Background

Malaria is a devastating infectious disease, which imposes an immense worldwide health burden [1]. Among the malaria pathogens, Plasmodium vivax has the largest geographical distribution and the greatest economic impact [2,3,4]. Basic research on P. vivax has been impeded by the lack of a continuous in vitro culture system, unlike the well-established continuous in vitro culture system for Plasmodium falciparum [5, 6]. Although P. vivax has been historically described as “benign” malaria, P. vivax malaria is increasingly recognized as a cause of severe morbidity and mortality [7, 8]. Individuals infected with P. vivax develop various clinical symptoms, such as fever, anemia, splenomegaly, or severe malaria [9]. However, the pathogenesis of severe vivax syndromes remains poorly understood and requires further research [10].

The spleen, which is the largest immune organ, is mainly involved in filtering the blood, capturing pathogens, and activating the adaptive immune response [11]. Plasmodium vivax infection can lead to significant changes in the spleen, which cause asymptomatic swelling or complications such as rupture and hypersplenism [12, 13]. The splenic fibroblasts (SFs) are essential for maintaining the structure and immune function of the spleen [14]; these are found extensively in connective tissue and secrete collagen I into the extracellular matrix (ECM) [15, 16]. Collagen I is the main component of ECM, which is critical for creating the microenvironment needed for immune response development [17, 18]. Furthermore, the collagen network contributes to the proper functioning of the immune system of the spleen [19]. In the spleen, SFs secrete and remain associated with collagen I; they combine with argyophilic reticular fibers, which strengthen the filtration beds [20]. The central function of the spleen—selective clearance of cells, microbes, and other particles from the blood—depends upon these filtration beds. Such functions of the spleen as phagocytosis and immunologic reactivity derive from clearance capacity of filtration beds [21]. In rodents infected with malaria parasites, iRBCs circulate in the bloodstream and pass through the spleen filtration bed, where they are diverted from arterioles to venous sinuses and eventually cleared by immune cells in the filtration bed [19], while the specific circulation and clearance mechanism of iRBCs in the human spleen remains unclear. Upon infection with Plasmodium, SFs are aberrantly activated and then evolve into barrier cells that interact with fibronectin and collagen I to form the blood-spleen barrier [22, 23]. The barrier cells protect RBCs from destruction by parasites but allow iRBCs to adhere and realize immune escape, which can sometimes lead to an imbalance in the immune response and even to severe disease [24, 25]. Thus, it is important to understand the changes in SFs and collagen I during Plasmodium infection.

The interaction between the malaria parasite and host cells mediated through ligand-receptor binding, especially involving malaria parasite export proteins, plays essential roles [26,27,28]. Malaria parasite export proteins are transported to the membrane surface of infected erythrocytes, which enhances the adhesion of infected erythrocytes to host cells and facilitates immune evasion by the parasite [29]. Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1) is expressed on the surface of mature infected erythrocytes and binds to endothelial cells, which allows evasion from circulation and destruction in the spleen [30, 31]. The VIR14 protein of the P. vivax vir gene family can be transported to the surface of reticulocytes and adheres to splenic fibroblasts (SFs) through the ICAM-1 receptor [23].

Similarly, the protein characterized by P. vivax tryptophan-rich antigens (PvTRAgs), as export protein, can be delivered to the surface of the erythrocyte membrane to bind with host cells [32, 33]. Recently, PvTRAg5 has been found to bind to erythrocyte receptors basigin and band 3 to facilitate parasite growth [34]. PvTRAg23 was found to interact with human SFs (HSFs) and reduce collagen I levels [35]. The reduced collagen I levels may potentially affect the hemofiltration function of the spleen, which facilitates parasite evasion. However, determining whether all PvTRAgs participate in changes in collagen I is interesting because of the elusive precise functions of PvTRAgs. Notably, there are 36 members of the PvTRAg family in the P. vivax reference genome Sal-I and 40 members of the PvTRAg family in the P. vivax reference genome PvP01, PvPAM, and PvW1.

This study aimed to identify the interactions between the PvTRAgs and SFs and assess their effect on collagen I levels in HSFs. Here, the activation of NF-κBp65 pathway results in the upregulation of inflammatory factor expression in response to PvTRAgs stimulation, which induces a decrease in collagen I levels. Our study expands the landscape of the functional roles of PvTRAgs and raises interesting hypotheses for the broader Plasmodium TRAg gene family.

Methods

Bioinformatics analysis

To acquire the information of pvtrag gene structure, the genes encoding pvtrag were retrieved from PlasmoDB (https://plasmodb.org/, PlasmoDB ID in Additional file 1: Table S1). The Simple Modular Architecture Research Tool (SMART) (http://smart.embl-heidelberg.de/) and the GPI Fungal Prediction Server (https://mendel.imp.ac.at/gpi/cgi-bin/gpi_pred_fungi.cgi) were used for the prediction of signal peptides and glycophosphatidylinositol-anchored (GPI-anchored) proteins.

PCR amplifcation of pvtrag

To amplify 36 pvtrag genes, genomic DNA of the P. vivax Sal-I strain was extracted. The primers used for amplification are described in Additional file 2: Table S1, and the genomic DNA was extracted as previously described [36]. The PCR reactions were conducted in a 10-μl reaction volume consisting of 1 μl genomic DNA as the template, 0.5 μl of each primer (5 μM), 2 μl 5 × TransStart® FastPfu Bufer (TransGen Biotech Co., Ltd., Beijing, China), 0.8 μl dNTPs (2.5 mM), 0.25 μl FastPfu DNA Polymerase (TransGen Biotech Co., Ltd.), and 4.95 μl nuclease-free water. The PCR amplification was performed in a Mastercycler (Eppendorf, Hamburg, Germany) under the following program: denaturation at 95 ℃ for 2 min, followed by 35 cycles of 95 ℃ for 20 s, 50 ℃ for 30 s, and 72 ℃ for 1 min, and a final extension at 72 ℃ for 5 min. The sizes of the PCR products were estimated using the (Trans2K Plus DNA marker (TransGen Biotech Co., Ltd.). All PCR products were sequenced by YiXin Biotechnology (Shanghai, China).

Expression and purification of recombinant proteins

To obtain recombinant PvTRAg proteins, Escherichia coli BL21 (DE3) pLysS cells (TransGen Biotechnology, Beijing, China) were used to express the corresponding proteins. Briefly, the pvtrag genes were amplified by PCR and cloned into the pET30 vector (Tianlin Bio, Wuxi). This vector adds six-histidine tags at the N- and C-terminal ends, which enables easier purification and immunodetection using monoclonal antibodies against the six-histidine tag. Next, the recombinant plasmids were transformed into E. coli BL21 (DE3) pLysS cells; 200 ml Luria-Bertani (LB) medium supplemented with 100 mg/ml ampicillin was inoculated with a transformed bacterial pre-culture and shaken at 37 ℃ until the cell density reached an OD600 of approximately 0.6–0.8; protein expression was induced with 0.1 mM isopropyl-β-D-thiogalactoside (IPTG) at 16 ℃ for 18–20 h [37]. Then, culture effluent was collected, and cells were harvested by centrifugation at 300 × g for 30 min at 4 ℃. The rPvTRAg proteins were purified under nondenaturing conditions by a biotechnology company (Youlong Bio, Shanghai, China) using a Ni agarose column. PvTRAgs were mixed with the protein loading buffer, boiled at 100 ℃ for 5 min, and separated in 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels. Next, they were stained with Coomassie Brilliant Blue (R250). For immunoblotting, PvTRAgs were detected using horseradish peroxide (HRP)-coupled anti-His antibody (1:5000, Abcam, Cambridge, MA, USA).

Cell culture

HSFs (ScienCell, CA, USA) were cultured in fibroblast medium (ScienCell, USA) containing 1% fibroblast growth supplement, 2% fetal bovine serum, and 1% antibiotic solution (P/S) at 37 ℃ in a humidified incubator containing 5% CO2 [38]. Prior to culturing HSFs, dishes need to be coated with polylysine (ScienCell, USA) for 2 h at room temperature to increase cell adhesion. The plates were rinsed twice with sterilized ultrapure water, and 8 ml complete medium was added to culture the cells. For cell passaging, a 95% fusion of cells is required.

Western blot analysis

HSFs were plated in 12-well culture plates and incubated at 37 ℃ in a cell incubator. When the cell density was about 70%, medium was removed and fresh medium containing PvTRAgs was added.

To analyze the binding capacity of PvTRAgs to HSFs, PvTRAgs were co-incubated with HSFs for 4 h. To assess the effects of PvTRAgs on collagen I synthesis, PvTRAgs were co-incubated with HSFs for 48 h. To examine the activation of cellular signaling, HSFs were treated with PvTRAgs for 0.5 h. The medium was discarded, and cells were washed once with ice-cold PBS after the experimental time period. Cells were then collected and resuspended in cold RIPA lysis buffer (Beyotime, China) containing protease and phosphatase inhibitors. Protein concentration was determined by BCA assay (Beyotime, China). Then, total protein (10–20 µg) was mixed with protein loading buffer, and all samples were boiled and separated on 10% SDS-PAGE. The separated proteins were transferred to polyvinylidene difluoride membranes. The membranes were blocked with closure buffer (1 × TBST, 5% milk powder) for 2 h at room temperature. They were then incubated with anti-His (1:1000, Abcam, Cambridge, MA, USA), anti-collagen I (1:1000, Abcam), anti-NF-κBp65 (1:1000, CST, Danvers, MA, USA), anti-phospho-NF-κBp65 (1:1000, CST), anti-FAK (1:1000, CST), anti-phospho-FAK (1:1000, CST), anti-p38 MAPK (1:1000, CST), anti-phospho-p38 MAPK (1:1000, CST), or anti-GAPDH (1:1000, Abcam) at 4 ℃ overnight. After washing with TBST, the membrane was incubated for 1 h with secondary antibody conjugated to HRP (1:5000, Abcam, Cambridge, USA). Bands were visualized by an ECL detection kit (XINSAIMEI, Suzhou, China), and protein expression was quantified with Image J (National Institutes of Health, MD, USA). Analysis of WB results was conducted using Image J software.

RNA extraction and quantitative real-time PCR

To examine the levels of cellular inflammatory factor mRNA, qPCR was performed as previously described [35]. PvTRAgs were co-incubated with HSFs after 48 h total RNA had been isolated from HSFs in accordance with the protocols of the manufacturer (Yeasen Biotechnology, Shanghai, China). The extracted total RNA was further processed to remove genomic DNA and then reverse transcribed into complementary DNA (cDNA) using 4 × HiFiScript RTMaster Mix, following the instructions for the RNA Reverse Transcription Kit (Yeasen Biotechnology, Shanghai, China). RT-PCR was performed in triplicate with three independent samples for each experimental group in a LightCycler480 II apparatus (Roche, USA) with SYBR Green Master Mix (No Rox) (Yeasen Biotechnology, Shanghai, China). The amplification program followed a two-step method. Thermal cycling conditions were as follows: pre-denaturation at 95 ℃ for 5 min, followed by 40 PCR cycles at 95 ℃ for 10 s and 60 ℃ for 30 s. The ratio of each target gene was determined using GAPDH as an internal control. The relative expression levels of genes were calculated from the quantification cycle (Cq) value and standardized by the 2−ΔΔCq method. Primers for amplified genes (IL-1β, IL-6, and TNF-α) are presented in Additional file 2: Table S2.

Statistical analysis

Immunoblotting and qRT-PCR analyses were repeated three times (n = 3). All statistical analyses were performed using GraphPad Prism 9.5 software (GraphPad, San Diego, CA, USA). Data from two or more groups were compared using one-way ANOVA, and differences between the two groups were analyzed using t-test. P < 0.05 was considered statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Results

Expression and purification of PvTRAgs

We summarized the structural features of PvTRAgs to investigate their molecular function. The smallest protein in the PvTRAg protein family was PvTRAg26 (223 amino acids; 26 kDa), and the largest protein was PvTRAg18 (2662 amino acids; 309 kDa). All PvTRAg proteins contain a tryptophan-rich structural domain in the C-terminal region. Characterization information for all members of the PvTRAgs is presented in Additional file 3: Table S3. We attempted to express and purify the recombinant PvTRAgs (rPvTRAgs) from E. coli to further characterize PvTRAgs function. Fusion proteins were successfully expressed for 23 of the 36 clones. The rPvTRAgs were purified under nondenaturing conditions (Fig. 1a). Corresponding immunoblotting assays were performed with anti-His tag monoclonal antibodies (Fig. 1b).

Fig.1
figure 1

SDS-PAGE and Western blot analysis of purified His-tag recombinant PvTRAgs. a SDS-PAGE analysis of purified PvTRAgs was performed under denaturing and reducing conditions. Gel protein content was detected by Coomassie blue staining. b Western blot analysis with anti-His antibody was performed to confirm the expression of His-tag on the recombinant protein. The molecular weight of the protein band includes the size of 6X His-tag. M: Protein marker; 1,2…36: PvTRAg proteins

Full size image

rPvTRAg treatment leads to different changes in collagen I levels in HSFs

The successfully expressed rPvTRAgs were co-incubated with HSFs for 4 h to explore the interaction between PvTRAgs and HSFs. Nine PvTRAgs (PvTRAg1, PvTRAg4, PvTRAg5, PvTRAg13, PvTRAg17, PvTRAg19, PvTRAg23, PvTRAg33, and PvTRAg34) were identified as capable of binding to HSFs (Fig. 2a).

Fig.2
figure 2

The levels of collagen I secreted by HSFs after stimulated by recombinant PvTRAgs. a Screening results of PvTRAg proteins binding to HSFs; b 5 × 106 HSFs were incubated with bindable recombinant proteins at the corresponding concentrations. After 48 h, the cells were lysed and the expression of collagen I in HSFs was detected by immunoblotting (left), and the gray values were compared between the two groups using GAPDH as an internal reference (right); c 5 × 106 HSFs were incubated with unbindable recombinant proteins at the corresponding concentrations. After 48 h, the cells were lysed and the expression of collagen I in HSFs was detected by immunoblotting, and the gray values were compared between the two groups using GAPDH as an internal reference. M: Protein marker; N: untreated HSF cells; 1,4…36: PvTRAg proteins. Statistical analyses were carried out by one-way ANOVA with Dunnett’s multiple comparisons test (ns P ≥ 0.05; *P < 0.05; **P < 0.01, ***P < 0.001, ****P < 0.0001)

Full size image

HSFs were treated with rPvTRAgs known to bind with cells for 48 h to explore whether other PvTRAgs interacting with HSFs affect collagen I. The results showed that only PvTRAg5 and PvTRAg23 significantly reduced the collagen I levels in HSFs (P < 0.0001, Fig. 2b). This result suggests that not all PvTRAgs (binding to HSFs) resulted in a decrease in collagen I levels.

HSFs were treated with rPvTRAgs (without binding to HSFs) to explore whether rPvTRAgs would act directly on HSFs. Despite certain proteins in the PvTRAg family not binding to HSFs, PvTRAg16, PvTRAg30, and PvTRAg32 also significantly reduced collagen I levels (P < 0.0001, Fig. 2c). These results provide strong evidence that PvTRAgs can specifically signal HSFs to decrease collagen I independent of receptor and ligand anchoring.

Type I collagen downregulation is linked to NF-κB signaling pathway activation

HSFs were treated with rPvTRAgs (PvTRAg5, PvTRAg16, PvTRAg30, and PvTRAg32) for 0.5 h to investigate the signaling pathway responsible for the reduction in collagen I levels. The results showed that three signaling pathways (NF-κBp65, FAK, and P38 MAPK) were activated after stimulation by PvTRAg5, PvTRAg16, PvTRAg30, and PvTRAg32 (Fig. 3a). HSFs were treated with appropriate signaling pathway inhibitors (Bay 11–7082, SB203580, and TAE226) to identify which pathways mediate the reduction in collagen I levels. After 1 h inhibitor treatment, HSFs were subsequently co-incubated with rPvTRAgs for 48 h. The results showed that collagen I levels increased after inhibition of the NF-κBp65 pathway, whereas they decreased after inhibition of the other pathways compared with the control group. Notably, the results are consistent with those for four proteins mentioned above (Fig. 3b–d). Therefore, only the NF-κBp65 pathway was found to be responsible for the decrease in collagen.

Fig. 3
figure 3

NF-κBp65 pathway mediates decreased collagen I. a The 5 × 106 HSFs were treated with 50 mg/ml rPvTRAg5, rPvTRAg16, rPvTRAg23, rPvTRAg30, and rPvTRAg32, and the treated cells were collected and incubated with the corresponding signaling pathway antibodies, respectively. The activation status of each group of pathways was detected by immunoblotting, and GAPDH was used as an internal reference. b HSFs were pretreated with the NF-κBp65 signaling pathway inhibitors for 1 h; c HSFs were pretreated with the FAK or P38 MAPK signaling pathway inhibitors for 1 h; 50 mg/ml rPvTRAg5, rPvTRAg16, rPvTRAg23, rPvTRAg30, and rPvTRAg32 stimulated the treated and untreated cells; 48 h later, the expression of collagen I was detected in each group by immunoblotting, and the comparison was made by using GAPDH as an internal involved group. Plus (+) indicates added; minus (−) indicates not added

Full size image

Expression of IL-1β, IL-6, and TNF-α decreases type I collagen in HSFs

The HSFs were treated with PvTRAgs for 48 h (PvTRAg5, PvTRAg16, PvTRAg30, and PvTRAg32) to assess the levels of inflammatory factors. Meanwhile, HSFs stimulated with PvTRAg1 (bound to HSFs) and PvTRAg2 (not bound to HSFs) were used as control groups.

The results showed that IL-1β, IL-6, and TNF-α levels were significantly higher in the groups simulated with rPvTRAg5, rPvTRAg16, rPvTRAg23, rPvTRAg30, and rPvTRAg32 than in the control group (P < 0.0001, Fig. 4). Overall, PvTRAgs (PvTRAg5, PvTRAg16, PvTRAg23, PvTRAg30, and PvTRAg32) activated the NF-κB p65 signaling pathway in HSFs. This activation triggered an inflammatory response, which ultimately led to a decrease in the collagen I levels.

Fig.4
figure 4

Elevated expression of IL-1β, IL-6, and TNF-α in HSFs. 5 × 106 HSFs were treated with 50 mg/ml rPvTRAg1, rPvTRAg2, rPvTRAg5, rPvTRAg16, rPvTRAg23, rPvTRAg30, and rPvTRAg32, respectively, and RNA was extracted from treated and untreated groups after 48 h. The expression of the relevant cytokines was detected by qPCR and compared with that of the control group. Three independent experiments have been performed. Statistical analyses were carried out by one-way ANOVA with Dunnett’s multiple comparisons test (ns P ≥ 0.05; *P < 0.05; **P < 0.01, ***P < 0.001, ****P < 0.0001)

Full size image

Discussion

TRAgs have been studied in a wide range of Plasmodium spp., but a clear molecular function for these antigens has not been determined [39]. Plasmodium vivax exhibits a higher abundance of TRAgs than other human malaria parasites, and their occurrence in clusters along chromosomes suggests expansion through gene duplication and diversification. Gene expansion is always accompanied by functional diversification [40]. Previously found stage-specific expression of pvtrag genes confirms their functional diversification in the life cycle of parasites [36]. Notably, PvTRAg23 has been identified to alter collagen I levels in HSFs. Therefore, we aimed to characterize 23 PvTRAgs by examining their ability to bind to HSFs and their impact on collagen levels. Our findings demonstrate that PvTRAgs can specifically reduce collagen I levels by activating the NF-κB p65 signaling pathway.

Binding of merozoite proteins to host cells is critical for their invasion, immune evasion [41, 42], and damage to host cells and tissues [43]. After infection of human erythrocytes, malaria parasite export proteins are transported to the iRBC membrane and act as mediators for interactions with host cells [44]. Similarly, PvTRAgs exhibit strong immunogenicity and partial binding to erythrocytes, which contributes to the pathogenesis of malaria [33]. Here, PvTRAgs were co-incubated with HSFs, and nine proteins were identified for binding to HSFs (Fig. 2a). Recent studies have shown that P. vivax-infected reticulocytes can adhere to HSFs via VIR14 and may avoid clearance by macrophages [23, 45]. Whether the binding of PvTRAg to HSFs contributes to P. vivax adhesion to HSFs remains to be determined and will not be discussed further here. Among the nine proteins proteins that could bind to HSFs, PvTRAg5 and PvTRAg23 decreased collagen I levels (Fig. 2b). This finding suggests that the receptor might not be structurally or functionally altered even in the presence of receptor–ligand interaction. Expression of the PvTRAg23 gene was high in Brazilian isolates; however, it was relatively low in Cambodian and East African isolates [46]. The PvTRAg23 gene expressions vary between different P. vivax strains as a contributing factor to antigenic variation [46,47,48], which may favor different immune evasion strategies for different P. vivax strains [49].

Although the specific binding of receptors and ligands is fundamentally important for various cellular processes and biochemical activities [50, 51], some proteins do not act through the formation of receptor–ligand complexes [52, 53]. Non-adhesive PvTRAgs may also play a crucial role in parasite growth [33]. The results show that three PvTRAgs (PvTRAg16, PvTRAg30, PvTRAg32) can directly act on HSFs to reduce the collagen I levels (Fig. 2c). These results indicate that the members of the PvTRAg protein family regulate collagen I levels through direct and indirect interactions with HSFs, which lead to a reduction in collagen I levels.

Fibroblasts produce large amounts of collagen I, which is the most abundant component of the ECM [15, 54]. The ECM can perform microbial recognition and phagocytosis in the immune response to infection [55] and can convey specific signals to immune cells encountering or navigating through it [56]. Furthermore, the collagen network contributes to the proper functioning of the immune system in the spleen [19, 20]. Overall, reduced expression of collagen I leads to altered ECM, which affects ECM-mediated functions in the spleen. Therefore, a decrease in collagen I levels by PvTRAgs can influence immune functions in the spleen, such as blood filtration and trapping of pathogens, which may ultimately contribute to parasite escape. Recently, a study has unequivocally shown that the spleen sustains a very large biomass of non-phagocytosed asexual P. vivax parasites and P. vivax malaria is mainly a cryptic erythrocytic infection of the spleen [57, 58]. Impaired splenic blood clearance contributes to parasite immune evasion and may favor the accumulation of malaria parasites in the spleen to enhance malaria parasite survival and replication in the spleen.

Altered cellular collagen levels are frequently accompanied by the activation of the NF-κBp65, FAK, and P38 MAPK signaling pathways [59,60,61]. PvTRAg23 mediates the activation of the NF-κBp65 signaling pathway, which leads to the upregulation of IL-1β, IL-6, and TNF-α [35]. This phenomenon inhibits the collagen I levels in HSFs [35]. In this study, the four other PvTRAgs, namely, PvTRAg5, PvTRAg16, PvTRAg30, and PvTRAg32, were also found to be involved in the reduction in collagen I levels in HSFs. The reduction in collagen I levels was determined to be mediated by the NF-κBp65 signaling pathway (Fig. 3a–c). Elevated levels of relevant inflammatory factors (IL-1β, IL-6, and TNF-α) were observed, as expected (Fig. 4). IL-1β, IL-6, and TNF-α have been found to inhibit the production of type I collagen [35]. The same situation was present in diabetic rat cardiomyocytes, where levels of TNF-α, IL-6, and IL-1b were negatively correlated with mRNA expression of collagen I [62]. Therefore, these inflammatory cytokines play a role in the regulation of collagen expression.

Despite our discovery of the function for PvTRAgs, some limitations need further improvement, which lays the groundwork for the exploration of the molecular functions of this unique family of Plasmodium protein. First, 13 recombinant PvTRAgs were unsuccessfully expressed. Thus, the functional characterization of these proteins needs further determination. Second, the receptor through which PvTRAgs bind to HSFs was not identified, and the role of ligand-receptor binding in causing a decrease in collagen I was unclear. Third, the activation mechanism of the NF-κBp65 signaling pathway by PvTRAgs was unknown.

Conclusions

Five members of the PvTRAg family of proteins (PvTRAg5, PvTRAg16, PvTRAg23, PvTRAg30, and PvTRAg32) are capable of activating NF-κBp65 in HSFs, which subsequently results in an inflammatory response and ultimately a decrease in collagen I levels. The reduced collagen levels caused by PvTRAgs may impact spleen functions, such as clearance and filtration, which may ultimately facilitate the realization of parasite escape.