Role of Plasmodium berghei cGMP-dependent Protein Kinase in Late Liver Stage Development^*

Parasite Transfection

The cKO targeting vector was constructed by cloning three PCR products into a vector that carries the two Flp recognition target sequences (FRT) sites and the human dihydrofolate reductase expression cassette (16). Primer sequences 5′-GCGGCCGCCACCAAGAGAGTAAACAC and 5′-GCGGCCGCCGAGCCCCAAAACATTAT were used to amplify a 2.0-kb fragment encompassing the 5′-untranslated reginon region and exons 1–3 of PbPKG. The product was cloned into the vector using NotI (underlined). Primer sequences 5′-CTCGAGAAAAAGGCACACATGTTTGC and 5′-CTCGAGAAGTAAAGAATAGCGAAAT were used to amplify a 3.0-kb fragment encompassing exons 4–5 and the 3′-untranslated reginon region of PbPKG. It was inserted into the previously generated plasmid using XhoI (underlined). Primer sequences 5′-GGTACCACAAATATTGTTTTTTCTTTTA and 5′-GAATTCGATATGTATGTCGGGGATTA were used to amplify a 0.5-kb fragment that was inserted into the previously derived plasmid using KpnI (underlined) and EcoRI (underlined). The final insert was released from the targeting construct using EcoRI. Transfections of the targeting plasmid into TRAP/FlpL(−) parasites were carried out using standard methodology (17). Transfected parasites were selected using pyrimethamine and cloned by limiting dilution.

Mosquito Cycles

Anopheles stephensi mosquitoes were fed on infected Swiss Webster mice. Mosquitoes were maintained at 20 °C until day 14 post-bloodmeal and then transferred to 25 °C. Sporozoites were dissected from their salivary glands at days 18–21 post-feeding. Infectivity of salivary glands was similar for cKO and control parasites at ∼8,000–10,000 sporozoites/mosquito.

Southern Hybridization and Diagnostic PCR

Parasite genomic DNA was digested with BamHI and SpeI before transfer to a nylon membrane. The membrane was probed with dioxygenin-labeled exon 5 (DIG High Prime DNA labeling and detection kit, Roche Applied Sciences). DNA hybridization was visualized using a chemiluminescent substrate following the manufacturer's instructions. PCR was performed using genomic DNA extracted from either erythrocytic stage parasites or sporozoites (QIAamp Blood and Tissue DNA kit, Qiagen). Sequences of PCR oligos 1 and 2 are as follows: 5′-GGAGGTATTTCAAAATTGTA and 5′-CCATATGATACTTCGTTTG. PCR was performed using the Expand High Fidelity PCR system (Roche Applied Sciences) with 1 cycle at 95 °C for 5 min followed by 35 cycles of 95 °C for 30 s, 48 °C for 30 s, and 68 °C for 10 min.

Reverse Transcriptase (RT)-PCR

RNA was extracted from HepG2 cells infected with P. berghei sporozoites using the Qiagen RNeasy kit, Qiagen. RNA was reverse-transcribed in the presence or absence of RT using the RETROscript kit, Ambion. The RT reaction was used as template in PCR reactions to amplify PbPKG and glyceraldehyde-3-phosphate dehydrogenase messages. PCR primers used to amplify PbPKG were 5′-CCATTTAGAATTGAGGGATAA and 5′-CACTATTTATAATGAAAAAGT. PCR primers used to amplify glyceraldehyde-3-phosphate dehydrogenase were 5′-ATGGCAATAACAAAAGTCGG and 5′-CCCCATGGAATTTGAGCT.

Infection of HepG2 Cells and Merosome Assay

Sporozoites were obtained at days 18–21, post-bloodmeal, from infected mosquitoes maintained as described above. They were added to HepG2 cells grown on collagen-coated coverslips. Cells were incubated for 40 or 65 h before fixation with 4% paraformaldehyde for 20 min followed by permeabilization with cold methanol for 10 min. Cells were stained with an anti-LS monoclonal antibody that recognizes parasite heat shock protein 70 (Hsp70) (18). The anti-Hsp70 antibody was diluted to 10 μg/ml in phosphate-buffered saline containing 1% bovine serum albumin and incubated with cells for 1 h at 37 °C. After washes with phosphate-buffered saline, the cells with treated with an anti-mouse fluorescein isothiocyanate-conjugated secondary antibody. Merosomes were obtained from infected cultures by collecting the medium 65 h post-infection. Merosomes were pelleted by a brief centrifugation, resuspended in a small volume of phosphate-buffered saline, and added to a coverslip. After paraformaldehyde fixation and methanol permeabilization, merosomes were stained with 4′,6-diamidino-2-phenylindole and examined using a fluorescent microscope.

Mouse Infections and Immunization Studies

Sporozoites were obtained at days 18–21, post-bloodmeal, from infected mosquitoes maintained as described above. They were injected intravenously into C57Bl/6 mice (6–8 weeks, females). To determine the pre-patent period of infection, injected mice were monitored daily through microscopic examination of Giemsa-stained thick blood smears. The first day post-infection, on which blood-stage parasites were detected, was determined to be the pre-patent period for that animal. For immunization studies, immunized and age-matched control animals were challenged with P. berghei NK65 sporozoites. The pre-patent period of infection in each animal was determined as described above.

Detection of Infectivity in Vivo Using Real Time PCR

The in vivo infectivity of CON and cKO sporozoites in the liver was determined as follows. Mice were infected by intravenous injection of 10,000 sporozoites of each parasite into groups of five C57Bl/6 animals. The parasite load within each mouse liver was determined 40 h post-infection by real time PCR (19). The standard curve was obtained using serial dilutions of a plasmid carrying the P. berghei 18 S rRNA gene. The parasite load of each sample was expressed as the plasmid DNA equivalent The primers used for amplification of P. berghei 18S rRNA were 5′-AAGCATTAAATAAAGCGAATACATCCTTC and 5′- GGAGATTGGTTTTGACGTTTATGT.

PbPKG Is Expressed during Liver Stage Development

PKG (PfPKG) is expressed during erythrocytic stages, primarily in rings, and in gametocytes (20, 21). Here, we investigated PbPKG expression in liver stages. PbPKG transcripts were clearly detected in early and late LSs developing in tissue-culture cells (Fig. 1). These results are consistent with microarray studies that found Plasmodium yoelii PKG expression during LS development in vivo (22). The temporal expression pattern of PbPKG, thus, suggested a possible role for PbPKG in liver stages. We examined its functional role in LS biology by producing loss-of-function mutants. However, as was reported earlier for PfPKG (14), PbPKG proved to be refractory to gene disruption, suggesting that it is essential for the erythrocytic cycle of the parasite. Therefore, we utilized a conditional mutagenesis approach for disrupting the PbPKG gene in the pre-erythrocytic stages of the parasite.

Conditional Mutagenesis of PbPKG Using the TRAP/FlpL System

We utilized a recently developed P. berghei conditional mutagenesis approach that uses stage-specific expression of a temperature-sensitive variant of the yeast recombinase, flippase (FlpL), to catalyze the excision of DNA placed between two FRT (16). FlpL catalyzes site-specific DNA recombination at a temperature range of 25–30 °C.

The PbPKG open reading frame consists of five exons, with exons 4 and 5 encoding four cGMP binding sites and a kinase domain (Fig. 2A). We modified the locus by inserting two FRT sites carried on a double crossover targeting plasmid called pKO. Homologous integration of the pKO plasmid would place one FRT site in the intron separating exon 3 and exon 4 and a second FRT site downstream of the 3′ expression sequence of the gene. The plasmid was transfected into parasites that express FlpL under the control of the sporozoite-specific promoter of the Thrombospondin-related anonymous protein gene (TRAP/FlpL), causing the excision of the FRT-flanked sequence specifically in sporozoites (Fig. 2A). Recombinant parasites were readily selected (Fig. 2C), showing that the 5′-FRT site located between exons 3 and 4 does not impair normal expression of PKG. A recombinant clone named “cKO” was isolated by limiting dilution for further analysis. A control clone named “CON” was selected after transfection of a derivative plasmid called pCON, which lacks the 5′-FRT site, into the same TRAP/FlpL parasite clone (Fig. 2B).

The cKO and CON clones were passaged into mosquitoes maintained at 20 °C until day 14 post-bloodmeal. On day 14, infected mosquitoes were transferred to 25 °C to increase the activity of FlpL and, hence, the efficiency of DNA excision at the PbPKG locus. Sporozoites that developed from the cKO and CON parasite populations were collected from mosquito salivary glands on days 18–21, post-bloodmeal. As shown in Fig. 2C, excision at the PbPKG locus was undetectable in the erythrocytic stages but was readily detected in the sporozoite population of the cKO clone. In contrast, the PbPKG locus was intact in both the erythrocytic stages and the sporozoite population of the CON clone, which lacks the 5′-FRT site. Henceforth, cKO parasites with a disrupted PbPKG locus are referred to as “PbPKG−,” and those with the intact locus are referred to as “PbPKG+.”

PbPKG Sporozoites Do Not Lead to a Blood Stage Infection but Develop into Late Liver Stages

To study the function of PbPKG in the parasite mammalian host, cKO and CON sporozoites were injected intravenously into mice. Excision of PbPKG in the cKO sporozoite population would generate parasites lacking PbPKG function (PbPKG−), whereas lack of excision would generate parasites in which PbPKG function was normal (PbPKG+). Although all animals infected with CON sporozoites became patent, i.e. displayed detectable blood-stage infection (p > 0.01%), by day 7, none of the animals infected with cKO population were patent at day 7 (Fig. 3, A and B). Four of the five mice infected with the cKO population stayed negative for the course of the experiment (animals were sacrificed at day 15).

One of the five animals infected with the cKO population became patent on day 8. PCR analysis of the genomic DNA of parasites collected from this animal detected only a ∼5.2-kb fragment corresponding to the non-excised version of the PbPKG cKO locus (Fig. 4). Therefore, this animal had been infected by the small fraction of the cKO sporozoite population that escaped DNA excision and carried an intact PbPKG locus (PbPKG+). The delayed patency displayed by the infected animal (day 8 versus day 5.8 on average for CON-infected animals) suggests that the percentage of sporozoites that escape excision, is small because a 10-fold decrease in sporozoite number generally leads to a delay of 1 day in the pre-patent period. In an independent experiment a higher infectious dose of the cKO sporozoite population also did not result in blood-stage infection by PbPKG− parasites (Fig. 3B). As before, the two animals that developed blood stage infections were infected only with PbPKG+ parasites (data not shown). These data confirm that PbPKG− sporozoites do not develop into erythrocytic stage parasites.

Next we examined if the lack of blood-stage infection by PbPKG− parasites was a result of impaired liver stage development. Using real-time PCR, we quantified liver infection in vivo 40 h post-infection of mice by cKO and CON sporozoites (Fig. 5A). We did not find a significant difference in parasite burden in the two groups of infected animals. These data suggested that cKO sporozoites are competent to initiate hepatocyte infection and undergo intrahepatic development. To examine liver stage development in greater detail, we used the cultured hepatoma cell line, HepG2. Development of cKO and CON parasites in HepG2 cells was followed by immunostaining of the cells with a parasite-specific antibody against Hsp70 (18). We observed that PbPKG cKO and CON sporozoites formed similar numbers of LSs at 40 h post-infection, consistent with the in vivo results (Fig. 5B). There were no significant morphological differences between cKO and CON LSs. However, at 72 h post-infection, the number of LSs in CON-infected cells declined significantly (Fig. 5B), and merosomes were released into the medium (65–72 h post-infection) (Fig. 5C). In contrast, in cKO-infected cells, the number of LSs remained high, and few merosomes were observed in the medium (Fig. 5C). These data suggest that cKO parasites are blocked at a late step in LS development.

Immunization with PbPKG cKO Sporozoites Protects from Malaria

Although early LS parasites are known to elicit a protective host response (4, 5, 23), the ability of late liver stages to elicit such immunity is unknown. We tested if immunization with cKO sporozoites would confer protection to immunized animals from a sporozoite challenge. Groups of C57Bl/6 mice, a mouse strain susceptible to P. berghei infection, were immunized and boosted with cKO sporozoites before infection with wild type sporozoites. We observed that whereas control non-immunized animals became patent, the immunized animals remained negative for the duration of the experiment (Table 1). These results suggest that the developmentally arrested LSs of cKO parasites are capable of eliciting a protective immune response from the host. That cKO LSs arrest later than previously identified LS mutants that induce immunity (3, 5, 7) implies that antigens present in late LSs, similar to those present in early LSs, might elicit a protective host immune response.