The P. falciparum α-tubulin-1 and β-tubulin are two components of the miroctubles and polar rings in the invasive apparatus. P. falciparum α-tubulin-1 and human α-tubulin sequences share >84% amino acid identity and 93% (422/453) sequence similarity ( Figure 1 ). P. falciparum β-tubulin-1 and human β-tubulin sequences share >85% amino acid identity and 94% (417/445) sequence similarity ( Figure 1 ). Therefore, the capacity of purified polyclonal anti-human α-tubulin and β-tubulin Ab to block P. falciparum oocyst formation in mosquitoes was examined using the standard membrane feeding assays (SMFA) (Zhang et al., 2015). These assays determine efficacy by counting the number of oocysts (intensity of infection) and prevalence of infection formed in midguts with and without experimental interventions. An irrelevant rabbit pAb (anti-V5) was used as a negative control. The mortalities of mosquitoes fed with or without pAb were below 10%, e.g., 0 to 2 dead mosquitoes per treatment, displaying no difference. The results show that pAb (0.01 mg/ml) against human α-tubulin significantly reduced the number of P. falciparum oocysts in An. gambiae midguts (p=0.015), compared to controls ( Figure 1 ). The infection prevalence decreased from 26% and 64% to 6% and 4% in two replicates. Oocysts were detectable in only one or two mosquitoes from the anti-α-tubulin experimental feeding groups. On the contrary, there is no significant difference in oocyst counts between pAb against human β-tubulin and the control mosquito midguts (p=0.23 and 0.35), indicating that the pAb against human β-tubulin did not prevent ookinetes from invading midguts ( Figure 1 ). Collectively, these data support the conclusion that α-tubulin-1 in sexual stage parasites plays key roles during malaria transmission to mosquitoes and is accessible to extracellular antibodies.

A lower gametocyte-to-ookinete conversion might cause the Ab-mediated reduction of oocysts. Therefore, we tested this possibility. The purified rabbit polyclonal Ab against human α-tubulin was added into the P. falciparum cultures containing 3-10% gametocytes. An equal amount of non-related purified rabbit Ab (anti-V5) was used as a negative control. After incubation, the gametocytes and ookinetes were stained with the Giemsa and examined microscopically. A statistically equivalent number of gametocytes and ookinetes were detected in both the experimental and control groups. The ookinete conversion rates (CR, calculated as the percentage of ookinetes among the total gametocytes) in the presence of anti-α-tubulin Ab and anti-V5 Ab (negative control) are shown in Table 1 . Three replicates showed no significance between the experimental and control groups (p=1). These data support the conclusion that Ab against α-tubulin-1 does not affect the conversion of P. falciparum gametocytes to ookinetes.

Next, we used commercially generated specific pAb against P. falciparum 3D7 α-tubulin-1 to examine their effects on P. falciparum transmission. The P. falciparum 3D7 Hsp70 gene was chosen as a negative control because it is also an abundant cytoplasmic protein. The pAb against P. falciparum α-tubulin-1 was generated in rabbits.

Since antibody titers in sera are related to antibody’s interactivity with antigens, the anti-sera were normalized with a pre-immune serum to the titer of 2x10⁵ (about 5-10 folds dilution). The sera were mixed with the same volume of cultured P. falciparum and used to infect An. gambiae with SMFA, as described previously. Pre-immune serum served as a negative control. The results showed that antiserum against α-tubulin-1 significantly (p=0.0046 and 0.0004 in two replicates) reduced by 3-5-fold the median intensity of infection of P. falciparum oocysts in mosquitoes in the two independent replicates when compared to the pre-immune serum ( Figure 2 ). The antisera against the Hsp70 did not significantly impact P. falciparum infection intensity (p=0.20 and 0.22 in two replicates).

We purified anti-P. falciparum α-tubulin-1 pAb to remove the serum factors. First, we analyzed the specificity of the purified antibodies. The SDS-PAGE was used to separate proteins in the lysates of cultured P. falciparum with ookinetes, P. falciparum α-tubulin-1-expressed Hi5, Hi5, human liver cells (HEK 293), and whole blood ( Figure 3 ). Immunoblotting analyses with the purified rabbit pAb specifically recognized α-tubulin-1 in P. falciparum ( Figure 3 ). Moreover, the transmission-blocking activity in SMFA of this Ab at different concentrations from 0 to 450 μg/mL against P. falciparum showed significantly reduced oocysts in numbers in a concentration-dependence ( Figure 3 ). As the concentration of Ab increased, the median number of oocysts and infection prevalence decreased from 3 and 69 to 0 and 29, respectively. The half-maximal inhibitory concentration (IC₅₀) is calculated to be 36 μg/mL.

Since the sequence between human α-tubulin and P. falciparum α-tubulin-1 share ~ about 84% identity, mAb was generated to evaluate the specific reaction and cross-reaction. The E. coli-expressed P. falciparum α-tubulin-1 was used to immunize mice. Sixteen hybridomas were generated that produced mAb (mAb) (Boster Bio, Pleasanton, CA). Using ELISA, the molecular interaction between mAb and bovine serum album (BSA), the lysates of uninfected human red blood cells (RBC), and P. falciparum-infected human red blood cells (Pf-iRBC) showed that two mAb (A3 and A16) bound to Pf-iRBC specifically ( Figure 4 ). The other fourteen mAb did not bind to Pf-iRBC lysate.

We evaluated the cross-reaction of mAb with human red blood cell lysate and human kidney cell (HEK293) lysates. A3 and A16 were confirmed to interact with Pf-iBRC lysate ( Figure 4 ). Also, A16 did not interact with RBC or HEK293 lysates, while A3 displayed a slight degree of cross-reaction with HEK293 lysate (p=0.1). Therefore, we used a Western blot to confirm the specificity of the mAb to tubulins in the lysate of P. falciparum-infected red blood cells (Pf-iRBC), human kidney cells (HEK293), recombinant P. falciparum α-tubulin-1 protein, mosquitoes, and human red blood cells (RBC). As expected, both A3 and A16 recognized α-tubulin-1 in P. falciparum-infected red blood cells and recombinant α-tubulin-1 protein from Hi5 cells ( Figure 4 ). Notably, they did not bind to tubulins in human cells and mosquitoes.

Next, we examined the efficiency of mAb in blocking P. falciparum transmission to An. gambiae by SMFA. Results showed that both A3 and A16 mAb (final concentration was ~ 10 μg/mL) significantly inhibited malaria transmission ( Figure 5 ). At the same time, mAb A4 did not specifically bind to P. falciparum lysate nor inhibit malaria transmission. Furthermore, the efficiency of mAb in blocking P. falciparum transmission to An. gambiae was examined using diluted mAb. Results confirmed that both A3 and A16 mAb significantly inhibited malaria transmission. A3 and A16 mAb inhibited P. falciparum transmission to An. gambiae with IC₅₀ of 12.2 μg/ml ( Figure 5 ) and 2.8 μg/ml ( Figure 5 ), respectively.

The sequence of P. falciparum α-tubulin-1 (P. falciparum 3D7; XP_001351911.1) was elongated with neutral GSGSGSG linkers at the C- and N-terminus to avoid truncated peptides. The elongated antigen sequences were converted into 15 amino acid peptides with a peptide-peptide overlap of 14 amino acids. The resulting custom peptide microarrays contained 1,191 different peptides printed in duplicate (2,382 peptide spots) and were framed by additional HA (YPYDVPDYAG, 106 spots) control peptides. Incubation of a custom peptide microarray with mouse IgG antibody A3 (D3D1) at concentrations of 1 µg/ml, 10 µg/ml, and 100 µg/ml in incubation buffer was followed by staining with the secondary and control antibodies (red = sample staining/green = control staining). We did not observe any strong interaction signals, suggesting that mAb A3 recognized a conformational epitope or an epitope formed by a peptide longer than 14 amino acids ( Figure 6 ).

We also mapped the epitope of mAb A16 (D3F4). Peptide microarray with mouse A16 mAb at concentrations of 1 µg/ml and 10 µg/ml in incubation buffer was followed by staining with the secondary and control antibodies. We observed a strong mAb response against a single epitope-like spot pattern formed by adjacent peptides with the consensus motif REDLAALEKD on α-tubulin-1 at high signal-to-noise ratios ( Figure 6B ). This sequence is one amino acid (bold L to M) difference at the center of this epitope between P. falciparum α-tubulin-1 and human α-tubulin.

We also determined the interaction between antibodies against P. falciparum α-tubulin-1 and impermeable ookinetes. We labeled customized polyclonal rabbit anti-α-tubulin-1 Ab with CF™ 568 dye and examined the antibody binding sites on live ookinetes under a fluorescence microscope. P. falciparum ookinetes were cultured in vitro (Zhang et al., 2015) and enriched by differential density centrifugation using 65% Percoll. Uninfected red blood cells and cell debris were sedimented to the bottom of the Percoll, while gametocytes and ookinetes were enriched at the interface of the Percoll and the culture medium ( Figure 7 ). The enriched gametocytes and ookinetes were washed and incubated with the CF™ 568-conjugated anti-α-tubulin Ab. After removal of the unbound Ab, the cells were deposited onto coverslips and counterstained with DAPI. Fluorescence microscopy revealed that anti-α-tubulin Ab is bound to the apical end of ookinetes ( Figure 7 ). Detailed examination of individual ookinetes revealed an intense fluorescent signal localized to apical ends (white arrows in Figure 7 , closeup), supporting the conclusion that anti-α-tubulin Ab recognized the α-tubulin-1 of the living ookinetes at their apical ends. Notably, we also saw that anti-α-tubulin-1staining was not concentrated at the extreme distal tip of the ookinetes ( Figure 7 , closeup of apical complex), but instead, the signal was localized to the apical ends, suggesting that Ab recognized α-tubulin-1 at apical polar rings of living ookinetes. Importantly, fixation with methanol and permeabilization of ookinetes revealed a signal that was distributed across the cells ( Figure 7 , permeable control). Substitution of anti-α-tubulin polyclonal Ab with an irrelevant (anti-V5) polyclonal Ab revealed no signal ( Figure 7 , Negative control).

Next, we examined the cells under a confocal microscope to obtain higher-resolution images. Gametocytes were observed in the bright field, but anti-α-tubulin-1 antibodies did not bind to them ( Figure 8 ), supporting the conclusion that α-tubulin-1 is not exposed at the gametocyte surface or that the host cell membrane covers gametocytes. Notably, anti-α-tubulin-1 antibodies were bound to a live ookinete ( Figure 8 ). Upon higher magnification, the invasive apparatus of the ookinete could be distinguished ( Figure 8 ), showing a protruding component was visible at one apical end of the ookinete. Two rings were detected at the apical complex. Indeed, the pixel intensity for the red color (α-tubulin-1) at the ookinete apical end is much stronger than in the other positions, and the apical polar rings were the strongest ( Figure 8 ). These imaging studies support the conclusion that α-tubulin-1 is exposed on live ookinete surface at apical rings of the invasive apparatus.

Since P. falciparum α-tubulin-1 in the invasive apparatus is accessible by large external molecules, we tested its interaction with An. gambiae midgut protein FREP1. The insect cell-expressed recombinant full-length FREP1 was used to coat the ELISA plates, followed by incubation with insect cell-expressed recombinant P. falciparum α-tubulin-1. The retained α-tubulin-1 was detected using a purified anti- α-tubulin-1 pAb. The insect cell-expressed chloramphenicol acetyltransferase (CAT) coated additional wells in the plates were used as the negative control. Since FREP1 can interact with P. falciparum lysate (Zhang et al., 2015), P. falciparum-infected blood cell lysates served as a positive control. Results showed that P. falciparum α-tubulin-1 bound FREP1 protein compared to background signals detected in the negative control wells (p<0.05) ( Figure 9 ). A midgut lysate mixed with Hi5-expressed α-tubulin-1 with a His-tag, pulled down by Ni-NTA beads, and detected by rabbit anti-FREP1 polyclonal antibodies, showed endogenous FREP1 was detected in the α-tubulin-1-pulldown fraction ( Figure 9 ), demonstrating that the strong interaction between α-tubulin-1 and FREP1 among the mixture of various proteins.

The full-length coding sequence (CDS) of P. falciparum HSP70 was amplified with primers of 5’-CGGGATCCATGGCTAACGCAAAAGCAAAGC-3’ and 5’-CGCAAGCTTTTAATCAACTTCTTCAACAGTTGGTC-3’. The full-length CDS of P. falciparum 3D7 α-tubulin-1 was amplified with primers of 5’-CGGGATCCATGAGAGAAGTAATAAGTATAC-3’ and 5’-GCTCTAGAATAGTCTGCCTCATATCCTTC-3’. The sequences underlined are restriction enzyme recognition sites. Depending on primer sequences, amplicons were digested with restriction enzymes BamHI and XbaI or BamHI and HindIII (New England BioLabs, MA). The column-isolated DNA fragments were ligated into a modified pFastBac1 plasmid having a 6xHis tag at the C-terminus, and positive recombinant plasmids were transformed into DH10Bac competent cells. Recombinant bacmids were extracted from white colonies. The sf9 cells were used to proliferate virus. The high-titer virus was used to infect Hi5 cells to express proteins. Following the manufacturer’s manual (Invitrogen, 2015), 0.5 mL 2x10⁵/mL Sf9 cells in a complete medium (Grace’s medium (ThermoFisher) containing 10% FBS) were deposited into one well of a 24-well plate. After incubation for 30 m, the supernatant was removed. Concurrently, one μL bacmid DNA (1 μg/μL) combined with 25 μL Grace’s medium was mixed gently, and 2 μL Cellfectin II (ThermoFisher) diluted with 25 μL Grace’s medium also was mixed gently. The DNA mixture and Cellfectin II mixture were combined and incubated for 30 m at RT and then added to the above-prepared Sf9 cells. After incubation for 5 h at 27°C, the transfection mixture was removed, and 0.5 ml complete growth medium (Grace’s medium with 10% FBS and 10 units of penicillin/ml and 10 µg/mL streptomycin) was added. After incubation at 27°C for three days, the medium containing recombinant baculovirus particles was collected and used to infect Hi5 to express recombinant proteins. After three passages (10 μl old cell culture to 200 μl new cell culture), expressed proteins were confirmed with ELISA by detecting the His-tag at the C-terminus of recombinant proteins.

The Hi5 cell-expressed recombinant α-tubulin-1 proteins were used to determine their specific interaction with FREP1 by ELISA. The expressed proteins in the cell lysate were normalized to 1 nM, and then a 96-well plate was coated overnight with 100 μl cell lysate to detect interactions with FREP1. After blocking the plate with 100 μl 0.2% BSA in 1×PBS, 100 μl of Hi5 cell-expressed recombinant FREP1 (Zhang et al., 2015) in PBS (1 nM) was introduced onto the wells and incubated for 1.5 h at RT. Purified anti-FREP1 Ab (3 μg/mL in PBS) followed by 100 μl 2^nd Ab (1:10,000 dilution in PBS) was used to detect any retained FREP1. The plates were developed with 100 µL of p-NPP solution (Sigma-Aldrich, St. Luis, MO) until the colors appeared. The absorbance at 405 nm was measured using a plate reader. Between each incubation, wells were washed with 100 µL PBST three times with 3 m incubation each time. For the control group, 1 nM of CAT protein (chloramphenicol acetyltransferase, containing a 6×His tag at the C-terminus) expressed in the same baculovirus-expression system (Thermo Fisher) was used to coat plate wells. The P. falciparum-infected blood cell lysate (1 mg/ml proteins) was used to coat wells as the positive control. Each sample was analyzed in triplicate, and the Mann-Whitney test was used to evaluate the difference between the negative control and experimental group.

For the pull-down, we prepared blood-fed mosquito midguts as described previously (Niu et al., 2021). About 400 μl of Hi5 cells lysis (1 mg/mL protein) containing the expressed α-tubulin-1 were mixed with 100 μL of midgut lysis from 100 blood-fed mosquito midguts and then incubated with 100 μL Ni-NTA bead for 24 hours on a rotary shaker (15 rotations per m). After centrifugation (300 g for 5 min), the pellet was washed with PBS three times, and supernatants were discarded. Subsequently, 100 μL SDS-PAGE loading buffer was added, boiled, and spun quickly to settle the beads. The supernatant was used for immunoblotting analysis and reacted with anti-FREP1 antibodies. Hi5 cell lysate expressed with the unrelated protein (CAT) was used as the negative control.

Plasmodium falciparum parasites (NF54 strain from MR4, Manassas, VA, the catalog is MRA-1000) were maintained in 5 mL RPMI 1640 medium (Life Technologies) supplemented with 10% heat-inactivated (56°C for 45 min) human AB+ serum (Interstate Blood Bank, Memphis, TN), 12.5 μg/ml hypoxanthine, and 4% hematocrit using a 6-well cell culture plate (Corning, NY) in a candle jar at 37°C (Zhang et al., 2015). The parasitemia and gametocytemia were analyzed daily by Giemsa staining of thin blood smears. When the parasitemia reached 5%, 0.5 ml of culture was transferred into 5 ml fresh complete RPMI-1640 supplemented with 4% hematocrit. The remaining culture was maintained for another ten days to two weeks before being used to infect mosquitoes. P. falciparum cultures harboring 2-5% stage V gametocytes were diluted 10-fold in RPMI 1640 supplemented with 20% heat-inactivated human AB+ serum and 50 μg/mL hypoxanthine and incubated at room temperature for 18-24 h to develop ookinetes.

IFAs were conducted independently with rabbit polyclonal Ab against human α-tubulin (ProteinTech, IL, USA) and against P. falciparum α-tubulin-1. Ab was labeled with CF™ 568 dye (Mix-n-Stain™ CF^® Dye Antibody Labeling Kits - CF^®568, Biotium Inc, CA). The cultured P. falciparum ookinete cell mixtures (10⁵ cells) were suspended at 200 µL in a complete RPMI-1640 medium and were deposited carefully on the top of 0.6 mL 65% Percoll (100% Percoll diluted with incomplete RPMI-1640) in a 1.5 mL plastic tube. After centrifugation at 2,000 rpm for 5 m, the cells at the interface of medium and 65% Percoll were collected and diluted 10-fold with incomplete RPMI-1640 medium and then collected by centrifugation at 2,000 rpm for 5 m. The cells were washed twice with 0.5 mL PBS through centrifugation (2,000 rpm for 5 m). Because the fluorescence is sensitive to light, the following steps were done in a dark room. The cells were suspended in 50 µL PBS and incubated with fluorescence-conjugated anti-α-tubulin Ab (200-fold dilution, final Ab concentration was ~1 µg/mL) for 45 m. The cells were collected and washed with 0.5 mL PBS twice by centrifugation (2,000 rpm for 5 m). Finally, the cells (~10⁴ cells) were suspended in 20 µL PBS, and 4 µL were deposited onto a slide coverslip. Immediately after drying, the cells on a coverslip were mounted on a slide with 5 μL of Vectashield mounting media (Vector Laboratories, Burlingame, CA). After incubation at 4°C for 2 h, the cells were examined using a Nikon Eclipse Ti-S fluorescence microscope. Cells treated similarly except using unrelated antibodies (anti-V5) were used as a negative control. Cells on slides were fixed with methanol and incubated with fluorescence-conjugated anti-α-tubulin Ab (200-fold dilution, final Ab concentration was ~1 µg/mL) as a positive control.

SMFA was performed as described (Niu et al., 2015; Zhang et al., 2015). The commercial rabbit polyclonal Ab against full-length human α-tubulin or β-tubulin (ProteinTech, IL, USA) were used in the malaria transmission-blocking assay. Approximately 6μL rabbit polyclonal Ab against human α-tubulin (0.3 mg/mL) was added into 200 μL P. falciparum-infectious blood (final Ab concentration was about 0.01 mg/mL) and fed to An. gambiae. The same amount of non-related purified rabbit pAb (anti-V5 tag) was used as a control. The number of oocysts in mosquito midguts was counted seven days late and analyzed statistically by the Mann-Whitney test that was implemented in GraphPad Prism software (Version 6h, San Diego, CA).

We commercially generated specific pAb against P. falciparum α-tubulin-1 (Boster Bio, Pleasanton, CA). The full-length P. falciparum α-tubulin-1 and Hsp70 genes were cloned from P. falciparum 3D7 and expressed in E. coli. The P. falciparum 3D7 Hsp70 gene was chosen as a negative control simply because it is also an abundant cytoplasmic protein. The pAb against P. falciparum α-tubulin-1 was generated in rabbits. The anti-sera were normalized with a pre-immune serum to the titer of 2x10⁵, mixed with the same volume of cultured P. falciparum, and fed 3-5-day-old female An. gambiae. The number of oocysts in the mosquito midguts was analyzed as described previously.

The rabbit pAb against P. falciparum 3D7 α-tubulin-1 was purified through protein A/G column and used in SMFA to rule out the possible effect of other components in sera. About 6μL rabbit pAb with different concentrations was mixed with 200 μL P. falciparum-infectious blood (final Ab concentrations were 0, 17, 50, 150, 450 μg/mL) and fed to An. gambiae. Seven days late, the number of oocysts in mosquito midguts was counted for statistical analysis by the Mann-Whitney test.

The half maximal inhibitory concentration (IC₅₀), e.g., the inhibition concentration that displayed 50% of infection intensity (the number of oocysts), was calculated using an online tool (APA, 2023).

We also generated mouse monoclonal antibodies (mAb) using E. coli-expressed recombinant full-length P. falciparum α-tubulin-1 through Boster Bio (Pleasanton, CA). The mAb were purified using protein A/G.

First, the numbers of gametocytes in cultures were counted in a blood smear under light microscopy. Approximately 1 mL of 15-day P. falciparum-cultured blood that harbors stage V gametocytes was collected by centrifugation (800x g for 4 m) and re-suspended into 10-fold ookinete cultures (RPMI-1640 supplemented with 20% heat-inactivated human AB+ serum and 50 μg/mL hypoxanthine). Approximately 3 µL and 5 µL of polyclonal anti-human α-tubulin Ab (0.3 mg/mL, Abcam) was then added into 150 µL P. falciparum-infectious blood (50- and 30-fold dilutions, final Ab concentrations were ~6 µg/mL and 10 µg/mL respectively) to a well in a 96 well plate. The parasites were then incubated at 21-23°C for about 16 hours. The same amount of non-related purified rabbit Ab (anti-V5) was used as a control. The number of ookinetes was counted, and the ookinete conversion rate (CR), the percentage of ookinetes among the total P. falciparum gametocytes, was calculated.

We mapped epitopes commercially (Pepperprint, Germany). A custom peptide microarray was pre-stained with the secondary antibody to investigate background interactions with the antigen-derived peptides that could interfere with the main assays. Subsequent incubation of further custom peptide microarrays with mouse IgG antibodies at concentrations of 1 µg/ml, 10 µg/ml, and 100 µg/ml was followed by staining with the secondary and control antibody. Read-out was performed with an Innopsys InnoScan 710-IR Microarray Scanner. The additional HA control peptides framing the peptide microarrays were subsequently stained with the control antibody as internal quality control to confirm assay performance and peptide microarray integrity. Microarray image analysis was done with PepSlide^® Analyzer. The intensity plots were correlated with peptides on the microarray to identify the epitopes of the antibody samples. For a better data overview, the baselines of the intensity plots were shifted up.