2.2 Sample preparation and MudPIT analysis

Purified zygote and ookinetes were used for MudPIT analysis using trypsin and proteinase K separately. For trypsin analysis, 1.15 × 10⁷ cells (zygotes) and 1.6 × 10⁶ cells (ookinetes) were used. For proteinase K analysis, 1.01 × 10⁷ cells (zygotes) and 1.5 × 10⁶ cells (ookinetes) were used. For whole cell fractionation and trypsin-based digestion, cell pellets were resuspended in 100 μL of 100 mM phosphate buffer pH 6.7, 5 mM MgCl₂, 250 mM sucrose, 5 mM CaCl₂. One microgram of trypsin (sequencing grade, Promega) was added and then incubated at 37°C for 30 min, followed by two-fold dilution with 100 μL of 100 mM phosphate buffer pH 6.7, then addition of 200 μL of 100 mM phosphate buffer pH 6.7, 5 mM MgCl₂, 250 mM sucrose, 2 mM CaCl₂. The mixture was centrifuged at 18 000 × g 30 min at 4°C, and the supernatant (extracellular, E) was obtained. The pellet was then resuspended in 100 μL of 10 mM Tris-HCl pH 8.5 and incubated for 1 h on ice to lyse cells. The sample was then centrifuged at 18 000 rcf for 30 min at 4°C, and the supernatant was transferred (soluble lysate, S). The pellet was resuspended in 100 μL of 0.1 M Na₂CO₃ pH 11.5, incubated on ice for 1 h, centrifuged at 18 000 rcf for 30 min at 4°C, and the supernatant was transferred (salt-eluted membrane fraction, SP). The pellet (membrane fraction, P) was dried via speed-vac.

S and SP fractions were subjected to protease digestion. Solid urea was added to 8 M, and TCEP was added to a final concentration of 5 mM, and incubated at 27°C for 30 min. The mixture was brought to 10 mM iodoacetamide (20 mM for P samples) and incubated at 27°C for 30 min in the dark. For SP, the pH was adjusted to 8.5 with 1 M Tris-HCl pH 8.5 and 1 μg of endoproteinase Lys-C (sequencing grade, Roche) was added to ookinete and zygote samples, respectively, and incubated overnight at 37°C, with shaking. Samples were diluted to 2 M urea with 100 mM Tris-HCl pH 8.5, CaCl₂ was added to 2 mM, 0.5 μg of trypsin was added, and incubated 37°C overnight, with shaking. Digestions were stopped by adding formic acid to 5%, and then stored at -80°C until analysis.

P fractions were subjected to cyanogen bromide-treatment followed by protease digestion. The P fraction was resuspended in 100 μL of 90% formic acid (this and all subsequent steps performed in a hood), 500 mg/mL CNBr, and incubated overnight in the dark. NH₄OH (30%) was then added dropwise until the pH was ~8.5 (300 μL). Tris-HCl pH 8.5 was added to a final concentration of 0.1 M. Solid urea was added to 8 M, TCEP was added to a final concentration of 5 mM, and incubated at room temperature (RT) for 30 min. The mixture was brought to 20 mM iodoacetamide and incubated at RT for 30 min in the dark. Lys-C (1 μg) was added and incubated overnight at 37°C. Samples were diluted to 2 M urea with 100 mM Tris-HCl pH 8.5. CaCl₂ was added to 2 mM, 0.5 μg of trypsin was added and incubated at 37°C overnight with shaking. Digestions were stopped by adding formic acid to 5%, and concentrated on a 4 mg SPEC-PT-C18 (Varian), then washed and eluted as directed by the manufacturer. Eluates were dried down via speed-vac, resuspended in 20 μL of MudPIT-buffer A (see below), and stored at -80°C until analysis.

Each digest was loaded onto a three-phase MudPIT column [25] as previously described [26]. Briefly, digests were loaded onto a multidimensional 250 μm id fused-silica column encountering 3.5 cm of C18 resin (5 μm, 110A, Aqua, Phenomenex) followed by 3.5 cm of SCX resin (5 μm, 110A, Partishere, Whatman) via a high pressure loading device. After loading, a 100 μm fused-silica analytical column with a 5 μm laser-pulled tip containing 10.5 cm on C18 resin was attached adjacent to the SCX-side of the multidimensional column via a union (Upchurch). This three-phase MudPIT column was then attached to an HP1100 HPLC (Agilent) and coupled to an LCQ Deca IT mass spectrometer (Thermo Electron). Online LC/LC-MS/MS was then carried out using salt steps, RP gradients, instrument settings, and data-dependent collection of MS/MS spectra as previously described [25, 26]. For all digests of ookinete supernatants, 6-step MudPIT runs were performed; for trypsin and proteinase K digests of subcellular fractions, 12-step MudPIT runs were performed. Collected MS/MS spectra were searched against the database described below using the algorithm SEQUEST [26].

A P. gallinaceum ORF database was created using current available genome databases (12/31/06 for contigs and 1/3/2007 for reads downloaded from ftp://ftp.sanger.ac.uk/pub/pathogens/Plasmodium/gallinaceum/). The nucleotide sequences of both reads and contigs were translated in all six reading frames, and a FASTA format database was then created for those entries containing at least 50 amino acids stretches using an in-house Perl script (J. Johnson, unpublished). This database was then combined with all chicken (Gallus gallus) protein entries (downloaded from ftp://ftp.ensembl.org/pub/current_gallus_gallus/data/fasta/pep/), to which was manually added a list of common contaminants (e.g., human keratin and trypsin), and the protein sequences for P. gallinaceum chitinase (PgCHT1 and PgCHT2) and all of the above sequences were reversed. This database was then used for searching MS/MS spectra. The program DTA select [27] was used to filter peptide identifications, requiring that all identifications pass standard XCorr (1.8, 2.5, 3.5 for 11, 12, and 13 charged peptides, respectively) and deltaCN (0.08) score thresholds [28]. By concatenating a reversed database, we estimated the false positive rate of the searches based on how many hits were obtained to reversed sequence. Additionally, identifications against chicken and contaminant entries were removed from the final list.

3.1 Total proteome of zygote, ookinete, and secretome

From the P. gallinaceum partial ORFs identified in the analysis of zygote, ookinete, and ookinete-secreted protein (ookinete culture supernatant), a total of 966 orthologous proteins of P. falciparum were identified. These predicted proteins account for 18.3% of the total predicted proteins in the P. falciparum genome. Among these identified 966 P. falciparum orthologs, 460 (47.6%) proteins had more than two spec count (spec count ranges from 2-397); another 506 proteins had only one spectrum counts. The details of identified proteins with sequence and spectrum counts presented as the Supporting Information. (Table S1).

The P. falciparum orthologs were classified according to their molecular function (class code 1-13) (Table 1). Forty percent of the orthologous proteins were hypothetical proteins matching to P. falciparum (36% hypothetical and 4% conserved hypothetical). Other important classes of proteins, belonging to cell surface (3%), signal transduction (3%), and protein fate (10%) accounts for 3-10% the total orthologs identified in P. falciparum. These classes of proteins are necessary for invasion, defense, and cell communication, which are critical for the successful transmission of the parasite.

Proteins with predicted secretory SP and TM domains are of particular importance due to their secretory fate or presence on the cell surface [23]. Few proteins (Chitinase, Pfs230, CTRP, WARP, and Pfs25) have been studied in detail in zygote and ookinete stages, some of which are characterized as malaria transmission-blocking vaccine candidates [4]. We have analyzed the presence of SP/TM domain(s) containing proteins in different functional classes of P. falciparum orthologs (Fig. 1). Surprisingly, a majority of SP and TM containing proteins are hypothetical, which suggests that a substantial number of hitherto uncharacterized proteins may have particularly interesting roles in parasite-vector interactions.

Open in a separate window

Figure 1

The bar diagram shows the percentage of P. falciparum orthologous proteins with secretory/cell surface associated domains among the different functional class of proteins. The P. falciparum orthologs were identified in the zygote, ookinete, and ookinete secretomes of P. gallinaceum.

In P. falciparum about 5268 predicted proteins are present in the genome, 65% of which are hypothetical or uncharacterized [30]. Earlier using similar MudPITapproaches, 2415 proteins from sprorozoites, merozoites, trophozoites, and gametocytes were described. This number was later increased to 2904 by adding proteins of rings, schizont, and gamete stages [31]. Le Roch et al. [13] reported that at least 4557 genes (88%) were transcribed in one or more stages of the life cycle of P. falciparum, but this analysis did not include analysis of proteins from gametocytes, zygotes, or ookinetes. In P. berghei, 1836 predicted proteins were detected with high confidence from asexual and ookinete stages of the parasite [18]. However, because the P. falciparum zygote and ookinete stages have not obtainable in sufficient number in vitro for proteomic analysis these stages have remained a “black box” in the life cycle of the human parasite.

3.2 Comparative proteome of zygote, ookinete, and secretome

In the present study, using proteins of P. gallinaceum zygote, ookinete, and culture supernatant, we identified 381-749 predicted orthologs shared by P. gallinaceum and P. falciparum in the zygote, ookinete, and ookinete-secreted protein proteomes (Fig. 2). These observations further demonstrated the close evolutionary affinity of P. falciparum and P. gallinaceum at the mosquito stages of both malaria parasites. The close evolutionary affinity of P. falciparum and P. gallinaceum allows for a more complete estimate of P. falciparum zygote and ookinete proteome. The identified zygote, ookinete, secreted proteome corresponded to 7-14.2% of the annotated protein databases from P. falciparum genome. Zygotes, ookinete, and ookinete-secreted proteins share only 16.9% of the identified P. falciparum orthologs. As reported earlier with P. falciparum, asexual blood stages, gametocytes, and sporozoites (but not zygotes and ookinetes) share 6% total predicted proteins [16], which indicates that each stage is highly specialized in its proteome. In zygotes, 418 P. falciparum orthologs were identified, of which 68 (16%) were unique to this stage. In ookinetes, 749 P. falciparum orthologs were identified, of which 343 (46%) proteins were unique to this stage. Zygotes and ookinete shared 338 (45-80%) proteins. In earlier published reports, 1147 proteins were detected in P. falciparum gametocytes, where 33% proteins were unique to the stage [16]. In P. falciparum sporozoites, of 1049 proteins identified, almost half (49%) were unique to the stage [16]. These findings, including our results, suggest that ookinete and sporozoites express a large number of unique proteins inside mosquito hosts. In the P. gallinaceum ookinete secretome, 381 P. falciparum orthologous proteins were detected, of which a remarkable 36% were not found in either zygote or in ookinete lysate.

Open in a separate window

Figure 2

Venn Diagram showing the unique and shared orthologs of P. falciparum found in zygotes (Z), ookinetes (O), and ookinete-secreted proteins (S) of P. gallinaceum by MudPIT analysis.

The number of unique proteins in the ookinete was higher than either zygote or ookinete secretome, hence it may be particularly interesting to investigate their role in host-parasite interactions. These results indicate that, like the asexual blood stages, gametocytes, and sporozoites that the midgut stages also have a largely unique proteome, likely related to the stage-specific functions including immune evasion, coping with the mosquito gut environment, and invasion of the peritrophic matrix.

This project was supported by U.S. Public Health Service grants R21AI053781 and R01AI45999 to J. M. V. from the National Institute of Health. The genome sequencing of P. gallinaceum was carried out by the Pathogen Sequencing Unit, Wellcome Trust Sanger Institute, supported by the Wellcome Trust, which provided permission to analyze this dataset at the whole genome level. This work was additionally supported by NIH grant P41 RR11823-10 to J. L. Y. and an NRSA fellowship 32AI062061 to G. C. We thank Dr. Fengwu Li and Karen Chin for their help and support in the laboratory. We especially thank Dr. Laurence Florens for her enormous support in initiating this project.