Globally, the World Health Organization estimates that, in 2019, 229 million clinical cases of malaria occurred and 409,000 people died of malaria (Varo et al., 2020; World Health Organization, 2020). Although tertian ague caused by Plasmodium vivax was once thought to be benign when compared with falciparum malaria induced by Plasmodium falciparum, recent reports found that, in many co-endemic malarious areas, the most commonly transmitted malaria parasite after P. falciparum parasite is P. vivax, and about a quarter of the cases in 2019 were due to P. vivax, mainly in Afghanistan and Pakistan (World Health Organization, 2020). P. vivax parasitemia is associated with substantial morbidity including a cumulative risk of severe anemia, hypohepatia, acute lung injury, hypersplenism, splenomegaly, and acute renal failure, whereas its pathogenesis is still unclear (Douglas et al., 2013; Commons et al., 2019). Among these clinical symptoms, splenomegaly is one of the most common features of malaria, and splenic rupture is more prevalent among infections caused by P. vivax compared with other species (Elizalde-Torrent et al., 2018). It has been found that plasmodium parasite evades clearance of host immune system in the erythrocytic stage to cause malaria symptoms (Bassat and Alonso, 2011; Douglas et al., 2013; Im et al., 2017; Commons et al., 2019). Clinical study has found that a large number of P. vivax—infected erythrocytes (Pv-iRBCs) were located in the red pulp of spleen, indicating that P. vivax could evade the immune clearance of spleen (Machado Siqueira et al., 2012). However, the precise mechanism is still unknown.

During the research of P. falciparum, it was found that the variant surface antigens of P. falciparum play an essential role in its immune evasion. For instance, as a member of the variant surface antigens, P. falciparum erythrocyte membrane protein-1 (PfEMP1) located at the knob on the infected erythrocyte surface is a ligand of several hose cell receptors, such as platelet endothelial cell adhesion molecule 1 (CD31) on the endotheliocyte surface, complement receptor type 1 on the erythrocyte surface, and intercellular adhesion molecule-1 (ICAM-1) (Chen et al., 2000). Therefore, P. falciparum—infected erythrocytes (Pf-iRBCs) could bind to uninfected erythrocytes to form rosettes and further bind to host vascular adhesins. Besides, increased rigidity and lowly deformability of Pf-iRBC further allow them to evade the host immune attack by adhering to the blood vessel endothelium instead of entering the spleen (Chen et al., 2000; Turner et al., 2013). However, Pv-iRBC cytoadhesion levels are 10-fold lower than those observed for Pf-iRBC, and Pv-iRBCs are still deformable, allowing them enter the spleen (Bozdech et al., 2008; Carvalho et al., 2010), indicating that the immune evasion mechanism of P. vivax might be different from that of P. falciparum.

Human splenic fibroblasts (HSF), also named as barrier cells, are deployed as diverse barriers in the splenic pulp. When the spleen is attacked by pathogens, these cells rapidly and greatly increase in number to form splenic filtration beds that are important in the increased capacities of spleen in clearance, phagocytosis, and immunological reactivity (Weiss, 1991). A non-lethal murine model of plasmodium, P. yoelii strain 17X, can induce spleen-resident fibroblasts to form physical barrier cells, causing the “open” circulation of the spleen to suddenly and temporarily change into a “closed” circulation (Weiss, 1989). Because of similarities between P. yoelii strain 17X infections in BALB/c mice and P. vivax infections, a hypothesis holds that Pv-iRBCs induce spleen structural remodeling to form barrier cells and, subsequently, selectively undergo cytoadherence to escape spleen clearance (del Portillo et al., 2004). However, this hypothesis has not been proved.

In present work, we found that P. vivax surface-related antigen (PvSRA), an exported antigen from P. vivax to the surface of the infected erythrocyte membrane, could bind with HSF. Considering the important role of surface antigen in the erythrocytic stage (Chen et al., 2000), together with the above hypothesis, we suspected that PvSRA might be involved in P. vivax immune evasion in the spleen. To investigate this speculation, RNA sequencing, protein microarray, and bioinformatics analysis technologies were applied to predict the potential function of PvSRA, and in vitro validations were further performed. The results indicated that recombinant PvSRA attracted HSF migration and interacted with HSF by targeting integrin β1 (ITGB1) along with some changes in HSF cell performance. This study indicated that PvSRA might indeed participate in the immune evasion of P. vivax in the spleen by changing HSF function through PvSRA–ITGB1 axis. This finding will cast new light on the research of the pathogenesis of P. vivax.

P. vivax is one of the most widely distributed human malaria parasites, and the risk of acute P. vivax attacks after splenectomy is increased, which even outweighs the risk of P. falciparum attack (Kho et al., 2019). However, mechanisms that underline immune evasion in P. vivax infection remain unclear. A hypothesis holds that Pv-iRBC can selectively cytoadhere to splenic fibroblasts and cause spleen structural remodeling and resulting in immune evasion from spleen clearance (del Portillo et al., 2004). In the present study, we provide evidence that PvSRA, a newly found exported protein from P. vivax to the surface of infected erythrocyte membrane, could interact with HSF and induce cell performances changes of HSF through interaction with ITGB1 on HSF cell surface.

As a blood filter, the spleen is mainly composed of the red pulp, white pulp, and their boundary called the marginal zone. Under normal circumstances, abnormal or aged erythrocytes are typically removed by the spleen through phagocytosis or antigen-specific immune response. However, it was found that large numbers of Pv-iRBC in the red pulp (Machado Siqueira et al., 2012), indicating that P. vivax could pass through endothelial slits, enter the spleen, and avoid the immune clearance of spleen (Handayani et al., 2009; Machado Siqueira et al., 2012). The rodent malaria P. yoelii has been extensively used to study molecular aspects of virulence in malaria, such as pre-erythrocytic stages of infection and pathogenesis of erythrocytic stages, mainly because of the existence of strains with different cellular tropisms, growth curves, and clinical outcomes, such as the reticulocyte-prone non-lethal P. yoelii 17X strain and the normocyte-prone lethal P. yoelii 17XL strain. Among these two strains, P. yoelii strain 17X infections in mouse spleen are similar to P. vivax infections in human spleen. Researchers used the Balb/c P. yoelii rodent malaria model to study the function and structure of the spleen, and they noticed lower numbers of parasites within macrophages in infections with 17X than in infections with 17XL. However, because these numbers were observed together with larger numbers of parasites in spleens infected with 17X and no differences were observed in either in vitro or in vivo phagocytic activity between the two infections, it can be assumed that the differences in the numbers of parasites within the macrophages are not due to an increased ability of macrophages to kill parasites during non-lethal infections. This finding provides further evidence of different spleen compartmentalization in animals infected with non-lethal strains (Martin-Jaular et al., 2011). Besides, it was found that non-lethal P. yoelii 17X strain induces remodeling of the spleen through the formation of a spleen tissue barrier of fibroblastic origin, causing the “open” circulation of the spleen to suddenly and temporarily change into a “closed” circulation (Weiss, 1989). Such barrier facilitates the channeling of blood from arterioles to venules and the adherence of infected reticulocytes to it thus physically protecting them from destruction by macrophages (Martin-Jaular et al., 2011). Because of similarities between P. yoelii strain 17X infections in BALB/c mice and P. vivax infections, a hypothesis holds that Pv-iRBC induces a spleen blood barrier of fibroblastic origin to which infected reticulocytes adhere in facilitating macrophage clearance escape (del Portillo et al., 2004).

HSF, which is the fibroblastic origin mentioned above, also known as splenic fibroblastic reticular cells, significantly contribute to the immunity of the organism because they augment the reticular cell fiber filtration beds of spleens when stressed by infection, imperfect erythrocytes, or immunological deficiency. Thus, they play essential roles in splenic clearance of the blood (Weiss, 1991). In addition, HSF provide a three-dimensional (3D) microenvironment and set up a niche for lymphoid cells, including T, B, and dendritic cells, through the expression of chemoattractants. This niche not only supports the onset and perpetuation of immune responses but also activates, enhances, and even inhibits immune responses (Mueller et al., 2007; Lukacs-Kornek et al., 2011). A recent study showed that plasma-derived extracellular vesicles from P. vivax patients are preferentially uptaken by HSF as compared with the uptake of extracellular vesicles from healthy individuals. After this uptake, Pv-iRBC has shown specific adhesion properties to these HSF (Toda et al., 2020). For extracellular vesicles that act as shuttle vectors or signal transducers that can deliver specific biological information and have progressively emerged as key regulators of organized communities of cells within multicellular organisms in health and disease (Simeone et al., 2020), we suspected that the cross-talk between Pv-iRBC and HSF might contribute to the immune evasion of P. vivax.

It is well known that variant surface antigens of P. falciparum, such as PfEMP1, play essential role during its immune evasion process. For instance, it mediates binding of infected erythrocytes to uninfected erythrocytes, and it is thought that P. falciparum evades the host immune system by shielding the infected erythrocytes and the newly released merozoites from host invasion-inhibitory antibodies in this way (Yam et al., 2017). Besides, PfEMP1 have also been found to interact with a variety of cell surface receptors on the endothelial cells, resulting in the adhesion of Pf-iRBC to endothelium (Yam et al., 2017). In addition, PfEMP1 is anchored at the erythrocytes membrane skeleton by knobs, macromolecular complexes that consist of knob-associated histidine-rich protein (KAHRP), whereas KAHRP binding with the erythrocyte membrane skeleton leads to an increased rigidity and further decreased the deformability of Pf-iRBC (Maier et al., 2008). Therefore, unlike normal erythrocytes that could pass through blood vessel wall into the spleen circulation through their deformability, Pf-iRBCs that adhered to endothelium are unable to enter the spleen, allowing P. falciparum to avoid immune clearance by the spleen (Chen et al., 2000; Turner et al., 2013). These findings showed that the exported antigen from plasmodium on infected erythrocyte membrane surface might play important role during immune evasion.

PvSRA, predicted as an extracellular protein, is a newly found erythrocytic stage antigen identified in our study, and it was confirmed as a merozoite surface protein that could be exported to the surface of infected erythrocyte membrane. Besides, in vitro cell experiment revealed that PvSRA could bind with HSF, indicating that PvSRA might mediate the binding of Pv-iRBC to the splenic fibroblasts in spleen. Considering the abovementioned P. vivax immune evasion hypothesis and the important role of exported antigen from plasmodium in the erythrocytic stage (Chen et al., 2000), we suspected that PvSRA might be involved in P. vivax immune evasion in the spleen. Because RNA-seq is a high-throughput technology used to provide a comprehensive view of entire transcriptome, and it enables a wide range of applications such as the discovery of novel genes, gene/transcript quantification, and differential expression and functional analysis (Wang et al., 2009; Wang et al., 2019), this technology was applied to test our speculation. The results showed that the expression of 639 of 24,647 genes was significantly regulated by PvSRA. GO and KEGG pathway enrichment analyses manifested that these DEGs are involved in various terms and signaling pathways. Interestingly, focal adhesion, extracellular matrix, and actin cytoskeleton related terms or pathway were founded in the enrichment results. Among them, focal adhesion, a specialized structure formed at the cell–extracellular matrix contacts points, where bundles of actin filaments are anchored to transmembrane receptors of the integrin family through a multi-molecular complex of junctional plaque proteins. Focal adhesion structurally links the cytoskeleton to the extracellular matrix and transmits mechanical signals (Mitra et al., 2005; Burkin et al., 2013). Thus, it not only plays an essential role during cell motility through the regulation of the actin cytoskeleton but also is involved in the regulation of other cell functions, such as cell proliferation and survival, through signal transmission (Mitra et al., 2005; Burkin et al., 2013). Extracellular matrix is a non-cellular 3D macromolecular network composed of collagens, proteoglycans/glycosaminoglycans, elastin, fibronectin, laminins, and several other glycoproteins. These components bind each other as and cell adhesion receptors forming a complex network into which cells reside in all tissues and organs. Cell surface receptors transduce signals into cells from extracellular matrix, which regulate diverse cellular functions, such as survival, growth, migration, and differentiation, and are vital for maintaining normal homeostasis, whereas the progression of several pathologic conditions is associated with the negative deregulation of extracellular matrix composition and structure (Theocharis et al., 2015). The actin cytoskeleton, a collection of actin filaments with their accessory and regulatory proteins, is the primary force-generating machinery in the cell. It was found that its generated force plays essential role in cell migration, cell shape, and mechanical properties of the cell surface, drive the intracellular motility and morphogenesis of membrane organelles, and allow cells to form adhesions with each other and with the extracellular matrix (Svitkina, 2018). Therefore, the changes of these three aspects of HSF seemed to agree with the hypothesis that P. vivax immune evasion in the spleen is mediated by affecting HSF.

It is well known that proteins rarely act alone as their functions tend to be regulated. Many molecular processes are carried out by molecular machines that are built from many protein components organized by their PPIs; however, abnormal or aberrant PPIs could cause diseases (Ryan and Matthews, 2005). Therefore, analyzing PPIs can help identify novel pathways to gain basic knowledge of the physiological activity or pathological action. Moreover, it can also help find new therapeutic targets of diseases and predict the potential function of newly found or unknown proteins (Berger and Iyengar, 2009). Protein microarray is a high-throughput method used not only to track interactions and activities of proteins but also to determine proteins function. Therefore, this technology was applied in the present work to search for substrates of PvSRA and explore the potential function of PvSRA through PPIs analysis. The results showed that substrates of PvSRA were not only distributed in different regions of the cell, such as membrane, cytoplasm, and nucleus, but also located at focal adhesion sites and extracellular vesicular exosome, indirectly indicating that PvSRA is capable of binding with or entering into some host cells through PPIs. To determine whether PvSRA induced HSF gene expression changes through protein interaction with some unknown substrates or receptors on the HSF membrane surface, we mainly focused on the PvSRA substrates located on the cell membrane surface in the present work. Among the surface receptors of PvSRA, ITGB1 was observed for its unique properties during BPs. Integrin family members are membrane receptors involved in cell adhesion and recognition in various processes including embryogenesis, hemostasis, tissue repair, and immune response (Wu et al., 2010). They link the actin cytoskeleton with the extracellular matrix and transmit signals bidirectionally between the extracellular matrix and cytoplasmic domains (Nix and Beckerle, 1997). One of the members of the integrin family, ITGB1, has several functions, including actin binding, cadherin binding, cell adhesion molecule binding, collagen binding involved in cell-matrix adhesion, fibronectin binding, laminin binding, protease binding, and protein tyrosine kinase binding (Mueller et al., 1999; Moog-Lutz et al., 2003; Cerecedo et al., 2006; Guo et al., 2014; Sipila et al., 2014; Zhao et al., 2016). Hence, it appears to be an important receptor of cellular adhesion molecules because it mediates adhesion between cell and cell or cell and extracellular matrix. In addition, it is primarily responsible for targeting integrin dimers to the appropriate subcellular locations. Function analysis and literature review showed that ITGB1 also participates in extracellular matrix organization, protein localization, cell cycle, cell proliferation and apoptosis, cell migration, and other BPs (Mueller et al., 1999; Miles et al., 2008; Juric et al., 2009; Zhao et al., 2016). In the current study, we confirmed that ITGB1 was expressed on HSF cell surface, indicating that Pv-iRBCs can bind with HSF through PvSRA–ITGB1 interaction and might further influenced HSF cell performance.

ITGB1-regulated subcellular locations in adhesive cells mainly include focal adhesions. In addition, the results of enrichment analysis showed that ITGB1 was involved in several microbial infections as a receptor for viruses, bacteria, and parasites. During these infections, ITGB1 mainly acts as a receptor of microbial composition or a constituent of focal adhesion structure to participate in the invasion process (Figueira et al., 2015; Olafsson et al., 2019). Interestingly, the RNA-seq results showed that the expression of genes involved in the extracellular matrix, focal adhesion, and actin cytoskeleton was significantly changed in the PvSRA-treated HSF group. This could demonstrate that Pv-iRBCs have the potential to induce the formation of focal adhesion structure on the HSF surface through PvSRA–ITGB1 interaction, followed by actin cytoskeleton remodeling and the alteration of HSF function, finally resulting in the escape from spleen clearance.

On the above basis of the bioinformation from RNA-seq and PPIs analyses, in vitro cell-based validations were further performed to verify the above speculation. The results demonstrated that PvSRA could bind to the HSF surface and attract HSF migration, indicating that Pv-iRBC could adhere to HSF after they enter the spleen. As a kinase that localizes to focal adhesion and regulates signaling in focal adhesion, FAK communicates signals between integrins and intracellular proteins and regulates diverse cellular processes such as cell proliferation, migration, actin cytoskeleton remodeling, and invasion (Rosen and Dube, 2006; Chen J.-S. et al., 2010; Schaller, 2010; Roa-Espitia et al., 2016). We observed that the expression of FAK was significantly enhanced by PvSRA treatment and, the MAPK signal pathway, a downstream signaling event of focal adhesion (Mitra et al., 2005), was also activated, indicating that PvSRA could induce focal adhesion formation on the HSF surface, followed by downstream signal activating. Besides, the results showed that PvSRA treatment induced HSF actin cytoskeleton remodeling, indicating that Pv-iRBC has the potential to alter the original cellular structure of HSF. We further tested protein expression of the representative DEGs involved in focal adhesion, extracellular matrix, and actin cytoskeleton of PvSRA-treated HSF. Among these genes, MMP1 is a prototypical MMP, primarily functioning to degrade collagen types 1 and 3 (Pardo and Selman, 2006). CPA3 is a member of the family of zinc metalloproteases, which are involved in the degradation of endogenous proteins (Roy et al., 2014). TFPI2 may play a role in the regulation of plasmin-mediated matrix remodeling (Du et al., 2003). COMP, an extracellular secreted glycoprotein, may play a role in the structural integrity via its interaction with other extracellular matrix proteins such as collagens and fibronectin (Haudenschild et al., 2011). ITGA8 is a cell surface protein that belongs to the alpha integrin family of transmembrane cell surface receptors, and it requires the formation of a heterodimer with ITGB1 for receptor function (Hung et al., 2018). MYH3 belongs to a group of proteins called myosin, which are involved in cell movement and transport of materials within and between cells (Li et al., 2016). The results from cell-based assays showed that, after PvSRA treatment, the expression of MMP1 and TFPI2 was obviously upregulated, whereas that of MYH3 and COMP was decreased and was consistent with the results obtained from RNA-seq. However, no changes were observed in the expression of CPA3 and ITGA8, indicating that the expression levels of these two genes were inconsistent with their protein expression levels. We also found that PvSRA has a significantly negative effect on the HSF cycle and protein expression of cell cycle–related gene Cyclin D1, which might further result in proliferation inhibition. These negative effect on HSF proliferation indicated that Pv-iRBC might inhibit the proliferation of the barrier cells after they enter the spleen. Besides, for HSF that mainly play its immune role through increasing in number to form splenic filtration beds when spleen is attacked by pathogens (Weiss, 1991), the present finding also indicated that Pv-iRBC might decrease the capacities of the spleen through inhibiting HSF proliferation. In present work, we also found that PvSRA treatment increased PD-L1 expression on the HSF surface. PD-L1 is a protein that allows some cells to escape the attack by the immune system. PD-L1 interacts with PD-1 on some important immune system cells such as T cells. It is acknowledged that PD-1/PD-L1 axis plays a crucial role in the progression of tumor by altering status of immune surveillance (Yi et al., 2018). Except for tumor progression, PD-L1 expressed on infected host cell also plays an essential role during extraneous pathogens invasion or infection process. For example, research has shown that lymphocytic choriomeningitis virus infection targeting of mouse splenic fibroblasts contributes to its chronic infection, because the upregulation expression of PD-L1 on infected splenic fibroblasts prevents severe immunopathology and architectural disruption of splenic fibroblast network in the spleen (Mueller et al., 2007). Because Pv-iRBC also has the potential to target HSF, we suspected that Pv-iRBC might also evade host immune attack in a similar way. These findings revealed that Pv-iRBC has the potential to decrease the immunological capacities of spleen through targeting HSF. However, the results of these in vitro experiments showed that the effect of PvSRA on HSF could be inhibited or weakened by blocking ITGB1 receptor on the HSF surface, demonstrating that ITGB1 is indispensable during PvSRA–HSF interaction. On the basis of above findings and in consideration of ITGB1 that was found to be involved in several microbial infections as a receptor for viruses, bacteria, and parasites (Figueira et al., 2015; Olafsson et al., 2019), we suspected that ITGB1 might also be invloved in P. vivax immune evasion in the spleen.

In addition to the bioinformation from RNA-seq and PPIs analyses, the results of in vitro assays contributed in improving the prediction of immune evasion process of P. vivax, which could be summarized as follows: after entering the spleen, Pv-iRBCs adhere to HSF through the ligand (PvSRA)–receptor (ITGB1) axis, followed by the formation of the focal adhesion structure on the HSF surface. Given that some of the constituents of focal adhesion participate in the structural link between membrane receptors and the actin cytoskeleton (Mitra et al., 2005; Burkin et al., 2013), the original shape and motility of HSF were affected and changed, which might induce the formation of barrier cells. Moreover, because some constituents of focal adhesion are signaling molecules (i.e., different protein kinases and phosphatases, substrates, and various adapter proteins) (Mitra et al., 2005; Burkin et al., 2013), focal adhesion might further modulate other capabilities of HSF, such as cell proliferation and extracellular matrix, then activating its downstream pathway (i.e., MAPK pathway). In addition, PvSRA might also have the possibility to enter into the intracellular or endonuclear environment through interacting with other receptors and then further affects or interferes with the signal transduction or genetic information processing of HSF through other PPIs. This would finally alter the immunoregulation role of HSF in the spleen to result in the immune evasion of P. vivax (Figure 8).

However, because the present study is mainly based on previous hypotheses and RNA-seq prediction and the prediction was only validated roughly by in vitro assays, there are some research limitations in present work. Because P. vivax parasites cannot be maintained reliably in continuous cultures in vitro, biology research on P. vivax faces different challenges. Therefore, some findings or predictions in present work mainly obtained through in vitro protein molecular cell experiments, lacking the verification through intercellular interaction experiments between Pv-iRBC and HSF. The recent development allows to purify P. vivax parasites in the ring stage and to develop them into late stages (Roobsoong et al., 2015; Prajapati et al., 2019). This will help address our future questions related to surface expression or discharge of PvSRA and to demonstrate a direct binding between Pv-iRBCs and HSF. Besides, because acetone could penetrate cell membranes, PvSRA location on Pv-iRBC membrane surface needed further verification by co-staining with erythrocyte surface proteins (e.g., Band 3 or Glycophorin A) without permeabilizing cells (Auffray et al., 2001). In addition, present research on P. vivax immune evasion mechanism is not deep enough; both the high-throughput analysis and in vitro validation were superficial. For instance, the present work mainly focused on focal adhesion, extracellular matrix, and actin cytoskeleton changes of HSF, and only partial related genes and pathway were verified in vitro; detailed relationship between signal pathway and cell performance changes has not been yet evaluated; the possibility that PvSRA might enter into HSF also needed further verification, and the prerequisite of this speculation is to make sure whether PvSRA could discharge or secrete from Pv-iRBC through testing the presence of PvSRA in the culture medium with ELISA or some other methods when culturing the P. vivax parasites in vitro. Furthermore, because there have been few related studies about relevant research, no appropriate control samples were included in present work. Previous studies have found that the cytoadhesion of Pv-iRBCs was, in part, mediated by VIR proteins localized at the surface of Pv-iRBCs (Fernandez-Becerra et al., 2009; Carvalho et al., 2010), and a recent research reveals that VIR14 protein (PlasmoDB ID: PVX_108770) mediates specific adhesion to HSF expressing ICAM-1 (Fernandez-Becerra et al., 2020), indicating that these VIR proteins could be taken as control samples in the following studies. Overall, the present study was mainly a predictive work, and the findings may provide a basis for further research in this field.