Plasmodium Alp5a and Alp5b are structurally similar to human Arp2 and Arp3, localize to the nucleus, and interact with each other

We started our study by in silico analysis of the paralogous genes Alp5a (PBANKA_0811800) and Alp5b (PBANKA_1007500). NCBI BLASTP analysis of P. berghei Alp5s revealed that Alp5a and Alp5b are closely associated with P. yoelii and P. falciparum, respectively (S1 Fig). The protein-protein interactions of both Alp5a and Alp5b were confirmed via the STRING database. An interactome of 11 proteins revealed that all the protein interactions were the same for Alp5a and Alp5b, except for PBANKA_1434600 (formin2, putative protein) in Alp5a (S2 Fig). All interacting proteins were classified as conventional or actin-like proteins, dynactin subunits, formins, or subunits of the Arp2/3 complex. Notably, the presence of capping protein subunits in the interactome further highlights the diversity of actin-associated regulators identified (S2 Fig). The predicted protein structures of Alp5a and Alp5b were generated using AlphaFold2 and evaluated with UCLA-SAVES. The highest-quality models were subsequently selected for further analysis. (Fig 1A and 1B). Alp5a and Alp5b were superposed with the human Arp3 and Arp2 structures via PyMOL. We observed the superposition of Alp5a and Alp5b with human Arp3 and Arp2, with RMSD values of 2.961 Å and 1.328 Å, respectively (Fig 1C and 1D). The alignment at residue Phe425 of ALP5a (atom no. 2249) and residue 282 of Arp3 (atom no. 6856) highlights a key region with an RMSD of 2.961 Å, indicating potential functional relevance (S3 Fig). To investigate the potential interaction between Alp5a-Alp5b and Alp5-actin, protein–protein docking was performed using the ClusPro platform, and the resulting complexes were visualized using PyMOL. Interaction scores between Alp5a and Alp5b suggest a potential protein–protein interaction (Fig 1E). To assess the interaction of Alp5 with actin, we generated a 3D model of actin isoforms (actin 1 and actin 2) (S4A and S4D Fig). Cluspro protein-protein docking studies indicated that Alp5a has a stronger affinity for both actin isoforms than Alp5b (S4B, S4C, S4E and S4F Fig). To further validate these findings, we conducted additional computational analyses using experimentally determined high-resolution crystal structures of P. falciparum actin 1 (PDB ID: 6I4H; 1.40 Å resolution) and P. berghei actin 2 (PDB ID: 6I4M; 1.87 Å resolution), obtained from the RCSB Protein Data Bank. To assess the accuracy of AlphaFold2-predicted models relative to these experimental structures, structural alignments and superpositions were carried out. Root-mean-square deviation (RMSD) values were calculated to quantify structural similarity. The AlphaFold2-predicted structure of P. berghei actin 1 exhibited strong concordance with the P. falciparum actin 1 crystal structure (6I4H), with an RMSD of 0.627 Å. Similarly, the AlphaFold2 model of P. berghei actin 2 closely aligned with its corresponding crystal structure (6I4M), with an RMSD of 0.796 Å (S5 Fig). These low RMSD values indicate a high degree of structural similarity and support the reliability of the AlphaFold2 predictions. Further docking studies using experimentally resolved actin structures revealed interaction patterns and docking scores comparable to those observed with AlphaFold2-predicted models (Fig 1F–1I). Subsequent docking of the Alp5a–Alp5b complex with F-actin suggested that the C-terminus of Alp5b may mediate the interaction with F-actin. In parallel, docking analyses with the Arp2/3 complex indicated that F-actin primarily interacts with the Arp3 subunit (Fig 1J–1K). To assess overall structural similarity, the Alp5a–Alp5b complex was individually aligned with Arp2/3 complex subunits. Structural superposition of Alp5a with Arp3 yielded a root-mean-square deviation (RMSD) of 2.905 Å, whereas alignment of Alp5b with Arp2 produced an RMSD of 1.899 Å (S6 Fig). These relatively low RMSD values indicate a notable degree of structural similarity, supporting the notion that the Alp5a–Alp5b complex may adopt a conformation resembling that of the Arp2/3 complex. The AlphaFold3 predicted template modeling (pTM) score and interface predicted template modeling (ipTM) score are summarized in S2 Table.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Plasmodium Alp5s are structurally conserved actin-related proteins (Arps).

(A) The cartoon structure of P. berghei Alp5a was visualized using PyMOL, with α-helices, β-sheets, and loop regions depicted to highlight the protein’s overall fold and key secondary structural features. (B) Cartoon structure of Alp5b. (C) The structural superposition of P. berghei Alp5a (green) with human Arp3 (chocolate) reveals a RMSD of 2.961 Å. This RMSD value quantifies the degree of structural similarity between the two proteins, indicating both conserved regions and conformational differences. (D) Structural superposition of P. berghei Alp5b (magenta) and human Arp2 (brown) reveals a RMSD of 1.328 Å, indicating a high degree of structural conservation. This similarity suggests that key functional domains are maintained across species, despite evolutionary divergence. (E) The interaction between P. berghei Alp5a and 5b, as predicted by ClusPro. The docking model illustrates the predicted binding interface and interaction sites between the two proteins, highlighting their potential functional relationships. (F) Predicted interaction between P. falciparum Actin I (PfActin I; PDB ID: 6I4H, pink) and P. berghei Alp5a (green). (G) Interaction between PfActin I (6I4H; pink) and PbAlp5b (magenta). (H) Modeled complex of PbActin II (PbActin II; PDB ID: 6I4M, salmon red) with PbAlp5a (green). (I) Predicted interaction between PbActin II (6I4M, salmon red) and PbAlp5b (magenta). (J) Predicted ternary complex of PbAlp5a (green) and PbAlp5b (magenta) interacting with F-actin (orange), illustrating their potential binding interfaces and relative orientations. (K) Predicted ternary complex of Arp2 (yellow) and Arp3 (sky blue) bound to F-actin (orange), representing a canonical actin-branching complex.

https://doi.org/10.1371/journal.ppat.1013687.g001

Next, to check the expression, localization, and interactions of Alp5, we endogenously tagged Alp5a and Alp5b with 3XHA and 3XHA-mCherry, respectively (S7A, S7B, S7D, and S7E Fig). Western blot analysis with an anti-HA antibody confirmed the expression of correctly sized Alp5 fusion proteins in the gametocytes of the transgenic parasites (S7C and S7F Fig). IFA revealed the expression of Alp5a and Alp5b in male and female gametocytes but not in the ring or trophozoite stages. Male and female gametocytes were immunostained with anti-tubulin and anti-g377 antibodies to confirm gamete-specific expression. Both Alp5a and Alp5b were expressed predominantly in the nuclei of gametocytes (Fig 2A and 2B). The signals of Alp5a and Alp5b localize from the nucleus to a microtubule-rich structure, likely the spindle or axoneme (Fig 2C and 2D). Diffuse signals were also observed in the parasite cytoplasm. The expression of Alp5a and Alp5b was also detected in zygotes, ookinetes, and oocysts (Fig 2E-2H). The expression of Alp5a and Alp5b was not detected in sporozoites or exo-erythrocytic forms (S8A and S8B Fig). To assess the colocalization of Alp5 and actin, nonactivated and activated gametocytes were immunostained with anti-HA and anti-actin antibodies. The colocalization of Alp5 and actin in activated gametocytes indicated their possible interaction (Fig 2I and 2J). To determine the endogenous interaction between Alp5a and Alp5b, the complex was immunoprecipitated from gametocytes of Alp5a-3XHA transgenic parasites using anti-HA magnetic beads. The blot was probed with an anti-HA antibody that detected the Alp5a band (Fig 2K). The blot was reprobed with an anti-Alp5b antibody, which detected the Alp5b band, indicating the interactions of Alp5a with Alp5b (Fig 2L). These results suggest that Alp5 is predominantly expressed in the sexual stages and that the Alp5a and Alp5b subunits interact with each other.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Expression and localization of Alp5s and their interaction.

(A and B) Representative confocal immunofluorescence images of male and female gametocytes immunostained with anti-HA, anti-tubulin and anti-g377 antibodies. The Alp5a-3XHA and Alp5b-3XHA-mCherry parasites show their localization. Nuclei were stained with Hoechst. DIC: differential interference contrast. (C and D) Confocal immunofluorescence analyses of microgametes after gametocyte activation. The cells were fixed and immunostained with anti-HA and anti-tubulin antibodies. Alp5a and Alp5b localize from the nucleus to a microtubule-rich structure, likely the spindle or axoneme. (E and F) Immunofluorescence analyses of Alp5a-3XHA and Alp5b-3XHA-mCherry revealed their localization at the zygote, retort, and ookinete stages. Parasites were immunostained with anti-HA and anti-P25 antibodies. (G and H) Expression of Alp5a-3XHA and Alp5b-3XHA-mCherry parasites in oocysts. The midgut oocysts were labeled with anti-HA and anti-CSP antibodies. (I and J) Male gametocytes of Alp5 transgenic parasites were immunostained with anti-HA and anti-actin antibodies at 0- and 15-min post-activation, and subsequently imaged using a confocal microscope. At 0 minutes post-activation (mpa), Alp5 was predominantly localized to the nucleus, whereas actin was distributed in both the cytosol and the nucleus. Following activation at 15 mpa, Alp5 and actin are redistributed around the nucleus and exhibit colocalization. (K and L) Interaction of Alp5s in the parasite. The protein complex was immunoprecipitated from gametocytes of Alp5a-3XHA transgenic parasites using anti-HA magnetic beads. The Alp5a and Alp5b bands were detected by immunoblotting with anti-HA and anti-Alp5b antibodies.

https://doi.org/10.1371/journal.ppat.1013687.g002

Alp5a and Alp5b are functionally redundant, and the Arp2/3 complex formed by these proteins is possibly essential for parasite survival‌

Since Alp5a and Alp5b are paralogous genes [44], we investigated their roles together. To understand the role of Alp5a and Alp5b in the parasite life cycle, we disrupted the gene in P. berghei via double crossover homologous recombination (S9A, S9B, S9E and S9F Fig). To restore gene function in the KO parasites, a complemented line was generated by reintroducing the Alp expression cassette (S9C, S9D, S9G, S9H and S9I Fig). We attempted to disrupt both genes simultaneously to check for redundancy in the functions of Alp5a and Alp5b. We applied two independent strategies for the double disruption of genes. First, we transfected an Alp5b-targeting construct into Alp5a KO schizonts. (S10A Fig). Twenty-four hours post-transfection, we observed a few parasites with red and green fluorescence, which were lost after WR drug selection. Next, we transfected P. berghei WT schizonts with Alp5a and Alp5b targeting constructs (S10B Fig). Parasites were selected using pyrimethamine drugs. We observed parasites with individual gene knockouts, but no double-deletion events were detected (Fig 3A). We treated Plasmodium blood-stage parasites with various inhibitors to further confirm the essential roles of actin and the Arp2/3 complex. We found that an Arp2/3 complex inhibitor (CK-666), an actin polymerization inhibitor (Cytochalasin D), and an actin polymerization inducer (Jasplakinolide) inhibited the growth and development of P. falciparum and P. berghei parasites (Figs 3B, 3C, S11A and S11B).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Alp5a and Alp5b are not individually required for asexual blood stage propagation, but simultaneously, they are indispensable for parasite viability.

(A) Parasites transfected with constructs targeting Alp5a and Alp5b, tagged with mCherry and GFP respectively, predominantly gave rise to single knockout populations expressing only one of the two fluorophores. Rare instances of parasites expressing both mCherry and GFP were observed within the same red blood cell; however, the lack of fluorescence colocalization suggests that these are not true double knockouts. Representative images were acquired using a widefield fluorescence microscope. (B) Effect of inhibitors on the growth of P. falciparum. Parasitemia was measured at 24, 48, and 72 h after treatment with the indicated concentrations of cytochalasin D, jasplakinolide, and CK-666. At 24 h post-treatment, no significant difference in parasitemia was observed between DMSO-treated controls and cultures treated with either cytochalasin D or jasplakinolide (P = 0.2631; one-way ANOVA). Inhibition was observed in the CK-666-treated cultures, DMSO vs CK-666, ***P = 0.0008; unpaired Student’s t test. A reduction in growth was observed at 48 h, DMSO vs cytochalasin D or jasplankinolide (500 nm-1 µM) ****P < 0.0001; one-way ANOVA, DMSO vs CK-666, **P = 0.0021; unpaired Student’s t test. The growth further decreased at 72 h. DMSO vs cytochalasin D or jasplankinolide (500 nm-1 µM), ****P < 0.0001 and DMSO vs CK-666 ***P = 0.0001; unpaired Student’s t test. Compared with DMSO, 100 nM jasplankinolide did not affect parasite growth at 48 h (P = 0.6447) or 72 h (P = 0.5590). The data are presented as the mean ± SEM and were analyzed via one-way ANOVA or unpaired Student’s t tests; n = 3 biological replicates. (C) P. berghei blood cultures were treated with the inhibitors cytochalasin D (1 µM), jasplankinolide (250 nM) and CK-666 (300 µM). Counting of ring, trophozoite, and schizont stages in P. berghei cultures after treatment. There was no effect on the ring (cytochalasin D, P = 0.1605; jasplankinolide, P = 0.6896; and CK-666, P = 0.0693) or trophozoite stage (cytochalasin D, P = 0.8053; jasplankinolide, P = 0.1031; and CK-666 (P = 0.4832). The number of schizonts decreased significantly in all the treated groups (cytochalasin D, *P = 0.0257; jasplankinolide, **P = 0.0082; and CK-666, **P = 0.0012). The data are presented as the mean ± SEM and were analyzed via unpaired Student’s t tests; n = 3 biological replicates. (D) Asexual blood-stage propagation of control and KO parasites. The data are presented as the mean ± SEM, with no significant differences (P = 0.9998; Brown-Forsythe ANOVA). (E) Estimation of gametocytemia. The data are presented as the mean ± SEM with no significant differences (P = 0.7776; Brown-Forsythe ANOVA). (F) The male and female gametocyte ratios were comparable in WT and KO parasites. The data are presented as the mean ± SEM, with no significant differences (P = 0.9864; one-way ANOVA). Data are presented from two biological replicates, each with 5 mice per group in D, E, and F. (G) Western blot analysis of F-actin and G-actin in blood-stage parasites. Parasites were homogenized in F-actin stabilization buffer, followed by centrifugation to separate F-actin from the G-actin pool. Immunoblotting was performed with an anti-actin antibody to detect actin. Western blot quantitation of F-actin and G-actin in blood-stage parasites and data are presented as the F/G actin ratio. There was no difference between WT and Alp5 KO parasites (P = 0.5903; one-way ANOVA). (H) Comparison of exflagellation centers in WT and Alp5 KO parasites revealed no difference (P = 0.1274, Brown Forsythe and Welch ANOVA). Counts were checked from 10 random fields from three biological replicates. The data are presented as the mean ± SEM.

https://doi.org/10.1371/journal.ppat.1013687.g003

For phenotypic analysis of the Alp5a and Alp5b KO parasites, we assessed asexual blood-stage propagation. Both KO lines exhibited normal asexual replication and produced gametocytes at rates comparable to WT parasites (Fig 3D and 3E). Next, we examined the proportions of male and female gametocytes and found that they were comparable in WT and KO parasites (Fig 3F). We also analyzed the F-actin/G-actin ratio in the KO parasites after separating them by ultracentrifugation, which was comparable to that in the WT parasites (Fig 3G). Next, we analyzed microgamete formation. There was no significant difference in the number of exflagellation centers between KO and WT parasites (Fig 3H). These results indicate that the Alps are not individually required for actin nucleation or blood-stage development but are essential as a complex for parasite survival.

Alp5s are essential for microgamete fertility, oocyst development, and malaria transmission

To analyze the development of Alp5 KO parasites in mosquitoes, we allowed mosquitoes to feed on the infected mice to facilitate transmission. We observed oocyst development on day 14 post-bloodmeal and found a significant reduction in oocyst size in both types of KO parasites (Figs 4A and S12A). Next, we analyzed the size and number of oocysts from days 8–18 post-feeding. The size and number of oocysts were significantly reduced in KO parasites. Interestingly, the number of oocysts decreased over time, with only a few remaining by day 18 (Figs 4B, 4C, S12B and S12C). Additionally, oocyst size did not increase, and sporogony was abolished in the KO parasites (Fig 4D). We crushed the midgut and salivary glands to estimate the sporozoite numbers. The sporozoites were completely absent in the Alp5 KO parasites (Fig 4E, 4F and 4G). All the KO-infected mosquitoes failed to transmit malaria to the mice in the biteback experiment (Fig 4H and Table 1). The complemented parasite lines successfully completed all developmental stages, confirming that the observed phenotype resulted from the deletion of the Alp5 gene (Fig 4). These results demonstrate that Alp5 is essential for oocyst development and sporogony, and that its deletion blocks malaria transmission.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. In vivo infectivity of Alp5 KO parasites in C57BL/6 mice. Sporozoites were inoculated into C57BL/6 mice via mosquito bites or intravenous injection. The presence of parasites in the blood was confirmed via a Giemsa-stained blood smear.

https://doi.org/10.1371/journal.ppat.1013687.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Alp5s are essential for oocyst development and malaria transmission.

(A) Representative widefield fluorescence microscopy images of mosquito midguts showing oocysts on the indicated days post-infection. (B) Quantification of oocyst numbers. There was a significant difference in oocyst number between WT and KO parasites on days 8 and 14 (****P < 0.0001). In contrast, no differences were detected between the complement lines and WT on days 8 (P = 0.8721) and 14 (P = 0.1075). Sixty midguts from all the groups were dissected each day. (C) Determination of the oocyst area. The data from independent clones were pooled, and a significant difference was detected between the WT and KO lines on days 8 and 14 (****P < 0.0001), whereas no differences were detected between the complemented lines and the WT line on days 8 (P = 0.1058) and 14 (P = 0.8983). The Kruskal‒Wallis test was used to determine the significance. A total of 200 oocyst areas were measured for each of the WT, Alp5a KO, and Alp5b KO lines, and 40 oocyst areas for each of the corresponding complemented lines, on the indicated days. Data represent three independent biological replicates. The triangle represents the mean individual count in each experiment, and each dot represents an individual count. (D) Widefield fluorescence microscopy of WT oocysts undergoing sporogony. No sporogony was observed in Alp5a or Alp5b KO oocysts. (E) Midgut sporozoite count. The data revealed a significant difference between the WT and knockout lines (****P < 0.0001; one-way ANOVA). No difference was observed between the WT and complemented lines (P = 0.7737). The total number of mosquitoes dissected in each group was 95 (WT GFP), 235 (Alp5b c1), 235 (Alp5b c2), 95 (Alp5b comp1), 70 (Alp5b comp2), 260 (Alp5a c1), 245 (Alp5a c2), 80 (Alp5a comp1), and 85 (Alp5a comp2). Data are presented as the mean ± SEM from three independent biological replicates. (F) Widefield fluorescence microscopy reveals sporozoite-specific fluorescence in WT but not KO-infected salivary glands. (G) Salivary gland sporozoite count. The total number of mosquitoes dissected in each group was 85 (WT GFP), 118 (Alp5b c1), 170 (Alp5b c2), 70 (Alp5b comp1), 100 (Alp5b c2), 135 (Alp5a c1), 235 (Alp5a c2), 70 (Alp5a comp1), and 70 (Alp5a comp2). The data revealed a significant difference between the WT and knockout lines (****P < 0.0001; one-way ANOVA). No difference was detected between the WT and complemented lines (P = 0.5317; one-way ANOVA). Data are presented as the mean ± SEM from three independent biological replicates. (H) Transmission of sporozoites to mice via mosquito bites (n = 5 mice per group) was assessed. Infection was not detected in mice bitten by mosquitoes infected with the KO parasites. The data represent pooled results from two independent biological replicates.

https://doi.org/10.1371/journal.ppat.1013687.g004

After the essential roles of Alp5a and Alp5b in oocyst development and malaria transmission were established, the formation of ookinetes was subsequently analyzed. We found no difference in ookinete number or morphology between WT and KO parasites (S13A and S13B Fig). A genetic cross was performed between the KO and WT parasites to analyze sex-specific defects in the KO parasites. On day 15, post-feeding, we dissected the midgut, and the development of oocysts was observed. Genetic crosses with WT parasites restored the KO parasite’s phenotype (Fig 5A), indicating a defect at the gamete stage. Plasmodium inherits many defects during mosquito stage development from one sex [11]. To investigate whether the functions of Alp5a and Alp5b are sex specific, we generated male and female gamete defective lines by knocking out the Pb48/45 [54] and Pb47 genes [55], respectively (S14 Fig). Next, we crossed KO parasites with gamete defective lines (Pb47 × Alp5b KO, Pb48/45 × Alp5b KO, Pb47 × Alp5a KO, Pb48/45 × Alp5a KO). Crossing Alp5 KO with Pb47 KO parasites resulted in robust oocyst formation, indicating that Alp5 KO parasites produce fertile female gametes (Figs 5B and S15). In contrast, the Alp5 KO × Pb48/45 KO cross showed no change in oocyst development, indicating defects in microgamete function in Alp5 KO parasites (Figs 5B and S15). Following genetic cross-experiments, oocysts were quantitatively assessed by counting and measuring their surface area, and sporogony progression was monitored. Sporogony of the restored oocysts proceeded normally (Fig 5C and 5D). On day 19 post-blood feeding, dissection of mosquito salivary glands revealed normal sporozoite loads in all experimental groups except the Alp5-KO lines crossed with a male gamete-defective parasite line (S16A and S16B Fig). The obtained salivary gland sporozoites were checked for infectivity in vivo and in vitro. Sporozoites were intravenously injected into C57BL/6 mice, and the onset of the pre-patent period was monitored starting from day three post-infection (Table 1). Groups with restored salivary gland sporozoite loads successfully infected hepatocytes and progressed normally to the blood stage (S17 Fig). Together, these results demonstrated that Alp5s play male-specific roles in gamete fertility.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Genetic cross revealed a defect in the male gamete in Alp5 KO parasites.

(A) Widefield fluorescence microscopy of oocysts at day 15 post-feeding following genetic crosses between WT-mCherry×Alp5b KO or WT-GFP × Alp5a KO parasites. (B) Genetic crosses of Alp5 KO parasites with male or female gamete-defective lines. The phenotype was restored in the KO lines crossed with the female gamete-defective lines. Genetic crosses with the male gamete-defective lines did not restore the KO phenotype. (C) Quantification of oocysts on day 15. Sixty-two midguts from each group were dissected and observed. Significant differences were observed in WT vs WT-mCherry×Alp5b KO (**P = 0.0038), WT vs WT-GFP × Alp5a KO (**P = 0.0019), WT vs Pb47 KO (****P < 0.0001), WT vs Pb48/45 KO (****P < 0.0001), WT vs Pb47 KO × Alp5b KO (**P = 0.0071), WT vs Pb48/45 KO × Alp5b KO (****P < 0.0001), WT vs Pb47 KO × Alp5a KO (**P = 0.0050), and WT vs Pb48/45 KO × Alp5a KO (****P < 0.0001). (D) Oocyst area on day 15 post-infection. A total of 210 oocyst areas were quantified in each group. Significant differences were observed in WT vs WT-mCherry×Alp5b KO (*P = 0.0377), WT vs WT-GFP × Alp5a KO (*P = 0.0203), WT vs Pb47 KO × Alp5b KO (*P = 0.0334), WT vs Pb48/45 KO × Alp5b KO (****P < 0.0001), WT vs Pb47 KO × Alp5a KO (*P = 0.0435), and WT vs Pb48/45 KO × Alp5a KO (****P < 0.0001). Unpaired Student’s t test was used to determine the statistical significance. The data were pooled from three independent biological replicates and are presented as the mean ± SEM in panels C and D.

https://doi.org/10.1371/journal.ppat.1013687.g005

Alp5s regulate DNA segregation during male gametogenesis

The male gametocyte undergoes three rapid rounds of DNA division during male gametogony, reaching an 8 N state. After division, this DNA distributed to the eight flagellar microgametes coincided with axoneme formation. To determine the defect in male gametogony, activated gametocytes were immunostained with an anti-tubulin antibody to identify activated microgametes and free microgamete. On the same slide, a single nucleated cell exhibiting a diffuse signal was identified as a female gametocyte (macrogamete). Despite no observable differences in the number of activated male and female gametes, we identified distinct patterns of DNA allocation within activated gametocytes. The DNA distribution was homogenous, flagellar and condensed in the WT parasites, with enlarged activated gametocyte nuclei. However, the DNA was residual and irregularly distributed in the Alp5 KO parasites (Fig 6A). The percentage of irregularly distributed Alp5a KO and Alp5b KO cells was greater than that of WT cells (Fig 6B). We found no difference in the DNA content of activated male gametocytes (enlarged nuclei coinciding with axonemes) or female gametocytes/gametes (single nuclei with diffuse tubulin signals). The microgametes from the KO showed a significant reduction in DNA content—approximately 50% less compared to the wild type WT. (Fig 6C and 6D). The DNA was quantified via the corrected total cell fluorescence (CTCF) method via ImageJ software. The reduced DNA content in the microgametes of Alp5 KO parasites did not affect the morphology of zygotes or ookinetes. Next, we checked the DNA content in zygotes and ookinetes, which was significantly reduced in Alp5 KO parasites (Figs 6E, 7A and 7B). The reduced DNA content in the ookinete did not affect its gliding ability (S13C Fig). Next, we analyzed the DNA within the oocysts at various time points using Hoechst staining. The stained oocysts were then imaged, and their DNA content was quantified using Fiji software. The DNA content in the KO oocysts was significantly reduced, indicating impaired development. In contrast, the DNA content in the WT and complemented lines was comparable (Fig 7C and 7D). The DNA content of oocysts derived from crosses between KO lines and female gamete-defective lines was comparable to that observed in WT. In contrast, crosses between KO lines and male gamete-defective lines showed no significant change in oocyst DNA content (Fig 7E and 7F). These results suggest that Alp5 proteins are essential for DNA segregation during male gametogenesis. Although Alp5 KO parasites can undergo meiosis and ookinete formation despite having reduced DNA content, they are arrested at the oocyst stage.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Impaired DNA segregation in Alp5 KO parasites.

(A) Gametocytes were activated and fixed at 15 mpa (minutes post-activation). DNA was stained with Hoechst, and an anti-tubulin antibody was used to identify activated male gametocytes. Images were acquired using a confocal microscope. (B) Different forms of microgametes were identified based on quantification of DNA localization patterns in activated male gametocytes. A total of 167 (WT), 192 (Alp5b KO), and 295 (Alp5a KO) male gametocytes were analyzed from three biological replicates. (C) Confocal IFA images of female gametocytes, activated male gametocytes and microgametes. Activated male gametocytes and microgametes were identified by anti-tubulin antibody staining. The female gametocytes were identified via diffuse tubulin staining. DNA was stained with Hoechst. (D) We quantified the DNA content of activated male and female gametocytes as well as microgametes, normalizing all measurements to the mean DNA content of female gametocytes from the same slide. Analysis included 70 female gametocytes, alongside male gametocytes from WT (68), Alp5b KO (69), and Alp5a KO (65) lines. Microgametes were similarly assessed with sample sizes of 92 (WT), 89 (Alp5b KO), and 64 (Alp5a KO). The data were comparable between WT and Alp5 KO lines in terms of the number of female gametocytes (P = 0.3126) and activated male gametocytes (P = 0.148). The DNA content significantly differed between WT and Alp5 KO microgametes (****P < 0.0001). Data from three independent biological replicates were analyzed using the Kruskal–Wallis test to assess statistical differences among groups. (E) Quantification of the zygote DNA content (a.u., arbitrary unit). We detected significant differences among the WT, Alp5a KO and Alp5b KO parasites (****P < 0.0001). The DNA content of 60 zygotes per group was analyzed. Data from three independent biological replicates were analyzed using Student’s t test to assess statistical differences.

https://doi.org/10.1371/journal.ppat.1013687.g006

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. DNA content is reduced in the ookinete and oocyst of Alp5 KO parasites.

(A) Representative confocal immunofluorescence images of ookinetes. Compared with WT ookinetes, Alp5 KO ookinetes presented a lower DNA content. (B) Quantification of ookinete DNA content (a.u., arbitrary unit). We detected significant differences between WT and Alp5 KO parasites (****P < 0.0001). The DNA contents of 70 (WT), 83 (Alp5b), and 83 (Alp5a) ookinetes were quantified. Data from three independent biological replicates were analyzed using the Kruskal–Wallis test to assess statistical differences among groups. (C) Widefield fluorescence microscopy images of Hoechst-stained oocysts showing the DNA content on different days. (D) The average DNA intensity of 50 oocysts was significantly different between WT and Alp5 KO parasites on days 8 (****P < 0.0001), 12 (****P < 0.0001) and 16 (****P < 0.0001). The DNA content was comparable between WT, ALP5b comp and ALP5a comp on day 8 (P = 0.7347), day 12 (P = 0.6874), and day 16 (P = 0. 8970). Data from three independent biological replicates were analyzed using the Kruskal–Wallis test to assess statistical differences among groups. (E) Widefield fluorescence microscopy images of Hoechst-stained oocysts illustrate the DNA content after the genetic crosses. The DNA content was restored in Alp5 KO parasites crossed with female gamete-defective lines. In contrast, genetic crosses with male gamete-defective lines did not rescue the DNA content. (F) The DNA content of 50 oocysts in Alp5 KO parasites was quantified and normalized to the mean DNA intensity of WT oocysts. A significant difference was observed in WT vs Pb47 KO × Alp5b KO (*P = 0.0330), WT vs Pb48/45 KO × Alp5b KO (****P < 0.0001), WT vs Pb47 KO × Alp5a KO (*P = 0.0172), and WT vs Pb48/45 KO × Alp5a KO (****P < 0.0001). Data from three independent biological replicates were analyzed using Student’s t test to assess statistical differences.

https://doi.org/10.1371/journal.ppat.1013687.g007

Ethics statement

All animal experiments were conducted in accordance with relevant guidelines and regulations and were approved by the Institutional Animal Ethics Committee of the CSIR-Central Drug Research Institute, India (IAEC Reference Nos: IAEC/2018/3 to IAEC/2023/15). Human red blood cells (RBCs) used for parasite culture were obtained under protocols approved by the Institutional Ethics Committee – Human Research, CSIR-Central Drug Research Institute (Approval No. CDRI/IEC/2017/A4).

Bioinformatics analysis

The protein sequences of Plasmodium berghei Alp5a and Alp5b were retrieved from PlasmoDB (https://plasmodb.org). The amino acid sequences were queried via NCBI protein BLAST (BLAST P) against model organisms. Owing to the unavailability of experimentally determined structures, the protein structures of PbAlp5a and PbAlp5b were predicted via AlphaFold 2 [62,63]. The quality of the predicted structure was assessed via SAVES v6.0 (https://saves.mbi.ucla.edu/). Further experimental high-resolution structures of P. falciparum actin 1 (6I4H with the resolution of 1.40 Å) and P. berghei actin 2 (6I4M with the resolution of 1.87 Å) were retrieved from the RCSB PDB database. To assess structural similarity and protein folding, the structure of Alp5a was superimposed onto the known human Arp3 protein structure, and Alp5b was superimposed onto the human Arp2 structure (human Arp2:6YW6 and Arp3:6UHC proteins from the PDB database https://www.rcsb.org/). Structural alignment and superposition were performed using pairwise alignment, employing the super command with an outlier rejection cutoff score of 2.0. The amino acid sequences of Alp5a and Alp5b were submitted to the STRING database to investigate potential interactions between these proteins [64]. To predict potential protein-protein interactions, docking studies were performed using the ClusPro server [65]. The workflow involves three primary steps: rigid-body docking using the PIPER algorithm to generate candidate conformations; clustering of low-energy poses based on structural similarity to identify the most representative complexes; and energy minimization to refine the selected models. Docked complexes were ranked according to cluster size and the weighted energy score of the lowest-energy member within each cluster. Subsequently, to predict the ternary complex, interaction studies were repeated using AlphaFold3 [66]. The top-ranking models were downloaded and visualized using PyMOL [67].

Parasites, mosquitoes and mice

Plasmodium berghei ANKA (MRA 311) and P. berghei ANKA GFP (MRA 867 507 m6cl1) were obtained from BEI Resources, USA. Anopheles stephensi mosquitoes were maintained in a walk-in environmental chamber at 28 °C, 80% relative humidity, and a 12:12 h light:dark cycle. Adults were provided with 10% (w/v) sucrose solution on cotton pads. Female Anopheles mosquitoes were allowed to probe and feed on Swiss mice to obtain a blood meal necessary for egg development. The parasites were transmitted to mosquitoes as previously described [68]. The infected mosquitoes were maintained in a walk-in environmental chamber set at 19 °C with 80% relative humidity and a 12-hour light/dark cycle. Swiss albino and C57BL/6 mice (6–8 weeks old) were used in this study. Chloroquine-sensitive P. falciparum, strain 3D7 (MRA-151, BEI Resources, USA) was used to assess the effects of the inhibitors. Parasites were cultured in human RBCs at 2% haematocrit in HEPES-modified RPMI-1640 medium (Sigma Aldrich, USA) supplemented with 1% glucose, 0.2% sodium bicarbonate, 100 µM hypoxanthine (Sigma-Aldrich #Cat. H9636), gentamycin (Gibco #Cat. 15750–060), (25 μg/ml) and 0.5% Albumax II (Gibco #Cat. 11021–037). The parasites were regularly synchronized with 5% D-sorbitol [69]. Human liver hepatocellular carcinoma (HepG2) cells (ATCC) were cultured in DMEM (Sigma-Aldrich #Cat. D5648) enriched with 10% FBS (Biological Industries #Cat. 04–121-1A), 0.2% NaHCO₃ (Sigma-Aldrich #Cat. S5761), 1% sodium pyruvate (Genetix #Cat. CC4016), and 1% penicillin‒streptomycin (Gibco #Cat. 15140–122) maintained at 37°C with 5% CO2.

Generation of Alp5b knockout and complemented parasite lines

To delete the Alp5b gene, the wild-type locus was replaced with a targeting cassette via double crossover homologous recombination. Two fragments, F1 (0.545 kb) and F2 (0.6 kb), were amplified from the 5’ and 3’ UTR regions via the primer sets 1611/1612 and 1613/1614, respectively. Fragments F1 and F2 were subsequently cloned into the pBC-GFP-hDHFR-yFCU vector at SalI and NotI/AscI, respectively. The final targeting cassette was separated from the vector backbone via digestion with XhoI/AscI and transfected into P. berghei ANKA schizonts as described previously [70]. Drug-resistant parasites that appeared after pyrimethamine selection were observed for GFP fluorescence. The genomic DNA was isolated from the GFP expressing drug-resistant parasites, and correct 5’ and 3’ site-specific integration was confirmed by PCR via the primer sets 1665/1225 and 1666/1215, respectively. The clonal lines were obtained by limiting dilution of the parasites and injecting them into Swiss mice. The clonal KO lines were confirmed via the primers 1671/1672. The Alp5b complemented line was generated by reintroducing the ORF via double crossover homologous recombination. A fragment encompassing the 5’UTR, ORF and 3’UTR was amplified via the primer set 1611/1614 and transfected into Alp5b KO schizonts. The transfected parasites were negatively selected with 5-fluorocytosine (Sigma-Aldrich #Cat. F7129) as previously described [71]. Gene restoration was confirmed by PCR via the primers 1671/1672.

Generation of Alp5a knockout and complemented parasite lines

To disrupt the Alp5a locus, two fragments, F3 (0.59 kb) and F4 (0.52 kb), were amplified via the primer sets 1895/1896 and 1897/1898, respectively. Fragments F3 and F4 were sequentially cloned into a pBC-mCherry-tgDHFR vector at KpnI/ClaI and NotI/AscI, respectively. The plasmid was linearized via KpnI/AscI and transfected into P. berghei ANKA schizonts. Genomic DNA was isolated from mCherry expressing pyrimethamine-resistant parasites, and 5’ and 3’ site-specific integration was confirmed by diagnostic PCR via primers 2089/1225 and 1913/2090, respectively. Clonal lines were obtained as described above and confirmed via the primers 2091/2092. To restore gene function, fragment F5 encompassing the Alp5a-expressing cassette (5’UTR + ORF + 3’UTR) was amplified via primers 2268/2269 and subsequently cloned into pBC-P230-mCherry-hDHFR at ClaI and EcoRI. The linearized cassette was transfected into Alp5a KO schizonts, which were selected via the WR99210 drug (Sigma-Aldrich #Cat. SML2976) as previously described [72]. Correct 5’ and 3’ site-specific integration was confirmed by diagnostic PCR via primer sets 2270/2271 and 1215/2272, respectively. The restoration of the ORF was confirmed via the primer set 2091/2092. The primer sequences are given in S1 Table.

Attempts to generate Alp5a/Alp5b double knockout parasites

Two strategies were employed to generate Alp5a/Alp5b double KO parasites. In the first strategy, the linearized cassette of Alp5b was transfected into Alp5a KO schizonts, which were selected with the WR99210 drug. In the second strategy, Alp5b and ALP5a targeting cassettes were cotransfected into P. berghei ANKA schizonts, which were selected with pyrimethamine (Sigma-Aldrich #Cat. 46706).

Generation of Alp5a and Alp5b transgenic parasites

For Alp5b tagging with 3XHA-mCherry, two fragments, F6 (0.54 kb) and F7 (0.60 kb), were amplified via the primer sets 1675/1676 and 1677/1614 and cloned into the pBC-3XHA-mCherry-hDHFR vector at XhoI/BglII and NotI/AscI, respectively. For Alp5a tagging with 3XHA, two fragments, F8 (0.61 kb) and F4 (0.52 kb), were amplified via the primers 2128/2129 and 1897/1898 and cloned into the pBC-3XHA-hDHFR vector at XhoI/BglII and NotI/AscI, respectively. The vector was linearized via XhoI/AscI and transfected into P. berghei ANKA schizonts as described above. Correct 5’ and 3’ site-specific integration was confirmed by diagnostic PCR via the primers 1724/1392 and 1215/1666 for Alp5b and 2130/1218 and 1215/2090 for Alp5a. The primer sequences are given in S1 Table.

Generation of gamete-defective lines

The 6-cysteine proteins, P47 (PBANKA_1359700) and P48/45 (PBANKA_1359600) targeting constructs for generating female and male gamete-defective lines, respectively, were obtained from the PlasmoGEM resource (https://plasmogem.umu.se/pbgem/) [73]. The plasmids were linearized via NotI, transfected into P. berghei ANKA schizonts and selected with pyrimethamine. Genomic DNA was isolated from drug-resistant parasites, and site-specific integration of the P47 and P48/45 targeting cassettes was confirmed by diagnostic PCR via primers 2232/2035 and 2229/2036, respectively. Clonal lines were obtained by limiting dilution of the parasites, and the absence of P47 and P48/45 ORFs was confirmed via primers 2233/2234 and 2230/2231, respectively. The primer sequences are given in S1 Table.

Asexual blood-stage propagation and gametocyte development

Two groups of Swiss mice (5 mice/group) were injected intravenously with 200 µl of blood containing 0.5% parasitemia. Blood growth and gametocyte development were monitored via Giemsa-stained blood smears.

Analysis of parasite development in mosquitoes

Swiss mice were injected with WT or KO parasites as described above. The mice positive for gametocytes were anesthetized and placed in starved mosquito cages for a blood meal. To analyze the different developmental stages of parasites, mosquitoes were dissected under a stereo-zoom microscope. Blood boluses were collected 10–20 min or 18–22 h postfeeding to analyze gametes and ookinetes, respectively, and slides were prepared for live imaging as previously described [71]. The blood boluses was crushed and diluted, and the ookinetes were counted via a hemocytometer. Mosquito midguts were collected on days 8–18 to determine the oocyst numbers and areas. The oocyst area was determined via Fiji software. Midgut and salivary gland sporozoites were enumerated on days 14 and 19–22 post-blood meal, respectively, as previously described [72].

Genetic cross

A genetic cross between the mutant and WT lines was performed. For this purpose, blood was collected from the mice infected with WT or KO parasites and mixed at a ratio of 1:1. Two hundred microliters of blood with 0.5% parasitemia was injected into a group of Swiss mice. Mosquitoes were allowed to probe for a blood meal in infected mice. Mosquitoes were dissected on days 15 and 19–22 after a blood meal to observe oocysts and salivary gland sporozoites, respectively.

Gametocyte purification and activation

Gametocytes were purified using a 50% Nycodenz gradient. Swiss mice were pre-treated with phenylhydrazine (6 mg/ml; Sigma-Aldrich, Cat. #114715) for three consecutive days to induce reticulocytosis, and subsequently injected intraperitoneally with 200 µl of blood containing 1% parasitemia. After 3–4 days, the presence of gametocytes was confirmed by Giemsa-stained blood smears. Infected blood was collected into RPMI medium, layered onto a 50% Nycodenz gradient, and centrifuged at 200 × g for 20 minutes. The gametocyte-rich layer appearing at the interphase was carefully collected and washed 2–3 times with PBS to remove residual Nycodenz. The purified gametocytes were activated using an exflagellation medium composed of 25 mM HEPES (Sigma-Aldrich, Cat. #83264) and 100 µM xanthurenic acid (Sigma-Aldrich, Cat. #D120804), supplemented with RPMI medium (Sigma-Aldrich, Cat. #23400–013) at pH 8.0. Parasite stages were observed either live or after fixation with 4% paraformaldehyde (PFA; Sigma-Aldrich, Cat. #HT5012).

Exflagellation centres

An exflagellation assay was conducted by incubating 10 µl of purified gametocytes with 40 µl of exflagellation medium at 20°C for 5–10 minutes. The exflagellation centres were observed under a phase contrast microscope at 100x magnification and counted per field.

Ookinete motility assay

Blood boluses were isolated, diluted, and mixed with an equal volume of Matrigel (BD Biosciences #Cat.356231). A 5 µl drop was placed on a slide, gently covered with a coverslip, and then incubated for 20 minutes at RT. After incubation, the slides were imaged under a Nikon Eclipse 80i microscope at 100x magnification, and a time-lapse video of 61 loops for 5 minutes was recorded.

Generation of the antibodies

Antibodies against Alp5b and P25 were generated in rats. KLH-conjugated peptides of Alp5b (Cys-SNKKKWITKNEYSANP) (GLS #Cat. 998808). and P25 (SPNTQCKNGFLAQMSN-Cys) (GLS #Cat.997976) were synthesized by GL Biochem. The peptides were immunized once with complete Freund’s adjuvant and twice with incomplete Freund’s adjuvant. The rats were bled, and the serum was collected for further analysis.

In vivo infectivity of sporozoites

Infected mosquitoes were used for sporozoite transmission 22 days post-blood meal, as previously described [74]. Briefly, C57BL/6 mice were anesthetized and kept on the mosquito cages. Fifty mosquitoes per C57BL/6 mouse were used. Infection was observed via Giemsa-stained blood smears.

In vitro development of exoerythrocytic stages

To observe the development of the parasite in hepatocytes, salivary gland sporozoites were added to HepG2 cells as previously described [75]. Briefly, HepG2 cells were seeded in a 48-well plate (55,000 cells/well) containing collagen-coated 9 mm glass coverslips. The next day, salivary gland sporozoites (5,000/well) were added to the wells, and the plate was spun at 310 × g for 4 minutes. The infected culture was kept in a CO₂ incubator set at 37°C with 5% CO₂. The medium was changed every 12 h, and the culture was fixed with 4% paraformaldehyde for 20 minutes at RT.

Immunofluorescence assay (IFA)

Parasites at different stages were harvested, and IFA was performed as previously described [76]. Briefly, parasites were fixed for 20 minutes at RT with 4% paraformaldehyde. Blood smears and sporozoite slides were permeabilized with 0.1% Triton X-100 (Sigma‒Aldrich #Cat. T8787) for 10 minutes at RT, and liver stage samples were permeabilized with chilled methanol. The samples were blocked with 1% BSA for 1 h at RT and incubated with primary antibodies for 1 h at RT or overnight at 4°C. The primary antibodies used were anti-CSP (diluted 1:1000), anti-HA (diluted 1:1000), anti-actin (diluted 1:1000), anti-g377 (diluted 1:100) [71], anti-tubulin (diluted 1:100), anti-P25 (diluted 1:100), and anti-MSP1 (diluted 1:5,000). The following secondary antibodies were used: Alexa Fluor 488-conjugated anti-rabbit IgG (diluted 1:1,000), (Invitrogen #Cat. A11008), Alexa Fluor 594-conjugated anti-rabbit IgG (diluted 1:1,000), (Invitrogen #Cat. A21442), Alexa Fluor 594-conjugated anti-rat IgG (diluted 1:1,000), (Invitrogen #Cat. A11007), Alexa Fluor 488-conjugated anti-rat IgG (diluted 1:1,000), (Invitrogen #Cat. A11006), Alexa Fluor 488-conjugated anti-mouse IgG (diluted 1:1,000), (Invitrogen #Cat. A11001) and Alexa Fluor 594-conjugated anti-mouse IgG (diluted 1:1,000), (Invitrogen #Cat. A11032). Nuclei were stained with Hoechst 33342 (Invitrogen #Cat. 62249), and samples were mounted with Prolong Diamond antifade reagent (Invitrogen #Cat. P36970). Images were acquired using a confocal laser scanning microscope (Olympus BX61WI) with FV1000 software and a UPlanSAPO 100 × /1.4 NA oil-immersion objective, a widefield fluorescence microscope (Nikon Eclipse 80i) with a Plan Fluor 100 × /1.30 NA oil objective, or a Leica DM3000 LED microscope with a 100 × /1.25 NA oil-immersion objective.

Analysis of DNA content

Developing male gametocytes were identified by strong anti-tubulin antibody staining, while female gametocytes were distinguished by their diffuse staining pattern. The zygote and ookinetes were immunostained with an anti-P25 antibody. For DNA content analysis of oocysts, mosquito midguts were collected at 8, 12, 15, and 16 days post-feeding, fixed in 4% paraformaldehyde for 30 minutes at room temperature, then washed with cold PBS. The samples were subsequently permeabilized and immunostained with an anti-CSP antibody. Nuclei were stained with Hoechst 33342 and mounted using ProLong Diamond antifade reagent. Fluorescence images were acquired and analyzed using Fiji software (ImageJ). Hoechst-stained DNA fluorescence intensity was quantified by measuring the total fluorescence within each nucleus after subtracting the background signal. The resulting fluorescence intensity values were normalized to those of female gametocytes or single-nucleated parasites. The software (https://theolb.readthedocs.io/en/latest/imaging/measuring-cell-fluorescence-using-imagej.html) was used to measure the fluorescence intensity.

Determination of the G-actin/F-actin ratio

The G-actin/F-actin ratio was determined via a G-actin/F-actin in vivo assay kit (Cytoskeleton, Inc. #Cat. BK037) according to the manufacturer’s instructions. Blood samples were lysed with saponin, and parasite pellets were washed washed four times with 1X PBS, resuspended in LAS2 buffer, and incubated at 37°C for 10 minutes. Following sequential centrifugation steps—first at 20,913 × g for 5 minutes and then ultracentrifugation at 100,000 × g for 1 hour—the supernatant (G-actin) and pellet (F-actin) fractions were separated. Both fractions were lysed in Laemmli buffer, subjected to 10% SDS-PAGE, and analyzed by western blotting to quantify actin forms.

Effects of inhibitors on erythrocytic schizogony

Synchronized P. falciparum parasites were incubated with different concentrations of cytochalasin-D (Sigma-Aldrich #Cat. C8273), jasplankinolide (Sigma-Aldrich #Cat. 420127), and the Arp2/3 complex inhibitor, CK-666 (Sigma-Aldrich #Cat. 182515) throughout the culture. A schizont culture was set up as described above to examine the effects of the inhibitors on the asexual blood stages of P. berghei. The inhibitors were added to the culture and incubated at 37°C for 22–24 h. The growth and morphology of the parasites were monitored via Giemsa-stained smears.

Immunoprecipitation

Purified gametocytes of the Alp5a-3XHA and WT strains were washed four times with DPBS and crosslinked with 5 mM DSP (dithiobis succinimidyl propionate; Thermo Scientific #Cat. 22586). The DSP-crosslinked samples were incubated for 30 minutes in the dark and centrifuged at 20,913 × g for 5 minutes. The pellet was lysed on ice in 500 µl of RIPA buffer for 30 minutes. The mixture was centrifuged at 20,913 × g for 15 minutes, and the supernatant was transferred to a fresh microcentrifuge tube. A total of 25 µl of anti-HA magnetic beads (Thermo Scientific #Cat. 88838X) was washed with 0.05% TBST four times in a magnetic stand (Stem Cell Technologies #Cat. 18000). The beads were incubated with parasite lysate for 30 min at RT on a tube rotator. After incubation, the beads were washed with TBST followed by ultrapure water, eluted in 2x Laemmli buffer (Bio-Rad #Cat. 1610737), and boiled for 5 minutes. The samples were resolved via 10% SDS‒PAGE, transferred to nitrocellulose membranes, and probed with anti-Alp5b and anti-HA antibodies.

Western blotting

Western blotting was performed as previously described [76]. Briefly, samples were resolved via 10% SDS‒PAGE and transferred onto a nitrocellulose membrane (Bio-Rad #Cat. 1620112). The membrane was blocked with 1% BSA/PBS and incubated with anti-HA (diluted 1:1,000), (CST #Cat. C29F4) or anti-Hsp70 (diluted 1:1,000) antibodies. The blot was washed 3–4 times with 1X PBST and then incubated with HRP-conjugated anti-rabbit (diluted 1:5,000) (Amersham Biosciences #Cat. NA934V) or anti-mouse IgG (diluted 1:5,000) (Amersham Biosciences #Cat. NA931V). The signals were detected via an enhanced chemiluminescence (ECL) substrate (Bio-Rad, #Cat. 170--5060) and imaged via a ChemiDoc XRS+ system (Bio-Rad, USA).

Software and statistics

All statistical analyses were conducted using GraphPad Prism 9 software. Comparisons involving more than two groups were performed using one-way ANOVA, while comparisons between two groups were made using Student’s t-test. Statistical significance is indicated as follows: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; n.s., not significant.