Dienoyl-CoA reductase (DECR) deficiency with hyperlysinemia is a rare disorder affecting the metabolism of polyunsaturated fatty acids and lysine. The molecular basis of this condition is currently unknown. We describe a new case with failure to thrive, developmental delay, lactic acidosis and severe encephalopathy suggestive of a mitochondrial disorder. Exome sequencing revealed a causal mutation in NADK2. NADK2 encodes the mitochondrial NAD kinase, which is crucial for NADP biosynthesis evidenced by decreased mitochondrial NADP(H) levels in patient fibroblasts. DECR and also the first step in lysine degradation are performed by NADP-dependent oxidoreductases explaining their in vivo deficiency. DECR activity was also deficient in lysates of patient fibroblasts and could only be rescued by transfecting patient cells with functional NADK2. Thus NADPH is not only crucial as a cosubstrate, but can also act as a molecular chaperone that activates and stabilizes enzymes. In addition to polyunsaturated fatty acid oxidation and lysine degradation, NADPH also plays a role in various other mitochondrial processes. We found decreased oxygen consumption and increased extracellular acidification in patient fibroblasts, which may explain why the disease course is consistent with clinical criteria for a mitochondrial disorder. We conclude that DECR deficiency with hyperlysinemia is caused by mitochondrial NADP(H) deficiency due to a mutation in NADK2.

INTRODUCTION

Mitochondrial fatty acid β-oxidation disorders are a group of well described inborn errors of metabolism characterized by impaired β-oxidation of saturated and unsaturated fatty acids. These disorders are caused by defects in enzymes or transporters that participate in fatty acid oxidation. In addition to these enzymes and transporters, auxiliary enzymes are needed for the metabolism of different (poly)unsaturated fatty acids (1). There is currently only one disorder associated with a defect in an auxiliary enzyme, i.e. 2,4-dienoyl-CoA reductase (EC 1.3.1.34; DECR) deficiency (MIM 222745) (2). DECR is crucial for the oxidation of polyunsaturated fatty acids and reduces conjugated Δ2,Δ4-dienoyl-CoAs to a Δ3-enoyl-CoA in an NADPH-dependent reaction (3). Two different DECRs have been characterized, the mitochondrial enzyme encoded by DECR1 (46) and the peroxisomal enzyme encoded by DECR2 (7,8).

Until now DECR deficiency has been reported in one patient presenting with failure to thrive, persistent hypotonia and microcephaly (2). Biochemically, this patient was characterized by elevated plasma levels of C10:2-carnitine (2-trans, 4-cis-decadienoylcarnitine) (2). The residual DECR activity in the patient's muscle and liver was 17 and 40%, respectively, using 2-trans,4-cis decadienoyl-CoA as a substrate (2). Interestingly, this DECR-deficient patient also displayed hyperlysinemia (2). Hyperlysinemia is caused by mutations in AASS, encoding α-aminoadipic semialdehyde synthase, the first enzyme in the lysine degradation pathway (9,10). However, the relation between the accumulation of C10:2-carnitine and lysine is currently unclear. We set out to define the molecular basis of DECR deficiency with hyperlysinemia in a newly diagnosed case.

RESULTS

Case report

The first patient with DECR deficiency was reported in 1990 (2). This patient died of respiratory acidosis at 4 months of age, and material for follow-up studies is not available. We now report the identification of a second case. This patient is a Hispanic male of non-consanguineous parents, who presented at 8 weeks of age with failure to thrive, microcephaly, central hypotonia and mild dysmorphic features. Metabolic evaluation revealed elevated plasma C10:2-carnitine at diagnosis (1.3 µmol/l) and follow-up (range 0.2–1.1 µmol/l, cutoff = 0.07 µmol/l). A retrospective analysis of the North Carolina newborn screening sample (taken at 25 h after birth) indicated a mild elevation of C10:2-carnitine (0.125 µmol/l, approximately twice the upper level of normal (11)). Plasma free carnitine was low at diagnosis (18 µmol/l free, 22 µmol/l total). Lysine was elevated in plasma (831 µmol/l at diagnosis, 135–759 µmol/l at follow-up), CSF (226 µmol/l) and urine (1779 mmol/mol creatinine). Saccharopine levels were normal. Lactate was elevated in arterial blood (1.8–7.7 mmol/l) and CSF (4.1 mmol/l). Unfortunately pyruvate levels were not obtained. Urinary organic acids showed persistent abnormalities, including elevations of lactic, pyruvic, ethylmalonic, glutaric, 2-OH glutaric and fumaric acid, suggestive of severe mitochondrial dysfunction. The urinary organic acid abnormalities were even more pronounced with illness. Incubation of cultured skin fibroblasts with linoleic acid and l-carnitine produced a 6-fold greater accumulation of C10:2-carnitine than in control cells. Lysosomal enzymes, very long-chain fatty acids, pipecolic acid, CSF neurotransmitters and arrayCGH were normal. Treatment consisted of dietary lysine restriction, caloric support, provision of medium-chain fatty acids and carnitine. At age 6 months, he had slow weight gain and poor head control, and was unable to sit. At age 18 months, he had nystagmus and abducens palsies, hypertonia and clonus. He developed choreoathetosis and intermittent lactic acidosis with minor illnesses. At 30 months, he developed pancreatitis and experienced further neurological decline after a surgical procedure. At 36 months, he developed generalized aminoaciduria, persistent metabolic acidosis and urinary concentration defect consistent with mixed type I and II renal tubular acidosis. At 4.5 years, he had severe encephalopathy, dystonia, spastic quadriplegia, cortical blindness, epilepsy and episodic central apnea. He died at age 5 with aspiration pneumonia. Multiple MRIs were consistent with progressive leukodystrophy, generalized cerebral atrophy, ventriculomegaly and bilateral basal ganglia T2 abnormalities. Autopsy showed ventriculomegaly and severe white matter gliosis particularly in frontal lobes and basal ganglia. Progressive encephalopathy, dystonia, renal tubular acidosis, intermittent lactic acidosis, the presence of leukodystrophy and basal ganglia lesions on MRI and gliosis on autopsy indicate that DECR deficiency results in a neurometabolic disease consistent with clinical criteria for a mitochondrial disorder (12). A diagnostic muscle biopsy for a mitochondrial disorder was proposed, but the parents did not consent because of the neurological progression. The parents provided written informed consent for this study.

DECR deficiency is not caused by mutations in DECR1

In the first reported patient, DECR activity was measured using 2-trans,4-cis decadienoyl-CoA, an intermediate in linoleic acid degradation (2). Since 2-trans,4-cis decadienoyl-CoA is not commercially available, we used sorboyl-CoA as a substrate to measure DECR activity in the second patient. Sorboyl-CoA is a validated DECR substrate (13,14). We found that residual DECR enzyme activity was 10% in lysates of this patient's fibroblasts (Fig. 1). Thus the biochemical features of DECR deficiency are similar between both patients. Next, we set out to determine the molecular basis of DECR deficiency in the second case. Surprisingly, Sanger sequencing of the coding region and intron/exon boundaries of the most likely candidate gene, DECR1, revealed no mutations. Subsequent sequence analysis excluded mutations in three other candidate genes, i.e. DECR2, AASS and MECR.

Figure 1.

DECR enzyme activity measurements. DECR enzyme activity in control (17 different controls, 21 independent measurements) and patient fibroblasts (7 independent measurements).

Figure 1.

DECR enzyme activity measurements. DECR enzyme activity in control (17 different controls, 21 independent measurements) and patient fibroblasts (7 independent measurements).

DECR deficiency is caused by a mutation in NADK2

Since our candidate gene approach failed to reveal the molecular basis of DECR deficiency, we used exome sequencing on DNA samples from the patient and parents. Variants were annotated and prioritized using the parents and in-house exomes as unaffected controls, and assuming a recessive mode of inheritance. We identified a homozygous nonsense mutation, c.1018C>T (NM_001085411.1), in exon 10 of NADK2, which leads to a premature stop codon at position 340 (p.R340X). Sanger sequencing confirmed that the mutation was homozygous in the patient and heterozygous in the parents (Supplementary Material, Fig. S1). Although the parents reported no consanguinity, the exome sequence data revealed a 20 Mb region of homozygosity around the location of NADK2 on chromosome 5. Expression analysis using fibroblast cDNA showed that the nonsense allele was expressed at reduced levels compared with controls, indicating that the mutated mRNA is subject to nonsense-mediated decay (Fig. 2A). Immunoblot analysis revealed that the NADK2 protein was absent in the patient's fibroblasts (Fig. 2B). The p.R340X nonsense mutation is not present in the SNP, 1000 genomes and exome variant databases. We conclude that the p.R340X mutation severely affects NADK2 mRNA and protein levels.

Figure 2.

DECR deficiency is caused by a mutation in NADK2. (A) Relative expression level of NADK2 normalized for cyclophilin B in control (two different controls, four measurements) and patient fibroblasts (two measurements). (B) Immunoblot analysis of fibroblast homogenates of the patient, two controls and a case of classic hyperlysinemia.

Figure 2.

DECR deficiency is caused by a mutation in NADK2. (A) Relative expression level of NADK2 normalized for cyclophilin B in control (two different controls, four measurements) and patient fibroblasts (two measurements). (B) Immunoblot analysis of fibroblast homogenates of the patient, two controls and a case of classic hyperlysinemia.

A mutation in NADK2 causes deficient mitochondrial NADP(H) levels

Recently, NADK2 was identified as the mitochondrial NAD kinase (15,16). NAD kinases (EC 2.7.1.23) catalyze the phosphorylation of NAD to yield NADP and are considered the sole biosynthetic source of NADP. Given the absence of a mitochondrial NADP transporter, humans and other animals require a mitochondrial NAD kinase in addition to the cytosolic enzyme (encoded by NADK). Thus the NADK2 deficiency is predicted to cause a mitochondrial NADP biosynthesis defect and as a consequence deficient mitochondrial NADP(H) levels. We analyzed NADP(H) levels in fractions enriched for cytosol and mitochondria isolated from control and patient fibroblasts. NADP(H) levels were normal in the cytosol-enriched fractions of the patient's fibroblasts, but were reduced in the mitochondria-enriched fraction (Fig. 3) confirming the identification of the first inherited defect in mitochondrial NADP biosynthesis.

Figure 3.

A mutation in NADK2 causes deficient mitochondrial NADP(H) levels. NADP(H) determination in mitochondrial and cytosolic fractions of control and patient fibroblasts. The error bars indicate SD. The control value was obtained from the analysis of five different cell lines. The value for the patient fibroblasts was obtained from five independent experiments.

Figure 3.

A mutation in NADK2 causes deficient mitochondrial NADP(H) levels. NADP(H) determination in mitochondrial and cytosolic fractions of control and patient fibroblasts. The error bars indicate SD. The control value was obtained from the analysis of five different cell lines. The value for the patient fibroblasts was obtained from five independent experiments.

NADK2 deficiency causes in vivo deficiency of NADP-dependent oxidoreductases

DECR and AASS are mitochondrial oxidoreductases that require NADPH as a cosubstrate. A mutation in NADK2 and as a consequence mitochondrial NADP(H) deficiency can therefore explain the increased levels of C10:2-carnitine and lysine, and thus the combined defect in polyunsaturated fatty acid oxidation and lysine degradation. Interestingly the presence of NADPH in the DECR enzyme assays did not restore activity (Fig. 1), despite the fact that DECR1 protein levels are not affected as judged from an immunoblot (Fig. 2B). This suggests that the prevailing mitochondrial NADPH deficiency leads to inactivation of DECR1. Experiments aimed to reactivate DECR activity by incubating lysates of the patient fibroblasts for up to 1 day with high levels of NADP(H) were unsuccessful. In contrast to DECR1, AASS protein levels are 3.1-fold reduced in fibroblasts of this patient (Figs 2B and 4B). Therefore NADPH deficiency most likely renders the AASS protein less stable.

Figure 4.

Rescue of deficient DECR activity in patient fibroblasts by wild-type NADK2. (A) DECR enzyme activity analysis after complementation of the NADK2-deficient fibroblasts with Myc-DDK-tagged NADK2. The DECR enzyme activity is displayed as the mean of three independent transfections for the control and four independent transfections for the patient fibroblasts. (B) Immunoblot analysis after complementation of the NADK2-deficient fibroblasts with Myc-DDK-tagged NADK2. (C) Quantification of AASS protein levels using immunoblot analysis after complementation of the NADK2-deficient fibroblasts with Myc-DDK-tagged NADK2. The AASS protein levels are displayed as the mean of three independent transfections for the control and four independent transfections for the patient fibroblasts.

Figure 4.

Rescue of deficient DECR activity in patient fibroblasts by wild-type NADK2. (A) DECR enzyme activity analysis after complementation of the NADK2-deficient fibroblasts with Myc-DDK-tagged NADK2. The DECR enzyme activity is displayed as the mean of three independent transfections for the control and four independent transfections for the patient fibroblasts. (B) Immunoblot analysis after complementation of the NADK2-deficient fibroblasts with Myc-DDK-tagged NADK2. (C) Quantification of AASS protein levels using immunoblot analysis after complementation of the NADK2-deficient fibroblasts with Myc-DDK-tagged NADK2. The AASS protein levels are displayed as the mean of three independent transfections for the control and four independent transfections for the patient fibroblasts.

Rescue of deficient DECR activity in patient fibroblasts by wild-type NADK2

In order to rescue the mitochondrial NADP biosynthesis defect, we transfected a plasmid encoding Myc-DDK-tagged NADK2 into control and NADK2-deficient fibroblasts (Fig. 4A). After transfection, DECR activity increased 2.2-fold in patient fibroblasts, whereas DECR activity in controls remained unchanged (Fig. 4B). In addition, AASS protein levels in the patient cell line increased by 1.8-fold upon introduction of Myc-DDK-NADK2 (Fig. 4C). Although the rescue was not complete due to limited transfection efficiency, this result unambiguously proves that the NADK2 mutations in our patient are responsible for the enzymatic defects in DECR deficiency with hyperlysinemia.

Impaired mitochondrial function due to NADK2 deficiency

NADPH is crucial for a wide variety of mitochondrial processes, including lipoic acid biosynthesis. Lipoic acid is a cofactor for all mitochondrial alpha-keto dehydrogenase complexes and the glycine cleavage system. Given the lactic acidosis in the patient, we measured the levels of lipoic acid in the E2 subunits of pyruvate dehydrogenase (DLAT) and alpha-ketoglutarate dehydrogenase (DLST) using immunoblot (Fig. 5A). Surprisingly, lipoic acid was normal in these E2 subunits, even when cells were cultured in serum free medium. This indicates that NADPH may not be essential for lipoid acid biosynthesis, at least in cultured fibroblasts.

Figure 5.

Impaired mitochondrial function due to NADK2 deficiency. (A) Immunoblot analysis of lipoic acid content in the E2 subunits of pyruvate dehydrogenase (DLAT) and alpha-ketoglutarate dehydrogenase (DLST) in patient and control fibroblast homogenates. Cells were also analyzed after culturing 2 days without fetal calf serum (FCS-) supplementation. (B) Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) in control and patient fibroblasts. The addition of oligomycin is marked by ‘a’, the addition of FCCP by ‘b’ and the addition of antimycin A by ‘c’. Maximal respiration is obtained after addition of FCCP.

Figure 5.

Impaired mitochondrial function due to NADK2 deficiency. (A) Immunoblot analysis of lipoic acid content in the E2 subunits of pyruvate dehydrogenase (DLAT) and alpha-ketoglutarate dehydrogenase (DLST) in patient and control fibroblast homogenates. Cells were also analyzed after culturing 2 days without fetal calf serum (FCS-) supplementation. (B) Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) in control and patient fibroblasts. The addition of oligomycin is marked by ‘a’, the addition of FCCP by ‘b’ and the addition of antimycin A by ‘c’. Maximal respiration is obtained after addition of FCCP.

To obtain a more general measure of mitochondrial function, we used extracellular flux analysis to measure the rates of oxygen consumption (OCR) and extracellular acidification (ECAR) during a mitochondrial stress test. This revealed that maximal oxygen consumption after the addition of FCCP was reduced in patient fibroblasts, which coincided with an overall elevated extracellular acidification (Fig. 5B). This shows that mitochondrial NADPH deficiency impairs mitochondrial processes beyond linoleic acid and lysine metabolism.

DISCUSSION

In this report we set out to define the molecular basis of DECR deficiency in a newly diagnosed case. Originally DECR deficiency was described as an inborn error of polyunsaturated fatty acid degradation (2). Unexpectedly, we found that this disease is not caused by mutations in DECR1, encoding the mitochondrial DECR. Instead exome sequencing revealed a causal mutation in NADK2. NADK2 encodes the mitochondrial NAD kinase, which is crucial for NADP biosynthesis. Subsequent studies showed a selective decrease in mitochondrial NADP(H) levels and as a consequence a deficiency of NADPH-dependent mitochondrial enzymes, such as DECR, but also AASS. As such the deficiency of the NADPH cosubstrate can explain the combined accumulation of C10:2-carnitine and lysine. These enzymatic defects were reversed upon overexpression of wild-type NADK2 in patients' fibroblasts, proving that the NADK2 mutations in this patient are responsible for the enzymatic defects in DECR deficiency with hyperlysinemia.

From a pathophysiological perspective, both patients had a severe and progressive neurometabolic disease. It seems unlikely that the defects in DECR1 and AASS are solely responsible for this disease. Indeed, the Decr1 knockout mouse displayed increased levels of C10:2-carnitine and fasting-induced hypoglycemia, a symptom common in other mitochondrial fatty acid β-oxidation defects, but appeared otherwise healthy (14). Hyperlysinemia caused by mutations in AASS is generally considered a benign metabolic variant (10,17). Although the severe disease might be caused by the combination of these two biochemical abnormalities, it is, however, more likely that any of the other mitochondrial NADP-dependent processes contribute to the disease. These include: reductive carboxylation in the citric acid cycle via NADP-dependent isocitrate dehydrogenase (IDH2) (18), reductive carboxylation of pyruvate for anaplerosis via mitochondrial NADP-dependent malic enzyme (ME3) (19), mitochondrial lipoic acid biosynthesis via the trans-2-enoyl-CoA reductase (MECR) (20), and mitochondrial metabolism of reactive oxygen species via glutathione reductase (GSR) (21). In addition, the mitochondrial cytochrome P450 enzymes depend on electrons from NADPH, which are transferred via ferredoxin reductase and ferredoxin 1. This mitochondrial NADPH deficiency is expected to impact on adrenal steroidogenesis, bile acid biosynthesis and vitamin D synthesis (22). More recently ferredoxin 2 (FDX2), which also accepts electrons from NADPH, was shown to function in iron-sulfur cluster biosynthesis (23). FDX2 depletion diminished levels of mitochondrial iron-sulfur cluster containing proteins such as aconitase, succinate dehydrogenase, cytochrome oxidase and complex I. This wide variety in mitochondrial NADPH-dependent processes suggests that NADK2 deficiency will result in general mitochondrial dysfunction. Indeed, failure to thrive, developmental delays, hypotonia, progressive encephalopathy, choreoathetosis, renal tubular acidosis, lactic acidosis and biochemical findings observed in DECR deficiency with hyperlysinemia are consistent with clinical criteria for mitochondrial disorders (12). Although we were unable to show a specific defect in lipoic acid biosynthesis, we provide evidence for decreased oxygen consumption and increased extracellular acidification in patient fibroblasts.

Newborn screening for DECR deficiency is included in the Recommended Uniform Screening Panel, presumably by using C10:2-carnitine as the primary marker (24). Using retrospective analysis we were able to show that C10:2-carnitine was mildly elevated in the newborn screening blood spot of this patient. Slight elevations of this analyte are not specific and sensitive for this disorder as similar elevations are seen in other fatty acid oxidation disorders (11). These fatty acid oxidation disorders, however, have their own specific marker acylcarnitine, and thus isolated elevation of C10:2-carnitine is suggestive for DECR deficiency. The screening may be improved by including lysine. Screening programs in the USA and elsewhere have included DECR since at least 2006, but to the best of our knowledge, no cases have been detected. Based on rarity of this disorder and the lack of effective treatment, the inclusion of this condition should be reconsidered. The identification of NADK2 as the affected gene, however, will now enable better diagnostics and genetic counseling for this condition.

In summary, we have found that a defect in the mitochondrial biosynthesis of NADP leads to DECR deficiency with hyperlysinemia. NADPH is not only crucial as the cosubstrate for DECR and AASS, but also acts as a molecular chaperone that activates and stabilizes these enzymes. In addition to polyunsaturated fatty acid oxidation and lysine degradation, NADPH also plays a role in various other mitochondrial processes, which may explain why the clinical course of this disease suggests a mitochondrial disorder.

MATERIALS AND METHODS

Exome sequencing

Approximately 1.5 µg of genomic DNA was sheared followed by ligation of system specific barcoded adaptors according to manufacturer's guidelines (Life Technologies, Fragment Library Preparation 5500 Series SOLiD™ Systems User Guide 4460960A). After eight cycles of amplification, three quality checked libraries were pooled with equimolar contributions. Whole exome capture was performed starting with 2 µg pooled library using system and barcode specific blockers, following supplier's protocol (Roche Nimblegen, SeqCap EZ Library SR User's Guide v3.0). Post capture PCR product was polyclonally amplified by means of Emulsion PCR and sequenced on a SOLiD 5500xl (Life Technologies). For each sample an average of 150 million barcode specific single-end reads (75 bp forward) and 50 million paired-end reads (75 bp forward and 35 bp reverse) were generated.

NGS analysis pipeline

Initial alignment of the raw sequence reads was performed in color space for the individual samples using the Lifescope aligner (v2.5.1) from Applied Biosystems (http://www.appliedbiosystems.com). Paired-end and single-end runs were aligned separately to the HG19 reference genome as provided by Lifescope. For paired-end reads duplicates were already marked by Lifescope; in case of single-end reads, duplicate marking was performed using Picards MarkDuplicates (v1.60). The obtained BAM files were merged into multisample BAM files, one for paired-end reads and one for single-end reads. Realignment was then performed using GATK (v2.2-5-g3bf5e3f) applying the intervals from the Nimblegen v3 target file and the known biallelic indels from the 1000G project. Recalibration was performed using GATK v1.6-5-g557da77 (25). Resulting final BAM files for the single- and paired-end reads were merged and variants were called using GATKs UnifiedGenotyper (v2.1-11-g13c0244) against a set of non-disease related in-house exomes (26). Variants were annotated and prioritized using KGGSeq v0.3 (http://statgenpro.psychiatry.hku.hk/limx/kggseq/) designating the parents and in-house exomes as unaffected samples (27). Settings were set as not to filter any variants using the strategies offered by KGGSeq. The dbSNP135 and ESP5400 databases were only used to obtain the allele frequencies for the observed variants. The resulting KGGSeq and GATK VCF file were combined to ease further analysis. The variant was found using a homozygous recessive model and sorting on the variant being disease causing as predicted by the training model implemented in KGGSeq. For more details and applied settings, see Supplementary Material.

Sanger sequencing

All exons, plus flanking intronic sequences of DECR1, DECR2 and AASS were sequenced after amplification by PCR from genomic DNA. All forward and reverse primers were tagged with a −21M13 or M13rev sequence, respectively. PCR fragments were sequenced in two directions using −21M13 and M13rev primers by means of BigDye Terminator Cycle Sequencing (v1.1, Applied Biosystems, Foster City, CA, USA) and analyzed on an Applied Biosystems 3130xl or 3730xl DNA analyzer. The c.1018C>T mutation in NADK2 was confirmed using Sanger sequencing. For this, exon 10 was amplified using a forward primer in intron 9 (5′-GGGGGCTACTGTTTATTTATGA-3′) and a reverse primer in intron 10 (5′-CTATCCCCCAGGACATGTTT-3′).

Mutation analysis was also performed on cDNA using exonic primers for DECR1 en MECR. PCR fragments were sequenced in both directions. For this, RNA was isolated from fibroblast pellets using Trizol extraction. cDNA was synthesized by using the Superscript II Reverse Transcriptase Kit (Invitrogen, Carlsbad, CA, USA). Quantitative real-time PCR analysis of DECR1 in fibroblasts was performed using the LC480 Sybr Green I Master mix (Roche).

DECR activity measurement

The CoA ester of sorbic acid was synthesized chemically using established methods (28). DECR activity was measured using 15 µM sorboyl-CoA. Assays were performed in 100 µl 100 mm potassium phosphate buffer (pH 7.4) with 125 µM NADPH and 50 µg/ml BSA. Reaction were terminated after 15 min of incubation at 37°C with 0.2 M HCl and subsequently neutralized. Reaction products were analyzed by reversed phase HPLC. Chromatographic separation of the reaction products was achieved on a Supelcosil LC-18S column (250 × 4.6 mm, 5 µm, Supelco) under gradient elution by a binary mixture of potassium dihydrogen phosphate (pH 5.9, 0.05 M) and methanol at a flow rate of 1.0 ml/min. A Shimadzu SPD-10A UV-VIS detector set at 260 nm was used for detection of the CoA ester.

Fibroblast fractionation

Fibroblast pellets were resuspended in 1 ml of MTE buffer (250 mm mannitol, 5 mm TRIS, 0.5 mm EDTA, pH 7.4). The cells were lysed using 10 passages through the cell cracker (European Molecular Biology Laboratory, Heidelberg, Germany) with a clearance of 6 µm. The homogenate was centrifuged at 1000g, after which the supernatant was transferred to a new tube and centrifuged at 10 000g. The resulting supernatant was considered the cytosolic fraction. The pellet was washed with 1 ml MTE buffer, and centrifuged at 3600g. This final pellet was resuspended in 250 µl MTE buffer and considered the mitochondrial fraction. All fractions were homogenized by sonication before further analysis. Purity of the fraction was assessed by the distribution of the marker enzymes glutamate dehydrogenase (GDH, mitochondrial) and phosphoglucoisomerase (PGI, cytosolic). The cytosolic fraction contained on average 88% of the total PGI activity with 6% of the GDH activity. The mitochondrial fraction contained 50% of the total GDH activity with 4% of the PGI activity. The purity of the fractions did not differ between control and patient fibroblasts.

NADP(H) measurement

One hundred µl of the cell fractions obtained by differential centrifugation was extracted by addition of 700 µl acetonitrile (ACN). The ACN extracts were evaporated to dryness and dissolved in 100 µl milli-Q water. This extract was used for the NADP(H) measurement using an enzymatic cycling assay (29). The assay mixture contained 100 mm TRIS buffer pH 8.0, 5 mm α-ketoglutarate, 5 U/ml glutamate dehydrogenase (from beef liver, Roche), 100 mm ADP, 5 mm glucose-6-phosphate, 5 U/ml glucose-6-phosphate dehydrogenase (from yeast, grade I, Roche), 25 mm ammonium acetate and 0.2 mg/ml BSA. This enzymatic system will cycle the added NADP(H) leading to a cumulative production of 6-phosphoglucono-δ-lactone. After 30 min of cycling the reaction was terminated by boiling for 5 min, which converts the lactone into 6-phosphogluconate. For quantification, a standard curve of NADP was used in the same cycling procedure. The produced 6-phosphogluconate was then analyzed spectrophotometrically (COBAS Fara centrifugal analyzer, Roche) by an endpoint method using NADPH production by 6-phosphogluconate dehydrogenase. The assay conditions were as follows, 25 mm Tris pH 7.7, 30 mm ammonium acetate, 0.1 mm EDTA, 0.2 mg/ml BSA and 0.5 mm NADP. The reaction was initiated by the addition of 20 U/ml 6-phosphogluconate dehydrogenase (from yeast, Roche).

Immunoblot

Fibroblast homogenates were prepared in PBS using sonication (twice on ice, 40 J at 8 W output). Samples were further prepared according to the instructions for electrophoresis of NuPAGE Bis-Tris mini gels (4–12%; Life Technologies) using NuPAGE LDS Sample Buffer for denaturation and NuPAGE Sample Reducing Agent for reduction of the protein disulfide bonds. Equal amounts of protein (25 µg) were loaded. An anti-NADK2 rabbit polyclonal antibody raised against amino acids 250–346 of the human protein was obtained from Abnova (dilution 1:1000; PAB23271). An anti-lipoic acid rabbit polyclonal antibody raised against lipoic acid linked to KLH was obtained from Calbiochem (Cat. no. 437695). The anti-DECR1 rabbit antiserum was a generous gift from Dr H. Schulz. An anti-AASS rabbit polyclonal antibody (1 in 500 dilution) raised against amino acids 528–649 of the protein was obtained from Sigma-Aldrich (HPA020728). Antibodies were visualized using IRDye 800CW or IRDye 680RD anti-rabbit secondary antibodies and the Odyssey Infrared Imaging System (Li-Cor Biosciences).

Fibroblast transfection

After reaching confluency, control and patients fibroblasts cells were isolated and transfected with a plasmid encoding Myc-DDK-tagged NADK2 (RC214147, Origene, Rockville, MD, USA). Briefly, 5 µg of the plasmid was transfected into 5 × 105 fibroblasts cells using the Nucleofector kit for Human Dermal Fibroblast (Lonza) as the transfection reagent and the Nucleofector Program U-023 according the manufacturer's instructions. After transfection the cells were immediately plated. Seven days after transfection, the cells were harvested and used for DECR activity assays and immunoblotting.

Extracellular flux analysis

To monitor the OCR and ECAR in intact human fibroblasts, we used the Seahorse Bioscience XF96 Extracellular Flux Analyzer (Seahorse Bioscience, North Billerica, MA, USA). For this, 1.5 × 104 fibroblasts were seeded in 200 µl of DMEM with 10% FCS and incubated overnight at 37°C in 5% CO2 atmosphere. After replacing the growth medium with 200 µl of bicarbonate and serum free DMEM, cells were preincubated at 37°C for 1 h before starting the analysis. OCR and ECAR were analyzed using a 2 min mix followed by 3 min measurement cycle. Oligomycin was used at a final concentration of 1.5 µM, carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP) at 1 µM and antimycin A at 2.5 µM. OCR and ECAR were normalized to protein content as measured using the BCA assay.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

AUTHORS' CONTRIBUTIONS

S.M.H., R.C.M.H. and R.J.A.W. conceived and designed the study. S.M.H. drafted the manuscript. S.D. carried out the biochemical studies. H.t.B. carried out the molecular genetic studies. E.J.B. and F.B. performed the exome sequencing. A.J., S.M.H. and A.H.C.v.K. performed the exome sequencing analysis. D.S.M., S.P.Y., D.M.F. and M.G.-C. were involved in the clinical evaluation, diagnosis and management of the patient. All authors read and approved the final manuscript.

FUNDING

This work was supported by the Netherlands Organization for Scientific Research (NWO Vidi grant 016.086.336 to S.M.H.).

ACKNOWLEDGMENTS

The authors thank Barbera van Schaik for initiating the implementation of an exome analysis pipeline and the patient's family who consented to testing in order to characterize this condition.

Conflict of Interest statement. None declared.

REFERENCES

1
Hiltunen
J.K.
Qin
Y.
beta-oxidation—strategies for the metabolism of a wide variety of acyl-CoA esters
Biochim. Biophys. Acta
 
2000
1484
117
128
2
Roe
C.R.
Millington
D.S.
Norwood
D.L.
Kodo
N.
Sprecher
H.
Mohammed
B.S.
Nada
M.
Schulz
H.
McVie
R.
2,4-Dienoyl-coenzyme A reductase deficiency: a possible new disorder of fatty acid oxidation
J. Clin. Invest.
 
1990
85
1703
1707
3
Schulz
H.
Kunau
W.H.
Beta-oxidation of unsaturated fatty acids—a revised pathway
Trends Biochem. Sci.
 
1987
12
403
406
4
Helander
H.M.
Koivuranta
K.T.
Horelli-Kuitunen
N.
Palvimo
J.J.
Palotie
A.
Hiltunen
J.K.
Molecular cloning and characterization of the human mitochondrial 2,4-dienoyl-CoA reductase gene (DECR)
Genomics
 
1997
46
112
119
5
Hirose
A.
Kamijo
K.
Osumi
T.
Hashimoto
T.
Mizugaki
M.
cDNA cloning of rat liver 2,4-dienoyl-CoA reductase
Biochim. Biophys. Acta
 
1990
1049
346
349
6
Koivuranta
K.T.
Hakkola
E.H.
Hiltunen
J.K.
Isolation and characterization of cDNA for human 120 kDa mitochondrial 2,4-dienoyl-coenzyme A reductase
Biochem. J.
 
1994
304
Pt 3
787
792
7
Fransen
M.
Van Veldhoven
P.P.
Subramani
S.
Identification of peroxisomal proteins by using M13 phage protein VI phage display: molecular evidence that mammalian peroxisomes contain a 2,4-dienoyl-CoA reductase
Biochem. J.
 
1999
340
Pt 2
561
568
8
Geisbrecht
B.V.
Liang
X.
Morrell
J.C.
Schulz
H.
Gould
S.J.
The mouse gene PDCR encodes a peroxisomal delta(2), delta(4)-dienoyl-CoA reductase
J. Biol. Chem.
 
1999
274
25814
25820
9
Sacksteder
K.A.
Biery
B.J.
Morrell
J.C.
Goodman
B.K.
Geisbrecht
B.V.
Cox
R.P.
Gould
S.J.
Geraghty
M.T.
Identification of the alpha-aminoadipic semialdehyde synthase gene, which is defective in familial hyperlysinemia
Am. J. Hum. Genet.
 
2000
66
1736
1743
10
Houten
S.M.
te Brinke
H.
Denis
S.
Ruiter
J.P.
Knegt
A.C.
de Klerk
J.B.
Augoustides-Savvopoulou
P.
Haberle
J.
Baumgartner
M.R.
Coskun
T.
et al.  
Genetic basis of hyperlysinemia
Orphanet J. Rare Dis.
 
2013
8
57
11
McHugh
D.
Cameron
C.A.
Abdenur
J.E.
Abdulrahman
M.
Adair
O.
Al Nuaimi
S.A.
Ahlman
H.
Allen
J.J.
Antonozzi
I.
Archer
S.
et al.  
Clinical validation of cutoff target ranges in newborn screening of metabolic disorders by tandem mass spectrometry: a worldwide collaborative project
Genet. Med.
 
2011
13
230
254
12
Wolf
N.I.
Smeitink
J.A.
Mitochondrial disorders: a proposal for consensus diagnostic criteria in infants and children
Neurology
 
2002
59
1402
1405
13
Hakkola
E.H.
Hiltunen
J.K.
The existence of two mitochondrial isoforms of 2,4-dienoyl-CoA reductase in the rat
Eur. J. Biochem.
 
1993
215
199
204
14
Miinalainen
I.J.
Schmitz
W.
Huotari
A.
Autio
K.J.
Soininen
R.
Ver Loren van
T.E.
Baes
M.
Herzig
K.H.
Conzelmann
E.
Hiltunen
J.K.
Mitochondrial 2,4-dienoyl-CoA reductase deficiency in mice results in severe hypoglycemia with stress intolerance and unimpaired ketogenesis
PLoS Genet.
 
2009
5
e1000543
15
Ohashi
K.
Kawai
S.
Murata
K.
Identification and characterization of a human mitochondrial NAD kinase
Nat. Commun.
 
2012
3
1248
16
Zhang
R.
MNADK, a novel liver-enriched mitochondrion-localized NAD kinase
Biol. Open
 
2013
2
432
438
17
Dancis
J.
Hutzler
J.
Ampola
M.G.
Shih
V.E.
van Gelderen
H.H.
Kirby
L.T.
Woody
N.C.
The prognosis of hyperlysinemia: an interim report
Am. J. Hum. Genet.
 
1983
35
438
442
18
Des Rosiers
C.
Fernandez
C.A.
David
F.
Brunengraber
H.
Reversibility of the mitochondrial isocitrate dehydrogenase reaction in the perfused rat liver. Evidence from isotopomer analysis of citric acid cycle intermediates
J. Biol. Chem.
 
1994
269
27179
27182
19
Des Rosiers
C.
Labarthe
F.
Lloyd
S.G.
Chatham
J.C.
Cardiac anaplerosis in health and disease: food for thought
Cardiovasc. Res.
 
2011
90
210
219
20
Miinalainen
I.J.
Chen
Z.J.
Torkko
J.M.
Pirila
P.L.
Sormunen
R.T.
Bergmann
U.
Qin
Y.M.
Hiltunen
J.K.
Characterization of 2-enoyl thioester reductase from mammals. An ortholog of YBR026p/MRF1'p of the yeast mitochondrial fatty acid synthesis type II
J. Biol. Chem.
 
2003
278
20154
20161
21
Andreyev
A.Y.
Kushnareva
Y.E.
Starkov
A.A.
Mitochondrial metabolism of reactive oxygen species
Biochemistry (Mosc)
 
2005
70
200
214
22
Ewen
K.M.
Kleser
M.
Bernhardt
R.
Adrenodoxin: the archetype of vertebrate-type [2Fe-2S] cluster ferredoxins
Biochim. Biophys. Acta
 
2011
1814
111
125
23
Sheftel
A.D.
Stehling
O.
Pierik
A.J.
Elsasser
H.P.
Muhlenhoff
U.
Webert
H.
Hobler
A.
Hannemann
F.
Bernhardt
R.
Lill
R.
Humans possess two mitochondrial ferredoxins, Fdx1 and Fdx2, with distinct roles in steroidogenesis, heme, and Fe/S cluster biosynthesis
Proc. Natl Acad. Sci. USA
 
2010
107
11775
11780
24
American College of Medical Genetics' Newborn Screening Expert Group
Newborn screening: toward a uniform screening panel and system
Genet. Med.
 
2006
8
Suppl. 1
1S
252S
25
McKenna
A.
Hanna
M.
Banks
E.
Sivachenko
A.
Cibulskis
K.
Kernytsky
A.
Garimella
K.
Altshuler
D.
Gabriel
S.
Daly
M.
et al.  
The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data
Genome Res.
 
2010
20
1297
1303
26
DePristo
M.A.
Banks
E.
Poplin
R.
Garimella
K.V.
Maguire
J.R.
Hartl
C.
Philippakis
A.A.
del Angel
G.
Rivas
M.A.
Hanna
M.
et al.  
A framework for variation discovery and genotyping using next-generation DNA sequencing data
Nat. Genet.
 
2011
43
491
498
27
Li
M.X.
Gui
H.S.
Kwan
J.S.
Bao
S.Y.
Sham
P.C.
A comprehensive framework for prioritizing variants in exome sequencing studies of Mendelian diseases
Nucleic Acids Res.
 
2012
40
e53
28
Rasmussen
J.T.
Borchers
T.
Knudsen
J.
Comparison of the binding affinities of acyl-CoA-binding protein and fatty-acid-binding protein for long-chain acyl-CoA esters
Biochem. J.
 
1990
265
849
855
29
Lowry
O.H.
Passonneau
J.V.
Schulz
D.W.
Rock
M.K.
The measurement of pyridine nucleotides by enzymatic cycling
J. Biol. Chem.
 
1961
236
2746
2755