The gene encoding region II (RII) of PocDBP (aa 182 to 506) was amplified using nested PCR. Genomic DNA was extracted from whole-blood samples from a parasite-infected patient, which was kindly provided by Jiangsu Institute of Parasitic Diseases, Wuxi, China. Primer sets for nested PCR were designed based on the pocdbp (PocGH01_00129200) sequence. The primers used were as follows: Nest 1 F: tcgcggatccgaattc GCTTTTAGAGATGTTCCTAATTATGG; Nest 1 R: ggtggtggtgctcgag TTTTATTCCTTTCTGCGCG; Nest 2 F: tcgcggatccgaattc AATATTACAAACAATGATGTAAATTATGT; Nest 2 R: ggtggtggtgctcgag TTTTATTCCTTTCTGCGCG. The restriction enzymes EcoRI and XhoI are indicated as italicized and underlined letters. PCR amplification was performed using high-fidelity Phusion DNA polymerase (New England Biolabs Inc., Ipswich, MA). Each reaction consisted of a total volume of 20 µl containing 7.5 mM MgCl₂, 2.5 mM dNTPs, 0.5 µl of sense and antisense primers (100 pmol/µl), 0.2 µl of high-fidelity Phusion DNA polymerase (2 U/µl) and 2 µl of gDNA as template. The PCR thermal cycling conditions were set as follows: initial denaturation at 95°C for 5 min, followed by 35 cycles of 95°C for 30 s, 59°C for 30 s and 72°C for 1 min and a final extension at 72°C for 10 min. Amplicons were gel purified using a DNA purification kit (Macherey–Nagel, Duren, Germany) according to the manufacturer’s instructions and ligated into the pET28a (+) expression vector (Novagen, Madison, WI) with a C-terminal His-tag. The obtained purified plasmid DNA sequence was confirmed by sequencing analysis and transformed into BL21 (DE3) competent cells (Novagen). Isopropyl-β-d-thiogalactopyranoside (IPTG; 1.0 mM; Sigma-Aldrich Co., St. Louis, MO) was used to induce recombinant protein expression. Protein solubilization, purification, and refolding were performed as previously described (Singh et al., 2001). The refolded proteins were eluted by ion-exchange chromatography (HiTrap™ SP FF; GE Healthcare Life Sciences, Chicago, IL) using 1 M NaCl. GST-His (glutathione S-transferase 6 X His tag) protein was expressed as per manufacturer’s protocol (GST Gene Fusion system, GE Healthcare Life Sciences, Uppsala, Sweden). Recombinant PvDBP-RII protein was expressed and purified as described elsewhere (Singh et al., 2001).

The refolded recombinant PocDBP-RII protein was separated by 8% SDS-PAGE and stained with 0.25% Coomassie brilliant blue (Sigma-Aldrich Co.). Briefly, 5 μg of refolded protein was incubated with 10 mM dithiothreitol (DTT, reducing condition) at 37°C for 1 hr, followed by the addition of 2× loading dye containing the reducing agent 2-mercaptoethanol; in addition, 5 μg of protein was processed without DTT (nonreducing condition) and with the 2× loading dye without reducing agent. The samples were heated at 100°C for 4 min. For the Western blot analysis, the proteins were electrotransferred to 0.45 μm PVDF membranes (Millipore, Bedford, MA) by electrophoresis in semidry transfer buffer (50 mM Tris, 190 mM glycine, 3.5 mM SDS, 20% methanol) with a continual current of 370 mA for 40 min using a semidry transfer system (ATTO Corp., Tokyo, Japan). Then, the membrane was incubated with blocking buffer (5% skim milk in PBS containing 0.2% Tween 20) and then incubated with a primary anti-penta-histidine antibody (1:2000) and rabbit immune serum (1:1000), followed by incubation with a secondary IRDye^® goat anti-rabbit antibody (1:10,000 dilution) (LI-COR^® Bioscience, Lincoln, NE). Data analysis was performed using an Odyssey infrared imaging system and the company-recommended software (LI-COR^® Bioscience).

One Japanese white rabbit was used to produce an anti-PocDBP-RII antibody. Two hundred and fifty micrograms of purified protein with complete Freund’s adjuvant (Sigma-Aldrich Co.) was injected subcutaneously, followed by treatment with 250 µg of incomplete Freund’s adjuvant for subsequent boosting. All immunizations were administered 3 times at 3-week intervals. Antisera were collected two weeks after the final boost. Anti-PvDBP-RII antibody generation was performed as described in our previous study (Han et al., 2016). Total IgG was purified from 1 mL of anti-PvDBP-RII and anti-PocDBP-RII rabbit immune serum by using a protein G HP column as per manufacturer’s protocol (GE Healthcare Life Sciences) as described elsewhere (Muh et al., 2018). All the experimental protocols were approved by the Kangwon National University Animal Care and Use Committee, and the experiments were conducted according to the Ethical Guidelines for Animal Experiments of Kangwon National University (KIACUC-16-0157).

Umbilical cord blood samples were collected in a 10 ml heparin tube (BD Vacutainer^®, Becton-Dickinson Co., Franklin Lakes, NJ). The relevant guidelines and regulations were followed to conduct all the experiments, and the human sample-related experimental protocols were approved by the Kangwon National University Hospital Ethical Committee (IRB No. 2014-08-008-006). Written informed consent was obtained from all the subjects.

Reticulocytes were enriched from umbilical cord blood using a cushion of 19% Nycodenz solution (Axis-Shield, Oslo, Norway) in high-KCl buffer via gradient centrifugation. Upon receipt, the fresh cord blood was washed twice with incomplete RPMI 1640 medium, and white blood cells (WBCs) were removed using an NWF filter (Zhixing Bio Co., Ltd., Bengbu, China). WBC-free packed cells were then resuspended in high-KCl buffer (115 mM KCl) (pH 7.4), followed by incubation at 4°C for 3 hr with rotation. Three milliliters of prewarmed Nicodenz solution (19%) was transferred into 15 ml tubes. Then, 5 ml of the RBC-high-KCl buffer mixture was poured on top of the Nicodenz cushion and centrifuged for 30 min at 3000 ×g without braking. The reticulocytes were harvested from the interface layer between Nicodenz and high-KCl buffer and washed three times with incomplete RPMI 1640 medium. Reticulocyte purity was determined from thin blood smears with new methylene blue staining by light microscopy and thiazole orange (TO) staining of the harvested reticulocytes. A total of 100,000 events were obtained per sample using a FACS Accuri™ C6 Flow Cytometer (Becton-Dickinson Co., Mansfield, NJ).

Enriched reticulocytes were prepared with up to 50% hematocrit. Then, the erythrocytes were washed with 500 μl of incomplete RPMI 1640 medium twice by centrifugation at 500 ×g for 3 min at 4°C. Then, the erythrocytes were treated with either neuraminidase (100 mU; from Vibrio cholerae, Sigma-Aldrich Co.), trypsin (0.5 mg; from bovine pancreas, Sigma-Aldrich Co.) or chymotrypsin (0.5 mg; from bovine pancreas, Sigma-Aldrich Co.) at 37°C on a rotator for 1 hr. After enzyme treatment, chymotrypsin- and trypsin-treated RBCs were incubated with a trypsin inhibitor from soybean (Glycine max) (Sigma-Aldrich Co.) at 37°C for 10 min and subsequently washed three times with 10 ml of incomplete RPMI 1640 medium. Packed cells were prepared at a concentration of 1 × 10⁶ cell/ml and used for flow cytometry analysis.

The erythrocyte binding assay was performed as described previously (Tran et al., 2005). Briefly, a gradient concentration of purified PocDBP-RII protein was incubated with 1 × 10⁶/ml cells or the same concentration of reticulocytes treated with each enzyme for 3 hr at 25°C. The PvDBP-RII and GST-His protein reticulocyte binding activities were used as the experimental controls. The samples were washed with 200 µl of PBS (1% BSA) two times, followed by incubation with diluted (1:50) mouse anti-penta-His Alexa Fluor 647-conjugated monoclonal antibody (Qiagen, Hilden, Germany) for 1 hr at 4°C in the dark. The samples were washed three times with PBS (1% BSA) and incubated with TO (Becton-Dickinson Co., San Jose, CA) for 30 min at 25°C. A total of 100,000 events were counted per sample using a FACS Accuri™ C6 Flow Cytometer (Becton-Dickinson Co.). FlowJo 7.6 (Treestar, Ashland, OR) was used to analyze the flow cytometric results. Unstained cells and cells singly stained with TO represented normocytes and reticulocytes, respectively.

Three-dimensional (3-D) structure modeling and validation of PocDBP-RII and PvDBP-RII were performed by homology-based modeler software. Satisfactory structural templates were explored and modeled using SWISS-MODEL (Biasini et al., 2014). The error residues were refined by using Galaxy Refine (Ko et al., 2012). Finally, all the structures were visualized by UCSF CHIMERA (Huang et al., 2014).

Data analysis was performed using GraphPad Prism (GraphPad Software, San Diego, CA). Student’s t-test was used to compare the experimentally measured values of different groups. Values of p < 0.05 indicated significant differences.

The pocdbp gene sequence encodes a moderate-sized protein (900 amino acids) with a predicted molecular weight of 103.22 kDa ( Figure 1A ). The gene consists of 5 exons encoding a signal sequence, a transmembrane domain and 22 cysteine residues. The putative functional domain site of PocDBP-RII is defined based on the PvDBP-RII homolog site as an erythrocyte-binding domain. Twelve cysteine residues are conserved in the RII domain of the pocdbp gene, similar to PvDBP-RII and PkDBP-α RII, which is probably indicative of similar functions. The sequence alignments of PvDBP-RII, PkDBP-α-RII, and PocDBP-RII were generated using Clustal W, revealing that PocDBP-RII shares 44.4% and 40.4% amino acid sequence identity with PvDBP-RII and PkDBP-α-RII, respectively ( Figure 1B ).

The three-dimensional structure analysis clearly indicated the similar shape of the binding pocket and structure between PvDBP-RII and PocDBP-RII ( Figure 2 ). Based on the electric charge, the structure is represented mainly by two distinct parts ( Figure 2 ). Positively charged PvDBP-RII structures are well known for binding to reticulocytes (Batchelor et al., 2011). Despite low sequence identity, the PocDBP-RII structure was very similar to that of PvDBP-RII ( Figure 2 ), which might indicate similar functional activities.

The recombinant PvDBP-RII and PocDBP-RII protein was purified from inclusion bodies after bacterial expression and refolded by rapid dilution as previously published (Singh et al., 2001). The recombinant PocDBP-RII protein was used for rabbit immunization. Evidence for refolding of the purified recombinant PvDBP-RII and PocDBP-RII protein was shown by SDS-PAGE analysis ( Figures 3A, B ) (Muh et al., 2018). Different mobilities of the refolded protein between reducing (DTT +) and nonreducing conditions (DTT −) indicated that the native protein had been formed correctly after refolding ( Figures 3A, B ). Anti-PocDBP-RII IgG and anti-His antibodies could be used to detect the recombinant proteins by Western blot analysis ( Figure 3C ). These results suggest that immune serum raised against PocDBP-RII can recognize the recombinant protein.

A fluorescence-activated cell sorting (FACS)-based binding assay was used for reticulocyte-binding activity evaluation. The PvDBP-RII and His-tagged GST proteins were used as positive and negative controls, respectively, for validation of binding activity. Reticulocytes (61.93% on average) were enriched from cord blood and used for the binding assay ( Figure 4A ). PvDBP-RII bound strongly to reticulocytes at a protein concentration of 1.25 μg/ml, and binding saturation was observed at a concentration of approximately 5 μg/ml ( Figure 4B ). The binding of PocDBP-RII to reticulocytes was shown to increase in a concentration-dependent manner and was saturated at a concentration of 10 μg/ml ( Figure 4B ). PvDBP-RII binding was detected at a concentration of 10 μg/ml, with a mean binding activity of 83.74% ± 1.24% with reticulocytes and 24.59% ± 2.90% with normocytes (p < 0.0001), which represented a 3.4-fold increase in reticulocytes ( Figure 4C ). The PocDBP-RII protein was bound explicitly to reticulocytes at a concentration of 10 μg/ml with a mean binding activity of 44.55% ± 5.44%, which was a 8.0 fold higher than the normocyte binding activity (p < 0.002) ( Figure 4C ). The GST-His protein normocyte binding activity was similar to that of reticulocytes and was below the cutoff percentage, which indicated that there was no binding activity with normocytes ( Figure 4C ). The GST-His protein binding activity was measured as a negative control.

The reticulocyte binding specificity was confirmed by an antibody inhibition assay. The anti-PocDBP-RII IgG antibody was able to inhibit reticulocyte binding in a concentration-dependent manner ( Figure 5 ). Binding inhibition was significantly different at concentration of 50 μg/ml and 100 μg/ml of anti-PocDBP-RII IgG (p = 0.0228 and p = 0.0496 respectively) as compared to preimmune rabbit IgG. Reticulocytes treated with trypsin, chymotrypsin, and neuraminidase were also used to check the binding specificity of the recombinant PocDBP-RII protein. The PvDBP-RII reticulocyte binding activity with chymotrypsin-treated reticulocytes was inhibited by three-fourths compared to that of untreated PvDBP-RII (percentage of relative binding, mean ± SD: 23.87% ± 7.49%); this binding activity was significantly different (p = 0.0196) from the binding activity for normal reticulocytes. A similar binding pattern was also observed for the PocDBP-RII protein. The PocDBP-RII binding activity (39.8% ± 9.19%) was significantly inhibited by chymotrypsin (p = 0.0226) ( Figure 6 ).