Characterisation of TcFUT1, a mitochondrial fucosyltransferase from Trypanosoma

Previous work has shown that the TbFUT1 and LmjFUT1 genes encode essential fucosyltransferases located inside the single mitochondria of the protozoan parasites Trypanosoma brucei and Leishmania major, respectively. However, nothing was known about the orthologous gene TcFUT1 or its gene product in Trypanosoma cruzi , aetiological agent of Chagas disease. In this study, we describe the overexpression of TcFUT1 with a C-terminal 6xMyc epitope tag in T. cruzi epimastigote cells. Overexpressed and immunoprecipitated TcFUT1 – 6xMyc was used to demonstrate enzymatic activity and to explore substrate specificity. This defined TcFUT1 as a GDP-Fuc : β Gal α 1 – 2 fucosyltransferase with a strict requirement for acceptor glycans with non-reducing terminal Gal β 1 – 3GlcNAc structures. This differs from the specificity of the T. brucei orthologue TbFUT1, which can also tolerate non-reducing terminal Gal β 1 – 4GlcNAc and Gal β 1 – 4Glc acceptor sites. Immunofluorescence microscopy using α -Myc tag antibodies also showed a mitochondrial location for TcFUT1 in T. cruzi epimastigote cells. Collectively, these results are like those described for TbFUT1 and LmjFUT1 from T. brucei and L. major , suggesting that FUT1 gene products have conserved function for across the trypanosomatids and may share therapeutic target potential.


Introduction
The protozoan parasite Trypanosoma cruzi is an insect-transmitted pathogen that causes Chagas disease, which is endemic throughout much of South and Central America.T. cruzi belongs to the kinetoplastid class and trypanosomatid sub-group of euglenozoan protozoa.The trypanosomatids include T. brucei, aetiological agent of human and animal African trypanosomiases, and Leishmania species that cause a variety of tropical and sub-tropical diseases in humans and animals.
For T. cruzi, human transmission occurs when an infected Triatominae insect vector feeds on human blood and leaves parasitecontaminated faeces on the skin.The metacyclic trypomastigote parasites therein enter the bite wound, or a nearby mucous membrane, and invade nucleated host cells where they differentiate into amastigote forms that replicate in the host cell cytoplasm.Some amastigote parasites differentiate into non-dividing trypomastigote forms that, upon host cell rupture, invade more host cells.Uptake of trypomastigote forms by a Triatominae insect vector during a bloodmeal leads to their differentiation to replicating epimastigote forms that colonise the insect midgut.Parasites that migrate to the insect colon differentiate into nondividing metacyclic trypomastigote forms, completing the lifecycle.About 6-7 million people are currently infected worldwide, mostly in Latin America, but migration from endemic countries has increased the number of cases detected in Canada, the United States of America, many European countries, Japan and Australia [1].
Studies on the biosynthesis of the nucleotide sugar GDP-β-L-Fucose (GDP-Fuc) in the kinetoplastids T. brucei [2] and L. major [3] have shown that it is essential.This suggested the presence of one or more essential GDP-Fuc dependent fucosyltransferases (FUTs).A single putative FUT gene (TbFUT1), belonging to the glycosyltransferase (GT) 11 family of glycosyltransferases [4], was found in the genome of T. brucei and shown to be essential for the bloodstream and procyclic life stage forms of the parasite [5].The L. major orthologue, (LmFUT1) was also shown to be essential for the promastigote form of L. major [3].TbFUT1 is also discussed in the context of the wider repertoire of T. brucei GTs in a recent review [7].
Recombinant TbFUT1 expressed in E. coli was used in activity assays and defined as a GDP-Fuc: βGal α1-2 fucosyltransferase, typical of GT11 enzymes, with an apparent preference for acceptor substrates containing a terminal Galβ1-3GlcNAc (lacto-N-biose, LNB) motif [5].Unusually, the subcellular localisation of TbFUT1 and LmjFUT1 was shown to be in the mitochondrion of these parasites [5,6].TbFUT1 and LmjFUT1 belong to a kinetoplastid-specific clade within the GT11 family.Each kinetoplastid species generally contains one, or at most two, FUT1 orthologous.Phylogenetic analysis of kinetoplastid FUT1 genes indicate they were inherited by horizontal gene transfer from bacteria via a nucleocytoplasmic large DNA virus [8].
Here we describe the characterisation of the orthologous T. cruzi gene product, TcFUT1.

TcFUT1-6xMyc overexpression construct
A plasmid containing a sequence encoding full-length TcFUT1 fused in-frame to a C-terminal 6xMyc epitope tag was purchased (GenScript).It contained an NdeI restriction site separating the tag and ORF (TcFUT1-6xMyc) and was flanked by a 5′ EcoRI site and a 3′ XhoI site.The TcFUT1-6xMyc sequence was PCR amplified using Q5 High-Fidelity polymerase and forward and reverse primers TGTCTAGAATTCATG AACAGCAGGG and CTCGAGTCAGAACAAAAACTCATCTCAGAAGAGG AT containing EcoRI and XhoI restriction sites (underlined).The purified PCR product was ligated into EcoRI and XhoI digested pTREX-G418r to make pTREX-G418r-TcFUT1-6xMyc.This cloned plasmid was sequenced and propagated by transfecting E. coli NEB 5-α (DH5α) cells, purifying the plasmid with a NEB Monarch Plasmid Miniprep kit, digesting the plasmid with NheI, inactivating the enzyme by incubation for 30 min at 80ºC, and purifying the linearized construct by ethanol precipitation and resuspension in sterile water.

Parasite transfection
Reagents and cuvettes from Human T Cell Nucleofector Kit (Lonza) were used.T. cruzi epimastigotes were grown to 5 × 10 6 cells/ml (mid log phase) and 5 × 10 7 cells were used per transfection.Cells were pelleted in 15 ml tubes (1550 x g, 10 min), resuspended in residual volume, transferred to microcentrifuge tubes and pelleted again (20 s, 16,000 g).The cell pellet was resuspended in 100 μl of Amaxa buffer (Lonza), transferred to a cuvette and mixed with 10 μg of NheI-linearized pTREX-G418r-TcFUT1-6xMyc.Electroporation was performed in an Amaxa Nucleofector II device (Lonza) using programme U-033.An electroporation without DNA was performed as a negative control.
Electroporated cells were added to 25 ml of pre-warmed growth medium and incubated for 24 h at 28 • C, 5 % CO 2 .Subsequently, G418 was added to a final concentration of 200 μg/ml and cultures were maintained for 2 weeks.Negative control cells died within one week of incubation in G418.After 2 weeks, pTREX-G418r-TcFUT1-6xMyc transfected and G418 selected cells were pelleted by centrifugation (1550 x g, 10 min) and resuspended in 10 ml of growth medium with fresh drug and maintained by dilution when they approached late log phase of growth.

Fucosyltransferase assays
Aliquots (20 μl) of the washed α-Myc agarose bead immunoprecipitate suspensions, corresponding to of 4 × 10 8 T. cruzi epimastigote cells, were incubated (12 h, room temperature, with shaking) in a final volume of 25 μl of 50 mM Tris-HCl, 25 mM KCl, pH 7.2, containing 0.3-0.4μCi of enzymatically synthesized GDP-[ 3 H]Fuc [12] that had been previously dried with a gentle nitrogen stream, and 1 mM synthetic saccharide acceptor (Dextra Laboratories and Toronto Research Chemicals).Reactions without acceptor and without acceptor and without enzyme source (i.e., using washed α-Myc agarose bead immunoprecipitate suspensions prepared from wild-type cell lysates) were performed as negative controls.Following incubation, reactions were stopped by cooling on ice and adding water to 200 μl final volume.The beads were sedimented (2000 g, 30 s) and the supernatants were desalted on mixed-bed ion-exchange columns made of 100 μl each of Chelex 100 (Na + ), over Dowex AG 50 W X12(H + ), over Dowex AG4(OH -) (BioRad), over QAE-Sephadex A25(OH -) (Merck).Columns were eluted four times with 400 μl of water.The flow through and the eluates were combined and a portion (30-40 %) dried in a centrifugal vacuum concentrator (Martin Christ) at 40ºC, 12,000 RPM.Dried material was redissolved in 150 μl of water, vacuum-dried again at 40ºC and redissolved in 5 μl 20 % 1-propanol and analysed by high-performance thin layer chromatography (HPTLC), together with non-radioactive and radioactive standards.HTLC was performed on 10 cm silica gel 60 HPTLC plates (Merck) with 1-propanol:acetone:H2O 9:6:4 (v:v:v) as mobile phase.All solvents used were AnalaR NORMAPUR grade (VWR).To improve resolution, in some cases plates were run twice before imaging.Non-radioactive sugars were visualized by orcinol/H 2 SO 4 staining, and radioactive material was detected by fluorography.

Orcinol staining
To stain non-radioactive carbohydrates, HPTLC plates were sprayed with orcinol/H 2 SO 4 solution, dried for 10 min, and heated with a heat gun until spots appeared.The staining solution was prepared by dissolving 180 mg of orcinol (Merck) in 5 ml of water, adding 75 ml ethanol, cooling on ice, and adding 10 ml concentrated sulfuric acid dropwise.The final solution was stored in the dark at kept 4 • C and brought to room temperature before use.

Fluorography
HPTLC plates with radioactive samples were sprayed three times with En 3 Hance Spray scintillation solution (PerkinElmer), leaving plates to dry after each spraying for 15 min.The dried plates were exposed to X-ray film (Sigma Carestream Biomax XAR film) and an intensifying screen (Kodak) at − 80 • C. The films were developed on a Protec ECO-MAX X Ray Film Processor.

Enzymatic characterization of fucosyltransferase reaction product
Fucosyltransferase reactions (25 μl) were inactivated by heating for 5 min at 60ºC and adjusted to 5 mM CaCl 2 and pH 5.5 by the addition of 4.4 μl 20 mM sodium acetate buffer, pH 4.5, 33.5 mM CaCl 2 .Reactions were further supplemented with 1 μl 3 mg/ml bovine serum albumin and either 2 μl α1-2 fucosidase from Xanthomonas manihotis (20 U/μl; NEB) or 2 μl water.Reactions were incubated at 37ºC for 12 h, then desalted and dried as described under Section 2.6.Half of the samples were used for HPTLC and fluorography.

Immunofluorescence microscopy
Round coverslips (13 mm diameter, WVR) were treated by dipping in 70 % ethanol, air drying, adding 150 μl of 0.1 mg/ml poly-L-lysine hydrobromide solution and incubating in the dark for 15 min at room temperature after which the solution was removed and coverslips washed twice with water and allowed to dry overnight.
Before initiating the process, T. cruzi epimastigotes cultures were supplemented for 1 h with 25 nM MitoTracker Red CMXRos (Thermo Fisher Scientific).Subsequently, 1 × 10 7 cells were harvested by centrifugation, resuspended in 1 ml ice-cold PBS and mixed with an equal volume of 8 % methanol-free paraformaldehyde in PBS.After fixation at room temperature for > 15 min, 10 μl of fixed cells (5 × 10 4 cells) were placed on the center of a treated coverslips and left for 12 h at room temperature to dry in the dark.Cells were re-hydrated by adding 200 μl of PBS and incubating for 2 min inside a humid chamber.The buffer was removed and cells permeabilized with 100 μl of 0.1 % Triton X-100 in PBS for 10 min, room temperature inside humid chamber.
Coverslips were washed once with 50 μl PBS and then blocked by incubating for 1 h at room temperature with 50 μl blocking buffer (5 % w/v fish skin gelatin, 10 % goat serum, 0.05 % Triton X-100, 0.05 % NaN 3 in PBS).Blocking buffer was then removed and coverslips washed once with 50 μl washing buffer (1 % w/v fish skin gelatin, 0.05 % Triton X-100, 0.05 % NaN 3 in PBS).Primary antibody, mouse monoclonal α-Myc (Cell Signaling Technology), was prepared by dilution (1:1000) in 0.2 µm filtered washing buffer.Aliquots (50 μl) were added to the coverslips and incubated inside humid chamber for 1 h, room temperature.Coverslips were then washed 3 times with 50 μl 0.2 µm filtered washing buffer.Aliquots (50 μl) of a 1:500 dilution of secondary antibody (goat α-mouse Alexa Fluor 488, Thermo Fisher Scientific) in 0.2 µm filtered washing buffer were added.Coverslips were incubated inside humid chamber for 1 h, room temperature and then washed 3 times with 50 μl 0.2 µm filtered washing buffer.A small drop of ProLong Gold antifade reagent with 4′,6-diamidino-2-phenylindole (Thermo Fisher Scientific) was spotted on the slide and the coverslip lowered gently onto the mounting agent.The slides were left overnight in the dark at room temperature and then sealed with nail varnish.Microscopy was performed on a Zeiss Axiovert 200 M microscope.

Identification and analysis of the TcFUT1 gene
The predicted amino acid sequence of TbFUT1 was used to search the TriTryp database [13] by BLASTp [14] in order to find orthologues in T. cruzi strain genomes.All available T. cruzi genomes contained one TbFUT1 orthologue, and for this study we selected the gene from T. cruzi Sylvio X10 strain (GeneDB ID: TCSYLVIO_003556) and named it TcFUT1.Analysis of the predicted amino acid sequence of TcFUT1 (307 amino acids) revealed that it contains all four GT11 family motifs [15,16] (Fig. 1A).However, as described for TbFUT1, motif IV appears to have evolved into a mitochondrial import sequence [5].

Expression and affinity enrichment of TcFUT1
Several attempts were made to express soluble, active TcFUT1 in E. coli and in mammalian Expi293F cells using a variety of constructs and conditions [12].These included conditions previously used to express soluble, active TbFUT1 and LmjFUT1 in E.coli [5,6].However, none of these approaches were successful for TcFUT1.Therefore, we resorted to endogenous overexpression of C-terminally 6xMyc epitope-tagged TcFUT1 in T. cruzi epimastigotes.The tagged TcFUT1-6xMyc construct was introduced into 18 S ribosomal RNA locus of T. cruzi epimastigote cells by homologous recombination following electroporation of a linearised pTREX-TcFUT1-6xMyc vector using G418 drug selection (see Materials and Methods), summarised in (Fig. 1B).This vector contains a polymerase I ribosomal RNA promoter and a short sequence (Ribo-HX1) from the upstream region of locus H1.8 of the ribosomal protein gene TcP2β [17] (Fig. 1B) and is designed to induce a strong expression of the gene of interest through integration of the plasmid into the 18 S ribosomal RNA gene locus [18].
Western blotting of total cell lysates of four separate TcFUT1-6xMyc transformed G418-selected T. cruzi populations with α-Myc antibody showed that TcFUT1 was expressed and appeared at the expected apparent MW of 42 kDa (Fig. 2A).Three of these cell cultures gave similar Western blots (Fig. 2A, lanes 2-4) while one (Fig. 2A, lane 1) showed an additional band of lower apparent MW (39 kDa), possibly the result of the removal of the predicted N-terminal mitochondrial targeting sequence.We then performed immunoprecipitation (IP) of detergent lysates of cells from cultures 1 and 4 with α-Myc agarose beads to capture, concentrate and partially purify TcFUT1-6xMyc.Aliquots of the insoluble, soluble and IP fractions were analysed by SDS-PAGE, Coomassie blue staining and α-Myc Western blotting (Fig. 2B), showing that the α-Myc agarose beads carried the majority of the soluble TcFUT1-6xMyc protein.

Fucosyltransferase assays with immunoprecipitated TcFUT1-6xMyc
The TcFUT1-6xMyc bead complexes from the IPs described above were used for on-bead fucosyltransferase assays with tritiated GDP-Fuc (GDP-[ 3 H]Fuc) alone and in the presence of lacto-N-biose (LNB) as a glycan acceptor substrate.The fucosyltransferase reaction products were separated by HPTLC and analysed by fluorography alongside authentic standards stained using orcinol reagent (Fig. 3A).When no enzyme was present, some tritiated L-fucose ([ 3 H]Fuc) was detected, indicating some degradation of the GDP-[ 3 H]Fuc to free [ 3 H]Fuc (Fig. 3A, lane 5).However, the [ 3 H]Fuc fluorographic signal was much stronger whenever TcFUT1-6xMyc containing agarose beads were present (Fig. 3A, lanes 1-4).Further, when the LNB acceptor was present, a radioactive component running slightly below the LNB standard appeared (Fig. 3A, lanes 1 and 3).These data show that, like TbFUT1 [5], TcFUT1-6xMyc transfers [ 3 H]Fuc from GDP-[ 3 H]Fuc to water (liberating free [ 3 H]Fuc) as well as to LNB.The enzymatic activity of TcFUT1-6xMyc bound to agarose beads appeared to be the same for the material from cultures 1 and 4, suggesting that processing to the lower molecular weight form does not affect activity.

The acceptor substrate specificity of TcFUT1-6Myc
To probe the acceptor substrate specificity of TcFUT1, protein-bead complexes from the IPs were incubated with GDP-[ 3 H]Fuc alone and in the presence of a panel of putative acceptor substrates selected from the literature as possible α1-2-FUT substrates [16,19,20].The effectiveness of each acceptor was assessed by the intensity of fluorographic signal for its radiolabelled product resolved by HPTLC.Partial degradation of the GDP-[ 3 H]Fuc to free [ 3 H]Fuc was observed as before (Fig. 3B, lane 10) but, again, the fluorographic signal for free [ 3 H]Fuc was much stronger when TcFUT1-6xMyc bead complexes were present (Fig. 3B, lanes 1-9).The best transfer of [ 3 H]Fuc to a glycan acceptor was seen with Galβ1-3GlcNAcβ1-O-Me (lacto-N-biose β-methyl glycoside; LNB-OMe) (Fig. 3B, lane 2), but Galβ1-3GlcNAc (LNB) and Galβ1-3GlcNAcβ1-3Galβ1-4Glc (lacto-N-tetraose) were also good acceptors (Fig. 3B, lanes 1 and 6).Based on the chemical structures of the assayed acceptors, we conclude that TcFUT1-6xMyc bead complexes were able to transfer [ 3 H]Fuc to carbohydrate acceptors containing non-reducing terminal Galβ1-3GlcNAc structures.Transfer did not occur to the β-methyl glycoside of galactose (Galβ1-OMe) (Fig. 3B, lane 8), or when the glycan contained a βGal residue bound to other positions (4 or 6) of GlcNAc (Fig. 3B, lanes 5 and 7) or to the 4-position of Glc (Fig. 3B, lane 4).When a βGal residue was linked to the 3-position of GalNAc, as in the Galβ1-3GalNAcβ1-OMe acceptor (Fig. 3B, lane 3), no product was detected either.This suggests TcFUT1 requires an equatorial hydroxyl group in at C-4 of the HexNAc residue the Galβ1-3HexNAc acceptor motif.As expected, no products other than free [ 3 H]Fuc were observed when no acceptor oligosaccharide was present in the reaction (Fig. 3B, lane 9).

Characterisation of the TcFUT1-6Myc and LNB reaction product
In another experiment, we first performed the assay (minus LNB-OMe acceptor) using IP beads from wild-type T. cruzi lysate (Fig. 4, lane 1) as well as IP beads from TcFUT1-6Myc overexpressing T. cruzi lysate (Fig. 4, lane 3).This confirmed that significant turnover of GDP-[ 3 H]Fuc to free [ 3 H]Fuc was dependent on TcFUT1-6Myc on the IP beads, and was not an artefact of incubating the beads with epimastigote lysate in general.As before, the inclusion of LNB-OMe with the TcFUT1-6Myc IP beads leads to the formation of a [ 3 H]Fuc-LNB-OMe product (Fig. 4, lane 2).When this reaction mixture was incubated with and without an α1-2 fucosidase from Xanthomonas manihotis [21], the [ 3 H]Fuc-LNB-OMe product remained stable in the absence of the fucosidase (Fig. 4, lane 4) but was completely degraded in the presence of the fucosidase (Fig. 4, lane 5).Since only the Gal residue of LNB-OMe has an -OH group at the 2-position, the positive digestion with X. manihotis α1-2 fucosidase, releasing free [ 3 H]Fuc, defines the product as [ 3 H] Fucα1-2Galβ1-3GlcNAcβ1-O-Me, and further defines TcFUT1 as a GDP-Fuc: βGal α1-2 fucosyltransferase.

Localisation of TcFUT1
Previously, the mitochondrial localization of TbFUT1 was determined by immunofluorescence microscopy, by co-localizing both Cterminally 3xMyc-tagged TbFUT1 and native TbFUT1 with the mitochondrial dye MitoTracker and with mitochondrial ATPase [5].The mitochondrial localization of LmjFUT1 was determined by immunofluorescence microscopy by co-localizing C-terminally HA-tagged LmjFUT1 with MitoTracker and by cryo-immuno-electron microscopy [6].
In our study, we used two cultures of T. cruzi epimastigotes overexpressing TcFUT1-6xMyc (cultures 1 and 4; see Fig. 1B, lanes 1 and 4) and performed localization using α-Myc antibody and MitoTracker by immunofluorescence microscopy.Since MitoTracker gives a very bright signal in T. cruzi epimastigotes, we first established conditions whereby there was no detectable signal leakage in the green α-Myc channel from the bright red (MitoTracker) channel using a no α-Myc primary antibody control experiment (Fig. 5A).Using these conditions, we then performed co-localisation analysis with α-Myc and MitoTracker for culture 1 and 4 cells (Fig. 5B).
Our results using MitoTracker were similar to those reported by other authors [22][23][24], whereby a very bright signal is observed corresponding to the location of the kinetoplast, along with less bright staining of a reticulated perinuclear network.The α-Myc staining for TcFUT1-6xMyc in these cells gave a similar reticulated labelling but a less pronounced labelling of the kinetoplast itself.These results are consistent with a mitochondrial location for TcFUT1.

Discussion
The results described in this paper extend the reports for T. brucei and L. major [5,6] to another important trypanosomatid human pathogen and associated disease; T. cruzi and Chagas disease.The fundamental details of the FUT1 genes and FUT1 gene products are conserved across these parasites, namely: (i) The trypanosomatid FUT1 genes are present in single copy per haploid genome and show synteny across the respective genomes [13].(ii) The FUT1 predicted amino acid sequences belong to the GT11 fucosyltransferase superfamily [4], except that motif IV of that superfamily appears to have evolved into an N-terminal mitochondrial import sequence.(iii) Native and/or C-terminally epitope tagged FUT1 proteins localise to the mitochondria in all three parasite species.(iv) The FUT1 gene was most likely acquired by a common ancestor of T. cruzi, T. brucei and L. major by horizontal transfer of a FUT1 gene of bacterial origin via a nucleocytoplasmic large DNA virus [8].(v) Recombinant FUT1 proteins are enzymatically active and display GDP-Fuc: βGal α1-2 fucosyltransferase activities with a preference for non-reducing terminal Galβ1-3GlcNAc acceptor sites.
Where the TcFUT1 properties described here diverge slightly from those of TbFUT1 is with respect to acceptor substrate specificity.Thus, whereas TcFUT1 is highly selective for non-reducing terminal Galβ1-3GlcNAc acceptor sites, TbFUT1 is less stringent and able to also use non-reducing terminal Galβ1-4Glc and Galβ1-4GlcNAc acceptor sites, albeit less efficiently.Whether this difference has any physiological relevance remains to be determined.
Whereas TbFUT1 and LmjFUT1 are known to be essential for T. brucei bloodstream and procyclic form and L. major promastigote cells, respectively, [5,6], the essentiality of TcFUT1 remains to be determined for T. cruzi.Nevertheless, given the five aforementioned shared characteristics, this seems likely and TcFUT1 could be considered as a potential drug target for Chagas disease, as well as for African trypanosomiases and for leishmaniases.However, the difficulty in obtaining enzymatically active TcFUT1 by heterologous recombinant expression in bacterial and eukaryotic protein expression systems [12], and the complexity of the existing radiochemical assay, that requires enzymatic synthesis of (currently commercially unavailable) GDP-[ 3 H] Fuc and chromatographic separation of radiolabelled substrates and products, are impediments to high-or medium-throughput chemical screening.While there is scope to replace the radiochemical assay with a GDP-Glo assay format [5,6,12], obtaining suitable quantities of recombinant, soluble and active FUT1s will require further attention.Although small amounts of active TbFUT1 and LmjFUT1 were obtained by bacterial expression [5,6], the low yields of soluble material and the presence of chaperone proteins in theses preparations suggests fundamental problems in recombinant FUT1 expression.One of these may be the requirement for co-expression of a partner protein (or proteins) that might assist in the folding and/or stabilisation of active FUT1.Indeed, recent data using blue native gels suggest that native TbFUT1 is present in a protein complex, consistent with this view (Samuel M. Duncan and Michael A. J. Ferguson, unpublished data).Future studies will include searches for putative FUT1 partner proteins and co-expression trials to try to enable FUT1s as tractable drug targets.

Fig. 1 .Fig. 2 .
Fig. 1.Schematic of TcFUT1 and the TcFUT1-6xMyc overexpression vector.Top: Schematic of TcFUT1, black area represents the predicted mitochondrial targeting sequence.The four characteristic GT11 family motifs are depicted as grey boxes (motifs I, II and III, amino acids 153-159, 197-207 and 265-273, respectively) or striped box (motif IV, amino acids 13-18) that appears to sit within the predicted mitochondrial targeting sequence.Below: Schematic of the pTREX-G418r-TcFUT1-6xMyc construct used to introduce TcFUT1-6xMyc into the 18 S ribosomal RNA locus of T. cruzi to drive overexpression.The construct contains a polymerase I ribosomal RNA promoter (Ribo) and an HX1 short sequence for a strong expression of the gene of interest.Integration occurs in the Ribo region upstream the ribosomal RNA locus.

Fig. 5 .
Fig. 5. TcFUT1-6xMyc localizes in the mitochondria of T. cruzi overexpressing cultures.Panel A: Preparations with no α-Myc staining were used to define the exposure conditions under which no signal was detected in the green (α-Myc) channel from crossover from the red MitoTracker channel.These conditions were used for the analyses in panel B. Panel B: TcFUT1-6xMyc over expressing culture 1 (upper panels) and culture 4 (lower panels) T. cruzi epimastigote cells stained with α-Myc after being treated with MitoTracker and counterstained for DNA with 4′,6-diamidino-2phenylindole.Comparable patterns were observed for α-Myc and MitoTracker, apart from the intense labelling of the kinetoplast by the latter.While some labelling of the kinetoplast with α-Myc was apparent, this staining was excluded from the interior of the kinetoplast.