The Mitochondrial Ornithine Transporter

Two isoforms of the human ornithine carrier, ORC1 and ORC2, have been identified by overexpression of the proteins in bacteria and by study of the transport properties of the purified proteins reconstituted into liposomes. Both transport l-isomers of ornithine, lysine, arginine, and citrulline by exchange and by unidirectional mechanisms, and they are inactivated by the same inhibitors. ORC2 has a broader specificity than ORC1, and l- and d-histidine, l-homoarginine, and d-isomers of ornithine, lysine, and ornithine are all substrates. Both proteins are expressed in a wide range of human tissues, but ORC1 is the predominant form. The highest levels of expression of both isoforms are in the liver. Five mutant forms of ORC1 associated with the human disease hyperornithinemia-hyperammonemia-homocitrullinuria were also made. The mutations abolish the transport properties of the protein. In patients with hyperornithinemia-hyperammonemia-homocitrullinuria, isoform ORC2 is unmodified, and its presence compensates partially for defective ORC1.

Rat liver mitochondria contain an ornithine-citrulline transport protein, often known as the ornithine carrier (ORC) 1 (1)(2)(3)(4)(5). The reconstituted purified protein is highly active in ornithinecitrulline exchange and is somewhat less active in unidirectional transport of ornithine. It transports lysine and arginine also but does not transport histidine. The positive charges of ornithine and lysine are compensated by cotransport of a proton with citrulline in exchange. Its affinity for ornithine is lower than for other substrates, and under saturating internal concentrations it is not dependent on the nature of the countersubstrate. The exchange reaction operates by a simultaneous (sequential) mechanism, and the unidirectional transport of ornithine or lysine (but not of citrulline) is compensated by a proton. The exchange of cytosolic ornithine for matrix citrulline is part of the urea cycle (1,3,6), and the ornithine-H ϩ exchange also has an important role in the catabolism of excess arginine and the biosynthesis of polyamines (4).
In Saccharomyces cerevisiae, the ornithine carrier is encoded by ARG11/ORT1, and its identity has been established by overexpression in bacteria, reconstitution, and study of its transport properties (7). Subsequently, the gene for the human mitochondrial ornithine carrier, which is defective in the hyperornithinemia-hyperammonemia-homocitrullinuria (HHH) syndrome (8), was identified by homology with the yeast sequence, and its function as an ORC was inferred from its ability to complement the defect of HHH fibroblasts by incorporating radioactive ornithine into cellular protein. The HHH phenotype is milder than those associated with a deficiency of an enzyme in the urea cycle, and it has been suggested that carriers with other functions can compensate for the defective protein by transporting ornithine (8).
As described below, the human and mouse genomes contain a spliced pseudogene that in man encodes a second isoform, ORC2 (9). The isoforms are expressed differently in various human tissues. Both were overexpressed in Escherichia coli. The purified proteins were reconstituted into liposomes, and their transport properties were characterized. The recombinant proteins have different substrate specificities, transport affinities (K m ), and specific activities (V max ). Five independent mutations associated with HHH were introduced singly into the isoform ORC1. They affected its transport properties severely. In the same patients with HHH, ORC2 is unaffected, and therefore, it is likely that it compensates for defective ORC1.

EXPERIMENTAL PROCEDURES
Sequence Search and Analysis-Data bases were screened with the sequence of ORC1 (GenBank TM /EBI accession number AF112968, gene name SLC25A15) with BLASTP and TBLASTN. The amino acid sequences were aligned with ClustalW (version 1.7).
Construction of Expression Plasmids-The coding sequences for ORC1 and -2 (GenBank TM /EBI accession number AF332005, gene name SLC25A2) were amplified from human liver cDNA by PCR. Total RNA populations extracted from the fibroblasts of patients with HHH were reverse transcribed, and the coding sequences for ORC1 mutants were amplified by PCR with the aid of a GeneAmp® RNA PCR core kit (Applied Biosystems, Foster City, CA). The cDNA for ORC2 was amplified from the same six samples. The mutant forms of ORC1 corresponded to 7 alleles from 6 HHH patients (10,11). Five of them were homozygous for the mutations G27R, R275Q, R179X, C861insG, and IVS5ϩ1g3a, and the other one was heterozygous for G190D and F188⌬. The recommended nomenclature for human gene mutations is that the adenosine of the first AUG codon of the mRNA from the * This work was supported by grants from the Ministero dell'Università e la Ricerca Scientifica e Tecnologica-Programmi di Ricerca di Interesse Nazionale (MURST-PRIN), MURST L.488/92 CO3 and CO4, MURST-Consiglio Nazionale delle Ricerche (CNR) L.95/95, the Centro di Eccellenza Geni in campo Biosanitario e Agroalimentare (CEGBA), the CNR target project on biotechnology, and the European Social Fund. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. This paper is dedicated to the memory of Professor Eraldo Antonini. ʈ To whom correspondence may be addressed. SLC25A15 gene (GenBank TM /EBI accession number NM_014252) is taken as nucleotide 1, and the corresponding methionine is taken as residue 1 of the protein. The oligonucleotide primers corresponded to the extremities of the coding sequences of ORC1 and -2 with additional NdeI and HindIII sites. The amplified products were cloned into the pRUN expression vector, which is derived from pKN172 (12), and the constructs were transformed into E. coli TOP 10 cells (Invitrogen). Transformants were selected on 2ϫ TY (16 g/liter tryptone, 10 g/liter yeast extract, 5 g/liter NaCl, pH 7.4) plates containing ampicillin (100 g/ml) and were screened by direct colony PCR and restriction digestion of plasmids.
Expression Analysis by Real Time PCR-Total RNAs from human tissues (Invitrogen) were reverse transcribed with the GeneAmp® RNA PCR core kit (Applied Biosystems) with random hexamers as primers. Primers and probes, based on the ORC1 and -2 cDNA sequences, were designed for real time PCRs with Primer Express (Applied Biosystems). The forward and reverse primers for ORC1 corresponded to nt 351-373 and 392-415, respectively, and those for ORC2 corresponded to nt 725-746 and 784 -812, respectively. The ORC1 FAM/MGB-and the ORC2 VIC/MGB-Dark Quencher-labeled probes corresponded to nt 374 -389 and 755-775 of the ORC1 and -2 cDNA sequences, respectively. Real time PCRs were performed in a MicroAmp® optical 96-well plate using the automated ABI Prism 7000 sequence detector system (Applied Biosystems). The reaction mixture (50 l) contained template (reverse transcribed first-stranded cDNA, 5 l), 1ϫ TaqMan Universal Master mix (Applied Biosystems), 200 nM probe for ORC1 or -2, and 900 nM each primer. To correct for differences in the amount of starting first-stranded cDNAs, the human ␤-actin gene was amplified in parallel as a reference endogenous housekeeping gene. The relative quantification of the two isoforms was performed according to the comparative method (2 Ϫ⌬⌬Ct ) (Ref. 13 and Applied Biosystems User Bulletin 2 P/N 4303859), with the heart ⌬Ct for ORC1 and the spleen ⌬Ct for ORC2 as the internal calibrator. 2 Ϫ⌬⌬Ct equals 2 Ϫ(⌬Ct sample Ϫ ⌬Ct calibrator) , where ⌬Ct sample is Ct sample Ϫ Ct reference gene and Ct is the threshold cycle, i.e. the PCR cycle number at which emitted fluorescence exceeds 10ϫ standard deviation of baseline emissions.
Bacterial Expression and Purification of ORC1, ORC2, and Mutated Forms of ORC1-ORC1 and its mutated forms as well as ORC2 were overexpressed as inclusion bodies in the cytosol of E. coli as described before (12), except that the host cells were E. coli CO214(DE3) (14,15). Control cultures with the empty vector were processed in parallel. Inclusion bodies were purified on a sucrose density gradient (12) and were washed at 4°C, first with TE buffer (10 mM Tris/HCl, 1 mM EDTA, pH 6.5), then twice with a buffer containing Triton X-114 (3%, w/v), 1 mM EDTA, and 10 mM HEPES pH 7.2, and finally with TE buffer. The proteins were solubilized in 1.8% sarkosyl (w/v). Small residues were removed by centrifugation (258,000 ϫ g for 1 h).
Reconstitution into Liposomes-Recombinant proteins in sarkosyl were reconstituted into liposomes in the presence or absence of substrates (16). The reconstitution mixture contained solubilized proteins (150 l with 1-1.2 g of protein), 10% Triton X-114 (70 l), 10% phospholipids as sonicated liposomes (80 l), 20 mM ornithine (except where indicated otherwise), 10 mM HEPES, pH 7.2, cardiolipin (0.8 mg, Sigma), and water to a final volume of 700 l. These components were mixed thoroughly, and the mixture was recycled 13 times through the same Amberlite column (Bio-Rad).
Transport Assays-External substrate was removed from proteoliposomes on columns of Sephadex G-75, which was pre-equilibrated with 50 mM sucrose and 10 mM HEPES, pH 7.2. Transport at 25°C was started by adding L- ]citrulline (PerkinElmer Life Sciences) to substrate-loaded proteoliposomes (exchange) or to empty proteoliposomes (unidirectional transport). In both cases, transport was terminated by addition of 15 mM pyridoxal 5Ј-phosphate (the "inhibitor-stop" method) (16). In controls, the inhibitor was added at the beginning 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 the radioactivity in the liposomes was measured (16). 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). Alternatively, the initial transport rate was calculated from the time course of isotope equilibration (16). The reconstituted proteins were assayed for other exchange activities by the inhibitor-stop method.
Other Methods-Proteins were analyzed by SDS-PAGE and stained ith Coomassie Blue dye. N-terminal sequencing was carried out as described before (17). The amount of ORC1 and -2 was estimated by laser densitometry of stained samples using carbonic anhydrase as the protein standard (17). The amount of protein incorporated into liposomes was measured as described previously (17). It varied between 24 and 32% of the protein added to the reconstitution mixture.

RESULTS
Identification of Human Isoform ORC2-By screening data bases with the sequence of ORC1, an intron-less gene interrupting the human protocadherin gene cluster (GenBank TM / EBI accession number AF332005, Ref. 9) was found to encode a protein 301 amino acids long that is 87% identical in sequence to ORC1. This protein is isoform ORC2.
Expression of the mRNAs for Human ORC1 and -2 in Various Tissues-The tissue distribution of mRNAs for isoforms ORC1 and -2 is summarized in Fig. 1. In all of these tissues, the ORC1 mRNA was expressed at higher levels than the ORC2 mRNA. The amount of the ORC1 mRNA in heart is about the same as that of the ORC2 mRNA in spleen, and as such this value served as an internal calibration in the relative quantification of the isoforms in various tissues. The ORC1 mRNA was expressed most strongly in liver, pancreas, testis, lung, and small intestine, and lower levels were detected in spleen, kidney, brain, and heart. The ORC2 mRNA was expressed in reasonable abundance in liver, testis, spleen, lung, pancreas, and small intestine and expressed poorly in other tissues. The ratio of ORC1 mRNA to ORC2 mRNA was 41 in brain, 93 in heart, 81 in kidney, 33 in liver, 65 in lung, 161 in pancreas, 38 in small intestine, 9.2 in spleen, and 18 in testis. However, because of the possibility of the operation of post-transcriptional mechanisms, these levels do not necessarily reflect the ratios of transport activities.
Bacterial Overexpression of ORC1 and -2-Both isoforms were expressed at high levels in E. coli C0214(DE3) (see Fig. 2, lanes 5 and 6), and about 70 mg of purified proteins were obtained per liter of culture. The proteins were not detected either in bacteria harvested immediately before induction of expression (lanes 1-3) or in control (with only vector) cells harvested after the induction (lane 4). Purified ORC1 and -2 (see Fig. 2, lanes 7 and 8) had apparent molecular masses of 31.5 and 31.0 kDa, respectively. The calculated values, including the initiator methionine, are 32714 and 32558 Da, respectively. All of the purified ORC1 mutant proteins gave single bands on SDS-PAGE. Their apparent molecular masses were 31.5 kDa for mutants G27R (Fig. 2, lane 9), G190D, R275Q, and F188⌬ (data not shown), 26.0 kDa for mutant IVS5ϩ1g3a Real time PCR experiments were conducted on cDNAs prepared by reverse transcription of total RNAs from various human tissues using specific primers and probes based on human ORC1 and -2. The relative quantification of human ORC1 (black bars) and ORC2 (gray bars) was performed by the comparative method (2 Ϫ⌬⌬Ct ). Human ␤-actin was employed as reference.
Transport Properties of Recombinant ORC1 and -2-In homo-exchange assays (same substrate inside and outside at external and internal concentrations of 1 and 10 mM, respectively), reconstituted ORC1 and -2 both catalyzed a [ 3 H]ornithine-ornithine exchange that was inhibitable by pyridoxal 5Ј-phosphate. They did not catalyze homo-exchanges for ADP, alanine, aspartate, ATP, carnitine, choline, citrate, glutamate, glutathione, malate, malonate, oxoglutarate, phosphate, proline, pyruvate, spermine, threonine, and valine. Control proteoliposomes made with boiled samples of ORC1 and -2 and with sarkosyl-solubilized material from bacterial cells either lacking expression vectors for ORC1 and -2 or harvested immediately before induction of expression had no [ 3 H]ornithineornithine exchange activity.
The kinetics of uptake into proteoliposomes of 1 mM [ 3 H]ornithine measured in the presence (exchange) or absence (unidirectional transport) of 10 mM internal ornithine are compared in Fig. 3. In both modes, isotopic equilibrium was approached exponentially in accord with transport by first-order kinetics. The ratio of maximal substrate uptake by exchange and by unidirectional transport was 9.0 for ORC1 and 11.3 for ORC2, agreeing with the value of 10 expected from intraliposomal concentrations at equilibrium of 1 and 10 mM for unidirectional transport and exchange, respectively. The addition of 10 mM unlabeled ornithine after incubation for 30 min (ORC1) or 120 min (ORC2) when radioactive uptake by the proteoliposomes had almost approached equilibrium caused an extensive efflux of radiolabeled ornithine from both ornithine-loaded and -unloaded proteoliposomes (data not shown). This efflux shows that the [ 3 H]ornithine taken up by unidirectional transport is released by exchange for externally added substrate. Therefore, ORC1 and -2 catalyze both the unidirectional transport of ornithine and ornithine-ornithine exchange, as reported also for the ornithine carrier from rat liver mitochondria and the recombinant yeast ortholog (1, 4, 7).
The substrate specificities of ORC1 and -2 were examined by measuring the uptake of [ 3 H]ornithine into proteoliposomes pre-loaded with potential substrates (Fig. 4). With ORC1, the highest activities were observed in the presence of internal L-ornithine, L-lysine, L-arginine, and L-citrulline (see Fig. 4A). The D-isomers were much less effective. The activities detected in the presence of internal histidine, carnitine, cysteine, phenylalanine, and valine and (data not shown) alanine, ␥-aminobutyrate, citrate, glutamine, leucine, proline, serine, and threonine were virtually the same as the activity observed without internal substrate. Virtually no activities were detected with internal spermine or spermidine, suggesting that they inhibit ornithine transport by ORC1. In contrast, with internal malate or phosphate, [ 3 H]ornithine uptake was greater than in the absence of internal substrate.
Isoform ORC2 behaved rather differently (see Fig. 4B), and high [ 3 H]ornithine transport was observed with both L-and D-isomers of internal ornithine, lysine, arginine, and histidine. The rate of the unidirectional transport of ornithine by ORC2 was about 25% of the rate of ornithine-ornithine exchange. Furthermore, with internal citrulline the rate of [ 3 H]ornithine uptake compared with that of ornithine uptake with no internal substrate was enhanced less with ORC2 than with ORC1. However, as observed with ORC1, the [ 3 H]ornithine uptake by ORC2 was the same with no internal substrate as with internal carnitine, cysteine, phenylalanine, and valine and (data not shown) alanine, ␥-aminobutyrate, glutamine, leucine, proline, and threonine; there was very low activity with internal spermine and spermidine and significant activity with internal malate and phosphate.  1 and 4) and with the coding sequence of ORC1 (lanes 2 and 5) or ORC2 (lanes 3 and 6). Samples were taken at the time of induction (lanes 1-3) and 5 h later (lanes 4 -6). The same number of bacteria was analyzed in each sample. Lanes 7 and 8, ORC1 and -2 purified from bacteria in lanes 5 and 6, respectively; lanes 9 -12, recombinant ORC1 mutants (G27R, IVS5ϩ1g3a, c861insG, and R179X, respectively) purified as was the wild-type protein reported in lane 7.
As ORC1 and -2 did not catalyze either [ 32 P]phosphatephosphate or [ 14 C]malate-malate homo-exchanges (see above), the effect of internal malate or phosphate on the uptake of [ 3 H]ornithine was investigated further. As shown in Table I, [ 3 H]ornithine uptake by ORC1 or -2 was enhanced not only with internal malate or phosphate but also with internal malonate, succinate, fumarate, aspartate, glutamate, and oxoglutarate in comparison with proteoliposomes with no internal substrate. In contrast, with asparagine, glutamine, and acetoacetate and (data not shown) propionate, butyrate, ␥-aminobutyrate, and 2-oxoisocaproate, the transport of [ 3 H]ornithine by both isoforms was virtually the same as in the absence of internal substrate. Results similar to those illustrated in Table  I  H]ornithine. Only at high concentrations with a K i higher than 10 mM did these compounds inhibit the carrier-mediated uptake of [ 3 H]ornithine. Therefore, malate, phosphate, and other dicarboxylates may regulate the rate of ornithine uptake by reacting with ORC1 and -2 on the internal surface of the proteoliposomes.
The [ 3 H]ornithine-ornithine exchange reactions catalyzed by ORC1 and -2 were inhibited strongly by pyridoxal 5Ј-phosphate (an inhibitor of several mitochondrial carriers) as well as by mercurials (mersalyl, p-chloromercuribenzene sulfonate, and mercuric chloride) and N-ethylmaleimide, which are powerful inhibitors of the ornithine carrier from rat liver (1). Both isoforms were inhibited strongly by spermine (Fig. 5), whereas carboxyatractyloside, bongkrekate, ␣-cyano-4-hydroxycinnamate, and bromocresol purple (inhibitors of other mitochondrial carriers) had little or no effect.
Kinetic Characteristics-The kinetic constants of ORC1 and -2 were determined from the initial transport rate of homoexchanges at various external labeled substrate concentrations in the presence of a constant saturating internal substrate concentration of 20 mM. With both isoforms, linear functions were obtained in double reciprocal plots. For ORC1, the specific activities (V max ) for homo-exchanges for ornithine, lysine, arginine, and citrulline were all about 3.0 Ϯ 0.4 mmol/min/g of protein, and the transport affinities (K m ) were 0.22 Ϯ 0.02, 0.80 Ϯ 0.06, 1.58 Ϯ 0.18, and 2.52 Ϯ 0.30 mM, respectively (in at least four experiments for each homo-exchange). For ORC2, the transport affinities were 0.40 Ϯ 0.06, 0.32 Ϯ 0.05, 0.71 Ϯ 0.09, and 1.28 Ϯ 0.14 mM, respectively, for L-isomers of ornithine, lysine, arginine, and histidine (in at least four experiments for each homo-exchange), and the V max values (1.2 Ϯ 0.2 mmol/min/g of protein) were again independent of the substrate. Citrulline, lysine, spermine, and spermidine inhibited competitively [ 3 H]ornithine uptake by both isoforms by increasing the apparent K m without changing the V max of ornithine uptake (data not shown). For ORC2, the inhibition constants (K i ) of citrulline, lysine, spermine, and spermidine were 8. 1 Ϯ 1.2, 0.37 Ϯ 0.04, 0.30 Ϯ 0.03, and 0.48 Ϯ 0.06 mM, respectively (in at least three experiments for each inhibitor). For ORC1, the K i values of citrulline, lysine, spermine, and  . Transport Properties of Mutant Forms of ORC1-The seven mutant forms of ORC1 were expressed in E. coli and reconstituted into liposomes just as well as the wild-type protein. Five of them, G27R, R275Q, R179X (residues 1-178 of ORC1), C861insG (residues 1-290), and IVS5ϩ1g3a (ORC1 lacking transmembrane segments 5 and 6), were incapable of catalyzing homo-exchanges of ornithine, arginine, lysine, and citrulline (Fig. 6). Even after a 30-min incubation in the presence of external substrate concentrations 15-fold higher than the K m , no activity was detected (data not shown). It is likely that their inability to transport ornithine and citrulline leads to the HHH syndrome. The other two mutant forms of ORC1, i.e. G190D and F188⌬, maintained a residual transport activity of ϳ35 and 10%, respectively. They occurred in different alleles of a patient who did not show a milder phenotype compared with that of the other patients studied in this work. He presented with neonatal onset hyperammonemia and developed spastic paraparesis in early adulthood (10). However, a clear genotypephenotype correlation is often lacking in human diseases. In all six patients, the sequence of ORC2 was unaltered, and, from semi-quantitative reverse transcription-PCR experiments, it appeared that expression levels in cultured fibroblasts do not differ significantly from the levels found in control fibroblasts. DISCUSSION Isoforms ORC1 and -2 have many transport properties in common. They both transport ornithine, lysine, arginine, and citrulline by exchange and unidirectionally, and they are inactivated by the same inhibitors. However, they also differ in a number of respects. First, ORC2 has a broader substrate specificity than ORC1. It transports L-and D-histidine, L-homoarginine, and D-isomers of ornithine, lysine, and arginine with the same efficiency as L-isomers. Second, the K m values of ORC2 for lysine and arginine are lower than for ORC1, and those for ornithine and citrulline are higher. Third, ORC2 is three times less active than ORC1.
The transport properties of rat liver ORC (1,2,5) are more similar to human ORC1 than to ORC2. Neither the native rat ORC nor the recombinant human ORC1 transports histidine; they both prefer to transport L-isomers of other amino acids, and their transport affinities for ornithine, lysine, arginine, and citrulline are very similar. Thus, as in human liver, ORC1 is probably the predominantly expressed isoform in rat liver.
Because ORC1 transports citrulline better than ORC2 and because it is more abundant than ORC2 in liver, ORC1 is likely to carry out the important function of exchanging cytosolic ornithine and mitochondrial citrulline. Thus, it probably provides an essential link in the urea cycle between enzyme activities in the cytosol and others in the mitochondrial matrix. The lower activity of ORC1 in patients with HHH syndrome supports this interpretation. Moreover, the transport of lysine by ORC1 provides a means for inhibition of urea synthesis by lysine observed in isolated hepatocytes (18). The phenotype of HHH patients is generally milder than those associated with defects in any urea cycle enzyme. In HHH patients, inactive ORC1 is found alongside fully active ORC2 in liver mitochondria and can compensate partially for ORC1. However, because it has a lower affinity for ornithine and citrulline and is expressed at a lower level, it cannot replace completely the function of ORC1 in the urea cycle. Since in mouse the second gene for ORC is a pseudogene (for the presence of a single nt deletion near the 5Ј-end causing a frameshift) (9), it would be interesting to delete the murine gene encoding for ORC1 and compare the resulting phenotype with that of a urea cycle enzyme knock-out mouse.
Isoform ORC1 and especially isoform ORC2 are likely to play a variety of important roles in cell metabolism. The import of lysine, arginine, and histidine into mitochondria is required for  mitochondrial protein synthesis, and import of ornithine is needed for the degradation of excess arginine. When the dietary content of arginine is high, the production of ornithine exceeds the requirement for citrulline formation. This excess is removed by ornithine aminotransferase (19,20) expressed in the mitochondrial matrix in response to the level of protein in the diet (21)(22)(23). In isolated rat liver mitochondria, ornithine aminotransferase catabolizes ornithine at a rate that is limited by transport of ornithine into the matrix (20). The hepatic enzyme is found in pericentral hepatocytes that contain glutamine synthetase and not in the periportal hepatocytes containing the urea cycle enzymes (24). Therefore, it seems possible that the ornithine-citrulline exchange activity of ORC1 and -2 occurs in mitochondria of periportal hepatocytes and that ornithine-H ϩ exchange is performed by the same transporters in mitochondria of the pericentral hepatocytes. Another possibility is that ORC2 is located mainly in mitochondria of pericentral hepatocytes and that ORC1 is in mitochondria of periportal hepatocytes.
The efflux of ornithine from mitochondria is required for polyamine biosynthesis from ornithine in the cytosol. When the diet content of arginine is low and/or in the tissues where the activity of arginase is negligible (25), ornithine made from glutamate inside mitochondria (21)(22)(23) has to be exported for polyamine biosynthesis. In intestinal mitochondria ornithine is made from glutamine and so could provide extra urea cycle intermediates for the liver (26). ORC1 and -2 are expressed in various tissues at different levels, and both isoforms are inhibited by spermine and spermidine. Thus, polyamines in the cytosol could control their own synthesis by reducing the amount of ornithine exported from the mitochondria via the ornithine carrier.
Another important observation is that the activities of ORC1 and -2 are stimulated by dicarboxylates, phosphate, and malate from the inside of the proteoliposomes, but as none of these compounds is transported by ORC1 and -2, the simplest explanation of this regulation is that malate and phosphate interact at a site distinct from the substrate-binding site. This site can bind C3-C5 dicarboxylates but not asparagine, glutamine, or monocarboxylates. Malate increases urea synthesis in isolated hepatocytes first by transporting into mitochondria reducing equivalents and the carbon skeleton of fumarate produced in the cytosol by argininosuccinate lyase and secondly by producing aspartate inside the mitochondria via the malate dehydrogenase and aspartate aminotransferase (27). As aspartate is a substrate of argininosuccinate synthetase, it is essential for the functioning of the urea cycle.
If ORC1 and -2 are reconstituted into liposomes with the same orientation as in the mitochondria, malate could stimulate ureogenesis directly by activating ORC1 and -2 on the matrix side of the inner mitochondrial membrane. Because the intramitochondrial concentration of phosphate is high and the affinity of the ornithine carrier for dicarboxylates other than malate is low, the physiological effect of these dicarboxylates on the activity of the ORC may be minor.