RESEARCH ARTICLE Human Mutation

OFFICIAL JOURNAL

Identification of Novel Mutations in the SLC25A15 Gene in

Hyperornithinemia-Hyperammonemia-Homocitrullinuria www.hgvs.org

(HHH) Syndrome: A Clinical, Molecular, and Functional

Study
Alessandra Tessa,1 Giuseppe Fiermonte,2 Carlo Dionisi-Vici,1 Eleonora Paradies,2 Matthias R. Baumgartner,3
Yin-Hsiu Chien,4 Carmela Loguercio,6 Helene Ogier de Baulny,7 Marie-Cecile Nassogne,8 Manuel Schiff,7
Federica Deodato,1 Giancarlo Parenti,5 S. Lane Rutledge,9 M. Antonia Vilaseca,10 Mariarosa A.B. Melone,6
Gioacchino Scarano,11 Luiz Aldamiz-Echevarrı́a,12 Guy Besley,13 John Walter,13 Eugenia Martinez-Hernandez,14
Jose M. Hernandez,15 Ciro L. Pierri,2 Ferdinando Palmieri,2  and Filippo M. Santorelli1
1
Molecular Medicine and Metabolism, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Children’s Hospital Bambino Gesù, Rome, Italy;
2
Department of Pharmaco-Biology, Laboratory of Biochemistry and Molecular Biology, University of Bari, Bari, Italy; 3Division of Metabolism,
Zurich University, Zurich, Switzerland; 4Medical Genetics, Taiwan University, Taipei, Taiwan; 5Pediatrics, Federico II University, Naples, Italy;
6
Internal Medicine and Hepatogastroenterology and Neurology, Second University of Naples, Naples, Italy; 7Division of Metabolism, R. Debre
Hospital, Paris, France; 8Pediatric Neurology, St-Luc Hospital, Bruxelles, Belgium; 9Department of Genetics, Alabama University, Birmingham,
Alabama; 10Biochemical Unit, Sant Joan de Deu Hospital, Barcelona, Spain; 11Medical Genetics, Azienda Ospedaliera Rummo, Benevento, Italy;
12 13
Laboratorio de Metabolismo, Hospital de Cruces, Bilbao, Spain; Willink Biochemical Genetics Unit, Royal Manchester Children’s Hospital,
14 15
Manchester, United Kingdom; Servei de Neurologia, Hospital de Sant Pau, Barcelona, Spain; Institut de Bioquı´mica Clı´nica, Corporació
Sanitària, Barcelona, Spain
Communicated by Ronald J. A. Wanders
Received 2 April 2008; accepted revised manuscript 24 September 2008.
Published online 25 February 2009 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/humu.20930

no clear-cut genotype–phenotype correlations. Although

ABSTRACT: Hyperornithinemia-hyperammonemia-homoci-
trullinuria (HHH) syndrome is an autosomal recessive
disorder of the urea cycle. With the exception of the French-
Canadian founder effect, no common mutation has been
detected in other populations. In this study, we collected 16
additional HHH cases and expanded the spectrum of
SLC25A15/ORC1 mutations. Eleven novel mutations were
identified including six new missense and one microrear-
rangement. We also measured the transport properties of the
recombinant purified proteins in reconstituted liposomes for
four new and two previously reported missense mutations
and proved that the transport activities of these mutant
forms of ORC1 were reduced as compared with the wild-
type protein; residual activity ranged between 4% and 19%.
Furthermore, we designed three-dimensional (3D)-modeling
of mutant ORC1 proteins. While modeling the changes in
silico allowed us to obtain new information on the
pathomechanisms underlying HHH syndrome, we found
Alessandra Tessa, Giuseppe Fiermonte, and Carlo Dionisi-Vici contributed equally
to this work.
Correspondence to: Ferdinando Palmieri, Department of Pharmaco-Biology,
Laboratory of Biochemistry and Molecular Biology, University of Bari, via Orabona, 4-
70125 Bari, Italy. E-mail: fpalm@farmbiol.uniba.it; and Filippo M. Santorelli, Molecular
Medicine, IRCCS Children’s Hospital Bambino Gesù, Piazza S. Onofrio, 4–00165 Rome,
Italy. E-mail: filippo3364@gmail.com
Contract grant sponsor: Ministero della Ricerca e dell’Università (MUR); Contract
grant sponsor: Italian National Research Council (CNR); Contract grant sponsor:
Istituto Superiore di Sanità; Pierfranco and Luisa Mariani Foundation ONLUS;
Contract grant sponsor: Telethon Foundation; Grant number: GGP06188.
patient metabolic alterations responded well to low-protein
therapy, predictions concerning the long-term evolution of
HHH syndrome remain uncertain. The preference for a
hepatic rather than a neurological presentation at onset also
continues, largely, to elude us. Neither modifications in
oxidative metabolism-related energy, such as those expected
in different mtDNA haplogroups, nor sequence variants in
SLC25A2/ORC2 seem to be crucial. Other factors, includ-
ing protein stability and function, and ORC1-ORC2
structural interactions should be further investigated.
Hum Mutat 30:741–748, 2009. & 2009 Wiley-Liss, Inc.
KEY WORDS: SLC25A15; ORC1; hyperornithinemia;
hyperammonemia; homocitrullinuria; HHH
Introduction
Hyperornithinemia-hyperammonemia-homocitrullinuria (HHH)
syndrome (MIM] 238970) is an autosomal recessive disorder of the
urea cycle [reviewed in Valle and Simell, 2001; Palmieri, 2008]. Since
its initial description in the late 1960s [Shih et al., 1969], fewer than
100 patients have been reported. The disease has a worldwide
distribution, although a founder effect has been observed in the
French-Canadian population from Quebec [Gatfield et al., 1975].
The period of onset of clinical symptoms in HHH syndrome
usually ranges from early infancy, including the neonatal period,
to childhood. However, descriptions of cases not diagnosed until
adulthood have been reported [Gatfield et al., 1975]. In the acute
phase, the disease is characterized by intermittent episodes of

& 2009 WILEY-LISS, INC.

hyperammonemia accompanied by vomiting, ataxia, lethargy,
confusion, and coma, as seen in other urea cycle defects [Valle and
Simell, 2001]. Remarkably, some HHH patients presented with
fulminant liver failure and were considered for liver transplanta-
tion [Fecarotta et al., 2006].
The chronic and progressive course of HHH includes an
aversion to protein-rich foods, coagulation abnormalities, hypo-
tonia, developmental delay, progressive encephalopathy with
mental regression, and early signs of motor dysfunction. Multiple
supratentorial stroke-like lesions have recently been reported [Al-
Hassnan et al., 2008]. The vast majority of patients respond well to
dietary and pharmacological therapy and the disorder is rarely
lethal: about 95% of patients survive after diagnosis and the
establishment of an appropriate dietary therapy. However, in early
adulthood most patients develop signs of pyramidal tract
dysfunction, often evolving into frank spastic paraparesis and
lower limb stiffness [Valle and Simell, 2001; Salvi et al., 2001b].
The SLC25A15 gene (MIM] 603861) was cloned in 1999 and a
‘‘common’’ mutation (p.F188del) in Québecois HHH patients was
identified [Camacho et al., 1999]. SLC25A15 encodes a 301–ami-
no-acid-long mitochondrial carrier protein, which catalyzes the
electroneutral exchange of ornithine for citrulline plus an H1 ion;
which is termed ORC1 [Fiermonte et al., 2003; Palmieri, 2004].
Dysfunction of ORC1 results in cytoplasmic ornithine accumula-
tion and reduced ability to synthesize citrulline, leading to
impaired ammonia detoxication and increased carbamoyl phos-
phate. Homocitrullinuria, a biochemical marker of the disease,
likely results from transcarbamoylation of lysine which, via ORC2,
enters the mitochondria by an ornithine/H1 exchange [Camacho
et al., 2003; Palmieri, 2004].
After the report of the first mutations in Québecois patients
[Camacho et al., 1999], the array of allelic variants associated with
the syndrome has expanded to Japan [Tsujino et al., 2000], Italy
[Salvi et al., 2001a], and elsewhere [Camacho et al., 2003, 2006;
Korman et al., 2004], demonstrating that HHH has a panethnic
distribution. With the exception of the French-Canadian founder
effect, no common mutation has been detected in other populations,
although the p.R179X variant is considered to be frequent in Japan
[Miyamoto et al., 2001]. Moreover, we and other investigators have
shown that drawing genotype–phenotype correlations is a somewhat
complex task [Tsujino et al., 2000; Miyamoto et al., 2001; Salvi et al.,
2001a; Korman et al., 2004; Camacho et al., 2006].
In this study, we investigated 16 additional HHH cases,
expanded the spectrum of SLC25A15 mutations, measured the
transport properties of the recombinant purified proteins in
reconstituted liposomes for new and two previously reported
missense mutations, and provided three-dimensional (3D)
modeling of mutant ORC1 proteins. Finally, we attempted to
correlate clinical, functional, and molecular features.
Patients and Methods
Patients and Sampling
Sixteen patients diagnosed with HHH in our centers were seen
at follow-up, at which time clinical, metabolic, and dynamic data
were collected again, adopting an approach used previously in an
earlier series of HHH patients [Salvi et al., 2001b]. After obtaining
approval from the ethics committees of our respective institutions,
we collected biological material from propositi in 13 families
(Families A–M), from 20 of their parents, and from six unaffected
siblings. Three families were of Italian ancestry (Families B, G, and
H), three were from Spain (Families K and M with Pakistani
742 HUMAN MUTATION, Vol. 30, No. 5, 741–748, 2009
ancestry, and Family E), two from Morocco (Families J and L),
and one each from Belgium (Patient A), Algeria (Patients C), the
United States (Patient I, of Greek origin), Taiwan (Patient F), and
France (Patient D, of African origin). All the patients had been
referred to us in the previous 2 years, presenting with the
characteristic metabolic triad (HHH). In one patient the
syndrome was diagnosed prospectively after genotyping an
affected sibling. Genomic DNA was extracted using standard
techniques from blood samples collected in EDTA after informed
consent had been obtained.
Molecular Genetic Analyses
The SLC25A15 gene was analyzed by PCR amplification and
direct sequencing using BigDye 3.1 chemistry (Applied Biosystems,
Foster City, CA) and intronic oligonucleotide primers, as described
[Salvi et al., 2001b]. The primers and PCR conditions for
amplification and sequencing of the human SLC25A2/ORC2 coding
sequence have been detailed elsewhere [Camacho et al., 2003].
The nomenclature of the mutations follows the format indicated
in consensus databases (www.hgvs.org/mutnomen) and refers to
the SLC25A15 cDNA sequence (GenBank reference sequence
NM_014252.2) in which we designated the A of the first methionine
as nucleotide11. The presence of the identified mutations was
confirmed by either PCR-restriction fragment-length polymorph-
ism (RFLP) analysis, or resequencing, or both. The control panel
for the identified SLC25A15 gene mutations consisted of 400 Italian,
120 Asian, and 200 North African chromosomes.
Cultured skin fibroblast polyA1RNA from three patients and
five controls was purified and reverse-transcribed using the First
Strand cDNA Synthesis Kit (Roche, Hamburg, Germany) accord-
ing to the manufacturer’s random primer protocol. For the
semiquantitative determination of SLC25A2/mRNA (ORC2;
MIM] 608157; GenBank reference sequence NM_031947.2), the
PCRs were performed in a total volume of 25 ml, each containing
200 ng of cDNA and 7.5 pmol of the specific primers ORC2-F1
(50 -CCACTGGGACCCTCAGACGA-30 ) and ORC2-R1 (50 -
CGGATCCAGGTTCACCAAGA-30 ).
The PCR conditions were set as follows: 35 cycles of 45 s at
941C, 45 s at 591C, and 45 s at 721C, followed by a final elongation
step at 721C for 7 min. For the amplification of the endogenous
gene (GAPDH) used as internal reference, we used primers F2 (50 -
CTTCCATCAGTCGGATACAC-30 ) and R2 (50 -TAGGAGGCTG-
CATCATCGT-30 ) and specific PCR conditions consisting of 35
cycles of 45 s at 941C, 45 s at 551C, and 45 s at 721C. Each sample
was run in triplicate, and results were reported as the mean values
of five independent experiments and referred to normal controls
whose expression was arbitrarily attributed the value of 1. The
results were analyzed using the Bio-Rad Versa Doc Imaging
System and Quantity One software (Bio-Rad, Hercules, CA) and
expressed as the ratio of signal intensity of the SLC25A2/mRNA
band to that of the GAPDH/mRNA band for each analyzed
sample. Student’s t-test was used for statistical analyses, setting the
level of significance at Po0.01.
The splicing effects of the intronic mutation in patients
were established using previously reported PCR conditions
and oligonucleotide primers [Salvi et al., 2001a]. Determination
of major mtDNA haplogroups in HHH patients was carried
out by specific endonuclease analyses as described elsewhere
[Torroni et al., 1996] using the revised Cambridge Sequence
as reference [Andrews et al., 1999]; mtDNA haplogroups
not classifiable in any specific haplogroup were classified as
‘‘Others’’ [Ruiz-Pesini et al., 2000].
Construction of Expression Plasmids
The coding sequence for ORC1 was amplified, as described
previously [Fiermonte et al., 2003]. The mutations p.M37R, p.L71Q,
p.G113C, p.T272I, p.M273K, and p.L283F were introduced into the
wild-type ORC1 cDNA by the overlap extension PCR method [Ho
et al., 1989] using oligonucleotides with appropriate mutations in
their sequences. The PCR products were cloned into the pRUN
expression vector [Fiermonte et al., 2003], and the constructs were
transformed into Escherichia coli TOP 10 cells (Invitrogen, Milan,
Italy). Transformants selected on 2  TY (16 g/L tryptone, 10 g/L
yeast extract, 5 g/L NaCl, pH 7.4) plates containing ampicillin
(100 mg/ml) were screened by direct colony PCR.
Bacterial Expression and Purification of ORC1 and
Mutated Forms of ORC1
ORC1 and its mutated forms were overexpressed as inclusion
bodies in the cytosol of E. coli, as previously described [Fiermonte
et al., 2003]. Control cultures with the empty vector were
processed in parallel. The inclusion bodies were purified by
sucrose-layer density-gradient centrifugation [Fiermonte et al.,
1993] and washed at 41C with TE buffer (10 mM Tris-HCl at pH
8.0, and 1 mM EDTA). They were then washed twice with a buffer
containing 10 mM piperazine-N,N0 -bis(2-ethanesulfonic acid)
(PIPES), at pH 7.0, 3% (w/v) Triton X-114, 1 mM EDTA,
20 mM Na2SO4, and finally with TE buffer again. The wild-type
and mutant ORC1 proteins were each solubilized in a buffer
containing 2.5% (w/v) sarkosyl, 1 mM EDTA, and 10 mM Tris-
HCl at pH 7.0. Residual material was removed by centrifugation
(258,000 g for 1 hr at 41C).
Reconstitution Into Liposomes and Transport
Measurements
Recombinant proteins were reconstituted into liposomes in the
presence of 20 mM ornithine, arginine, lysine, or citrulline
substrates, independently [Palmieri et al., 1995]. External substrate
was removed from proteoliposomes on columns of Sephadex G-75
(Sigma-Aldrich, St. Louis, MO), which were preequilibrated with
50 mM sucrose and 10 mM HEPES; at pH 7.2. Transport at 251C
was started by adding L-[3H]ornithine, L-[3H]lysine, L-[3H]arginine,
or L-[14C]citrulline (PerkinElmer Life Sciences, Boston, MA) to
substrate-loaded proteoliposomes and was terminated by adding
20 mM pyridoxal 50 -phosphate (the ‘‘inhibitor-stop’’ method)
[Palmieri et al., 1995]. In controls, the inhibitor was added at the
beginning, together with the radioactive substrate. All transport
measurements were carried out at the same internal and external pH
value of 7.2. Finally, the external substrate was removed and
radioactivity in the liposomes was measured [Palmieri et al., 1995].
The experimental values were corrected by subtracting control
values and the transport activities were calculated by taking into
account the efficiency of reconstitution (i.e., the yield of successfully
incorporated protein). The initial transport rate was calculated from
the radioactivity taken up by proteoliposomes in 2 min (in the initial
linear range of substrate uptake).
Other Methods
Proteins were separated by SDS-PAGE and stained with
Coomassie Blue dye. The amount of recombinant protein was
estimated from Coomassie Blue-stained SDS-PAGE gels with the
Bio-Rad GS-700 Imaging Densitometer (Bio-Rad Laboratories, San
Francisco, CA) using carbonic anhydrase as the standard [Fiermonte
et al., 1998]. The extent of incorporation of the recombinant
proteins into liposomes was determined by centrifugation for 30 min
at (3  105)g (90,000 rpm) at 41C of 30 ml of the transport-active
proteoliposomes, prepared as described above, diluted with TE
buffer up to 500 ml. The pellet containing proteoliposomes was
delipidated; the lipid-free proteins were subsequently separated by
SDS-PAGE and quantified by densitometry [Cappello et al., 2007].
This value (i.e., the share of successfully incorporated protein)
ranged from 24 to 32% of the protein added to the reconstitution
mixture for both the wild-type and mutated forms of ORC1.
The homology model of ORC1 was built by comparison with
the crystallographic structure of bovine AAC1 [Pebay-Peyroula
et al., 2003] using the computer application MODELLER
(www.salilab.org/modeller). Crucial to the accuracy of compara-
tive modeling is the quality of the pairwise alignment, in this case
between the ORC1 and AAC1 template sequences, which was
performed using the PSIPRED protein structure prediction server
(www.psipred.net/psiform.html). The secondary structure of
AAC1 was used to weight the gap penalties, as described
previously [Cappello et al., 2006]. One thousand energy mini-
mization steps were performed to generate 10 minimized models
for the wild-type ORC1 and each of the six ORC1 mutants. The
structural properties of the ORC1 models with the best energy
function were evaluated using protein analysis tools available on
the WHAT IF Web server [Vriend, 1990]. Final models were
examined with PyMOL (www.pymol.org), VMD (www.ks.uiuc.
edu/research/vmd/) and Swiss PDB Viewer (spdbv.vital-it.ch) and,
where side-chain packing led to clashes, alternative side-chain
rotamers were evaluated.
Results
Table 1 lists the main clinical and molecular features of the 16
new HHH patients in 13 families at follow-up. Serum levels of
ornithine and ammonium as well as of urinary homocitrulline
were, by definition, elevated in all except for Patient 15. In the
latter case only DNA was available and a prospective diagnosis was
performed after genotyping his brother (Patient 14) who presents
the metabolic triad of the HHH syndrome.
Figure 1A shows the morbidity map of the SLC25A15 gene
mutations, including those identified in this study. Molecular
analyses identified 11 novel conventional mutations and two
already-described variants in a total of 31 HHH alleles. In one
patient we detected an unclear rearrangement (c.6211578_??del/
1578insAA) in intron 5 in sequences potentially affecting splicing,
but this remained untested. Messenger RNA studies in cultured
primary cells showed that the 5611G4T mutation resulted in
altered splicing with skipping of exon 3 (not shown).
In 15 of the 16 patients investigated, no mutations were found
screening the entire sequence of SLC25A2/ORC2, a potential
modifier of the clinical phenotype [Camacho et al., 2003;
Fiermonte et al., 2003]. We did not analyze Patient 15, the
brother of Patient 14. Also, semiquantitative RT-PCR showed that
levels of the ORC2/ORC1 mRNA ratio did not significantly differ
in three HHH patients (genotypes p.R179X, p.G27R, and
p.R275Q) [Salvi et al., 2001b] when compared to normal, age-
matched controls after correction for GAPDH levels (Fig. 1B).
mtDNA haplogroup reconstruction revealed that 12 patients
(80%) harbored the major European haplogroup H without
evident correlation with clinical and metabolic features. Less
common were the haplogroups T (13%) and ‘‘Others’’ (7%).
To gain insight into the functional consequences of HHH
mutations, four missense variants (namely, p.M37R, p.L71Q,
HUMAN MUTATION, Vol. 30, No. 5, 741–748, 2009 743
744
HUMAN MUTATION, Vol. 30, No. 5, 741–748, 2009

Table 1. Clinical and Molecular Features in New Hyperornithinemia-Hyperammonemia-Homocitrullinuria (HHH) Patients

Patient/ Family/ Lethargy Seizures/ Pyramidal Liver Mutation Mutation

Sex AF Origin A.O. coma myoclonus signs MR signs Coagulopathy Ammonia Ornithine Homocitrulline Orotate AST/ALT (cDNA level)a (protein level)

1/M 6y A/Belgium Infancy No No No Mild Yes Yes 137 951 1157 Increased 172/235 c.110T4G p.M37R
2/F 2 mz B/Italy Neonatal Yes No na na No No na na na na na p.212T4A p.L71Q
3/M 5y C/Algeria Infancy No No Yes Yes No No 370 822 10 4.03 25/45 p.733A4T p.K245X
4/M 3 C/Algeria pr.d (2 m) No No Yes Yes No No 125 887 12 207 61/54 p.733A4T p.K245X
5/F 6y D/Senegal Neonatal Yes Yes Yes Yes No Yes 700 509 Increased 109 66/40 c.535C4T p.R179X
6/F 6y D/Senegal Neonatal Yes No Yes Mild No No 100 290 Increased 1.1 na c.535C4T p.R179X
7/M 2 y E/Spain Infancy na na na Yes na na 75 419 1654 11.7 na c.646G4A p.G216S
8/F 2y F/Taiwan Infancy Yes na na na Yes (11) Yes (11) 321 450 Increased 4.7 2112/2331 c.815C4T p.T272I
9/M 41 y G/Italy Childhood Yes No Yes No No Yes 235 216 na 0.9 25/99 c.79G4A p.G27R
10/M 54 y H/Italy Adulthood Yes No Yes No Yes (11) Yes 222 537 Increased 12 1315/1930 c.5611G4T/ Frameshift
intronic change
11/F 1m I/U.S.-Greece Neonatal Yes na na na no Yes 400 370 na 78 na c.525insC/c.847C4T p.S175fsX192/
p.L283F
12/M 1.5 y J/Morocco Infancy Yes No Yes Yes Yes (11) Yes (11) 200 700 139 600 na c.208_209delCAinsTT p.A70L
13/F 3.8 y K/Pakistan Infancy No No No Yes Mild No 62 471 Increased Increased Increased c.564C4G p.F188L
14/M 2y L/Morocco Infancy Yes No No No Yes (11) Yes 96 493 Increased 75 na/880 c.535C4T p.R179X
15/M 4y L/Morocco pr.d (6 y) na na na na na na na na na na na c.535C4T p.R179X
16/F 57 y M/Pakistan Adulthood Yes Yes Yes No No No 306 na na na na c.208_208delCAinsTT p.A70L
a
GenBank reference sequence NM_014252.2.
M, male; F, female; AF, age at follow-up; AO, age at onset; m, months; y, years; z, age at death; 11, severe; pr.d, prospective diagnosis; na, not available; MR, mental retardation. AST, aspartate transaminase; ALT, alanine transaminase.
p.T272I, and p.L283F) detected in new patients were introduced
in the pRUN expression vector and the corresponding mutant
proteins were overexpressed in E. coli, purified, and compared to
wild-type ORC1. We also functionally analyzed p.G113C and
p.M273K, which we had reported previously in an atypical HHH
patient [Fecarotta et al., 2006]. We chose not to test the potential
missense p.A70L and the p.G216S, which were unlikely to provide
new information, or mutation p.F188L. The latter concerns the
same residue that is affected in the ‘‘common’’ Québecois
mutation and whose functional relevance has already been
established [Fiermonte et al., 2003].
Mutant proteins were not detected either in bacteria harvested
immediately before induction of expression, as shown for the
wild-type ORC1 (lane 2, Fig. 2) or in control (with only vectors)
cells harvested after induction (lane 3, Fig. 2). All of the purified
ORC1 mutant proteins as well as the wild-type protein displayed
single bands on SDS-PAGE with an apparent molecular mass of
30 kDa (lanes 5–11, Fig. 2). The identities of all proteins were
confirmed by N-terminal sequencing.
Equivalent amounts of purified recombinant proteins were
reconstituted into liposomes, and their ability to transport the
known substrates of ORC1 (ornithine, arginine, lysine, and
citrulline) was tested in homo-exchange experiments (i.e., with
the same substrate inside and outside the phospholipid vesicles).
Figure 3 shows that the wild-type protein efficiently transported
ornithine, arginine, lysine, and citrulline, whereas the mutant
proteins exhibited very low transport activities despite normal
insertion in the liposomal membrane. It is worth pointing out that
the insertion of the recombinant proteins into liposomes is
scrambled [Fiermonte et al., 1998, 2003]. In our experimental
conditions, ‘‘normal’’ means that the percentage of incorporation
of the mutant proteins into liposomes is comparable to that of the
wild-type protein. In particular, p.M37R was virtually incapable of
catalyzing any ornithine, arginine, lysine, or citrulline homo-
exchange. All the transport activities of the other five mutant
forms of ORC1 (namely, p.L71Q, p.G113C, p.T272I, p.M273K,
and p.L283F) were markedly reduced as compared with the wild-
type protein (residual activity, range 4–19%).
ORC1 is a member of the mitochondrial carrier protein family
[Palmieri, 2004], the prototype of which is the ADP/ATP carrier
whose crystal structure is available [Pebay-Peyroula et al., 2003]. We
therefore built a comparative model of ORC1 (Fig. 4) based on the
3D structure of the carboxyatractyloside-ADP/ATP carrier complex.
This structure consists of a barrel of six transmembrane a-helices
(H1–H6), surrounding a cavity open toward the cytosol, and three
short a-helices (h12, h34, and h56) parallel to the membrane plane on
the matrix side. The prolines of the signature motifs (PFDTMK,
PTELVK, and PVDCIK, at positions 29–34, 126–131, and 229–234,

Figure 1. A: Morbidity map of the SLC25A15 gene. The map illustrates the exon-intron structure of SLC25A15; arrows indicate the HHH-
associated variants. Novel mutations identified in this study are in bold. B: Semiquantitative determination of SLC25A2/mRNA in cultured skin
fibroblasts from three HHH patients and four age-matched controls. Genotypes of HHH cells (p.R179X, p.G27R, and p.R275Q) have previously
been reported [Salvi et al., 2001a]. Size and level of SLC25A2/ORC2 are not influenced by mutations in the SLC25A15 gene. The values in the
histograms are means7SD of at least three separate experiments carried out in triplicate. [Color figure can be viewed in the online issue, which
is available at www.interscience.wiley.com.]

HUMAN MUTATION, Vol. 30, No. 5, 741–748, 2009 745

Figure 2. Overexpression in Escherichia coli and purification of
ORC1 and ORC1 mutants. Proteins were separated by SDS-PAGE and
stained with Coomassie Blue dye. Lane M, markers (bovine serum
albumin, carbonic anhydrase, and cytochrome c); lanes 1–4,
Escherichia coli C0214 (DE3) containing the expression vector without
(lanes 1 and 3) and with (lanes 2 and 4) the coding sequence of ORC1.
Samples were taken at the time of induction (lanes 1 and 2) and 5 hr
later (lanes 3 and 4). The same number of bacteria was analyzed in
each sample. Lane 5, ORC1 purified from bacteria in lane 4; lanes
6–11, recombinant ORC1 mutants (p.M37R, p.L71Q, p.G113C, p.T272I,
p.M273K, and p.L283F, respectively) purified as was the wild-type
protein reported in lane 5.
respectively) sharply kink the odd-numbered transmembrane a-
helices, and the charged residues of the motifs form a salt bridge
network (D31-K131, K34-D231, and K234-E128) that closes the cavity on
the matrix side of the protein. All the mutations analyzed in the
reconstituted system are shown in the structural model of ORC1
(Fig. 4) and are located in the transmembrane a-helices. Specifically,
p.M37R is found in H1, p.L71Q in H2, and p.G113C in H3, whereas
p.T272I, p.M273K, and p.L283F are located in H6.
Discussion
Our study corroborates the view that whole-gene sequencing
remains the only correct approach to obtain a full molecular
identification in HHH patients, unless they are of French-
Canadian origin [Gatfield et al., 1975; Camacho et al., 1999].
Complete molecular analysis is essential for a definite diagnosis,
although determination of the functional consequences of mutant
alleles and biochemical and molecular analyses of potential genetic
modifiers might help in further understanding of the pathome-
chanisms. This goal is crucial for the development of future
therapeutic strategies, particularly for the prevention of long-term
motor disability [Valle and Simell, 2001; Salvi et al., 2001b].
In 13 new HHH families (with a total of 16 index cases) we
identified 13 mutations, 11 of which are novel. These findings add to
the growing list of SLC25A15 variants. The following newly
identified mutations, p.K245X, c.5611G4T, p.S175fsX192, c.6211
578_??del/1578insAA, are clearly pathogenic since they predict early
translation termination or frameshift of the coding sequence. The
new homozygous microrearrangement, c.208_209delGC/insTT, may
either predict a missense variant (p.A70L) or alter the correct
splicing as determined by in silico analyses (www.rulai.cshl.edu/cgi-
bin/tools/ESE/esefinder.cgi,rulai.cshl.edu/new_alt_exon_db2/
HTML/score.html), but this was not tested experimentally. The six
missense mutations, p. M37R, p.L71Q, p.F188L, p.G216S, p.T272I,
and p.L283F, concern residues that are highly conserved in ORC1 of
746 HUMAN MUTATION, Vol. 30, No. 5, 741–748, 2009
Figure 3. Transport assays of wild-type and mutant ORC1 proteins.
Recombinant, reconstituted wild-type, and mutant ORC1 were
assayed for their ability to catalyze the homo-exchanges [3H]or-
nithine/ornithine (black bars), [3H]arginine/arginine (white bars),
[3H]lysine/lysine (gray bars), and [14C]citrulline/citrulline (hatched
bars). Proteoliposomes were preloaded internally with 20 mM
substrate, and transport was started by adding 0.2 mM [3H]ornithine,
1.5 mM [3H]arginine, 0.8 mM [3H]lysine, and 2.5 mM [14C]citrulline,
respectively. The reaction time was 2 min. The values are means7 SD
of at least three separate experiments carried out in duplicate.
different species, cosegregate with the disease in all the families
investigated, and were not detected in a large panel of ethnically-
matched control chromosomes. In addition, four of these mutations
are demonstrated here to be disease-causing since they are
deleterious for ORC1 function in liposomes reconstituted with the
recombinant mutant proteins (Fig. 3).
In the patients analyzed, there is no direct correlation between
the site and type of mutation, predicted length of the mutant
protein, and clinical features. For example, the potential p.A70L
(c.208_209delGC/insTT) mutation in two unrelated cases
presented as a severe hepatic phenotype with onset in infancy in
Patient 12 (Family J) and as a later-onset neurological disease in
Patient 16 (Family M). Interestingly, both Patients 10 and 16, who
are now ages 54 and 57 years, respectively, presented the first signs
of HHH syndrome around the age of 20 years and did not receive
a metabolic diagnosis until they were in their 50 s. Moreover, we
identified the homozygous p.R179X mutation in a metabolically
and clinically healthy member (Patient 15) in family L, although
the elder sibling (Patient 14) presented with severe liver
abnormalities. Nevertheless, careful follow-up in this asympto-
matic sibling is necessary to detect clinical manifestations of the
disease. It is worthwhile to note that the same mutation has
previously been associated with progressive neurological impair-
ment in an Italian patient [Salvi et al., 2001b] and with hypotonia
and lethargy in Japanese children [Tsujino et al., 2000]. Clotting
problems were observed in 8 of the 14 index cases (57%) for
whom this observation was obtained. Although it might be related
to transient liver dysfunction, in most cases coagulopathy
persisted with specific abnormalities of coagulation factors
[Dionisi-Vici et al., 1987]. However, the association of coagulo-
pathy with HHH remains to be clarified.
The mitochondrial ADP/ATP carrier and the other members of
the family are highly homologous and must have the same overall
structure [Palmieri, 2008]. Therefore, homology models of
mitochondrial carriers have been built for several members of the
family [Tonazzi et al., 2005; Morozzo della Rocca et al., 2005;
Cappello et al., 2006; Robinson and Kunji, 2006; Palmieri, 2008]. In
Figure 4. Missense mutations in patients with HHH syndrome. A: Lateral view of the structural-comparative model of human ORC1 (cartoon
representation). B: A view of the six amino acid substitutions from the cytoplasmic side. In (A) and (B), the positions of the six new missense
mutations indicated by their number in the ORC1 sequence (in stick representation) are shown; the transmembrane helices are colored as
follows: H1, red; H2, light brown; H3, green; H4, emerald-green; H5, cyan; and H6, blue; and purple surfaces highlight the salt bridge network
between residues K34 and D231, K234 and E128, and K131 and D31.

an attempt to understand the cause of the disease at the molecular
level and to obtain clues regarding correlation with the clinical
features, the missense mutations shown in this work to cause carrier
dysfunction are seen in light of the comparative model of ORC1
based on the available crystallographic structure of the ADP/ATP
carrier [Pebay-Peyroula et al., 2003]. In the mutant p.M37R
detected in Patient 1, R37 interacts with D31 and K34, which are
involved in the network salt bridges D31–K131 and K34–D231, or with
D42 located in the matrix loop connecting H1 and H2 (Fig. 4).
Consequently, R37 may cause loss of ORC1 transport activity either
because it interferes with the formation of important salt bridges or
because it stiffens the protein structure near D42. A different
mechanism explains the pathogenic role of the p.L71Q variant.
Residue L71 is oriented toward the membrane lipids. Generally,
amino acids interacting with the membrane bilayer tolerate
substitution quite well, as shown by cysteine-scanning mutagenesis
of the oxoglutarate carrier [Cappello et al., 2006, 2007]. However,
p.L71Q substitution may disturb the H2 package because of its
membrane-facing position and hydrophilic nature [Kyngäs and
Valjakka, 1998]. A plausible explanation for the dysfunction of
p.G113C is that inter–a-helical movements, which must occur in
the catalytic cycle of the carrier, are hindered by the bulkier side
chain of cysteine. A steric impediment might also explain the effects
of the p.T272I mutation. It is likely that at position 272 isoleucine
interferes with the formation of crucial bonds due to steric
hindrance and its hydrophobic nature. A similar effect may be
produced by the mutation p.M273K since M237 also interacts with
F30 and its substitution with a positively charged residue may
disturb the D31-K131 salt bridge. Finally, regarding the p.L283F
detected in Patient 11, in the structural model of ORC1 (Fig. 4) L283
is located near the C-terminus of H6 on the membrane’s cytosolic
side. It has been suggested that residues at the H6 C-terminus play a
role in early recognition of the substrate in both the oxoglutarate
carrier and carnitine/acylcarnitine carrier [Morozzo della Rocca
et al., 2005; De Lucas et al., 2008]. Thus, the mutation may impede
substrate recognition by inducing a stronger interaction between
H6 and H1 and/or a distortion of H6 in this region. Comparison of
the above-mentioned effects of ORC1 mutations on transport
activity with the clinical phenotype observed in HHH patients does
not allow at present to substantiate the claim that the location of
the defect within the protein might explain the onset and severity of
the clinical features. Also, it is hard to imagine compounds with the
potential of correcting the carrier structural changes in HHH
patients. Similarly, the prospective role of modifiers remains largely
unclear. Neither modifications in oxidative metabolism related
energy, such as those expected in different mtDNA haplogroups
[Ruiz-Pesini et al., 2000], nor sequence variants in ORC2, seem to
play an important role, as previously suggested [Camacho et al.,
2003]. From this perspective, other factors, including protein
stability and function, and ORC1–ORC2 structural interactions
should be further investigated.
In summary, we have presented clinical, molecular, and
functional data on novel variants observed in 16 HHH patients.
While modeling the mutations in silico allowed us to obtain new
information on the mechanisms underlying HHH syndrome, we
found no clear-cut genotype–phenotype correlation. Although the
metabolic alterations of these patients respond well to low-protein
therapy, we remain less able to predict the long-term evolution of
HHH syndrome. Also, the preference for a hepatic rather than a
neurological presentation at onset continues largely to evade our
understanding. As proposed in argininemia disorder, another urea
cycle defect characterized by pyramidal signs [Marescau et al.,
1990], we cannot exclude that some uncommon biochemical
abnormalities, such as those related to polyamine metabolism,
might be involved in progressive lower limb dysfunction in HHH
syndrome [Shimizu et al., 1990]. Conversely, we are still unable to

HUMAN MUTATION, Vol. 30, No. 5, 741–748, 2009 747

offer any plausible explanation for the fulminant, hepatitis-like
presentation observed in a subset of SLC25A15 variants.
Acknowledgments
We thank Catherine J. Wrenn for her expert editorial assistance. This work
was supported in part by grants from the Ministero della Ricerca e
dell’Università (MUR) (to G.F. and F.P.); Italian National Research Council
(CNR) (to F.P.); Istituto Superiore di Sanità (to F.P. and F.M.S.); and the
Pierfranco and Luisa Mariani Foundation ONLUS (to F.M.S.).
References
Al-Hassnan ZN, Rashed MS, Al-Dirbashi OY, Patay Z, Rahbeeni Z, Abu-Amero KK.
2008. Hyperornithinemia-hyperammonemia-homocitrullinuria syndrome with
stroke-like imaging presentation: clinical, biochemical and molecular analysis. J
Neurol Sci 264:187–194.
Andrews RM, Kubacka I, Chinnery PF, Lightowlers RN, Turnbull DM, Howell N.
1999. Reanalysis and revision of the Cambridge reference sequence for human
mitochondrial DNA. Nat Genet 23:147.
Camacho JA, Obie C, Biery B, Goodman BK, Hu CA, Almashanu S, Steel G, Casey R,
Lambert M, Mitchell GA, Valle D. 1999. Hyperornithinaemia-hyperammonae-
mia-homocitrullinuria syndrome is caused by mutations in a gene encoding a
mitochondrial ornithine transporter. Nat Genet 22:151–158.
Camacho JA, Rioseco-Camacho N, Andrade D, Porter J, Kong J. 2003. Cloning and
characterization of human ORNT2: a second mitochondrial ornithine
transporter that can rescue a defective ORNT1 in patients with the
hyperornithinemia-hyperammonemia-homocitrullinuria syndrome, a urea cycle
disorder. Mol Genet Metab 79:257–271.
Camacho JA, Mardach R, Rioseco-Camacho N, Ruiz-Pesini E, Derbeneva O, Andrade
D, Zaldivar F, Qu Y, Cederbaum SD. 2006. Clinical and functional
characterization of a human ORNT1 mutation (T32R) in the hyperornithine-
mia-hyperammonemia-homocitrullinuria (HHH) syndrome. Pediatr Res
60:423–429.
Cappello AR, Curcio R, Miniero DV, Stipani I, Robinson AJ, Kunji ERS, Palmieri F.
2006. Functional and structural role of amino acid residues in the even-
numbered transmembrane a-helices of the bovine mitochondrial oxoglutarate
carrier. J Mol Biol 363:51–62.
Cappello AR, Miniero DV, Curcio R, Ludovico A, Daddabbo L, Stipani I, Robinson AJ,
Kunji ERS, Palmieri F. 2007. Functional and structural role of amino acid residues
in the odd-numbered transmembrane a-helices of the bovine mitochondrial
oxoglutarate carrier. J Mol Biol 369:400–412.
De Lucas JR, Indiveri C, Tonazzi A, Perez P, Giangregorio N, Iacobazzi V, Palmieri F.
2008. Functional characterization of residues within the carnitine/acylcarnitine
translocase RX(2)PANAAXF distinct motif. Mol Membr Biol 25:152–163.
Dionisi-Vici C, Bachmann C, Gambarara M, Colombo JP, Sabetta G. 1987.
Hyperornithinemia-hyperammonemia-homocitrullinuria syndrome: low crea-
tine excretion and effect of citrulline, arginine, or ornithine supplement. Pediatr
Res 22:364–367.
Fecarotta S, Parenti G, Vajro P, Zuppaldi A, Della Casa R, Carbone MT, Correra A,
Torre G, Riva S, Dionisi-Vici C, Santorelli FM, Andria G. 2006. HHH syndrome
(hyperornithinaemia, hyperammonaemia, homocitrullinuria), with fulminant
hepatitis-like presentation. J Inherit Metab Dis 29:186–189.
Fiermonte G, Walker JE, Palmieri F. 1993. Abundant bacterial expression and
reconstitution of an intrinsic membrane-transport protein from bovine
mitochondria. Biochem J 294:293–299.
Fiermonte G, Dolce V, Palmieri F. 1998. Expression in Escherichia coli, functional
characterization, and tissue distribution of isoforms A and B of the phosphate
carrier from bovine mitochondria. J Biol Chem 273:22782–22787.
Fiermonte G, Dolce V, David L, Santorelli FM, Dionisi-Vici C, Palmieri F, Walker JE.
2003. The mitochondrial ornithine transporter. Bacterial expression, reconstitu-
tion, functional characterization, and tissue distribution of two human
isoforms. J Biol Chem 278:32778–32783.
Gatfield PD, Taller E, Wolfe DM, Haust MD. 1975. Hyperornithinemia, hyper-
ammonemia, and homocitrullinuria associated with decreased carbamyl
phosphate synthetase I activity. Pediatr Res 9:488–497.
Ho SN, Hunt HD, Horton RM, Pullen JK, Pease LR. 1989. Site-directed mutagenesis
by overlap extension using the polymerase chain reaaction. Gene 77:51–59.
748 HUMAN MUTATION, Vol. 30, No. 5, 741–748, 2009
Korman SH, Kanazawa N, Abu-Libdeh B, Gutman A, Tsujino S. 2004.
Hyperornithinemia, hyperammonemia, and homocitrullinuria syndrome with
evidence of mitochondrial dysfunction due to a novel SLC25A15 (ORNT1) gene
mutation in a Palestinian family. J Neurol Sci 218:53–58.
Kyngäs J, Valjakka J. 1998. Unreliability of the Chou-Fasman parameters in predicting
protein secondary structure. Protein Eng 11:345–348.
Marescau B, De Deyn PP, Lowenthal A, Qureshi IA, Antonozzi I, Bachmann C,
Cederbaum SD, Cerone R, Chamoles N, Colombo JP, et al. 1990. Guanidino
compound analysis as a complementary diagnostic parameter for hyperargini-
nemia: follow-up of guanidino compound levels during therapy. Pediatr Res
27:297–303.
Miyamoto T, Kanazawa N, Kato S, Kawakami M, Inoue Y, Kuhara T, Inoue T,
Takeshita K, Tsujino S. 2001. Diagnosis of Japanese patients with HHH
syndrome by molecular genetic analysis: a common mutation, R179X. J Hum
Genet 46:260–262.
Morozzo della Rocca B, Miniero DV, Tasco G, Dolce V, Falconi M, Ludovico A,
Cappello AR, Sanchez P, Stipani I, Casadio R, Desideri A, Palmieri F. 2005.
Substrate-induced conformational changes of the mitochondrial oxoglutarate
carrier: a spectroscopic and molecular modelling study. Mol Membr Biol
2005;443–452.
Palmieri F, Indiveri C, Bisaccia F, Iacobazzi V. 1995. Mitochondrial metabolite carrier
proteins: purification, reconstitution and transport studies. Methods Enzymol
260:349–369.
Palmieri F. 2004. The mitochondrial transporter family (SLC25): physiological and
pathological implications. Pflugers Arch 447:689–709.
Palmieri F. 2008. Diseases caused by defects of mitochondrial carriers: a review.
Biochim Biophys Acta 1777:564–578.
Pebay-Peyroula E, Dahout-Gonzalez C, Kahn R, Trézéguet V, Lauquin GJ, Brandolin
G. 2003. Structure of mitochondrial ADP/ATP carrier in complex with
carboxyatractyloside. Nature 426:39–44.
Robinson AJ, Kunji ER. 2006. Mitochondrial carriers in the cytoplasmic state
have a common substrate binding site. Proc Natl Acad Sci USA 103:
2617–2622.
Ruiz-Pesini E, Lapena AC, Diez-Sanchez C, Perez-Martos A, Montoya J, Alvarez E,
Diaz M, Urries A, Montoro L, Lopez-Perez MJ, Enriquez JA. 2000. Human
mtDNA haplogroups associated with high or reduced spermatozoa motility. Am
J Hum Genet 67:682–696.
Salvi S, Dionisi-Vici C, Bertini E, Verardo M, Santorelli FM. 2001a. Seven
novel mutations in the ORNT1 gene (SLC25A15) in patients with
hyperornithinemia, hyperammonemia, and homocitrullinuria syndrome. Hum
Mutat 18:460.
Salvi S, Santorelli FM, Bertini E, Boldrini R, Meli C, Donati A, Burlina AB, Rizzo C,
Di Capua M, Fariello G, Dionisi-Vici C. 2001b. Clinical and molecular findings
in hyperornithinemia-hyperammonemia-homocitrullinuria syndrome. Neurol-
ogy 57:911–914.
Shih VE, Efron ML, Moser HW. 1969. Hyperornithinemia, hyperammonemia, and
homocitrullinemia: a new disorder of amino acid metabolism associated with
myoclonic seizures and mental retardation. Am J Dis Child 117:83–92.
Shimizu H, Maekawa K, Eto Y. 1990. Abnormal urinary excretion of polyamines in
HHH syndrome (hyperornithinemia associated with hyperammonemia and
homocitrullinuria). Brain Dev 12:533–535.
Tonazzi A, Giangregorio N, Indiveri C, Palmieri F. 2005. Identification by site-
directed mutagenesis and chemical modification of three vicinal cysteine
residues in rat mitochondrial carnitine/acylcarnitine transporter. J Biol Chem
280:19607–19612.
Torroni A, Huoponen K, Francalacci P, Petrozzi M, Morelli L, Scozzari R, Obinu D,
Savontaus ML, Wallace DC. 1996. Classification of European mtDNAs from an
analysis of three European populations. Genetics 144:1835–1850.
Tsujino S, Kanazawa N, Ohashi T, Eto Y, Saito T, Kira J, Yamada T. 2000. Three novel
mutations (G27E, insAAC, R179X) in the ORNT1 gene of Japanese patients
with hyperornithinemia, hyperammonemia, and homocitrullinuria syndrome.
Ann Neurol 47:625–631.
Valle D, Simell O. 2001. The hyperonithinemias. In: Scriver CR, Beaudet AL, Sly WS,
Valle D, editors. The metabolic and molecular basis of inherited disease, 8th
edition. New York: McGraw-Hill, Inc. p 114–186.
Vriend G. 1990. WHAT IF: a molecular modeling and drug design program. J Mol
Graph 8:52–56.