The biochemical properties of the mitochondrial thiamine pyrophosphate carrier from Drosophila melanogaster Domenico Iacopetta 1, *, Chiara Carrisi 2, *, Giuseppina De Filippis 2 , Valeria M. Calcagnile 3 , Anna R. Cappello 1 , Adele Chimento 1 , Rosita Curcio 1 , Antonella Santoro 1 , Angelo Vozza 3 , Vincenza Dolce 1 , Ferdinando Palmieri 3 and Loredana Capobianco 2 1 Department of Pharmaco-Biology, University of Calabria, Arcavacata di Rende, Cosenza, Italy 2 Department of Biological and Environmental Sciences and Technologies, University of Salento, Lecce, Italy 3 Department of Pharmaco-Biology, University of Bari, Italy Introduction Several cofactors (i.e. coenzymes and prosthetic groups) are essential for the functioning of important metabolic processes occurring in mitochondria. Although most of these cofactors have to be imported from the cytosol into mitochondria, very little is known about the molecular basis of their transport across the mitochondrial membrane. Thiamine pyro- phosphate (ThPP) is a fundamental coenzyme of vari- ous cytosolic and mitochondrial reactions. It is synthesized in the cytosol [1,2], and is required in the Keywords CG2857 and CG6608; Drosophila melanogaster; mitochondria; proteomics; thiamine pyrophosphate carrier Correspondence L. Capobianco, F. Palmieri or V. Dolce, Department of Biological and Environmental Sciences and Technologies, University of Salento, 73100 Lecce, Italy; Department of Pharmaco-Biology, University of Bari, 70125 Bari, Italy; Department of Pharmaco-Biology, University of Calabria, Rende 87036 (CS), Italy Fax: +39 0 832 298 626; +39 0 80 5442 770; +39 0 984 493 270 Tel: +39 0 832 298 864; +39 0 80 5443 323; +39 0 984 493 177 E-mail: loredana.capobianco@unile.it; fpalm@farmbiol.uniba.it; vdolce@unical.it *These authors contributed equally to this work. (Received 28 July 2009, revised 15 December 2009, accepted 17 December 2009) doi:10.1111/j.1742-4658.2009.07550.x The mitochondrial carriers are a family of transport proteins that shuttle metabolites, nucleotides and cofactors across the inner mitochondrial mem- brane. The genome of Drosophila melanogaster encodes at least 46 mem- bers of this family. Only five of these have been characterized, whereas the transport functions of the remainder cannot be assessed with certainty. In the present study, we report the functional identification of two D. mela- nogaster genes distantly related to the human and yeast thiamine pyrophos- phate carrier (TPC) genes as well as the corresponding expression pattern throughout development. Furthermore, the functional characterization of the D. melanogaster mitochondrial thiamine pyrophosphate carrier protein (DmTpc1p) is described. DmTpc1p was over-expressed in bacteria, the puri- fied protein was reconstituted into liposomes, and its transport properties and kinetic parameters were characterized. Reconstituted DmTpc1p trans- ports thiamine pyrophosphate and, to a lesser extent, pyrophosphate, ADP, ATP and other nucleotides. The expression of DmTpc1p in Saccha- romyces cerevisiae TPC1 null mutant abolishes the growth defect on fer- mentable carbon sources. The main role of DmTpc1p is to import thiamine pyrophosphate into mitochondria by exchange with intramitochondrial ATP and ⁄ or ADP. Abbreviations MCF, mitochondrial carrier family; NDP, nucleoside diphosphate; NMP, nucleoside monophosphate; NTP, nucleoside triphosphate; Pi, phosphate; PPi, pyrophosphate; Th, thiamine; ThMP, thiamine monophosphate; ThPP, thiamine pyrophosphate; Tpc, thiamine pyrophosphate carrier. 1172 FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS cytosol for the activity of transketolase, and in the mitochondria for the activity of pyruvate-, oxogluta- rate- and branched chain keto acid dehydrogenases. To our knowledge, in Drosophila melanogaster, only the pyruvate dehydrogenase complex has been charac- terized among the ThPP-dependent enzymes [3]. Exper- iments performed with intact rat liver mitochondria have led to the proposal of the existence of different transport systems for thiamine, ThPP ⁄ thiamine (Th) exchange, ThPP and thiamine monophosphate (ThMP) transport, and ThMP uniport or ThMP ⁄ phosphate (Pi) exchange [4–6]. To date, only the yeast ThPP car- rier (Tpc1p) and the human Tpc have been identified as being responsible for the mitochondrial transport of ThPP and ThMP [7,8]. In particular, the human Tpc encoded by the SLC25A19 gene was previously indi- cated as the deoxynucleotide carrier [9], and then ascertained to be the human Tpc [8]. Tpc1p and Tpc belong to the mitochondrial carrier family (MCF) [10– 12]. Family members have a tripartite structure con- sisting of three tandemly repeated sequences of  100 amino acids in length. Each repeat contains two hydrophobic stretches that span the membrane as a-helices and a characteristic sequence motif [10]. An analysis of the D. melanogaster genome has led to the identification of 46 possible MCF members [13]. To date, five D. melanogaster mitochondrial carriers have been identified by their high similarity with orthologs in other organisms. They are the two isoforms of the ADP ⁄ ATP translocase [14–16], the carnitine ⁄ acylcarni- tine [17,18], citrate [13] and mitoferrin carriers [19]. In the present study, we report the identification of two D. melanogaster genes, CG6608 and CG2857, which are related to the human thiamine pyrophosphate car- rier (TPC) and yeast TPC1 genes, as well as the expression profile of the corresponding transcripts in different developmental stages. Moreover, in the pres- ent study, we provide evidence that DmTpc1p (encoded by CG6608) is the transporter of ThPP. DmTpc1p over-expressed in Escherichia coli and recon- stituted into phospholipid vesicles transports ThPP across liposomal membranes with high affinity. Fur- thermore, the expression of DmTpc1p in a yeast mutant laking TPC restores the growth defect on fermentable substrates. Results Identification and characterization of DmTPC cDNAs The protein sequence of the human Tpc encoded by the SLC25A19 gene [8,9] was used to search the Fly- Base database (http://flybase.org) for homologous sequences. Three putative transcripts corresponding to D. melanogaster genes CG6608 and CG2857 were iden- tified. The CG6608 gene encodes for two transcripts (CG6608-RA and CG6608-RB), whereas CG2857 is an intronless gene coding for only one transcript. The two transcripts of the CG6608 gene contained the same 999 bp ORF encoding a putative protein of 332 amino acid residues (henceforth named DmTpc1p) with a cal- culated molecular mass of 36.7 kDa (Fig. 1). The CG2857 gene containing a 972 bp ORF encoded a putative protein of 323 amino acid residues (henceforth named DmTpc2p) with a calculated molecular mass of 36.4 kDa (Fig. 1). DmTpc1p and DmTpc2p share 39% of identical amino acids. They have 33% and 31% sequence iden- tity and 53% and 51% sequence similarity to human Tpc. The D. melanogaster proteins were used to screen yeast databases for homologous sequences. The closest relative of DmTpc1p and DmTpc2p in Saccharomy- ces cerevisiae is YPR011c whose function is not yet known (26% and 23% sequence identity, respectively), followed by yTpc1p encoded by the YGR096w gene (24% and 21% sequence identity, respectively), which has been demonstrated to be the transporter of ThPP [7] (Fig. 1). DmTpc1p and DmTpc2p belong to the MCF because their amino acid sequences are composed of three tandem repeats of  100 amino acids, each containing two transmembrane a-helices, linked by an extensive loop, and a conserved signature motif [10]. Expression of D. melanogaster TPC transcripts in various developmental stages To determine the expression levels of transcripts corre- sponding to the CG6608 and CG2857 genes, we per- formed a semi-quantitative RT-PCR analysis on total RNAs from wild-type embryos, larvae, pupae and adults, using primers based on sequence retrieved from FlyBase. A PCR product of the predicted size was detected at high levels in embryos and adult flies, although a weaker but significant signal was found in larvae and pupae (Fig. 2) for transcripts CG6608-RA and CG6608-RB. The significance of these two tran- scripts, which have arisen from alternative splicing of the 5¢-UTR, is not yet known. However, the 5¢-UTR of eukaryotic mRNAs can play a role in the post- transcriptional regulation of gene expression through the modulation of translation efficiency and message stability [20]. No visible band of expression was found for the CG2857-RA transcript. Furthermore, any attempt to amplify the coding sequence corresponding to the D. Iacopetta et al. Mitochondrial transport of thiamine pyrophosphate FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS 1173 CG2857-RA transcript failed (data not shown). A con- trol RT-PCR was carried out using specific primers for Rp49 (Fig. 2). Bacterial expression of DmTpc1p DmTpc1p, the only protein encoded by both tran- scripts of the CG6608 gene, was expressed at high lev- els in E. coli BL21(DE3) (Fig. 3, lane 4) to identify its biochemical function. It accumulated as inclusion bodies and was purified as described previously [9] (Fig. 3, lane 5). The apparent molecular mass of the CG6608 - RA CG6608 - RB CG2857 - RA RP 49 ELPA Fig. 2. Expression of the DmTPC transcripts during development. Ethidium bromide staining of the RT-PCR products obtained using specific primers for D. melanogaster transcript TPCs and cDNA from Oregon R embryos (E), larvae (L), pupae (P) and adults (A). As a control for the RNA integrity, the Rp49 was amplified. 97 kDa 66.2 kDa 45 kDa 31 kDa 21.5 kDa 14.4 kDa 12345M Fig. 3. Expression in E. coli and purification of DmTpc1p. Proteins were separated by SDS-PAGE and stained with Coomassie blue dye. Lane M, markers (phosphorylase b, serum albumin, ovalbumin, carbonic anhydrase, trypsin inhibitor and lysozyme); lanes 1–4, E. coli BL21(DE3) containing the expression vector, without (lanes 1 and 3) and with the coding sequence for DmTpc1p (lanes 2 and 4). Samples were taken at the time of induction (lanes 1 and 2) and 4 h later (lanes 3 and 4). The same number of bacteria was analy- sed in each sample. Lane 5, purified DmTpc1p (5 lg) originating from bacteria shown in lane 4. Fig. 1. Comparison of predicted Tpc proteins from various species. Alignment of D. melanogaster, Homo sapiens and S. cerevisiae proteins. The accession numbers for the different sequences used in the alignment are NP_650034 (DmTpc1p); NP_611977 (DmTpc2p); NP_068380 (hTpc); NP_015336 (YPR011c); NP_011251 (yTpc1p). Dashes denote gaps. Asterisks and dots indicate residues in all five sequences, which are identical and conserved, respectively. Mitochondrial transport of thiamine pyrophosphate D. Iacopetta et al. 1174 FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS recombinant protein was  37 kDa (the calculated value with initiator methionine was 37 499 Da). The identity of the purified protein was confirmed by N- terminal sequencing. The protein was not detected in bacteria harvested immediately before the induction of expression (Fig. 3, lane 2), nor in cells harvested after induction but lacking the coding sequence in the expression vector (Fig. 3, lane 3). Approximately 90 mg of purified protein was obtained per litre of culture. Functional characterization of recombinant DmTpc1p DmTpc1p was reconstituted into liposomes, and its transport properties were tested in homo-exchange (i.e. same substrate inside and outside) experiments. Using external and internal substrate concentrations of 1 and 5mm, respectively, the reconstituted protein catalyzed an active dATPaS-[ 35 S] ⁄ dATP exchange but not homo-exchanges for malate, oxoglutarate, citrate, car- nitine, glutamate and aspartate (data not shown). No dATPaS-[ 35 S] ⁄ dATP exchange was observed with DmTpc1p that had been boiled before incorporation into liposomes, nor by reconstitution of sarcosyl-solu- bilized material from bacterial cells either lacking the expression vector for DmTpc1p or harvested immedi- ately before the induction of expression. The substrate specificity of recombinant DmTpc1p was examined in detail by measuring the uptake of dATPaS-[ 35 S] into proteoliposomes preloaded with various substrates. As shown in Fig. 4, the highest activity was observed in the presence of internal ThPP. ADP and dADP were also transported at a consider- able rate. Significant activities were also observed with internal pyrophosphate (PPi), nucleoside diphosphates (NDPs), nucleoside triphosphates (NTPs), dNDPs and dNTPs of the bases A, G, U or C. Furthermore, no significant exchange activity was found using Th, ThMP, adenosine, Pi, nucleoside monophosphates (NMPs) and dNMPs of the bases A, G, U or C. No activity was observed with guanosine, cytidine, uridine, oxoglutarate, citrate, adenosine 3¢,5¢-diphosphate and CoA (data not shown). The substrate that was best transported comprised ThPP, followed by ADP and dADP, which were transported with a slightly higher efficiency than PPi and the remaining NDPs, NTPs, dNDPs and dNTPs. Consistently, dATPaS-[ 35 S] uptake in the presence of 5 mm ADP inside the proteoliposomes was strongly inhibited by the external addition of ThPP, ADP and dADP (Fig. 5A). A lower inhibition was found with PPi, NDPs, NTPs, dNDPs and dNTPs of the bases A, G, U or C. Almost no effect was exerted by external Th, ThMP, adenosine, Pi, NMPs and dNMPs of the base A, G, U or C. The reaction catalyzed by reconstituted DmTpc1p was completely inhibited by p-chloromercuribenzene sulfonate and bathophenanthroline (strong inhibitors of several mitochondrial carriers) and, to a lesser extent, by pyridoxal 5¢-phosphate, mersalyl and mercu- ric chloride (other strong inhibitors of many mitochon- drial carriers) (Fig. 5B). No significant inhibition was observed with N-ethylmaleimide. The different inhibi- tory potency of SH reagents may be explained, at least in part, by the different microenvironment surrounding the reactive cysteine(s). Carboxyatractyloside and bon- gkrekate, powerful inhibitors of the mitochondrial ADP ⁄ ATP carrier [21,22], were partly effective on DmTpc1p (50% and 30% inhibition, respectively). A specific inhibitor of the mitochondrial citrate carrier, 1,2,3-benzenetricarboxylate, strongly reduced dATP ⁄ ADP exchange. No significant inhibition was observed with butylmalonate and phenylsuccinate (i.e. inhibitors of other characterized mitochondrial carriers) (Fig. 5B). 0 1 2 3 4 Adenosine Pi PPi Th ThMP ThPP AMP GMP CMP UMP ADP GDP CDP UDP ATP GTP CTP UTP dAMP dGMP dCMP dUMP dADP dGDP dCDP dATP dGTP dCTP dUTP V (µmol·min –1 × mg protein) Fig. 4. Substrate specificity of DmTpc1p. Liposomes reconstituted with DmTpc1p were preloaded internally with various substrates (concentration 5 m M). Transport was started by addition of 125 lM dATPaS-[ 35 S] and terminated after 2 min. Similar results were obtained in at least four independent experiments. D. Iacopetta et al. Mitochondrial transport of thiamine pyrophosphate FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS 1175 Kinetic characteristics of recombinant DmTpc1p The uptake of 0.5 mm dATPaS-[ 35 S] into proteolipo- somes was measured either as uniport (in the absence of internal substrate) or as exchange (in the presence of internal 5 mm ADP) (Fig. 6A). The uptake of dATP by exchange followed a first-order kinetics (rate constant 0.016 min )1 ; initial rate 1.47 lmolÆmin )1 Æmg protein )1 ) with isotopic equilibrium being approached exponentially (Fig. 6A). By contrast, no dATPaS-[ 35 S] uptake was observed without an internal substrate, indicating that DmTpc1p does not catalyze the unidi- rectional transport (uniport) of dATP, but only the exchange reaction. The uniport mode of transport was further investigated by measuring the efflux of dATPaS-[ 35 S] from prelabeled active proteoliposomes because it provides a more convenient assay for unidi- rectional transport [23]. In the absence of external sub- strate, no efflux was observed even after incubation for 60 min (Fig. 6B), whereas extensive efflux occurred upon addition of external ThPP. A significant efflux of dATPaS-[ 35 S] from prelabeled proteoliposomes was observed after the addition of external UTP or ATP. These results demonstrate that reconstituted DmTpc1p catalyzes an obligatory exchange reaction of 0 20 40 60 80 100 0 20 40 60 80 100 Adenosine Pi PP Th ThMP ThPP AMP GMP CMP UMP ADP GDP CDP UDP ATP GTP CTP UTP dAMP dGMP dCMP dUMP dADP dGDP dCDP dATP dGTP dCTP dUTP Inhibition (%) HgCl2 pCMBS Mersalyl N-ethylmaleimide PLP BAT BTA PHS BMA CAT BKA Inhibition (%) A B Fig. 5. Effect of inhibitors on the dATPaS-[ 35 S] ⁄ ADP exchange mediated by DmTpc1p. Proteoliposomes were preloaded internally with 5 m M ADP; transport was initiated by adding 125 lM dATPaS- [ 35 S] and terminated after 2 min. (A) Effect of external substrates. The external substrates (concentration 0.5 m M) were added together with dATPaS-[ 35 S]. (B) Effect of mitochondrial carrier inhib- itors. Thiol reagents were added 2 min before the labeled sub- strate; the other inhibitors were added together with dATPaS-[ 35 S]. The final concentrations of the inhibitors were 10 l M carboxyatract- yloside (CAT) and bongkrekic acid (BKA); 0.1 m M p-chloromercuri- benzene sulfonate (pCMBS), mersalyl and mercuric chloride (HgCl 2 ); 2 mM N-ethylmaleimide (NEM), benzene-1,2,3-tricarboxy- late (BTA), butylmalonate (BMA) and phenylsuccinate (PHS); 10 m M pyridoxal 5¢-phosphate (PLP) and bathophenanthroline (BAT). The extent of inhibition (%) from a representative experiment is reported. Similar results were obtained in at least five experiments. 0 50 100 150 200 250 0 20 40 60 80 100 A B Time (min) dATP uptake µmol·mg protein –1 0 20 40 60 0 2000 4000 6000 8000 10 000 12 000 Time (min) dATP efflux (cpm x 10 3 ) Fig. 6. Kinetics of dATPaS-[ 35 S] transport in proteoliposomes recon- stituted with DmTpc1p. (A) Uptake of dATP. A concentration of 500 l M dATPaS-[ 35 S] was added to proteoliposomes containing 5m M ADP (exchange, )or5mM NaCl and no substrate (uniport, ). Similar results were obtained in three independent experiments. (B) Efflux of dATPaS-[ 35 S] from proteoliposomes reconstituted in the presence of 5 m M ADP. The internal substrate pool was labeled with dATPaS-[ 35 S] by carrier-mediated exchange equilibration. Then the proteoliposomes were passed through Sephadex G-75. dATP aS-[ 35 S] efflux was initiated by adding Hepes 10 mM (pH 6.9), without (•) or with 0.5 m M dithioerythritol ( ), 2 mM ThPP with 0.5 m M dithioerythritol ( ), 2 mM UTP with 0.5 mM dithioerythritol (.)or2m M ATP with 0.5 mM dithioerythritol (¤). Similar results were obtained in five independent experiments. Mitochondrial transport of thiamine pyrophosphate D. Iacopetta et al. 1176 FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS substrates. In another set of experiments, the addition of 5 mm ThPP following a 60 min incubation, during which dATPaS-[ 35 S] uptake by proteoliposomes had almost reached equilibrium, caused an extensive efflux of radioactive compound. This efflux shows that the dATPaS-[ 35 S] taken up by proteoliposomes is released in exchange for externally added ThPP. Therefore, ThPP is transported by reconstituted DmTpc1p not only when it is inside liposomes, but also when added externally. The kinetic constants of the recombinant purified DmTpc1p were determined by measuring the initial transport rate at various external dATPaS-[ 35 S] con- centrations in the presence of a constant saturating internal concentration (5 mm) of ADP. The K m and V max values (measured at 25 °C) were 107.6 ± 0.4 lm and 1.73 ± 0.12 lmolÆmin )1 Æmg protein )1 , respectively (means of 30 experiments). The activity was calculated by taking into account the amount of DmTpc1p recov- ered in the proteoliposomes after reconstitution. Sev- eral external substrates were competitive inhibitors of dATPaS-[ 35 S] uptake (Table 1) because they increased the apparent K m without changing V max (not shown). These results confirm that the affinity of DmTpc1p for ThPP is higher than that for dADP, UTP and ATP. Furthermore, the K i value of ThPP is more than 200- fold lower than that of AMP. DmTpc1p functions as a ThPP transporter in S. cerevisiae The yeast TPC1 null mutant does not grow on thia- mine-less synthetic minimal medium supplemented with fermentable carbon sources [7]. This phenotype is explained by the ability of Tpc1p to import ThPP into mitochondria. Thus, the expression of a mitochondrial carrier protein that recognizes ThPP as a substrate should mitigate or abolish the growth defect of the tpc1D knockout. The DmTpc1p expressed in tpc1D cells via the yeast vector pYES2 fully restored growth of the tpc1D strain on galactose (Fig. 7), indicating that DmTpc1p imports ThPP into yeast mitochondria. By contrast, when the tpc1D cells were transformed with the empty vector, no growth restoration was observed. Discussion In the present study, DmTpc1p (encoded by the CG6608 gene) was shown, by direct transport assays, to transport ThPP after expression in E. coli and reconstitution into liposomes. This approach, which has previously been used for the identification of mito- chondrial carriers from high eukaryotes [10], yeast [24] and plants [25], revealed that DmTpc1p is different from any previously described mitochondrial carrier protein. On the basis of the transport properties and kinetic characteristics of DmTpc1p reported in the present study, this protein is the D. melanogaster mito- chondrial transporter for ThPP. Furthermore, comple- mentation of the yeast TPC null mutant by the expression of DmTpc1p clearly indicates that DmTpc1p is able to transport ThPP into mitochondria. The related sequence DmTpc2p (encoded by the CG2857 gene) could not be functionally characterized because no corresponding cDNA was generated by RT-PCR in any developmental stage analysed. The absence of transcripts of the intronless CG2857 gene was not unexpected because its structure clearly indi- cates that it is a paralogous gene, produced by retro- transposition, of the pre-existing ‘parent’ gene CG6608 [26,27]. Indeed, a virtual screening of the expressed sequence tag databases showed that CG2857, similar to the OXPHOS paralogous genes [27], is expressed (at very low levels) only in testis [27,28]. Table 1. Competitive inhibition by various substrates of dATPaS- [ 35 S] uptake in proteoliposomes containing recombinant DmTpc1p. The values were calculated from Lineweaver–Burk plots of the rate of dATPaS-[ 35 S] versus substrate concentrations. The competing substrates at appropriate constant concentrations were added together with 0.005–1.25 m M dATPaS-[ 35 S] to proteoliposomes containing 5 m M ADP. The data represent the mean ± SD of at least three different experiments. Substrate K i (mM) ThPP 0.010 ± 0.002 dADP 0.10 ± 0.01 UTP 0.23 ± 0.03 ATP 0.28 ± 0.04 AMP 2.51 ± 0.37 Fig. 7. The yeast tpc1D strain is fully complemented by the gene for DmTpc1. Four-fold serial dilutions of wild-type, tpc1D,DmTPC1- pYES2 tpc1D and pYES2 tpc1D cells were plated on solid thiamine- less synthetic minimal medium supplemented with 2% galactose. The plates were incubated at 30 °C for 4 days. D. Iacopetta et al. Mitochondrial transport of thiamine pyrophosphate FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS 1177 DmTpc1p and DmTpc2p, which share 39% identity, have a higher degree of sequence identity with the unknown yeast protein YPR011c (26% and 23%, respectively) than with the yeast Tpc1p encoded by the YGR096w gene [7]. However, a phylogenetic analysis (Fig. 8) carried out using several Tpc sequences, as well as other mitochondrial carriers, revealed that DmTpc1p, DmTpc2p, yeast Tpc1p [7] and human Tpc [8–9] are monophyletic, whereas the yeast protein YPR011c clusters with the Grave’s disease carrier (and its yeast homologue leu5p) and SLC25A42 [29–31]. The biochemical properties of the recombinant reconstituted DmTpc1p are different from the human and yeast Tpc proteins in several respects: DmTpc1p catalyzes an obligatory counter-exchange; the substrate that is more efficiently transported is ThPP; the affinity of DmTcp1 for this substrate is very high (K i for ThPP, 10 lm), a value that is 20-fold lower than that measured in yeast (no data is available in humans); the D. melanogaster protein is unable to transport ThMP; effective counter-substrates for ThPP probably are ATP (NTPs), ADP (NDPs) and PPi; the K i for dATP is similar to that determined for the human carrier encoded by SLC25A19 [9], whereas it is five-fold lower than that determined for yeast [7]; and 1,2,3-benzene- tricarboxylate, a known inhibitor of the citrate carrier, strongly reduces the dATP ⁄ ADP exchange rate to 15%. Because ThPP is produced in the cytosol by thia- mine pyrophosphokinase [1,2], the primary function of DmTpc1p is to catalyze the uptake of ThPP into mito- chondria. However, given that DmTpc1p functions by a counter-exchange mechanism, the carrier-mediated uptake of ThPP requires the efflux of a counter- substrate. The internal counter-ion for exchange could be either ADP or most likely ATP. Thus, in the resting state, the intramitochondrial ATP ⁄ ADP ratio is  4 [32] and the rate of exchange of external ThPP for internal ATP is favored by the high amount of ATP generated by oxidative phosphorylation. Therefore, the physiological role of the DmTpc1p is probably to cata- lyze the uptake of ThPP into the mitochondrial matrix in exchange for internal ATP. DmTpc1p is crucial for mitochondrial metabolism because ThPP is an essential coenzyme for the E1 components of pyruvate dehydrogenase and oxogluta- rate dehydrogenase, which are located in the mito- chondrial matrix. In agreement with its importance in mitochondrial metabolism, DmTpc1p is localized in the mitochondria, as revealed by immunofluorescence analysis (V. Dolce & L. Capobianco, unpublished data) and is expressed during all stages of develop- ment. Mutations of SLC25A19 cause lethal Amish microcephaly, which is characterized by severe congen- ital microcephaly, elevated levels of a-ketoglutarate in urine, almost no orientation to sight or sound and no motor development. Studies using TPC1 null mutants of D. melanogaster could help to gain insight into the molecular and cellular pathogenetic mechanisms of Amish microcephaly. Indeed, although the investiga- tion of rodent models is sometimes of significant impact, invertebrate models offer several advantages (i.e. short life span, large number of offspring and numerous genetic techniques, amongst others) that can 0.1 yNdt2p yNdt1p yAnt1p yTpc1p DmTpc2p DmTpc1p YPR011c yLeu5p hSLC25A42 hGDC yAAC3 yAAC2 yAAC1 hAAC4 hAAC3 hAAC2 hAAC1 hACP3 hACP2 hACP1 yPTP hPICB hPiCA yGgc1p hSAMC yRim2p ySam5p hANC hTPC Fig. 8. Phylogenic tree of amino acid sequences of mitochondrial transporters from various organisms. The unrooted dendogram orig- inated from an alignment performed by CLUSTALW (http://www.ebi. ac.uk/clustalw) using the default options. Branch lengths are drawn proportional to the amount of sequence change. The bar indicates the number of substitutions per residue, with 0.1 corresponding to a distance of ten substitutions per 100 residues. The tree was visu- alized using DENDROSCOPE software [38]. The proteins have the accession numbers: DmTpc1p, NP_650034; DmTpc2p, NP_611977; hAAC1, NP_001142; hAAC2, NP_001143; hAAC3, NP_001627; hAAC4, NP_112581; hACP1, NP_998816; hACP2, NP_077008; hACP3, NP_001006643; hANC, NP_006349; hGDC, NP_689920; hPiCA, NP_005879; hPiCB, NP_002626; hSAMC, NP_775742; hSLC25A42, NP_848621; hTpc, NP_068380; yAAC1, NP_013772; yAAC2, NP_009523; yAAC3, NP_009642; yAnt1p, NP_015453; yGgc1p, NP_010083; yLeu5p, NP_011865; yPTP, NP_012611; yNdt1p, NP_012260; yNdt2p, NP_010910; YPR011c, NP_015336; yRim2p, NP_009751; ySam5p, NP_014395; yTpc1p, NP_011251. Dm, D. melanogaster; h, human; y, yeast; AAC, ADP ⁄ ATP carrier; ACP, ATP-Mg ⁄ Pi carrier; ANC, peroxisomal adenine nucleotide carrier; GDC, Graves’ disease carrier; PiC, phosphate carrier; SAMC, S-adenosylmethionine carrier; SLC25A42, CoA and adeno- sine 3¢,5¢-diphosphate carrier; Ant, peroxisomal adenine nucleotide transporter; Ggc, GTP ⁄ GDP carrier; Leu5, accumulation of CoA in the matrix; yPTP, phosphate transport carrier; Ndt, NAD + carrier protein; Rim, pyrimidine nucleotides carrier; YPR011c. Mitochondrial transport of thiamine pyrophosphate D. Iacopetta et al. 1178 FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS address some important issues underlying neurological disease [33]. Experimental procedures Computer search for DmTPC genes The D. melanogaster genome annotated in the FlyBase (http://flybase.org) was screened with the human sequence of the mitochondrial Tpc also known as deoxynucleotide carrier [8,9] with the aid of tblastn (http://blast.ncbi. nlm.nih.gov/blast.cgi). Amino acid sequences were aligned with clustalw (http://www.ebi.ac.uk/tools/clustalw2/index. html). Construction of the expression plasmid coding for DmTpc1p Total RNA was extracted from Oregon R adult flies using RNeasy Mini Kit (Qiagen, Valencia, CA, USA) and reverse transcribed as described previously [13]. The coding region for DmTpc1p was amplified from first strand cDNA (100 ng) by PCR with 5¢-GCGGTAACCACAGGCTC-3¢ (sense primer) and 5¢-CTAATGATGATGATGATGGAA GCGCACCTGCTTGAGCT-3¢ (antisense primer) of the D. melanogaster transcript CG6608-RA. The forward and reverse primers carried NdeI and HindIII restriction sites, respectively, as linkers. The reverse primer also carried a DNA sequence coding for six histidines followed by a stop codon. The reaction product was recovered from agarose gel, cloned in the expression vector pMW7 [34] and trans- formed into E. coli TG1 cells. Transformants, selected on LB plates containing ampicillin (100 lgÆmL )1 ), were screened by direct colony PCR, and by restriction digestion of purified plasmids. The sequences of the inserts were verified. Expression analysis by semiquantitative RT-PCR Total RNA was extracted from Oregon R embryos, larvae, pupae and adult flies using RNeasy Mini Kit (Qiagen) and reverse transcribed as described previously [13]. The constit- utive ribosomal gene (Rp49) was used as an internal con- trol. The sense and antisense gene-specific primers were: CG6608-RA, sense 5¢-AGGCATGATACTAAATGCCAT TGAA-3¢ and antisense 5¢-TCCAGAACTGACAAATGC CGTAC-3¢; CG6608-RB, sense 5¢-GTGGAGCATGATAC TTAAATGCCA-3¢, and antisense 5¢-TCCAGAACTGACA AATGCCGTAC-3¢; CG2857-RA, sense 5¢-CTCTTCTACA AGTACCTCAACGCGG-3¢ and antisense 5¢-TTCTCCCA AGATACTAATGCTTGCC-3¢; Rp49, sense 5¢-ATGACC ATCCGCCCAGCATACA-3¢ and antisense 5¢-TTGGTG AGGCGGACCGACAG-3¢. The PCR products were analy- sed by 1% agarose gel electrophoresis. Band intensities were quantified using quantity one 1-D Analysis Software (Bio-Rad, Hercules, CA, USA). Bacterial expression and purification of DmTpc1p The over-expression of DmTpc1p as inclusion bodies in the cytosol of E. coli was accomplished as described previously [35]. Control cultures with the empty vector were processed in parallel. Inclusion bodies were purified on sucrose den- sity gradient and washed at 4 °C, first with TE buffer (10 mm Tris ⁄ HCl, pH 8), then twice with a buffer contain- ing Triton X-114 (2%, w ⁄ v) and 10 mm Hepes (pH 6.9) and, finally, with Hepes 10 mm (pH 6.9). Proteins were sol- ubilized in 2.5% sarkosyl (w ⁄ v) and DmTpc1p was purified by centrifugation and Ni + -NTA-agarose affinity chroma- tography, as described previously [9]. Reconstitution into liposomes and transport assays The recombinant protein in sarkosyl was reconstituted into liposomes in the presence or absence of substrates [23]. The reconstitution mixture contained purified proteins (150 lL with 0.8–1 lg of protein), 10% Triton X-114 (90 lL), 10% phospholipids as sonicated liposomes (90 lL), 5 mm ADP (except where indicated otherwise), 10 mm Hepes (pH 6.9) and water to a final volume of 700 lL. These components were mixed thoroughly, and the mixture was recycled 13 times through the same Amberlite column (Bio-Rad). The external substrate was removed from proteolipo- somes on a Sephadex G-75 columns pre-equilibrated with 50 mm NaCl and 10 mm Hepes (pH 6.9) [23]. Transport at 25 °C was started by adding dATPaS-[ 35 S] (Perkin Elmer, Boston, MA, USA) at the indicated concentrations. The carrier-mediated transport was terminated by addition of 30 mm pyridoxal 5¢-phosphate and 10 mm bathophenanthr- oline. In control samples, the inhibitors were added at time 0 according to the inhibitor stop method [23]. All transport measurements were carried out at the same internal and external pH values. Finally, the external substrate was removed, and the radioactivity in the liposomes was mea- sured [23]. The experimental values were corrected by sub- tracting control values. The initial transport rate was calculated from the radioactivity taken up by proteolipo- somes after 1 min (in the initial linear range of substrate uptake). For efflux measurements, proteoliposomes contain- ing 5 mm ADP were labeled with 20 lm dATPa S-[ 35 S] by carrier-mediated exchange equilibration [23]. After 60 min, external substrate was removed by exclusion chromato- graphy in the presence of a reversible inhibitor (0.1 mm p- chloromercuribenzene sulfonate) to avoid efflux of internal substrate. Efflux was started by adding Hepes 10 mm (pH 6.9) without or with 0.5 mm dithioerythritol or unlabeled external substrate in the presence of 0.5 mm dithioerythritol. D. Iacopetta et al. Mitochondrial transport of thiamine pyrophosphate FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS 1179 In all cases, the transport was terminated by adding the inhibitors indicated above. Complementation of a yeast mutant lacking TPC1 by DmTPC1 BY4741 (wild-type) and tpc1D yeast strains were provided by the EUROFAN resource center EUROSCARF (Frank- furt, Germany). In the tpc1D mutant, the tpc1 (YGR096w) locus of S. cerevisiae strain BY4741 (MATa; his3D1; leu2D0; lys2D0; ura3D0) was replaced by kanMX4. The coding sequence of DmTpc1p was cloned into the BamHI-EcoRI sites of the expression vector pYES2 that had been previously modified by cloning a DNA sequence coding for the V5 epitope and six histidines into XhoI-XbaI sites (DmTPC1-pYES2). This plasmid was introduced into the tpc1D yeast strain, and trasformants were selected for uracil auxotrophy. Wild-type, tpc1D, DmTPC1-pYes2 tpc1D and pYes2 tpc1D strains were grown in rich medium con- taining 2% bactopeptone and 1% yeast extract, synthetic complete medium or thiamine-less synthetic minimal med- ium [36]. All media were supplemented with 2% glucose or 2% galactose. Other methods Proteins were analysed by SDS-PAGE and stained with Coomassie blue dye. The N-termini were sequenced, and the amount of pure DmTpc1p was estimated by laser densi- tometry of stained samples using carbonic anhydrase as a protein standard. The amount of protein incorporated into liposomes was measured as described previously [37]. Approximately 20% of Dm Tpc1p was reconstituted. Acknowledgements This work was supported by grants from the Ministero dell’Universita ` e della Ricerca (MIUR) and Apulia Region Neurobiotech (Progetto Strategico 124). We gratefully thank Dr Daniela Fiore for helpful discussion. References 1 Deus B & Blum H (1970) Subcellular distribution of thiamine pyrophosphokinase activity in rat liver and erythrocytes. Biochim Biophys Acta 219, 489–492. 2 Hohmann S & Meacock PA (1998) Thiamin metabo- lism and thiamin diphosphate-dependent enzymes in the yeast Saccharomyces cerevisiae: genetic regulation. Biochim Biophys Acta 1385, 201–219. 3 Katsube T, Nomoto S, Togashi S, Ueda R, Kobayashi M & Takahisa M (1997) cDNA sequence and expres- sion of a gene encoding a pyruvate dehydrogenase kinase homolog of Drosophila melanogaster. DNA Cell Biol 16, 335–339. 4 Barile M, Passarella S & Quagliariello E (1986) Uptake of thiamin by isolated rat liver mitochondria. Biochem Biophys Res Commun 141, 466–473. 5 Barile M, Passarella S & Quagliariello E (1990) Thiamine pyrophosphate uptake into isolated rat liver mitochondria. Arch Biochem Biophys 280, 352–357. 6 Barile M, Valenti D, Brizio C, Quagliariello E & Passarella S (1998) Rat liver mitochondria can hydrolyse thiamine pyrophosphate to thiamine mono- phosphate which can cross the mitochondrial membrane in a carrier-mediated process. FEBS Lett 435, 6–10. 7 Marobbio CM, Vozza A, Harding M, Bisaccia F, Palmieri F & Walker JE (2002) Identification and reconstitution of the yeast mitochondrial transporter for thiamine pyrophosphate. EMBO J 21, 5653–5661. 8 Lindhurst MJ, Fiermonte G, Song S, Struys E, De Leonardis F, Schwartzberg PL, Chen A, Castegna A, Verhoeven N, Mathews CK et al. (2006) Knockout of Slc25a19 causes mitochondrial thiamine pyrophosphate depletion, embryonic lethality, CNS malformations, and anemia. Proc Natl Acad Sci USA 103, 15927–15932. 9 Dolce V, Fiermonte G, Runswick MJ, Palmieri F & Walker JE (2001) The human mitochondrial deoxynu- cleotide carrier and its role in the toxicity of nucleoside antivirals. Proc Natl Acad Sci USA 98, 2284–2288. 10 Palmieri F (2004) The mitochondrial transporter family (SLC25): physiological and pathological implications. Pflugers Arch 447, 689–709. 11 Palmieri F (2008) Diseases caused by defects of mito- chondrial carriers: a review. Biochim Biophys Acta 1777, 564–578. 12 Palmieri F & Pierri CL (2009) Structure and function of mitochondrial carriers - Role of the transmembrane helix P and G residues in the gating and transport mechanism. FEBS Lett, doi:S0014-5793(09)00851-5. 13 Carrisi C, Madeo M, Morciano P, Dolce V, Cenci G, Cappello AR, Mazzeo G, Iacopetta D & Capobianco L (2008) Identification of the Drosophila melanogaster mitochondrial citrate carrier: bacterial expression, reconstitution, functional characterization and develop- mental distribution. J Biochem 144, 389–392. 14 Zhang YQ, Roote J, Brogna S, Davis AW, Barbash DA, Nash D & Ashburner M (1999) stress sensitive B encodes an adenine nucleotide translocase in Drosophila melanogaster. Genetics 153, 891–903. 15 Rikhy R, Ramaswami M & Krishnan KS (2003) A temperature-sensitive allele of Drosophila sesB reveals acute functions for the mitochondrial adenine nucleo- tide translocase in synaptic transmission and dynamin regulation. Genetics 165, 1243–1253. 16 Brand MD, Pakay JL, Ocloo A, Kokoszka J, Wallace DC, Brookes PS & Cornwall EJ (2005) The basal Mitochondrial transport of thiamine pyrophosphate D. Iacopetta et al. 1180 FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS proton conductance of mitochondria depends on adenine nucleotide translocase content. Biochem J 392, 353–362. 17 Hartenstein K, Sinha P, Mishra A, Schenkel H, Torok I & Mechler BM (1997) The congested-like tracheae gene of Drosophila melanogaster encodes a member of the mitochondrial carrier family required for gas-filling of the tracheal system and expansion of the wings after eclosion. Genetics 147, 1755–1768. 18 Oey NA, Ijlst L, van Roermund CW, Wijburg FA & Wanders RJ (2005) dif-1 and colt, both implicated in early embryonic development, encode carnitine acylcar- nitine translocase. Mol Genet Metab 85, 121–124. 19 Metzendorf C, Wu W & Lind MI (2009) Overexpres- sion of Drosophila mitoferrin in l(2)mbn cells results in dysregulation of Fer1HCH expression. Biochem J 421, 463–471. 20 Payton SG, Haska CL, Flatley RM, Ge Y & Matherly LH (2007) Effects of 5¢ untranslated region diversity on the posttranscriptional regulation of the human reduced folate carrier. Biochim Biophys Acta 1769, 131–138. 21 Fiore C, Trezeguet V, Le Saux A, Roux P, Schwimmer C, Dianoux AC, Noel F, Lauquin GJ, Brandolin G & Vignais PV (1998) The mitochondrial ADP ⁄ ATP carrier: structural, physiological and pathological aspects. Biochimie 80, 137–150. 22 Klingenberg M (1989) Molecular aspects of the adenine nucleotide carrier from mitochondria. Arch Biochem Biophys 270, 1–14. 23 Palmieri F, Indiveri C, Bisaccia F & Iacobazzi V (1995) Mitochondrial metabolite carrier proteins: purification, reconstitution, and transport studies. Methods Enzymol 260, 349–369. 24 Palmieri F, Agrimi G, Blanco E, Castegna A, Di Noia MA, Iacobazzi V, Lasorsa FM, Marobbio CM, Palmieri L, Scarcia P et al. (2006) Identification of mitochondrial carriers in Saccharomyces cerevisiae by transport assay of reconstituted recombinant proteins. Biochim Biophys Acta 1757, 1249–1262. 25 Palmieri F, Rieder B, Ventrella A, Blanco E, Do PT, Nunes-Nesi A, Trauth AU, Fiermonte G, Tjaden J, Agrimi G et al. (2009) Molecular identification and functional characterization of Arabidopsis thaliana mito- chondrial and chloroplastic NAD+ carrier proteins. J Biol Chem 284, 31249–31259. 26 Tripoli G, D’Elia D, Barsanti P & Caggese C (2005) Comparison of the oxidative phosphorylation (OXPHOS) nuclear genes in the genomes of Drosophila melanogaster, Drosophila pseudoobscura and Anopheles gambiae. Genome Biol 6, R11. 27 Porcelli D, Barsanti P, Pesole G & Caggese C (2007) The nuclear OXPHOS genes in insecta: a common evo- lutionary origin, a common cis-regulatory motif, a com- mon destiny for gene duplicates. BMC Evol Biol 7, 215. 28 Parisi M, Nuttall R, Edwards P, Minor J, Naiman D, Lu J, Doctolero M, Vainer M, Chan C, Malley J et al. (2004) A survey of ovary-, testis-, and soma-biased gene expression in Drosophila melanogaster adults. Genome Biol 5, R40. 29 Zarrilli R, Oates EL, McBride OW, Lerman MI, Chan JY, Santisteban P, Ursini MV, Notkins AL & Kohn LD (1989) Sequence and chromosomal assignment of a novel cDNA identified by immunoscreening of a thy- roid expression library: similarity to a family of mito- chondrial solute carrier proteins. Mol Endocrinol 3, 1498–1505. 30 Prohl C, Pelzer W, Diekert K, Kmita H, Bedekovics T, Kispal G & Lill R (2001) The yeast mitochondrial car- rier Leu5p and its human homologue Graves’ disease protein are required for accumulation of coenzyme A in the matrix. Mol Cell Biol 21, 1089–1097. 31 Fiermonte G, Paradies E, Todisco S, Marobbio CM & Palmieri F (2009) A novel member of solute carrier family 25 (SLC25A42) is a transporter of coenzyme A and adenosine 3¢,5¢-diphosphate in human mitochon- dria. J Biol Chem 284, 18152–18159. 32 Letko G, Kuster U, Duszynski J & Kunz W (1980) Investigation of the dependence of the intramitochond- rial [ATP] ⁄ [ADP] ratio on the respiration rate. Biochim Biophys Acta 593, 196–203. 33 Rubin GM & Lewis EB (2000) A brief history of Drosophila’s contributions to genome research. Science 287, 2216–2218. 34 Fiermonte G, Palmieri L, Dolce V, Lasorsa FM, Palmieri F, Runswick MJ & Walker JE (1998) The sequence, bacterial expression, and functional reconstitution of the rat mitochondrial dicarboxylate transporter cloned via distant homologs in yeast and Caenorhabditis elegans. J Biol Chem 273, 24754–24759. 35 Fiermonte G, Walker JE & Palmieri F (1993) Abundant bacterial expression and reconstitution of an intrinsic membrane-transport protein from bovine mitochondria. Biochem J 294 (Pt 1), 293–299. 36 Sherman F (1991) Getting started with yeast. Methods Enzymol 194, 3–21. 37 Dolce V, Scarcia P, Iacopetta D & Palmieri F (2005) A fourth ADP ⁄ ATP carrier isoform in man: identifica- tion, bacterial expression, functional characterization and tissue distribution. FEBS Lett 579, 633–637. 38 Huson DH, Richter DC, Rausch C, Dezulian T, Franz M & Rupp R (2007) Dendroscope: an interactive viewer for large phylogenetic trees. BMC Bioinformatics 8, 460. D. Iacopetta et al. Mitochondrial transport of thiamine pyrophosphate FEBS Journal 277 (2010) 1172–1181 ª 2010 The Authors Journal compilation ª 2010 FEBS 1181 . The biochemical properties of the mitochondrial thiamine pyrophosphate carrier from Drosophila melanogaster Domenico Iacopetta 1, *, Chiara Carrisi 2, *,. (1997) The congested-like tracheae gene of Drosophila melanogaster encodes a member of the mitochondrial carrier family required for gas-filling of the tracheal system and expansion of the wings. mitochondria. Although most of these cofactors have to be imported from the cytosol into mitochondria, very little is known about the molecular basis of their transport across the mitochondrial membrane. Thiamine