A novel mutation in the SLC25A15 gene in a Turkish patient with HHH syndrome: functional analysis of the mutant protein.

The hyperornithinemia-hyperammonemia-homocitrullinuria syndrome is a rare autosomal recessive disorder caused by the functional deficiency of the mitochondrial ornithine transporter 1 (ORC1). ORC1 is encoded by the SLC25A15 gene and catalyzes the transport of cytosolic ornithine into mitochondria in exchange for citrulline. Although the age of onset and the severity of the symptoms vary widely, the disease usually manifests in early infancy. The typical clinical features include protein intolerance, lethargy, episodic confusion, cerebellar ataxia, seizures and mental retardation. In this study, we identified a novel p.Ala15Val (c.44C>T) mutation by genomic DNA sequencing in a Turkish child presenting severe tantrum, confusion, gait disturbances and loss of speech abilities in addition to hyperornithinemia, hyperammonemia and homocitrullinuria. One hundred Turkish control chromosomes did not possess this variant. The functional effect of the novel mutation was assessed by both complementation of the yeast ORT1 null mutant and transport assays. Our study demonstrates that the A15V mutation dramatically interferes with the transport properties of ORC1 since it was shown to inhibit ornithine transport nearly completely.

Introduction

The hyperornithinemia–hyperammonemia–homocitrullinuria (HHH) syndrome (OMIM#238970), also called ornithine carrier deficiency, is a rare autosomal recessive urea cycle disorder [7,38]. Age of onset and severity of the symptoms vary widely. The disease usually manifests in early infancy or childhood; however, cases of adult onset have also been reported [13,36]. The typical clinical features of the syndrome include lethargy, episodic confusion or coma due to postprandial hyperammonemia, protein intolerance, vomiting, spastic paraplegia, cerebellar ataxia, seizures, mental retardation and fulminant liver failure [7,38]. The great majority of affected patients can be treated with pharmacological and dietary interventions. HHH syndrome is caused by the functional deficiency of the 301 amino acid long mitochondrial ornithine carrier 1 (ORC1), a transmembrane solute carrier protein encoded by the SLC25A15 (solute carrier family 25, member 15) gene (OMIM#603861) located on chromosome 13q14 [3,25]. Although a founder effect was reported in the French-Canadian population [3], there exist many allelic variants worldwide [5,16,20,34,36,37] pointing out the diversity of mutations and panethnic distribution of the disease. ORC1 catalyzes the transport of cytosolic ornithine into the mitochondrial matrix in exchange for mitochondrial citrulline and hence plays a crucial role in the urea cycle, ammonium detoxification and arginine synthesis [9,23]. The impaired transport of ornithine causes its accumulation in the cytosol and disruption of the urea cycle, leading to hyperornithinemia and hyperammonemia. In addition, ornithine deficit in the mitochondria allows the reaction of carbamoyl phosphate with lysine causing homocitrullinemia (Fig. 1). The genotype–phenotype correlation in HHH syndrome is extremely weak [4,16,20,34,36,37]. This fact is partially due to a gene redundancy effect resulting from the existence of a second mitochondrial ornithine carrier (ORC2) with 87% amino acid identity to ORC1 and similar function [4,9]. In addition to a possible rescue mechanism by ORC2, the weak genotype–phenotype correlation may be caused by varying effects of mutations on the protein function.

Fig. 1

The key role of the mitochondrial ornithine carrier in the urea cycle. Abbreviations: ARG1, arginase 1; ASL, argininosuccinate lyase; ASS, argininosuccinate synthase; CPS1, carbamoyl phosphate synthase 1; ORC1, mitochondrial ornithine carrier isoform 1; OTC, ornithine transcarbamoylase.

In this study, we identified a novel c.44C > T (p.A15V) mutation in the SLC25A15 gene in a 9-year-old Turkish patient clinically diagnosed with HHH syndrome. This mutation was found to be deleterious to protein function as assessed by two different approaches: complementation of the yeast ORT1 null mutant and direct measurement of the transport activity.

Materials and methods

Subjects

A 9-year-old male Turkish patient was born at term to a healthy 31-year-old mother by spontaneous vaginal delivery with a birth weight of 4400 g (9.12 pounds) and no history of perinatal complications. He started to experience speech problems when he was 3.5 years old and was hospitalized at the age of 6 years upon his first attack with severe tantrum, confusion, gait disturbances and loss of speech abilities. Upon physical examination upward paralysis, aphasia, bilateral flask paralysis in the upper and lower extremities and deep tendon reflexes were noticed. Cranial CT, MRI, EMG and fundoscopic findings were normal. The patient appeared to have an intolerance for protein-rich food and presented hyperornithinemia (380 μmol/L in plasma; N: 40–160 μmol/L), hyperammonemia (300 μmol/L in plasma; N: < 50 μmol/L) and homocitrullinuria (385 μmol/mmol creatinine, N: 5.3–175 μmol/mmol). The patient’s family history revealed that his older brother developed similar clinical symptoms at the age of 5 years and died at the age of 18 years as a result of fulminant hepatic failure. Based on the characteristic metabolic triad and family history, the patient was clinically diagnosed with HHH syndrome. For the past 3 years, he has shown remarkable clinical and biochemical improvement, without any complications, after being placed on a protein-restricted diet (1 g/kg per day) supplemented with citrulline (100 mg/kg/day) and sodium benzoate (250 mg/kg/ day).

Molecular analysis

Blood samples were collected upon signed informed consent of the individuals participating in the study; sample collection and molecular analyses were performed in accordance with the Haliç University Human Research Ethics Board protocols. Following genomic DNA extraction from blood samples of the patient and his parents (High Pure PCR Template kit, ROCHE, Germany), all the 7 exons of the SLC25A15 gene (NG_012248.1) were amplified under the optimized PCR conditions using 13 intronic primer pairs designed by utilizing PRIMER3 software. The last exon (2914 bp) was amplified with 7 overlapping primers (Supplementary material). Due to the existence of an SLC25A15 pseudogene on the Y chromosome, designed primers were cross-checked for specificity using Primer Blast. The amplicons were purified (High Pure PCR Product Purification Kit, ROCHE, Germany) and sequenced by Iontek Inc. (Istanbul, Turkey) using BigDye 3.1 (Applied Biosystems, Foster City, USA). One hundred healthy control chromosomes of Turkish origin were also sequenced.

Expression of ORC1 and A15V ORC1 in Escherichia coli and Saccharomyces cerevisiae

The coding sequence of the SLC25A15 gene was amplified by PCR from human liver cDNA. Forward and reverse oligonucleotide primers were synthesized corresponding to the extremities of the coding sequence (Accession No. NM_014252) with additional NdeI and HindIII restriction sites as linkers. The Ala15Val mutation was introduced into the wild-type ORC1 cDNA by overlap-extension PCR [14]. The amplified products were cloned into the pMW7 vector for expression in E. coli. The ORC1-pYES2 and A15V ORC1-pYES2 plasmids were constructed by cloning the coding sequences of wild-type and mutant ORC1 into the yeast pYES2 expression vector (Invitrogen) under the control of the GAL10 promoter [19]. In this case, the reverse primer contained the hemagglutinin (HA) tag at the 3′ terminus of the sequences before the stop codon. All the constructs were transformed into E. coli TG1 cells [22] (Invitrogen). Transformants were selected as described [19] and the sequences of inserts were verified.

BY4742 (wild-type) and ORT1∆ S. cerevisiae strains were provided by the EUROFAN resource center EUROSCARF. The ORT1∆ strain, transformed with ORC1-pYES2 and A15V ORC1-pYES2 as described [19], and the wild-type strain were grown in 2% ethanol-supplemented SC medium [35] until the early exponential phase was reached; then 0.4% galactose was added for 5 h to induce the expression of wild-type or A15V ORC1. The amounts of ORC1, A15V ORC1 and porin in yeast mitochondria, isolated by standard procedures, were determined by immunoblotting, as described previously [31].

The overexpression of wild-type and A15V mutant ORC1 as inclusion bodies in the cytosol of E. coli C0214 (DE3) was accomplished as described [8]. Control cultures with the empty vector were processed in parallel. Inclusion bodies were purified on a sucrose density gradient and the bacterially expressed proteins were purified as described in Agrimi et al. [1] and solubilized in 2.2% (w/v) sarkosyl (sodium N-laurylsarcosinate). The amount of purified wild-type and A15V ORC1 was estimated by laser densitometry of Coomassie Blue-stained SDS–PAGE gels, using carbonic anhydrase as protein standard.

Reconstitution of the bacterially expressed wild-type and A15V mutant ORC1 into liposomes and transport assays

The solubilized recombinant proteins were reconstituted into liposomes by cyclic removal of the detergent with a hydrophobic column of Amberlite beads (Fluka) [9,28]. The initial mixture used for reconstitution was prepared by adding its components in the following order: solubilized purified protein (about 5 μg), 1% Triton X-114, 1% egg yolk phospholipids in the form of sonicated liposomes [2], 20 mM l-ornithine, 10 mM HEPES (pH 7.2), 1.0 mg of cardiolipin and water to a final volume of 700 μl. After vortexing, this mixture was recycled 13 times through the Amberlite column (3.5 × 0.5 cm) pre-equilibrated with a buffer containing 10 mM HEPES, pH 7.2, and 10 mM NaCl. The amount of protein incorporated into liposomes was measured as previously described [18]. With both the wild-type and A15V mutant ORC1, the fraction of protein successfully reconstituted was about 20% of the protein added to the reconstitution mixture in all experiments.

External substrate was removed from proteoliposomes on a Sephadex G-75 column, pre-equilibrated with 10 mM HEPES and 50 mM NaCl at pH 7.2. Transport at 25 °C was initiated by adding l-[3H]ornithine (American Radiolabeled Chemicals) to l-ornithine-loaded proteoliposomes. Transport was terminated, at the indicated times, by the addition of 15 mM pyridoxal 5′-phosphate and 18 mM bathophenanthroline, which in combination and at high concentrations completely inhibit the activity of several mitochondrial carriers [12,39]. In controls, the inhibitors were added at the beginning together with the radioactive substrate (the “inhibitor stop” method [26]). Finally, the external radioactivity was removed by a Sephadex G-75 column, and the radioactivity in the proteoliposomes was measured [11]. The experimental values were corrected by subtracting control values.

Results

Molecular identification of the mutation

The molecular analysis of the patient’s genomic DNA revealed that he is homozygous for a novel mutation. The identified mutation is a C > T transition at position 44 on ORC1 coding DNA sequence, corresponding to an alanine to valine substitution at the 15th position on the ORC1 protein (p.A15 > V). This new variant was not present in the Human Gene Mutation Database (www.hgmd.org), the LOVD database (http://www.lovd.nl) or the EVS database (http://evs.gs.washington.edu). The identified mutation has been submitted to the LOVD database. The healthy parents of the patient proved to be heterozygous for the same mutation (Fig. 2). The newly identified variant was not detected in 100 Turkish control chromosomes.

Fig. 2

The mother (A) and father (B) are both heterozygous (C/T) for the c.44C > T mutation, and the patient is homozygous (T/T). Arrows indicate the position of the mutation.

Functional analysis of the A15V mutant ORC1

The conservation of the altered alanine residue across species suggests that its mutation might disrupt protein function (Fig. 3). To assess the pathogenic potential of the c.44C > T mutation, two completely different approaches were used. The first approach, which to our knowledge has not yet been applied to putative HHH syndrome-causing mutations, is based on the observations that ORC1 is the closest human relative of the S. cerevisiae Ort1p [29,30], and that the S. cerevisiae ORT1 null mutant does not grow on SC medium without arginine, a phenotype explained by its inability to transport ornithine across the mitochondrial membrane [6,30]. We therefore set out to investigate whether complementation of the ORT1∆ yeast strain with wild-type or A15V mutant ORC1 could mitigate or abolish the growth defect of the ORT1 knockout in the absence of arginine. ORC1 expressed in ORT1∆ cells via the multicopy yeast vector pYES2 restored growth of the ORT1∆ strain on arginine-less SC medium almost to wild-type levels (Fig. 4B). In contrast, the growth phenotype of ORT1∆ cells in the absence of arginine was not restored at all by complementing the knockout strain with the A15V mutant ORC1 in the same vector (Fig. 4B), indicating that A15V mutant ORC1 is unable to transport ornithine across the mitochondrial membrane. It is worth mentioning that immunodecoration of mitochondria isolated from ORT1∆ cells transformed with wild-type or A15V mutant ORC1 revealed that the two heterologously expressed proteins were present in similar amounts.

Fig. 3

Sequence alignment of ORC1 across species, showing amino acids 13–34. The arrow indicates the conserved alanine residue.

Fig. 4

Effect of complementing ORT1Δ cells with wild-type or A15V ORC1 on growth. (A) and (B) Four-fold serial dilutions of wild-type cells (WT), ORT1Δ cells and ORT1Δ cells transformed with the ORC1-pYES2 or A15V ORC1-pYES2 plasmid were plated on SC medium with (A) or without (B) arginine, supplemented with 2% ethanol and 0.4% galactose; the plates were incubated for 60 h at 30 °C. (C) Expression levels of ORC1, A15V ORC1 and porin in ORT1Δ cells. Mitochondria (30 μg of protein) were obtained from ORT1Δ cells transformed with the ORC1-pYES2 plasmid (WT) or the A15V ORC1-pYES2 plasmid (A15V). Mitochondrial proteins were separated by SDS–PAGE, transferred to nitrocellulose and immunodecorated with the anti-hemagglutinin (HA) or the anti-porin antibody.

The second approach to investigate whether a mutation affects protein function consisted in measuring the activity of the mutant protein by direct transport assays [9,36]. For this purpose, wild-type and mutant ORC1 were overexpressed in E. coli, purified and reconstituted into liposomes. We then followed the time course of the [3H]ornithine/ornithine exchange in proteoliposomes (Fig. 5). The wild-type protein catalyzed an efficient uptake of [3H]ornithine into proteoliposomes similar to that reported previously [9]. In contrast, negligible transport activity was detected with the A15V mutant ORC1, although the amount of protein inserted in the liposomal membrane was the same for both recombinant proteins. The inhibition of the initial transport rate of the [3H]ornithine/ornithine exchange caused by the mutation was 96.9 ± 2.0% in four experiments. Therefore, we conclude that the transport activity of the A15V mutant ORC1 was almost abolished.

Fig. 5

Transport assays of wild-type and A15V mutant ORC1. At time zero 1 mM [3H]ornithine was added to liposomes reconstituted with the recombinant wild-type (filled squares) or A15V mutant ORC1 (filled circles) and containing 20 mM ornithine. At the indicated times, the uptake of the labeled substrate was terminated by adding 15 mM pyridoxal 5′-phosphate and 18 mM bathophenanthroline. Similar results were obtained in four independent experiments.

Discussion

The diagnosis of HHH syndrome can be safely based on the presence of specific metabolic abnormalities. However, in some cases, clinical diagnosis becomes difficult due to slightly higher ornithine levels in the initial phases of the disease and to the fact that in the later stages hyperornithinemia can be regarded as secondary to hepatic failure. An additional complicating factor is the age of symptom onset, which may range from the neonatal period to adulthood. In this respect, molecular analysis of the SLC25A15 gene is to be considered as part of a rapid and definitive diagnosis. Identification of the disease at its early stages is extremely important, because hepatic failure can be prevented with basic treatment (e.g., protein-restricted diet, and supplementation with arginine and sodium benzoate). Apart from the French-Canadian founder mutation (p.F118del), no other common mutation in the ORC1 gene has been identified. Therefore, screening the whole coding sequence of the gene by direct sequencing is essential for the molecular analysis of HHH syndrome. Although there is no clear genotype–phenotype correlation, identification of novel mutations along with comprehensive clinical and biochemical findings may pave the way to explore potential pathogenetic molecular mechanisms.

In this study, we have identified a novel p.A15V mutation in a Turkish patient with childhood onset presenting distinct biochemical and clinical features of HHH syndrome. The parents of the patient were heterozygous for the same mutation. The pathogenicity of the novel A15V mutation was confirmed by using complementation of the ORC1∆ yeast strain and direct transport assays. Indeed, complementation with the A15V mutant ORC1 could not restore the yeast null mutant growth defect on arginine-free medium, and the transport assay, which measures the activity of the mutant protein, showed a dramatic inhibition of the transport activity of the A15V mutant ORC1. Until now, the function of HHH syndrome-causing mutations has been evaluated mainly by determining the ability of mutant ORC1 to complement the defect of HHH fibroblasts in incorporating radioactive ornithine into cellular protein via intramitochondrial conversion to glutamate and proline [3,5]. Compared to this method, yeast complementation analysis, which has been described in this study for the first time, has the advantage of avoiding the use of cultured skin fibroblasts from patients and of providing information on changes in intracellular localization and stability of ORC1 variants. Both methods are simple and capable of distinguishing between ORC1 mutations with normal or impaired function. However, direct assay of transport activity in reconstituted liposomes, which has already been applied to ORC1 mutations [9,36] as well as to mutations of other mitochondrial carriers [10,17,32,40], is more quantitative though somewhat more specialized, compared to the above-mentioned methods. Although it may seem surprising that a single amino acid replacement affects ORC1 function, this has been observed with pathogenic mutations in other proteins including mitochondrial carriers. For example, A281V SLC25A20 has been shown to be responsible for CAC deficiency [15], G177A SLC25A19 for congenital Amish microcephaly [24,33] and P206L SLC25A22 for neonatal epileptic encephalopathy with suppression bursts [21]. The substituted alanine residue of ORC1 discussed herein is located in the first transmembrane α-helix of the protein adjacent to a highly conserved glycine-rich region, which is also present in the symmetric third and fifth transmembrane α-helices. This region has been proposed to be involved in opening and closing the mitochondrial carriers on the cytosolic side [27]. It is likely therefore that the A15V mutation disturbs important conformational changes that take place in the catalytic cycle of the carrier to accomplish ornithine translocation across the inner mitochondrial membrane [25]. Although an in-depth understanding of the structure of ORC1 (or other mitochondrial carriers) in their different conformational states would be required to validate this hypothesis, the A15V mutation provides an excellent explanation for HHH syndrome in our patient.

Including the ORC1 mutation described in this study, which to our knowledge is the first report on HHH syndrome in the Turkish population, 33 mutations in the SLC25A15 gene causing this disorder have been identified until now. It is apparent that mutational screening of the SLC25A15 gene is crucial for an accurate diagnosis of the disease, for advancing knowledge of the worldwide SLC25A15 mutation spectrum and for a better understanding of the molecular basis of the disease, all of which represent essential information for establishing future therapeutic strategies.

Conflict of interest

The authors declare that they have no conflict of interest.