Identification and Molecular Characterization of a Gene Encoding a Protective Leishmania amazonensis Trp-Asp (WD) Protein

Parasite culture and antigen preparation. L. amazonensis (MHOM/BR/77/LTB0016), L. major (MRHO/SU/P/LV39), Leishmania chagasi (MHOW/BR/00/1669), and Leishmania mexicana (a fresh patient isolate) parasites were maintained by regular passage through either hamsters (L. chagasi) or BALB/c mice (all other species). Promastigotes were cultured at 23°C in Schneider's Drosophila medium (Invitrogen Corporation, Carlsbad, Calif.) supplemented with 20% fetal bovine serum. For immunization study, L. amazonensis promastigotes of less than three in vitro passages were used for animal infection. L. amazonensis metacyclics were purified by negative selection with the 3A1 monoclonal antibody (a gift of David Sacks, National Institute of Allergy and Infectious Diseases, Bethesda, Md.) as described previously (9). L. amazonensis and L. mexicana amastigotes were cultured at 32°C in complete Schneider's medium (pH 5.0) supplemented with 20% fetal bovine serum as described previously (14). To prepare promastigote and amastigote lysates, parasites were suspended in phosphate-buffered saline and subjected to three freeze-thaw cycles and a 15-min sonication in an ice bath prior to storage at −70°C.

Mice and hamsters.Female BALB/c mice and female Syrian golden hamsters were purchased from Harlan Sprague-Dawley (Indianapolis, Ind.). All animals were maintained under specific-pathogen-free conditions. Mice were 4 weeks of age when DNA immunizations were initiated. Hamsters were 4 weeks of age when recombinant protein was injected. Animal protocols were approved by the Animal Care and Use Committee of the University of Texas Medical Branch, Galveston (UTMB).

Screening of amastigote cDNA library.A λ-Zap (Stratagene, La Jolla, Calif.) amastigote cDNA expression library (a gift of David Mosser, University of Maryland) was screened according to the manufacturer's protocol. A total of ∼2 × 10⁶ plaques were screened. Duplicate filters were probed with sera from naive and infected (8 weeks) BALB/c mice at a dilution of 1:500 for the primary antibody incubation. After incubation with goat anti-mouse-alkaline phosphatase (1:1,000, Bio-Rad, Hercules, Calif.), filters were developed by using the alkaline phosphatase substrate kit (Bio-Rad). Positive clones were selected based on reaction with infected mouse, but not naive mouse, serum. A total of five individual clones were selected for further screening. These clones were subjected to secondary and tertiary screening for confirmation, and then the phagemid was in vivo excised from the λ-Zap expression vector to form the pBK-CMV phagemid by using the manufacturer's helper phage excision protocol. Clones were sequenced at the UTMB Protein Chemistry Laboratory with a 373XL automated DNA sequencer (ABIPrism PEBiosystems).

The pBK-CMV clone containing the LAWD gene did not contain the full mRNA, as indicated by the missing spliced leader sequence. To obtain the full mRNA sequence, we performed PCR on cDNA with a spliced leader primer (5′-CTAACGCTATATAAGCTTCAGTTTCTG-3′) and a gene-specific internal primer (5′-GTCACCACGGCAGCTTCC-3′). Cycling conditions were 95°C for 2 min, followed by 30 cycles of 94°C for 1 min, 52°C for 30 s, and 72°C for 1 min. The PCR product was ligated into the EcoRV site of the pT7Blue-3 vector (Novagen, Madison, Wis.) and then transformed into competent DH10B Escherichia coli (Invitrogen). Positive clones were confirmed via sequencing.

Northern blot analysis.Total RNAs for amastigotes and promastigotes were purified by using Tri-Reagent (Sigma, St. Louis, Mo.). RNA samples (10 μg) were analyzed on standard MOPS (morpholinepropanesulfonic acid)-formaldehyde-1% agarose gels. The RNAs from the gel were transferred to a nitrocellulose membrane (MSI, Westborough, Mass.) and UV cross-linked by using a UV Stratalinker (Stratagene). Blots were hybridized by using PerfectHyb Plus (Sigma) according to the manufacturer's protocol for hybridization and washing. The DNA probe was generated by digesting the pBK-CMV clone containing the LAWD gene with BamHI and NcoI (equivalent to nucleotides 363 to 851 of the open reading frame [ORF]) and then radioactively labeled by using the Random Primers DNA Labeling System (Invitrogen). Washed blots were exposed to Kodak BioMax Film at −70°C for 20 h. For control of loading, the membranes were probed with a 525-bp fragment of the Ldp 23 gene, which is known to be expressed comparably in both promastigotes and amastigotes (7).

Recombinant LAWD (rLAWD) protein.The whole gene insert from the pBK-CMV clone containing the LAWD gene was subcloned into the pET32a plasmid (Invitrogen) for His tag protein expression. To ensure that the ORF would be read in frame, the plasmid was first digested with BamHI and then blunt-ended by a Klenow reaction (Invitrogen). This product was purified by agarose gel electrophoresis with the QIAquick gel extraction kit (Qiagen, Valencia, Calif.), digested with XhoI, and repurified by agarose gel electrophoresis. This DNA product was ligated into a pET32a vector that had been digested with EcoRV and XhoI and dephosphorylated with calf intestinal phosphatase (New England Biolabs, Beverly, Mass.). Competent E. coli BL21 (Stratagene) was transformed with the ligated pET32a-LAWD, and positive clones were confirmed through sequencing. Recombinant protein was generated by growing the bacteria at 22°C for 4 h in the presence of 1 mM IPTG (isopropyl-β-d-thiogalactopyranoside) prior to purification of the His-tagged LAWD protein via Ni²⁺-charged agarose (Sigma).

Generation of hamster antisera against rLAWD.His tag-purified rLAWD was resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), stained with 0.05% brilliant blue R-250 (Fisher Scientific) in 50% methanol, and destained with 10% acetic acid-5% methanol. The appropriately sized band was cut out and frozen in liquid nitrogen. The frozen gel fragment was homogenized by using a motorized pestle and resuspended in phosphate-buffered saline. This mixture was mixed 1:1 with Ribi adjuvant system (Corixa Corporation, Seattle, Wash.) prior to delivery. Five hamsters were subcutaneously injected with 25 μg of rLAWD emulsified with the adjuvant in five sites in the back at 2-week intervals for a total of four separate injections. Hamster antibody responses to rLAWD were evaluated by direct enzyme-linked immunosorbent assay (ELISA). All five hamsters had detectable titers against rLAWD, and the two with the highest titers were selected for immunoglobulin G (IgG) isolation with the Immunopure (A) IgG purification kit (Pierce). The elution fractions with the highest concentrations of IgG were used for Western blot and immunoelectron microscopy experiments.

rLAWD ELISA.To assess parasite-specific antibody titers, Immulon 4 microplates (Dynatech Labs) were coated overnight at 4°C with either amastigote lysate (50 μg/ml), rLAWD (10 μg/ml), or His-tagged purified pET Induction Control J (Novagen) to serve as a negative control. After blocking, the plates were incubated with individual serum samples (1:50 to 1:6,400) for 1 h at 37°C and then with horseradish peroxidase-conjugated goat anti-human IgG (1:1,000, Sigma) for 1 h at 37°C. Color was developed with ImmunoPure ABTS [2,2′-azinobis(3-ethylbenzthiazolinesulfonic acid)] tablets and hydrogen peroxide (Pierce) according to the standard protocol.

Western blotting.All SDS-PAGE analyses were performed on Protean III minigel electrophoresis equipment with precast SDS-polyacrylamide gels (Bio-Rad). Parasites (2 × 10⁸ cells/well) were lysed in 0.5% Triton X-100 containing 1 mM phenylmethylsulfonyl fluoride and 1% protease inhibitor cocktail (Sigma) for 30 min at room temperature. Samples were prepared and electrophoresed according to a standard protocol (4). Blotting of proteins onto polyvinylidene difluoride membranes (Bio-Rad) was performed for 1 h at 100 V. The membranes were blocked with 5% Blotto and incubated with the primary antibody (1:500) for 6 h at room temperature and then with horseradish peroxidase-conjugated goat anti-hamster IgG (1:5,000; Sigma) for 1 h at room temperature. The blot was developed by using ECL (Amersham Pharmacia Biotech, Little Chalfont, England) and developed on Kodak BioMax film. Purified naive hamster IgG was used as a negative control.

Immunoelectron microscopy.Amastigotes and promastigotes were fixed for 1 h at room temperature in 0.1% glutaraldehyde-2.5% formaldehyde-0.03% trinitrophenol in 0.05 M cacodylate buffer, pH 7.3. Cells were embedded in LR White resin as previously described (27). The ultrathin sections were cut on a Sorvall MT-6000 ultramicrotome (RMC, Tucson, Ariz.) and incubated with purified hamster IgG specific to rLAWD (1:100) for 1 h at room temperature and then at 4°C overnight. After washing, sections were incubated with rabbit anti-hamster IgG (heavy plus light chains) (1:50; Pierce) for 1 h at room temperature and then incubated with 15-nm-diameter colloidal gold-labeled anti-rabbit IgG (heavy plus light chains) (1:20; Amersham Biosciences, Piscataway, N.J.) for 1 h at room temperature. Immunostained grids were stained with 0.4% lead citrate prior to evaluation with a Philips 201 electron microscope at 60 kV (Philips Electron Optics, Eindhoven, The Netherlands). Purified naive hamster IgG was used as a negative control.

Human blood samples.Human serum samples (n = 37) (a generous gift of Gregory Lanzaro, University of California, Davis) were collected from Nicaraguan leishmaniasis patients diagnosed with atypical cutaneous leishmaniasis by indirect immunofluorescent-antibody test at the Ministry of Health in Managua, Nicaragua. Additionally, human serum samples (n = 7) (a generous gift of Sergio Mendonca, Instituto Oswaldo Cruz, Fiocruz, Brazil) were collected in Brazil from patients diagnosed with either mucocutaneous (5 samples) or atypical cutaneous (2 samples) leishmaniasis. Serum samples (n = 10) from individuals with no previous history of leishmaniasis were used as negative controls. Human serum samples from Trypanosoma cruzi patients were a generous gift of Nisha Garg, UTMB.

Purification and stimulation of CD4⁺ LN T cells from L. amazonensis-infected mice.CD4⁺ T cells were purified from draining lymph nodes (LN) of BALB/c mice that were infected for 4 to 5 weeks by positive selection with magnetic beads (Dynal, Lake Success, N.Y.). Briefly, LN cells (pooled from five mice per group) were incubated with appropriate amounts of anti-CD4 monoclonal antibody-coated beads for 20 min at 4°C with gentle shaking. The cell rosettes were collected by using a magnetic separator, and bound cells were separated from beads by using DETACHaBEAD mouse CD4 (Dynal). CD4⁺ LN T cells (10⁵) were cultured with either medium alone, 10 μg of rLAWD per ml, irrelevant His-tagged protein (negative control), or amastigote lysate (positive control) in the presence of 5 × 10⁵ irradiated syngeneic splenocytes of naive BALB/c mice for 4 days in 96-well plates. The levels of IFN-γ and IL-10 were measured by specific ELISA, as previously described (6).

DNA vaccine construct.For DNA vaccination, the LAWD gene was subcloned from pBK-CMV-LAWD into pcDNA3.1/Hygro(−) (Invitrogen) at the BamHI and KpnI restriction sites. The ligation reaction product was transformed into competent E. coli DH10B (Invitrogen), and positive clones were confirmed through sequencing. Large-scale DNA purification was done by two cycles of CsCl gradient purification. Samples were tested for lipopolysaccharide contamination by the limulus amebocyte lysate test (BioWhitaker, Walkersville, Md.); all samples had <3 endotoxin units per 100 μg of DNA. DNA purity was also determined by measuring the spectrophotometric A₂₆₀/A₂₈₀ ratio; all samples had a ratio of 1.9 or higher.

Immunization and challenge in BALB/c mice.Mice (five per group) were immunized in five locations with a total of 100 μg of DNA (50 μg of LAWD and 50 μg of IL-12) per mouse: four injections in both sides of the inner and outer thigh muscles of the hind legs (∼50 μl/site) and one subcutaneous injection in the left hind foot (∼5 μl/site). Mice were boosted twice at 3-week intervals and then challenged 3 weeks after the last immunization with 2 × 10⁵ metacyclic promastigotes in the right hind foot. Coinjection with an IL-12 gene adjuvant was used as previously described (6). Mice injected with the empty vector only were included as a control. Lesion development was monitored with a digital caliper (Control Company, Friendswood, Tex.). Tissue parasite burdens were measured via a limiting-dilution assay as previously described (32).

Statistical analysis.Data are presented as the mean ± standard deviation, and data were considered significant at a P value of ≤0.05. Significance for parasite burdens, lesion size, and direct ELISA experiments was determined by Student's two-tailed t test, while cytokine ELISA data were analyzed by analysis of variance.

The full L. amazonensis LAWD mRNA sequence has been submitted to GenBank (accession no. AY457171 ).

Identification, cloning, and sequence analysis of the LAWD gene.Sera from L. amazonensis-infected BALB/c mice (8 weeks postinfection) were used to screen an amastigote cDNA expression library. Positive clones were isolated, and pBK-CMV phagemids were generated by helper phage excision. These phagemids were sequenced, and one was selected for characterization because it has not previously been reported. The sequence obtained from the phagemid was used to design an internal reverse primer that was used in conjunction with the spliced leader primer for generating the 5′ region of the mRNA. Thus, the entire mRNA sequence was deduced by aligning the overlapping region of the original clone sequence with the PCR-generated 5′ region. The methionine start codon was identified at nucleotide position 297 after the spliced leader and was located near a putative Kozak sequence (28), with the −3 position relative to the start codon being an A as predicted for 95% of the eukaryotic mRNA sequences (17). The ORF was determined to be 2,025 nucleotides, which translates into a predicted protein of 75 kDa (674 amino acids) and a computer-predicted pI of 6.55. Upstream of the poly(A) site were five large polypyrimidine tracts, which are known to play a role in stability of mRNA and the regulation of gene expression in trypanosomes (2).

Although sequences of the LAWD gene show no significant nucleotide or protein homology to other known proteins in GenBank, TIGR, or EMBL-EBI databases, significant homology to genes sequenced in the Leishmania genome projects was found (http://www.sanger.ac.uk ). For Leishmania infantum, only partial shotgun sequence homologies were found (LI0153c01.q1k and LI0291f05.q1k), but for L. major, a gene designated Het-e2c*4 is the homologue for the gene we have designated the LAWD gene. This Het-e2c*4 misnomer is based on the predicted translated protein's homology with Podospora anserina beta-transducin-like protein HET-E2C, which is homologous only in the WD portion of the protein. This homology suggested that this predicted protein was a WD protein and implied no correlation of function with the HET-E2C protein. For this reason, the Het-e2c*4 gene of the L. major Friedlin genome is referred to as the L. major LAWD gene in this publication.

The L. major LAWD gene is located on chromosome 6 of the Friedlin genome. Analysis of the raw sequences surrounding this gene (16.5 kb) confirmed our Southern blot results (data not shown) that only one LAWD gene copy was present in the genome. The gene sequences surrounding the LAWD gene also did not show significant homologies to known proteins. An 85% homology between L. amazonensis and L. major for alignment of both nucleotides (not shown) and amino acids (Fig. 1) was determined.

• Open in new tab
• Download powerpoint

FIG. 1.

Analysis of LAWD predicted translated sequences. (A) Amino acid sequence alignment of the L. amazonensis (La) and L. major (Lm) LAWD proteins. Presented are the deduced amino acid sequences from both species. Identical amino acids are marked by dashes. Amino acid deletions are marked by asterisks. The seven WD domains are approximated by the grey boxes. Prospective Ser/Thr phosphorylation sites are marked by diamonds, and the potential glycosylation site is underlined. (B) WD motif alignment. The alignment of the seven WD repeat domains from L. amazonensis LAWD is juxtaposed against the WD consensus sequence (31). The residues for the conserved A, B, and C β-strands of each of the seven repeats (grey boxes in panel A) are shown. Amino acids commonly found at specific positions within the repeat are shown at the bottom. Conserved residues are in boldface.

The LAWD proteins of both L. major and L. amazonensis have seven WD domains extending from the middle of the protein to the C terminus (Fig. 1). There is one potential N-glycosylation site for L. amazonensis (Fig. 1A), but this site is not present in L. major. Most notable are the numerous potential phosphorylation sites (Fig. 1A), which remain conserved in both species. Only those Ser and Thr residues that had a NetPhos 2.0 score of <95% for being phosphorylated by protein kinase C (36), protein kinase A (30), and casein kinase 1 (29) are indicated.

RNA expression of the LAWD gene.Transcription of the LAWD gene was confirmed by Northern blot analysis of total RNAs extracted from both Old World and New World species of Leishmania, as well as both promastigotes and amastigotes of L. amazonensis (Fig. 2). The probe, which consisted of the 5′ portion of the ORF that did not encode the WD portion, recognized one band that was slightly larger than 3 kb, which is consistent with the size of the full-length mRNA transcript. Gene homologues to the LAWD gene were detected in L. mexicana, L. chagasi, and L. major (Fig. 2B), as well as in Leishmania pifanoi and Leishmania braziliensis (data not shown). The levels of expression of the LAWD gene were comparable in L. amazonensis promastigotes and amastigotes (Fig. 2A), as well as in log-phase and stationary-phase promastigotes (data not shown).

• Open in new tab
• Download powerpoint

FIG. 2.

RNA and protein expression of LAWD. Both L. amazonensis amastigote (lane A) and promastigote (lane P) RNAs (A), as well as those from other Leishmania species (B), were probed with both LAWD gene-specific DNA and a loading control gene, p23. For detection of LAWD protein in parasites, 2 × 10⁸L. amazonensis promastigotes per well were separated by SDS-PAGE prior to immunoblot analysis. Polyclonal antiserum raised against rLAWD detected two protein bands. One band was the approximate predicted size, 75 kDa, and the other band was slightly larger, possibly due to posttranslational modifications such as glycosylation or phosphorylation.

Identification and localization of native Leishmania LAWD protein.Purified hamster anti-LAWD IgG, generated by using rLAWD, recognized two protein bands in Western blots of L. amazonensis lysates. A protein band at 75 kDa was detected, which corresponded to the predicted size of the LAWD protein (Fig. 2C). Purified IgG from naive hamster serum showed no binding to either of these two bands (data not shown). The antiserum also detected a larger band, suggesting a possible posttranslational modification (glycosylation or phosphorylation) of the LAWD protein. Anomalous migration of the LAWD protein due to a highly basic or acidic protein composition seemed unlikely because the protein had a neutral pI and showed no other potential modification sites within the LAWD ORF. This doublet was recognized in both L. amazonensis promastigote and amastigote lysates, as well as in soluble and membrane fractions (data not shown), suggesting that the protein may be found free in the cytoplasm as well as in membrane-bound vacuoles or organelles. The LAWD protein itself appears to be soluble, as suggested by hydrophobicity and hydrophilicity plots (data not shown), and is less likely to be a transmembrane protein. In some Western blotting experiments, a 30-kDa band was recognized by the antiserum; the presence of this band corresponded with an absence or reduction of the aforementioned larger bands (data not shown), suggesting that this protein may be rapidly degraded or highly vulnerable to Leishmania proteases.

To localize LAWD in the parasite, we performed immunoelectron microscopy. The LAWD protein localized mainly to the flagellar pocket region in both amastigotes (Fig. 3) and promastigotes (data not shown). The label was found both around the outskirts of the flagellar pocket (Fig. 3A and B) and on the internal portion of the flagella (Fig. 3C and D), while it was not found on the external flagella in promastigotes. The flagellar pocket is the major source of secretion for Leishmania, but it is also an important sorting compartment (reviewed in reference 21). Proteins may be transported by an indirect route from the Golgi apparatus to the lysosome-multivesicular tubule via the flagellar pocket (24). The localization of LAWD to the flagellar pocket implies that it may either be secreted, play a role in the sorting process, or be sorted itself within the flagellar pocket. Although some scattered gold particles were found in the cytoplasmic and nuclear regions, this staining was not significantly different from that of the naive hamster IgG control. Some parasites were not labeled, suggesting that the protein may not be constantly produced or may be rapidly degraded.

• Open in new tab
• Download powerpoint

FIG. 3.

Immunoelectron microscopic detection of LAWD in L. amazonensis amastigotes. Antiserum generated in hamster against rLAWD was used for immunoelectron microscopy. In postembedding staining with polyclonal anti-LAWD, the label is mostly detected in the flagellar pocket (FP). Morphological markers are the nucleus (N) and the kinetoplast (K).

Recognition of rLAWD by human patient sera.Because LAWD was identified by using infected mouse sera, we wanted to examine whether Leishmania patient sera would also recognize this protein. Serum samples (n = 44) from patients diagnosed with cutaneous or mucocutaneous leishmaniasis were first screened for reactivity to soluble parasite antigens by direct ELISA. Of these samples, 40 reacted significantly with leishmanial antigens compared to normal control samples and were subsequently tested for reactivity with rLAWD by direct ELISA (Fig. 4). A significant difference was found between the patient sera and the control sera (P < 0.01). None of the sera samples (n = 7) from T. cruzi patients showed reaction with rLAWD (data not shown), suggesting parasite-specific recognition of LAWD in leishmaniasis patients.

• Open in new tab
• Download powerpoint

FIG. 4.

Patient sera are capable of recognizing rLAWD in direct ELISA. Serum samples (n = 40) from patients diagnosed with mucocutaneous and cutaneous leishmaniasis were tested for reactivity with recombinant LAWD protein by direct ELISA. The lines represent the averages for the groups. (When patient sera were compared to the 10 samples from humans not exposed to Leishmania, the P value was <0.01.).

CD4⁺ T cells of infected BALB/c mice recognize rLAWD.Recognition of rLAWD by sera derived from infected mice and human patients implies that LAWD is a competent B-cell antigen. To test LAWD's potential as a T-cell antigen, we isolated CD4⁺ T cells from the draining LN of L. amazonensis-infected BALB/c mice (4 to 5 weeks postinfection) and stimulated the cells with 10 μg of rLAWD per ml, irrelevant His tag protein (a negative control), and parasite lysate (a positive control). Supernatants were collected at 72 h and were assayed for IFN-γ and IL-10 by sandwich ELISA. No IL-10 was detected for the rLAWD or irrelevant His tag protein samples (data not shown). The rLAWD protein induced significant IFN-γ production in CD4⁺ T cells (Fig. 5) compared to the negative control (P < 0.01). Note that the parasite lysate was used merely as a positive control to demonstrate detection of cytokines, and thus its IFN-γ levels should not be directly compared with those of rLAWD. Th1 cytokines, like IFN-γ, have been implicated in protection against L. amazonensis, and the ability of LAWD to stimulate such high production of this cytokine implies that it has potential to be a protective antigen.

• Open in new tab
• Download powerpoint

FIG. 5.

Recombinant LAWD is able to stimulate CD4⁺ cells to produce IFN-γ. CD4⁺ cells were isolated from the draining LN of mice infected with L. amazonensis for 4 to 5 weeks and then stimulated with 10 μg of rLAWD, an irrelevant His-tagged recombinant protein (a negative control), or amastigote parasite lysate (a positive control). Shown are averages for two independently repeated experiments. (When IFN-γ levels were compared between LAWD and the negative control, the P value is <0.01.).

DNA vaccination.Given the high levels of IFN-γ but no IL-10 production in cells reactive to rLAWD, we then tested the efficacy of DNA vaccination with LAWD DNA. BALB/c mice (five per group) were vaccinated three times with the empty vector (a negative control) or with plasmids encoding LAWD and murine IL-12 at 3-week intervals and then challenged with 2 × 10⁵L. amazonensis metacyclic promastigotes three weeks after the final immunization. As shown in Fig. 6, mice coinjected with LAWD and an IL-12 gene showed a marked delay in lesion development for up to 10 weeks postinfection (P < 0.01). Limiting-dilution assays were performed on foot tissues to confirm that this reduced lesion size coincided with a reduction in parasite burden. This delayed lesion development correlated with an approximately 2-log-unit reduction in parasite quantities (Fig. 6). Similar results were observed in an independent experiment, in which the lesion development was monitored for 8 weeks, and there was a 5-log-unit reduction in parasite quantities in mice given LAWD plus IL-12 (data not shown).

• Open in new tab
• Download powerpoint

FIG. 6.

DNA vaccination with LAWD and IL-12 genes delays lesion development for L. amazonensis challenge. BALB/c mice (five per group) were infected with 2 × 10⁵L. amazonensis metacyclic promastigotes. The lesion development was monitored for 8 weeks, at which time parasite burdens in foot tissues were assessed; the numbers are means ± standard deviations for the groups. There is a statistically significant difference in the lesion size for the two infection groups (*, P < 0.01).