The use of Tn5 transposable elements in a gene trapping strategy for the protozoan Leishmania

International Journal for Parasitology 37 (2007) 735–742 www.elsevier.com/locate/ijpara The use of Tn5 transposable elements in a gene trapping strategy for the protozoan Leishmania Eliane C. Laurentino a,1, Jeronimo C. Ruiz a,1,2, Loislene O. Brito a, Michael Fiandt b, Liliana M. Nicoletti a, M.C. Jamur a, C. Oliver a, Luiz R.O. Tosi a, Angela K. Cruz a,* a Departamento de Biologia Celular e Molecular e Bioagentes Patogênicos, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Av. Bandeirantes, 3900, Ribeirão Preto, 14049-900 São Paulo, Brazil b EPICENTRE Biotechnologies, 726 Post Road, Madison, WI 53713, USA Received 25 September 2006; received in revised form 17 December 2006; accepted 22 December 2006 Abstract The use of transposable elements as a gene-trapping strategy is a powerful tool for gene discovery. Herein we describe the development of a transposable system, based on the bacterial Tn5 transposon, which has been used successfully in Leishmania braziliensis. The transposon carries the neomycin phosphotransferase gene, which is expressed only when inserted in-frame with a Leishmania gene present in the target DNA. Four cosmid clones from a L. braziliensis genomic library were used as targets in transposition reactions and four insertional libraries were constructed and transfected in L. braziliensis. Clones resistant to G418 were selected and analysed by immunofluorescence in order to identify the subcellular localisation of the protein coded by the trapped gene. A definitive subcellular localisation for neomycin phosphotransferase/targeted protein fusion was not obtained in any of the four Leishmania clones investigated. However, the constructed transposable element is highly efficient considering the frequency of insertion in large targets and is therefore a useful tool for functional genetic studies in Leishmania. Our data confirm the utility of the Tn5 transposon system for insertion of sequencing priming sites into target DNA. Furthermore, the high frequency of insertion and even distribution are important in studying genomic regions bearing long and polymorphic repetitive sequences.  2007 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Leishmania braziliensis; Tn5; Transposon; In vitro transposition; Gene trapping 1. Introduction The Leishmania genus consists of more than 20 different species that are the causative agents of leishmaniasis, a human disease with a broad spectrum of clinical manifestation such as visceral, cutaneous or mucosal lesions. According to the World Health Organization (WHO), the different species of Leishmania parasites infect between 1.5 and 2 million people per year in tropical * Corresponding author. Tel.: +55 16 3602 3318; fax: +55 16 3633 1786. E-mail address: akcruz@fmrp.usp.br (A.K. Cruz). 1 These authors contributed equally to this work. 2 Present address: Centro de Pesquisas René Rachou, FIOCRUZ, Belo Horizonte, Brazil. and temperate regions of the world (WHO, May 2000). No effective vaccines against leishmaniasis are available yet (Handman, 1997) and treatment relies on highly toxic chemotherapeutic agents (Papadopoulou and Ouellette, 1993). Over the last few years several genetic tools have been introduced which allow manipulation and analysis of the Leishmania genome (Clayton, 1999). These tools include different expression vectors, positive and negative selectable markers, reporter genes, methods for gene replacement and functional complementation, and transposon-based mutagenesis, among others (Cruz et al., 1991; LeBowitz et al., 1992, 1993; Ryan et al., 1993; Ha et al., 1996; Gueiros-Filho and Beverley, 1997; Roy et al., 2000; Tosi and Beverley, 2000; Augusto et al., 2004; Denise et al., 2004). 0020-7519/$30.00  2007 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijpara.2006.12.021 736 E.C. Laurentino et al. / International Journal for Parasitology 37 (2007) 735–742 The availability of tools and strategies for the genetic evaluation of trypanosomatids and their medical relevance led the WHO/Special Programme for Research and Training in Tropical Diseases (TDR) to launch a collaborative project for the complete sequencing of the Leishmania genome (Leishmania Genome Network, LGN, http:// www.who.int/topics/leishmaniasis/en/). The reference strain chosen for the project was Leishmania major Friedlin, whose sequencing has recently been completed (Ivens et al., 2005). The Leishmania infantum genome has already been completed and so has the genome of Leishmania braziliensis (five times the genome coverage; http://www. sanger.ac.uk/Projects/L_braziliensis/). A pilot comparative study on the genomes of L. major and L. braziliensis revealed differences that reinforce the relevance of further functional/comparative analyses of the Leishmania species (Laurentino et al., 2004). Many networks are being established for the postgenomic analysis of the Leishmania, Trypanosoma cruzi and Trypanosoma brucei genomes. These studies include the use of large-scale sequence annotation and database building, microarray analysis, transcriptomics, proteomics and the continued development of new tools for direct and reverse genetics (Beverley, 2003). The development of methods to assess gene function on a genomic scale in the post-genomic age has become necessary due to the number of genes annotated as having an unknown function. Since the gene annotation process is considered to be imperfect, wide functional testing must also be conducted. The use of transposable elements in a gene-trapping strategy is a powerful tool for gene discovery, since genes are trapped regardless of their transcriptional activity or in silico detection (Kumar et al., 2004). Therefore, the generation of Leishmania insertional libraries is a promising route for studying gene content. Herein we describe the development of a transposable system based on the bacterial Tn5 transposon (Reznikoff, 2003). Considering the frequency of insertion, the constructed transposable element is highly efficient and constitutes a useful tool for the conduct of functional studies in Leishmania. 2. Materials and methods 2.1. Parasite strain, culture and genomic library A strain of Leishmania (Viannia) braziliensis (MHOM/ BR/75/M2904) was used in our studies. Promastigotes were cultured at 26 C in M199 medium (HyClone) supplemented with 10% heat-inactivated FBS (Gibco), 100 lM adenine, 10 lg/mL hemin, 40 mM Hepes (N-2-hydroxyethylpiperazine-N 0 -ethanesulfonic acid, pH 7.4), penicillin 50 U ml 1, streptomycin 50 lg ml 1 and 2% human urine (Kapler et al., 1990; Armstrong and Patterson, 1994). A L. braziliensis genomic library was constructed with high molecular weight genomic DNA, which was partially digested by Sau3AI, repaired and ligated into the shuttle cosmid, cLHYG, at the BamHI restriction site, as previously described (Ryan et al., 1993). 2.2. Construction of Tn_neosat transposon and in vitro transposition reaction An XhoI 1.7 kb fragment containing the neomycin phosphotransferase gene was obtained from the pHM3SAT2 vector (Goyard et al., 2001). The cohesive ends were filled in and the fragment was further digested with XbaI and cloned into pMOD-3 < R6K ori/MCS (EPICENTRE Biotechnologies, Madison, WI, USA). This strategy allowed the in-frame cloning and the exclusion of the initiation codon of the marker gene. The generated transposon vector, pMOD_neosat, was electroporated into Transformax EC100 Eletrocompetent Escherichia coli (Epicentre, Fig. 1). pMOD_neosat was amplified by PCR to obtain the linear transposon (Tn_neosat), following the manufacturer’s instructions (Epicentre). The linear fragment was used in a 10 ll transposition reaction containing 0.2 ll of target DNA (30 ng), an equimolar amount of the Tn_neosat and the EZ-Tn5 Transposase enzyme (Epicentre). The reaction mixture was incubated for 2 h at 37 C. Electroporation was performed using 1 ll of the mixture in the same electrocompetent cells described above. Transformed clones were selected in the presence of ampicillin and streptothricin (100 ll/ml ampicillin – AMP; 60 lg/ml streptothricin – SAT) and arrayed in 96-well microtiter dishes. For each of the original cosmids there is an array of clones which constitute the insertional library. 2.3. Sequencing and analysis Sequencing reactions used Big Dye terminator chemistry with either M13 forward or reverse primers. Single-pass sequencing was carried out on an ABI3100 sequencing apparatus (Applied Biosystems). A PHRED (Ewing and Green, 1998; Ewing et al., 1998) quality value of 20, which corresponds to an error probability of 1/100 in the base call at each position of the read, was used. Different algorithms, such as BLAST (Altschul et al., 1990) and FASTA (Pearson and Lipman, 1988), were used for sequence similarity searches against different databases (DB), including the non-redundant protein DB from the National Center for Biotechnology Information (NCBI) and the L. major genome DB (ftp:// ftp.sanger.ac.uk/pub/databases/L.major_sequences/). The default filter (DUST) was turned off in order to prevent the break of long high segment score pairs (HSPs) into smaller ones due to low-complexity segments and repeats. The annotation and graphical output were performed using ARTEMIS (http://www.sanger.ac.uk/Software/Artemis/) and in-house developed PERL (Practical Extraction and Report Language) scripts to analyse and format the results. All in silico searches and analyses described 737 E.C. Laurentino et al. / International Journal for Parasitology 37 (2007) 735–742 Tn_neosat 500 1000 EM-7 1500 2000 MCS ME ME vector SAT AG ‘NEO vector Fig. 1. Structure of the Tn_neosat transposon. The transposon length is 2169 bp and contains two inverted repeats at both ends of the linear fragment, called ME (mosaic ends), represented by an arrowhead. The transposon bears the reporter gene neomycin phosphotransferase in which an initiation codon is missing (‘NEO), a splicing acceptor site for Leishmania (AG), a bacterial promoter (EM-7) and the streptothricin acetyl transferase gene (SAT). The internal region is a 1.7 kb XhoI fragment, extracted from pMH3Sat2, which was blunt-ended and digested with XbaI for cloning into SmaI and XbaI sites at the MCS of the pMOD3 < R6Kgamma ori/MCS > vector (Epicentre). The position of the forward and reverse primers is depicted by the thin horizontal lines at both ends of pMOD backbone (vector). MCS, multiple cloning site. above, together with specific pipelines and PERL scripts developed during this work, were run in a local server using an in-house copy of DBs and software. A pipeline was established for the identification of Tn_neosat fusion events. The pipeline takes as input any sequence-trace data. PHRED software (Ewing and Green, 1998; Ewing et al., 1998) was used for the basecalling. CROSS_MATCH (Phil Green, www.phrap.org) was used for rapid nucleic acid sequence comparison, for database searches and for masking the 35-nucleotide-sequence that flanks the transposon insertion site (described in Fig. 2). A PHRED quality of 20 (one error/100 bp) and a 20-nucleotide sequence with 90% of identity were used as cut-offs for PHRED and CROSS_MATCH, respectively. The high quality data were submitted to BLASTP for the localisation of the insertion site and the determination of fusions, and to BLASTX for heterologous protein identification. PERL Scripts were written for data report extraction, DB storage and graphic report generation. 2.4. Transfection experiments and recovery of cosmid DNA from Leishmania Clonal line LB 2904 (L. (V.) braziliensis) was transfected by electroporation with 20 lg of DNA from recombinants bearing fusions or cLHYG with no insert as control. Two different protocols were used for transfection, which was conducted with low or high voltage. For the high-voltage protocol, Leishmania parasites were grown to mid-log phase, treated and electroporated as described (van den Hoff et al., 1992; Robinson and Beverley, 2003). Transfection experiments conducted with the low-voltage protocol were performed as previously described (Kapler et al., 1990). Following electroporation, cells were transferred to M 199 (1·), incubated at 26 C overnight and plated on semi-solid medium (Kapler et al., 1990). Drugs for selection were Hygromycin B (16 lg/ml) and G418 (8 lg/ml). Recovery of cosmid DNA from transfected resistant cells was performed as previously described (Ryan et al., 1993). 3’-end 35 nucleotides 3’“forward primer” NEO transposon insertion site transposon insertion site CCAACGTACCCACCACGAACGCAGCGACAGCGAGGTTTGCCAAGCGCCGGT CA target DNA sequence P T Y P P R T Q R Q R G L P S A G In silico translation of the target DNA sequence Fig. 2. Schematic representation of the process of localisation of the transposon insertion site. Between the forward primer and the insertion site there are 35 nucleotides belonging to the Tn_neosat, which were identified and masked by CROSS-MATCH. The sequence from this point on was translated in silico and submitted to BLASTP against proteome databases to identify a possible Leishmania gene fusion with NEO. 738 E.C. Laurentino et al. / International Journal for Parasitology 37 (2007) 735–742 2.5. Immunolocalisation Promastigotes were pelleted by centrifugation (2000g, 10 min), washed three times in PBS, resuspended in 60 ll of PBS (106 cells/60 ll PBS) and placed on coverslips (in 24-well plates) pretreated with 2% Biobond (Electron Microscopy Sciences, Hatfield, PA) for 15 min. Samples were then rinsed twice in PBS and cells were fixed in 2% paraformaldehyde (Ladd Research Industries, Burlington, NY) in methanol (5 min/ 20 C). Cells were permeabilised with 0.2% Triton X-100 in PBS (10 min), washed five times in PBS and treated with 0.1 M glycine in PBS (3 min). The cells were blocked for 30 min at room temperature in PBS + 1% BSA and then incubated with the antibody against the endoplasmic reticulum (15 lg/ml GRP78, Santa Cruz Biotechnology, Santa Cruz, CA) and/or anti-neomycin phosphotransferase (anti-NPT II, Europa Bioproducts). Samples were subsequently incubated with the specific secondary antibody (1:300 diluted in PBS): anti-goat TritC for anti-GRP78 (Jackson ImmunoResearch) and/or anti-rabbit Alexa 488 for anti-NPTII (Invitrogen Molecular Probes, Carlsbad, CA). The nucleus and kinetoplast were stained using 4 0 ,6-diamidino-2-phenylindole (DAPI). After extensive washing, coverslips were mounted onto slides with Fluoromont G (EM Sciences) and analysed by confocal microscopy (Leica, model TCS_NT). 3. Results 3.1. Transposition and sequence analysis Four recombinant cosmids, selected from the L. (V.) braziliensis genomic library, were targeted in transposition reactions with Tn_neosat. End sequencing was used to map Table 1 Leishmania braziliensis genes present in the cosmid insertional libraries generated Recombinant cosmid 10A11 10D02 Chromosome assignment 2 5 Predicted genesa LbrM02.0190, LbrM02.0210, LbrM02.0230, LbrM02.0250, LbrM02.0200, LbrM02.0220, LbrM02.0240, LbrM02.0260 LbrM05.0330, LbrM05.0350, LbrM05.0370, LbrM05.0390, LbrM05.0340, LbrM05.0360, LbrM05.0380, LbrM05.0400 LbrM14.0440, LbrM14.0460, LbrM14.0480, LbrM14.0500 10E07 14 LbrM14.0430, LbrM14.0450, LbrM14.0470, LbrM14.0490, 10D10 34 LbrM34.1340, LbrM34.1350, LbrM34.1360, LbrM34.1370, LbrM34.1380, LbrM34.1390 a Information extracted from L. braziliensis genome version 2 available at http://www.sanger.ac.uk/Projects/L_braziliensis/index.shtml. cosmids c10D02, c10D10, c10E07 and c10A11 within the genome (Table 1). A double-antibiotic selection was used in order to minimise the recovery of insertion events in the vector backbone. Therefore, the transposon-recipient cosmids always had an intact ampicillin resistance gene from the vector backbone and a streptothricin resistance gene provided by Tn_neosat. An average of 600 insertion events corresponding to AMP/SAT resistant E. coli transformants was obtained from each reaction. Analysis of these 600 events revealed one insertion at every 67 bp. Insertions into c10D02 were sequenced using a transposon-specific primer annealing to the 3 0 end of the transposon (Figs. 1 and 2). A pipeline for mapping and identifying the transposon insertion site and for the analysis of sequence data was developed. We analysed 376 insertional events into c10D02 and 329 (87.5%) of those occurred in the 33.7 kb insert. When aligned to the current annotation of the L. major genome (v5.2) these events mapped to a contiguous sequence in chromosome 5. Six putative genes with predicted function and two hypothetical genes conserved among trypanosomatids were annotated in this locus (Fig. 3). As depicted in Fig. 3, insertion events were evenly distributed (v2 test; P > 0.01). In the corresponding genomic region, open reading frames (ORFs) LbrM05.0380 and LbrM05.0390 are annotated as microtubule-associated repetitive protein-1 (MARP-1)-like proteins. They are similar to those of L. major, T. brucei and T. cruzi (CAB89601, AL354513.2 and AAD51095, respectively). It is noteworthy that these microtubule-associated proteins bear eight almost perfect repeats of 38 amino acid residues. These repeats are present in LbrM05.0380 and LbrM05.0390 and do not seem to impede insertion events from happening in the region. In fact, the insertion event shown in Fig. 4 happened in one of these repeats. Therefore, our data also indicate the utility of the Tn5 transposon system for high throughput sequencing. The high frequency and the even distribution of insertion observed are important to study those genomic regions bearing long and polymorphic repetitive sequences that may not be correctly assembled by currently available software. 3.2. Mapping Tn_neosat insertion events and detection of gene fusion The localisation of the insertion sites was performed on the 329 transpositions within the insert. The forward primer used to sequence and identify the transposon site of insertion is localised 35 nucleotides upstream from the mosaic end (Figs. 1 and 2). In order to precisely identify the insertion sites, this region of 35 nucleotides was used as a tag to identify the first nucleotide of the insertion site. The process of identification and masking of this tag was carried out using CROSS-MATCH. The sequences identified were translated in silico using the +1 frame from the insertion site and further submitted to BLASTP against E.C. Laurentino et al. / International Journal for Parasitology 37 (2007) 735–742 739 Fig. 3. Schematic representation of the Tn5 insertion events within c10D02. Schematic representation of a 34 kb fragment from Leishmania braziliensis chromosome 5 that has been the targeted region of transposition. The insertion sites mapped in the region are depicted as vertical short lines on the upper part of the figure. The scale for the fragment is shown below the insertion sites (k, for thousand bp). In the lower part of the figure the boxes represent those predicted genes mapped on the L. braziliensis genome Version 2.0 (available at http://www.sanger.ac.uk/Projects/L_braziliensis/index.shtml). The annotated open reading frames are: LbrM05.0320, DNA replication licensing factor, putative; LbrM05.0330, hypothetical protein, conserved; LbrM05.0340, dual specificity phosphatase-like protein; LbrM05.0350, TRYR trypanothione reductase, LbrM05.0360, ATP-dependent RNA helicase, putative; LbrM05.0370, hypothetical protein, conserved; LbrM05.0380, microtubule-associated protein, putative; LbrM05.0390, microtubule-associated protein, putative; LbrM05.0400, protein kinase, putative. Fig. 4. LB_10D02-C08/7 transfectant analysed by confocal microscopy using the GRP 78 and anti-NPT II antibodies. (A) A fragment of LbrM05 0380 (MARP1-like protein) where the insertion event has occurred. (B) Differential interference contrast image. (C) Merged immunodetection images using anti-reticulum (GRP78, red fluorescence) and anti-NEO (green fluorescence) antibodies. The genomic and kDNA were visualised by DAPI staining (blue fluorescence). The arrowhead points to an example of co-localisation between the fusion protein and endoplasmic reticulum. the proteome databases of L. major, T. brucei andT. cruzi. This comparison allowed the identification of in-frame NEO insertions. Considering its annotation, 54.8% of this locus corresponds to coding sequences. Our data show that the 6235 bp of sequences generated from these insertion events covers 31.7% of the coding regions within this locus, and 2492 bp (39.9%) correspond to sequences generated from in-frame insertions. The v-square test of the data indicates that these values are not significantly different from what was expected (P < 0.001). 740 E.C. Laurentino et al. / International Journal for Parasitology 37 (2007) 735–742 Fig. 5. Schematic representation of Leishmania braziliensis chromosome 14 represented in cosmid c10E07. Schematic representation of the 42 kb insert from c10E07 that has been the targeted region of transposition. The scale for the fragment is shown in the top horizontal line (k, for thousand bp). Boxes represent those predicted genes mapped on the L. braziliensis genome Version 2.0 (available at http://www.sanger.ac.uk/Projects/L_braziliensis/ index.shtml). Fifty-five insertion events have been sequenced and are shown as vertical short lines on the lower part of the figure. The annotated open reading frames are: LbrM14.0420, LbrM14.0430, LbrM14.0440, LbrM14.0450, LbrM14.0460, LbrM14.0470, LbrM14.0480, LbrM14.0490, LbrM14.0500. Possible L. braziliensis genes not predicted by the annotation process, CDS01 and CDS02, are represented in black boxes. In the nonredundant bank of NCBI, CDS01 matches with a glycoside hydrolase (ZP_00651374.1, 39% similarity) and CDS02 with a reverse transcriptase of Trypanosoma cruzi (CAB41693.1, 55% similarity). The in silico analysis of transposon insertions within c10D02 revealed 15 predicted proteins with at least one fusion with NEO. Among the 15 predicted trapped proteins, four are shared by all trypanosomatids analysed (T. brucei, T. cruzi, Leishmania infantum, L. braziliensis and L. major), and two of those are exclusive to the three Leishmania species studied. 3.3. Transfection and immunolocalisation of fusions Two strategies were employed in order to confirm the fusion with NEO in vivo. In the first strategy, a protein fusion was initially predicted by in silico analysis and only those recombinants bearing a potential fusion were transfected into the parasite for functional analysis. In silico prediction of fusions in c10D02 suggested the existence of 48 different insertion events leading to potential fusions between the 15 trapped proteins and NEO. We selected the c10D02-C08/ 7event, corresponding to the gene for MARP1-like protein (Accession No. CAB89601; Affolter et al., 1994; Gull, 1999) for transfection into the parasite. The transfectant obtained survived in G418 supplemented media, confirming the predicted protein fusion. This transfectant was analysed by immunofluorescence to identify the subcellular localisation of the fusion product. Immunodetection using anti-neomycin phosphotransferase II antibody (anti-NPT II) showed a diffuse distribution throughout the cytoplasm and rare colocalisation with the endoplasmic reticulum (Fig. 4). Cosmid c10D02-C08/7 was extracted from the parasite and resequenced; the predicted insertion into the MARP-1-like protein was confirmed. In the second strategy to detect fusion proteins, a pool of recombinant cosmids, which were submitted to the transposition reaction, was transfected into the parasite. Pools containing 50 and 100 clones from the insertional library were tested. An advantage of this strategy is that it allows the identification of genes that would be missed in the annotation process. Using the insertional library of cosmid c10E07, six transfectants were obtained in a 50-cosmid pool experiment. These transfectants were rescued in semi-solid media containing hygromycin B and subsequently transferred to liquid media with G418 (8 lg/ml); one clone was resistant to this drug. Such recovery indicates a fusion where NEO is being expressed in frame with a resident gene. The result suggests that it is possible to rescue missed genes with the insertional library pool approach. In fact, in the L. braziliensis annotated genome, the region represented in cosmid c10E07 contains a long region (9 kb), between genes LbrM14.0480 and LbrM14. 0490, with no predicted ORFs. The presence of two novel genes may have been overlooked (Fig. 5, CDS01, CDS02) during the annotation process, but computational parameters do not strongly suggest the presence of genes in this region. Therefore, the potential existence of these genes needs functional validation and the c10E07 insertion library generated in this study is a valuable tool for such investigation. 4. Discussion The increased need for reliable tools to conduct functional genomics studies led us to develop and evaluate a novel transposon system based on Tn5 in vitro in L. braziliensis DNA. We selected four cosmids of a L. braziliensis genomic library and followed two experimental strategies. The in silico search for fused genes with NEO is useful to study predicted genes of interest. The blind systematic transfection of pools of target DNA permits the rescue of non-annotated genes in vivo. The success of this strategy using pools of 50 cosmids is a major contribution of this work. In spite of their limitations, both approaches are complementary. Therefore, by analysing sequence data we can detect potential fusions and the transfection of pre-characterised DNA will allow further functional studies. On the other hand, as each gene is rescued, mainly based on the G418 resistance (Augusto et al., 2004), the transfection of pools of DNA allows the detection of non-annotated genes. The transposon’s ability to be inserted evenly throughout the target DNA with a reasonable frequency is an essential feature for any application of such a system. The frequency of one insertion for each 67 bp in large targets, such as recombinant cosmids, makes the systematic analysis of sequencing data from target DNA submitted to transposition reaction a reliable approach. In addition, Tn5 does not use a specific site for insertion, a relevant feature of E.C. Laurentino et al. / International Journal for Parasitology 37 (2007) 735–742 the system to achieve an even allocation of insertion events. As shown here, even through repetitive sequences a random distribution is obtained. Therefore, the presented system is undoubtedly useful for the study of large molecules such as cosmids or BACs (Wechter et al., 2002). A pipeline was created using in-house programs developed in our laboratory for management of the input data and contributed significantly to the identification of the insertion site and the detection of in-frame fusions. We have demonstrated that Tn_neosat can be used for systematic gene trapping. Several conditions need to be fulfilled for the generation of a stable fusion. These include the correct orientation of the reporter gene (NEO) and the maintenance of a reading frame, the level of expression and product stability. Moreover, the inserted element must not interfere with protein folding and sorting. We have evaluated subcellular distribution of two transfectants; both failed to maintain the correct subcellular location and possibly the folding of the original protein. Similar results have been described before with a modified mariner transposon (Augusto et al., 2004). However, other studies have successfully determined subcellular localisation of a Leishmania gene with green fluorescent protein (GFP) fusions in an extrachromosomal environment (Dubessay et al., 2006). These data show that the overexpression induced by drug pressure needed to keep the episome may or may not interfere in the correct folding and localisation of the fused product. Therefore, the usefulness of the strategy will vary depending on the structure and function of the targeted protein. Genomic integration of fused genes could be an alternative approach in these cases. In spite of the high levels of protein detected, the corresponding transcript was present at extremely low levels (data not shown), suggesting low stability of these mRNAs. This could be due to the loss of regular control and synthesis and degradation of these mRNAs. Improvements of the current tool must be carried out to prevent mislocalisation of the fused products. An alternative could be the association of the mobile element with a small tag. This technique was devised by Ross-Macdonald et al. (1999) in a largescale analysis in yeast. A transposon containing a tag smaller than the usual engineered transposons might not interfere with the folding, localisation and function of the fused proteins. The improvement of this system will be important for proteomic studies of Leishmania or any other organism. If the system is improved to localise and consequently to infer protein function, the in-frame insertions obtained by transposition can be used for systematic screening of numerous potential fusions. Considering the high frequency of insertion, the characterisation of genomic regions using transposons has an enormous potential for annotation of the L. braziliensis genome using an anchoring process, which would allow future comparative and functional studies. Large-scale functional investigation such as the one here presented will be particularly useful to validate and add information to the assembly and annotation of the L. braziliensis genome, 741 which has come to completion (Peacock et al., unpublished data). Given the depth of the genome coverage and the usefulness of L. major genome information to the L. braziliensis genome assembly, annotation inaccuracy may occur and can be solved by functional analysis. Furthermore, this pilot project has generated ready-to-use tools for functional analysis and the insertion libraries from the four recombinant cosmids are available upon request. The successful use of a mariner-based transposon designed for gene knockout in L. major has been recently described (Squina et al., 2006) and transposons can be especially designed either for gene trapping or knockouts. Nevertheless, the Tn5 transposon system presented here can be used for comparative and functional analyses of genomes and for knocking out genes. It is possible to use the insertion libraries generated in this study as a tool to target and truncate a resident gene. This is possible because any of the insertion events that happened at the 5 0 end of the coding sequence will lead to loss of the gene function. Therefore, a linear fragment from a selected clone may be transfected into the parasite for gene replacement. In some cases, when gene dosage is relevant, a phenotype will come out on the heterozygous line. Acknowledgements We thank Tânia Paula de Aquino Defina for all the sequencing data, Viviane Ambrósio Trombela for technical assistance and Márcia S.Z.Graeff for confocal microscopy assistance. We thank Renato Mortara for helpful discussions and F. Hyde for valuable comments. This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP- 99/12403-3) and Conselho Nacional de Desenvolvimento Cientı́fico e Tecnológico (CNPq - 301157/2003-0). ECL and JCR were supported by fellowships from FAPESP (00/10059-2 and 01/134619). JCR received financial support from the UNICEF/ UNDP/WORLD BANK/WHO Special Programme for Research and Training in Tropical Diseases (TDR). We used information from the L. braziliensis sequence data produced by the Pathogen Sequencing Unit at the Wellcome Trust Sanger Institute, available from the website http://www.sanger.ac.uk/Projects/L_braziliensis. References Affolter, M., Hemphill, A., et al., 1994. The repetitive microtubuleassociated proteins MARP-1 and MARP-2 of Trypanosoma brucei. J. Struct. Biol. 112 (3), 241–251. Altschul, S.F., Gish, W., et al., 1990. Basic local alignment search tool. J. Mol. Biol. 215 (3), 403–410. Armstrong, T.C., Patterson, J.L., 1994. Cultivation of Leishmania braziliensis in an economical serum-free medium containing human urine. J. Parasitol. 80 (6), 1030–1032. Augusto, M.J., Squina, F.M., et al., 2004. Specificity of modified Drosophila mariner transposons in the identification of Leishmania genes. Exp. Parasitol. 108 (3–4), 109–113. Beverley, S.M., 2003. Protozomics: trypanosomatid parasite genetics comes of age. Nat. Rev. Genet. 4 (1), 11–19. 742 E.C. Laurentino et al. / International Journal for Parasitology 37 (2007) 735–742 Clayton, C.E., 1999. Genetic manipulation of kinetoplastida. Parasitol. Today 15 (9), 372–378. Cruz, A., Coburn, C.M., et al., 1991. Double targeted gene replacement for creating null mutants. Proc. Natl. Acad. Sci. USA 88 (16), 7170– 7174. Denise, H., Coombs, G.H., et al., 2004. Generation of Leishmania mutants lacking antibiotic resistance genes using a versatile hit-andrun targeting strategy. FEMS Microbiol. Lett. 235 (1), 89–94. Dubessay, P., Blaineau, C., et al., 2006. Cell cycle-dependent expression regulation by the proteasome pathway and characterization of the nuclear targeting signal of a Leishmania major Kin-13 kinesin. Mol. Microbiol. 59 (4), 1162–1174. Ewing, B., Green, P., 1998. Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Res. 8 (3), 186–194. Ewing, B., Hillier, L., et al., 1998. Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res. 8 (3), 175–185. Goyard, S., Tosi, L.R., et al., 2001. New Mos1 mariner transposons suitable for the recovery of gene fusions in vivo and in vitro. Gene 2801–2802, 97–105. Gueiros-Filho, F.J., Beverley, S.M., 1997. Trans-kingdom transposition of the Drosophila element mariner within the protozoan Leishmania. Science 276 (5319), 1716–1719. Gull, K., 1999. The cytoskeleton of trypanosomatid parasites. Annu. Rev. Microbiol. 53, 629–655. Ha, D.S., Schwarz, J.K., et al., 1996. Use of the green fluorescent protein as a marker in transfected Leishmania. Mol. Biochem. Parasitol. 77 (1), 57–64. Handman, E., 1997. Leishmania vaccines: old and new. Parasitol. Today 13 (6), 236–238. Ivens, A.C., Peacock, C.S., et al., 2005. The genome of the kinetoplastid parasite, Leishmania major. Science 309 (5733), 436–442. Kapler, G.M., Coburn, C.M., et al., 1990. Stable transfection of the human parasite Leishmania major delineates a 30-kilobase region sufficient for extrachromosomal replication and expression. Mol. Cell. Biol. 10 (3), 1084–1094. Kumar, A., Seringhaus, M., et al., 2004. Large-scale mutagenesis of the yeast genome using a Tn7-derived multipurpose transposon. Genome Res. 14 (10A), 1975–1986. Laurentino, E.C., Ruiz, J.C., et al., 2004. A survey of Leishmania braziliensis genome by shotgun sequencing. Mol. Biochem. Parasitol. 137 (1), 81–86. LeBowitz, J.H., Cruz, A., et al., 1992. Thymidine kinase as a negative selectable marker in Leishmania major. Mol. Biochem. Parasitol. 51 (2), 321–325. LeBowitz, J.H., Smith, H.Q., et al., 1993. Coupling of poly(A) site selection and trans-splicing in Leishmania. Genes Dev. 7 (6), 996–1007. Papadopoulou, B., Ouellette, M., 1993. Frequent amplification of a short chain dehydrogenase gene in methotrexate resistant Leishmania. Adv. Exp. Med. Biol. 338, 559–562. Pearson, W.R., Lipman, D.J., 1988. Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA 85 (8), 2444–2448. Reznikoff, W.S., 2003. Tn5 as a model for understanding DNA transposition. Mol. Microbiol. 47 (5), 1199–1206. Robinson, K.A., Beverley, S.M., 2003. Improvements in transfection efficiency and tests of RNA interference (RNAi) approaches in the protozoan parasite Leishmania. Mol. Biochem. Parasitol 128 (2), 217–228. Ross-Macdonald, P., Coelho, P.S., et al., 1999. Large-scale analysis of the yeast genome by transposon tagging and gene disruption. Nature 402 (6760), 413–418. Roy, G., Dumas, C., et al., 2000. Episomal and stable expression of the luciferase reporter gene for quantifying Leishmania spp. infections in macrophages and in animal models. Mol. Biochem. Parasitol. 110 (2), 195–206. Ryan, K.A., Dasgupta, S., et al., 1993. Shuttle cosmid vectors for the trypanosomatid parasite Leishmania. Gene 131 (1), 145–150. Squina, F.M., Pedrosa, A.L., et al., 2006. Shuttle mutagenesis and targeted disruption of a telomere-located essential gene of Leishmania. Parasitology, 1–12. Tosi, L.R., Beverley, S.M., 2000. cis and trans factors affecting Mos1 mariner evolution and transposition in vitro, and its potential for functional genomics. Nucleic Acids Res. 28 (3), 784–790. van den Hoff, M.J., Moorman, A.F., et al., 1992. Electroporation in ‘intracellular’ buffer increases cell survival. Nucleic Acids Res. 20 (11), 2902. Wechter, W.P., Begum, D., et al., 2002. Physical mapping, BAC-end sequence analysis, and marker tagging of the soilborne nematicidal bacterium, Pseudomonas synxantha BG33R. Omics 6 (1), 11–21.