INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The catecholamines adrenaline, noradrenaline, dopamine, and the tryptophan derivative serotonin function as neurotransmitters of the monoaminergic neurons and as hormones in the control of physiological processes like glucose storage and metabolism, thermoregulation, and blood pressure. Changes in synaptic concentration or temporary availability of monoamines are associated with mental dysfunction, neuropsychiatric disorders, and drug addiction (10). Furthermore, altered plasma concentrations can result in physiological dysfunction (28). Tight control of the levels of synaptic and circulating catecholamines is thus essential for proper neuronal signaling and maintenance of internal homeostasis.

Two uptake systems that clear extracellular monoamines have been described. The neuronal uptake-1 system is present at presynaptic nerve endings and mediates the reuptake of released monoamines from the synaptic cleft. Uptake-1 is a high-affinity and Na⁺- and Cl-dependent system mediated by the noradrenaline (Net), dopamine (Dat), and serotonin (Sert) transporter proteins (reviewed in reference 1). Targeted inactivation experiments with mice have shown that the uptake-1 transporter proteins are a target for antidepressant and psychostimulatory treatments and are pivotal in the control of synaptic catecholamine concentrations and prevention of neurobehavioral changes (3, 14, 37).

The extraneuronal uptake-2 system was originally discovered in myocardial cells of the rat heart but has also been identified in vascular and nonvascular smooth muscle cells like those in the uterus, as well as in human central nervous system glial and kidney carcinoma cell lines (19, 24, 26, 31, 33). Uptake-2 can be discriminated from uptake-1 in substrate specificity and transport kinetics (reviewed in reference 33). In addition, corticosteroids, -haloalkylamines, and O-methylated catecholamines are inhibitors of uptake-2 but not of uptake-1. The cyanine derivative disprocynium 24 (D24) was isolated as a highly potent uptake-2 inhibitor in vitro (23). However, the application of D24 in vivo to study uptake-2 was revealed to be limited, as it was shown previously that D24 blocks not only uptake-2 but also other transport mechanisms that clear catecholamines (8, 11).

Recently, molecular identification of the uptake-2 transporter protein has been reported. Called extraneuronal monoamine transporter (EMT) in humans and organic cation transporter 3 (OCT3) in rats, the protein is predicted to contain a 12-transmembrane domain structure (18, 36). The mouse homolog of the EMT and OCT3 genes, called Orct3 (locus name Slc22a3), was isolated from the critical region of the natural embryonic lethal mouse mutant t^w73 and shown to be tightly linked to the closely related Orct1/Slc22a1 and Orct2/Slc22a2 organic cation transporter genes (35). This physical linkage is conserved in humans and suggests that these genes have evolved from a common ancestor. Further evidence for this comes from in vitro studies in which it has been shown that the Orct1, Orct2, and Orct3 proteins can all transport catecholamines and the neurotoxin MPP⁺ (1-methyl-4-phenylpyridinium) (4, 16-18). However, transport inhibition studies have shown that only Orct3 is sensitive to all uptake-2 antagonists, including O-methylisoprenaline, with nearly identical kinetics (15, 18). In mice, the Orct1, Orct2, and Orct3 genes have clearly distinct expression profiles. Orct1 is expressed in liver, kidney, and intestine, whereas Orct2 expression is restricted to the kidney and brain (21, 27). In contrast, Orct3 expression is seen in a wide range of tissues. The highest levels of expression are found in skeletal muscle and in the heart and uterus, for which uptake-2 activity has been described previously (33, 35). Similarly, high expression of the human homolog is found in aorta, prostate, adrenal gland, skeletal muscle, and liver (35). During mouse embryonic development, Orct3 is expressed in the early postimplantation embryo (38). At later stages, expression is restricted to the labyrinth layer of the placenta, in which Orct3 is coexpressed with the gene for the monoamine oxidase A (Maoa) metabolizing enzyme (34). Thus, both the in vitro studies and the expression data have provided strong evidence that Orct3 is the molecular component of the extraneuronal monoamine transport (uptake-2) system.

To test whether Orct3 has a major role in uptake-2 activity in any particular organ in vivo, we have generated mice deficient for Orct3 by homologous recombination in embryonic stem (ES) cells. These mice are viable and fertile and show no obvious physiological defect and no significant imbalance of two tested monoamines, noradrenaline and dopamine. However, using MPP⁺ as a substrate, we show that Orct3 is an essential component for uptake-2 function in the adult heart and placenta but not in other adult organs. These data establish the presence of uptake-2 in the heart and identify the placenta as a novel uptake-2 site of action, where it functions at the fetoplacental interface.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Orct3 gene targeting. For constructing the targeting vector, a 17.5-kb BglII fragment spanning the first exon of the mouse Orct3 gene was isolated from the genomic BAC clone 228C21 (Research Genetics, Inc.) (35). To generate plasmid p16BglII, a 1.4-kb fragment from the 3' end of the subclone was removed using Asp718, followed by religation. In p16BglII, a 4.3-kb BamHI fragment containing exon 1, and 0.65 kb of upstream and 2.9 kb of downstream genomic sequences, was replaced by the pGKneo/pGKtk selection cassette flanked by loxP sites (9). The targeting vector was linearized with XhoI, and 50 µg was used in an electroporation of E14 (129/Ola) mouse ES cells. One hundred ninety-two individual colonies were screened for homologous recombination by hybridization with an 0.8-kb Asp718-SpeI 3' external probe of BglII-digested genomic DNA. Six homologous recombinant clones were identified (i.e., a 3% recombination frequency). The integrity of the homologous recombination event was verified with a SalI-BglII 5' external probe in an SpeI digest (data not shown). The subclonal appearance of the targeted allele is caused by a contamination of the clones with surviving nontargeted ES cells (Fig. 1B). Four individually targeted ES cell clones were transiently transfected with a CRE expression plasmid to remove the neomycin/tk selection cassette. Individual clones were picked and analyzed for CRE-mediated recombination. Upon karyotyping, two independently derived ES cell clones with the Orct3 (positive or negative) genotype were used for blastocyst injections to generate chimeric mice. Both clones resulted in germ line transmission and yielded similar results. The mutant Orct3 allele was bred into an FVB/N genetic background for analysis and is available for distribution.

View larger version (24K): [in this window] [in a new window]   FIG. 1.   Disruption of the mouse Orct3 gene in ES cells. (A) Orct3 gene targeting strategy. The top line is a schematic overview of the Orct3 gene locus. Exon 1 is shown as a gray box. The arrow marks the wild-type Orct3 promoter and the direction of transcription. Restriction enzyme sites for Asp718 (A), BamHI (B), BglII (Bg), and SpeI (Sp) are indicated. In the targeting construct, a 4.3-kb BamHI genomic fragment containing exon 1 was replaced by the selection cassette (neo) flanked by loxP sites (arrowheads). Following homologous recombination, the selection cassette was removed by CRE-mediated recombination, leaving a single loxP site in place of the 4.3-kb BamHI fragment containing exon 1. wt, wild type. (B and C) Genotype analysis of the targeted ES cell clones (B) and the CRE-transfected ES cells (C). An 800-bp 3' external Asp718-SpeI fragment was used as a probe on BglII-digested ES cell genomic DNA, detecting a 17-kb wild-type (+/+), a 9.5-kb homologous recombined (+/neo), and a 12.5-kb floxed (+/) allele.

Genomic Southern blot analysis. ES cell genomic DNA was isolated by proteinase K digestion (100 µg/ml) in 100 mM Tris (pH 8.0)-5 mM EDTA-200 mM NaCl-0.2% sodium dodecyl sulfate. Five micrograms of DNA was used for digestion and run on a 0.6% agarose gel. Upon transfer of the DNA to a Hybond-N⁺ nylon membrane (Amersham), hybridization was performed under Church hybridization conditions (6).

Northern blot analysis. Total RNA of mouse tissues was isolated by the lithium chloride extraction method (2). Analysis of the RNAs by Northern blotting was done as described previously (35). The Orct3 probe was a 2.7-kb fragment from the 3' end of the Orct3 cDNA; mouse Pai-1 and human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNAs were used as probes for loading control.

MPP⁺ transport studies. The mice used for the time curve were 7-week-old wild-type FVB/N female mice. Four mice were injected for the 5- and 10-min points, and three mice were injected for the 15- and 60-min points. For the adult transport experiment, homozygous mutant Orct3 or wild-type littermates of a mixed genetic background (25% 129/Ola, 75% FVB/N) were injected at the age of 12 weeks. Seven mice of each genotype were used. Timed matings were performed for the placental transport studies. Three pregnant mothers were injected at 15.5 days postcoitum (dpc) of gestation, resulting in a total of eight wild-type and six mutant embryos. [³H]MPP⁺ (methyl-4-phenylpyridinium acetate, N-methyl-³H labeled, 77.5 Ci/mmol) was obtained from NEN Life Science Products, Inc., and was diluted 1:150 with cold MPP⁺ iodide (Research Biochemicals International) to a final concentration of 0.2 mg/ml in 0.9% NaCl. After anesthesia with methoxyflurane (Medical Developments), animals were injected intravenously in the tail vein with 1.0 mg of the drug per kg of body weight. The mice were killed at the indicated time points by axillary bleeding after renewed anesthesia. Plasma and organs were collected, weighed, and frozen till further processing. Intestinal contents were separated from intestinal tissues. For the placental transport studies, placentas, embryos, and amniotic fluid as well as maternal blood were collected. Tissues were homogenized in 4% bovine serum albumin, and concentrations of [³H]MPP⁺-derived radioactivity were measured by liquid scintillation counting.

Statistical analysis. All values are given as means ± standard deviations (SD). The two-tailed unpaired Student t test was used to assess the significance of differences between data sets. Differences were considered to be statistically significant when P was <0.01.

Orct3 antiserum. Three polyclonal antibodies were raised in rabbits: (i) one against a C-terminal peptide at amino acids 524 to 542, synthesized onto a lysine tree; (ii) one against a fusion protein consisting of the large extracellular loop between TM1 and TM2 fused to maltose binding protein; and (iii) one against a similar fusion protein, which contained the large intracellular loop between TM6 and TM7. These regions are the least conserved among the Orct1, Orct2, and Orct3 family members. Only the first antigen gave an antiserum which recognized the Orct3 protein in lysates from cells transfected with the wild-type Orct3 gene in a Western blot assay. However, this antiserum was unable to detect Orct3 in placental or adult organ lysates. The Orct3 gene is heavily glycosylated in vivo, and Western blotting produces bands only if transfected cells are treated with tunicamycin (5 µg/ml).

HPLC analysis of noradrenaline and dopamine. E12.5 embryos and placentas with a wild-type genotype and an Orct3-null genotype were obtained from heterozygous matings. The samples were prepared as described previously (32). Noradrenaline and dopamine concentrations were measured by high-performance liquid chromatography (HPLC) (Gynkotec; Separations) with electrochemical detection (Antec Leyden) using 3,4-dihydroxybenzylamine as an internal standard. Protein concentrations were determined as described previously (32), and noradrenaline and dopamine concentrations were expressed per milligram of protein.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Orct3 gene targeting. To inactivate the Orct3 gene, a two-step targeting approach was used. First, a 4.3-kb genomic BamHI fragment that contains the complete first exon of the Orct3 gene and 650 bp of upstream sequences was replaced with a pGKneo/pGKtk selection cassette flanked by loxP sites (Fig. 1A). Successful homologous recombination was monitored by Southern blotting with a 3' external probe, detecting a 17-kb wild-type (+/+) BglII fragment and a 9.5-kb fragment of the recombined allele (+/neo [Fig. 1B]). Subsequently, four independently targeted clones were transiently transfected with a CRE-recombinase expression vector to remove the pGKneo/pGKtk selection cassette. CRE-mediated recombination resulted in an Orct3 knockout allele (+/) that could be identified as a 12-kb BglII fragment (Fig. 1C). Thus, the resulting Orct3 knockout allele lacked exon 1 and contained a single loxP site replacing a 4.3-kb genomic BamHI fragment (Fig. 1A).

The Orct3 targeted allele is a complete null. It was anticipated that removal of the first exon and 650 bp of upstream sequences would result in the complete absence of Orct3 gene expression. To test this, RNA was isolated from various organs of adult wild-type and Orct3 homozygous knockout mice and analyzed by Northern blotting (Fig. 2A). Wild-type mice show high expression in heart and skeletal muscle. No Orct3 expression was detected in these tissues of homozygous mutant mice. Also, Orct3 expression was absent in brain tissue, which shows low levels of expression in wild-type mice. In placentas of homozygous mutant embryos, no Orct3 expression was detected at 11.5 days of gestation (Fig. 2B). At later time points, however, expression of an aberrant Orct3 transcript was observed. Figure 2B shows expression of this aberrant transcript, which is approximately 700 bp shorter than the wild-type transcript, in 12.5- to 17.5-dpc placentas. Reverse transcription-PCR with different sets of intron-spanning Orct3 oligonucleotide primers confirmed the presence of an Orct3 transcript in the homozygous mutant placentas and showed that exons 3 to 11 of the Orct3 gene were contained within this mRNA (data not shown and map in Fig. 3A). A 5' rapid-amplification-of-cDNA-ends experiment was performed with Orct3-specific oligonucleotides on homozygous mutant placenta RNA. A unique 158-bp sequence that is not present in the wild-type Orct3 transcript was identified. This sequence was mapped to intron 2 of the Orct3 gene and contains a consensus splice donor sequence by which it splices to exon 3 of the Orct3 gene (data not shown). The 158-bp sequence does not reintroduce an ATG translation start codon, which was removed with exon 1, into the mutant transcript. The first in-frame ATG codon is within the fourth transmembrane domain, and if it were used, the predicted protein would contain only 8 of the 12 transmembrane domains of the wild-type protein. Transient overexpression of the mutant RNA in transfected cell lines confirmed that the short transcript is incapable of generating a protein product (Fig. 3). These results show that the targeting strategy resulted in the complete absence of Orct3 gene expression in adult mice and in early placentas, whereas at later stages in embryonic development an aberrant, noncoding Orct3 mRNA was expressed in the placenta.

View larger version (61K): [in this window] [in a new window]   FIG. 2.   Gene expression analysis in Orct3 mutant mice. (A) Northern blot analysis of Orct3 expression in different organs of wild-type and homozygous mutant adult mice. GAPDH was used as a loading control. (B) Orct3 expression in wild-type and homozygous mutant mouse placentas at different stages of development. The 3.5-kb wild-type and 2.8-kb aberrant transcripts are indicated. Pai-1 hybridization was used as a loading control.

View larger version (42K): [in this window] [in a new window]   FIG. 3.   The knockout (KO)-specific 2.8-kb transcript is not translated. (A) Exon map of the knockout-specific RNA 2.8-kb transcript and the wild-type (wt) mRNA. The 2.8-kb transcript lacks exons 1 and 2 but contains a novel exon (gray box labeled 1') from intron 2 spliced onto exons 3 to 11 (there are 11 exons in wild-type Orct3). The first in-frame ATG codon is within the fourth transmembrane domain, and if it were used, the predicted protein would contain only 8 of the 12 transmembrane domains of the wild-type protein. The position of the 4.3-kb deleted sequence that spans the Orct3 promoter is indicated by the dotted line. (B) RNase protection assay (RPA) of transiently transfected human embryonic kidney 293 cells. Four constructs were used: Orct3, wild-type Orct3/Slc22a3 cDNA; Orct3myc, wild-type Orct3 cDNA with a Myc tag inserted into a HindIII site located six codons before the translation stop; Orct3KO, the knockout-specific cDNA; and Orct3KOmyc, the knockout-specific cDNA containing a Myc tag inserted into a HindIII site located six codons before the translation stop. All constructs were driven by a cytomegalovirus promoter and enhancer and generated abundant RNA. 13.5 dpc, wild-type placental RNA that serves as a control for Orct3 production; nt, nontransfected control cells. (C) Western blotting using antiserum raised to an Orct3 peptide as described in Materials and Methods. Only the wild-type cDNA is translated; the wild-type Orct3 Myc-tagged protein cannot be recognized by the Orct3 antiserum, because the Myc tag is inserted in the epitope recognized by the antiserum. (D) Western blotting of the same samples using an anti-Myc antiserum; only the wild-type Myc-tagged protein is recognized.

Orct3-null mice are viable and show no obvious phenotype. Mice heterozygous for the two independent Orct3 null alleles were bred to establish two independent Orct3 knockout mouse lines that behaved similarly in all tests (see Materials and Methods). Mice homozygous for the targeted allele appeared in normal Mendelian ratios (data not shown), indicating that Orct3 is dispensable for embryonic development. Orct3-null mice appear normal in stature and have a normal life span (the oldest mice in the colony are now 15 to 16 months). Both male and female Orct3-null mice are fertile with normal breeding behavior. In addition, female mice show normal maternal nurturing behavior and have reared the same sizes of litters as did wild-type mice. Finally, the Orct3-null mice show no abnormal behavior under routine housing and handling, indicating a degree of tolerance to normal stress. Histological examination of Orct3^/ placentas, which are the highest Orct3-expressing organs at any stage, and of the heart, which is the highest Orct3-expressing organ in adult mice, similarly revealed no cellular or structural alterations (data not shown). The heart in Orct3-null mice was of a normal size range, color, and appearance. In addition, the weight of the heart in Orct3-null mice was unchanged from that of wild-type mice (Table 1).

Impaired uptake-2 activity in adult Orct3-deficient mice. To study uptake-2 in Orct3-deficient mice, we designed a protocol measuring uptake-2-mediated accumulation of the neurotoxin MPP⁺ after intravenous injection. MPP⁺ has been described elsewhere as a good uptake-2 substrate and is not subject to metabolism in vivo, in contrast to the monoamines (22, 25). Monoamines are converted into antagonistic metabolites for uptake-2, which complicates their use as substrates in the analysis of uptake-2 activity in vivo (33). To measure a maximal effect in primary uptake, we first determined the temporal curve of [³H]MPP⁺ accumulation. Wild-type FVB/N female mice were injected intravenously with 1.0-mg/kg [³H]MPP⁺, and MPP⁺ concentrations in the heart, liver, and plasma were determined 5, 10, 15, and 60 min after injection (Fig. 4). In both heart and liver, an increase in MPP⁺ concentration was seen up to 15 min. At later time points, the MPP⁺ levels decreased in both organs, following the rapid decline in MPP⁺ plasma concentrations (Fig. 4). Based on these data in subsequent studies, the primary uptake of [³H]MPP⁺ was determined 10 min after intravenous injection.

View larger version (19K): [in this window] [in a new window]   FIG. 4.   Time curve of MPP+ accumulation in heart and liver tissues of wild-type female mice. Levels of MPP+ in the heart () and liver () (plotted on the primary y axis in nanograms per gram of tissue) are indicated at different time points after intravenous injection. The MPP+ plasma levels are shown by the dashed line () and plotted on the secondary y axis in nanograms per milliliter.

View larger version (37K): [in this window] [in a new window]   FIG. 5.   MPP+ transport in adult Orct3-deficient mice. (A and B) Concentrations of MPP+ in the heart and liver (plotted on the primary y axis in nanograms per gram of tissue) of adult wild-type (black bars) and Orct3 mutant (white bars) females (A) and males (B). The levels in plasma are also shown and are plotted on the secondary y axis in nanograms per milliliter. The asterisk indicates a statistically significant difference. P values were <0.0001 for both male and female hearts. (C and D) MPP+ distribution in different organs of adult wild-type (black bars) and Orct3 mutant (white bars) female (C) and male (D) mice. small int., small intestine.

Orct3 transports MPP⁺ at the fetoplacental interface. We next addressed the role of Orct3 in the placenta. Pregnant females of an Orct3 heterozygous cross were injected at 15.5 dpc with 1.0 mg of [³H]MPP⁺ per kg. The MPP⁺ concentrations were determined in embryos, placentas, and amniotic fluid 10 min after injection (Fig. 6). The highest levels of MPP⁺ were detected in the placenta, intermediate levels were detected in the embryo, and the lowest levels were detected in the amniotic fluid. In embryos, a threefold reduction in MPP⁺ accumulation, from 64.7 ± 22.7 ng/g in wild-type mice to 20.4 ± 4.4 ng/g in mutants, was detected (P < 0.001). However, no differences in [³H]MPP⁺ accumulation were found in placentas and amniotic fluid of both groups. Since Orct3 is not expressed in embryonic tissue, these data indicate that Orct3 is the rate-limiting step in MPP⁺ transport from the placenta to the fetus but does not play a major role in placental uptake from the maternal circulation. These results identify the placenta as a novel uptake-2 site of action.

View larger version (26K): [in this window] [in a new window]   FIG. 6.   Placental MPP+ transport. MPP+ distribution in wild-type (black bars) and Orct3 mutant (white bars) placentas (plotted on the primary y axis) and embryos and amniotic fluid (plotted on the secondary y axis). The asterisk indicates a statistically significant difference. The P value was <0.001 for embryos.

Noradrenaline and dopamine levels in Orct3-null mice. Embryos and placentas at 12.5 dpc with a wild-type and an Orct3-null genotype were obtained from a heterozygous mating. Noradrenaline and dopamine steady-state levels were measured by HPLC. Table 2 shows that, although both noradrenaline and dopamine levels are reduced by approximately 50% in Orct3-null embryos, the results are not statistically significant (P = 0.1015). Samples with the same genotype and obtained from the same litter showed a very large variation in noradrenaline and dopamine steady-state levels for which there is no current explanation. An independent examination of monoamine levels in heart tissue has also found a large variation among animals with the same genotype, with no significant difference between wild-type and Orct3-null mice (B. Giros and S. Gautron, personal communication)

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In this report, we have described the targeted inactivation of the mouse extraneuronal monoamine transporter gene Orct3, which has been shown to function as uptake-2. Removal of exon 1 abolished transcription in tissues of adult mice but resulted in an aberrant noncoding transcript in the placenta. We have analyzed uptake-2 activity in Orct3-null mice using the neurotoxin MPP⁺ as a substrate and identified a transport phenotype in the adult heart and in the embryonic placenta. Despite the large differences in MPP⁺ transport in embryos and adult hearts, an overt physiological or neural phenotype in Orct3 mutant mice that may be associated with altered extracellular monoamine concentrations was not observed. In addition, no significant differences in the steady-state levels of noradrenaline and dopamine could be detected in either embryos or placentas.

Targeted inactivation of the Orct3 gene was performed by deletion of a 4.3-kb genomic fragment that contains the Orct3 gene promoter and exon 1. The Orct3 promoter consists of two closely linked transcriptional start sites that were mapped 150 and 170 bp upstream of the beginning of the CpG island in which exon 1 is embedded and were fully removed by the deletion (data not shown). Orct3 expression was completely abolished in all tissues of adult homozygous mutant mice. Later-stage placentas from 12.5 dpc to term, however, express an aberrant, noncoding transcript. This transcript is not driven by the wild-type Orct3 gene promoter, as it was fully included in the targeted deletion. These data indicate that deletion of the wild-type transcriptional start sites activated an ectopic promoter sequence to drive expression of the aberrant transcript. Alternatively, deletion of the wild-type promoter may have fortuitously joined together a DNA sequence that can drive expression of the aberrant transcript. The expression, however, of the aberrant nontranslated transcript is tissue specific and temporally restricted in development.

To study the role of Orct3 in uptake-2 function in vivo, we performed transport studies using the neurotoxin MPP⁺ as a substrate. MPP⁺ has been described elsewhere in the etiology of 1,2,3,6-methylphenyltetrahydropyridine (MPTP)-induced Parkinson's disease (reviewed in reference 10). To exert its neurotoxic effect on the central catecholaminergic neurons, MPTP requires conversion to MPP⁺ by monoamine oxidase B (Maob) in glial cells. Subsequently, MPP⁺ has to leave the glial cells, upon which it is transported into the catecholaminergic neurons by the neuronal dopamine uptake system. Previously, it has been shown by in vitro cell culture that MPP⁺ is able to make use of the extraneuronal monoamine transport system to exit human glial cells (24). In this report, we have shown that Orct3 transports MPP⁺ in vivo in mice, raising the possibility that Orct3 is an important mediator of MPTP neurotoxicity. In addition to the effect in the central nervous system, it has been reported that systemic injection of MPP⁺ or MPTP results in severe depletion of heart noradrenaline concentrations (12). The cardiac depletion is resistant to desipramine and GBR12909, suggesting that there is no involvement of the uptake-1 transporters Net and Dat (13, 20). Our transport studies show reduced MPP⁺ uptake in hearts of Orct3-deficient mice, which suggests that Orct3 might be involved in the action of MPP⁺-MPTP-mediated depletion of cardiac noradrenaline.

Uptake-2 has been identified in various organs, like the heart and uterus. Both organs express high levels of Orct3. However, our results identified a major effect of Orct3 deficiency on MPP⁺ uptake in only adult hearts. MPP⁺ accumulation was reduced by 72% in Orct3 mutant hearts in males and females. This reduction is identical, despite the fact that nearly twofold-lower levels of MPP⁺ were measured in wild-type males than in the females (Fig. 5A and B). This indicates that the relative contribution of uptake-2 in MPP⁺ transport is not different between males and females. A second difference in the amount of MPP⁺ accumulation was found in hearts of inbred FVB/N females and wild-type females of a mixed genetic background (Fig. 4 and 5A). The difference in MPP⁺ accumulation might be explained by a differential contribution of different genetic backgrounds. However, there is also an age difference between the two cohorts (7 and 12 weeks, respectively), which may be an important contributor. In the uterus, a nearly twofold reduction in MPP⁺ accumulation was observed between wild-type and Orct3^/ females, but this was not statistically significant (Fig. 5C). In pigs, it was demonstrated that uptake-1 and uptake-2 activity in the uterine artery vary during the estrous cycle (7). Since the mice in our experiments were not synchronized, this could explain the large variation seen for the uterus that may potentially mask an effect of Orct3 deficiency. Other Orct3-expressing organs showed no difference in MPP⁺ uptake. This might be caused by a functional redundancy with other transporter genes. Orct3 is only moderately expressed in the kidney, for instance, whereas the closely related organic cation transporter gene Orct2 is found at high levels (21). As Orct2 is also capable of MPP⁺ transport, expression of this gene could contribute significantly to the total MPP⁺ accumulation in the kidney (17). Similarly, the Orct1 gene is highly expressed in the liver, which may account for the MPP⁺ uptake measured in this tissue (Fig. 4 and 5).

The placental MPP⁺ transport studies show that Orct3 transports MPP⁺ between the placenta and fetus, but not to the maternal circulation, and identify the placenta as a novel uptake-2 site of action (Fig. 6). It has been reported previously that during development embryos show a high monoamine turnover compared with any other physiological condition seen for adults (30). For sheep, it has been determined that nearly 50% of the total in utero monoamine clearance was mediated by the placenta. Only part of the placental activity was assigned to the action of neuronal monoamine transporters, as was determined by inhibition mediated by cocaine, which is a neuronal monoamine transporter antagonist (5). Using RNA in situ hybridization, we have recently described the cellular expression pattern of Orct3 and the monoamine-degrading enzyme gene Maoa in the mouse placenta. The two genes are expressed in a similar pattern in the labyrinth layer (34), indicating that they could form a functional monoamine degradation pathway. This interpretation is supported by the transport studies presented here that have identified an Orct3-mediated uptake-2 activity at the fetoplacental interface.

The MPP⁺ transport studies show that Orct3 is an essential component in vivo for the transport activity of the extraneuronal monoamine clearance system known as uptake-2_. Transport defects in Orct3-deficient mice were observed in embryonic development. However, despite a broad expression pattern in adult animals, an essential function is seen in only one adult organ, the heart. Surprisingly, despite significant differences in MPP⁺ uptake, the Orct3-null mice show no overt neural or physiological dysfunction as embryos or adults that may indicate a monoamine imbalance. In addition, we have been unable to identify a significant difference in the levels of two tested monoamines in Orct3-null embryos that show a 65% reduction in MPP⁺ levels. Thus, the functional significance of the role of the Orct3 gene in monoamine transport remains unclear. These results do not dismiss a function for the Orct3 gene in extraneural monoamine transport, but they indicate the possibility that its role is more complex than has been predicted.