Overexpression of hepatocyte EphA2 enhances liver-stage infection by Plasmodium vivax

The liver is the first destination of malaria parasites in humans. After reaching the liver by the blood stream, Plasmodium sporozoites cross the liver sinusoid epithelium, enter and exit several hepatocytes, and eventually invade a final hepatocyte host cell. At present, the mechanism of hepatocyte invasion is only partially understood, presenting a key research gap with opportunities for the development of new therapeutics. Recently, human EphA2, a membrane-bound receptor tyrosine kinase, was implicated in hepatocyte infection by the human malaria parasite Plasmodium falciparum and the rodent parasite Plasmodium yoelii, but its role is not known for Plasmodium vivax, a major human parasite whose liver infection poses a specific challenge for malaria treatment and elimination. In this study, the role of EphA2 in P. vivax infection was investigated. It was found that surface expression of several recombinant fragments of EphA2 enhanced the parasite infection rate, thus establishing its role in P. vivax infection. Furthermore, a new permanent cell line (EphA2Extra-HC04) expressing the whole extracellular domain of EphA2 was generated. This cell line supports a higher rate of P. vivax infection and is a valuable tool for P. vivax liver-stage research.

www.nature.com/scientificreports/ for P. vivax. Here, the role of EphA2 in P. vivax hepatocyte infection was explored. Mammalian cell transfection was used to induce the expression of six different partial and full-length constructs of EphA2 on the surface of HC-04 17 . The susceptibility of these transfected cells to P. vivax infection was evaluated.

Results
Recombinant EphA2 constructs and expression in HC-04. EphA2 (Fig. 1) is a 976 amino acid-long single-transmembrane human protein present in the plasma membrane of cells. Its extracellular N-terminal portion consists of a conserved ligand-binding domain (LBD) followed by a cysteine-rich (CysRich) domain, and two fibronectin III (FN-III) repeats that are linked to the transmembrane domain (TMD). The cytoplasmic C-terminal portion following the TMD contains a kinase domain, a sterile α-motif domain (SAM), and a PDZbinding motif 18 . To investigate the role of EphA2 in P. vivax sporozoite infection of hepatocytes, we generated six recombinant EphA2 constructs ( Fig. 1), each having a distinct domain composition. These constructs were designed for surface expression using the pDisplay™ mammalian expression vector 19 . All constructs carry an N-terminal HA-tag, which enables immuno-fluorescence staining. Except for the full-length construct (HA-FL), all constructs are anchored to the plasma membrane by the TMD of the platelet-derived growth factor receptor (PDGFR) which is the default TMD of the pDisplay™ vector. The full-length construct, HA-FL, on the other hand, was generated by cloning the full-length gene (without the signal peptide) into the pDisplay™ plasmid with a stop codon before the PDGFR TMD coding sequence, resulting in a construct that is nearly identical to the endogenous EphA2. After transfection of the HC-04 cell line with these plasmids, the expression of each recombinant protein was confirmed by Western blot (Fig. 2a). Immuno-fluorescence assays (IFAs) against the extracellular HA-tag performed without cell-permeabilization showed a peripheral pattern of protein expression for all constructs, confirming their presence on the plasma membrane (Fig. 2b). The transfection efficiency was 20-30% as measured by flow cytometry (Supplementary Fig. S1).

Expression of recombinant EphA2 can enhance P. vivax infection of HC-04.
To determine the effect of each EphA2 construct on P. vivax infection, plasmids were freshly transfected into the HC-04 cell line 24 h before initiating the hepatocyte infection assay. Therefore, for each sporozoite batch, new transfection was performed. Ten different clinical isolates of P. vivax were tested, and the highest number of liver-stage parasites was observed in HC-04 transfected with the HA-Extra construct (Fig. 2c, Supplementary Table S1). This construct induced on average a twofold increase in the infection rate relative to the mock transfection control with the empty pDisplay™ plasmid. Two other constructs, HA-LBD/FN-III and HA-2XFN-III, also led to a significant but less pronounced increase. Lastly, cells transfected with the CysRich and LBD constructs did not differ in their susceptibility to P. vivax relative to the mock control. These results suggest that the extracellular domain of  www.nature.com/scientificreports/ EphA2 is functional for P. vivax infection without the endogenous TMD and the cytoplasmic portion, and that larger extracellular constructs appeared to have a more significant impact on sporozoite invasion. Interestingly, no enhancement was detected when the full-length construct (HA-FL) was used. This could be due to protein truncation or degradation, as indicated by the presence of multiple bands on Western blot (Fig. 2a).
Generation of a new transgenic HC-04 cell line with improved P. vivax susceptibility. The in vitro study of P. vivax liver-stage development is highly challenging due to several factors, from sporozo-  www.nature.com/scientificreports/ ite production, which requires a robust mosquito colony and fresh blood from P. vivax malaria patients, to the extremely low sporozoite infection rates in hepatic cell lines 17,20,21 . Given the enhancement of infection achieved by the transient transfection above, we sought to improve the HC-04 cell line for P. vivax research. Using a CRISPR/Cas9 system, we successfully inserted the HA-Extra construct into the genome of HC-04 at the AAVS1 integration site (Fig. 3a). Two clones of the transgenic cell line (EphA2Extra-HC04), 4D11 and 1C9, were obtained. Integration and protein expression were confirmed by PCR and Western blot, respectively (Fig. 3b). IFAs showed circumferential labeling of the HA-tag, confirming proper protein trafficking (Fig. 3c). Like in the transient transfection experiment (Fig. 2b), these EphA2Extra-HC04 transgenic clones offered approximately twofold higher infection rates for P. vivax sporozoites than the original HC-04 cell line (Fig. 4a, Supplementary  Table S2). Figure 4b shows representative parasites on day 4 post infection in EphA2Extra-HCO4 and original HC-04. The size distribution of day 4-parasites in the EphA2Extra-HCO4 clones and the original HC-04 were shown ( Fig. 4c, Supplementary Fig. S2). At this timepoint, the parasites were generally smaller than 15 µm and too early to be size-differentiated into growing schizonts and hypnozoites. IFAs also revealed nuclear division ( Supplementary Fig. S3). Together these data indicate that EphA2Extra-HC04 supports normal parasite development.

Discussion
The research on Plasmodium hepatocyte infection and development has been severely limited relative to the research on the blood stages. Plasmodium sporozoites recognize a few hepatocyte surface proteins during invasion. Among these proteins is EphA2, a receptor tyrosine kinase, first identified as a P. yoelii and P. falciparum receptor through antibody array screening 13 . In our study, we investigated and demonstrated a functional role of the extracellular domain of EphA2 in P. vivax liver-stage infection, using an in vitro culture system of the hepatic HC-04 cell line. Through transient transfection of plasmids encoding different EphA2 constructs, we could achieve 20-30% transfection efficiency. Despite the low efficiency, the full extracellular construct (HA-Extra), which contains LBD, CysRich, and two FN-III repeats, could enhance the P. vivax infection two-fold. Smaller constructs containing LBD and FNIII or two FNIII's could also promote infection, but to a smaller extent. In contrast, LBD or CysRich alone was not sufficient. Therefore, larger constructs of EphA2 tend to perform better. The reason for this is unclear, but it is possible that the larger constructs may increase the accessibility to the binding site due to its increased protrusion from the cell surface. It is also possible that there are more binding sites for the parasite ligand on the larger proteins. EphA2 is known to have multiple binding sites on different extracellular subdomains for its natural ligand EphrinA1 [22][23][24] . Of note, the full-length construct did not promote productive invasion. It is possible that the native EphA2 TMD or the cytoplasmic portion present in the full-length protein hinders the interaction of EphA2 and its parasite ligand, perhaps by affecting the quaternary structure of EphA2. The SAM domain in the C-terminal half of EphA2 has been shown to inhibit the protein's dimerization 25 . The molecular interaction between hepatocyte EphA2 and parasite ligand(s) awaits further investigation. Because the transient transfection of EphA2 extracellular domains could enhance P. vivax infection at 20-30% transfection efficiency, we thought that higher infection rates could be obtained through stable transfection. Therefore, we used the CRISPR/Cas9 technology to introduce the HA-Extra construct to the genome of HC-04. www.nature.com/scientificreports/  www.nature.com/scientificreports/ However, to our surprise, this new cell line (EphA2Extra-HC04) was still only two-fold more susceptible to P. vivax than the original HC-04. It is thus possible that because sporozoites are motile, they can search for a suitable host cell, in which case increasing the proportion of suitable cells by genome integration would not lead to a further enhancement. Additionally, some parasites may not depend strongly on EphA2 if there is an alternative invasion pathway utilizing a different receptor protein.
Although the twofold increase in P. vivax susceptibility of EphA2Extra-HC04 is modest, it is highly valuable for P. vivax exoerythrocytic stage research. The number of sporozoites has been the single most important limiting factor for research of P. vivax pre-erythrocytic stages. By reducing the number of sporozoites by half to achieve the same invasion rate, researchers can broaden compound screening for drug development and test more antibodies to identify new vaccine targets. It is possible that overexpressing EphA2 with other host receptors such as SR-BI 7 could further promote of P. vivax liver infection.
Our findings further suggest that P. vivax liver-stage development could be supported in vitro by EphA2Extra-HC04 cells. The size distribution of parasites in the engineered cell line was similar (clone 4D11) or greater (clone 1C9) to that of the parasites in the original HC-04 cell. There was also evidence of parasite nuclear division. However, due to the small size of the parasite at the early timepoint of parasite examination (4 days after inoculation), we were not able to reliably differentiate the growing schizonts from the dormant hypnozoites. Therefore, it remains unclear whether the introduction of EphA2-Extra would perturb the growing-schizont: hypnozoite ratio.
In summary, this study provides the first evidence supporting the role of EphA2 in P. vivax infection of hepatocytes. It suggests that EphA2 may be a common hepatocyte receptor across rodent and human malaria. The study also established a new cell line that supports a superior P. vivax infection rate. This new cell line is valuable for vaccine development, antimalarial compound screening, and fundamental research into parasite biology. Female Anopheles dirus were fed on P. vivax blood through membrane feeding as previously described 26 . Blood was collected from P. vivax patients during October, 2017-February, 2020. Each week a new batch of mosquitoes were produced for the experiments. Only An. dirus that fed on P. vivax mono-infection blood was used in the study. P. vivax sporozoites were harvested from salivary glands of the infected An. dirus on day 14-21 post feeding. Briefly, the mosquitoes were subjected to surface cleansing by 4-step sequential dipping: (i) 70% (v/v) ethanol, (ii) 10% (v/v) penicillin/streptomycin in sterile water, (iii) 25 μg/mL fungizone in sterile water, and (iv) RPMI1640 incomplete medium containing 1% (v/v) Pen/Strep, pH 8.0 (dissecting medium). The mosquito salivary glands were harvested under a stereomicroscope and kept in a 1.5 mL microcentrifuge tube containing 50 µL of cold dissecting medium. The salivary glands were washed once with dissecting medium and centrifuged at 13,000×g for 30 s. The salivary glands were resuspended in 50 µL of dissecting medium and the sporozoites were released by using a sterile pestle. The sporozoites were counted using a hemocytometer and diluted to 5 × 10 5 sporozoites/mL. The culture supernatant of the transfected HC-04 was replaced with 100 µL of sporozoite suspension which yielded the final of 5 × 10 4 sporozoites/well. The culture was maintained at 37 °C under 5% CO 2 with daily change of culture medium. The parasite culture was harvested on day 4 post infection by fixing with 4% (w/v) paraformaldehyde for 30 min at the room temperature. The immunofluorescence staining was performed for quantification of liver-stage parasites.   For imaging, localization of recombinant EphA2 fragments and transgenic EphA2 in HC04 cells as well as liver-stage parasites were acquired using a Zeiss LSM-700 laser scanning confocal microscope. The numbers of liver-stage parasites were quantified using an Olympus IX73 inverted microscope as determined by UIS4 and DAPI.

Data availability
The sequence of human EphA2 is publicly available on the NCBI database (Accession Number NM_004431.5, https:// www. ncbi. nlm. nih. gov/ nucco re/ NM_ 004431.5). The liver-stage infection datasets generated and/or analyzed during the current study are available upon request and will be provided by the corresponding author W.R. www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.