The novel Plasmodium berghei protein S14 is essential for sporozoite gliding motility and infectivity

The Plasmodium life cycle alternates between a mosquito and a mammalian host, which involves invasive and replicative stages. When a mosquito probes for the blood meal in an infected mammalian host, it ingests the gametocytes, which develop further into gametes that fuse to form the zygote. The zygote transforms into an ookinete, which establishes an oocyst in the mosquito midgut that produces thousands of midgut sporozoites. Following oocyst rupture, sporozoites are released into the hemolymph and further invade the mosquito salivary gland (Douglas et al., 2015; Klug and Frischknecht, 2017). Sporozoites transform into liver stages after transmission to the mammalian host, forming thousands of merozoites that initiate the erythrocytic cycle (Prudêncio et al., 2006). Sporozoites are highly motile cells and rely on gliding motility to travel through different host species. Gliding motility is powered by an actomyosin motor and requires the coordinated action of a multiprotein machine known as the glideosome (Ferreira et al., 2021). The glideosome is located between the inner membrane complex (IMC) and parasite plasma membrane (PPM) (Keeley and Soldati, 2004). The gliding-associated proteins (GAPs) tether MyoA to the IMC and hold the PPM and the IMC together (Boucher and Bosch, 2015). The transmembrane protein thrombospondin-related anonymous protein (TRAP) connects the submembrane motor to the extracellular environment (Frénal et al., 2017).

Plasmodium sporozoites can invade target cells in the mosquito and the mammalian hosts and many proteins that have been implicated in this process are expressed specifically in sporozoites. TRAP was the first protein found to be involved in sporozoite gliding motility and host cell invasion. TRAP knockout (KO) sporozoites failed to invade the salivary gland, and hemolymph sporozoites were nonmotile (Sultan et al., 1997). TRAP cytoplasmic tails incorporate a C-terminal tryptophan residue that is crucial for interaction with aldolase, which connects with an actin-myosin-based motor (Braumann et al., 2023; Buscaglia et al., 2003; Heintzelman, 2015). TREP (also known as S6 and UOS3) is also a TRAP family adehesin, and, by disruption, it has been shown that it is likely to have a role in parasite adhesion and gliding motility (Combe et al., 2009; Mikolajczak et al., 2008; Steinbuechel and Matuschewski, 2009). The micronemal-localized protein MAEBL has been found to be essential for sporozoite attachment to the salivary gland surface and infection (Kariu et al., 2002). The other proteins implicated in parasite motility and host cell invasion include MAEBL (Saenz et al., 2008), TLP (Heiss et al., 2008), RON4 (Giovannini et al., 2011), GEST (Talman et al., 2011), circumsporozoite protein (CSP), the small solute transporter PAT (Kehrer et al., 2016) and claudin-like apicomplexan microneme protein (CLAMP) (Loubens et al., 2023). The individual functions of these proteins are known. However, how they interact with each other to coordinate gliding motility and invasion is poorly understood.

The above proteins are specific to sporozoites; however, several proteins function in merozoites, ookinetes and sporozoites. Proteins such as GAP40, GAP45 and GAP50, together with the myosin A tail domain interacting protein (MTIP), are known to cluster MyoA with the IMC (Daher and Soldati-Favre, 2009; Frénal et al., 2017; Pazicky et al., 2020; Poulin et al., 2013; Ripp et al., 2022). In P. falciparum, GAP45 is myristoylated and palmitoylated and required for membrane targeting (Qian et al., 2022; Rees-Channer et al., 2006). GAP45 in Toxoplasma gondii is involved in the recruitment of the motor complex (Frénal et al., 2010, 2017). P. falciparum (Pf)GAP50 is a transmembrane protein that anchors the invasion machinery in the IMC (Baum et al., 2006; Yeoman et al., 2011). In T. gondii, GAP45 and GAP50 form a complex with MyoA and its light chain, MLC1 (Frénal et al., 2017).

Many glideosome and surface adhesins have been found to be essential for parasite motility (Frénal et al., 2010). The exact or complete composition of the glideosome, however, is still unknown. Here, we identify and characterize a novel and essential component of the Plasmodium motility apparatus. We selected S14 from the transcriptome list of the top 25 of Plasmodium yoelii transcripts that were highly upregulated in sporozoites in a suppressive subtraction hybridization (Kaiser et al., 2004). The S genes, S2 (also known as CSP; PF3D7_0304600) (Klug and Frischknecht, 2017), S5 (PF3D7_0408600) (Engelmann et al., 2009), TREP (PF3D7_1442600) (Steinbuechel and Matuschewski, 2009), TRAP (also known as S8 and SSP2; PF3D7_1335900) (Sultan et al., 1997) and S15 (also known as SIMP; PF3D7_0814600) (Singh et al., 2022) are implicated in sporozoite gliding motility. In this study, we investigated the role of S14 in the P. berghei life cycle. We demonstrate that S14 might be an IMC-associated protein. We disrupted the gene and found that S14 is essential for sporozoite gliding motility, host cell invasion and malaria transmission.

We identified S14 (PBANKA_0605900) as a gene that was upregulated in sporozoites from transcriptome data of Plasmodium yoelii and identified its ortholog in P. berghei. Bioinformatic analysis using InterPro, SMART, CD-Search (NCBI), SignalP and TMHMM revealed that S14 lacks conserved domains, a signal peptide and transmembrane domains. SMART recognized a coiled-coil region along the acidic E-stretch from position 152 to 192, representing the low complexity of the protein but no recognizable known protein domains. Using NCBI protein BLAST search, we found that S14 is conserved within the genus Plasmodium, even to a small degree in Hepatocystis sp. ex Piliocolobus tephrosceles 2019. OrthoMCL also revealed that S14 is conserved within the genus Plasmodium. Related proteins are absent in other Apicomplexa like Toxoplasma and Cryptosporidium. Sequence similarity using self-organizing maps (SOM) revealed that S14 is mainly restricted to Plasmodium (Fig. S1A). Sequence alignments in UniProt revealed 42.6% similarity between P. berghei and P. falciparum S14 proteins, and 87.21% similarity between P. berghei and P. yoelii (Fig. S1B).

We started our study by validating the transcripts of S14 in different stages of the parasites using quantitative real-time PCR (qRT-PCR). We found that S14 was predominantly transcribed in midgut and salivary gland of P. berghei sporozoites (Fig. 1A). To investigate further whether S14 transcripts are translated, we endogenously tagged the S14 gene with 3×HA–mCherry using double crossover homologous recombination (Fig. S2A). The correct site-specific integration was confirmed by diagnostic PCR (Fig. S2B). First, we checked for mCherry expression in the blood stages, and the signal was not detected (Fig. S3A). Next, we initiated the mosquito cycle and monitored mCherry expression in sporozoites and liver stages. We found mCherry expression in sporozoites but not in the liver stages (Fig. 1B,C; Fig. S3B). The tagging of S14 with 3×HA–mCherry did not affect sporozoite development in the mosquito, in vitro gliding motility or infection in mice inoculated intravenously (Fig. S2C–E). Analysis of the mCherry pattern on sporozoites revealed localization on the membrane of the sporozoites (Fig. 1D). Furthermore, an immunofluorescence assay with anti-CSP and anti-mCherry antibodies confirmed S14 localization on the membrane (Fig. 1E). Expression of the S14–3×HA–mCherry fusion protein (∼66.5 kDa) was also confirmed by western blotting using an anti-mCherry antibody. The appearance of an extra band with a higher molecular mass in immunoblotting was possibly due to the post-translational modification of S14 (Fig. 1F). Similar electromobility shifts of GAP45 caused by the palmitoylation have been reported (Rees-Channer et al., 2006). These results indicate that S14 is a sporozoite-specific membrane protein; however, further investigation is needed to confirm its post-translational modification status.

To investigate the role of S14 in the P. berghei life cycle, we disrupted the gene by double-crossover homologous recombination (Fig. S4A). The drug-resistant parasites expressing GFP indicated successful transfection (Fig. S4B). We obtained two clonal KO parasite lines (denoted c1 and c2) by limiting dilution of the parasites, integration, and the absence of the wild-type (WT) locus was confirmed by diagnostic PCR (Fig. S4C). Finally, two S14 KO clonal lines were confirmed by Southern blotting, which showed the modified locus in the KO parasites (Fig. S4D).

The S14-complemented parasite line was also generated to check the specificity of the phenotype (Fig. S4E). First, a new S14 KO parasite with cytosine deaminase and uridyl phosphoribosyl transferase (yFCU) was generated, and then an S14 expression cassette consisting of the 5′UTR, ORF and 3′UTR was amplified and transfected into S14 KO (yFCU) parasites. After recombination, the WT locus was amplified in S14-complemented (hereafter S14 comp) parasites (Fig. S4F).

Next, we checked the propagation of the asexual intraerythrocytic cycle of KO parasites, which was comparable to that of WT GFP parasites (Fig. S5). To analyze the phenotype of S14 KO parasites in the mosquito stages, we transmitted them to mosquitoes by allowing them to probe for a blood meal on infected Swiss mice. We observed the mosquito midgut and salivary glands on day 14 and day 19 post a blood meal. We found that oocyst formation and development of sporozoites were comparable to those of WT GFP parasites (Fig. 2A–D). However, no sporozoite-associated GFP signals were observed in salivary glands, and the number of salivary glands sporozoites per mosquito was severely reduced (154±68; mean±s.e.m.) (Fig. 2E,F). Genetic complementation of the KO parasites restored salivary gland sporozoite numbers to a level similar to that for WT GFP parasites (Fig. 2F). We counted the sporozoite numbers in the mosquito hemolymph to investigate whether the KO sporozoites failed to egress from the oocyst or could not invade the salivary gland. We found a higher accumulation of hemolymph sporozoites in KO-infected mosquitoes compared to the WT GFP-infected mosquitoes (Fig. 2G), suggesting that mutant sporozoites invaded the salivary gland in highly reduced numbers. These results indicate that mutant sporozoites egress from the oocysts normally but invade the salivary gland in highly reduced numbers.

To evaluate the ability of S14 KO-infected mosquitoes to transmit malaria, C57BL/6 mice were subjected to bites from mosquitoes infected with S14 KO or WT GFP parasites. We found that mice inoculated with WT GFP or S14 comp sporozoites became patent on day 3, whereas S14 KO sporozoites failed to initiate blood-stage infection (Table 1). Next, we checked whether this in vivo infectivity defect was due to negligible salivary gland sporozoite load or whether mutant sporozoites failed to infect mice. For this, we intravenously inoculated C57BL/6 mice with 5000 hemolymph sporozoites and found that all the mice inoculated with KO sporozoites failed again to initiate blood-stage infection (Table 1).

To determine the stage-specific defect, we performed in vitro assays. First, we infected HepG2 cells with hemolymph sporozoites and observed the exoerythrocytic form (EEF) development at 40 hours post infection (hpi). We found EEFs in culture inoculated with WT GFP sporozoites but not in the KO-infected cells (Fig. 3A; Fig. S6A). Next, we checked the invasion capacity of S14 KO hemolymph sporozoites using double immunostaining with anti-CSP antibody, which differentiates the outside versus the inside of sporozoites (Rénia et al., 1988). S14 KO sporozoites showed a complete failure to invade hepatocytes (Fig. 3B; Fig. S6B), explaining their inability to establish an infection in mice. Next, we performed a development assay to check the ability of mutant sporozoite to transform into a bulb-like structure. For this, we incubated the mutant sporozoites in a transformation medium for 4 h. We found that S14 KO sporozoites retained the ability to transform into bulb-like structures (Fig. 3C,D). These data demonstrate that S14 KO sporozoites lost their infectivity to mammalian hosts due to the inability of sporozoites to invade cells.

S14 KO sporozoites being invasion deficient indicates that either S14 interacts with host receptors or cannot generate power to invade cells. As S14 lacks a signal sequence and transmembrane domain, we analyzed whether S14 is present on the plasma membrane or IMC. We treated S14–3×HA–mCherry hemolymph sporozoites with Triton X-100 to remove the plasma membrane. The Triton X-100-treated and untreated sporozoites were immunostained with anti-CSP and anti-mCherry antibodies. The CSP signal was lost in Triton X-100-treated sporozoites, whereas the S14–3×HA–mCherry signal was retained (Fig. 4A). Western blotting confirmed the immunofluorescence assay result, as the mCherry signal was detected in Triton X-100-treated sporozoites (Fig. 4B). To further confirm the IMC localization of S14, we generated antibodies against two IMC proteins, MTIP (Bergman et al., 2003) and GAP45 (Gaskins et al., 2004), and performed an immunofluorescence assay. The MTIP and GAP45 signals were retained in Triton X-100-treated sporozoites and colocalized with the S14–mCherry signal (Fig. 4A). This result indicates that S14 might be an IMC-associated protein.

Plasmodium parasites actively invade host cells in a manner powered by gliding motility (Frénal et al., 2017). Next, we checked the gliding motility of WT GFP and S14 KO hemolymph sporozoites. WT GFP sporozoites glided normally, whereas S14 KO sporozoites were found to be nonmotile (Fig. 5A). We counted the sporozoites associated with or without CSP trails to quantify the percentage of gliding sporozoites;∼53% of WT GFP sporozoites were associated with CSP trails, whereas no sporozoites associated with CSP trails were observed in S14 KO sporozoites (Fig. 5B). To pinpoint further the gliding patterns of WT and KO sporozoites, the movement of sporozoites were analyzed (Fig. 5C; Movie 1). Whereas 56% of the WT GFP sporozoites showed continuous motion, most of the S14 KO sporozoites (61%) remained attached and showed no movement (Fig. 5D). The rest of the S14 KO sporozoites showed patch, helical and waver movement (Wichers-Misterek et al., 2023). Next, we checked whether the S14 gliding function is linked to its secretion beyond the sporozoite membrane. Analysis of S14–3×HA–mCherry transgenic sporozoites revealed that S14 was not secreted beyond the sporozoite membrane (Fig. S7A,B). These data demonstrate that S14 is essential for the sporozoite gliding motility.

Plasmodium sporozoites exhibit a substrate-dependent gliding motility for which surface and IMC proteins are employed. We analyzed the expression and organization of two IMC proteins, MTIP and GAP45, and two surface proteins, CSP and TRAP, in S14 KO sporozoites. Immunostaining revealed an intact IMC and surface organization in S14 KO sporozoites (Fig. 6A). We quantified the fluorescence signals from immunostained sporozoites and found no difference between WT and KO parasites (Fig. 6B–E)

This study identified a novel Plasmodium protein, S14, which lacks signal peptide and transmembrane domains. However, it was found to be an IMC-associated protein that colocalized with the IMC protein GAP45. We found that S14 is a gliding-associated protein, with a core function in gliding motility and host cell invasion. Deletion of S14 resulted in the accumulation of a higher number of sporozoites in the hemolymph, which failed to invade the salivary gland and hepatocytes. RMgm-4741 (https://www.pberghei.eu/index.php?rmgm=4741) describes an S14 KO in P. berghei. However, the phenotype of that strain was not analyzed. RMgm-4742 (https://www.pberghei.eu/index.php?rmgm=4742) describes a replacement of P. berghei S14 with that from P. falciparum. The phenotype was not analyzed in detail, but the database describes normal numbers of oocysts and strongly reduced numbers of salivary gland sporozoites. No liver infection or blood infection was observed after the intravenous injection of sporozoites. These results indicate that the P. falciparum (PF3D7_1207400) ortholog cannot complement P. berghei S14 (PBANKA_0605900) during sporozoite production and liver infection. Although the P. falciparum and P. berghei proteins share 42.6% similarity (apart from an 80-amino-acid stretch in the center of the protein), a P. berghei mutant with P. falciparum S14 replacing the endogenous P. berghei S14 gene suffered strongly reduced salivary gland invasion (see https://www.pberghei.eu/index.php?rmgm=4742). Here, we found that the inability of S14 KO sporozoites to invade the host cell was due to impaired gliding motility. Overall, these results indicate that S14 is essential for gliding motility and invasion of Plasmodium sporozoites.

We propose that S14 works with other glideosome-associated proteins and facilitates gliding motility. As well as S14, several other parasite proteins play a role in the gliding motility and invasion of both mosquito and mammalian hosts, such as surface protein TRAP (Sultan et al., 1997), claudin-like apicomplexan microneme protein (CLAMP) (Loubens et al., 2023) and TREP (Steinbuechel and Matuschewski, 2009). S14 is not a surface protein with an extracellular domain, and its host cell invasion defect was due to impaired gliding. Like S14, limp mutant sporozoites were found to be impaired in gliding motility and host cell invasion (Santos et al., 2017). However, several sporozoite proteins, such as SPECT, SPECT2 and PLP1, differ from S14, as sporozoite mutant for these they invade the host cell but have a deficiency in cell traversal (Ishino et al., 2004, 2005b; Risco-Castillo et al., 2015). Sporozoites with mutant surface proteins, such as P36 and P52, show normal gliding motility and cell traversal, suggesting that these proteins play a role in host cell invasion by interacting with host cell receptors (Arredondo et al., 2018; Ishino et al., 2005a; Manzoni et al., 2017).

The parasite gliding machinery consists of the atypical myosin MyoA, MTIP and ELC, the glideosome-associated proteins GAP40 and GAP45, and the transmembrane protein GAP50 (Bergman et al., 2003; Frénal et al., 2010; Pazicky et al., 2020). MyoA interacts with F-actin, which connects with surface proteins through aldolase (Huynh and Carruthers, 2006; Jewett and Sibley, 2003; Sultan et al., 1997). In T. gondii, GAP45 plays a vital role in maintaining the close association of the IMC to the plasma membrane (Frénal et al., 2010). GAP45 also interacts with GAP50 through its C-terminal region, supporting its function as the anchor of the motor complex in the IMC.

Deletion of S14 resulted in the accumulation of a higher number of sporozoites in hemolymph, indicating the dispensable role of S14 during the egress of sporozoites from oocysts. The conditional deletion of GAP45 during P. falciparum asexual blood stages has revealed its role in the invasion but not in egress (Perrin et al., 2018). These results indicate that a functional motor complex is not required for egress from red blood cells (RBCs), which plays a crucial role in invasion (Perrin et al., 2018). GAP40 and GAP50 and members of the GAPM family play important roles in the biogenesis of IMCs during intracellular replication of T. gondii. Parasites lacking GAP40 or GAP50 start replication but fail to complete it, implying that they have a structural role in maintaining the stability of the developing IMC during the asexual life cycle of T. gondii replication (Harding et al., 2016). It has been shown that IMC is crucial for the anchorage and stabilization of the glideosome (Opitz and Soldati, 2002) and is required during the invasion of the host cell (Bargieri et al., 2013; Egarter et al., 2014; Meissner et al., 2013; Togbe et al., 2008). S14 deletion does not affect GAP45, MTIP, CSP and TRAP expression and localization, suggesting that it performs motor-related functions only. These results indicate that the S14 is an IMC-associated protein in sporozoites and is essential for sporozoite gliding motility.

P. berghei ANKA (MRA 311) and P. berghei ANKA GFP (MRA 867 507 m6cl1) were obtained from BEI resources, USA. Anopheles stephensi mosquitoes were reared at 28°C and 80% relative humidity and kept under a 12-h-light–12-h-dark cycle as previously described (Gupta et al., 2020). Female Swiss albino and C57BL/6 mice (6–8 weeks old) were used for parasite infections. All animal procedures were approved by the Institutional Animal Ethics Committee at CSIR-Central Drug Research Institute, India (IAEC/2013/83 and IAEC/2023/15). Human liver hepatocellular carcinoma (HepG2) cells (ATCC) were regularly maintained in DMEM (Sigma, USA) supplemented with 10% FBS (Sigma, USA), 0.2% NaHCO₃ (Sigma, USA), 1% sodium pyruvate (Genetix, India), and 1% penicillin-streptomycin (Invitrogen, USA) at 37°C with 5% CO₂. We routinely tested cell lines for mycoplasma contamination.

The P. berghei S14 (PBANKA_0605900) amino acid sequence was retrieved from PlasmoDB (https://plasmodb.org/plasmo/app). The NCBI Protein Blast search tool was used to find similarities with other apicomplexan parasites (https://blast.ncbi.nlm.nih.gov/Blast.cgi). To identify conserved domains in S14, InterPro (https://ebi.ac.uk/interpro/), SMART (http://smart.embl-heidelberg.de/) and CD-Search (https://ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) were used. OrthoMCL (https://orthomcl.org/orthomcl/app) was used to identify ortholog groups. Sequence alignment and sequence similarity matrix were generated using UniProt align (https://www.uniprot.org/align). The presence of the signal peptide and transmembrane domain was predicted using SignalP (https://services.healthtech.dtu.dk/services/SignalP-6.0) and TMHMM (https://services.healthtech.dtu.dk/services/TMHMM-2.0/) (Sonnhammer et al., 1998).

For the absolute quantification of S14 transcripts, a standard was generated by amplifying a 120 bp fragment within the S14 ORF (PBANKA_0605900) using primers 1001 and 1002 (sequences are given in Table S1). The amplified product was cloned into the pCR 2.1-TOPO vector. For the normalization of transcripts, Hsp70 was used (Choudhary et al., 2019). Total RNA was isolated from blood stage schizonts (Schz), liver stages (LS), midgut (MG), and salivary gland (SG) sporozoites using TRIzol reagent (Takara Bio, Japan) and an RNA isolation kit (Genetix, India) following the manufacturers' instructions. cDNA was prepared by reverse transcription using a Superscript cDNA synthesis kit. Real-time PCR was carried out using SYBR green reagent (Takara Bio, Japan), and the ratio of transcript numbers of S14 and Hsp70 was used to determine the copy number.

For the endogenous tagging of S14 with 3×HA–mCherry, two fragments, F1 (0.74 kb) and F2 (0.65 kb), were amplified using primers 1010 and 1011, and 1005 and 1006 (Table S1) and cloned into plasmid pBC-3XHA-mCherry-hDHFR (Kota Arun Kumar, School of Life Sciences, University of Hyderabad Hyderabad, India) at the XhoI/BglII and NotI/AscI sites, respectively. The plasmid was linearized using XhoI/AscI and transfected into P. berghei ANKA schizonts as previously described (Janse et al., 2006). Correct 5′ and 3′ site-specific integration was confirmed by diagnostic PCR using primers 1007 and 1392, and 1215 and 1008, respectively (Table S1). The clonal lines were obtained by limiting dilution of the parasites and analyzed for expression and localization.

P. berghei S14 (PBANKA_0605900) was disrupted by double-crossover homologous recombination. For this, two fragments, F3 (0.73 kb) and F4 (0.637 kb), were amplified using P. berghei ANKA genomic DNA with primers 1003 and 1004, and 1005 and 1006, respectively (Table S1). The fragments F3 and F4 were cloned sequentially into the pBC-GFP-hDHFR vector (Robert Menard, Institut Pasteur, Paris, France) at the XhoI/ClaI and NotI/AscI sites, respectively, and finally linearized with XhoI/AscI and transfected into P. berghei schizonts (Janse et al., 2006). The drug-resistant GFP-expressing parasites were confirmed for 5′ and 3′ site-specific integrations using primers 1007 and 1225, and 1215 and 1008, respectively (Table S1). To generate an S14 KO complemented parasite line, another S14 KO parasite line was generated with the hDHFR:yFCU selection cassette (generated in house). A fragment consisting of the S14 5′UTR, ORF and 3′UTR was amplified using primers 1003 and 1006 (Table S1) and transfected into S14 KO parasites. Parasites containing restored S14 loci were selected by negative selection using a 5-fluorocytosine (5-FC) drug (Sigma, USA) as previously described (Srivastava and Mishra, 2022). The clonal lines were obtained by limiting dilution of the parasites, and the absence of S14 ORF was confirmed using primers 1010 and 1011 (Table S1). Furthermore, two clonal lines were also confirmed by Southern blotting as described previously (Narwal et al., 2022). Fragment F3 was used as a probe to detect the band in a Southern blot.

The S14 KO clonal lines were first analyzed for asexual blood-stage propagation, and for this, 200 µl of infected RBCs (iRBCs) with 0.2% parasitemia was intravenously injected into a group of mice. Parasitemia was monitored daily by Giemsa staining of blood smears. Next, we initiated infection with KO parasites in Anopheles stephensi mosquitoes as previously described (Narwal et al., 2022). On days 14 and 19, midgut and salivary glands were observed for infection, and sporozoite numbers were counted. The hemolymph sporozoites were collected and counted as previously described (Mastan et al., 2017).

To determine the in vivo infectivity of KO sporozoites, C57BL/6 mice were either infected by mosquito bite or by intravenously injecting hemolymph sporozoites. Ten mosquitoes per mouse were used for the bite experiment and allowed to probe for the blood meal for 20 min. The appearance of parasites in the blood was observed by making Giemsa-stained blood smears.

The in vitro infectivity of sporozoites was tested by infecting HepG2 cells as previously described (Narwal et al., 2022). A total of 50,000 cells/well were seeded in 48-well plates containing sterilized coverslips pretreated with collagen. Hemolymph sporozoites (10,000 sporozoites/well for the invasion assay or 5000 sporozoites/well for EEF development) were added to the HepG2 monolayers, and the plate was centrifuged at 310 g for 4 min and incubated at 37°C in a CO₂ incubator. The culture was fixed using 4% paraformaldehyde (PFA) at 1 hpi for the invasion assay and 40 hpi for the EEF development assay.

WT GFP and S14 KO hemolymph sporozoites were incubated in a DMEM containing with 2 mM L-glutamine and 4.5 g/l glucose, supplemented with 10% FBS (Sigma, USA), 500 U/ml penicillin-streptomycin (Thermo Fisher Scientific, USA) and 1.25 μl/ml fungizone as previously described (Kaiser et al., 2003). The sporozoites were incubated at 37°C in a CO₂ incubator for 4 h and then fixed using 4% PFA.

Affinity-purified polyclonal rabbit antibodies against P. berghei MTIP and GAP45 were developed by GenScript Inc., Piscataway, USA, against the peptide sequences CVNKDDRKIYFDEKS and CHKYENDSDKLETGS, respectively.

Sporozoites (3×10⁴) were collected and treated with 1.0% Triton X-100 diluted in PBS and incubated on ice for 30 min as previously described (Bergman et al., 2003). After incubation, sporozoites were spun at 13,800 g for 20 min at 4°C. Both treated and untreated sporozoites were washed three times with PBS and fixed with 2% PFA diluted in PBS or resuspended in sample buffer for western blotting.

To quantify sporozoite gliding motility, a glass eight-well chamber slide was coated with 10 μg/ml anti-CSP monoclonal antibody (3D11; Victor Nussenzweig, Department of Pathology, New York University, USA) in PBS overnight, and the assay was performed as described previously (Stewart and Vanderberg, 1988). Hemolymph or salivary gland sporozoites collected in 3% BSA in DMEM were added at 5000 per well and incubated for 1 h at 37°C in a CO₂ incubator. After incubation, sporozoites were fixed with 4% PFA, blocked using 3% BSA in PBS, and incubated with biotin-labeled anti-CSP antibody (Aastha Varshney, Division of Molecular Microbiology and Immunology, CSIR-Central Drug Research Institute, India, dilution, 1:100), and signals were revealed using streptavidin–FITC (dilution, 1:500; Invitrogen, USA). To visualize the S14-associated mCherry trails, the glass slide was coated with a 10 μg/ml anti-mCherry antibody developed in rabbit (cat. no. NBP2-25157, Novus Biologicals, USA). Signals were revealed using biotin-labeled anti-mCherry antibody (as above, dilution, 1:100). Trails associated with sporozoites were counted using a Nikon 80i fluorescence microscope. For live imaging, sporozoites were dissected in DMEM containing 3% BSA and purified by centrifuging at 60 g for 4 min. The sporozoites were imaged using a Nikon Eclipse 80i microscope and movement patterns were analyzed as previously described (Srivastava et al., 2024; Wichers-Misterek et al., 2023).

Fixed sporozoites were washed with PBS, permeabilized using 0.1% Triton X-100 for 15 min at room temperature, and then blocked with 1% BSA in PBS for 1 h at room temperature. Then, sporozoites were incubated with anti-mCherry antibody developed in rabbit (as above, 1:500), anti-CSP mouse monoclonal (3D11; dilution, 1 µg/ml; Yoshida et al., 1980), anti-MTIP (GenScript; dilution, 1 µg/ml), anti-GAP45 (GenScript; dilution, 1 µg/ml) or anti-TRAP (dilution, 1:200; Mishra et al., 2023 preprint) antibodies. The signals were revealed using Alexa Fluor 594-conjugated or Alexa Fluor 488-conjugated antibodies (dilution, 1:1000; Invitrogen, USA). For the staining of blood stage parasites, a blood smear was incubated with anti-mCherry developed in rabbit (as above; Novus Biologicals, USA) and anti-Hsp70 mouse monoclonal (4C9; dilution, 1 µg/ml; Tsuji et al., 1994) antibodies. For the staining of EEFs, fixed cultures were washed with PBS, and staining with anti-UIS4 (dilution, 1:1000; Mueller et al., 2005) or anti-CSP (dilution, 1 μg/ml; Yoshida et al., 1980) antibodies was performed as previously described (Narwal et al., 2022). Nuclei were stained with Hoechst 33342 (Sigma-Aldrich, USA) and mounted using Prolong diamond anti-fade reagent (Invitrogen, USA). The images were acquired using a confocal laser scanning microscope with a UPlanSAPO 100×/1.4 oil immersion objective (Olympus BX61WI) at an additonal 4× digital magnification. The fluorescence intensity was quantified using NIH ImageJ (Schindelin et al., 2012).

To analyze the secretion of sporozoite protein in the medium, sporozoites were incubated in complete medium (DMEM with 10% FBS) at 37°C for 35 min, as previously described (Srivastava et al., 2024). The samples were centrifuged at 18,400 g, the supernatant and pellets were separated, and Laemmli buffer was added for western blotting.

Sporozoites were pelleted by centrifuging at 18,400 g for 4 min and resuspended in Laemmli buffer. Immunoblotting was performed as previously described (Narwal et al., 2022). Briefly, samples were resolved by 10% SDS-PAGE, transferred to a nitrocellulose membrane (Bio-Rad, USA), and blocked with 1% BSA. The blot was incubated with an anti-mCherry (cat. no. NBP2-25157, Novus Biologicals, USA; dilution, 1:1000,) or anti-CSP (dilution, 1 μg/ml; Yoshida et al., 1980) antibodies. The membrane was washed and incubated with HRP-conjugated anti-rabbit or anti-mouse IgG (dilution, 1:5000, Amersham Biosciences, UK). The signals were detected using ECL chemiluminescent substrate (Thermo Scientific, USA) in a ChemiDoc XRS+ System (Bio-Rad, USA). Full uncropped immunoblot images can be found in Figs S8 and S9.

GraphPad Prism 9 software was used to generate graphs and to calculate P-values (two-tailed unpaired Student's t-test or ordinary one-way). P-values are denoted as *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, n.s.; not significant.

We thank Dr Pratik Narain Srivastava for performing initial bioinformatic studies. We thank BEI Resources, USA, for the parasite strains and plasmids and Dr Kota Arun Kumar (University of Hyderabad, India) for the pBC-3XHA-mCherry-hDHFR vector. We thank Dr Photini Sinnis (Johns Hopkins University, USA) for the anti-UIS4 antibody. We acknowledge the THUNDER (BSC0102) and MOES (GAP0118) Intravital and Confocal microscopy facility of CSIR-CDRI. The Council of Scientific and Industrial Research, University Grants Commission, and Indian Council of Medical Research, Government of India research fellowships supported A.G., A.V., N., R.G. and S.K.N. This manuscript is CDRI communication no. 10786.