Plasmodium microtubule-binding protein EB1 is critical for partitioning of nuclei in male gametogenesis

ABSTRACT Sexual reproduction of the malaria parasites is critical for their transmission to a mosquito vector. Several signaling molecules, such as kinases and phosphatases, are known to regulate this process. We previously demonstrated that Plasmodium falciparum (Pf) Ca2+-dependent protein kinase 4 (CDPK4) and serine/arginine-rich protein kinase 1 (SRPK1) are critical for axoneme formation during male gametogenesis, with genetic deletion of either gene causing a complete block in parasite transmission to the mosquito. A comparative phospho-proteome analysis of Pfcdpk4− and RNA-seq analysis of Pfsrpk1− gametocytes showed that these kinases regulate similar biological processes linked to both microtubule (MT) dynamics and cell motility. One of these proteins was a nuclear MT-associated End Binding protein 1 (EB1), which was hypophosphorylated in Pfcdpk4− gametocytes. To study the functional relevance of EB1, we created gene deletion parasites for EB1. We further demonstrate that Pfeb1− parasites like WT NF54 parasites proliferate normally as asexuals and undergo gametocytogenesis and gametogenesis. Strikingly, these parasites suffer a severe defect in nuclear segregation and partitioning of nuclei into emerging microgametes. Further genetic crosses utilizing male- and female-sterile parasites revealed that Pfeb1− parasites only suffer a male fertility defect. Overall, our study reveals an essential function for PfEB1 in male gamete nuclear segregation and suggests a potential therapeutic avenue in the design of transmission-blocking drugs to prevent malaria transmission from humans to mosquito. IMPORTANCE Gametogenesis and subsequent gamete fusion are central to successful transmission of the malaria parasites to a female Anopheles mosquito vector and completion of the sexual phase of the parasite life cycle. Male gametogenesis involves the formation of axonemes inside male gametes from male gametocytes via active cytoskeleton remodeling. The tubulin and tubulin-binding proteins are, thus, attractive anti-malarial drug targets. In the present study, we demonstrate that a microtubule-binding protein PfEB1 is essential for male gamete fertility, specifically for the inheritance of nuclei from activated male gametocytes. Targeting PfEB1 function may provide new avenues into designing interventions to prevent malaria transmission and disease spread.

meal. Mosquitoes are infected only by the parasite stages that arise after fertilization, and thus, malaria transmission is dependent on successful gamete development.
The morphologically distinct changes during gametocytogenesis require active cytoskeletal remodeling, and male gametes depend on microtubule (MT)-driven motility to fertilize females. MTs are cytoskeletal filaments with a diameter of ~25 nm, composed of alternating rings of α-and β-tubulin dimers (1). Plasmodium tubulin is more similar to that of plants than to mammalian tubulin, and inhibitors targeting parasite tubu lin specifically disrupt their growth without affecting human MTs (2). More recently, antibodies targeting Pf α-tubulin I have shown strong transmission reducing efficacy (3). Therefore, cytoskeletal proteins such as MTs and their associated proteins (MAPs) are attractive drug or vaccine targets, especially during sexual-stage development. Tubulinbinding proteins in eukaryotic cells facilitate the assembly and disassembly of MTs, which drive cell division, differentiation, and motility (1,4,5). In eukaryotes, MTs also form the scaffolds of the mitotic and meiotic spindles of dividing cells (6). Plasmodium invasive stages and gametocytes maintain cell shape and rigidity with a pellicle undergirding the plasma membrane, which is composed of a network of subpellicular microtubules (SPMTs) (7) and associated proteins (8) underneath a double membrane known as the inner membrane complex (9,10). Two additional populations of microtubules are found in the gametocyte nuclear spindle or hemi-spindle (11) and in the cytoplasm. While the cytoplasmic microtubules are short lived and only appear in early-stage gametocytes on the opposite side of the cell from the developing pellicle (10), the nuclear microtubules are present in stages III/IV gametocyte (7,12).
Male gametogenesis is a rapid process and culminates in the formation of eight flagellated microgametes that egress from the infected RBC (exflagellation) (13). Microtubule-binding proteins impact flagellar formation and length via involvement in the equilibrium of assembly and disassembly of tubulin at the plus end (14,15). In Apicomplexan parasites, the MT-binding apicortin protein complex comprise of a doublecortin domain and a partial p25α domain, which exhibit MAP-like functions (16). In rodent malaria parasite, Plasmodium yoelii (Py) (Pyp25α) is crucial for exflagellation and, thus, essential for microgametogenesis (17). A recent study reported the nucleation of SPMTs in P. falciparum gametocytes at the outer centriolar plaque, a non-mitotic microtubule organizing center (MTOC) embedded in the nuclear membrane of the parasite (12). This study also reported that classical mitotic machinery components, including centriolar plaque proteins, Pfcentrin-1 and -4, microtubule-associated protein, end-binding protein-1 (EB1), a kinetochore protein, PfNDC80, and centromere-associated protein, PfCENH3, are involved in the nuclear microtubule assembly/disassembly (12).
End-binding EB1 proteins are microtubule plus-end-tracking proteins (+TIPs) (18), which play a crucial role in regulating MT dynamics by binding growing MT ends and interacting with a network of +TIPs (18). In budding yeast mutants, BIM1 (an EB1 homolog) plays a role in MT search and capture, cell polarization, and chromosome stability (19). BIM1 gene deletion results in increased net polymerization (19). EB1 demonstrates a critical role in mitosis by aiding in kinetochore clustering and MT stability during segregation of chromosomes toward cell poles (20). However, in plant cells, EB1 colocalizes with MTs, demonstrates +TIP abilities, and influences MT dynamics by impacting polymerization rates (21). In the Apicomplexan parasite, Toxoplasma gondii (Tg), TgEB1 is localized to the nucleus and tightly regulates MT dynamics (22). EB1 has recently been localized to the full length of the nuclear MT bundles of P. falcipa rum during asexual and sexual erythrocytic stages (12), suggesting a possible role in chromosomal segregation during these developmental stages.
We recently demonstrated that both Pfcdpk4 − parasites (23) and Pfsrpk1 − para sites (24) exhibit severe defects in exflagellation and microgametogenesis. Also, our transcriptomic and proteomic studies on both these parasite gene knockout lines revealed that they have a strong dysregulation in the expression of genes encoding transcripts for biological processes linked to MT and cell motility (23,24). Interestingly, PfEB1 was hypo-phosphorylated in Pfcdpk4 − parasites at S 15 (23), and its transcripts were severely downregulated in Pfsrpk1 − parasites (24). Therefore, we sought to determine PfEB1 function during sexual stages. We created gene deletion parasites (Pfeb1 − ) via CRISPR/Cas9-mediated transgenesis. Pfeb1 − parasites showed normal asexual bloodstage growth and underwent normal gametocytogenesis. Strikingly, Pfeb1 − parasites exhibited a robust block in transmission to mosquitoes. Careful examination of microgametes revealed a severe defect in partitioning of nuclei in microgametes in Pfeb1 − parasites, while macrogametes formed normally. Further genetic crosses involving male-only and female-only sterile parasite lines demonstrated that PfEB1 is critical for male gamete fertility only.

PfEB1 is not required for intra-erythrocytic parasite development
PfEB1 is encoded on chromosome 3 with gene identifier PF3D7_0307300 (https:// plasmodb.org/plasmo/app/record/gene/PF3D7_0307300). PfEB1 domain analysis using SMART (http://smart.embl-heidelberg.de/) revealed that it contains a calponin homology (CH) domain at the N terminus and a namegiving EB1 domain at the C terminus, separated by a linker region (Fig. 1A). The linker region of PfEB1 is longer than that of Human EB1 (HsEB1) (Fig. 1A). Further structural analysis using Alphafold (https:// alphafold.ebi.ac.uk/) revealed a conserved EB1 structure for PfEB1 (Fig. 1B). A sequence alignment of the various Plasmodium spp. revealed that EB1 shows high degree of conservation in its CH domain and EB1 domain including a conserved serine residue at amino acid position 15 (S 15 ) (Fig. S1). The CH domain of HsEB1 is responsible for the MT binding observed from EB1 localization to MT plus ends (25) and while the EB1 domain binds to adenomatous polyposis coli-binding domain (26,27). For studying the role of EB1 in parasite development, endogenous PfEB1 gene deletion was achieved using CRISPR/Cas9 (Fig. 1C). A set of diagnostic PCRs with oligonucleotides specific to the PfEB1 locus and its upstream (5′) and downstream (3′) regions were used for confirmation of gene deletion parasites (Pfeb1 − ) ( Fig. 1C to E). Two clones for Pfeb1 − parasites (clones 3D4 and 11C4) were used for phenotypic analysis. Pfeb1 − parasites (clones 3D4 and 11C4) as well as wild-type (WT) NF54 parasites were used for a comparative asexual parasite growth assay. Growth of the parasites was monitored over two continuous asexual replication cycles with Giemsa-stained thin smears prepared from in vitro cultures every 48 hours. Microscopic enumeration of parasitemias indicated that Pfeb1 − parasites grow similarly to WT PfNF54 parasites ( Fig. 2A), demonstrating that PfEB1 is not critical for asexual blood-stage development and replication.

Pfeb1 − parasites undergo gametocytogenesis and exhibit normal exflagella tion
Pfeb1 − parasites were next analyzed for their ability to undergo gametocytogenesis. On day 15 of in vitro culture, gametocytemia for both Pfeb1 − parasites (clones 3D4 and 11C4) and WT PfNF54 parasites were scored using Giemsa-stained culture smears and micro scopic inspection. Pfeb1 − gametocytemias were similar to WT PfNF54 (Fig. 2D), indicated by their ability to undergo gametocytogenesis, develop through all five gametocyte stages, and further develop into mature male and female gametocytes ( Fig. 2B and C). Immunofluorescence assays (IFAs) were performed to better visualize male and female stage V gametocytes using anti-PfP230p (28) and anti-Pfg377 antibodies (29), respec tively. This revealed normal differentiation of Pfeb1 − parasites into both genders and further confirmed results obtained from Giemsa-stained smears (Fig. 2C). To next analyze the ability of Pfeb1 − mature gametocytes to undergo gametogenesis, day 15 WT PfNF54 and Pfeb1 − parasites were activated by addition of O + human serum and a decrease in temperature from 37°C to room temperature. Wet mounts were prepared using activated gametocyte cultures and then viewed under bright-field microscopic illumination at 40× magnification to measure the exflagellation centers. Pfeb1 − parasites exhibited similar numbers of exflagellation centers as WT parasites, indicating no exflagellation defects ( Fig. 3A). We next performed IFAs on activated gametocytes and free microgametes using anti-tubulin antibodies. This revealed that microgamete formation in Pfeb1 − parasites was similar to WT PfNF54 parasites ( Fig. 3B and C). During microgametogenesis, male gametocytes rapidly undergo three rounds of DNA replication (8N), which is allocated in eight flagellar microgametes along with axoneme formation. Therefore, we next examined the DNA allocation in free microgametes ( Fig. 3B and C). Strikingly, however, examination of free microgametes for WT PfNF54 and Pfeb1 − parasites for DNA staining revealed that the majority of Pfeb1 − microgametes were DAPI negative ( Fig. 3D), suggesting a possible male fertility defect. To determine a possible defect in female gametogenesis, we performed IFAs using Pfs25 antibodies, which mark activated female gametocytes and macrogametes. These revealed that Pfeb1 − parasites form normal macrogametes like WT PfNF54 parasites (Fig. 3E).

The Pfeb1 − parasites do not transmit to the mosquito vector due to reduced male fertility
Next, Pfeb1 − gametocytes were examined for their transmissibility to female Anopheles stephensi mosquitoes. Infectious blood meals were prepared for WT PfNF54 and Pfeb1 − stage V gametocytes using standard procedures, and gametocytes were then fed to mosquitoes through membrane feeders. Mosquito midguts, dissected and analyzed under bright-field microscope at 10× magnification on day 7 post-feed, revealed a severe reduction in oocyst numbers for Pfeb1 − parasite infections in comparison to well-infected WT controls (Fig. 4A). Since P. falciparum male and female gametocytes exist together, it is not feasible to determine a sex-specific fertility and transmission defect; genetic crosses were performed between Pfeb1 − parasites and male-sterile Pfcdpk4 − parasites (23) and female-sterile Pfmacfet − parasites (30). For this, the stage V gametocytes from both Pfeb1 − parasites and Pfcdpk4 − parasites or Pfeb1 − parasites and Pfmacfet − parasites were mixed and fed to the same mosquitoes. The genetic cross between Pfeb1 − parasites and Pfcdpk4 − parasites showed no transmission, while the genetic cross between Pfeb1 − parasites and Pfmacfet − parasites showed productive transmission. This demonstrated a Research Article mBio male-gamete-specific fertility defect in Pfeb1 − gametocytes (Fig. 4B). Taken together, the data show that PfEB1 is critical for parasite transmission to the mosquito vector via a crucial role during microgametogenesis.

DISCUSSION
To complete the sexual phase of the P. falciparum life cycle, a subset of asexually replicating parasites commit to the sexual pathway, followed by differentiation into gametocytes and uptake by Anopheline mosquitoes. Environmental triggers within the mosquito midgut activate male gametocyte differentiation into eight flagellated microgametes in a rapid process called exflagellation (13). Microgametes show vigorous motility and, upon encountering a female, initiate fertilization to enable the union of the two haploid genomes and sexual recombination. In Plasmodium, the microtubules play important structural roles in the invasive forms (31,32) as well as in gametocytes, power the motility of microgametes, and play critical role in mitosis (33,34). During mitosis in Plasmodium, the chromosomes are present in an uncondensed state, and the nuclear membrane remains intact as nuclear microtubules capture the kinetochores (34,35). An Representative images are shown. Parasite DNA was visualized with DAPI (blue). Scale bar = 5 µm. Merge I indicates the merged images for red and green panels.
Merge II indicates the merged images for red, green, and blue panels. DAPI, 4′,6-diamidino-2-phenylindole; DIC, differential interest contrast. Male and female gametocytes are indicated using sex symbols shown on the left side of the image panels. (D) On day 15 of in vitro culture, gametocytemia were measured using thin Giemsa-stained smears. Data from three biological replicates were averaged and presented as mean ± standard deviation. ns, not significant.
Research Article mBio electron-dense acentriolar centrosome called the centriolar plaque, which is embedded in a pore in the nuclear envelope, serves as MTOC and is likely present throughout the cell cycle (34,36,37). Exflagellation is an astounding cellular process producing eight motile exflagellae, each with a full genome equivalent, from a 1N haploid microgamete following the assembly and stabilization of the nuclei with their associated nuclear envelope non-mitotic MTOC, outer centriolar basal body, and axoneme (38). Proteins involved specifically with basal body function during male gametogenesis of Plasmo dium berghei (Pb) and most likely for P. falciparum include SAS6 (39), SAS4 (40), and Kinesin-8B (41). Our study reveals a critical function for nuclear MAP EB1 in nuclear partitioning during P. falciparum microgametogenesis, which is crucial for male gamete fertility. The EB1 family proteins are evolutionarily conserved and have been identified in every organism and nearly all cell types (18). EB proteins comprise of ~300 amino acids with an N-terminal CH domain, a linker region, C-terminal domain, and overlapping coiled-coil and EBH domains, with the CH and coiled-coil domains being conserved in EB1 proteins from diverse organisms (20). However, EB1 is the only member of the +TIP family with an ortholog in P. falciparum (12). PfEB1 contains a CH domain and an EB1 domain separated by a linker region. The linker region between the CH and EB1 domains in P. falciparum is larger compared to that in HsEB1, despite having similar respective positioning of their EB1 domains and CH domains of similar sizes. HsEB1 is heavily modified at numerous sites (Y 6 , S 7 , S 9 , S 40 , K 60 , K 66 , Y 71 , K 76 , K 83 , K 89 , K 95 , K 100 , Y 119 , K 122 , Y 124 ,   (42). So far, no PTMs have been reported for PfEB1, although in our previous data sets, we found S 15 is phosphorylated in gametocyte stages and hypo-phosphorylated in Pfcdpk4 − parasites, which exhibited a severe male gametogenesis defect (23). The Pfcdpk4 − gametocytes also show hypo-phosphorylation of several parasite kinases, including ARK2, RIO1, and SRPK2 (23). Yeast EB1 (BIM1) is phosphorylated by Aurora kinase homolog lpl1 (43), while HsEB1 co-immunoprecipitates with Aurora kinase B (44). Also, HsEB1 is required to enrich Aurora kinase B at inner centromeres in an MT-dependent manner, resulting in phosphorylation of both kinetochore and other chromatin sub strates (45). A recent study in P. berghei also demonstrated the interaction between ARK2 and EB1 (46). It is reasonable to propose that PfCDPK4-mediated phosphorylation of ARK2 and/or ARK2-mediated phosphorylation of EB1 at S 15 may be regulating its microgametogenesis function. This hypothesis is further supported by a recent study demonstrating the role of EB1 S 15 in spindle-kinetochore attachment during microgame togenesis in P. yoelii (47). It will be interesting to explore the role of S 15 phosphorylation on PfEB1 for its relevance during parasite transmission. PfEB1 contributes to the functions of an MTOC embedded in the nuclear membrane, which stabilizes and controls the elaborate MT network in gametocytes (12). PfEB1 associates with nuclear MTs, where they bind along the full length of the MT bundles, Research Article mBio rather than being strictly localized to the plus end (12). Captured chromatin on the nuclear MT bundles in gametocytes has been indicated by the elongated localization of EB1 in conjunction with the localized regions of chromatin to the nuclear MTs (12).
Further heterologous expression experiments have demonstrated that PyEB1 has an intrinsic MT-lattice binding property (47). The Pfeb1 − parasites described herein showed that PfEB1 is not required for asexual blood-stage proliferation and gametocytogenesis. Strikingly, the majority of the male Pfeb1 − gametes lack a nucleus, indicating a defect in nuclear division in the absence of EB1, which may result in a motility defect or early death in these microgametes. It is possible that some parasites can compensate for the defects in MT stabilization in the absence of EB1, but there is a delay in mitosis completion, which results in the absence of nuclei in majority of the microgametes. This would be in stark contrast to PbEB1 as Pbeb1 − parasites form normal microgametes and do transmit to the mosquitoes (46). Interestingly, the Pyeb1 − parasites like Pfeb1 − also exhibit a nuclear segregation defect in microgametes which is restored upon comple mentation with PfEB1 (47). This reflects a species-specific function for EB1 in different malaria parasites. Human Kinesin Family Member 18B (KIF18B) is an EB1-binding protein localized to the nucleus during interphase, where it is enriched on astral MT plus ends (48). Furthermore, Kif18B regulates astral MT organization and length during early mitosis (48). Similarly, P. berghei kinesin-8B is essential for microgamete formation due to its key role in basal body formation and assembly, as well as structural organization of the axoneme (41). Taken together, kinesin-8B is predicted to interact with EB1 during microgametogenesis. Proper kinetochore assembly is essential for ensuring chromosome segregation during mitosis and, thus, is an essential step in parasite transmission. This process is regulated by centromere proteins CENP-A and CENP-C, kinesins, and MAPs (49). Kinesin-8s, specifically, are associated with MTs and influence MT dynamics (41). The Plasmodium genome encodes kinesin-8X and kinesin-8B, the latter of which is essential in microgamete development due to its role in the formation of basal bodies and axoneme development (41). In related Apicomplexan parasite, T. gondii, TgEB1 regulates nuclear divisions by controlling spindle dynamics and its association to the kinetochore NDC80 complex (22). Indirectly, TgEB1 secures kinetochore organization and promotes accurate chromosome segregation (22). Both in P. falciparum (12) and P. berghei (46), EB1 associates with kinetochore marker NDC80 (46), while PbEB1 also associates with basal body marker SAS4 (46). Another basal body marker PbSAS6 is known to have role in nuclear allocation in microgametes (39). Since PfEB1 associates with kinetochore (12) and Pfeb1 − parasites show an abnormal nuclear allocation, we hypothesize that PfEB1 may function in endomitosis during microgametogenesis. The high-resolution microscopy and live-cell imaging experiments would be required to confirm this hypothesis.
In summary, our study demonstrates a crucial role for PfEB1 during male gametogen esis and parasite transmission to the mosquito and links a nuclear microtubule-binding protein to DNA segregation during microgametogenesis.

P. falciparum culture and transfection
Standard procedures were followed to culture P. falciparum parasites (WT PfNF54 and Pfeb1 − ) as asexual blood stages, which received complete RPMI 1640 media, supplemen ted with 10% (vol/vol) human serum every 24 hours. O + human RBCs (Valley Biomedical, VA, USA) and O + human serum (Valley Biomedical, VA, USA or Interstate Blood Bank, TN, USA) were used to generate in vitro gametocytes using methods published elsewhere (50). PfEB1 Gene deletion parasites, Pfeb1 − , were generated via CRISPR/Cas9 strategy. Recombinant parasites were drug selected, PCR genotyped, and then cloned by limiting dilution. The clonal parasites were confirmed by a set of genotyping PCRs (Fig. 1D and E). Functional assays used two individual clones for Pfeb1 − parasites (clones 3D4 and 11C4).

Measurement of asexual blood-stage growth and gametocyte development
WT PfNF54 and Pfeb1 − parasites were synchronized at ring stages using 5% sorbitol, and plating was done at equal parasitemia (1%) in six-well plates for comparative analysis of growth rates. Parasite growth for two consecutive asexual replication cycles was quantified by scoring parasitemia on thin Giemsa-stained smears. Likewise, gametocy temia was quantified on day 15 of in vitro culture on thin Giemsa-stained smears for comparison of sexual development.

Indirect immunofluorescence assays
IFAs were performed on asexual and sexual blood-stage parasites and exflagellating microgametocytes using thin smears prepared on Teflon-coated slides as described elsewhere (50). Antigens were visualized using anti-species antibodies. Images were acquired using a 100× 1.4 NA objective 90 (Olympus) on a Delta Vision Elite High-Reso lution Microscope (GE Healthcare Life Sciences).

Statistical analysis
Data collected were expressed as mean ± SD. One-way analysis of variance with post hoc Bonferroni multiple comparison test or unpaired two-tailed Student's t test were used, as indicated, to determine statistical differences. Statistical significances were calculated using GraphPad Prism 9.4, with values of P < 0.05 being considered significant. Significance is represented in the figures as follows: ns, not significant, P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001. The authors declare no competing financial or non-financial interests.

DATA AVAILABILITY
All other relevant data are available from the authors upon reasonable request.