Introduction

Parasites infect hundreds of millions of people every year and collectively represent one of the principal causes of human misery. Among the protozoa, the Trypanosomatidae family comprises a large number of species responsible for diseases such as sleeping sickness (Trypanosoma brucei) and Leishmaniasis (Leishmania spp) (Barrett et al. 2003). Leishmania donovani and Trypanosoma cruzi are obligatory intracellular parasites that cause Kala Azar and Chagas disease in humans, respectively, killing thousands of patients annually (Barrett et al. 2003; Singh et al. 2006). These organisms cycle between insect vectors and mammalian hosts (Herwaldt 1999). Thus, these parasites encounter dramatic environmental changes in temperature, pH and nutrients. The parasites respond to these changes by differentiating into forms that are highly adapted to each environment (Rosenzweig et al. 2008; Mukkada et al. 1985).

Amino acids play a vital role in the life cycle of these parasites, some serving as alternative carbon sources and energy reserves (Mukkada et al. 1974; Opperdoes and Coombs 2007; Pereira et al. 2000) as well as precursors for biosynthesis of key molecules (Gaur et al. 2007; Roberts et al. 2004) in addition to participating in osmoregulation (Blum 1996). Hence, there are certain amino acids essential to Leishmania and Trypanosoma but non- or semi-essential to the host, which consequently represent potential targets for new drugs (Opperdoes and Coombs 2007). Amino acid permeases supply parasite cells with amino acids and accordingly, are key players in the mechanisms underlying adaptation to vector and host environments. Indeed, many permeases involved in such processes became essential during parasite evolution as transport systems supplanted biosynthetic pathways (Ginger 2006).

Earlier studies indicated that biochemically homologous transporters of specific amino acids are found across the various genera of Trypanosomatidae, suggesting that the transporters serve an evolutionarily conserved function (Mazareb et al. 1999; Pereira et al. 1999; Silber et al. 2006; Silber et al. 2002; Zilberstein and Gepstein 1993). Subsequent genomic analyses identified several members of this amino acid/auxin permease (AAAP, TC 2.A.18) gene family in T. cruzi and L. donovani (Bouvier et al. 2004; Akerman et al. 2004). Further detailed analyses revealed the significance of gene rearrangements during the evolution of these molecules by transpositive duplication, tandem duplication and descent (Jackson 2007).

Previously, we cloned and characterized a member of the AAAP family from L. donovani (LdAAP3) and T. cruzi (TcAAAP411) that functions as a high affinity arginine-specific transporter (Shaked-Mishan et al. 2006; Carrillo et al. 2010). Arginine transport is regulated by its availability and by metabolic pathways that require arginine as a precursor (Pereira et al. 2002; Darlyuk et al. 2009). Notably, the response of LdAAP3 to amino acid availability is identical to that reported for the mammalian cation amino acid transporter 1 (CAT1). However, CAT1 transports cationic amino acids in general (lysine and arginine) whereas LdAAP3 and TcAAAP411 translocate only arginine (Hatzoglou et al. 2004; Kandpal et al. 1995; Shaked-Mishan et al. 2006; Carrillo et al. 2010). Lysine is an essential amino acid for most eukaryotes. To date, only plants have been shown to synthesize this amino acid from aspartic acid (Azevedo et al. 1997; Stepansky et al. 2005; Galili et al. 2005). Indeed, trypanosomatids lack the key lysine synthesis enzymes found in plants (Opperdoes and Coombs 2007) and therefore must acquire lysine from their environments via a transporter. In light of our discovery that trypanosomatids, unlike their mammalian or vector hosts, separate lysine from arginine transport, we hypothesized that this parasite feature could play an important role in its life cycle and moreover, provide a new therapeutic approach to managing trypanosomatid pathogens. With these issues in mind, we have identified and characterized high affinity lysine-specific transporters in L. donovani and T. cruzi. Our present study represents the first characterization of a lysine transporter gene in parasitic protozoa.

Materials and methods

Materials

3H-labeled amino acids and 32−P-dCTP were from Amersham. The antibiotics G418 and Hygromycin B as well as medium 199 were from Sigma; fetal calf serum was from Biological Industries, Inc. Rabbit polyclonal anti-HA IgG was from Santa Cruz and the fluorochrome-conjugated secondary antibodies were from Jackson. All other reagents were analytical grade.

Phylogenetic analysis

The phylogenetic tree was produced using three different programs. Probcons with default options produced the initial multiple alignment (Do et al. 2005), which was subsequently fed into the FastTree phylogenetic software (Price et al. 2009). Local support values were computed using FastTree to indicate whether each split in the inferred topology is correct. Then tree visualization was done using Dendroscope (Huson et al. 2007).

Parasites culture

A cloned line of L. donovani 1SR was used in all experiments (Saar et al. 1998). Cultures were maintained by inoculating growth medium with single colonies of promastigotes from medium 199 agar plates. Promastigotes were grown at 26°C in medium 199 supplemented with 10% fetal calf serum. T. cruzi epimastigotes of the CL Brener strain (starting with 106 cells per mL) were cultured at 28°C in plastic flasks (25 cm2) containing 5 mL of LIT medium supplemented with 10% fetal calf serum, 100 U/mL penicillin, and 100 μg/mL streptomycin (Camargo 1964). The parasites were subcultured every 7 days and counted using a hemocytometric chamber.

Yeast strains and growth conditions

Strain 22∆6AAL (Fischer et al. 2002) was transformed with pDR195, pDR195-LdAAP7 and pDR195-TcAAP7 according to Dohmen et al. 1991. Plasmid expression was selected by growth on minimal medium (0.17% yeast nitrogen source without amino acids and without ammonium sulphate, 2% glucose) with 1 g/L urea as the nitrogen source and 1 g/L ‘Lys-Asp’ (rich media) or 100 μM lysine (selective media). S. cerevisiae strain 22∆7AA was used for lysine transport assays (Fischer et al. 2002). This strain was grown in minimal media supplemented with 1 g/L urea.

DNA and RNA work

S. cerevisiae complementation assays

PCR-amplified LdAAP7 or TcAAP7-ORF (GeneDB systematic IDs LinJ32_V3.2800 and Tc00.1047053511545.80, respectively) was cloned into the yeast expression vector pDR195 (Rentsch et al. 1995) between XhoI and NotI or SpeI and NsiI, respectively.

Northern blot analysis

Total RNA was extracted and subjected to northern blotting as described in Barak et al. (2005). Membranes were probed with 32−P dCTP-PCR amplified LdAAP7 and TcAAP7 ORFs (primer sequences in Table 1).

Table 1 Primers
Full size table

AAP7 over expression

For LdAAP7 over expression, the LdAAP7 ORF was cloned into the pNUS HnN expression vector (Tetaud et al. 2002) between the 5′ XhoI and 3′ KpnI sites (Table 1). Expression plasmid was transfected into L. donovani 1SR using standard electroporation conditions (LeBowitz et al. 1990; Jiang et al. 1999). Transfected colonies were selected on medium 199 agar plates containing 50 μg/mL G418. For TcAAP7 overexpression, the entire coding sequence of TcAAP7 (1392 bp) was cloned into pTREX (Vazquez and Levin 1999) or was fused to the 3′ end of the GFP gene present in the pTREX-GFP expression vector. The expression plasmids were transfected into T. cruzi as follows: 108 parasites grown at 28°C in LIT medium were harvested by centrifugation, washed with PBS, and resuspended in 0.35 mL of electroporation buffer (PBS containing 0.5 mM MgCl2 and 0.1 mM CaCl2). This cell suspension was mixed with 50 μg of plasmid DNA in 0.2 cm gap cuvettes (Bio-Rad Laboratories). The parasites were electroporated using a single pulse of (400 V, 500 μF) with a time constant of about 5 ms. Stable cell lines were achieved after 30 days of growth in the presence of 500 μg/mL G418 (Calbiochem).

LdAAP7 gene replacement

861 bp of the gene 5′ flanking region was cloned upstream to the hygromycin resistance cassette between the 5′ SalI and 3′ HindIII sites in pKOH plasmid (Ruepp et al. 1997). In addition, 929 bp of the gene 3′ flanking region was cloned downstream of the same hygromycin resistance cassette between the 5′ BamHI and 3′ XbaI sites. The fragment containing these LdAAP7 5′ and 3′ flanking regions surrounding the hygromycin resistance cassette was amplified by PCR and subsequently electroporated into L. donovani promastigotes. Transfected colonies were selected on medium 199 agar plates containing 50 μg/mL hygromycin B. A second fragment containing 5′ and 3′ flanking regions surrounding a G418 resistance cassette was created the same way using pKON plasmid (Ruepp et al. 1997). Heterrosygous colonies from step one were electroporated with the second construct and doubly transfected colonies selected on medium 199 containing 50 μg/mL G418 and hygromycin (see Table 1 for primers). Insertion of the antibiotic resistant markers at the correct location on the genome was validated by PCR in which the reverse primers are targeted downstream to the inserted 3′UTR (see Table 1 for primers and Fig. 5a, b).

AAP7 cellular localization

Localization in L. donovani was pursued using immunofluorescence technique as follows: The N terminus of the LdAAP7 ORF was fused to a hemagglutinin tag. This chimera was cloned into the pNUS-HnN expression vector (Tetaud et al. 2002) between the 5′ XhoI and 3′ KpnI sites. L. donovani 1SR was transfected with the plasmid using standard electroporation conditions (LeBowitz et al. 1990; Jiang et al. 1999). Transfected parasites were selected on medium 199 agar plates containing 50 μg/mL G418. Mid-log L. donovani promastigotes expressing pNUS HnN HA-AAP7 were washed twice with PBS, fixed in 1% formaldehyde/PBS on a slide for 10 min and then permeabilized by exposure to 0.2% Triton X-100/PBS for 10 min. Then blocking solution (10% milk/PBST) was added and the cells were incubated for 30 min at room temperature. Then cells were incubated for 1 h at room temperature with polyclonal rabbit anti-HA (1:200) and a further 1 h in the dark with secondary polyclonal goat anti-rabbit IgG Dy-light 549 (1:500; red, Jackson Inc.). Finally, cells were washed with PBST and 5 μl of 0.5 μg/mL supplemented DAPI (Fluka). Fluorescence analyses were carried out using a fluorescent microscope (Axiovert 200 M-Zeiss).

Cellular localization in T. cruzi was done as follows: Freshly grown T. cruzi epimastigotes transfected with pTREX-GFP::TcAAP7 were washed twice with PBS. Cells were allowed to settle for 30 min at room temperature onto poly-l-lysine coated coverslips before fixation with 3% formaldehyde in PBS at room temperature for 15 min. DAPI (1.5 μg/mL) was supplemented to visualize the DNA. Cells were observed using an Olympus BX60 fluorescence microscope. Images were recorded using an Olympus DP71 camera and processed using the Olympus DP software.

Transport assays

S. cerevisiae transport studies

Logarithmically growing S. cerevisiae strain 22∆7AA cells were harvested at an OD600nm of 0.8, washed twice with water and resuspended in 0.6 M sorbitol to a final OD600nm of 8. Before transport measurements, the cells were supplemented with 100 mM glucose and 50 mM potassium phosphate pH 4.5 and incubated for 5 min at 30°C. To start the reaction, 130 μL of this cell suspension was added to 130 μL of the same buffer (0.6 M sorbitol, 100 mM glucose, 50 mM potassium phosphate pH 4.5) containing 37 to 92.5 kBq labeled 3H l-lysine and appropriate amounts of unlabelled lysine. Samples were removed after 30, 60, 120, 180 and 300 s, transferred to 4 mL of ice-cold potassium phosphate buffer (50 mM, at appropriate pH), filtered onto glass fiber filters and washed with 9 mL of the same buffer. The amount of tritium on each filter was determined by liquid scintillation spectrometry. The transport activity of S. cerevisiae mutants transformed with the empty vector pDR195 served as background and was subtracted from the observed transport measurements. At least three independent repeats of each transport measurement were performed and the mean calculated. When investigating pH dependence, the pH was adjusted prior to the transport experiment; cells resuspended to a final OD600nm of 8 in 0.6 M sorbitol and diluted to a final OD600nm of 6 by adding 0.33 vol of 200 mM phosphate buffer (pH 4.5–7.5) containing 0.6 M sorbitol.

L. donovani transport studies

Logarithmically growing promastigotes were washed twice in ice cold Earl’s buffer and concentrated to 108 per mL. This cell suspension was mixed with reaction mix (Earl’s buffer, 5 mM glucose, 10 mM Tris and 10 mM succinate) at the appropriate pH to a final volume of 570 μl. Cells were pre-incubated for 10 min at 30°C before transport was started. Transport startled by supplementing 30 μl of lysine buffer (3H l-lysine plus non-labeled l-lysine at the appropriate concentration) at the appropriate pH. 100 μl samples were taken at 30, 60, 120, 180 and 300 s and placed directly on 24 mm GF/C glass microfiber filters (Whatman 1822 024). Filters were washed twice with ice cold earl’s buffer at the appropriate pH and soaked in scintillation liquid. The amount of tritium on each filter was determined by liquid scintillation spectrometry.

T. cruzi transport studies

Epimastigotes (107 parasites) were grown for the indicated periods, harvested by centrifugation at 8,000×g for 30 s, washed twice with phosphate-buffered saline (PBS) at pH 7.0 and resuspended in 0.1 mL PBS. Transport was started by adding 0.1 mL of transport mixture that contained 300 μM (or indicated concentration) 3H l-lysine (Perkin Elmer; 2 μCi per assay). Following incubation for the indicated time at 28°C, cells were centrifuged as above and washed twice with 1 mL of ice-cold PBS. Pellets were re-suspended in 0.2 mL of 0.2 N NaOH/0.1% SDS and radioactivity measured using UltimaGold XR liquid scintillation cocktail (Packard Instrument Co., Meriden, CT, USA). A control transport experiment in the presence of 10 mM l-lysine was performed to assess non-specific transport and carry over. Each transport assay was performed at least in triplicate. Cell viability was assessed by direct microscopic examination. Kinetic constants were calculated following the procedures of Lineweaver and Burk as described by Dixon and Webb (Dixon and Webb 1964).

Results

Functional expression in S. cerevisiae and sequence analysis

We took advantage of S. cerevisiae mutant strains impaired in the transport of various amino acids (see list of mutants in Table 2 of Shaked-Mishan et al. 2006) to screen and functionally characterize lysine transporters from L. donovani and T. cruzi. In the published genome of Leishmania infantum (http://www.tritrypDB.com), the most closely related specie to L. donovani, there are twenty-five putative AAPs (Akerman et al. 2004). Using L. infantum genome, in this work ten genes were amplified from L. donovani, cloned and expressed in the different mutant yeast strains. Among these genes, LdAAP7 (LinJ32_V3.2800) and its ortholog gene from T. cruzi (Tc00.1047053511545.80), enabled growth only of the S. cerevisiae strain 22∆6AAL, a lysine auxotroph (Fischer et al. 2002), on selective lysine concentrations (Fig. 1a). To determine transport characteristics, LdAAP7 and TcAAP7 were expressed in another S. cerevisae strain called 22∆7AA that has an additional HIP1 mutation and in contrast to 22∆6AAL is not a lysine auxotroph and do not require the supplementation of the dipeptide ‘lys-asp’ for its growth (Fischer et al. 2002). As expected, LdAAP7 and TcAAP7 expressing 22∆7AA cells took up 3H-l-lysine in a time dependent manner whereas cells transformed with empty vector took up negligible amounts of lysine (Fig. 1b) supporting the observation that these AAP7s function as lysine transporters. Kinetic analysis of initial lysine transport rates indicated a K m of 7.36 ± 3.6 μM for LdAAP7 (Fig. 1c) with optimal transport at pH 4.5 (Fig. 1d). Notably none of the amino acids listed in Table 2, even when supplemented at 10- and 50-fold concentrations, inhibited LdAAP7-mediated lysine transport significantly.

Table 2 Specificity of lysine transport
Full size table
Fig. 1
figure 1

Leishmania donovani AAP7 and Trypanosoma cruzi AAP7 mediate lysine transport in S. cerevisiae. S. cerevisiae strains were transformed with either pDR195-LdAAP7 (filled circles), pDR195-TcAAP7 (filled squares) or pDR195 (open circles). a Growth on non-selective (minimal medium with 1 g/L urea and the dipeptide ‘lys-asp’—left) and lysine selective (minimal medium with 1 g/L urea—right) media (strain 22∆6AAL). Number of cells plated is indicated above each column. b Time course of lysine transport (strain 22Δ7AA). Transport assays were performed with 1 × 106 cells mL−1 at pH 4.5 and 11 μM 3H l-lysine. c Kinetic analysis of LdAAP7-mediated initial lysine transport (strain 22Δ7AA). The data are mean values of three independent experiments ± SD. Initial transport rate was determined after 5 min of incubation, at indicated lysine concentrations, pH 4.5 and 30°C. d pH profile of LdAAP7-mediated lysine transport (strain 22Δ7AA). The data are mean values ± SD (n = 3). One hundred per cent transport corresponds to 11.2 ± 4 pmol lysine per minute per 106 cells. Initial transport rate was determined after 5 min of incubation at the time points indicated in panel B, and at indicated pH, 20 μM 3H l-lysine and 30°C

Full size image

LdAAP7 (accession number ABD64602) is a single copy gene on chromosome 32 of L. donovani. The encoded protein is 504 amino acids long and contains 11 predicted trans-membrane domains and as such belongs to the large amino acid/auxin permease family (AAAP; TC 2.A.18;(Busch and Saier 2003; Akerman et al. 2004). According to TriTrypDB (http://www.tritrypDB.com) this gene is syntenic with L. infantum LinJ32_V3.2800, L. major LmjF32.2660, L. braziliensis LbrM32_V2.2900, T. brucei Tb11.01.7500 and T. cruzi Tc00.1047053511545.80. Therefore, we named these genes LinAAP7, LmjAAP7, LbrAAP7, TbAAP7 and TcAAP7 (formerly, TcAAAP545), respectively. The amino acid sequences of LdAAP7 and LinAAP7 are identical and henceforth we describe characterization of only LdAAP7. Phylogenetic analysis performed using the amino acid sequences of these trypanosomatid AAP7 proteins and various plant, bacterial, yeast and mammalian lysine transporters revealed that the putative AAAPs from all members of the trypanosomatid family form a closely related group (Fig. 2). Additionally, the analysis showed that lysine transporters from Arabidopsis thaliana (ATF/AAAP) are contiguous to trypanosomatid AAP7s (Rentsch et al. 2007), supporting that the latter function as lysine transporters. Most of the other lysine transporters in the analysis belong to the APC super family and accordingly, appear quite distinct from the putative trypanosomatid permeases.

Fig. 2
figure 2

Phylogenetic analysis of global lysine transporters. Radial phylogenetic tree of 19 lysine transporter genes including the putative amino acid permeases from Trypanosomatidae. Trypanosomatid transporter sequences with the following ID numbers were taken from GeneDB (http://www.genedb.org): LinAAP7-LinJ32_V3.2800, LmAAP7- LmjF32.2660, LbAAP7- LbrM32_V2.2900, TcAAP7- Tc00.1047053511545.80 and TbAAP7- Tb11.01.7500. Other lysine transporter sequences with the following accession numbers were taken from Genbank (http://www.ncbi.nlm.nih.gov): from S. cerevisiae, Gap1- P19145, Can1- XP_714306.1 and Lyp1- CAA47729; from bacteria, LysP (E. coli) -NP_416661.1, LysP (Corynebacterium glutamicum)- NP_600195.1; from A. thaliana, LHT1- AAC49885.1, AAT1- NP_193844.2, AAP5- NP_175076.2, AAP6- NP_199774.1; from mammals (Rattus norvegicus), CAT1- XP_859317.1, CAT2 -AAD40315.1, CAT2B- NP_072141.2 and CAT3 NP_058913.1. Colors indicate different phyla; trypanosomatids are colored violet, plants green, bacteria cyan, yeasts blue and mammals are colored orange. Local support values are shown for major clusters. See text for details (color figure online)

Full size image

Over expression and subcellular localization

To confirm that TcAAP7 and LdAAP7 mediate lysine transport also in parasite cells, we expressed them ectopically in epimastigotes and promastigotes, respectively, and subjected the transgenic parasites to transport analysis. With regards to growth rate and morphology, the phenotypes of TcAAP7 and LdAAP7 over expressing parasites were identical to those of wild type parasites (data not shown). In line with the S. cerevisiae data demonstrating that TcAAP7 functions as a lysine permease, T. cruzi parasites transfected with pTREX-TcAAP7 exhibited a transport rate about 50-fold higher than parasites transfected with empty vector (Fig. 3c). As expected, Northern blot analysis confirmed elevated TcAAP7 mRNA expression (Fig. 3a). In contrast, ectopic expression of LdAAP7 in L. donovani promastigotes grown in medium 199 did not increase lysine transport activity despite the elevated LdAAP7 expression evidenced by Northern blot analysis (Fig. 3b, c). We reasoned that this could be due to saturating levels of lysine in the growth medium (that contains 0.35 mM lysine) and therefore repeated the experiment under reduced lysine conditions. However, no significant change in lysine transport was observed in L. donovani promastigotes over expressing LdAAP7 grown in medium 199 containing no lysine and dialyzed fetal bovine serum (not shown).

Fig. 3
figure 3

LdAAP7 and TcAAP7 mediate lysine transport in L. donovani promastigotes and T. cruzi epimastigotes, respectively. a Northern blot analysis of total RNA samples obtained from wild type T. cruzi transfected with pTREX (control) or pTREX-TcAAP7 (OE). The upper band in both lanes is the endogenous TcAAP7 mRNA whereas the lower band present only in lane 2 is pTREX-TcAAP7 mRNA. b Northern blot analysis of total RNA extracted from wild type L. donovani promastigotes transfected with pNUS HnN (Control) or pNUS HnN-AAP7 (OE). Transcript sizes are indicated. The three ribosomal RNA bands serve as loading controls. c Initial rate of lysine transport in log phase epimastigotes and promastigotes versus AAP7 over expressing parasites. For T. cruzi, transport experiments were performed over 20 min at 150 μM 3H l-lysine, pH 7 and 28°C. 100% transport corresponds to 0.022 ± 0.0021 nmol/min per 108 cells. For L. donovani, transport experiments were performed over 5 min at 10 μM 3H l-lysine, pH 5 and 30°C. 100% transport corresponds to 0.9 ± 0.2 nmol/min per 108 cells

Full size image

Next, we confirmed the substrate specificity of lysine transport in wild type L. donovani promastigotes and T. cruzi epimastigotes (Table 2) and in T. cruzi parasites over expressing TcAAP7 (data not shown). Twelve amino acids were assayed as competitors in transport experiments, but none inhibited lysine transport significantly in either parasite. Notably, TcAAP7 over expressing parasites displayed the same substrate specificity as wild type epimastigotes. In summary, these data show that LdAAP7 and TcAAP7 function as lysine-specific permeases in L. donovani promastigotes and T. cruzi epimastigotes, respectively. Notably, the identical amino acid specificity of LdAAP7 both in yeast and promastigotes strongly suggest that in promastigotes, LdAAP7 is the sole lysine transporter.

To investigate the subcellular localization of these AAP7 proteins, HA-tagged LdAAP7 was expressed in L. donovani promastigotes and GFP-tagged TcAAP7 in T. cruzi epimastigotes. In L. donovani promastigotes the transporter was localized to the surface membrane and flagella (Fig. 4a). However, in T. cruzi epimastigotes the transporter was localized mainly to a membrane-bound structure or invagination close to the kinetoplast, the latter corresponding to the flagellar pocket or associated structures, such as the cytostome or contractile vacuole (Fig. 4b, c). These data are consistent with the recently published localization of a putrescine-cadaverine transporter (Hasne et al. 2010). Of note, parasites over expressing GFP-tagged TcAAP7 were demonstrated to exhibit elevated lysine transport levels similar to parasites over expressing TcAAP7, confirming the functionality of the GFP-tagged permease (data not shown).

Fig. 4
figure 4

Cellular localization of AAP7 in L. donovani and T. cruzi. a Immunofluorescence images of L. donovani promastigotes expressing HA-tagged LdAAP7. Cells were stained for HA (red) and with DAPI (blue). Scale indicates 10 μm. b, c Fluorescence images of T. cruzi epimastigotes expressing GFP-tagged TcAAP7 (green). Cells were stained with DAPI (blue). Arrow indicates the flagellar pocket (FP), plasma membrane (PM), the positions of the nucleus (N) and kinetoplast (K) are also indicated (color figure online)

Full size image

LdAAP7 is essential for L. donovani survival

To determine if LdAAP7 is essential, we carried out gene replacement as described originally by Cruz et al. (Cruz et al. 1991). This procedure involves two steps, as the gene must be deleted from each Leishmania allele (Fig. 5a). The LdAAP7 ORF present on one allele was replaced with the gene coding for hygromycin resistance to generate heterozygous mutants. Subsequently, the second allele was replaced with the gene coding for G418 resistance. This step yielded only 10 colonies that were resistant to both antibiotics. PCR was employed to confirm insertion of both antibiotic resistance genes in the expected orientation within the L. donovani promastigote genome (Fig. 5b). This notwithstanding, an LdAAP7 ORF could still be amplified by PCR (Fig 5b) and Northern blot analysis evidenced LdAAP7 RNA expression, though at significantly lower levels than that observed in wild type parasites (Fig. 5c). Analysis of Solexa sequencing runs performed using genomic DNA from L. donovani 1SR (the strain used in this work) has indicated trisomy of a few L. donovani chromosomes, in particular chromosome 31 (Myler, P.J., personal communication). However, such data indicates there are only two alleles of chromosome 32 that encodes LdAAP7. Therefore, we suspect that our gene replacement procedure induced duplication of the region coding for this transporter, suggesting that LdAAP7 is indeed an essential Leishmania gene. In line with this premise, the duplicated gene was found to be functional, with the rate of lysine transport in the mutants comparable to that observed in wild type parasites (Fig. 5d). Of note, the lower LdAAP7 mRNA expression but almost wild type lysine transport displayed by the mutants suggests that translational or posttranslational up-regulation of LdAAP7 expression is occurring in the mutant.

Fig. 5
figure 5

Elimination of LdAAP7 from L. donovani genome. a Strategy of knockout (KO) procedure. Primer locations in the LdAAP7 ORF and in the G418/Hyg fragments replacing the LdAAP7 ORF are indicated as well as expected PCR fragment lengths. b PCR on genomic DNA extracted from wild type promastigotes (AAP7/AAP7), promastigotes after the first step (AAP7/Hyg) or promastigotes after the second step (G418/Hyg). Reaction made with primers targeted to: forward: middle of G418 resistance gene and reverse: downstream to the 3′ flanking region (Left), Forward: middle of hygromycin, reverse: downstream to the 3′ flanking region (Middle) and forward: middle of LdAAP7 ORF, reverse: downstream to the 3′ flanking region (right). c Northern blot analysis on total RNA extracted from same cells. Membrane was probed with LdAAP7 ORF. d Lysine transport of logarithmic phase wild type promastigotes (filled squares) and promastigotes after the first step of the knock out (filled diamonds). Transport assays were carried out at 10 μM 3H l-lysine and pH 7

Full size image

Since lysine is an essential amino acid and the aforementioned replacement study suggests that LdAAP7 is an essential gene, we conclude that LdAAP7 is the sole lysine transporter in L. donovani promastigotes. Hence, lysine transport analysis in intact parasite cells reflects LdAAP7 activity. When performing the same gene replacement procedure with T. cruzi epimastigotes, the first TcAAP7 allele was successfully replaced by a G418 resistance marker but replacement of the second allele failed.

Regulation and kinetic analyses of lysine transport

Assuming that LdAAP7 is the only lysine permease in L. donovani, we carried out more detailed functional analysis of Leishmania lysine transport. Transport of lysine by L. donovani promastigotes was found to increase linearly with time (Fig. 5d). The K m of lysine transport in intact promastigotes was 3 ± 0.4 μM with a V max of 0.27 ± 0.03 nmol/min per 108 cells (Fig. 6a); this matches the K m value determined using S. cerevisiae expressing LdAAP7 (Fig. 1c). However, in the range tested, the optimum pH for lysine transport by promastigotes was pH 6.5, which differs from that determined using S. cerevisiae expressing LdAAP7 (Figs. 6b, 1d, respectively). To ascertain if glucose influences lysine transport we suspended promastigotes in Earl’s salt solution with and without 5 mM glucose. No significant effect of glucose on lysine transport was observed (Fig 6c). Then, to study the role of cations (Na+ and K+) in lysine transport, promastigotes were suspended in a solution at pH 7 in which both potassium and sodium had been replaced by choline (Mazareb et al. 1999). The initial transport rate was unchanged in the choline solution indicating that LdAAP7 is a cation-independent lysine transporter (Fig. 6c).

Fig. 6
figure 6

Lysine transport in L. donovani promastigotes and T. cruzi epimastigotes. a Kinetic analysis of lysine transport in L. donovani promastigotes. The data are mean values of three independent repeats ± SD. Initial transport rate was determined over 5 min of incubation, at the indicated lysine concentrations, pH 7 and 30°C. b pH optimum of l-lysine transport in L. donovani promastigotes. Initial transport rate was done as indicated for panel B. Transport assays were carried out at 5 μM 3H l-lysine, pH 4.5–7 and 30°C. The data are mean values of four independent repeats ± SD. 100% corresponds to 0.4 ± 0.12 nmol/min per 108 cells. c The effect of glucose and cations on lysine transport in L. donovani promastigotes. 100% transport corresponds to lysine transport in Earl’s buffer containing 5 mM glucose. Transport assays were carried out at 5 μM 3H l-lysine and pH 7. Results indicate the mean values of three independent repeats ± SD. d Kinetic analysis of lysine transport in T. cruzi epimastigotes. Initial rates of lysine transport (V 0) were measured as a function of lysine concentration in the range 1–30 μM. The data are mean values of three independent repeats ± SD. e pH optimum of lysine transport in T. cruzi. Initial transport velocities (V 0) were measured at pH ranging from 4 to 8. 100% transport corresponds to 0.023 nmol/min per 108 cells. Obtained values were evaluated using a one-way ANOVA followed by a Bonferroni’s multiple comparison test between all pHs and all the comparisons P values were non-significant (P < 0.001)

Full size image

In the insect vector, Leishmania promastigotes exist in two forms, proliferating non-virulent procyclic promastigotes and non-dividing virulent metacyclic promastigotes. In axenic culture, log phase cells correspond to procyclics whereas metacyclics develop during late stationary phase (da Silva and Sacks 1987; Sacks and Perkins 1984). To assess if lysine transport changes during promastigote development we determined the transport rate of cultures at different stages of axenic growth. Accordingly, a culture was inoculated at 5 × 105 promastigotes per mL and subsequently lysine transport assayed after 48 and 72 h. We have shown previously that after 4 days such a culture reaches stationary phase as indicated by expression of the metacyclic-specific gene SHERP1 (Saxena et al. 2007). Thus, after 48 and 72 h the cultures contain mid- and late log phase cells, respectively. Lysine transport decreased dramatically after 72 h, indicating that lysine transport is indeed influenced by development (Fig. 7a). For comparison, we checked the influence of development on arginine and proline transport and found that transport of these amino acids was reduced as cultures aged but less so (Fig. 7b, c, respectively).

Fig. 7
figure 7

Effect of culture age on lysine transport in L. donovani promastigotes. Initial transport rates were measured after 48 h (log phase) and 72 h (stationary phase) of growth for a 1 μM 3H l-lysine, b l-arginine, and c l-proline at pH 7 and 30°C. Promastigote initial cell density was 5 × 105 cells/mL

Full size image

Although we did not manage to establish if TcAAP7 is the only lysine permease in T. cruzi, we proceeded to characterize lysine transport by this parasite. Unlike what we observed for L. donovani, in T. cruzi epimastigotes lysine transport increased linearly with time only for the first 10 min, the transport rate dependent on lysine concentration but saturating at lysine concentrations over 100 μM. V max and K m values, 0.024 ± 0.001 nmol/min per 108 cells and 23.4 ± 2.3 μM, respectively, were generated from Lineweaver–Burk plots of concentration-dependent initial influx rates (V i) (Fig. 6d). Lysine transport was assessed in epimastigote cells after starvation for 2 h in PBS and in similarly starved cells but when the PBS had been supplemented either with glucose, lysine, arginine, proline or glycine (Table 3). We found that lysine transport increased 140% after 2 h of starvation in PBS and about 400% when the PBS contained glucose. Of note, parasites starved in PBS-lysine exhibited lysine transport rates decreased by fourfold supporting the existence of lysine sensing mechanisms; this effect was not observed for the other amino acids tested. Next, we examined if pH influences T. cruzi lysine transport. In accordance with previous studies concerning the pH dependence of amino acid transport in T. cruzi epimastigote cells (Canepa et al. 2004; Canepa et al. 2005; Canepa et al. 2009; Pereira et al. 1999; Silber et al. 2006; Tonelli et al. 2004) and unlike what we observed for L. donovani, lysine transport was constant between pH 4 and 8 (Fig. 6e). Finally, in a similar manner to that described for L. donovani above, we examined if T. cruzi lysine transport is influenced by development. As seen with L. donovani, lysine transport rates decreased with culture age; lysine transport was 11-fold lower on culture day 14 relative to day 4 (Table 3).

Table 3 Regulation of lysine transport in T. cruzi
Full size table

In summary, in L. donovani promastigotes lysine transport increases linearly with time and is mono specific, pH dependent, glucose independent and influenced by developmental stage. In contrast, in T. cruzi epimastigotes lysine transport is pH independent, glucose dependent and influenced by developmental stage.

Discussion

Lysine is used mostly for protein synthesis and is an essential amino acid for non-plant eukaryotic cells. Accordingly most eukaryotic cells, including Leishmania and Trypanosoma, acquire it from their environments using specific permeases. Here we identified and characterized genes that encode lysine permeases in L. donovani and T. cruzi, LdAAP7 and TcAAP7, respectively. Both proteins are mono-specific, high affinity, and low capacity transporters. LdAAP7 translocates lysine against its concentration gradient in a K+ and Na+ independent mechanism; with [lysine]out = 25 μM and [lysine]in = 3.8 mM (Darlyuk et al., 2009). The specificity of these trypanosomatid permeases for lysine supports our original premise that, unlike higher eukaryotes, Leishmania and Trypanosoma separate lysine from arginine transport completely (Shaked-Mishan et al. 2006). Mammalian cells, including macrophages that host Leishmania, translocate both amino acids through cation amino acid transporters (Christensen 1990; Closs et al. 1993; Kim et al. 1991).

Leishmania donovani LdAAP7 is syntenic with L. infantum LinAAP7, L. major LmjF32.2660 and L. braziliensis LbrM32_V2.2900 (http://www.tritrypDB.com). Thus we suggest that these latter proteins are also lysine transporters. This is corroborated by our finding that TbAAP7 mediates lysine transport (Inbar et al. unpublished). Further support for the premise that trypanosomatid lysine transporters differ from those in other eukaryotes is provided by our phylogenetic analysis, which highlights that trypanosomatid permeases belong to the AAAP family whereas the permeases of other eukaryotes are members of the APC family.

Our attempts to delete LdAAP7 from the L. donovani promastigote genome failed due to gene duplication; Cells after two gene replacements were resistant to both G418 and hygromycin. PCR indicates both antibiotics at the correct orientation with an additional allele that contains AAP7 ORF. Moreover, Solexa sequencing analyses indicate that the L. donovani 1SR chromosome 32 that carries LdAAP7 is present at only two alleles (Myler P.J., personal communication). Therefore, we surmise that the LdAAP7 duplication observed in the knockout experiments suggests that this gene is essential for parasite survival and is the only lysine transporter expressed in promastigotes. Of note, this finding that L. donovani promastigotes express only one lysine transporter represents another difference in lysine transport between these parasites and their hosts. For in general, higher eukaryotes encode at least two distinct transporters that can translocate lysine; for example, Lyp1p, and Gap1p in S. cerevisiae and CAT 1, 2 and 3 in mammalian cells (Ito and Groudine 1997; Reviewed in Malandro and Kilberg 1996). We suspect that arginine and not lysine is the priority substrate of the arginine/lysine transporters of higher eukaryotes and this could explain why these organisms evolved to encode more than one such transporter.

Several attempts to delete both copies of TcAAP7 in T. cruzi epimastigotes failed. This failure could be interpreted as preliminary evidence for the uniqueness and essential nature of the T. cruzi lysine permease; further studies must validate this interpretation. This notwithstanding, another indication that the T. cruzi lysine transport system is unusual is its pH-independence. Moreover, an excess of lysine in starvation medium inhibited lysine transport whereas glucose promoted transport. These observations point to the existence of energy-dependent substrate-sensing mechanisms that regulate T. cruzi lysine transport. It is notable that the emerging phenotype of T. cruzi lysine transport resembles glucose-dependent arginine homeostasis in L. donovani (Darlyuk et al. 2009). A further indication of the unique nature of TcAAP7 is its location next to the flagellar pocket. The flagellar pocket constitutes an invagination of the plasma membrane where the flagellum exits the cytoplasm and is involved in exocytosis, endocytosis, cell polarity and cell division (Field and Carrington 2009). Several flagellar pocket-associated proteins have been identified, however, this is the first report of an amino acid permease concentrated close to this structure. The specific cellular location of TcAAP7, which contrasts with the uniform distribution of LdAAP7, raises the possibility that metabolite intake from the extracellular media is somehow facilitated by this membrane structure in T. cruzi epimastigotes.

Overexpressing TcAAP7 in T. cruzi epimastigotes increased lysine transport almost by 50-fold, validating its function as a lysine transporter in parasite cells. In contrast, L. donovani promastigotes over expressing LdAAP7 exhibited no increase in lysine transport. These findings suggest that Leishmania promastigotes tightly regulate their cellular pool of lysine, which accords with the observation that the concentration of cellular lysine remains stable in these parasites even during starvation (Darlyuk et al. 2009) and development (Goldman, Rentsch and Zilberstein, unpublished). Plants also control tightly cellular lysine concentrations. The critical nature of this lysine regulation is confirmed by the abnormal phenotypes exhibited by transgenic plants engineered to have high levels of free lysine in vegetative tissues (Galili et al. 2005). Given that L. donovani appear to control cellular lysine levels more tightly than T. cruzi, it is perhaps surprising that our kinetic analysis data indicate that LdAAP7 and TcAAP7 have similar biochemical characteristics, namely similar K m and specificity. However, as described above, T. cruzi lysine transport does appear to exhibit some unusual regulatory and localization properties that could account for the global differences in lysine pool regulation between the two parasites.

Notably, both T. cruzi epimastigotes and L. donovani promastigotes reduced lysine transport as cultures aged. We observed that stationary phase parasites transported less proline and arginine as well but lysine transport in particular was almost completely shut down. It was shown in microorganisms such as yeast and bacteria that the level of general protein synthesis declines when cells enter the stationary phase (Boucherie 1985; Braun et al. 1996; Goldberg and St John 1976). Unlike other amino acids such as proline and arginine (Darlyuk et al. 2009), lysine does not appear to be involved in other processes beyond protein synthesis. Thus, if lysine transport would not be reduced in stationary phase its cellular levels will rise. Given the toxic effect of high lysine concentrations on numerous organisms, we suspect that stationary phase cells have evolved ways to shut down lysine transport.

In summary, lysine is an essential amino acid for trypanosomatids and its transport is mediated by lysine permeases belonging to the AAAP family. We show here that Trypanosomatids, unlike other eukaryotes, possess transporters dedicated to lysine. A potential ramification of this finding is that parasite lysine transporters could serve as vehicles for selective drug delivery or drug targets.