Genetic sequence characterization and naturally acquired immune response to Plasmodium vivax Rhoptry Neck Protein 2 (PvRON2)

To evaluate the genetic diversity of pvron2 (sequence from nucleotide 5481–6240 from the pvron2 gene), a total of 36 P. vivax isolates were collected from patients in Manaus, Amazonas-Brazil at Fundação de Medicina Tropical Dr. Heitor Vieira Dourado (FMT-HVD) (CAAE-0044.0.114.000-11/CAAE 54234216.1.0000.0005) and in Mâncio Lima and Acrelândia, Acre State (Fig. 1) between 2011 and 2013 and in 2015 (936/CEP, 2010 and 1169/CEPSH, 2014).

Humoral immune responses to a recombinant PvRON2 (rPvRON2) antigen in Manaus (− 03°06′26″S; 60°01′34″W) (CAAE-0044.0.114.000-11/CAAE 54234216.1.0000.0005) and in Itaituba (04°16′34″S; 55°59′01″W), a gold mining area located in the most southwestern part of the State of Pará, were assessed (Fig. 1). The Annual Parasite Incidence designates Itaituba as an area with a high risk of malaria transmission (102.0 cases/1000 inhabitants per year) (CAAE—001.219.346-15). In contrast, Manaus presents a low malaria transmission risk (5.8 cases/1000 inhabitants per year).

Samples of peripheral blood were collected from the following three groups: (1) individuals acutely infected with P. vivax in Itaituba (n = 93); (2) individuals who were not infected but previously had malaria (n = 97); and (3) acutely infected individuals from Manaus (n = 124). All individuals were previously diagnosed by thick blood film microscopy screens, and P. vivax mono infection was also determined by nested PCR, as previously described [20].

The PvRON2 gene has a length of 6612 bp. A synthetic gene fragment was designed based on the nucleotide sequence of the pvron2 Sal1 strain (PlasmoDB PVX_117880). This sequence is only partially located in the PvRON2 region that binds to AMA1 during the MJ formation. Nevertheless, the selected region presented a high score in the antigenicity analysis, according to the Kolaskar and Tongaonkar method, using the program Immune Epitope database and Analyses Resource (IEDB Analysis Resource) [21].

Genomic DNA from P. vivax isolates collected in Manaus was extracted using standard phenol–chloroform methods [22]. The DNA templates from the isolates collected in Mâncio Lima and Acrelândia were isolated from 200 µL of whole blood using QIAamp DNA blood kits (Qiagen, Hilden, Germany), with a final DNA elution volume of 200 µL, according to the manufacturer’s instructions. DNA samples were stored at − 20 °C prior to use.

DNA samples of P. vivax were used as the template for the amplification of the pvron2 759pb sequence. Three DNA fragments were PCR-amplified to obtain the complete fragment. The oligonucleotide sequences used in this study are listed in Additional file 1. The PCR conditions used to amplify fragments one and three were as follows: 1 cycle of 5 min at 95 °C followed by 35 cycles of 30 s at 95 °C, 45 s at 60 °C, 1 min at 72 °C and a final cycle of 5 min at 72 °C. To amplify fragment two, the PCR conditions were as follows: 1 cycle of 5 min at 94 °C followed by 35 cycles of 30 s at 94 °C, 30 s at 60 °C, 45 s at 72 °C and a final cycle of 5 min at 72 °C. The reactions were performed using a reaction mixture containing 2.5 mM MgCl₂, 0.5 mM each dNTP (Invitrogen), 0.5 units of Platinum Taq polymerase (Invitrogen) and 1 µM of each oligonucleotide primer in a final volume of 50 µL. The purified PCR product was sequenced using 3730xl DNA Analyzer (Applied Biosystems).

Amplified pvron2 sequences from 36 Brazilian Amazon isolates were analysed. The fragments were assembled using CodonCode Aligner v. 6.0.2, and the sequence data was deposited in the GenBank (accession numbers are listed in Additional file 2). Single nucleotide polymorphisms (SNPs) were identified in the alignment of 103 pvron2 sequences from 7 other countries (Thailand, Mexico, Mauritania, China, Peru, Colombia and North Korea), which were previously deposited in the PlasmoDB [23], and GenBank [24] databases. All sequences were compared to the P. vivax reference sequence Sal-1 strain (PlasmoDB: PVX117880) using CLC Sequence Viewer 7.

A PvRON2 gene fragment encoding amino acid residues 1828–2080 was obtained from the genomic DNA of the P. vivax Sal I strain. This fragment was codon optimized for E. coli and subsequently cloned into the pGEX 4T-1 vector (synthetized by GenScript USA Inc. Piscataway, New Jersey). For recombinant GST-PvRON2 protein expression, the pGEX4T-1_pvron2 plasmid was transformed into competent STAR BL21(DE3) E. coli cells by heat shock [25]. Then, the bacteria cultures were grown according to the manufacturer’s instructions. Briefly, 100 mL of overnight culture was transferred into 3 L of LB containing ampicillin (100 µg/mL) and incubated at 37 °C with shaking. When the culture reached an OD₆₀₀ of 0.6, protein expression was induced by the addition of 0.8 mM IPTG for 4 h at 37 °C. The culture was pelleted by centrifugation (6000×g, 10 min), resuspended in 25 mL of lysis buffer (10 mM Tris HCl pH 8.0, 150 mM NaCl, 1 mM EDTA) in the presence of 1× Complete Protease Inhibitor Cocktail (Roche, Mannheim, Germany) and incubated for 1 h on ice. The sample was then disrupted using an M-110 L Pneumatic High Shear Fluid Processor (Microfluidcs). Next, cell fragments were pelleted by 30 min of centrifugation at 10,000×g. Finally, the supernatant was collected and analysed using SDS-PAGE and Western Blot to confirm protein expression.

Naturally, acquired IgG and IgM antibodies against rPvRON2 were measured in the plasma samples by direct enzyme-linked immunosorbent assay (ELISA). Plasma samples from infected (n = 93) and non-infected (n = 97) individuals from Itaituba, and infected individuals from Manaus (n = 124) were evaluated for the presence of IgG antibodies to rPvRON2. The same groups were evaluated for the presence of IgM antibodies to rPvRON2, as follows: infected individuals from Itaituba (n = 56), non-infected individuals from Itaituba (n = 97), and infected individuals from Manaus (n = 68). Samples from non-infected individuals from a non-endemic region were used as negative controls (n = 21).

High-protein binding 96-well ELISA plates were coated with 50 μL of rPvRON2 at 5 μg/mL in 0.05 M carbonate-bicarbonate, pH 9.6, overnight at 4 °C. Then, plasma samples (100 µL) diluted 1:100 were added to each well and incubated for 1 h at room temperature. For the detection of bound antibodies, the samples were incubated with a 1:2000 dilution of peroxidase-conjugated goat anti-human IgG or IgM (Sigma). The optical density (OD) was measured at 490 nm using CLARIOstar data analysis.

To avoid a bias in the results caused by the possible reactivity of the GST tag during protein expression, the excess band size was measured on the SDS PAGE gel using ImageJ. To calculate the protein surplus, plates were coated with GST. The values obtained in each sample reaction to GST were subtracted from the value obtained in the reaction against PvRON2.

All plates tested were normalized using the values of the anti-GST controls (4 well per plate). The cutoff value was calculated as the mean plus three standard deviations of the negative control. The reactivity indices (RIs) were obtained from the ratio of the absorbance values of each sample and the cutoff value. The prevalence of IgG and IgM against the rPvRON2 antigen was considered positive if the (RI) values were higher than 1.0.

The detection of IgG subclasses was performed as mentioned above, except that IgG1 (HRP), IgG2 (HRP), IgG3 (HRP) and IgG4 (HRP) specific secondary monoclonal mouse anti-human antibodies (abcam) diluted 1:2000 were used. The results were expressed as the RI ± SEM (standard error of the mean).

The plasma levels of the cytokines IL-6, IL-10, IFN-γ and TNF were quantified by flow cytometry using the BD IL-6, IL-10, IFN-γ, TNF Human Flex Set (BD Bioscience Pharmingen, San Diego, Ca, USA) following the instructions provided by the manufacturer. Data analyses were performed using the FACSDiva software (BD Biosciences, San Jose, CA, USA). The cytokine concentrations in each sample were determined based upon standard curves. The plasma cytokine concentrations for each sample were extrapolated from the standard curves, and the data were expressed as ρg/mL.

As the antibody levels were not normally distributed, nonparametric tests were used. Spearman’s correlation was applied to assess the association between antibody levels with the following parameters: age, parasitaemia, platelets, RBC, haematocrit, haemoglobin, IL-6, IL-2, IL-10, IL-4, TNF and IFN-γ. Correlation networks were generated by the analysis of relationships among each mediator measured in the plasma samples. The systemic levels of each mediator were input in the R software (v. 3.4.3). Initially, pairwise Spearman’s correlation coefficients were calculated using the R programming language. Along with the Spearman rank-order correlation coefficient, the p value to test for non-correlation was evaluated using p ≤ 0.05 as a cutoff. Moreover, based on the Spearman correlation coefficient, the same software was applied to identify links (edges) of interaction between the mediators (nodes). The correlation strength is represented by edge transparency and width; positive correlations are represented by red edges, and negatives correlations are represented by blue edges. Following this approach, each mediator was selected as a target, and the R software was used to perform a search within the other mediators for those that were associated with the target, in terms of correlation strength. As a result, the features related to the selected target were linked. This process was repeated for each mediator, and the result was the inferred network among the input values. To analyse the structure of the networks; the graphics of the network analysis were customized using Cytoscape software (v 3.5.1). The prefuse force-directed layout was applied, which, in the equilibrium states for the system of forces, represents the correlation strength, the edges with uniform length, and nodes that are not connected by an edge tend to be drawn further apart.

Fisher’s test was applied to compare the prevalence of positive responses, and differences between medians and cytokine levels were assessed with the Mann–Whitney U-test using GraphPad Prism. A p-value of p < 0.05 was considered significant.