomal half-ABC transporter, P70R, which shares only 25.2% amino acid identity with ALDP (6). The function of the P70R protein is however unknown. In fibroblasts, the ABCD4 gene was expressed at the same levels in CCER, AMN-C, AMN and controls (Table 3). In contrast, ABCD4 mRNA levels were significantly reduced in the normal WM of CCER (2.9-fold, P=0.04) and AMN-C (2.2-fold, P=0.01) patients. ABCD4-gene expression was reduced only 1.6-fold in the two AMN samples (Table 3 and Fig. 3). The differences in ABCD4-gene expression among the CCER, AMN-C and AMN samples were not statistically significant. However, ABCD4-gene expression in WM clearly tended to be correlated with ALD phenotype, in contrast to what was observed in fibroblasts. The expression levels of the ABCD4 gene in the GM from patients with CCER, AMN-C and AMN were the same as in age-matched controls (Table 3).

The BG1 gene.

The BG1 gene is expressed in tissues affected by ALD pathology and encodes a non-peroxisomal synthetase which converts VLCFA to its CoA derivatives (32). BG1-mRNA levels were significantly lower in CCER (2-fold) and AMN-C (2.3-fold) than in control WM. In the two AMN samples, BG1 gene expression was decreased only 1.4-fold (Table 3 and Fig. 4). As for the ABCD4 gene, the limited number of samples made it impossible to reach statistical significance, but there was a trend indicating that expression of the BG1 gene was correlated with cerebral demyelination. The levels of expression of the BG1 gene in the GM from patients with CCER, AMN-C and AMN were the same as that in age-matched controls (Table 3).

The VLCS gene.

The VLCS enzyme is the only VLACS identified in peroxisomes. VLCFA-synthetase activity is reduced in peroxisomes from ALD fibroblasts (26,27), and it has long been considered that the in vivo activity of this enzyme may play a role in the accumulation of VLCFA observed in ALD tissues. However, the WM from CCER, AMN-C and AMN patients and age-matched controls expressed the VLCS gene at the same levels (Table 3). In the GM, the expression of the VLCS gene was significantly lower in the CCER and AMN-C samples than in the controls (1.8- and 1.5-fold lower, respectively) (Table 3).

In summary, the expression of the ABCD2 gene remained unchanged in the WM from ALD patients, whatever their phenotype. In contrast, ABCD4-gene expression tended to be correlated with ALD phenotype. The BG1-gene expression tended to be correlated with the presence or absence of cerebral demyelination rather than phenotype per se. There was no modification of the expression of genes of which the mRNA levels could be reliably quantified in fibroblasts. In the GM, none of the genes studied showed any significant change in expression, apart from the VLCS gene.

VLCFA concentrations in the normal-appearing WM of ALD patients with CCER, AMN-C and AMN phenotypes

Saturated and mono-unsaturated VLCFA levels were the same in the WM from child and adult controls (data not shown). Saturated C24 : 0 levels remained unchanged in the normal WM of all ALD patients (data not shown). In contrast, saturated C26 : 0–C30 : 0 fatty acid levels were increased in the normal WM from all ALD patients. In comparison with control values, C26 : 0 levels were increased 3-fold in CCER normal white matter and 1.9-fold in AMN-C or AMN normal WM (Fig. 5A).

Concentrations of the mono-unsaturated fatty acids C22 : 1, C24 : 1 (Fig. 5B) and C25 : 1 were decreased 2-fold in CCER, 1.4-fold in AMN-C and remained unchanged in AMN samples. The saturated VLCFA that accumulate in ALD are mostly of endogenous origin (33,34) and elongation of saturated VLCFA is enhanced in ALD fibroblasts (35). The reduction of mono-unsaturated VLCFA observed in the WM from CCER and AMN-C patients may reflect substrate competition between saturated and mono-unsaturarted long chain fatty acids for the two elongases (ELOVL1 and ELOVL3) that display chain-length specificity toward VLCFA (36). Both saturated and unsaturated fatty acids are chain-lengthened by the same elongation system (36,37).

DISCUSSION

The phenotypic expression of ALD varies widely, and there are fundamental clinical and neuropathological differences between the cerebral and the AMN phenotypes (1). Because ALDP interacts both with itself and with other peroxisomal ABC transporters, it has been suggested that ABCD1 gene mutations causing premature termination of translation could generate a truncated ALDP with dominant-negative activity that modulates the severity of ALD disease. The existence of such a mechanism has been suggested in a large ALD kindred, in which all female carriers exhibited complete penetrance of an ABCD1 gene mutation (delta26, 369–394) resulting in an N-terminal truncated ALDP lacking the first 65 amino acids (38). Our data indicate that ABCD1 gene mutations leading to truncated ALDP are unlikely able to cause variations in the clinical ALD phenotype. ABCD1 gene expression was strongly decreased in brain and fibroblasts from all ALD patients with frame shift, nonsense and insertion/deletion mutations. Marked down-regulation of dominant-negative ABCD1 transcripts can be expected to produce a considerable reduction in the quantity of truncated ALDP, preventing adverse effects on other peroxisomal ABC transporters. With the exception of the delta26 mutation, this is consistent with the observation that all the frame shift, nonsense and insertion/deletion mutations of ABCD1 gene so far identified result in the absence of detectable levels of ALDP in fibroblasts (39). The down-regulation of dominant-negative ABCD1 transcripts is probably mediated by the NMD RNA surveillance pathway that reduces the abundance of transcripts containing a premature termination codon (40). The protective effects of the NMD pathway against the production of faulty proteins have been suggested or demonstrated in several other hereditary disorders, including β-thalassemia (40).

In the absence of a mouse model of ALD reproducing the phenotype variability observed in ALD patients, the search for ALD-modifying genes remains particularly challenging. We focused our study on two sets of genes because of a priori hypotheses about their etiological role in ALD: the family of peroxisomal ABC transporters to which the ABCD1 gene belongs and two VLCAS which play a role in VLCFA metabolism. We made the assumption that if any of these genes play a role in the expression of phenotype, then differences in the expression of these genes should be detectable in the normal WM of ALD patients with different phenotypes. The limited number of brain samples did not allow to reach the power of association studies using DNA variants. However, our study allows to identify two genes of which the levels of expression tend to be correlated with the severity of the disease or phenotypes. However, our study did not allow to clarify whether these changes were the consequence or the cause of disease severity/phenotype variability.

The ABCD2 gene is mainly expressed in neurons and its expression is not modified in the ALD mouse (23,41). However, ubiquitous up-regulation of the ABCD2 gene can compensate for the defective function of the ABCD1 gene in the ALD mouse that develops an AMN-like phenotype resulting from neuro-axonal degeneration in the spinal cord and peripheral nerves (23). The level of ABCD2 transcripts was the same in the normal WM from patients with different ALD phenotypes and did not differ from that in controls. It is therefore unlikely that variation in ABCD2-gene expression contributes to the susceptibility of developing cerebral demyelination. This is in agreement with the absence of ABCD2 allele/phenotype co-segregation in a large ALD kindred (42).

The over-expression of ABCD3 also restores VLCFA ß-oxidation in human ALD fibroblasts and in the liver of ALD mouse, indicating that the functions of the PMP70 and ALDP overlap (43,44). Abcd3 −/− mice exhibit impaired peroxisomal β-oxidation of phytanic and pristanic acids, but with normal VLCFA metabolism (45). Gene-microarray analysis of multiple sclerosis lesions showed that the expression of ABCD3 is reduced 1.7–2.4-fold in chronic or acute plaques (46). The expression of the ABCD3 gene was not modified in the normal WM of ALD patients. Like the ABCD2 gene, it is unlikely that the ABCD3 gene acts as a modifier gene in ALD through modification of its expression. However, this study does not exclude the possibility that subtle changes in PMP70 activity might interfere with the severity of the disease. In addition, the expression of the ABCD3 gene in demyelinated lesions of ALD warrants further study.

A current hypothesis is still that ALDP is involved in some manner with the transport of VLCFA in peroxisomes. The activity of the enzyme that catalyzes the first step in VLCFA β-oxidation, i.e. the activation of VLCFA to their CoA derivatives, is decreased in peroxisomes from ALD fibroblasts (26,27). This step is catalyzed by the VLCS enzyme. However, ALDP does not itself have VLACS activity. How the loss of ALDP function results in reduced peroxisomal VLACS activity is unknown. The absence of any change in the expression of the VLCS gene in the normal WM of CCER, AMN-C and AMN patients is similar to that observed in ALD mouse tissues (32,47). Because of its very low level of expression in fibroblasts, we were not able to reliably quantify the expression of the VLCS gene in ALD fibroblasts, but others have reported no substantial modification of its expression in CCER and AMN fibroblasts (32). Taken together, these data indicate that VLCS-gene expression is unlikely to contribute to ALD-phenotype variability. This concurs with the study of the ALD/Vlcs double knockout mouse that does not display a more severe phenotype than the ALD mouse (48).

In contrast, the expression of the ABCD4 and BG1 genes did correlate with the susceptibility to develop cerebral demyelination, and there was a trend indicating that ABCD4-gene expression in the normal WM correlated with the CCER, AMN-C and AMN phenotypes. In the AMN brain samples, the levels of expression of the ABCD4 and BG1 genes were not however any higher than the mean values observed in the WM of adult controls. It is therefore not clear whether the modifications in the expression of ABCD4 and BG1 genes result from metabolic changes or are primary determinants of the ALD phenotype. Little is known about ABCD4 gene function. In contrast to ALDRP and PMP70, we have been unable to demonstrate the heterodimerization of its products (P70R) with ALDP by using the two-hybrid system and co-immunoprecipitation (unpublished data). Significantly, the ABCD4 gene is the only peroxisomal ABC transporter gene of which the expression parallels the expression of the ABCD1 gene in brain, with higher expression in the WM than in the GM. ABCD4-gene function deserves clearly further study to determine its role in ALD pathogenesis.

The expression of the BG1 gene is not modified in the brain and spinal cord of the ALD mouse (32,49). In accordance with these results, the expression of BG1 was decreased only 1.4-fold in the normal WM from AMN patients. BG1 transcripts were however significantly reduced in the WM from CCER and AMN-C patients. In human ALD brain, VLCFA accumulate in the cholesterol ester (ChE) fraction (1), and the rate of the reverse reaction is exceedingly low (50). The formation of VLCFA-containing ChE could not only trap the VLCFA leading to increased levels, but also impair cellular processes at the plasma membrane by modifying the cholesterol/ChE ratio (51). From studies performed in the mouse neuroblastoma cell line Neuro2a, it was concluded that BG1 is not required for the incorporation of C24:0 into ChE. It is noteworthy that saturated C24 : 0 does not accumulate in the normal WM from CCER, AMN-C and AMN patients. However, levels of C26 : 0–C30 : 0 fatty acids increase 2–3-fold. The down-regulation of BG1-gene expression may reflect attempts to reduce the trapping of VLCFA in ChE and the deleterious effects that result from imbalance of the cholesterol/CheE ratio at the plasma membrane. Down-regulation of the BG1 gene may also decrease the incorporation of VLCFA into complex lipids, such as gangliosides. The presence of excess VLCFA in gangliosides is thought to trigger the inflammatory response that occurs in cerebral ALD (5254). Association studies to determine whether genetic variants of the BG1 and ABCD4 genes are correlated with CCER, AMN-C and AMN phenotypes are under investigation.

No correlation has been found between the phenotype and the levels of saturated VLCFA in ALD fibroblasts or lymphocytes (55,56). However, a relationship does exist between the degree of VLCFA accumulation and the levels of inflammatory cytokine/gene expression in demyelinated areas (57). VLCFA accumulation was shown to be detrimental to adrenocortical cells in culture (58) and to model membrane (59), but the pathogenic role of saturated VLCFA accumulation remains elusive. This has led to envisage that the high tissue VLCFA levels observed in mouse and human ALD could be a secondary biochemical event and that the relationship between ALDP function and VLCFA metabolism is indirect (41). Saturated VLCFA accumulate in the normal WM of ALD patients before the onset of any detectable demyelinating process, as reported recently by others (57). More importantly, our data demonstrate for the first time that the accumulation of saturated VLCFA and the compensating decrease in mono-unsaturated VLCFA in the WM correlate with the ALD phenotype. These results indicate that the threshold of VLCFA accumulation in the WM probably plays a crucial role in the initial demyelinating process. The incorporation of VLCFA in components of the multilamellar myelin sheaths might indeed destabilize ALD myelin (59). An unknown triggering factor may first alter the ALD myelin, which could make it more or less sensitive to breakdown because of the levels of VLCFA accumulation. Subtle changes in intracellular VLCFA concentrations may also influence the lipid pathways to which these fatty acids are assigned, resulting in decreased survival of oligodendrocytes or abnormal interactions between glial cells.

MATERIALS AND METHODS

Tissue and cell-line specimens

Skin fibroblasts from six CCER, five AMN-C, five AMN and 10 control subjects (Table 4) were kindly provided by Professor R.J.A Wanders (AMC, Amsterdam, The Netherlands).

Brain tissues from ALD patients and age-matched controls were obtained from the Brain and Tissue Bank for Developmental Disorders at the University of Maryland, Baltimore, MD, USA. Frozen blocks of normal-appearing GM and WM were dissected from frontal, parietal or occipital lobes from 19 controls and 17 ALD patients and used for mRNA extraction, biochemical studies and mutational analysis (Table 2). The control group included nine children (4.5–12 years) and 10 adults (27–47 years) who died from non-neurological diseases. The median age at death was not different between the controls and the ALD patients (P=0.2). The postmortem interval up to autopsy was shorter for the ALD patients than for the controls (9±7.5 and 15±3.8, P=0.06). ALD brain tissues were obtained from six children with CCER, nine adults with AMN-C and two adults with AMN, but no cerebral demyelination. Apart from one patient (AMN-C4), all the children and adults with cerebral ALD had the conventional ‘parieto-occipital’ form of cerebral ALD.

Histopathology

Tissue slices were fixed in 10% buffered formalin, embedded in paraffin and sectioned at 4 µm thickness. Then sections were examined using standard histological stains H&E and LFB. Rabbit anti-ionized calcium-binding adapter molecule 1 (Iba1) antibody (dilution 1 : 400, Imai, Japan (60)) was used to detect microglial activation.

Mutation analysis

Screening for ABCD1 mutations was done using direct sequencing of genomic DNA extracted from brain samples. Both sense and anti-sense strands were systematically sequenced using an ABI 3100 capillary electrophoresis analyzer (Applied Biosystems, Applera). Intronic and/or exonic sets of oligonucleotide primers were designed according to the published genomic sequence of the ABCD1 gene (GenBank accession no. U52111). The oligonucleotide primers and PCR conditions (‘touch down’ PCR) are available from the authors upon request.

ABCD1-gene mutations in the fibroblasts were identified as previously described (61).

Real-time RT–PCR

The theoretical and practical aspects of real-time quantitative RT–PCR using the ABI Prism 7700 Sequence Detection System (Perkin–Elmer Applied Biosystems) have been described in detail elsewhere (62). Briefly, total RNA is reverse-transcribed before real-time PCR amplification. Quantitative values are obtained from the Ct number at which the increase in the signal associated with exponential growth of PCR products begins to be detected using PE Biosystems analysis software, according to the manufacturer's manuals.

The precise amount of total RNA added to each reaction mix (based on optical density) and its quality (i.e. lack of extensive degradation) are both difficult to assess. We therefore also quantified transcripts of the two endogenous RNA controls RPLP0 (also known as 36B4) which encodes human acidic ribosomal phosphoprotein P0 and PPIA (peptidylprolyl isomerase A) that encodes cyclophilin A, and each sample was normalized on the basis of its RPLP0 (or PPIA) content. The ratio of RPLP0/PPIA expression did not show significant variation in the WM and GM or fibroblast of controls and ALD patients (data not shown).

The results were therefore expressed as N-fold differences in target gene expression relative to the RPLP0 gene, termed Ntarget, and determined by the formula: Ntarget=2ΔCtsample, where the ΔCt value of the sample was determined by subtracting the Ct value of the target gene from the Ct value of the RPLP0 gene.

The Ntarget values of the samples were subsequently normalized to a ‘basal mRNA level’, i.e. normalized to the smallest amount of target gene mRNA detectable and quantifiable by real-time quantitative RT–PCR assays based on fluorescence TaqMan® methodology (target gene Ct value=40; Ntarget value=1).

Primers and probes for RPLP0, PPIA and the six target genes were chosen with the assistance of the Oligo 5.0 computer program (National Biosciences, Plymouth, MN, USA). The nucleotide sequences of primers and oligonucleotide hybridization probes are available from the authors upon request.

To avoid amplification of contaminating genomic DNA, one of the two primers was placed at the junction between two exons. The thermal cycling conditions comprised an initial denaturation step at 95°C for 10 min and 50 cycles at 95°C for 15 s and 65°C for 1 min.

Statistical analysis

The non-parametric Mann–Whitney U-test (Statview 1996, Abacus Concepts, Inc., Berkeley, CA, USA) was used to evaluate the variation in gene expression between samples. Differences between two set of data were judged significant at confidence levels >95% (P<0.05).

Fatty acid analysis of brain samples

Frozen blocks of normal-appearing WM (frontal lobes) from two CCER, five AMN-C, two AMN and five age-matched controls were used to determine the fatty acid content. Samples were pounded and homogenized by sonication in 1 ml of PBS. Total lipids were extracted using 2 : 1 (vol/vol) chloroform/methanol and the Folch method (63). VLCFA concentrations were measured/analyzed by gas chromatography/mass spectrometry as described (64).

ACKNOWLEDGEMENTS

We thank P. Bougnères for his commitment to the study, R. Vigorito and R. Zielke from the Brain and Tissue Bank for Developmental Disorders (University of Maryland, Baltimore, MD, USA) for their collaboration and R.J.A. Wanders and H.R. Waterham for providing ALD fibroblasts (AMC, Amsterdam, The Netherlands). We also thank S. Guidoux and F. Fouquet for their technical assistance. This work was funded by grants from the EU-project LSHM-CT-2004-502987 and the European Leukodystrophy Association (ELA).

Figure 1. Representative ‘normal-appearing’ white matter from ALD patients and controls. LFB and H&E staining showed no significant difference between the ‘normal-appearing’ white matter from ALD patients and the controls. Staining with the microglial marker Iba1 (in red, Cy3) reveals the same number of microglia in white matter from ALD patients and controls. However, the microglial cells from the ALD white matter are less ramified.

Figure 1. Representative ‘normal-appearing’ white matter from ALD patients and controls. LFB and H&E staining showed no significant difference between the ‘normal-appearing’ white matter from ALD patients and the controls. Staining with the microglial marker Iba1 (in red, Cy3) reveals the same number of microglia in white matter from ALD patients and controls. However, the microglial cells from the ALD white matter are less ramified.

Figure 2. Expression of the ABCD1, ABCD2, ABCD3, ABCD4, VLCS and BG1 genes in normal WM, GM and fibroblasts. mRNA levels were measured by quantitative RT–PCR and normalized to RPLP0 and to the sample that displayed the lowest detectable and quantifiable level of gene expression (the value 1 on the left ordinate corresponds to a Ct value of 40). A scale indicating qualitative mRNA abundance is included on the right ordinate to assist the reader in interpreting the figure.

Figure 2. Expression of the ABCD1, ABCD2, ABCD3, ABCD4, VLCS and BG1 genes in normal WM, GM and fibroblasts. mRNA levels were measured by quantitative RT–PCR and normalized to RPLP0 and to the sample that displayed the lowest detectable and quantifiable level of gene expression (the value 1 on the left ordinate corresponds to a Ct value of 40). A scale indicating qualitative mRNA abundance is included on the right ordinate to assist the reader in interpreting the figure.

Figure 3.ABCD4-gene expression in the WM from ALD patients with CCER, AMN-C and AMN phenotypes. The levels of ABCD4 mRNA are decreased in the WM from CCER and AMN-C (*P<0.05).

Figure 3.ABCD4-gene expression in the WM from ALD patients with CCER, AMN-C and AMN phenotypes. The levels of ABCD4 mRNA are decreased in the WM from CCER and AMN-C (*P<0.05).

Figure 4.BG1-gene expression in the WM from ALD patients with CCER, AMN-C and AMN phenotypes. The levels of BG1 mRNA are decreased in the WM from CCER and AMN-C (*P<0.01).

Figure 4.BG1-gene expression in the WM from ALD patients with CCER, AMN-C and AMN phenotypes. The levels of BG1 mRNA are decreased in the WM from CCER and AMN-C (*P<0.01).

Figure 5. VLCFAs levels in the normal WM from ALD patients. (A) Saturated VLCFAs (C26 : 0–C30 : 0) levels in the WM of patients with CCER, AMN-C, AMN and controls. Results are expressed in µg/mg protein. (B) Nervonic acid (C24 : 1) levels in the white matter of patients with CCER, AMN-C, AMN phenotypes and controls. Results are expressed in µg/mg protein.

Figure 5. VLCFAs levels in the normal WM from ALD patients. (A) Saturated VLCFAs (C26 : 0–C30 : 0) levels in the WM of patients with CCER, AMN-C, AMN and controls. Results are expressed in µg/mg protein. (B) Nervonic acid (C24 : 1) levels in the white matter of patients with CCER, AMN-C, AMN phenotypes and controls. Results are expressed in µg/mg protein.

Table 1.

Mutations identified in ALD patients

Patient No. Mutation Amino acid alteration Type of mutation at the protein level Tissue sample 
CCER1 521A>G Y174C Missense  
CCER2 1414insC fsE471 Frame shift  
CCER3 Unknown Unknown Unknown Fibroblast 
CCER4 411G>A W137X Nonsense  
CCER5 1961T>C L654P Missense  
CCER6 529C>T Q177X Nonsense  
CCER7 901-1G>A fsE300 Frame shift  
CCER8 796G>A G266R Missense  
CCER9 1822G>A G608S Missense Brain 
CCER10 1390C>A R464X Nonsense  
CCER11 253-254insC fsP84 Frame shift  
CCER12 619_627del S207_A209del Deletion  
AMN-C1 1414-1415insC fsE471 Frame shift  
AMN-C2 1661G>A R554H Missense  
AMN-C3 1585delG fsG528 Frame shift Fibroblast 
AMN-C4 1661G>A R554H Missense  
AMN-C5 1825G>A E609K Missense  
AMN-C6 919C>T Q307X Nonsense  
AMN-C7 1850G>A R617H Missense  
AMN-C8 887A>G Y296C Missense  
AMN-C9 965T>C L322P Missense Brain 
AMN-C10 1390C>T R464X Nonsense  
AMN-C11 [1165C>T;1224+1GT>TG] [R389C;fSE408] Missense; frame shift  
AMN-C12 1661G>A R554H Missense  
AMN-C13 [1997A>C;2007C>G] [Y666S;H669Q] Missense  
AMN-C14 1755delG fsH586 Frame shift  
AMN1 529C>T Q177X Nonsense  
AMN2 1999C>G H667D Missense  
AMN3 1415delAG fsE471 Frame shift Fibroblast 
AMN4 337delC fsA112 Frame shift  
AMN5 310C>T R104C Missense  
AMN6 919C>T Q307X Nonsense  
AMN7 323C>T S108L Missense Brain 
Patient No. Mutation Amino acid alteration Type of mutation at the protein level Tissue sample 
CCER1 521A>G Y174C Missense  
CCER2 1414insC fsE471 Frame shift  
CCER3 Unknown Unknown Unknown Fibroblast 
CCER4 411G>A W137X Nonsense  
CCER5 1961T>C L654P Missense  
CCER6 529C>T Q177X Nonsense  
CCER7 901-1G>A fsE300 Frame shift  
CCER8 796G>A G266R Missense  
CCER9 1822G>A G608S Missense Brain 
CCER10 1390C>A R464X Nonsense  
CCER11 253-254insC fsP84 Frame shift  
CCER12 619_627del S207_A209del Deletion  
AMN-C1 1414-1415insC fsE471 Frame shift  
AMN-C2 1661G>A R554H Missense  
AMN-C3 1585delG fsG528 Frame shift Fibroblast 
AMN-C4 1661G>A R554H Missense  
AMN-C5 1825G>A E609K Missense  
AMN-C6 919C>T Q307X Nonsense  
AMN-C7 1850G>A R617H Missense  
AMN-C8 887A>G Y296C Missense  
AMN-C9 965T>C L322P Missense Brain 
AMN-C10 1390C>T R464X Nonsense  
AMN-C11 [1165C>T;1224+1GT>TG] [R389C;fSE408] Missense; frame shift  
AMN-C12 1661G>A R554H Missense  
AMN-C13 [1997A>C;2007C>G] [Y666S;H669Q] Missense  
AMN-C14 1755delG fsH586 Frame shift  
AMN1 529C>T Q177X Nonsense  
AMN2 1999C>G H667D Missense  
AMN3 1415delAG fsE471 Frame shift Fibroblast 
AMN4 337delC fsA112 Frame shift  
AMN5 310C>T R104C Missense  
AMN6 919C>T Q307X Nonsense  
AMN7 323C>T S108L Missense Brain 

All mutation designations conform to the nomenclature described by Antonarakis and den Dunnen (30,31). Nucleotides are numbered from the first ATG (position 387), according to the cDNA sequence of GenBank accession no. NM_000033. For frame shift (fs) mutations, the last unchanged amino acid is indicated.

Table 2.

Clinical characteristics of ALD patients and controls from whom brain samples were obtained

Phenotype No. Age of death Age of onset Gender PMI Brain localization 
Child control Cc6 4.5 — 18 Frontal 
(n=9) Cc7 4.5 — 15 Frontal 
 Cc8 4.5 — 17 Parietal 
 Cc9 — 20 Parietal 
 Cc10 5.5 — 17 Frontal 
 Cc11 12 — 16 Frontal 
 Cc12 13 — 18 Frontal 
 Cc13 — 21 Parietal 
 Cc14 9.5 — 17 Parietal 
Adult control Ac6 27 — 10 Frontal 
(n=10) Ac7 28 — Parietal 
 Ac8 31 — 15 Frontal 
 Ac9 37 — 12 Parietal 
 Ac10 38 — 11 Frontal 
 Ac11 40 — 23 Parietal 
 Ac12 42 — 15 Frontal 
 Ac13 43 — Frontal 
 Ac14 46.5 — Parietal 
 Ac15 47 — 13 Frontal 
CCER (n=6) CCER7 12 Frontal 
 CCER8 Frontal 
 CCER9 4.5 Frontal 
 CCER10 10 8.5 Frontal 
 CCER11 13 10 14 Frontal 
 CCER12 13 Frontal 
AMN-Csss AMN-C6 28 24 11 Frontal 
(n=9) AMN-C7 28.5 22 18 Frontal 
 AMN-C8 33 22 Frontal 
 AMN-C9 39 36 Frontal 
 AMN-C10 39 28 Frontal 
 AMN-C11 39 36 Frontal 
 AMN-C12 43 31 Parietal 
 AMN-C13 47 42 24 Frontal 
 AMN-C14 48 46 14 Parietal 
AMN AMN6 30 27 40 Frontal 
(n=2) AMN7 47 25 22 Frontal 
Phenotype No. Age of death Age of onset Gender PMI Brain localization 
Child control Cc6 4.5 — 18 Frontal 
(n=9) Cc7 4.5 — 15 Frontal 
 Cc8 4.5 — 17 Parietal 
 Cc9 — 20 Parietal 
 Cc10 5.5 — 17 Frontal 
 Cc11 12 — 16 Frontal 
 Cc12 13 — 18 Frontal 
 Cc13 — 21 Parietal 
 Cc14 9.5 — 17 Parietal 
Adult control Ac6 27 — 10 Frontal 
(n=10) Ac7 28 — Parietal 
 Ac8 31 — 15 Frontal 
 Ac9 37 — 12 Parietal 
 Ac10 38 — 11 Frontal 
 Ac11 40 — 23 Parietal 
 Ac12 42 — 15 Frontal 
 Ac13 43 — Frontal 
 Ac14 46.5 — Parietal 
 Ac15 47 — 13 Frontal 
CCER (n=6) CCER7 12 Frontal 
 CCER8 Frontal 
 CCER9 4.5 Frontal 
 CCER10 10 8.5 Frontal 
 CCER11 13 10 14 Frontal 
 CCER12 13 Frontal 
AMN-Csss AMN-C6 28 24 11 Frontal 
(n=9) AMN-C7 28.5 22 18 Frontal 
 AMN-C8 33 22 Frontal 
 AMN-C9 39 36 Frontal 
 AMN-C10 39 28 Frontal 
 AMN-C11 39 36 Frontal 
 AMN-C12 43 31 Parietal 
 AMN-C13 47 42 24 Frontal 
 AMN-C14 48 46 14 Parietal 
AMN AMN6 30 27 40 Frontal 
(n=2) AMN7 47 25 22 Frontal 

PMI, post-mortem interval up to autopsy (in hours).

Table 3.

mRNA levels of ABCD1, ABCD2, ABCD3, ABCD4, BG1 and VLCS in fibroblast, WM and GM from ALD patients and controls

Gene (I) Child controlmedian±SD (II) Adult controlmedian±SD (III) CCERmedian±SD (IV) AMN-Cmedian±SD (V) AMNmedian±SD P-value Mann–Whitney test 
      III/I or III/I+II IV/II or IV/I+II V/I+II 
Fibroblast         
ABCD1 1±0.22 1±0.22 0.18±0.38 0.51±0.47 0.13±0.57 0.03 ns ns 
ABCD3 1±0.46 1±0.43 0.82±0.32 1.24±0.27 0.84±0.17 ns ns ns 
ABCD4 1±0.31 1±0.25 0.74±0.31 0.85±0.20 0.89±0.24 ns ns ns 
WM         
ABCD1 1±0.22 1±0.22 0.31±0.27 0.51±0.33 0.40±0.02 0.0006 0.003 na 
ABCD2 1±0.33 1±0.50 0.89±0.23 0.79±0.39 0.75±0.29 ns ns na 
ABCD3 1±0.19 1±0.18 0.87±0.09 0.73±0.28 0.73±0.21 ns 0.03 na 
ABCD4 1±0.36 1±0.20 0.34±0.20 0.45±0.25 0.61±0.08 0.04a 0.01a na 
BG1 1±0.35 1±0.42 0.50±0.16 0.43±0.12 0.68±0.29 0.003 0.0002 na 
VLCS 1±0.44 1±0.46 1.20±0.38 1.70±0.66 0.70±0.29 ns ns na 
GM         
ABCD1 1±0.25 1±0.44 0.41±0.26 0.68±0.41 0.41±0.27 0.002 0.02 na 
ABCD2 1±0.45 1±0.33 0.76±0.11 0.86±0.26 1.07±0.10 0.01a nsa na 
ABCD3 1±0.25 1±0.13 1.04±0.11 0.87±0.33 0.87±0.25 ns ns na 
ABCD4 1±0.32 1±0.52 0.78±0.25 1.12±0.46 1.69±0.21 ns ns na 
BG1 1±0.31 1±0.79 0.74±0.10 0.66±0.80 1.05±0.06 ns ns na 
VLCS 1±0.09 1±0.19 0.54±0.12 0.65±0.25 0.90±0.16 0.002a 0.01a na 
Gene (I) Child controlmedian±SD (II) Adult controlmedian±SD (III) CCERmedian±SD (IV) AMN-Cmedian±SD (V) AMNmedian±SD P-value Mann–Whitney test 
      III/I or III/I+II IV/II or IV/I+II V/I+II 
Fibroblast         
ABCD1 1±0.22 1±0.22 0.18±0.38 0.51±0.47 0.13±0.57 0.03 ns ns 
ABCD3 1±0.46 1±0.43 0.82±0.32 1.24±0.27 0.84±0.17 ns ns ns 
ABCD4 1±0.31 1±0.25 0.74±0.31 0.85±0.20 0.89±0.24 ns ns ns 
WM         
ABCD1 1±0.22 1±0.22 0.31±0.27 0.51±0.33 0.40±0.02 0.0006 0.003 na 
ABCD2 1±0.33 1±0.50 0.89±0.23 0.79±0.39 0.75±0.29 ns ns na 
ABCD3 1±0.19 1±0.18 0.87±0.09 0.73±0.28 0.73±0.21 ns 0.03 na 
ABCD4 1±0.36 1±0.20 0.34±0.20 0.45±0.25 0.61±0.08 0.04a 0.01a na 
BG1 1±0.35 1±0.42 0.50±0.16 0.43±0.12 0.68±0.29 0.003 0.0002 na 
VLCS 1±0.44 1±0.46 1.20±0.38 1.70±0.66 0.70±0.29 ns ns na 
GM         
ABCD1 1±0.25 1±0.44 0.41±0.26 0.68±0.41 0.41±0.27 0.002 0.02 na 
ABCD2 1±0.45 1±0.33 0.76±0.11 0.86±0.26 1.07±0.10 0.01a nsa na 
ABCD3 1±0.25 1±0.13 1.04±0.11 0.87±0.33 0.87±0.25 ns ns na 
ABCD4 1±0.32 1±0.52 0.78±0.25 1.12±0.46 1.69±0.21 ns ns na 
BG1 1±0.31 1±0.79 0.74±0.10 0.66±0.80 1.05±0.06 ns ns na 
VLCS 1±0.09 1±0.19 0.54±0.12 0.65±0.25 0.90±0.16 0.002a 0.01a na 

For each gene, results were normalized so that the median of the child or adult control values equalled 1. SD, standard deviation; ns, not statistically significant; na, not applicable (number of patients=2).

aFor ABCD4 gene in the WM, ABCD2 and VLCS genes in the GM the Mann–Whitney test was performed between CCER versus child control and AMN-C versus adult control, because the expression of these genes were not identical in child and adult controls.

Table 4.

Clinical characteristics of ALD patients and controls from whom fibroblast samples were obtained

Phenotype No. Age at sampling Gender 
Child control (n=5) Cc1 10 
 Cc2 
 Cc3 
 Cc4 
 Cc5 
Adult control (n=5) Ac1 29 
 Ac2 30 
 Ac3 32 
 Ac4 40 
 Ac5 25 
CCER (n=6) CCER1 
 CCER2 
 CCER3 10 
 CCER4 
 CCER5 10 
 CCER6 
AMN-C (n=5) AMN-C1 36 
 AMN-C2 41 
 AMN-C3 25 
 AMN-C4 30 
 AMN-C5 16 
AMN (n=5) AMN1 38 
 AMN2 40 
 AMN3 33 
 AMN4 32 
 AMN5 28 
Phenotype No. Age at sampling Gender 
Child control (n=5) Cc1 10 
 Cc2 
 Cc3 
 Cc4 
 Cc5 
Adult control (n=5) Ac1 29 
 Ac2 30 
 Ac3 32 
 Ac4 40 
 Ac5 25 
CCER (n=6) CCER1 
 CCER2 
 CCER3 10 
 CCER4 
 CCER5 10 
 CCER6 
AMN-C (n=5) AMN-C1 36 
 AMN-C2 41 
 AMN-C3 25 
 AMN-C4 30 
 AMN-C5 16 
AMN (n=5) AMN1 38 
 AMN2 40 
 AMN3 33 
 AMN4 32 
 AMN5 28 

References

1
Moser, H.W., Smith, K.D., Watkins, P.A., Powers, J. and Moser, A.B. (
2001
) X-linked adrenoleukodystrophy. In Scriver, C.R. (ed.),
The Metabolic & Molecular Bases of Inherited Disease
 , 8th edn. McGraw-Hill, New York, Vol.
II
, pp.
3257
–3301.
2
Dubois-Dalcq, M., Feigenbaum, V. and Aubourg, P. (
1999
) The neurobiology of X-linked adrenoleukodystrophy, a demyelinating peroxisomal disorder.
Trends Neurosci.
 ,
22
,
4
–12.
3
Mosser, J., Douar, A.M., Sarde, C.O., Kioschis, P., Feil, R., Moser, H., Poustka, A.M., Mandel, J.L. and Aubourg, P. (
1993
) Putative X-linked adrenoleukodystrophy gene shares unexpected homology with ABC transporters.
Nature
 ,
361
,
726
–730.
4
Holzinger, A., Kammerer, S., Berger, J. and Roscher, A.A. (
1997
) cDNA cloning and mRNA expression of the human adrenoleukodystrophy related protein (ALDRP), a peroxisomal ABC transporter.
Biochem. Biophys. Res. Commun.
 ,
239
,
261
–264.
5
Gartner, J., Moser, H. and Valle, D. (
1992
) Mutations in the 70K peroxisomal membrane protein gene in Zellweger syndrome.
Nat. Genet.
 ,
1
,
16
–23.
6
Holzinger, A., Kammerer, S. and Roscher, A.A. (
1997
) Primary structure of human PMP69, a putative peroxisomal ABC-transporter.
Biochem. Biophys. Res. Commun.
 ,
237
,
152
–157.
7
Shani, N., Jimenez-Sanchez, G., Steel, G., Dean, M. and Valle, D. (
1997
) Identification of a fourth half ABC transporter in the human peroxisomal membrane.
Hum. Mol. Genet.
 ,
6
,
1925
–1931.
8
Berger, J., Molzer, B., Fae, I. and Bernheimer, H. (
1994
) X-linked adrenoleukodystrophy (ALD): a novel mutation of the ALD gene in 6 members of a family presenting with 5 different phenotypes.
Biochem. Biophys. Res. Commun.
 ,
205
,
1638
–1643.
9
Pujol, A., Hindelang, C., Callizot, N., Bartsch, U., Schachner, M. and Mandel, J.L. (
2002
) Late onset neurological phenotype of the X-ALD gene inactivation in mice: a mouse model for adrenomyeloneuropathy.
Hum. Mol. Genet.
 ,
11
,
499
–505.
10
Lu, J.F., Lawler, A.M., Watkins, P.A., Powers, J.M., Moser, A.B., Moser, H.W. and Smith, K.D. (
1997
) A mouse model for X-linked adrenoleukodystrophy.
Proc. Natl Acad. Sci. USA
 ,
94
,
9366
–9371.
11
Kobayashi, T., Shinnoh, N., Kondo, A. and Yamada, T. (
1997
) Adrenoleukodystrophy protein-deficient mice represent abnormality of very long chain fatty acid metabolism.
Biochem. Biophys. Res. Commun.
 ,
232
,
631
–636.
12
Forss-Petter, S., Werner, H., Berger, J., Lassmann, H., Molzer, B., Schwab, M.H., Bernheimer, H., Zimmermann, F. and Nave, K.A. (
1997
) Targeted inactivation of the X-linked adrenoleukodystrophy gene in mice.
J. Neurosci. Res.
 ,
50
,
829
–843.
13
Ewart, G.D., Cannell, D., Cox, G.B. and Howells, A.J. (
1994
) Mutational analysis of the traffic ATPase (ABC) transporters involved in uptake of eye pigment precursors in Drosophila melanogaster. Implications for structure-function relationships.
J. Biol. Chem.
 ,
269
,
10370
–10377.
14
Hettema, E.H., van Roermund, C.W., Distel, B., van den Berg, M., Vilela, C., Rodrigues-Pousada, C., Wanders, R.J. and Tabak, H.F. (
1996
) The ABC transporter proteins Pat1 and Pat2 are required for import of long-chain fatty acids into peroxisomes of Saccharomyces cerevisiae.
EMBO J.
 ,
15
,
3813
–3822.
15
Shani, N., Watkins, P.A. and Valle, D. (
1995
) PXA1, a possible Saccharomyces cerevisiae ortholog of the human adrenoleukodystrophy gene.
Proc. Natl Acad. Sci. USA
 ,
92
,
6012
–6016.
16
Swartzman, E.E., Viswanathan, M.N. and Thorner, J. (
1996
) The PAL1 gene product is a peroxisomal ATP-binding cassette transporter in the yeast Saccharomyces cerevisiae.
J. Cell Biol.
 ,
132
,
549
–563.
17
Shani, N., Sapag, A., Watkins, P.A. and Valle, D. (
1996
) An S. cerevisiae peroxisomal transporter, orthologous to the human adrenoleukodystrophy protein, appears to be a heterodimer of two half ABC transporters: Pxa1p and Pxa2p.
Ann. N.Y. Acad. Sci.
 ,
804
,
770
–772.
18
Liu, L.X., Janvier, K., Berteaux-Lecellier, V., Cartier, N., Benarous, R. and Aubourg, P. (
1999
) Homo- and heterodimerization of peroxisomal ATP-binding cassette half-transporters.
J. Biol. Chem.
 ,
274
,
32738
–32743.
19
Smith, K.D., Kemp, S., Braiterman, L.T., Lu, J.F., Wei, H.M., Geraghty, M., Stetten, G., Bergin, J.S., Pevsner, J. and Watkins, P.A. (
1999
) X-linked adrenoleukodystrophy: genes, mutations, and phenotypes.
Neurochem. Res.
 ,
24
,
521
–535.
20
Guimaraes, C.P., Domingues, P., Aubourg, P., Fouquet, F., Pujol, A., Jimenez-Sanchez, G., Sa-Miranda, C. and Azevedo, J.E. (
2004
) Mouse liver PMP70 and ALDP: homomeric interactions prevail in vivo.
Biochim. Biophys. Acta
 ,
1689
,
235
–243.
21
Flavigny, E., Sanhaj, A., Aubourg, P. and Cartier, N. (
1999
) Retroviral-mediated adrenoleukodystrophy-related gene transfer corrects very long chain fatty acid metabolism in adrenoleukodystrophy fibroblasts: implications for therapy.
FEBS Lett.
 ,
448
,
261
–264.
22
Kemp, S., Wei, H.M., Lu, J.F., Braiterman, L.T., McGuinness, M.C., Moser, A.B., Watkins, P.A. and Smith, K.D. (
1998
) Gene redundancy and pharmacological gene therapy: implications for X-linked adrenoleukodystrophy.
Nat. Med.
 ,
4
,
1261
–1268.
23
Pujol, A., Ferrer, I., Camps, C., Metzger, E., Hindelang, C., Callizot, N., Ruiz, M., Pampols, T., Giros, M. and Mandel, J.L. (
2004
) Functional overlap between ABCD1 (ALD) and ABCD2 (ALDR) transporters: a therapeutic target for X-adrenoleukodystrophy.
Hum. Mol. Genet.
 ,
13
,
2997
–3006.
24
Watkins, P.A. (
1997
) Fatty acid activation.
Prog. Lipid Res.
 ,
36
,
55
–83.
25
Hirsch, D., Stahl, A. and Lodish, H.F. (
1998
) A family of fatty acid transporters conserved from mycobacterium to man.
Proc Natl Acad. Sci. USA
 ,
95
,
8625
–8629.
26
Wanders, R.J., van Roermund, C.W., van Wijland, M.J., Schutgens, R.B., van den Bosch, H., Schram, A.W. and Tager, J.M. (
1988
) Direct demonstration that the deficient oxidation of very long chain fatty acids in X-linked adrenoleukodystrophy is due to an impaired ability of peroxisomes to activate very long chain fatty acids.
Biochem. Biophys. Res. Commun.
 ,
153
,
618
–624.
27
Lazo, O., Contreras, M., Hashmi, M., Stanley, W., Irazu, C. and Singh, I. (
1988
) Peroxisomal lignoceroyl-CoA ligase deficiency in childhood adrenoleukodystrophy and adrenomyeloneuropathy.
Proc. Natl Acad. Sci. USA
 ,
85
,
7647
–7651.
28
Steinberg, S.J., Kemp, S., Braiterman, L.T. and Watkins, P.A. (
1999
) Role of very-long-chain acyl-coenzyme A synthetase in X-linked adrenoleukodystrophy.
Ann. Neurol.
 ,
46
,
409
–412.
29
Min, K.T. and Benzer, S. (
1999
) Preventing neurodegeneration in the Drosophila mutant bubblegum.
Science
 ,
284
,
1985
–1988.
30
Antonarakis, S.E. (
1998
) Recommendations for a nomenclature system for human gene mutations. Nomenclature Working Group.
Hum. Mutat.
 ,
11
,
1
–3.
31
den Dunnen, J.T. and Antonarakis, S.E. (
2000
) Mutation nomenclature extensions and suggestions to describe complex mutations: a discussion.
Hum. Mutat.
 ,
15
,
7
–12.
32
Jia, Z., Pei, Z., Li, Y., Wei, L., Smith, K.D. and Watkins, P.A. (
2004
) X-linked adrenoleukodystrophy: role of very long-chain acyl-CoA synthetases.
Mol. Genet. Metab.
 ,
83
,
117
–127.
33
Brown, F.R., III, Van Duyn, M.A., Moser, A.B., Schulman, J.D., Rizzo, W.B., Snyder, R.D., Murphy, J.V., Kamoshita, S., Migeon, C.J. and Moser, H.W. (
1982
) Adrenoleukodystrophy: effects of dietary restriction of very long chain fatty acids and of administration of carnitine and clofibrate on clinical status and plasma fatty acids.
Johns Hopkins Med. J.
 ,
151
,
164
–172.
34
Rizzo, W.B., Avigan, J., Chemke, J. and Schulman, J.D. (
1984
) Adrenoleukodystrophy: very long-chain fatty acid metabolism in fibroblasts.
Neurology
 ,
34
,
163
–169.
35
Kemp, S., Valianpour, F., Denis, S., Ofman, R., Sanders, R.J., Mooyer, P., Barth, P.G. and Wanders, R.J. (
2005
) Elongation of very long-chain fatty acids is enhanced in X-linked adrenoleukodystrophy.
Mol. Genet. Metab.
 ,
84
,
144
–151.
36
Leonard, A.E., Pereira, S.L., Sprecher, H. and Huang, Y.S. (
2004
) Elongation of long-chain fatty acids.
Prog. Lipid Res.
 ,
43
,
36
–54.
37
Bourre, J.M., Daudu, O. and Baumann, N. (
1976
) Nervonic acid biosynthesis by erucyl-CoA elongation in normal and quaking mouse brain microsomes. Elongation of other unsaturated fatty acyl-CoAs (mono and poly-unsaturated).
Biochim. Biophys. Acta
 ,
424
,
1
–7.
38
O'Neill, G.N., Aoki, M. and Brown, R.H., Jr (
2001
) ABCD1 translation-initiator mutation demonstrates genotype–phenotype correlation for AMN.
Neurology
 ,
57
,
1956
–1962.
39
Kemp, S., Pujol, A., Waterham, H.R., van Geel, B.M., Boehm, C.D., Raymond, G.V., Cutting, G.R., Wanders, R.J. and Moser, H.W. (
2001
) ABCD1 mutations and the X-linked adrenoleukodystrophy mutation database: role in diagnosis and clinical correlations.
Hum. Mutat.
 ,
18
,
499
–515.
40
Holbrook, J.A., Neu-Yilik, G., Hentze, M.W. and Kulozik, A.E. (
2004
) Nonsense-mediated decay approaches the clinic.
Nat. Genet.
 ,
36
,
801
–808.
41
McGuinness, M.C., Lu, J.F., Zhang, H.P., Dong, G.X., Heinzer, A.K., Watkins, P.A., Powers, J. and Smith, K.D. (
2003
) Role of ALDP (ABCD1) and mitochondria in X-linked adrenoleukodystrophy.
Mol. Cell. Biol.
 ,
23
,
744
–753.
42
Holzinger, A., Mayerhofer, P.U., Maier, E.M., Roscher, A.A. and Berger, J. (
2003
) Evidence against the adrenoleukodystrophy-related gene acting as a modifier of X-adrenoleukodystrophy.
Adv. Exp. Med. Biol.
 ,
544
,
95
–96.
43
Braiterman, L.T., Zheng, S., Watkins, P.A., Geraghty, M.T., Johnson, G., McGuinness, M.C., Moser, A.B. and Smith, K.D. (
1998
) Suppression of peroxisomal membrane protein defects by peroxisomal ATP binding cassette (ABC) proteins.
Hum. Mol. Genet.
 ,
7
,
239
–247.
44
Netik, A., Forss-Petter, S., Holzinger, A., Molzer, B., Unterrainer, G. and Berger, J. (
1999
) Adrenoleukodystrophy-related protein can compensate functionally for adrenoleukodystrophy protein deficiency (X-ALD): implications for therapy.
Hum. Mol. Genet.
 ,
8
,
907
–913.
45
Silva-Zolezzi, I., Hebron, K., Mihalik, S., Thomas, G., Valle, D. and Jimenez-Sanchez, G. (
2003
) Abcd3−/− mice have a non-shivering thermogenesis defect related to a disturbance in fasting fuel homeostasis.
Proceedings of the 53rd Annual Meeting of The American Society of Human Genetics
 . The American Society of Human Genetics, Los Angeles, California, Suppl. to Vol.
73
, p.
199
.
46
Lock, C., Hermans, G., Pedotti, R., Brendolan, A., Schadt, E., Garren, H., Langer-Gould, A., Strober, S., Cannella, B., Allard, J. et al. (
2002
) Gene-microarray analysis of multiple sclerosis lesions yields new targets validated in autoimmune encephalomyelitis.
Nat. Med.
 ,
8
,
500
–508.
47
Heinzer, A.K., Kemp, S., Lu, J.F., Watkins, P.A. and Smith, K.D. (
2002
) Mouse very long-chain acyl-CoA synthetase in X-linked adrenoleukodystrophy.
J. Biol. Chem.
 ,
277
,
28765
–28773.
48
Heinzer, A.K., Watkins, P.A., Lu, J.F., Kemp, S., Moser, A.B., Li, Y.Y., Mihalik, S., Powers, J.M. and Smith, K.D. (
2003
) A very long-chain acyl-CoA synthetase-deficient mouse and its relevance to X-linked adrenoleukodystrophy.
Hum. Mol. Genet.
 ,
12
,
1145
–1154.
49
Fraisl, P., Forss-Petter, S., Zigman, M. and Berger, J. (
2004
) Murine bubblegum orthologue is a microsomal very long-chain acyl-CoA synthetase.
Biochem. J.
 ,
377
,
85
–93.
50
Ogino, T. and Suzuki, K. (
1981
) Specificities of human and rat brain enzymes of cholesterol ester metabolism toward very long chain fatty acids: implication for biochemical pathogenesis of adrenoleukodystrophy.
J. Neurochem.
 ,
36
,
776
–779.
51
Simons, K. and Ehehalt, R. (
2002
) Cholesterol, lipid rafts, and disease.
J. Clin. Invest.
 ,
110
,
597
–603.
52
Powers, J.M., Liu, Y., Moser, A.B. and Moser, H.W. (
1992
) The inflammatory myelinopathy of adreno-leukodystrophy: cells, effector molecules, and pathogenetic implications.
J. Neuropathol. Exp. Neurol.
 ,
51
,
630
–643.
53
Tagawa, Y., Laroy, W., Nimrichter, L., Fromholt, S.E., Moser, A.B., Moser, H.W. and Schnaar, R.L. (
2002
) Anti-ganglioside antibodies bind with enhanced affinity to gangliosides containing very long chain fatty acids.
Neurochem. Res.
 ,
27
,
847
–855.
54
Schmidt, S., Marrosu, G.M., Kolsch, H., Haase, C.G., Ferenczik, S., Sokolowski, P., Kohler, W., Schmidt, M., Papassotiropoulos, A., Heun, R. et al. (
2003
) Genetic variations and humoral immune responses to myelin oligodendroglia glycoprotein in adult phenotypes of X-linked adrenoleukodystrophy.
J. Neuroimmunol.
 ,
135
,
148
–153.
55
Boles, D.J., Craft, D.A., Padgett, D.A., Loria, R.M. and Rizzo, W.B. (
1991
) Clinical variation in X-linked adrenoleukodystrophy: fatty acid and lipid metabolism in cultured fibroblasts.
Biochem. Med. Metab. Biol.
 ,
45
,
74
–91.
56
Moser, A.B., Kreiter, N., Bezman, L., Lu, S., Raymond, G.V., Naidu, S. and Moser, H.W. (
1999
) Plasma very long chain fatty acids in 3,000 peroxisome disease patients and 29,000 controls.
Ann. Neurol.
 ,
45
,
100
–110.
57
Paintlia, A.S., Gilg, A.G., Khan, M., Singh, A.K., Barbosa, E. and Singh, I. (
2003
) Correlation of very long chain fatty acid accumulation and inflammatory disease progression in childhood X-ALD: implications for potential therapies.
Neurobiol. Dis.
 ,
14
,
425
–439.
58
Whitcomb, R.W., Linehan, W.M. and Knazek, R.A. (
1988
) Effects of long-chain, saturated fatty acids on membrane microviscosity and adrenocorticotropin responsiveness of human adrenocortical cells in vitro.
J. Clin. Invest.
 ,
81
,
185
–188.
59
Ho, J.K., Moser, H., Kishimoto, Y. and Hamilton, J.A. (
1995
) Interactions of a very long chain fatty acid with model membranes and serum albumin. Implications for the pathogenesis of adrenoleukodystrophy.
J. Clin. Invest.
 ,
96
,
1455
–1463.
60
Imai, Y., Ibata, I., Ito, D., Ohsawa, K. and Kohsaka, S. (
1996
) A novel gene iba1 in the major histocompatibility complex class III region encoding an EF hand protein expressed in a monocytic lineage.
Biochem. Biophys. Res. Commun.
 ,
224
,
855
–862.
61
Ligtenberg, M.J., Kemp, S., Sarde, C.O., van Geel, B.M., Kleijer, W.J., Barth, P.G., Mandel, J.L., van Oost, B.A. and Bolhuis, P.A. (
1995
) Spectrum of mutations in the gene encoding the adrenoleukodystrophy protein.
Am. J. Hum. Genet.
 ,
56
,
44
–50.
62
Bieche, I., Parfait, B., Le Doussal, V., Olivi, M., Rio, M.C., Lidereau, R. and Vidaud, M. (
2001
) Identification of CGA as a novel estrogen receptor-responsive gene in breast cancer: an outstanding candidate marker to predict the response to endocrine therapy.
Cancer Res.
 ,
61
,
1652
–1658.
63
Folch, J., Lees, M. and Sloane Stanley, G.H. (
1957
) A simple method for the isolation and purification of total lipides from animal tissues.
J. Biol. Chem.
 ,
226
,
497
–509.
64
Lagerstedt, S.A., Hinrichs, D.R., Batt, S.M., Magera, M.J., Rinaldo, P. and McConnell, J.P. (
2001
) Quantitative determination of plasma c8–c26 total fatty acids for the biochemical diagnosis of nutritional and metabolic disorders.
Mol. Genet. Metab.
 ,
73
,
38
–45.