C26:0-Carnitine Is a New Biomarker for X-Linked Adrenoleukodystrophy in Mice and Man

X-linked adrenoleukodystrophy (ALD), a progressive neurodegenerative disease, is caused by mutations in ABCD1 and characterized by very-long-chain fatty acids (VLCFA) accumulation. Virtually all males develop progressive myelopathy (AMN). A subset of patients, however, develops a fatal cerebral demyelinating disease (cerebral ALD). Hematopoietic stem cell transplantation is curative for cerebral ALD provided the procedure is performed in an early stage of the disease. Unfortunately, this narrow therapeutic window is often missed. Therefore, an increasing number of newborn screening programs are including ALD. To identify new biomarkers for ALD, we developed an Abcd1 knockout mouse with enhanced VLCFA synthesis either ubiquitous or restricted to oligodendrocytes. Biochemical analysis revealed VLCFA accumulation in different lipid classes and acylcarnitines. Both C26:0-lysoPC and C26:0-carnitine were highly elevated in brain, spinal cord, but also in bloodspots. We extended the analysis to patients and confirmed that C26:0-carnitine is also elevated in bloodspots from ALD patients. We anticipate that validation of C26:0-carnitine for the diagnosis of ALD in newborn bloodspots may lead to a faster inclusion of ALD in newborn screening programs in countries that already screen for other inborn errors of metabolism.


Introduction
X-linked adrenoleukodystrophy (ALD) is a progressive neurodegenerative disorder caused by mutations in the ABCD1 gene [1]. The disease is characterized by impaired degradation of very long-chain fatty acids (VLCFA; >C22) [2,3], resulting in VLCFA accumulation in plasma and tissues [4]. ALD affects approximately 1 in 17.000 males [5] and has been diagnosed in all geographic regions and ethnic groups. There is no evidence that the prevalence varies with ethnic background [6]. Patients with ALD are asymptomatic at birth [7]. A study in neurologically asymptomatic young boys with ALD revealed that 80% had unrecognized adrenal insufficiency [8]. Virtually all male patients with ALD eventually develop progressive myelopathy (adrenomyeloneuropathy, AMN). The onset is typically between 20-30 years of age [7,9]. This myelopathy progresses gradually, eventually causing severe disability. Recently, we showed that women with ALD are not merely carriers, but that >80% also develop signs of myelopathy [10]. A subset of male patients, however, develops a fatal cerebral demyelinating disease (cerebral ALD). The age of onset of cerebral ALD cannot be predicted. A newborn male patient has a 35-40% risk to develop cerebral ALD between the ages of 3 and 18 years, but cerebral ALD can also occur in adulthood [7,9,11]. Allogeneic hematopoietic stem cell transplantation (HSCT) is curative for cerebral ALD provided the procedure is performed in an early stage of the disease before extensive MRI white matter abnormalities are present [12]. Unfortunately, this therapeutic window is often missed.
If newborn screening for ALD is implemented cerebral ALD can be identified and treated in an early stage, thus substantially increasing the likelihood of a better outcome. This is one of the reasons why ALD is being added to an increasing number of newborn screening programs. For example, New York State initiated ALD newborn screening in 2014 and in Europe the Netherlands will start ALD newborn screening in the near future. In August 2015, the United States advisory committee on heritable disorders in newborns and children recommended ALD to be added to the recommended uniform screening panel (RUSP). Currently ALD newborn screening involves quantification of 1-hexacosanoyl-2-lyso-sn-3-glycero-phosphorylcholine (C26:0-lysoPC) by liquid chromatography-tandem mass spectrometry (LC-MS/MS) in dried blood spots [13]. While the method shows high sensitivity and specificity, the need for liquid chromatography to separate the different lysoPC species hampers high throughput screening. The procedure has been slightly modified [14,15] to combine the existing method for acylcarnitine analysis and C26:0-lysoPC, which enables simultaneous extraction and screening for peroxisomal disorders, mitochondrial fatty acid oxidation disorders and organic acidurias. Recently, a method was developed that enables the quantification of amino acids, acylcarnitines, succinylacetone, and C26:0-lysoPC in a single dried bloodspot punch [16].
The neurometabolic consequences of ALD protein (ALDP) deficiency and the subsequent increase in VLCFA levels have not yet been resolved at the cellular level. The absence of a ALD mouse model that mimics the disease in humans has been a major limitation in unraveling the mechanism underlying the pathogenesis of ALD. The Abcd1 y/knockout mouse develops a mild phenotype that resembles the myelopathy of ALD, but only at around 20 months of age [17]. There is indirect evidence that an increase in VLCFA levels in Abcd1 knockout mice may result in an earlier onset of disease. For example, Abcd1 knockout mice that were put on a high-fat diet showed structural abnormalities in the adrenal gland on electron microscopy [18]. In readily accessible material from patients, such as plasma, fibroblasts, or blood cells, total C26:0 levels do not correlate with phenotype [19]. However, biochemical analysis of normal-appearing grey and white matter that was dissected from frontal, parietal or occipital lobes from 17 ALD patients and 19 age-matched controls revealed a correlation between VLCFA levels and clinical phenotype [20]. Support for a toxic effect of VLCFA came from the demonstration that C24:0-lysoPC injection into wild type mouse brain resulted in widespread microglial activation and apoptosis [21]. Importantly, C16:0-lysoPC injections did not cause these effects. Studies using rat neural cells revealed that exposure of oligodendrocytes and astrocytes to VLCFA, but not C16:0, caused cell death [22]. Challenging neural cells with VLCFA induced depolarization of mitochondria in situ and caused deregulation of intracellular calcium homeostasis [22]. Long before the first neuropathological lesions are detectable in the spinal cord of Abcd1 knockout mice there is oxidative damage to proteins [23].
With the aim to generate an ALD mouse with increased VLCFA levels in the central nervous system, we developed an Abcd1 knockout mouse with a Cre-inducible ELOVL1 transgene: the key enzyme in VLCFA synthesis [24,25]. Based on the above studies we anticipated that increased VLCFA levels in the central nervous system could impact the clinical phenotype of the Abcd1 knockout mouse and may result in the identification of new biomarkers.

Patient and Control Samples
Samples from ALD patients were available from a previous study that was approved by the Institutional Review Board (Medisch Ethische Toetsings Commissie) of the Academic Medical Center (BEZA trial [26]). Written informed consent was received from each patient. Samples for control values were obtained from the Laboratory Genetic Metabolic Diseases from anonymized samples from individuals who were examined in our institute for therapeutic control or exclusion of inherited metabolic disease. All samples were collected according to the institutional guidelines for sampling.

Ethics Statement and Animal Care
Mice were bred in the animal housing facility of the AMC, had ad libitum access to water and standard rodent food, and were kept on a 12 hour light and dark cycle. For the care and use of animals utilized in this research, we monitored the animals twice per week. None of animals showed severe illness or died during the experiments. A protocol for early euthanasia/ humane endpoints is performed if one of the following criteria is met: the loss of body weight more than 15% or a wound that cannot be improved after medication. For tissue collection, the mice were sacrificed by introduction of 100% carbon dioxide into a bedding-free cage with the lid closed at a rate sufficient to induce rapid anesthesia, with death occurring within 2.5 minutes. All animal experiments were approved by the institutional review board for animal experiments at the Academic Medical Center, University of Amsterdam (Amsterdam, The Netherlands).

Generation of ELOVL1 Knock-In Mice
The ELOVL1 transgenic construct and mouse model were commercially generated by GenOway (Lyon, France). GenOway's validated Rosa26 "Quick Knock-in™" approach was used to introduce a single copy of the human ELOVL1 transgenic cassette into the Rosa26 locus on chromosome 6 through homologous recombination in embryonic stem cells. Briefly, the human ELOVL1 cDNA fragment was subcloned into a GenOway's G136 plasmid containing the human growth factor polyA cassette. Then the ELOVL1-polyA cassette was subcloned into GenOway's G212 vector downstream of the pCAG promoter and the transcriptional STOP cassette flanked by two loxP sites. The linearized construct (Fig 1) was transfected into mouse 129SV ES cells according to GenOway's standard electroporation procedures (i.e. 5×10 6 ES cells in presence of 40 μg of linearized plasmid, 260 volt, 500 μF). Positive selection was started 48 hours after electroporation by addition of 200 μg/ml of G418. Approximately 223 G418-resistant clones were screened for the correct homologous recombination event at the Rosa26 locus by PCR and southern blotting. Two correctly recombined ES clones were used for injection into C57BL/6J blastocysts. Injected blastocysts were re-implanted into OF1 pseudo-pregnant females and allowed to develop to term. A total of 9 male chimeric mice were produced from the 2 injection sessions. Four highly chimeric males (displaying 80% chimerism) were each mated with 2 C57BL/6J Flp-deleter females to allow germline excision of the neomycin selection cassette. Agouti F1 progeny were screened for assessing the Flp-mediated excision event of the neomycin cassette on the ELOVL1 allele by PCR and southern blotting, using genomic DNA isolated from tail biopsies. As there was mosaicism for the Flp ELOVL1 allele in F1 mice, these mice were then crossed with wild-type C57BL/6J mice to generate a pure line of Flp excised heterozygous ELOVL1 knock-in mice. PCR and southern blot screening confirmed the successful generation of 3 males and 2 females heterozygous for the Rosa26 neomycin-excised ELOVL1 knock-in allele. These mice are referred to as Rosa26 CAG-STOPflox/flox-ELOVL1TG .

Generation of Abcd1-CMV-ELOVL1 Mice
The C57BL6/J CMV-ELOVL1 strain was crossed with our in-house in-bred Swiss Abcd1 knockout strain (defined as 10 generations back-crossed to Swiss, followed by >20 generations of brother-sister crossings) to obtain Abcd1 knockout mice ubiquitously overexpressing ELOVL1 on a mixed Swiss, C57BL/6J background (referred to as Abcd1;CMV-ELOVL1).

Purification of Mouse Liver Microsomes
Microsomes were isolated from livers from wild type and ubiquitous ELOVL1 over-expressing Abcd1 knockout mice (Abcd1 y/-;CMV-ELOVL1 +/-) by differential centrifugation. Livers were washed with ice-cold homogenization buffer containing 250 mM sucrose, 2 mM EDTA, 2 mM DTT and 5 mM MOPS (pH 7.4), minced and homogenized with 20 strokes of a dounce homogenizer while kept on ice. A post-nuclear supernatant was produced by centrifugation at 600 g for 10 min. The supernatant was centrifuged at 22,500 g for 10 min and the pellet was discarded. To obtain a microsomal fraction, the supernatant was centrifuged for 1 h at 100,000 g. To remove any residual fatty acids the pellet was resuspended in homogenization buffer containing 10 mg/mL methyl-β-cyclodextrin and sonicated four times for 5 seconds at 7W. The microsomal membranes were collected by centrifugation at 100,000 g for 1 h. Finally, the microsomes were resuspended in homogenization buffer and stored at -80°C in 100 μL aliquots until further use. All steps were carried out at 4°C. Protein concentration was determined using Pierce 1 BCA protein assay (Thermo Scientific, Waltham, MA, USA) with human serum albumin as standard.

Non-Targeted Metabolic Profiling
Bloodspots were generated by collecting 50 μL blood from the vena saphena from C57BL6/6 and Swiss mice. All subsequent steps were carried out at Metabolomic Discoveries GmbH (Potsdam, Germany; www.metabolomicdiscoveries.com) essentially as described [29], but with improvements. Discs of 6 mm diameter were extracted with 1 ml 80% (v/v) methanol and sonicated for 1 h at 4°C. 100 μl was used for HPLC-TOF-MS analyses. LC separation was performed using a SeQuant ZIC HILIC column (Merck) operated by an Agilent 1290 UPLC system (Agilent, Santa Clara, CA, USA). The LC mobile phase was A) 10 mM ammonium acetate in 95% (v/v) acetonitrile and 5% (v/v) water B) 10 mM ammonium acetate in 95% (v/v) water with a gradient from 0% B to 90% over 5 min, to 95% at 6.5 min and 100% at 8 min, subsequently equilibrate. The flow rate was 400 μl/min, injection volume 1 μl. Mass spectrometry was performed using a high-resolution 6540 QTOF/MS Detector (Agilent, Santa Clara, CA, USA) operated in positive and negative ionization modes with a mass accuracy of <2ppm. For UPLC-QTOF/MS, metabolites were identified or putatively annotated in comparison to Metabolomic Discoveries' database entries of authentic standards and IDEOM database entries through peak mass within 5 ppm mass accuracy and retention time.

Total Fatty Acid Analysis
Fatty acids were analyzed by electrospray ionization mass spectrometry (ESI-MS) as described previously [30].

LysoPC and Acylcarnitine: Sample Preparation
A combined punch of a dried bloodspot (¼ inch in diameter) and 10 μL of internal standard were extracted with 0.5 mL of methanol by ultrasonication for 5 min. The extract was dried (N 2 , 40°C) and reconstituted in 50 μL of methanol. 10 μL was used for analysis.
For mouse tissues, 0.2 mg and 10 μL of internal standard were extracted with 0.5 mL of acetonitrile by ultrasonication for 5 min. After full speed centrifugation (5 min, 4°C) the supernatant was dried (N 2 , 40°C) and reconstituted in 50 μL of methanol. 10 μL was used for analysis.

UPLC-MS/MS
The ACQUITY UPLC system (Waters, Milford, MA, USA) consisted of a binary solvent manager, a vacuum degasser, a column heater and sample manager. The column temperature was maintained at 50°C. The samples were injected onto a Kinetex C8 column, 50 × 2.1 mm, 2.6 μm particle diameter (Phenomenex, Torrance, CA, USA). The acylcarnitines were separated by a linear gradient between solution A (0.1% formic acid in H 2 O) and solution B (0.1% formic acid in methanol). The gradient was as follows: at T = 0 min: 36% A, 64% B, flow 0.4 mL/min towards T = 6 min: 0% A, 100% B, flow 0.4 mL/min; T = 6-11 min: 0% A, 100% B, flow 0.4 mL/min, and T = 11-11.1 back to 36% A, 64% B, flow 0.4 mL/min. A Quattro Premier XE (Waters, Milford, MA, USA) was used in the positive electrospray ionization mode. Nitrogen was used as desolvation gas (900 L/h) and cone gas (50 L/h). Desolvation temperature was 350°C, capillary voltage was 3.5 kV and the source temperature was 130°C. Argon was used as collision gas (2.5 x 10e-3 mbar). For the very long-chain acylcarnitines and lysophosphatidylcholines (lysoPC) multiple reaction monitoring (MRM) traces were acquired with optimized cone voltage and collision energy for each transition with a dwell time of 0.01 s ( Table 1).

Data Analysis
Data are expressed as means ± SD of measurements. Statistical comparisons were performed using the unpaired student's test, and significance was defined as: Ã = P < 0.05, ÃÃ = P < 0.01 and ÃÃÃ = P < 0.001.

Results and Discussion
Generation of ELOVL1 Transgenic Mice ELOVL1 is the essential enzyme in the elongation of C22:0 to C26:0 [24,25,32]. To investigate the in vivo effect of ELOVL1 overexpression on VLCFA homeostasis, we generated a Creinducible ELOVL1 transgenic mouse (Rosa26 CAG-STOPflox/flox-ELOVL1TG ). The ELOVL1 transgene consists of a loxP flanked STOP element positioned between the ELOVL1 coding DNA sequence and its upstream chicken β-actin/CMV immediate early enhancer fusion promoter ( Fig 1A).
As a proof of concept that enhanced ELOVL1 expression results in increased ELOVL1 protein levels and increased VLCFA levels, homozygous Rosa26 CAG-STOPflox/flox-ELOVL1TG mice were crossed with a Cre-driver strain that expresses Cre recombinase under the control of a human cytomegalovirus (CMV) promoter (CMV-Cre +/-). F1 mice were either heterozygous for the activated ELOVL1 transgene (CMV-ELOVL1 +/-) in all tissues or negative for the activated ELOVL1 transgene (controls). The efficacy of transgene activation was assessed in mouse liver homogenates generated from CMV-ELOVL1 +/and control mice. An organelle fraction was prepared from which microsomes were isolated.
Protein blot analysis revealed a strong increase in ELOVL1 protein in livers derived from CMV-ELOVL1 +/mice, but not in controls (Fig 1B). ELOVL1 was present exclusively in the organelle fraction, with a strong expression in the microsomes, indicating correct targeting to the endoplasmic reticulum [33]. While only a very faint ELOVL1 signal was detectable in the microsomal fraction of the control mice, a strong ELOVL1 signal was present in CMV-ELOVL1 +/mice (ratio 1:20). In control mice ELOVL1 was only detectable in purified microsomes. Taken together, these data show that activation of the ELOVL1 transgene results in a marked increase in ELOVL1 expression and that the protein is localized correctly to the ER membrane. Microsomes from control and CMV-ELOVL1 +/mice were used to determine the fatty acid elongation capacity for both long-chain fatty acids (C16:0-CoA) and VLCFA (C22:0-CoA) (Fig 1C). No difference between control and CMV-ELOVL1 +/mice was seen when microsomes were incubated with C16:0-CoA. This was expected as C16:0-CoA is a substrate for the long-chain fatty acyl elongase, ELOVL6 [24,34]. In contrast, microsomes from CMV-ELOVL1 +/mice had a 25-fold higher elongation activity for C22:0-CoA, which is in agreement with the substrate specificity of ELOVL1 [24,32].
To assess the biochemical effect of ELOVL1 over-expression in Abcd1 knockout mice, the CMV-ELOVL1 line was crossed with Abcd1 knockouts. ELOVL1 protein and VLCFA levels were measured in wild type, Abcd1 y/knockout and Abcd1 y/-;CMV-ELOVL1 +/mice. Protein blot analysis of brain, adrenal gland and testis demonstrated a strong expression of transgenic ELOVL1 (Fig 1D).

Possibly Early Lethal Phenotype in Abcd1;CMV-ELOVL1 Mice
Breeding results with the Abcd1;CMV-ELOVL1 line provided potential evidence for the early phenotype in Abcd1 y/-;CMV-ELOVL1 +/male mice. Homozygous Abcd1 knockout females (Abcd1 -/-) were crossed with males heterozygous for the activated ELOVL1 transgene (CMV-ELOVL1 +/-). In the F1, an equal genotype distribution was expected (25% for each genotype) with females being either Abcd1 +/or Abcd1 +/-;CMV-ELOVL1 +/and males being either Abcd1 y/or Abcd1 y/-; CMV-ELOVL1 +/-. Genotype analysis of 31 pups from 4 independent breeding pairs, however, revealed a significant underrepresentation in the number of Abcd1 y/-;CMV-ELOVL1 +/males. The expected and actual genotype-distribution is presented in Fig 2A. Although 25% of the newborn mice were expected to be Abcd1 y/-;CMV-ELOVL1 +/males, only 6% males with this genotype were born (Chi 2 P = < 0.001). Additional evidence for a possibly early phenotype came from a second breeding strategy. Homozygous Abcd1 -/knockout females were crossed with male mice homozygous for the activated ELOVL1 transgene (CMV-ELOVL1 +/+ ). The F1 expected genotypes were Abcd1 +/-;CMV-ELOVL1 +/-(females) and Abcd1 y/-;CMV-ELOVL1 +/-(males) with an equal female to male ratio. Genotyping analysis of 17 pups from 4 independent mating pairs, however, revealed an actual female to male ratio of 3:1 (Chi 2 P = < 0.001) (Fig 2B). Taken together, these data indicate the onset of a possibly early lethal phenotype. There are at least three possible explanations for these observations, namely: 1) compared to normal sperm cells, ELOVL1 over-expressing sperm cells may be affected in their motility, which could affect successful fertilization, 2) Abcd1 y/-;CMV-ELOVL1 +/males may already die in the prenatal stage, or 3) Abcd1 y/-;CMV-ELOVL1 +/pups may die immediately after birth. Alternatively, we cannot rule out the possibility that the mixed C57BL6/Swiss background of the Abcd1;CMV-ELOVL1 line affects the early phenotype. This is nicely illustrated in Fig 2C  that shows the results of an untargeted comprehensive metabolite profiling of bloodspots derived from C57BL6/J and Swiss wild type and Abcd1 knockouts. We observed strong differences in the metabolome between the two mouse strains. The untargeted metabolomics data set was used for hierarchical clustering using MetaboAnalyst (http://metaboanalyst.ca/). In one dimension, this analysis perfectly clustered the four different groups of mice according to the background strain, i.e. Swiss or C57BL6/J (Fig 2C). This highest difference between Swiss and C57BL6/J mice was observed for serotonin. Interestingly, clustering revealed specific differences between Swiss and C57BL6/J mice in different lipid species and acylcarnitines. Both in Swiss and C57BL6/J Abcd1 knockouts C26:0-lysoPC and to a lesser extend C24:0-lysoPC were highly elevated when compared to wild types.
However, because tissue VLCFA levels were uniformly elevated in a manner that does not reflect the biochemical signature of ALD (Fig 1E), we determined that the Abcd1;CMV-E-LOVL1 model had limited utility, and decided not to pursue studies with this line any further.
Interestingly, the analysis of C26:0 levels in whole brain homogenates from Abcd1 knockouts with ubiquitous ELOVL1 over-expression (Abcd1 y/-;CMV-ELOVL1 +/-) and Abcd1 knockout with ELOVL1 overexpression restricted to oligodendrocytes (Abcd1 y/-;Cnp-ELOVL1 +/-) revealed similar C26:0 levels (20-fold and 15-fold, respectively). This indicates that oligodendrocytes are largely responsible for the total VLCFA synthesis in brain. These data are in agreement with earlier work that reported similar C26:0 levels in whole brain lysates and in purified myelin derived from oligodendrocyte-specific peroxisome knockout mice [36]. Taken together, these data strongly indicate that oligodendrocytes are largely responsible for the total VLCFA degradation and VLCFA synthesis activity of the brain.

C26:0-lysoPC Is Highly Elevated in Central Nervous Tissue
The currently available biomarkers for ALD are excellent for establishing the diagnosis (for instance, total C26:0 in plasma or blood cells), but do not correlate with, phenotype, disease severity or pattern of progression. In clinical trials C26:0 is therefore a biomarker of doubtful significance, which implies that new biomarkers are urgently needed. In recent years, it has been demonstrated that 1-hexacosanoyl-sn-glycero-3-phosphocholine (C26:0-lysoPC) is already elevated at birth in blood of newborns with ALD by 5-fold [13][14][15]. We investigated whether C26:0-lysoPC levels are directly associated with elevated total C26:0 levels observed in our mouse models and measured C26:0-lysoPC in mouse brain and spinal cord. Analysis of brain and spinal cord samples from wild type, Abcd1 y/knockout and Abcd1 y/-;Cnp-ELOVL1 +/mice revealed that C26:0-lysoPC is highly elevated in Abcd1 y/knockouts and Abcd1 y/-;Cnp-ELOVL1 +/mice (Fig 4). Compared to wild type mice, in Abcd1 knockouts C26:0-lysoPC levels were increased 4-fold in brain and 6-fold in spinal cord, and in Abcd1 y/-;Cnp-ELOVL1 +/-C26:0-lysoPC levels were increased 13-fold in brain and 24-fold in spinal cord.

C26:0-Carnitine Is Highly Elevated in Central Nervous Tissue
In 2003, increased C24:0-carnitine and C26:0-carnitine levels were reported in plasma and bloodspots from patients with either a peroxisomal biogenesis disorder or D-bifunctional protein deficiency [37]. No increase was reported in plasma from ALD patients. Importantly, bloodspots from ALD patients were not included in the study. Significantly elevated VLCFA acylcarnitines were also reported in urine samples from patients with a peroxisomal biogenesis disorder [38]. This latter study also did not include samples from ALD patients. We investigated whether VLCFA-acylcarnitines are also elevated in our ALD mouse models. Analysis of metabolome between both background strains. The color in the heat map reflects the global metabolite abundance level according to the z-score. The 50 most significant different metabolites are shown (P < 0.05) as determined by a Welch's ttest.
doi:10.1371/journal.pone.0154597.g002 C26:0-Carnitine Is a New Biomarker for ALD brain and spinal cord samples from wild type, Abcd1 y/knockout and Abcd1 y/-;Cnp-ELOVL1 +/mice revealed that C26:0-carnitine levels are highly elevated in Abcd1 y/knockouts and Abcd1 y/-;Cnp-ELOVL1 +/mice (Fig 5). Compared to wild type mice, in Abcd1 y/knockouts C26:0-carnitine levels were increased 10-fold in brain and 9-fold in spinal cord, and in Abcd1 y/-;Cnp-ELOVL1 +/-C26:0-carnitine levels were increased 40-fold in brain and 33-fold in spinal cord. In the USA, New York State has initiated ALD newborn screening in 2014 and in Europe the Netherlands will start ALD newborn screening in the near future. The current method for diagnosing ALD is sensitive and specific, but it involves a dedicated and separate analysis for C26:0-lysoPC. In the search for new biomarkers for ALD that are more sensitive than total C26:0 and easier to measure than C26:0-lysoPC we compared C26:0 levels with C26:0-lysoPC and C26:0-carnitine in mouse and then expanded this analysis to human plasma and bloodspots.
Previously, it was demonstrated that brain VLCFA levels correlate with the clinical phenotype [20]. Biochemical analysis of normal-appearing grey and white matter that was dissected from frontal, parietal or occipital lobes from 17 ALD patients and 19 age-matched controls revealed that in comparison with age-matched controls, C26:0 levels were increased 3-fold in cerebral ALD patients, and 1.9-fold in AMN patients. These results suggest that the accumulation in brain VLCFA levels precedes histopathological alterations and are an important factor in the development of cerebral disease. Our data demonstrate that, at least in mice, total plasma C26:0 levels do not correlate with C26:0 levels present in either spinal cord or brain. Interestingly, both C26:0-lysoPC and C26:0-carnitine levels in mouse bloodspots correlate with the C26:0, C26:0-lysoPC and C26:0-carnitine levels found in brain and spinal cord. We conclude that C26:0-lysoPC and C26:0-carnitine are new biochemical biomarkers for ALD that reflect the VLCFA levels present in the central nervous system. Further studies are warranted to investigate whether bloodspot C26:0-lysoPC and/or C26:0-carnitine levels correlate with disease severity in ALD patients.

Conclusions
We anticipated that increased VLCFA levels in the central nervous system of Abcd1 knockout mice play a role in the clinical phenotype of the Abcd1 knockout mouse and may lead to the identification of new biomarkers. These are urgently needed as C26:0 levels do not correlate with disease severity. To this end, we developed an Abcd1 knockout mouse with a Cre-inducible ELOVL1 transgene. Our data demonstrate that enhanced in vivo expression of ELOVL1 results in increased VLCFA synthesis and consequently increased VLCFA levels. Transgenic mice displayed highly elevated levels of C26:0-lysoPC in central nervous tissue. Furthermore, we identified a new biomarker, C26:0-carnitine, in brain and spinal cord tissue from Abcd1; Cnp-ELOVL1 mice. Interestingly, highly elevated levels of C26:0-carnitine were also demonstrable in easily accessible material, i.e. bloodspots of both transgenic mice and ALD patients with the AMN phenotype. Our finding that C26:0-carnitine is elevated in dried bloodspots from males with ALD could, after validation of these results in a newborn bloodspots, result in a simple addition of C26:0-carnitine to the often already existing high throughput screen for mitochondrial fatty acid disorders thereby eliminating the need for a separate C26:0-lysoPC analysis. We anticipate that this may lead to a faster inclusion of ALD in newborn screening programs in countries that already screen for other inborn errors of metabolism.