A novel Plasmodium yoelii pseudokinase, PypPK1, is involved in erythrocyte invasion and exflagellation center formation

Malaria parasites proliferate by repeated invasion of and multiplication within erythrocytes in the vertebrate host. Sexually committed intraerythrocytic parasites undergo sexual stage differentiation to become gametocytes. After ingestion by the mosquito, male and female gametocytes egress from erythrocytes and fertilize within the mosquito midgut. A complex signaling pathway likely responds to environmental events to trigger gametogenesis and regulate fertilization; however, such knowledge remains limited for malaria parasites. Several pseudokinases are highly transcribed at the gametocyte stage and are possible multi-functional regulators controlling critical steps of the life cycle. Here we characterized one pseudokinase, termed PypPK1, in Plasmodium yoelii that is highly expressed in schizonts and male gametocytes. Immunofluorescence assays for parasites expressing Myc-tagged PypPK1 confirmed that PypPK1 protein is expressed in schizonts and sexual stage parasites. Transgenic ΔpPK1 parasites, in which the PypPK1 gene locus was deleted by the CRISPR/Cas9 method, showed significant growth defect and reduced virulence in mice. In the blood stage, ΔpPK1 parasites were able to egress from erythrocytes similar to wild type parasites; however, erythrocyte invasion efficacy was significantly reduced. During sexual stage development, no clear changes were seen in male and female gametocytemias as well as gametocyte egress from erythrocytes; but, the number of exflagellation centers and oocysts were significantly reduced in ΔpPK1 parasites. Taken together, PypPK1 has an important role for both erythrocyte invasion and exflagellation center formation.


Introduction
Malaria remains a heavy burden on human society worldwide, causing > 200 million cases and > 400,000 deaths annually [1]. Malaria parasites proliferate by repeated invasion of and multiplication within erythrocytes in mammalian hosts. This asexual blood stage is responsible for malaria clinical symptoms and pathogenesis. During asexual proliferation, some parasites undergo sexual stage differentiation and matured gametocytes circulate in the host peripheral blood in anticipation of ingestion by a mosquito [2]. Although sexual stage parasites do not contribute to malaria pathogenesis, they are essential for transmission to other hosts by mosquito vectors. Therefore, sexual development in both mammalian hosts and vector mosquitoes has been studied extensively to identify novel drug and vaccine targets to block parasite transmission [3].
Pseudokinases lack one or several catalytic residues, but still have significant roles in signaling pathways, such as binding to and modulating functions of other kinases and phosphatases, and competing with signal-related molecules by binding to their substrates [14]. Several pseudokinases have been characterized in apicomplexan parasites. In Toxoplasma gondii, a pseudokinase TgBPK1 is involved in cyst wall formation and infectivity to the host [15]. Another pseudokinase TgROP5 forms a complex with TgROP17 and TgROP18 kinases, which interact with host immunity-related GTPases (IRG) and regulate acute symptoms [16,17]. Malaria parasites possess 8 pseudokinases [18], all of which are poorly characterized with the exception of one pseudotyrosine kinase which was reported to be an exported protein to the infected erythrocyte and has a scaffolding role for serine repeat antigen 5 (SERA5) and protein phosphatase type 1 (PP1) [19].
Plasmodium yoelii PY17X_1220300 is a pseudokinase which is highly transcribed at the schizont and male gametocyte stages. Disruption of its ortholog in P. berghei (PBANKA_1217100) showed a slow growth phenotype during asexual replication in mice [20]. An exhaustive mutagenesis analysis suggested that its ortholog in P. falciparum (PF3D7_0321400) also has a slow growth phenotype in in vitro culture conditions [21,22]. To further characterize the role of this pseudokinase we took advantage that it is dispensable during the asexual stage, and evaluated PY17X_1220300 for its expression using epitope-tagged transgenic P. yoelii lines and its biological roles in the asexual and gametocyte stages using gene deletion parasite lines.

Parasites and experimental animals
The P. yoelii 17XL line was maintained in 6-8 week old female ICR or BALB/c mice (Japan SLC, Hamamatsu, Japan). Animal experiments were approved by the Animal Care and Use Committee of Nagasaki University (Permit number: 1403031120-5).

Plasmid construction and transfection
A CRISPR/Cas9 plasmid (pDC2-Cas9-gRNA-hdhfr, a kind gift from the Wellcome Genome Campus Advanced Course) [28], was digested with BamHI, and then a DNA fragment containing the P. yoelii U6 promoter and its terminator PCR-amplified from P. yoelii gDNA was ligated using an In-Fusion HD cloning kit (Takara Bio Inc., Shiga, Japan), yielding pDC2-cam-Cas9-PyU6-hDHFR. To generate plasmids to knockout the PY17X_1220300 gene locus or to tag the C-terminal end with Myc epitopes, pDC2-cam-Cas9-PyU6-hDHFR was digested with BbsI and ligated with the gRNA component targeting PY17X_1220300.
Transfection to P. yoelii was performed as described with minor modification [29]. Schizonts were enriched by density gradient centrifugation using the Histodenz solution (1.077 g/mL) and were mixed with human T cell nucleofector solution (Lonza, Basel, Switzerland) containing 20 μg of the circular plasmids and electroporated using a Nucleofector™ 2b device (Lonza; U-33 program). Drinking water containing 0.07 g/mL pyrimethamine (Fukuzyu Pharmaceutical Co., LTD, Japan) was administrated orally from one day post transfection and parasites emerged following drug selection were passaged to a new mouse, then cloned by limiting dilution. To confirm the gene deletion and the insertion of the epitope tag sequence, gDNA of transfectants were extracted using a QIAamp DNA Blood Mini Kit (Qiagen), and PCR was performed with specific diagnostic primer pairs (Table S1), because off-target effects have not been reported [30].

Gametocyte preparation and ookinete induction
Wild type or ΔPypPK1 parasite-infected blood was collected by cardiac puncture and the gametocyte-enriched fraction (together with schizonts) was obtained by Histodenz density gradient centrifugation described above. To induce gametogenesis, fertilization, and ookinete transformation, enriched gametocytes were incubated for 16 h at 24°C in ookinete culture medium (RPMI1640 medium supplemented with 0.225% sodium bicarbonate, 20% fetal calf serum, 25 mM HEPES, 50 μg/mL hypoxanthine, and 100 μM xanthurenic acid (XA)) [31]. The ookinete fraction was enriched by density gradient centrifugation as described above.

RNA isolation, cDNA synthesis, and reverse-transcription polymerase chain reaction (RT-PCR)
Parasite total RNA was extracted using TRIzol (Thermo Fisher, Waltham, MA) and SV total RNA isolation system (Promega) following the manufacturers' protocols. Complementary DNA (cDNA) synthesis was carried out using SuperScript III reverse transcriptase (Invitrogen) and random hexamers according to the company's instruction. RT-PCR was performed with primers listed in the table S1 using Tks GFlex polymerase (Takara Bio Inc.). Pypkac was used as a positive control.

Asexual parasite growth assay and mouse survival assay
Parasite-infected erythrocytes (10 6 ) were intravenously injected into mice (6-weeks old female BALB/c, n = 5) and parasitemias were monitored daily by microscopic observation of Giemsa-stained thin blood smears. To determine parasitemias, at least 10,000 erythrocytes or 100 parasites were counted and differences were examined by oneway ANOVA followed by a post-hoc Tukey's multiple comparison test using Prism 6 software (GraphPad Software Inc., San Diego, CA). To evaluate parasite virulence, 10 6 parasite-infected erythrocytes were intravenously injected into mice (6-week old female BALB/c, n = 10), then survival was monitored every day up to 14 days post-inoculation. Mouse survival rates were shown as Kaplan-Meier curves and significant differences were evaluated by the Log-rank (Mantel-Cox) test implemented in Prism 6 software. To examine erythrocyte preference, selectivity index (SI) was calculated as described [36].

In vitro and in vivo erythrocyte invasion assay
In vitro invasion assays were performed as described [29]. In brief, 10 7 enriched schizonts described above were resuspended in complete culture medium for P. yoelii (PyCM: RPMI1640 medium supplemented with 0.225% sodium bicarbonate, 1% AlbuMax I, 25 mM HEPES, 50 μg/mL hypoxanthine, and 10 μg/mL gentamicin) and free merozoites were obtained by filtration through a filter unit (1.2 μm pore size; Sartorius Stedim Biotech, Göttingen, Germany) at 15°C. Purified free merozoites (approximately 1 × 10 6 ) were mixed with uninfected mouse erythrocytes pre-warmed to 37°C (approximately 5 × 10 6 ) in 300 μL of PyCM and incubated with 5% O 2 , 5% CO 2 , and 90% N 2 gas mixture at 37°C for 1 h using a shaking incubator at 1000 rpm. Mixtures were then centrifuged, and the pellets resuspended with fresh PyCM to a final hematocrit of 1.66% and transferred to a 96-well plate. Parasites were further incubated for 18 h with the above-mentioned gas mixture. Giemsa-stained blood smears were prepared and at least 10,000 erythrocytes were counted to calculate parasitemia. Significant differences were evaluated by one-way ANOVA followed by Tukey's multiple comparison test.
For in vivo invasion assays, 10 8 enriched matured schizonts were intravenously inoculated into mice (n = 5) and parasitemias were monitored independently for each stage by microscopic examination of Giemsa-stained blood smears at 2 h post-inoculation. At least 10,000 erythrocytes were counted to calculate parasitemia and significant differences were evaluated by Kruskal-Wallis one-way ANOVA followed by Dunn's multiple comparison test using Prism 6 software.

Assays to evaluate gametocyte egress, exflagellation, and oocyst numbers
Gametocyte egress assays were performed as described [31]. Briefly, gametocytes were enriched as described in section 2.5 and gametocyte egress was induced by incubating in ookinete culture medium containing XA. After 15 or 30 min incubation, cultures were fixed with PFA and thin blood smears were prepared to evaluate male or female gametocyte egress, respectively. The blood smears were immunostained with anti-TER-119 (erythrocyte marker), α-tubulin II antiserum (male gametocyte/microgamete marker), and anti-Pys25 antiserum (female gametocyte/macrogamete marker). The male or female gametocyte egress ratios were obtained by dividing the number of parasites with α-tubulin II-or Pys25-positive and TER-119 negative by the number of α-tubulin II-or Pys25-positive parasites, respectively. To quantitate exflagellation, 3 μL of parasite-infected blood was resuspended in 57 μL of ookinete culture medium and incubated at RT for 5 min. The mixture was then applied to a hemocytometer and the number of exflagellation centers per 1 × 10 4 erythrocytes were counted from day 2 to 4 post-infection [31]. The numbers of exflagellated male gametocytes were estimated from male gametocytemia, egress ratio, and exflagellated ratio at day 3. Then, the observed numbers of exflagellation centers were divided by the estimated number of exflagellated male gametocytes. The ratio of exflagellation centers per exflagellated male gametocyte were normalized by the average of the pPK1-myc parasites. Significant differences in gametocyte egress ratios, exflagellation ratios, and the number of exflagellation centers were evaluated by one-way ANOVA followed by Tukey's multiple comparison test. Anopheles stephensi mosquitoes were fed on parasite-infected ICR mice. Fully engorged mosquitoes were maintained at 24°C with a 12 h light/dark cycle. The number of midgut oocysts was counted from at least 35 mosquitoes under the microscope on day 11 post-feeding [37] and significant differences were evaluated by Kruskal-Wallis one-way ANOVA followed by Dunn's multiple comparison test.

Time lapse imaging of exflagellation
Blood from parasite-infected mice were collected from tail veins on day 3 post-inoculation, mixed with ookinete culture medium, applied to C-Chip hemocytometer slides (NanoEnTek Inc., Seoul, South Korea), and observed using a confocal microscope system (A1 Rsi; Nikon) with a 60 × magnification lens. Video images were obtained every 0.1 s, up to 20 min using a CCD camera (ORCA-R2), cropped, and analyzed using Fiji software [38].

PypPK1 is conserved among Plasmodium spp. and lacks a catalytic aspartate and a Mg binding site
PY17X_1220300 encodes a 319 amino acid protein with an expected molecular weight of 38 kDa. The CD-Search webware predicted a protein kinase catalytic domain from amino acid positions (aa) 36 to 278 (Evalue = 6.63e −22 ; Fig. 1A) [39], and a signal peptide sequence or transmembrane region was not predicted [25,26]. An active protein kinase has several critical features for phosphorylation activity such as a catalytic aspartate and a Mg binding site. The corresponding regions of PY17X_ 1220300 were not conserved with the known P. yoelii protein kinases, PyCDPK4, PyCDPK3, PyPKAc, PyGSK3, and PyMAPK2, suggesting that PY17X_1220300 does not possess phosphorylation activity (Fig. 1B). The relevant regions are widely conserved within orthologs in other Plasmodium spp. (Fig. 1C and Fig. S1), suggesting that PY17X_1220300 is a pseudokinase with a conserved role across Plasmodium. Taken together, we designated this protein as P. yoelii pseudo Protein Kinase 1 (PypPK1).

PypPK1 is localized at the apical side of daughter merozoites
Myc epitope tags were introduced at the C-terminus of PypPK1 via homologous recombination and the resulting pPK1-myc transgenic P. yoelii clones were verified by diagnostic PCR ( Fig. 2A and B). Western blot analysis of pPK1-myc clones with anti-Myc antibody revealed a single band around 40 kDa, consistent to the calculated molecular weight of this protein, 40.2 kDa (Fig. 2C). Protein extracts from wild type parasites did not show any bands, thus excluding non-specific reaction of the anti-Myc antibody. Western blot analysis using anti-AMA1 antibody ensured that the loaded protein amounts were similar (Fig. 2C  bottom). IFA of pPK1-myc schizont-stage parasites showed that pPK1myc signals were more closely co-localized with signals for the microneme marker AMA1 and the inner membrane complex marker MTIP than the signal for the dense granule marker EBL (Fig. 2D). These results suggest that pPK1-myc is expressed at the schizont stage of P. yoelii, consistent to the peak transcription pattern at this stage of P. berghei during the asexual cycle, and localized at the apical side of each daughter merozoite.

Deletion of pPK1 reduced P. yoelii activity to invade host erythrocytes and its virulence
To examine the function of pPK1, two clones of P. yoelii ppk1 deletion mutants (ΔpPK1 #1 and #2) were independently generated using CRISPR/Cas9 methodology (Fig. 3A). Deletion of the pPK1 gene locus and the loss of mRNA were confirmed by diagnostic PCR and RT-PCR, respectively ( Fig. 3B and C). ΔpPK1 parasites showed a significant growth defect and a significant number of mice infected with this parasite survived longer than mice infected with the wild type parasite (Fig. 4A and B). Because pPK1 is expressed in daughter merozoites, we examined the effect on erythrocyte egress and invasion by parasites. Firstly, fully matured schizonts were purified and injected into mice and non-egressed schizonts and newly invaded ring stage parasites were monitored 2 h later by microscopic observation of Giemsa-stained blood Fig. 4. Phenotypes of ΔpPK1 parasites during asexual stage development. (A) Parasitemia curves of wild type (WT) parasites and two ΔpPK1 parasite clones #1 and #2 (n = 5). Parasitemias were monitored every day after inoculation of 1 × 10 6 parasites. Asterisks indicate that the ΔpPK1 parasitemia was significantly lower than WT parasitemia at each examined day (p < .01 by Tukey's multiple comparison test). (B) Kaplan-Meier curves with 10 BALB/ c mice for each group. Mouse survival rates with ΔpPK1 parasites were significantly higher than for WT parasites (* p < .001 by Mental-Cox test). (C) in vivo erythrocyte egress and invasion assay. Egress efficacy and parasite invasion ability was evaluated by schizont (left) and ring (right) stage parasitemias at 2 h post inoculation of 1 × 10 8 schizonts into a mouse. Ring stage parasitemias of ΔpPK1 #1 and #2 parasites were significantly lower than that of WT parasites by Dunn's multiple comparison test). n.s.; not significant. (D) in vitro erythrocyte invasion assay. The invasion efficacy of ΔpPK1 parasites was significantly lower than that of WT parasites by Tukey's multiple comparison test. (E) Microorganelle protein secretion assay for WT and ΔpPK1 #1. DIC, differential interference contrast image; MSP1 merozoite surface protein 1 (green); AMA1 or EBL (red). MSP1 and AMA1 or EBL images were merged with DAPI nucleus signals (Merge). Scale bar 2 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) smears. The percentage of non-egressed schizonts was not significantly different between wild type and ΔpPK1 parasites (Fig. 4C left); however, the ratio of newly invaded ring stages of ΔpPK1 parasites was significantly lower than for wild type parasites (Fig. 4C right). In addition, in vitro erythrocyte invasion assays revealed that the parasitemias of ΔpPK1 #1 and #2 parasites were 0.53 ± 0.02% and 0.39 ± 0.08%, respectively, significantly lower than the parasitemia of the parental wild type parasites (1.07 ± 0.10%) (Fig. 4D). These in vivo and in vitro results indicate that pPK1 is not crucial for erythrocyte egress, but partly contributes to erythrocyte invasion. To investigate whether the reduction in erythrocyte invasion efficacy is due to an altered erythrocyte preference by ΔpPK1 parasites, a selectivity index (SI) was calculated at day 3 post-inoculation ( Table 1). The SI value of parasites that can only invade immature erythrocytes is high (e.g. P. yoelii 17XNL and P. vivax), whereas the SI value of parasites that invade both immature and mature erythrocytes is low (e.g. P. yoelii 17XL and P. falciparum) [34,36]. The SI value of ΔpPK1 #1 and #2 parasites were 2.97 and 3.03, respectively, slightly higher than the value of wild type parasites (1.79), but the difference was not significant. Thus, the effect of PypPK1 gene deletion on erythrocyte preference appears to be marginal or absent.
Because phosphorylation signals are involved in organelle discharge, we investigated whether depletion of pPK1 affected the secretion of two invasion related organelle molecules, AMA1 and EBL [29,40]. Signals of both proteins were detected on the merozoite surface of wild type and ΔpPK1 parasites (Fig. 4E), suggesting that pPK1 has no or a limited role in AMA1 and EBL secretion in the P. yoelii 17XL line.

pPK1 plays an important role for exflagellation center formation
Next, we examined the expression of pPK1 in sexual stage parasites using pPK1-myc parasites. Western blot analysis was performed for enriched gametocytes from mice treated with sulfadiazine for 2 days before collection. Only a few asexual stage parasites (< 1% of gametocytes) were observed by microscopy after sulfadiazine treatment. A single~40-kDa band, similar to the band from asexual stage parasites, was detected with anti-Myc antibody (Fig. 5A). IFA against α-tubulin IIpositive male and PyS25-positive female gametocytes revealed that pPK1-myc fluorescent signals overlapped with the DAPI-nucleus signals in both gametocytes. The fluorescent signal was also detected at the nucleus of microgametes (Fig. 5B). A patchy fluorescent pattern was detected along the ookinete plasma membrane and in this parasite stage the pPK1-myc signals did not overlap with nucleus signals (Fig. 5B). These data indicate that PypPK1 is expressed in sexual stage parasites and its localization differed among stages.
To investigate a role of pPK1 in sexual stage parasites, firstly gametocytemias were compared between pPK1-myc and ΔpPK1 parasites by counting immuno-stained parasites labeled with male and female gametocyte markers. Male gametocytemias of ΔpPK1 #1 and #2 parasites on day 3 post-infection were 0.08 ± 0.03% and 0.10 ± 0.01%, respectively, which were not significantly different from that of pPK1-myc parasites (0.11 ± 0.03%). Female gametocytemias of ΔPypPK1 #1 and #2 parasites were 0.12 ± 0.04% and 0.11 ± 0.03%, respectively, and the difference from the pPK1-myc parasites (0.14 ± 0.02%) was not significant (Fig. 6A). Secondly, we compared the ratio of gametocyte egress by counting male or female parasites within or outside the erythrocyte after XA activation. The signals of non-egressed gametocytes (with anti-α-tubulin II or anti-PyS25) were co-localized with the erythrocyte marker TER-119 signals, whereas the signals of egressed gametocyte were not co-localized with TER-119 signals (Fig. S3). The egress ratios were calculated for male and female gametocytes independently at 15 and 30 min post XA induction, respectively (Fig. 6B). Male gametocyte egress rates of ΔpPK1 #1 and #2 parasites were 54 ± 4% and 53 ± 4%, respectively, which was not significantly different from that of pPK1-myc parasites (57 ± 3%). Female gametocyte egress rates of ΔpPK1 #1 and #2 parasites were 33 ± 5% and 34 ± 5%, respectively, and no significant difference was detected from that of pPK1-myc parasites (34 ± 2%). Egressed male gametocytes initiate three rounds of endomitosis to produce eight microgametes, followed by axoneme assembly and exflagellation to release eight motile male gametes [41]. Thus, thirdly, we evaluated exflagellation rates using the ratio of the number of exflagellated male gametocytes (Fig. 6C left) per the number of egressed male gametocytes (Fig. 6C right). The exflagellation rates of ΔPypPK1 #1 and #2 parasites were 31.5 ± 8.4% and 23.1 ± 2.7%, respectively, and not significantly different from that of pPK1-myc parasites (33.4 ± 5.3%) (Fig. 6D). Fourthly, the effect on the exflagellation center formation was examined; that is, the clustering of exflagellating microgametes with surrounding erythrocytes. The number of exflagellation centers per erythrocyte at day 3 after parasite inoculation to mice was normalized by the male gametocytemia, the egress rate, and the exflagellation rate (Fig. 6E). The normalized percentage of exflagellation centers per exflagellated male gametocyte were 43 ± 17 and 19 ± 21 for ΔPypPK1 #1 and #2 parasites, respectively, significantly lower than the value for pPK1-myc parasites (100 ± 22%). Microscopic observation revealed that some ΔpPK1 parasites successfully formed exflagellation centers (Fig. 7A middle; Movie S2), and some ΔpPK1 parasites failed to form (Fig. 7A bottom; Movie S3), whereas the exflagellation center was always formed for wild type parasites (Fig. 7A top and Movie S1). Lastly, the number of mosquito midgut oocysts were compared between ΔpPK1 and pPK1-myc parasites. The numbers of oocysts per mosquito were counted at day 11 after blood feeding. The median and interquartile range of oocyst numbers of ΔpPK1 #1 and #2 parasites were 0.0 (0.0-3.0) and 0.0 (0.0-2.0), which were significantly lower than that of pPK1-myc parasites (4.5, 1.0-13.5) (Fig. 6F), consistent with the decreased number of exflagellation center formation.

Deletion of pPK1 did not affect the microgamete surface localization of PyS230 and PyMiGS
Eksi et al. (2006) reported that disruption of the microgamete surface protein Pfs230, a soluble protein with an N-terminal signal peptide sequence and an array of 6-cysteine motif domains, in P. falciparum resulted in a significant reduction in exflagellation center formation [42], a phenotype similar to our observation for the P. yoelii ΔpPK1 parasites. P. yoelii microgamete surface protein (PyMiGS) is a protein with an Nterminal signal peptide sequence, aspartyl protease-like domain, and one transmembrane region at the C-terminal side and is expressed within the osmiophilic body of male gametocytes and on the microgamete surface [31]. The deletion mutant of PyMiGS in P. yoelii showed an impaired exflagellation activity [31]. Thus, we examined whether the microgamete surface localization of PyS230, a P. yoelii ortholog of Pfs230, and PyMiGS is affected in ΔpPK1 parasites. IFA against microgametes under non-permeabilized condition was able to detect PyS230 and PyMiGS on the surface of ΔpPK1 microgametes similar to wild type parasites (Fig. 7B), indicating that pPK1 did not affect the surface localization of PyS230 and PyMiGS. Antibody against a cytoplasmic protein α-tubulin II did not stain microgametes under non-reducing conditions, confirming that the plasma membrane was intact (Fig. 7C).

Discussion
In this study we analyzed a Plasmodium yoelii pseudokinase (PypPK1; PY17X_1220300) that is highly expressed in both schizont and male gametocyte stages of P. berghei. Deletion of PypPK1 resulted in a significant growth defect and lower erythrocyte invasion efficacy compared to the parental parasite line. Although no differences were seen in male and female gametocytemias and gametocyte egress rates, the number of exflagellation centers and midgut oocysts were significantly reduced in the PypPK1 deletion mutant lines. We showed that PypPK1 participates in regulation of parasite erythrocyte invasion. During this process, merozoites sequentially secret microorganelle proteins such as AMA1 from micronemes and EBL from dense granules, in the case of the P. yoelii 17XL line [29,40]. AMA1 secretion is regulated by a hierarchical phosphorylation cascade by PKAc (cAMP-dependent protein kinase catalytic subunit) and GSK3 (glycogen synthase kinase 3) [43,44]. Thus we thought that PypPK1 could be involved in the process to discharge these microorganelles. Indeed,  recently reported that when PDEβ, a molecule responsible for cAMP regulation and located upstream of PKA in the same signal cascade, was conditionally disrupted in P. falciparum, the pPK1 ortholog was phosphorylated > 2-fold relative to the control [45]. This suggested possible involvement of pPK1 in a cAMP signaling cascade. If this is the case, secretion of microorganelle proteins could be affected; however, we did not see clear differences for at least the secretion of AMA1 and EBL in the PypPK1-deleted P. yoelii 17XL line. Recent phosphor-proteome analysis of a P. falciparum line, in which the PKAc gene was conditionally disrupted, did not detect pPK1 as a substrate of PKA [46], which was not inconsistent to our result. Together, pPK1 appears to be involved in processes not linked to microorganelle discharge events.
PypPK1 signals were localized at the apical side of the merozoite and a punctate pattern was seen at the ookinete stage; however, they were colocalized with nuclei at the gametocyte and gamete stages, even though a nuclear localization sequence was not predicted [27]. Colocalization with nucleus signals were not seen at the merozoite and ookinete stages. Nucleus localization of pseudokinases without NLS have been reported in other organisms. For example, the pseudokinase Tra1 of Saccharomyces cerevisiae is a phosphatidylinositol 3-kinase-related kinase family protein without NLS but is transported to and anchored at the nucleus using its C-terminal phosphoinositide 3-kinase domain [47]. Another example is a vaccinia B12 pseudokinase, which interplays with B1 kinase and has a potent inhibitory activity [48]. PypPK1 signals overlapping with the nucleus at gametocyte and gamete stages may indicate that this protein is anchored and has a role at the nucleus at these stages. This is the second report after Pfs230 to exhibit a phenotype of reduced exflagellation center formation in malaria parasites, without reducing the number of exflagellated male gametocytes. Aberrant morphology and motility were not clearly observable for ΔPypPK1 male gametocytes (Movie S2, S3), thus this is due to the reduced adherence activity of the microgametes to erythrocytes (Movie S3). Templeton et al. (1998) showed that human erythrocyte adherence to P. falciparum microgametes is sialic acid and glycophorin A dependent and that infectivity to mosquitoes is significantly reduced when sialic acids are removed from gametocyte-infected human erythrocytes [49]. Thus, we interpreted that the reduction of exflagellation formation observed in ΔPypPK1 parasite lines participated in the reduced oocyst numbers. However, because we did not assess the effect of ΔPypPK1 on steps after exflagellation center formation and before oocyst formation, other steps could also participate in the reduced oocyst numbers. The reduced activity of microgametes to adhere to erythrocytes could be due to the surface exposure of parasite adhesins; however, this in unlikely the case because we found two such candidate proteins PyS230 and PyMiGS were detected on the surface of the microgametes of ΔpPK1 parasites. Thus at least pPK1 is not responsible for the surface localization of these proteins. If they are responsible for erythrocyte adherence, their function may be affected in ΔPypPK1; for example, through altered posttranslational modification or complex formation with other molecules.
In conclusion, we characterized a novel pseudokinase pPK1 in P. yoelii and showed that it possesses roles in merozoite invasion and exflagellation center formation. Further studies to determine the mode of action of this and other pseudokinases are of paramount interest and would provide potential targets to fight against malaria.