Amplification of L. donovani CPN10 DNA

An alignment of the known eukaryotic CPN10 amino acid sequences revealed four limited stretches of sequence conservation (not shown). Oligonucleotides CPN10.1 to CPN10.4 were designed according to the codon bias of Leishmania spp.. Using L. donovani genomic DNA as template, combinations of the aforementioned primers were tested in a polymerase chain reaction. Only the combination of primers CPN10.1 and CPN10.4 yielded an amplification product of the expected size (not shown). The amplification product was then subcloned into the vector pBluescript and subjected to sequence analysis using standard pBluescript sequencing primers. The sequence analysis confirmed that the amplification product harboured probable CPN10 sequences (not shown).

Isolation of a CPN10 genomic DNA clone

The amplified putative CPN10 gene fragment was used to screen a L. donovani genomic DNA cosmid library [3]. Positive cosmids were subjected to a restriction digest and Southern Blot analysis to identify common subfragments which harboured the CPN10 gene(s). A 1.5 kb XhoI I fragment which hybridised with the CPN10 probe was subcloned into pBluescript and subjected to sequence analysis. The sequence (GenBank AF394959) contains an uninterrupted open reading frame of 303 bp encoding a putative 10.7 kDa polypeptide. A sequence comparison by BLAST search confirmed this putative polypeptide as related to eukaryotic and prokaryotic co-chaperonins. Figure 1A shows the position of the L. donovani CPN10 in the lineage of eukaryotic CPN10 proteins. L. donovani CPN10 groups within the kinetoplastid homologs. Figure 1B shows an alignment with selected eukaryotic CPN10 polypeptides. The high degree of sequence conservation between LdCPN10 and its counterparts in higher eukaryotes indicates a conserved function.

Southern Analysis

Within the haploid set of chromosomes, L. donovani CPN10 appears to be single copy. This was determined by Southern Blot analysis of L. donovani genomic DNA, digested with 7 restriction endonucleases (Figure 2A). The pattern of restriction fragments does not indicate the presence of additional gene copies. We also performed pulsed field gel electrophoresis (Figure 2B) to determine the chromosomal localisation of the CPN10 gene. The CPN10 specific probe hybridised to a chromosomal band of approximately 1,100 kb. This size corresponds to chromosomes 26 or 27b. This was confirmed when sequencing of the L. major genome was finished; two copies of CPN10 are found on chromosome 26 http://www.genedb.org/genedb/Search?name=LmjF26.0620&organism=leish.

RT-PCR analysis

Northern analysis of L. donovani RNA using a CPN10-specific probe did not yield any bands in the expected size range. To ascertain that the putative CPN10 gene encodes a stable RNA, we performed RT-PCR. Whole cell RNA was reversely transcribed using an oligo-dT primer. The first strand cDNA was used as template in a PCR amplification using primers CPN10-1 and CPN10-4. Negative controls included an amplification of material from a cDNA synthesis mix that had been inactivated at 65°C for 10 min (Figure 2C, lane 1), and an amplification with an equivalent amount of whole cell RNA as template (lane 3), to exclude DNA contaminations in the RNA preparation. Both reactions did not give rise to an amplification product. The reaction with cDNA as template (lane 2), however, produced a ~350 bp PCR product, much the same as the reaction with genomic DNA as template (lane 4). We conclude that stable, polyadenylated RNA is produced from the CPN10 gene.

Production of anti-CPN10 antibodies

The CPN10 open reading frame was subcloned into pJC45, a derivative of pJC40 [4, 12], for expression in E. coli strain Bl21 (DE3) [13]. His-tagged CPN10 was purified by metal chelate chromatography (Figure 3) and used to immunise laying hens. Antibodies were prepared from egg yolk before and after immunisation to yield pre-immune and CPN10-specific antibody preparations. In Western Blot analyses, these antibodies faithfully recognised recombinant rCPN10 (not shown).

LdCPN10 expression is induced in the amastigote

Using the anti-CPN10 antibodies, we performed immunoblot analysis on lysates of L. donovani promastigotes before and after heat shock, and of axenically cultured amastigotes (Figure 4B). In promastigotes cultured at 25°C, we observe only a faint band in the expected size range. After a 24 h treatment of the promastigotes at 37°C, a signal is observed. The CPN10 signal increases strongly during in vitro amastigote differentiation. Figure 4A shows identical samples separated by SDS-PAGE and Coomassie Blue staining. We conclude that Leishmania donovani CPN10 is expressed preferentially in the amastigote stage, induced at least in part by the elevated temperature.

CPN10 is the co-chaperonin in L. donovani

By analogy, CPN10 is expected to act as co-chaperonin to the 60 kDa chaperonin, CPN60. However, there are two diverged members of the CPN60 gene family in L. donovani, CPN60.1 and CPN60.2. Of these, only CPN60.2 could be shown to encode a protein which localised to the mitochondrion [4]. We therefore tested whether CPN10 interacts with CPN60.2 protein in L. donovani and performed co-immune precipitation. L. donovani cell lysates were incubated with anti-CPN10 antibodies, anti-chicken IgG (rabbit) and Protein-A agarose. The precipitated material was then analysed by immunoblot using anti-CPN60.2 antibody. As shown in Figure 4C, the anti-CPN10 precipitate contains a protein which reacts with the anti-CPN60.2 antibodies. We conclude that CPN60.2 is the bona fide chaperonin and that CPN10 serves as its co-chaperonin.

Subcellular localisation of LdCPN10

Since CPN10 serves as co-chaperone to CPN60 we expected CPN10 to localise to the kinetoplast/mitochondrion complex inside the leishmaniae. To test this prediction, the CPN10 coding sequence was fused to a GFP (Green Fluorescent Protein) coding DNA in the expression vector pIRmcs3-. The linearised expression construct was recombined into the rRNA locus of L. donovani to yield stable expression of the CPN10::GFP chimera. The recombinant parasites were cultivated as promastigotes and analysed by fluorescence microscopy. The results are shown in Figure 5. We observe brightly fluorescent tubular structures in the promastigotes. No fluorescence is observed in the cytoplasm. This indicates that the CPN10 sequence can quantitatively direct GFP into a subcellular compartment or organelle.

To identify the subcellular compartment that harbours CPN10, we performed immune electron microscopy using sections of heat stressed promastigotes and of axenic amastigotes. Figure 6 shows the result. We find the gold particles associated both with tubular structures and with the kinetoplast. We conclude that CPN10 is a mitochondrial protein.

When we analysed the subcellular localisation of CPN60.2 in L. donovani, this chaperonin, too, localised to a tubular subcellular compartment. However, we did not find it associated with the kinetoplast. Therefore, the exact subcellular localisation was still unclear. Using the strain which overexpressed the CPN10::GFP chimera, we performed a co-immune electron microscopy. We used anti-GFP mAB in conjunction with anti-mouse immunogold particles (5 nm). Then the sections were co-stained using anti-CPN60 antibody, anti-chicken IgG (rabbit), and protein A gold (10 nm). As shown in Figure 7, CPN10::GFP and CPN60 colocalise to the same tubular compartment. This shows that CPN60.2 is indeed a mitochondrial protein in Leishmania.

Parasite strains and culture

For our analyses we used the Lo8 strain of L. donovani, a gift from D. Zilberstein. Promastigotes were routinely cultivated at 25°C in supplemented M199 medium [3]. Cell density was monitored using a Schaerfe Systems CASY Cell Counter. In vitro promastigote to amastigote differentiation was performed as described [3].

PCR amplification of CPN10 DNA

Four primers were delineated from an amino acid sequence comparison of known GroES and CPN10 proteins. The extreme bias in Leishmania towards G and C residues in the wobble positions of codons was used to create primers with minimal degeneracy:

primer CPN10.1: CCGCTGTTCG ACCGCGTGCT GG

primer CPN10.2: GGCGGCATCR TGCTGCC

primer CPN10.3: CGGCAGCAYG ATGCCGCC

primer CPN10.4: CAGCACCTTG TCGCCCACCT TCAC

100 ng of L. donovani genomic DNA was mixed with 40 ng of oligonucleotide primer pairs, 25 nmoles of each dNTP, 10 × reaction buffer, and 1 unit of Taq DNA polymerase (Beckman) in a 50 μl reaction. The reaction mix was incubated for 1 min at 95°C, 1 min at 55°C, and 1 min at 72°C. The cycle was repeated 35 times. 10% of the reaction was analysed on a 1% agarose gel. Negative controls included each primer by itself and omission of template DNA. Bands absent from the negative controls were recovered by preparative agarose gel electrophoresis and by using glass beads adsorption (PureGen Kit, Biozyme). PCR products were subcloned using the TA cloning kit (Invitrogen) and subjected to sequence analysis..

Reverse transcriptase (RT-) PCR

Total RNA from promastigotes stage parasites strain Lo8 was prepared using a Ribolyser Kit (Hybaid). cDNA synthesis was carried out using the First Strand synthesis kit (Pharmacia) with the enclosed (dT)₁₈ primer and 5 μg of total RNA. The cDNA was then amplified enzymatically using the primers CPN10.1 and CPN10.4 The amplification was carried out as described above.

Library screening

We used a cosmid library of L. donovani previously described in [16]. The library was screened by hybridisation with digoxigenin-labeled DNA probes derived from the subcloned amplification products. Hybridisation was performed at 65°C in HYB 9 solution (Biozyme) for 16 h. Washes were performed at decreasing SSC concentrations (6 × to 0.2 × SSC, 0.5 % SDS) at 65°C. Filters were developed using anti-digoxigenin FAB/AP conjugate (Boehringer Mannheim) followed by colorimetric staining with NBT/BCIP. Positive cosmid clones were picked, rescreened, and verified in an amplification reaction using CPN10-specific primers.

Subcloning and sequence analysis

Cosmid DNA from positive clones was subjected to restriction endonucleased digest and Southern Blot to determine suitable restriction endonucleases. Cosmid DNA was then digested at a preparative scale and subcloned into an appropriately cleaved pBluescript KS+ vector (Stratagene) or pJC45 vector [4]. Plasmid subclones were then subjected to bidirectional primer walking sequence analysis, starting from the known sequences of the PCR amplificates. Sequencing was performed on an Applied Biosystems Model 370 sequencer using the DyeTerminator Kit.

Sequence evaluation

Contig alignment, in silico translation, and sequence alignment were performed using the MacMolly Tetris and the MacVector software packages. Phylogenetic analyses were performed using the clustalW algorithm included with MacVector at default settings.

Recombinant Expression of CPN10

The putative open reading frame of CPN10 was amplified using specific primers that modify the start codon sequence into a NdeI site and the stop codon sequence into an EcoRI site. The amplification products were cleaved with NdeI and EcoRI and ligated into the expression vector pJC45 [4], between the NdeI site and the EcoRI site. The expression plasmids were transformed into the bacterial strain BL21 (DE3) [pAPlacIQ], a gift from Olivier Payet, CNRS Toulouse. Bacteria were grown in CircleGrow medium (BIO 101) supplemented with 50 μg/ml Ampicillin and 10 μg/ml Kanamycin to OD600 = 0.5. IPTG was added to 1 mM and incubation was continued for 1 h. The purification of His-tagged proteins from bacterial lysates by metal chelate chromatography has been described [12, 14].

To express CPN10::GFP chimera, the CPN10 coding sequence, minus the stop codon, was enzymatically amplified to create BamHI and HindIII sites at the 5' and 3' ends, respectively. Digested with BamHI and HindIII, the amplification products were ligated between the BamHI and HindIII sites of the yeast plasmid p426/PQ25 [17]. By this procedure, CPN10 and GFP coding sequences were fused in frame. The chimeric CPN10::GFP coding sequence was then excised with BamHI and XhoI and ligated into pIRmcs3- [18]. The plasmid pIRCPN10::GFP and the parent plasmid pIRmcs3- were linearised using SwaI and transfected into L. donovani by electroporation. Recombinant parasites were placed under ClonNAT (Werner Bioreagents) selection (100 μg/ml) and analysed by fluorescence microscopy.

Immunisation and antibody preparation

The immunisation of laying hens and the preparation of antibodies from egg yolk has been described [19].

Immunoblot analysis

SDS-PAGE and Western Transfer were performed as described [14, 19]. Briefly, membranes were treated with blocking buffer (4% skim milk powder and 0.1% Tween 20 in DPBS), with antibodies at appropriate dilutions in blocking buffer, and with secondary antibody/alkaline phosphatase conjugate (Dianova) in blocking buffer. Blots were stained with nitrobluetetrazolium chloride and 5-bromo-4-chloro-3-indolyl phosphate.

Immune precipitation

1 × 10⁸ promastigotes were harvested by centrifugation and lysed in 500 μl solubilising buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 1% NP-40, and 2.5 μg × ml-1 each of aprotinin, leupeptin, pepstatin, and antipain). Lysates were centrifuged at 13000 × g, 4°C. Soluble proteins were incubated with 20 μl anti-CPN10 antibody for 1 h at 4°C. 350 μl of protein A-sepharose slurry were preincubated with anti-chicken IgG from rabbit and added. The slurry was further incubated at 4°C for 2 h. The immunoabsorbent was centrifuged and washed three times in washing buffer A (10 mM Tris-HCl pH 7.5, 150 mM NaCl, 0,2% NP-40, 2 mM EDTA), twice in washing buffer B (as above, except NaCl is 500 mM) and once in washing buffer C (10 mM Tris-HCl pH 7.5). For analysis on SDS-PAGE 250 μl of SDS sample buffer was added and the samples were heated for 5 minutes at 95°C. After centrifugation, 50 μl of the supernatant was loaded onto a 12% PA gel.

Immune electron microscopy

Immune electron microscopy was performed essentially as described [3]. LR-White embedded microsections were treated with chicken antibody at 1:500 (anti-CPN10) or 1:2000 (anti-CPN60.2) in PBS (0.1 % Tween 20, 1% BSA). Anti-chicken IgG (rabbit) and Protein-A immunogold particles (10 nm) were used for detection. For detection of CPN10::GFP chimera, anti-GFP monoclonal antibody and 5 nm anti-mouse immunogold conjugate (Sigma) were used.

Fluorescence microscopy

L. donovani promastigotes expressing CPN10:GFP chimera were grown to 2 × 10⁷ cells ml^-1. 1 ml of parasite culture was subjected to centrifugation for 10 min at 1500 × g. The supernatant was discarded and the cell pellet was resuspended in 200 μl of Dulbecco's PBS. 2 μl was applied to a multiwell microscope slide and subjected to fluorescence microscopy at 63-fold magnification. Samples were analysed in bright field and in the FITC channel using a Hitachi Model monochrome CCD camera. Images were taken and merged using the Improvision Open Lab^® software package and exported in TIFF format. Cropping and juxtapositioning were done using Adobe Photoshop^® software.

Pulsed Field Gel Electrophoresis

Leishmania cells were harvested by centrifugation and washed twice in PBS. Following centrifugation, parasites were resuspended in PBS and mixed with an equal volume of prewarmed 1,5 InCert agarose (FMC BioProducts, Rockland). This mixture was aliquoted in block formers (2 × 107 parasites per block, Pharmacia). To lyse parasites, the agarose blocks were incubated in 2 mg/ml proteinase K (1% Laurylsarkosyl, 0,5 m EDTA pH 9.0) at 37°C for 48 h, and stored in 0.5 EDTA at 4°C.

PFGE was performed at 13°C in 0,25 × TBE buffer under the following conditions: intervall in sec: 100-10 log; switching angle: -110 lin; voltage: 200-150 log (Rotaphor, Biometra).

Saccharomyces cerevisiae strain YPH80 chromosomes (New England Biolabs) were used as size standards.

After staining with ethidium bromide (1 μg ml-1) and strand breakage by a 5 min exposure to UV radiation (254 nm), the gel was blotted onto a positively charged nylon membrane (Qiagen) by alkaline transfer [20]. The membrane was hybridised with a digoxigenin-labeled CPN10 probe. Blots were stained as described above.

Digital imaging

Primary experimental data were digitalised either on a flatbed scanner (Microtek Scanmaker 8700) or on a 35 mm film scanner (NIKON LS-4000). Phosphor imager (Molecular Dynamics) data were imported directly as TIFF images. Images were cropped and optimised for colour saturation using Adobe Photoshop^® software. No filters or other image altering functions were employed on the images or parts thereof. Images were combined with vector graphics and text using ClarisDraw^® software, version 1.0d.