Molecular identification of a novel carnitine transporter specific to human testis. Insights into the mechanism of carnitine recognition.

l-Carnitine is an essential component of mitochondrial fatty acid beta-oxidation and plays a pivotal role in the maturation of spermatozoa within the male reproductive tract. Epididymal plasma contains the highest levels of l-carnitine found in the human body, and initiation of sperm motility occurs in parallel to l-carnitine increase in the epididymal lumen. Using a specific carrier, epididymal epithelium secretes l-carnitine into the lumen by an active transport mechanism; however, the structure-activity relationship comprising the carnitine-permeation pathway is poorly understood. We discovered a novel carnitine transporter (CT2) specifically located in human testis. Analyzing the primary structure of CT2 revealed that it is phylogenetically located between the organic cation transporter (OCT/OCTN) and anion transporter (OAT) families. Hence, CT2 represents a novel transporter family. When expressed in Xenopus oocytes, CT2 mediates the high affinity transport of l-carnitine but does not accept mainstream OCT/OCTN cationic or OAT anionic substrates. We synthesized and tested various carnitine-related compounds and investigated the physicochemical properties of substrate recognition by semi-empirical computational chemistry. The data suggest that the quaternary ammonium cation bulkiness and relative hydrophobicity be the most important factors that trigger CT2-substrate interactions. Immunohistochemistry showed that the CT2 protein is located in the luminal membrane of epididymal epithelium and within the Sertoli cells of the testis. The identification of CT2 represents an interesting evolutionary link between OCT/OCTNs and OATs, as well as provides us with an important insight into the maturation of human spermatozoa.

L-Carnitine ([-],␤-hydroxy-␥-N-trimethylaminobutyric acid) is a highly polar, water-soluble quaternary amine that exists as a zwitterion under physiological conditions. In eukaryotic cells, L-carnitine is biologically important for mitochondrial ␤-oxidation of long-chain fatty acids and to generate ATP (1). In eukaryotic metabolism, a major function of L-carnitine is to promote the translocation of fatty acids across the mitochondrial inner membrane as ␤-hydroxyl O-acyl esters of L-carnitine.
Carnitine has been linked to the regulation of spermatozoa motility in several mammalian species. L-Carnitine is secreted from mammalian epithelium into epididymal plasma and ultimately into spermatozoa, where it accumulates as free and acetylated L-carnitine (2,3). The concentration of L-carnitine in epididymal plasma and spermatozoa range as high as 2-100 mM, 2000-fold higher than circulating blood levels (10 -50 M). Previous studies have shown that the role of L-carnitine in the epididymis is to foster fertilizing and maturation of spermatozoa (2,4). Spermatozoa produced in testis are immobile and infertile and must undergo postgonadal modification in the epididymis to become fertile (2). Within the epididymal lumen, it has been shown that the initiation of sperm motility occurs in parallel with increase in L-carnitine (2). Regulated by androgens, previous studies using micropuncture and in vitro microperfusion demonstrate that the epididymal epithelium secrete L-carnitine into the epididymis lumen by a saturable process (2,5,6). However, the molecular nature of the carnitine-specific transporter(s) in human epididymis is poorly understood.
Molecular cloning techniques have led to considerable progress in the identification of various organic ion transporters. Organic ion transporters, located in various tissues, play important roles in the transport of a wide variety of structurally diverse xeno-and endobiotics. Koepsell et al. (8) and others identified the multispecific organic cation transporter (OCT) 1 family. The OCT family mediates the excretion of diverse cationic substrates in tissues such as the kidney, intestine, placenta, and liver (7,8). Successful cloning followed the discovery of the OCT family. We, and others, have identified a multispecific organic anion transporter (OAT) family that mediates the transport of organic anions in a variety of tissues, i.e. OAT1 (kidney and brain) (9,10), OAT2 (liver) (11), OAT3 (kidney and brain) (12), OAT4 (kidney and placenta) (13), and a urate transporter (URAT1) located in the kidney (14). As members of the OCT family, the novel organic cation transporter (OCTN) family has been described (OCTN1 (Ref. 15), OCTN2 (Ref. 16), and OCTN3 (Ref. 17)). OCTNs are structurally similar to the OCT family and possess the ability to transport a variety of prototypical organic cations, such as tetraethylammonium cation (TEA), 1-methylphenyl-1,2,3,6-tetrahydropyridine, and choline.
Ubiquitously expressed in a wide variety of tissues, OCTN2 (16) is a member of the OCT/OCTN family and was initially described as accepting an assortment of organic cations; however, a study by Tamai et al. (18) has revealed that human OCTN2 mediates Na ϩ -dependent transport of L-carnitine. About the same time, we isolated OCTN2 cDNA from rat intestine (19) and found it to mediate the high affinity transport of L-carnitine. Consequently, we made a proposal for designating this transporter as "CT1" (for carnitine transporter 1) (19). Wu et al. (20) suggested that OCTN2 (CT1) mediate organic cations without involving Na ϩ , whereas OCTN2 (CT1) could accept L-carnitine but only in the presence of extracellular Na ϩ . However, a more rigorous investigation regarding the exact mechanism(s) of recognition for cationic and zwitterionic substrates has remained somewhat vague.
We have identified and characterized a novel carnitine transporter, expressed exclusively in human testis, that we call CT2 (carnitine transporter 2). Compared with OCT/OCTN and OAT family members that have broad substrate specificity, CT2 has been unique because it displays substrate selectivity. We determined the subcellular localization of CT2 by immunohistochemistry. The CT2 protein was localized in the luminal membrane of human epididymis and supports the idea that CT2 constitutes an L-carnitine secretion mechanism from epididymal epithelium into the lumen. Therefore, the function of CT2 may be relevant in the maturation of human spermatozoa. At the level of amino acid sequence, CT2 is distinct from OCT/ OCTN and OAT family members. We suggest that CT2 be a novel organic ion transporter family from which the OCT/ OCTNs and OATs evolved. Our experimental and semi-empirical computational analyses (for a summary of structure-activity relationship descriptors, see Ref. 44) provide insight into the mechanism of carnitine recognition by CT2, and we have compared CT2 versus OCTN2 (CT1) substrate specificity.

Cloning of Human CT2
A human expressed sequence tag (EST) from human testis cDNA libraries (GenBank TM /EBML/DDBJ accession no. AA778598) showing nucleotide sequence similarity to rat OAT1 (9, 10) was obtained from the Integrated and Molecular Analysis of Genomes and their Expression (IMAGE) data base (cDNA clone number 1048962). The [ 32 P]dCTPlabeled probe was synthesized from the clone (T7 Quick Prime, Amersham Biosciences) and was used to screen a human testis cDNA library. As described previously, a nondirectional cDNA library was prepared from the human testis poly(A) ϩ RNA (CLONTECH) with the Superscript TM Choice system (Invitrogen) (9,13). The cDNAs were ligated into ZipLox EcoRI arms (Invitrogen). Replicated filters of the phage library were hybridized overnight at 37°C in a hybridization solution (50% formamide, 5ϫ standard saline citrate (SSC) (1ϫ SSC ϭ 0.15 M NaCl and 0.015 M sodium citrate), 3ϫ Denhardt's solution, 0.2% SDS, 10% dextran sulfate, 0.2 mg/ml denatured salmon sperm DNA, 2.5 mM sodium pyrophosphate, 25 mM MES, and 0.01% Antifoam B, pH 6.5), and washed at 37°C in 0.1ϫ SSC and 0.1% SDS. The cDNA inserts (CT2) in positive ZipLox phages were recovered in the expression vector, pZL1 (Invitrogen), by in vivo excision.

Sequence Analyses
Sequence analyses and determination of genomic organization were performed with web-based data base searches through the National Center for Biotechnology Information (www.ncbi.nim.nih.gov) and the DNASIS programs (Hitachi Software Engineering). Multiple sequence alignment and topology prediction were done with the DNASIS program. The phylogenetic analyses were performed with the CLUSTAL program (clustalw.genome.ad.jp/) and displayed with the TreeView drawtree program (taxonomy.zoology.gla.ac.uk/rod/treeview.html).

Functional Expression of CT2 in Xenopus Oocytes
As previously described, complementary RNA (cRNA) synthesis and uptake measurements were performed (9). Briefly, the cDNA was linearized with SpeI, and the cDNA insert transcribed in vitro with T7 RNA polymerase (Stratagene) in the presence of RNA cap analog (Amersham Biosciences). The resultant cRNA was purified by multiple phenol/chloroform extractions and precipitated with ethanol.
The kinetic parameters for the uptake of L-carnitine via CT2 were estimated with the following equation where v is the uptake rate of L-carnitine (picomol/oocyte/h), [S] is the substrate concentration (M) in the medium, and K m is the Michaelis-Menten constant (M). Inhibition constants (K i ) were calculated as previously described (19).
In Na ϩ substitution experiments, Na ϩ in ND96 bath was replaced iso-osmotically with equimolar Li ϩ or N-methyl-D-glucamine (NMDG). For the inhibition study, uptake rate of 50 nM [ 3 H]L-carnitine by oocytes injected with water or CT2 cRNA were measured for 1 h in the absence or presence of 5 or 50 M test compound in ND96 solution.
To examine trans-stimulatory effects on the efflux of radiolabeled substrates, CT2 expressed oocytes were incubated with [ 3 H]L-carnitine (50 nM) for 90 min and transferred into the medium with or without unlabeled L-carnitine. The radioactivity in the medium and within the oocytes was determined after the 90-min incubation period.

Computational Analyses
A Compaq Deskpro EN computer with a Pentium III processor was used. Chemical structures were drawn with CS Chem-Draw Ultra ® (version 6.0.1, Cambridge Soft Corp.) and copied into CS Chem3D Ultra ® (version 6.0, Cambridge Soft Corp.). For each molecule, a molecular mechanics minimization was performed with a root-meansquare of 0.001. Gaussian 98W ® (version 5.4, revision A.9) was installed, and molecular analyses were used to compute, at the semiempirical Austin Model 1 level of theory set at "tight convergence setting," optimized structures (25). Subsequently, optimization using Hartree-Fock/321G was performed set at "tight convergence" and theoretical atomic charges, electron density, and dipoles (Debye) were computed. Afterward, the property server (44) was used to compute ClogP and the Connolly accessible area (Å 2 ).

Northern Blot Analysis
A human 16-lane, Multiple-Tissue Northern (MTN TM ) blot (CLONTECH) was used for the Northern blot analysis of CT2. We used a [ 32 P]dCTP-labeled CT2 cDNA fragment (877-1317) as the probe. According to the instructions from the manufacturer, the master blot filter was hybridized with the probe for 1 h at 65°C. The filter was finally washed under a highly stringent condition (0.1ϫ SSC and 0.1% SDS at 65°C).

Immunohistochemistry
Corresponding to the 14 amino acids of the COOH terminus of CT2, we generated a rabbit anti-CT2 polyclonal antibody raised against a keyhole limpet hemocyanin-conjugated synthesized peptide, KTEAIT-PRDSGLGE. The human testis or epididymal tissues were obtained from a patient who had died from bronchial asthma. As described previously, paraffin sections (3 m) were processed for light microscopic immunohistochemical analysis (14). Regarding the absorption experiments, the tissue sections were treated with the primary antibody in the presence of antigen peptides (50 g/ml).

Statistical Analysis
The experiments were performed using three different batches of oocytes, and the results from the experiments are expressed as mean Ϯ S.E. Statistical significance was judged from Student's t tests. Differences were considered significant at a level of p Ͻ 0.05.

RESULTS
Structural Features of Human CT2-An EST data base search identified an EST (GenBank™/EMBL/DDBJ accession number AA778598) exhibiting sequence similarity to rat OAT1 (9,10). Corresponding to the EST as our probe, a human testis cDNA library was screened with a cDNA fragment. As a result, a cDNA encoding a novel carnitine transporter (designated as CT2) was isolated. CT2 cDNA consisted of 2048 base pairs, had an open reading frame of 1632 base pairs, and included the termination codon. The open reading frame was flanked by a 284-bp-long 5Ј-noncoding sequence and a 132-bp-long 3Ј-noncoding sequence. The CT2 cDNA encoded a 543-amino acid protein. Fig. 1a shows the deduced amino acid sequence of CT2 aligned with those of human OAT1 (26,27), OCT1 (28), and OCTN2 (CT1) (16). The amino acid sequence of CT2 shows 36, 38, and 37% identities to human OAT1, OCT1, and OCTN2 (CT1), respectively. Kyte-Doolittle hydropathy plot analysis (29) predicted 12 membrane-spanning domains in CT2 (Fig. 1,  a and b), which are similar to other organic ion transporter family members. Analogously modeled to known mammalian transporters with 12 transmembrane domains, the CT2 amino and carboxyl termini were oriented toward the cytoplasmic side of the membrane. N-Glycosylation sites (residue 74) and protein kinase C-dependent phosphorylation sites (residues 14, 142, 285, 322, 516, and 535) were also predicted in the CT2 sequence (Fig. 1a). Phylogenetic analysis with the CLUSTAL program revealed CT2 to exhibit remote similarity to each member of OCT/OCTNs and OATs (Fig. 2). Hence, CT2 could not be assigned to any of the known families and represents a novel organic solute transporter class.
Genomic Organization of Human CT2 Gene-Using the CT2 cDNA nucleotide, a human genome data base (30) search revealed that the human gene encoding this transporter has been entirely sequenced. The gene is ϳ50 kbp long and located on chromosome 6q21-22.1. The location was confirmed by fluorescence in situ hybridization using CT2 cDNA as a probe (data not shown). Using the cloned CT2 cDNA with the reported genomic sequence, alignment of the nucleotide sequence enabled us to deduce the exon-intron gene organization. The CT2 gene consists of 10 exons and 9 introns (Fig. 3). The size of each exon, intron, and nucleotide sequence of the splice junctions is given in Table I. The 5Ј and 3Ј termini for each intron possess the consensus sequence for RNA splicing (gt/ag). The translation start site ATG is present in exon 2, and the translation termination site TAA is present in exon 10 (Fig. 3). Exon 1 does not code for the protein.
When extracellular Na ϩ was replaced with Li ϩ or NMDG at equimolar concentration, the uptake was reduced to ϳ50% (Fig. 4e), suggesting that the interaction of carnitine with CT2 is not completely dependent on extracellular Na ϩ . To determine the role of extracellular pH on carnitine transport, the uptake of [ 3 H]L-carnitine was measured with solutions of various pH (Fig. 4f). The rate of CT2-specific transport gradually increased as the pH was increased from 5.5 to 8.5. The transport rate for L-carnitine increased almost 2-fold between pH 5.5 and 8.5. Next, to determine whether CT2 was an exchanger or a facilitated transporter, the efflux of radioactivity from oocytes preloaded with 50 nM [ 3 H]L-carnitine was measured in the absence or presence of extracellular L-carnitine (0.5 and 5 M) (Fig. 4g). Radioactive efflux, essentially 40%, was detected for 60 min in the absence of extracellular L-carnitine from oocytes expressing CT2. The efflux result demonstrates that CT2 mediates bidirectional permeation of carnitine across the plasma membrane. The efflux was not trans-stimulated by extracellular L-carnitine, supporting the notion that the CT2-mediated carnitine permeation occurs by facilitated diffusion but not by an exchange mechanism like OCT1 (7), OAT1 (9, 10), and URAT1 (14).
Substrate Selectivity of Human CT2-To test CT2 substrate specificity, we performed competition experiments. We used various unlabeled carnitine-related molecules to test for competition with [ 3 H]L-carnitine uptake (Fig. 5). Distinctly different from OCT/OCTNs and OATs, CT2 showed unique substrate selectivity. Including structural analogs of L-carnitine, the uptake of 50 nM [ 3 H]L-carnitine by CT2 was measured in the presence or absence of non-radiolabeled compounds (5 or 50 M). As shown in Fig. 5, acetyl-L-carnitine, acetyl-DL-carnitine, octanoyl-L-carnitine, and betaine significantly inhibit CT2-mediated [ 3 H]L-carnitine uptake. In contrast, lysine, leucine, glycine, choline, taurine, GABA, and trimethyllysine did not produce any noticeable effects. Among the precursors of L-carnitine synthesis, only betaine showed an inhibitory effect on CT2-mediated L-carnitine uptake. Fig. 5 illustrates that CT2 failed to accept TEA or choline, which are prototypical substrates for OCT/OCTNs, and para-aminohippurate and estrone sulfate, which are typical substrates for OATs. Our results demonstrate that the CT2-mediated transport process does not encompass broad substrate specificity.
We tested whether or not CT2 could accept various OCTN2 (CT1) substrates (16,31). As shown in Fig. 6, L-carnitine uptake by CT2 was not inhibited by cationic xenobiotics such as cephaloridine, procainamide, desipramine, and quinidine. Although both transporters are able to mediate L-carnitine transport, the results show that the property of substrate recognition by CT2 is distinctly different from OCTN2 (CT1).
Examining the mechanism(s) of substrate recognition by CT2, we synthesized various L-carnitine analogous compounds (see "Experimental Procedures"). We tested the ability of carnitine-related compounds (20 M) to compete with [ 3 H]L-carnitine (20 nM) for the transport process by CT2 (Fig. 7). In addition, to compare the mechanisms of substrate recognition between CT2 and OCTN2 (CT1), we expressed rat OCTN2 (CT1) (19) in oocytes and performed the same inhibition study. As shown in Fig. 7, among 20 synthetic compounds, 1-5A, 7A-10A, and 13 possessed inhibitory effects on both OCTN2 (CT1)-and CT2-mediated carnitine transport. The inhibition of carnitine transport caused by these compounds was in the range of 15-40%. Furthermore, compounds 6A, 11A, 12A, 17, and 18 showed inhibitory effect on carnitine uptake by CT2 but not on OCTN2 (CT1). The inhibitory potency of 6A on CT2mediated transport was less than 11A, 12A, 17, and 18, although it was statistically significant. These results help to reveal the subtle physicochemical differences between substrate recognition of CT2 and OCTN2 (CT1).
Tissue Distribution of Human CT2-The expression of CT2 messenger RNA (mRNA) in human tissues was investigated by high stringent Northern blot analysis (Fig. 8). The CT2-specific mRNA was exclusively detected in human testis and not in heart, brain, placenta, lung, liver, skeletal muscle, kidney, pancreas, spleen, thymus, prostate, ovary, small intestine, colon, or peripheral blood leukocytes. Two transcript sizes (2.1 and 1.8 kb) were identified in the human testis.
Immunohistochemical Localization of CT2 in Human Testis and Epididymis-For the determination of localization in adult human tissues, we generated a rabbit polyclonal antibody raised against COOH terminus of CT2. Using immunohistochemistry techniques, the CT2 protein was detected in Sertoli cells (Fig. 9, a and b); however, the CT2 protein was not detected in germinal cells or in the interstitial tissues such as Leydig cells, blood vessels, and lymph vessels. In the epididymis, CT2 immunoreactivity was prominent in the epithelial cells of the epididymal tubules (Fig. 9d). In a high magnification view (Fig. 9e), CT2 protein was detected in the luminal membrane of principal cells but not in epithelial cell stereocilium or basal cells of epididymal epithelium. Anatomically, the epididymis is divided into three regions: the caput, the corpus, and the cauda epididymis (32). CT2 immunoreactivity was observed mainly in the caput and corpus regions, whereas CT2 was barely detected in the cauda epididymis regions. CT2 immunoreactivity was faintly present in the luminal membrane of the ductus efference, the rete testis, and spermatozoa (data not shown). In absorption experiments where the tissue sections were treated with primary antibodies in the presence of antigen peptides (50 g/ml), immunostaining was not observed and confirms the specificity of the immunoreactions (Fig. 9, c and f). DISCUSSION We describe a testis-specific carnitine transporter (CT2) representing an evolutionary change that shares only 34 -37% homology with the other OCT/OCTN and OAT families (Fig. 2). Compared with the multispecific organic ion transporter family members, CT2 has restricted substrate selectivity. CT2 is capable of accepting L-carnitine and analogous compounds but not mainstream OCT/OCTN and OAT substrates (Figs. 5 and 6). Hence, CT2 is unique to other carnitine transporters, OCTN2 (CT1) (16,18,19) and OCTN3 (17), which function as multispecific organic cation transporters. Furthermore, in contrast to OCTN2 (CT1) (17,19,20,33), CT2 accepts zwitterions without Na ϩ suggesting a difference in OCTN2 (CT1) and CT2 substrate recognition mechanism(s).
L-Carnitine and carnitine-related compounds are physiological substrates for CT2 (Fig. 5). L-Carnitine is a ␤-hydroxy ␥-butyrobetaine and possesses a hydroxyl group at the ␤-carbon, a trimethylammonium moiety at the ␥-carbon, and a carboxylic acid at the ␣-carbon. Consequently, L-carnitine exists as a zwitterion at physiological pH (1). In our present studies, CT2-mediated L-carnitine uptake was inhibited by D-carnitine, acetyl-L-carnitine, octanoyl-L-carnitine, and betaine (a precursor of carnitine biosynthesis) (Fig. 5). Hence, compounds containing two or three carbons between the carbonyl moiety and the positively charged quaternary ammonium nitrogen can inhibit L-carnitine uptake. In contrast, trimethyllysine and lysine, an intermediate and a precursor of L-carnitine biosynthesis, respectively, had no apparent inhibitory effect on CT2mediated carnitine uptake. Trimethyllysine and lysine have a carboxylic anion and cationic nitrogen separated by four methylene (-CH 2 -) units; at physiological pH, trimethyllysine and lysine have molecular formal charges of ϩ1. GABA, like carnitine, is a butanoic acid; however, GABA does not contain a permanent quaternary ammonium cation. On the other hand, choline is a mainstream OCT/OCTN substrate with a trimethyl quaternary ammonium cation but does not possess a carbonyl moiety. Choline has a molecular formal charge of ϩ1 and did not inhibit carnitine uptake with CT2. Hence, the data support the notion that CT2 substrate recognition requires the presence of a carbonyl and quaternary ammonium nitrogen separated by three or fewer carbons.
The results summarized in Fig. 7 afford the following observations and conclusions. 1) Carnitine-related molecules not containing a ␤-hydroxy moiety (2A-12A) can inhibit OCTN2 (CT1) and CT2 carnitine uptake. Exceptions to conclusion one  6A, 10A, 11A, and 12A. Once the substituents on the quaternary ammonium cation has become bulky, i.e. 6A has a tripropyl, the molecule will not readily inhibit OCTN2 (CT1) and CT2 carnitine uptake. Furthermore, once the length of the carbon chain has reached hexyl, 10A-12A, Tris OCTN2 (CT1) will not be as tolerant to the bulky quaternary ammonium cation as will CT2. 2) Carnitine-related molecules with a ␤-hydroxy moiety and a quaternary ammonium cation larger than trimethyl-(i.e. triethyl-, tripropyl-), the butane and pentane series (13)(14)(15)(16)(17)(18)(19)(20), will not readily inhibit OCTN2 (CT1) and CT2 carnitine uptake. However, compounds 17 and 18 do in fact demonstrate minor differences between OCTN2 (CT1) and CT2 substrate selectivity. It is noteworthy that comparison of the butane and pentane results (14-16 versus 17-20) suggests CT2 can tolerate a pentyl chain with more quaternary ammonium cation hydrophobicity and bulkiness than can OCTN2 (CT1). The butyl series is not as accommodating as the pentyl series, presumably because of the differences in relative degrees of freedom and the fact that the pentyl series forms a hydrogen bond in a seven-membered ring whereas the butyl series forms a six-member ring. Hence, the quaternary ammonium cation bulkiness (size) and relative hydrophobicity are important factors that inherently influence the charge on the nitrogen atom and triggers substrate interactions (34).
Immunohistochemistry revealed CT2 to be present in the luminal surface of epithelial cells of human epididymis (Fig. 9). Consistent with previous investigations where L-carnitine uptake was particularly active in isolated epididymal tubules (2,5,6,32,36), the immunohistochemical result allows one to propose that a CT2-mediated pathway constitutes a mechanism for L-carnitine transport in the luminal membrane of epididymis. In the epididymal lumen, concentrated L-carnitine goes through the sperm plasma membrane by passive diffusion (3) and serves as accessible energy storage when needed (37). When spermatozoa enter the epididymis, they are immotile and L-carnitine content is low. During their transit through the epididymis, spermatozoa initiate flagellar motion in parallel with accumulation of high concentrations (mM) of free L-carnitine from the luminal fluid (2). Therefore, the roles of carnitine are associated with improving sperm quality and fertility. There is a significant positive correlation between carnitine concentration in the genital tract, number of spermatozoa, and percentage of motile normal spermatozoa (38,39). Furthermore, evidence indicates that clinical administration of L-carnitine or acetyl-L-carnitine to infertile male patients (e.g. idio- pathic oligoasthenospermia) was followed by increase in sperm number and motility (40 -42). Thus, it is plausible that CT2mediated L-carnitine transport is required for the maturation of spermatozoa and that the human CT2 gene be a potential target for male infertility screening and treatment. Recently, a sperm-specific calcium channel (CatSper) was identified as a requirement for sperm motility and male fertility and represents an excellent molecular target for developing non-hormonal contraceptives (43). CT2 appears to be an additional example of a novel target for developing new antifertility agents. CT2 is only expressed in testis, consistent with its being involved in regulating spermatozoa motility and with the fact that a specific blocker may not influence other tissues.
Essential in the regulation of spermatogenesis, CT2 was also present in Sertoli cells (Fig. 9, a and b), which suggests a potential role of CT2 in these cells. Palmero et al. (35) provides support that L-carnitine facilitates lipid metabolism in Sertoli cells and denote the involvement of L-carnitine in the regulation of Sertoli cell function related to germ cell nutrition.
In summary, we cloned a novel carnitine transporter (CT2) specifically expressed in human testis. CT2 substrate selectivity and substrate recognition mechanism is distinct from other multispecific OCT/OCTN and OAT family members. CT2 represents an interesting evolutionary link between OCT/OCTNs and OATs, and also provides significant insight into the role of carnitine in the maturation of human spermatozoa.