Abstract

Cofactor disorders of mitochondrial energy metabolism are a heterogeneous group of diseases with a wide variety of clinical symptoms, particular metabolic profiles and variable enzymatic defects. Mutations in NFU1, BOLA3, LIAS and IBA57 have been identified in patients with deficient lipoic acid-dependent enzymatic activities and defects in the assembly and activity of the mitochondrial respiratory chain complexes. Here, we report a patient with an early onset fatal lactic acidosis presenting a biochemical phenotype compatible with a combined defect of pyruvate dehydrogenase (PDHC) and 2-ketoglutarate dehydrogenase (2-KGDH) activities, which suggested a deficiency in lipoic acid metabolism. Immunostaining analysis showed that lipoylated E2-PDH and E2-KGDH were extremely reduced in this patient. However, the absence of glycine elevation, the normal activity of the glycine cleavage system and the normal lipoylation of the H protein suggested a defect of lipoic acid transfer to particular proteins rather than a general impairment of lipoic acid biosynthesis as the potential cause of the disease. By analogy with yeast metabolism, we postulated LIPT1 as the altered candidate gene causing the disease. Sequence analysis of the human LIPT1 identified two heterozygous missense mutations (c.212C>T and c.292C>G), segregating in different alleles. Functional complementation experiments in patient's fibroblasts demonstrated that these mutations are disease-causing and that LIPT1 protein is required for lipoylation and activation of 2-ketoacid dehydrogenases in humans. These findings expand the spectrum of genetic defects associated with lipoic acid metabolism and provide the first evidence of a lipoic acid transfer defect in humans.

INTRODUCTION

Cofactor disorders of mitochondrial energy metabolism constitute a heterogeneous and emerging group of diseases with a wide variety of clinical symptoms associated with particular metabolic profiles and variable enzymatic defects. Although most of the genes involved in mitochondrial cofactor biosynthesis still remain to be elucidated, an increasing number of them have been recently identified and described to be involved in human diseases (1–9).

Lipoic acid is an essential cofactor necessary for the activity of four mitochondrial enzyme complexes: pyruvate dehydrogenase (PDHC), 2-ketoglutarate dehydrogenase (2-KGDH), branched-chain 2-keto acid dehydrogenase (BCKDH) and the glycine cleavage system (GCS) (10). Lipoic acid is covalently attached to a conserved lysine residue of the E2 subunits of PDHC, 2-KGDH and BCKDH complexes and to the H protein of the GCS. The biosynthesis of lipoic acid is not completely understood, but current models propose a multistep process that involves the mitochondrial fatty acid synthetase II (mtFASII) pathway (10) and the iron-sulfur cluster biosynthesis (11). Bacterial lipoic acid metabolism has been well characterized, and more recent studies in yeast have just begun to elucidate the molecular basis of lipoic acid synthesis in eukaryotes. According to these models, octanoic acid conjugated with an acyl carrier protein (ACP) derived from the mtFASII pathway is transferred to target apoproteins and acts as the substrate for sulfur insertion by the iron-sulfur protein lipoic acid synthetase (LIAS) (10). Homologs of the prokaryotic enzymes in the pathway of lipoic acid synthesis have been identified in both yeast and mammals (10). However, in contrast to bacteria, eukaryotic cells cannot use exogenously supplied lipoic acid and depend exclusively on de novo intramitochondrial synthesis (12–14).

Recent studies have reported disease-causing mutations in genes encoding proteins involved in the maturation of the mitochondrial iron-sulfur clusters: NFU1 (MIM 608100), BOLA3 (MIM 613183), IBA57 (MIM 615316) and LYRM4 (MIM 613311) (1,2,7–9). All these proteins but ISD11 (9), encoded by LYRM4, have been shown to result in a deficiency of lipoic acid because of the fact that LIAS requires iron-sulfur cluster prosthetic groups (15). In addition, a patient with mutations affecting LIAS (MIM 607031) has also been reported (4). Clinically, all these patients presented with a fatal encephalopathy, some of them associated with pulmonary hypertension (1,16). The biochemical phenotype was consistent with deficient lipoic acid-dependent enzymatic activities, variable lactic acidemia and hyperglycinemia (1,2,4,7,8).

We have identified an individual with mutations in lipoyltransferase 1 gene (LIPT1, MIM 610284) associated with encephalopathy, pulmonary hypertension and a particular biochemical phenotype compatible with a specific defect in the lipoylation and activation of the 2-ketoacid dehydrogenase complexes with normal GCS and no evidence of glycine elevation in body fluids. This is the first report of a defect in lipoic acid transfer associated with human disease.

RESULTS

Clinical and biochemical findings suggestive for a combined defect of 2-ketoacid dehydrogenase complexes

The severe clinical presentation associated with lactic acidosis, increase of α-alanine and of 2-ketoglutarate led us to suspect a mitochondrial disease, but the enzyme activities and assembly of the respiratory chain complexes in muscle biopsy and skin fibroblasts were normal (Table 1, Supplementary Material, Fig. S1). However, PDHC activity was clearly reduced (Table 1), whereas the partial PDHC reactions catalyzed by pyruvate decarboxylase (PDH-E1) and dihydrolipoamide dehydrogenase (PDH-E3) subunits were normal. Dihydrolipoamide acetyl transferase (PDH-E2) activity was not measured, but analysis of the cDNA sequence of DLAT (MIM 608770), encoding for PDH-E2, did not reveal any mutation. Moreover, sequence analysis of the genes encoding for the remaining PDHC subunits [PDHA1 (MIM 300502), PDHB (MIM 179060), DLD (MIM 238331) and PDHX (MIM 608769)] did not reveal sequence variations, excluding a primary PDHC deficiency. The rates of oxidation of 14-C-substrates (pyruvate and leucine) in fibroblasts were low, whereas the rates of oxidation of glutamate and succinate were slightly below the reference range (Table 1).

Table 1.

Mitochondrial energy metabolism enzyme activities

Enzyme activities in muscle biopsy Patient Controls 
 PDHC total 0.2 0.9–3.5 
 PDH-E1 0.2 >0.06 
 PDH-E3 228 63–143 
 Complex I + III 32 12–56 
 Complex II 4–10 
 Complex II + III 7–24 
 Complex III 146 55–259 
 Complex IV 208 59–170 
 Citrate synthase 150 71–200 
Enzyme activities in fibroblasts 
 PDHC total 0.1 0.3–2.6 
 Complex II + III 2–17 
 Complex IV 41 17–48 
Substrate oxidation rates 
 [1 -14C] pyruvate 4.5 8–36 
 [2 -14C] pyruvate 0.3 2–12 
 [1 -14C] leucine 0.1 0.17–0.48 
 [1 -14C] glutamate 3.7 4.5–25 
 [2,3 -14C] succinate 0.03 0.04–0.28 
GCS activity in liver necropsy 73 45–195 
Enzyme activities in muscle biopsy Patient Controls 
 PDHC total 0.2 0.9–3.5 
 PDH-E1 0.2 >0.06 
 PDH-E3 228 63–143 
 Complex I + III 32 12–56 
 Complex II 4–10 
 Complex II + III 7–24 
 Complex III 146 55–259 
 Complex IV 208 59–170 
 Citrate synthase 150 71–200 
Enzyme activities in fibroblasts 
 PDHC total 0.1 0.3–2.6 
 Complex II + III 2–17 
 Complex IV 41 17–48 
Substrate oxidation rates 
 [1 -14C] pyruvate 4.5 8–36 
 [2 -14C] pyruvate 0.3 2–12 
 [1 -14C] leucine 0.1 0.17–0.48 
 [1 -14C] glutamate 3.7 4.5–25 
 [2,3 -14C] succinate 0.03 0.04–0.28 
GCS activity in liver necropsy 73 45–195 

PDHC and mitochondrial respiratory chain activities in muscle are expressed as nmol/min/mg protein; PDHC and mitochondrial respiratory chain activities in fibroblasts are expressed as mU/mg protein; substrate oxidation rates are expressed as nmol/h/mg protein; GCS activity is expressed as µkat/kg protein. Altered values are outlined in bold.

Altogether these findings led us to speculate that a deficiency of a common cofactor of the 2-ketoacid dehydrogenase complexes might be the cause of the disease. The two candidates were lipoic acid and thiamin as they are known to be necessary for the activity of these enzymatic complexes. However, the low PDHC activity, measured in the presence of thiamin in the incubation mixture, made the diagnosis of thiamin deficiency unlikely (3).

Defective lipoylation of the E2 subunits of the PDH and 2-KGDH complexes was deficient whereas it was normal for the H protein of the GCS

To determine whether lipoic acid metabolism was impaired in this individual, we analyzed fibroblasts from this patient by immunostaining using an antibody that specifically recognizes lipoic acid bound to proteins. Immunohistochemical analysis showed complete absence of protein-bound lipoic acid (Fig. 1A). Immunoblot staining in control cell lines detected two predominant lipoylated proteins of 65 and 50 kDa, corresponding to lipoic acid-bound PDH-E2 and 2-KGDH-E2, respectively (Fig. 1B) (17). In contrast, a complete absence of lipoylated E2 subunits of both PDHC and 2-KGDH was observed in the patient's fibroblasts, whereas PDH-E2 protein content was normal (Fig. 1B).

Figure 1.

Immunofluorescence staining and immunoblot analysis of protein-bound lipoic acid. (A) Immunofluorescence staining with an antibody that specifically recognizes lipoic acid bound to proteins (anti-LA) showed complete absence of protein lipoylation in fibroblasts from the affected individual compared with control cells. (B) Immunoblot analysis performed in extracts obtained from skin fibroblasts of the affected individual showed a severe decrease in the lipoylated forms of E2 subunits of PDHC (LA-PDH-E2) and 2-KGDH (LA-KGDH-E2), whereas the E2 subunit of PDHC (PDH-E2) was in normal amounts, as well as GAPDH used as a control protein. P1: affected individual; C1 and C2: controls.

Figure 1.

Immunofluorescence staining and immunoblot analysis of protein-bound lipoic acid. (A) Immunofluorescence staining with an antibody that specifically recognizes lipoic acid bound to proteins (anti-LA) showed complete absence of protein lipoylation in fibroblasts from the affected individual compared with control cells. (B) Immunoblot analysis performed in extracts obtained from skin fibroblasts of the affected individual showed a severe decrease in the lipoylated forms of E2 subunits of PDHC (LA-PDH-E2) and 2-KGDH (LA-KGDH-E2), whereas the E2 subunit of PDHC (PDH-E2) was in normal amounts, as well as GAPDH used as a control protein. P1: affected individual; C1 and C2: controls.

Patients carrying mutations in genes causing disruption of lipoic acid biosynthesis (NFU1, BOLA3, LIAS and IBA57) also showed high levels of glycine in plasma, urine and cerebrospinal fluid (CSF) owing to a defect in the lipoylation of the H protein of the CGS (1,2,4,7,8). Interestingly, the patient reported here showed no evidence of glycine elevation. Accordingly, the lipoylation of the H protein and the enzymatic activity of the GCS measured in liver necropsy were completely normal (Table 1, Fig. 2). In the western blot studies (Fig. 2), a patient carrying NFU1 mutations was used as negative control and, in contrast to the patient described here, he showed complete absence of H protein lipoylation as well as of E2 subunits of PDHC and 2-KGDH. The particular biochemical phenotype of our patient, together with the protein lipoylation pathway proposed in yeast models (18), raised the rationale for the investigation of the LIPT1 in this patient, the human homolog of yeast LIP3 (Table 2, Fig. 3).

Table 2.

Enzymes involved in lipoic acid biosynthesis and transfer

 E. coli S. cerevisiae H. sapiens 
Lipoic acid synthetase LipA LIP5 LIAS 
Lipoyl-octanoyl-protein ligases LipB LIP2 LIPT2 
 LplA LIP3 LIPT1 
 E. coli S. cerevisiae H. sapiens 
Lipoic acid synthetase LipA LIP5 LIAS 
Lipoyl-octanoyl-protein ligases LipB LIP2 LIPT2 
 LplA LIP3 LIPT1 
Figure 2.

Immunoblot analysis of protein-bound lipoic acid in liver tissue from the affected individual. Protein extracts obtained from liver necropsy were analyzed using an antibody that specifically recognizes lipoic acid bound to proteins. Immunoblot analysis showed a specific defect in the lipoylation of the E2 subunits of PDHC (LA-PDH-E2) and 2-KGDH (LA-KGDH-E2) but not of the H protein of the GCS (LA-GCS-H). ATP5A1 was used as a loading control. A protein extract obtained from a NFU1-deficient individual was used as a reference sample for known lipoylation defects in the E2 subunits of PDHC and 2-KGDH, and the H protein of the GCS. C1 and C2: controls; P1: affected individual; P-NFU1: individual carrying NFU1 mutations.

Figure 2.

Immunoblot analysis of protein-bound lipoic acid in liver tissue from the affected individual. Protein extracts obtained from liver necropsy were analyzed using an antibody that specifically recognizes lipoic acid bound to proteins. Immunoblot analysis showed a specific defect in the lipoylation of the E2 subunits of PDHC (LA-PDH-E2) and 2-KGDH (LA-KGDH-E2) but not of the H protein of the GCS (LA-GCS-H). ATP5A1 was used as a loading control. A protein extract obtained from a NFU1-deficient individual was used as a reference sample for known lipoylation defects in the E2 subunits of PDHC and 2-KGDH, and the H protein of the GCS. C1 and C2: controls; P1: affected individual; P-NFU1: individual carrying NFU1 mutations.

Figure 3.

Proposed model for lipoic acid biosynthesis and transfer. According to previous observations in yeast, octanoic acid, conjugated with an ACP derived from the mtFASII pathway, is transferred to the H protein of GCS by the action of the octanoyl-lipoyl-protein ligase LIP2 (LIPT2 in humans). The octanoylated H protein might be the substrate for sulfur insertion by LIP5, the yeast homolog of LIAS in humans. LIP3 is required for the octanoylation and, therefore, the subsequent lipoylation of the E2 components of 2-ketoacid dehydrogenase complexes. Human homologs of yeast LIP5, LIP2 and LIP3 proteins are highlighted in bold. NFU1, BOLA3 and IBA57, three Fe-S cluster proteins associated with human disease and known to be involved in Fe-S delivery to LIAS, are also depicted. The LIPT1 mutations reported in this work cause a severe mitochondrial disease characterized by a specific defect in the lipoylation of the E2 subunits of the 2-ketoacid dehydrogenase complexes but not of the H protein of the GCS.

Figure 3.

Proposed model for lipoic acid biosynthesis and transfer. According to previous observations in yeast, octanoic acid, conjugated with an ACP derived from the mtFASII pathway, is transferred to the H protein of GCS by the action of the octanoyl-lipoyl-protein ligase LIP2 (LIPT2 in humans). The octanoylated H protein might be the substrate for sulfur insertion by LIP5, the yeast homolog of LIAS in humans. LIP3 is required for the octanoylation and, therefore, the subsequent lipoylation of the E2 components of 2-ketoacid dehydrogenase complexes. Human homologs of yeast LIP5, LIP2 and LIP3 proteins are highlighted in bold. NFU1, BOLA3 and IBA57, three Fe-S cluster proteins associated with human disease and known to be involved in Fe-S delivery to LIAS, are also depicted. The LIPT1 mutations reported in this work cause a severe mitochondrial disease characterized by a specific defect in the lipoylation of the E2 subunits of the 2-ketoacid dehydrogenase complexes but not of the H protein of the GCS.

Identification of mutations in LIPT1

We next analyzed the coding region of LIPT1 (RefSeq NM_001204830.1) for the presence of mutations. Two heterozygous nucleotide substitutions, c.212C>T and c.292C>G, predicted to change serine 71 to phenylalanine (p.Ser71Phe) and arginine 98 to glycine (p.Arg98Gly), respectively, were identified in the affected individual (Fig. 4). In silico analysis (Polyphen2®) predicted a deleterious effect for both mutations. Because the parents of this patient were healthy, we hypothesized an autosomal recessive inheritance for this disease. To demonstrate whether the c.212C>T and c.292C>G substitutions were in separate alleles, we amplified by PCR the LIPT1-coding region and ligated into a pcDNA3.1 vector. Colony screening of transformed cells showed that the c.212C>T and the c.292C>G mutations segregated alternatively and were, therefore, in different alleles (Supplementary Material, Fig. S2). Carrier rate for the c.212C>T substitution was analyzed by PCR followed by restriction fragment length polymorphism (RFLP), using the fact that this mutation removes a restriction site for MnlI. No carrier was found in a control group of 130 alleles from healthy Spanish population (Supplementary Material, Fig. S3). Interestingly, the c.292C>G substitution has already been reported in the SNPs public databases (rs137973334) in 1 out of 758 alleles of European population but, according to the 1000 genomes population genetics project, no homozygous individual for this change has been reported. These data, together with the highly deleterious effect predicted by the in silico analysis of this substitution, suggest that the c.292C>G nucleotide change might be a disease-causing mutation rather than a rare polymorphic variant.

Figure 4.

Sequence analysis of LIPT1. (A) Sequence analysis of LIPT1 identified two heterozygous substitutions, c.212C>T (p.Ser71Phe) and c.292C>G (p.Arg98Gly), which affects amino acids located close to phylogenetically conserved domains of the LIPT1 protein. (B) Multi-sequence alignment of the LIPT1 protein in human and other organisms. The c.212C>T and c.292C>G substitutions are indicated. These residues are closely located to conserved biotin/lipoate protein ligase domain (BLP_LplA_LipB) of the LIPT1 protein. (C) Protein extracts obtained from liver tissue of the affected individual showed normal levels of LIPT1 protein expression. ATP5A1 was used as a control protein. P1: affected individual; C: control.

Figure 4.

Sequence analysis of LIPT1. (A) Sequence analysis of LIPT1 identified two heterozygous substitutions, c.212C>T (p.Ser71Phe) and c.292C>G (p.Arg98Gly), which affects amino acids located close to phylogenetically conserved domains of the LIPT1 protein. (B) Multi-sequence alignment of the LIPT1 protein in human and other organisms. The c.212C>T and c.292C>G substitutions are indicated. These residues are closely located to conserved biotin/lipoate protein ligase domain (BLP_LplA_LipB) of the LIPT1 protein. (C) Protein extracts obtained from liver tissue of the affected individual showed normal levels of LIPT1 protein expression. ATP5A1 was used as a control protein. P1: affected individual; C: control.

LIPT1 mutations are disease-causing by impairing protein lipoylation

To test the effect of these mutations on LIPT1, we analyzed protein expression levels by western blot in liver tissue extracts using an antibody against LIPT1. Results showed normal content of LIPT1 protein in the patient liver extract, suggesting that the p.Ser71Phe and the p.Arg98Gly mutations may not affect protein stability (Fig. 4C).

Therefore, to demonstrate that LIPT1 mutations are disease-causing, we performed complementation studies in fibroblasts derived from this patient and tested the functional rescue of the defective protein lipoylation. Fibroblasts were electroporated with plasmids encoding for the wild-type LIPT1, as well as for the p.Ser71Phe and the p.Arg98Gly mutants. Transfection was verified by PCR analysis of DNA obtained from stably transfected cell lines using LIPT1 and vector-specific primers (Supplementary Material, Fig. S4). Stable cell lines were grown on glass coverslips and analyzed by immunostaining for protein-bound lipoic acid (Fig. 5). Results showed that cells expressing wild-type LIPT1 had a marked increase of lipoylation of mitochondrial proteins, as seen by co-localization of protein-bound lipoic acid and ATP5A1, used as a mitochondrial control protein, which is not altered in our patient. In contrast, cells transfected with the empty vector or plasmids encoding for p.Ser71Phe and p.Arg98Gly LIPT1 mutants showed absence of protein-bound lipoic acid staining (Fig. 5).

Figure 5.

Rescue of protein lipoylation in the affected individual with LIPT1 mutations. Complementation experiments in fibroblasts confirmed the pathogenicity of LIPT1 mutations. Fibroblasts from the affected individual were stably transfected with plasmids encoding wild-type and mutant LIPT1, followed by immunofluorescence analysis of protein-bound lipoic acid. ATP5A1 was used as a mitochondrial marker. Wild-type but not mutant LIPT1 restores lipoylation of mitochondrial proteins, as demonstrated by co-localization with ATP5A1.

Figure 5.

Rescue of protein lipoylation in the affected individual with LIPT1 mutations. Complementation experiments in fibroblasts confirmed the pathogenicity of LIPT1 mutations. Fibroblasts from the affected individual were stably transfected with plasmids encoding wild-type and mutant LIPT1, followed by immunofluorescence analysis of protein-bound lipoic acid. ATP5A1 was used as a mitochondrial marker. Wild-type but not mutant LIPT1 restores lipoylation of mitochondrial proteins, as demonstrated by co-localization with ATP5A1.

LIPT1 mutations impair lipoic acid-dependent enzymatic activities

To ensure that the recovery observed in the lipoylation of mitochondrial proteins upon wild-type LIPT1 expression also resulted in a functional rescue of the biochemical phenotype, PDHC activity and substrate oxidation rates were tested in the cells transfected with the wild-type and mutant LIPT1 (Fig. 6). The activity of PDHC and the oxidation rates of pyruvate and leucine were increased upon transfection with the wild-type LIPT1 but not with a plasmid encoding for the p.Arg98Gly-mutated protein. Intriguingly, patient's fibroblasts expressing the p.Ser71Phe mutant protein did not show any increase in the activity of PDHC and pyruvate oxidation but restored leucine oxidation rate to ∼50% of the levels seen in wild-type LIPT1-transfected cells (Fig. 6).

Figure 6.

Rescue of lipoic acid-dependent enzyme activities in the affected individual with LIPT1 mutations. Complementation experiments in fibroblasts confirmed the functional impairment caused by LIPT1 mutations. Fibroblasts from the affected individual were stably transfected with plasmids encoding for wild-type and mutant LIPT1. The activity of PDHC as well as the oxidation rates of pyruvate and leucine was analyzed in these cells. Transfection with wild-type LIPT1 but not with plasmids encoding the p.Ser71Phe- and the p.Arg98Gly-mutated proteins increased the oxidation rate of both substrates.

Figure 6.

Rescue of lipoic acid-dependent enzyme activities in the affected individual with LIPT1 mutations. Complementation experiments in fibroblasts confirmed the functional impairment caused by LIPT1 mutations. Fibroblasts from the affected individual were stably transfected with plasmids encoding for wild-type and mutant LIPT1. The activity of PDHC as well as the oxidation rates of pyruvate and leucine was analyzed in these cells. Transfection with wild-type LIPT1 but not with plasmids encoding the p.Ser71Phe- and the p.Arg98Gly-mutated proteins increased the oxidation rate of both substrates.

Cellular localization and expression of LIPT1

To understand the regulation and the physiological behavior of LIPT1, we have determined the cellular localization of this protein and analyzed its expression levels in a series of human control tissues. Permeabilized COS7 cells containing the mitochondrial and the cytosolic fractions as well as the whole cell lysate were analyzed by immunostaining. According to the in silico prediction analysis (Mitoprot®), sub-cellular fractionation in COS7 cells showed that LIPT1 is exclusively located in mitochondria because it co-fractionates with mitochondrial but not with cytosolic markers (Fig. 7A). We also demonstrated that LIPT1 is ubiquitously expressed in a wide variety of human tissues and follows a similar pattern to those seen for the corresponding lipoylated proteins (Fig. 7B).

Figure 7.

Localization and expression of LIPT1 in human tissues. (A) Subcellular fractionation in COS7 cells indicates the predominant mitochondrial localization of LIPT1. LIPT1 co-fractionated with mitochondrial lipoylated proteins (LA-PDH-E2 and LA-KGDH-E2) but not with GAPDH (cytosol). (B) Immunoblot analysis using LIPT1 antibody was performed in extracts obtained from human control tissue samples of pediatric age. Results showed ubiquitous expression of LIPT1 protein, being higher in liver, kidney, heart, and brain. Lipoylated proteins were detected using anti-LA antibody. GAPDH was used as loading control.

Figure 7.

Localization and expression of LIPT1 in human tissues. (A) Subcellular fractionation in COS7 cells indicates the predominant mitochondrial localization of LIPT1. LIPT1 co-fractionated with mitochondrial lipoylated proteins (LA-PDH-E2 and LA-KGDH-E2) but not with GAPDH (cytosol). (B) Immunoblot analysis using LIPT1 antibody was performed in extracts obtained from human control tissue samples of pediatric age. Results showed ubiquitous expression of LIPT1 protein, being higher in liver, kidney, heart, and brain. Lipoylated proteins were detected using anti-LA antibody. GAPDH was used as loading control.

DISCUSSION

We report on the identification of mutations in LIPT1 as the cause of defective lipoic acid transfer to specific target proteins in a patient with an early-onset fatal disease leading to death on the ninth day of life. Extreme bradycardia, hypertonia, dystonic movements and pulmonary hypertension were the main clinical symptoms. Lipoic acid is an essential cofactor required for the activity of four mitochondrial enzymes involved in the oxidative decarboxylation of pyruvate, 2-ketoglutarate and branched-chain amino acids and in the glycine cleavage (10). Recent studies have reported disease-causing mutations in genes encoding for proteins involved in lipoic acid biosynthesis: NFU1, BOLA3, LIAS and IBA57 (1,2,4,7,8). Patients carrying mutations in these genes showed a similar clinical and biochemical phenotype resulting in early-onset fatal disease characterized by reduced levels of protein-bound lipoic acid (1,2,4,7,8), low PDHC activity (1,2,4,7), low pyruvate (1,4) and leucine oxidation (1) rates and increase of glycine in serum and CSF (1,2,4,7,8), and low or undetectable GCS in liver extracts (1). Moreover, all the defects but LIAS (4), which catalyzes the final step of lipoic acid synthesis, are associated with multiple respiratory chain deficiencies (1,2,7,8) as those proteins are involved in the iron-sulfur cluster biogenesis, which are essential prosthetic groups of the respiratory chain complexes I, II and III (19,20). As it has been reported in LIAS deficiency (4), our patient showed normal respiratory chain activities in muscle and fibroblasts, but sequence analysis excluded mutations in LIAS (data not shown). The clinical picture of the patient resembles that of NFU1 defects, presenting with fatal infantile encephalopathy and pulmonary hypertension (1,16). The fact that lipoic acid is essential for the activity of several key enzymes of the mitochondrial energy metabolism explains the severe course of the disease in patients with defects in the metabolism of this cofactor. In addition to the patient described here, pulmonary hypertension has also been reported in some individuals carrying NFU1 and BOLA3 mutations. To our knowledge, this finding has never been reported in primary deficiencies of PDHC, 2-KGDH, BCKDH or GCS. Therefore, it seems that lipoic acid itself or the underlipoylated proteins could be responsible for the pulmonary hypertension. In fact, it has been reported in cellular models that a partial reduction of protein-bound lipoic acid levels by the downregulation of LIAS generates oxidative stress (17), which to some extend could contribute to pulmonary hypertension in these patients (21). Another attractive hypothesis could be related to a defect in heme biosynthesis, which is a cofactor of nitric oxide synthase. This enzyme is involved in the formation of the pulmonary vasculature, and its deficiency is associated with pulmonary hypertension (22). The biosynthesis of heme requires succinyl-CoA as a precursor (23). Therefore, the deficient lipoylation and the subsequent deficiency in the activity of PDHC and 2-KGDH could potentially decrease the efficiency of Krebs cycle and limit the availability of succinyl-CoA for heme biosynthesis. However, this hypothesis was not sustained by experiments in HeLa cells depleted for IBA57, ISCA1 and ISCA2 that showed completely normal levels of heme regarding the strong defect of protein lipoylation, making this hypothesis unlikely (24).

The particular biochemical phenotype with normal glycine as well as normal GCS in liver pointed to a deficiency of lipoic acid transfer to particular target proteins rather than a general impairment of lipoic acid biosynthesis. Lipoic acid biosynthesis is well known in E. coli where, in addition to LIAS homolog (LipA), two protein ligases (LipB and LplA) have been described to be involved in octanoyl and lipoyl transfer to target proteins (10). Homologs of these proteins have also been identified in eukaryotes (Table 2). Experiments in yeast strains deficient in the homologs of LIAS (LIP5) or in the lipoyltransferase LIPT2 (LIP2) demonstrated that proteins encoded by LIP5 and LIP2 are absolutely required for the lipoylation of the E2 subunits of 2-ketoacid dehydrogenase complexes, as well as for the lipoylation of the H protein. In contrast, deletion of LIP3, the homolog of human LIPT1, only abrogated the lipoylation of the E2 subunits but not that of the H protein, suggesting a specific role for LIP3 in the lipoylation of the E2 subunits of 2-ketoacid dehydrogenase complexes (18). These observations suggest a model in which the lipoylated H protein may act as an intermediate for the lipoylation of the E2 subunits of 2-ketoacid dehydrogenase proteins. However, a direct transfer of lipoic acid from H protein to the E2 subunits has not been demonstrated, but this process is known to be mediated by LIP3. Recently, a study performed in yeast provided more detailed information of LIP3 function (25). It demonstrated that LIP3 is, in fact, an octanoyl-CoA transferase required for the octanoylation and, therefore, the subsequent lipoylation of the E2 components of 2-ketoacid dehydrogenase complexes (Fig. 3). Therefore, the identification of LIPT1 mutations in humans together with the proposed role for the yeast homolog, LIP3, could potentially explain the biochemical phenotype of the patient we report here.

The normal levels of LIPT1 detected in this individual suggested that the identified missense mutations may not affect the stability of the protein. However, the crystal structure of the bovine LIPT1 homolog showed that the amino acids modified in our patient are located in a small β-sheet comprising three strands, which contain important residues for lipoic acid binding. In particular, the arginine located at position 98 is a residue conserved in all organisms that has been demonstrated to be involved in lipoic acid binding (26). These observations are in agreement with the hypothesis that the mutations identified in the proband could impair lipoic acid binding or transfer to target proteins. To this effect, functional complementation studies in patient's fibroblasts demonstrated that LIPT1 mutations are disease-causing and that functional LIPT1 is required for the lipoylation and activation of a particular set of lipoic acid-dependent enzymes in humans. However, the fact that the expression of the mutated p.Ser71Phe protein in patient's fibroblasts partially restored the oxidation rate of leucine but not that of pyruvate or PDHC activity made us to hypothesize that this residue could be involved in the interaction of LIPT1 to particular target proteins, such as the E2 subunit of BCKDH, rather than in lipoic acid binding.

The identification of mutations in LIPT1 as a cause of a fatal mitochondrial disease highlights the proposed pivotal role of LIPT1 in lipoic acid transfer to key enzymes of the mitochondrial energy metabolism (10,18,25). However, the precise biological function of LIPT1 as well as the mechanisms of lipoic acid binding and transfer to target proteins is not well known in humans. Here, we provide new insights concerning the physiological regulation of LIPT1 by demonstrating that this protein is localized in the mitochondria and is ubiquitously expressed in a wide variety of tissues.

In summary, this report expands the spectrum of genetic alterations associated with lipoic acid biosynthesis and provides the first evidence of a lipoic acid transfer defect in humans. The biochemical phenotype of the patient we report here provides new insights for the diagnosis and for the physiological mechanisms underlying lipoic acid biosynthesis and transfer to target proteins. We demonstrated that LIPT1 is a mitochondrial protein that may play a specific role in the lipoylation of the E2 subunits of the 2-ketoacid dehydrogenase complexes but not of the H protein of the GCS.

Although more detailed studies will be needed to completely delineate the protein lipoylation pathway in humans, our observations suggest a strong parallelism between humans and yeast in this metabolic process. Finally, this report highlights the importance of an accurate biochemical characterization and the usefulness of the metabolic knowledge in other model organisms to identify the genetic causes of human diseases.

MATERIALS AND METHODS

Patients' clinical history

The affected individual, a girl, presented an early-onset fatal disease, leading to death on the ninth day of life. She was the first child of healthy non-consanguineous parents and was born after a normal pregnancy and delivery. The family reported a previous spontaneous miscarriage. Birth weight (3180 g), head circumference (34.5 cm) and Apgar scores (9/10/10) were within the normal limits. She had an initial symptom-free period but, at two days of life, she presented sudden clinical deterioration and severe bradycardia. She showed moderate jaundice, dystrophy (12% weight loss), hypoactivity, weak cry and slight dehydration. A perinatal infection was suspected, and intravenous therapy with antibiotics was started. Two days later, a sudden clinical deterioration owing to extreme bradycardia led to artificial ventilation. Treatment with vasoactive drugs was started, and she was then transferred to our hospital. At admission, neurological examination revealed generalized hypertonia and dystonic movements of hands and feet. Her thumbs were remarkably adducted. Metabolic investigation in plasma revealed severe lactic acidosis (lactate 32 mmol/L, controls: 1.3–2.3) and liver dysfunction (ALT 206 IU/L, controls: 2–42; AST 277 IU/L, controls: 4–77; prothrombin time 69%, controls: 80–120; bilirubin 111 μmol/L, controls: 68–102; conjugated bilirubin 12 μmol/L, controls: <5). Amino acid analysis showed general hypoaminoacidaemia with high levels of α-alanine (901 μmol/L, controls: 190–337) and proline (448 μmol/L, controls: 90–270) in agreement with the previously detected lactic acidosis and liver dysfunction. The urinary organic acid profile showed a huge excretion of lactate (10 317 mmol/mol creatinine, controls: 15–60) and 2-ketoglutarate (1879 mmol/mol creatinine, controls: 10–271). Brain imaging was normal, but echocardiogram indicated severe pulmonary hypertension with left ventricle dilatation. ECG showed no paroxysms. The child underwent continuous arteriovenous hemodiafiltration, and therapy with biotin, thiamin and l-carnitine was started. Forty-eight hours later, lactate dropped to 6.3 mmol/L. However, no clinical improvement was observed. Moreover, thrombopenia (51 000 platelets/mm3), facial myoclonus and increasing generalized hypertonia with brisk deep tendon reflexes appeared. Limitation of ventilatory support led to progressive deterioration, cardiorespiratory arrest and death at 9 days of life. The histopathologic study of the brain showed patched periventricular gliosis and moderate astrocytosis with gemistocytic astrocytes in basal ganglia.

Biochemical and enzymatic studies

Organic acids in urine were analyzed as trimethylsilyl-esther derivatives by gas chromatography–mass spectrometry as previously described (27). Amino acids in plasma and urine were derivatized with phenylisothiocyanate (PITC) according to the Pico Tag method of Waters and analyzed by HPLC (Waters, Manchester, UK).

PDHC activity and substrate oxidation rates were analyzed in fibroblast by measuring 14CO2 production from [1-14C]-pyruvate in PDHC activity and [1-14C]-pyruvate, [1-14C]-leucine and [14C]-glutamate for substrate oxidations (3).

Respiratory chain activities were determined in skeletal muscle as described (28–30). GCS activity in liver necropsy was measured as previously described (31).

Mutation analysis

RNA was extracted from fibroblasts of the affected individual using QIAshredder and RNeasy kits (Qiagen, Hilden, Germany). Single-stranded cDNA was obtained using oligo-dT primers and M-MLV reverse transcriptase, RNase H Minus, Point Mutant (Promega, Madison, WI, USA) according to the manufacturer's protocol. cDNA was amplified by PCR followed by Sanger sequencing. Primers are listed in Supplementary Material, Table S1. Carrier rate for the c.212C>T substitution was analyzed by PCR followed by RFLP using MnlI (New England Biolabs, Ipswich, MA, USA).

Cell culture and generation of stable cell lines

Patient fibroblasts were maintained in DMEM containing 4.5 g/l glucose and supplemented with 10% fetal calf serum, 1 mm glutamine and 1% penicillin–streptomycin.

LIPT1-coding region was amplified by PCR and ligated into a pcDNA3.1 vector (Invitrogen, Paisley, UK). Patient fibroblasts were electroporated with plasmids encoding for wild-type LIPT1, as well as for the p.Ser71Phe and the p.Arg98Gly mutants (Gene Pulser Xcell eukaryotic system, BioRad) and grown with medium supplemented with 1 mg/ml geneticin (G418). Stably transfected cells were verified by PCR using vector-specific primers (Supplementary Material, Fig. S4).

Immunofluorescence

Fibroblasts grown onto glass coverslips were fixed in ice-cold 4% formaldehyde for 15 min at room temperature. Fixed cells were permeabilized and blocked (0.1% Triton X-100, 1% BSA) for 15 min at room temperature and incubated with the indicated antibodies for 1 h at room temperature. Appropriate secondary antibodies (Santa Cruz Biotechnology, Heidelberg, Germany) were incubated for 45 min at room temperature. Coverslips were mounted using Ultracruz Mounting Medium (Santa Cruz Biotechnology) and analyzed under fluorescence microscope (Eclipse 50i, Nikon instruments Inc., Melville, USA). Antibodies used were anti-protein-conjugated lipoic acid (Calbiochem, Darmstadt, Germany) and ATP5A1 (Invitrogen).

Sub-cellular fractionation

COS7 cells were permeabilized with digitonin, and the cell lysate was centrifuged for 10 min at 15.000g. Pellet (containing mitochondria) and supernatant fraction (containing cytosol) were analyzed by western blot using the indicated antibodies.

Western blot and BN-PAGE

Fibroblasts and tissues obtained from patient and control individuals were homogenized in SETH buffer (10 mm Tris–HCl pH 7.4, 0.25 m sucrose, 2 mm EDTA, 5 × 104 U/l heparin). Cleared lysates were subjected to SDS–PAGE and electroblotted, and proteins were visualized by immunostaining with specific antibodies followed by colorimetric detection (Opti-4CNTM Substrate Kit, Bio-Rad, USA). The assembly of mitochondrial respiratory chain complexes was analyzed by BN-PAGE as described (32). Protein-conjugated lipoic acid antibody (Calbiochem), LIPT1 (Sigma, St. Louis, MO, USA) and GAPDH (Santa Cruz Biotechnology) were used in this study. Antibodies against the mitochondrial respiratory subunits NDUFA9 (complex I), SDHA (complex II), UQCRC2 (complex III) and ATP5A1 (complex V) were from Invitrogen. Anti-COX-subunit Va (complex IV) was from MitoSciences (Eugene, Oregon, USA).

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

FUNDING

This research was supported by the Instituto de Salud Carlos III (FISPI12/01138, FIS PI08/90348, FIS PI08/0307) and the Centro de Investigación Biomédica en Red de Enfermedades Raras (CIBERER), an initiative of the Instituto de Salud Carlos III (Ministerio de Ciencia e Innovación, Spain).

ACKNOWLEDGEMENTS

This work was performed in the context of the Medicine PhD Program of the University of Barcelona (UB). We are grateful to the family involved in this study. We thank Dr W. Ruitenbeck, for kindly providing antibodies against PDHC.

Conflict of Interest statement. None declared.

REFERENCES

1
Navarro-Sastre
A.
Tort
F.
Stehling
O.
Uzarska
M.A.
Arranz
J.A.
Del Toro
M.
Labayru
M.T.
Landa
J.
Font
A.
Garcia-Villoria
J.
, et al.  . 
A fatal mitochondrial disease is associated with defective NFU1 function in the maturation of a subset of mitochondrial Fe-S proteins
Am. J. Hum. Genet.
 , 
2011
, vol. 
89
 (pg. 
656
-
667
)
2
Cameron
J.M.
Janer
A.
Levandovskiy
V.
Mackay
N.
Rouault
T.A.
Tong
W.H.
Ogilvie
I.
Shoubridge
E.A.
Robinson
B.H.
Mutations in iron-sulfur cluster scaffold genes NFU1 and BOLA3 cause a fatal deficiency of multiple respiratory chain and 2-oxoacid dehydrogenase enzymes
Am. J. Hum. Genet.
 , 
2011
, vol. 
89
 (pg. 
486
-
495
)
3
Mayr
J.A.
Freisinger
P.
Schlachter
K.
Rolinski
B.
Zimmermann
F.A.
Scheffner
T.
Haack
T.B.
Koch
J.
Ahting
U.
Prokisch
H.
, et al.  . 
Thiamine pyrophosphokinase deficiency in encephalopathic children with defects in the pyruvate oxidation pathway
Am. J. Hum. Genet.
 , 
2011
, vol. 
89
 (pg. 
806
-
812
)
4
Mayr
J.A.
Zimmermann
F.A.
Fauth
C.
Bergheim
C.
Meierhofer
D.
Radmayr
D.
Zschocke
J.
Koch
J.
Sperl
W.
Lipoic acid synthetase deficiency causes neonatal-onset epilepsy, defective mitochondrial energy metabolism, and glycine elevation
Am. J. Hum. Genet.
 , 
2011
, vol. 
89
 (pg. 
792
-
797
)
5
Ylikallio
E.
Suomalainen
A.
Mechanisms of mitochondrial diseases
Ann. Med.
 , 
2012
, vol. 
44
 (pg. 
41
-
59
)
6
Quinzii
C.M.
Hirano
M.
Primary and secondary CoQ(10) deficiencies in humans
Biofactors
 , 
2011
, vol. 
37
 (pg. 
361
-
365
)
7
Haack
T.B.
Rolinski
B.
Haberberger
B.
Zimmermann
F.
Schum
J.
Strecker
V.
Graf
E.
Athing
U.
Hoppen
T.
Wittig
I.
, et al.  . 
Homozygous missense mutation in BOLA3 causes multiple mitochondrial dysfunctions syndrome in two siblings
J. Inherit. Metab. Dis.
 , 
2013
, vol. 
36
 (pg. 
55
-
62
)
8
Ajit Bolar
N.
Vanlander
A.V.
Wilbrecht
C.
Van der Aa
N.
Smet
J.
De Paepe
B.
Vandeweyer
G.
Kooy
F.
Eyskens
F.
De Latter
E.
, et al.  . 
Mutation of the iron-sulfur cluster assembly gene IBA57 causes severe myopathy and encephalopathy
Hum. Mol. Genet.
 , 
2013
, vol. 
22
 (pg. 
2590
-
2602
)
9
Lim
S.C.
Friemel
M.
Marum
J.E.
Tucker
E.J.
Bruno
D.L.
Riley
L.G.
Christodoulou
J.
Kirk
E.P.
Boneh
A.
Degennaro
C.
, et al.  . 
Mutations in LYRM4, encoding iron-sulfur cluster biogenesis factor ISD11, cause deficiency of multiple respiratory chain complexes
Hum. Mol. Genet.
 , 
2013
, vol. 
22
 (pg. 
4460
-
4473
)
10
Hiltunen
J.K.
Autio
K.J.
Schonauer
M.S.
Kursu
V.A.
Dieckmann
C.L.
Kastaniotis
A.J.
Mitochondrial fatty acid synthesis and respiration
Biochim. Biophys. Acta.
 , 
2010
, vol. 
1797
 (pg. 
1195
-
1202
)
11
Booker
S.J.
Cicchillo
R.M.
Grove
T.L.
Self-sacrifice in radical S-adenosylmethionine proteins
Curr. Opin. Chem. Biol.
 , 
2007
, vol. 
11
 (pg. 
543
-
552
)
12
Sulo
P.
Martin
N.C.
Isolation and characterization of LIP5. A lipoate biosynthetic locus of Saccharomyces cerevisiae
J. Biol. Chem.
 , 
1993
, vol. 
268
 (pg. 
17634
-
17639
)
13
Morris
T.W.
Reed
K.E.
Cronan
J.E.
Jr.
Lipoic acid metabolism in Escherichia coli: the lplA and lipB genes define redundant pathways for ligation of lipoyl groups to apoprotein
J. Bacteriol.
 , 
1995
, vol. 
177
 (pg. 
1
-
10
)
14
Yi
X.
Maeda
N.
Endogenous production of lipoic acid is essential for mouse development
Mol. Cell. Biol.
 , 
2005
, vol. 
25
 (pg. 
8387
-
8392
)
15
Duin
E.C.
Lafferty
M.E.
Crouse
B.R.
Allen
R.M.
Sanyal
I.
Flint
D.H.
Johnson
M.K.
[2Fe-2S] to [4Fe-4S] cluster conversion in Escherichia coli biotin synthase
Biochemistry
 , 
1997
, vol. 
36
 (pg. 
11811
-
11820
)
16
Seyda
A.
Newbold
R.F.
Hudson
T.J.
Verner
A.
MacKay
N.
Winter
S.
Feigenbaum
A.
Malaney
S.
Gonzalez-Halphen
D.
Cuthbert
A.P.
, et al.  . 
A novel syndrome affecting multiple mitochondrial functions, located by microcell-mediated transfer to chromosome 2p14–2p13
Am. J. Hum. Genet.
 , 
2001
, vol. 
68
 (pg. 
386
-
396
)
17
Padmalayam
I.
Hasham
S.
Saxena
U.
Pillarisetti
S.
Lipoic acid synthase (LASY): a novel role in inflammation, mitochondrial function, and insulin resistance
Diabetes
 , 
2009
, vol. 
58
 (pg. 
600
-
608
)
18
Schonauer
M.S.
Kastaniotis
A.J.
Kursu
V.A.
Hiltunen
J.K.
Dieckmann
C.L.
Lipoic acid synthesis and attachment in yeast mitochondria
J. Biol. Chem.
 , 
2009
, vol. 
284
 (pg. 
23234
-
23242
)
19
Lill
R.
Hoffmann
B.
Molik
S.
Pierik
A.J.
Rietzschel
N.
Stehling
O.
Uzarska
M.A.
Webert
H.
Wilbrecht
C.
Mühlenhoff
U.
The role of mitochondria in cellular iron-sulfur protein biogenesis and iron metabolism
Biochim. Biophys. Acta.
 , 
2012
, vol. 
1823
 (pg. 
1491
-
1508
)
20
Rouault
T.A.
Biogenesis of iron-sulfur clusters in mammalian cells: new insights and relevance to human disease
Dis Model Mech.
 , 
2012
, vol. 
5
 (pg. 
155
-
164
)
21
Wong
C.M.
Bansal
G.
Pavlickova
L.
Marcocci
L.
Suzuki
Y.J.
Reactive oxygen species and antioxidants in pulmonary hypertension
Antioxid Redox Signal.
 , 
2013
, vol. 
18
 (pg. 
1789
-
1796
)
22
Han
R.N.
Stewart
D.J.
Defective lung vascular development in endothelial nitric oxide synthase-deficient mice
Trends Cardiovasc. Med.
 , 
2006
, vol. 
16
 (pg. 
29
-
34
)
23
Ye
H.
Rouault
T.A.
Human iron-sulfur cluster assembly, cellular iron homeostasis, and disease
Biochemistry
 , 
2010
, vol. 
49
 (pg. 
4945
-
4956
)
24
Sheftel
A.D.
Wilbrecht
C.
Stehling
O.
Niggemeyer
B.
Elsässer
H.P.
Mühlenhoff
U.
Lill
R.
The human mitochondrial ISCA1, ISCA2, and IBA57 proteins are required for [4Fe-4S] protein maturation
Mol. Biol. Cell.
 , 
2012
, vol. 
23
 (pg. 
1157
-
1166
)
25
Hermes
F.A.
Cronan
J.E.
The role of the Saccharomyces cerevisiae lipoate protein ligase homologue, Lip3, in lipoic acid synthesis
Yeast.
 , 
2013
, vol. 
30
 (pg. 
415
-
427
)
26
Fujiwara
K.
Hosaka
H.
Matsuda
M.
Okamura-Ikeda
K.
Motokawa
Y.
Suzuki
M.
Nakagawa
A.
Taniguchi
H.
Crystal structure of bovine poyltransferase in complex with lipoyl-AMP
J. Mol. Biol.
 , 
2007
, vol. 
371
 (pg. 
222
-
234
)
27
Ribes
A.
Riudor
E.
Briones
P.
Christensen
E.
Campistol
J.
Millington
D.S.
Significance of bound glutarate in the diagnosis of glutaric aciduria type I
J Inherit. Metab. Dis.
 , 
1992
, vol. 
15
 (pg. 
367
-
370
)
28
Montero
R.
Artuch
R.
Briones
P.
Nascimento
A.
García-Cazorla
A.
Vilaseca
M.A.
Sánchez-Alcázar
J.A.
Navas
P.
Montoya
J.
Pineda
M.
Muscle coenzyme Q10 concentrations in patients with probable and definite diagnosis of respiratory chain disorders
Biofactors
 , 
2005
, vol. 
25
 (pg. 
109
-
115
)
29
Rustin
P.
Chretien
D.
Bourgeron
T.
Gérard
B.
Rötig
A.
Saudubray
J.M.
Munnich
A.
Biochemical and molecular investigations in respiratory chain deficiencies
Clin. Chim. Acta.
 , 
1994
, vol. 
228
 (pg. 
35
-
51
)
30
Srere
P.A.
Citrate synthase
Method. Enzymol.
 , 
1969
, vol. 
13
 (pg. 
3
-
11
)
31
Tada
K.
Kure
S.
Non-ketotic hyperglycinaemia: molecular lesion, diagnosis and pathophysiology
J. Inherit. Metab. Dis.
 , 
1993
, vol. 
16
 (pg. 
691
-
703
)
32
Wittig
I.
Braun
H.P.
Schägger
H.
Blue native PAGE
Nat. Protoc.
 , 
2006
, vol. 
1
 (pg. 
18
-
28
)

Supplementary data