Mapping of B-cell epitopes on the N- terminal and C-terminal segment of nucleocapsid protein from Crimean-Congo hemorrhagic fever virus

Crimean-Congo hemorrhagic fever virus (CCHFV) is a tick-borne pathogen that causes severe disease in humans. CCHFV is widely distributed in more than 30 countries and distinct regions, which means that it poses a serious threat to human health. The nucleocapsid protein (NP) encoded by the CCHFV S gene is the primary detectable antigen in infected cells, which makes it an important viral antigen and a clinical diagnostic target. In this study, the modified biosynthetic peptide (BSP) method was used to identify the fine epitopes on the N- and C- terminals of NP from the CCHFV YL04057 strain using rabbit antiserum against CCHFV-NP. Nine epitopes were identified: E1a (178NLILNRGG185), E1b (184GGDENP189), E2 (352PLKWGKK358), E3 (363FADDS367), E4 (399NPDDAA404), E5a (447DIVASEHL454), E5b (452EHLLHQSL459), E6 (464SPFQNAY470) and E7 (475NATSANII482). Western blotting analysis showed that each epitope interacted with the positive serum of sheep that had been naturally infected with CCHFV. Amino acid sequence alignment between each epitope and their homologous proteins showed that they were almost 100% conserved among 12 CCHFV sequences from different lineages, except for epitopes E1a, E1b and E2. Three-dimensional structural modeling analysis showed that all identified epitopes were located on the surface of the NP “head” domain. This study identified fine epitopes on the N- and C- terminals of NP, which will increase the understanding of the structure and function of NP, and it could lay the foundation for the design and development of a CCHFV multi-epitope peptide vaccine and detection antigen.


Plasmids and antibodies
pET-32a-NP1 (aa 1-200), pGEX-KG-NP2 (aa 170-305) and pET-32a-NP3 (aa 286-482) were previously constructed and stored by this research group [22]. The prokaryotic expression plasmid pXXGST-3 was donated by Professor Wan-xiang Xu from Shanghai Institute of Planned Parenthood Research. The rabbit pAbs against CCHFV-NP were donated by Professor Fei Deng from Wuhan Institute of Virology, Chinese Academy of Sciences. The sheep serum samples used in the study were collected in 2005 in Bachu County and kindly provided by Professor Yujiang Zhang from Xinjiang Centers for Disease Control and Prevention [23]. The serum sample of CCHFV-infected sheep was previously identified as positive using an immunofluorescent assay (IFA) and indirect reverse transcription polymerase chain reaction (RT-PCR) [23]. In our previous study, we identified the antigenicity of epitopes on NP (237-305) using the same sera of sheep naturally infected with CCHFV [19]. Serum samples of a healthy rabbit and a healthy sheep with no history of CCHFV infection were used as negative controls in the western blotting assay. Escherichia coli BL21 (DE3) competent cells were used to express 16/8mer peptides fused with a truncated GST188 protein (i.e., with the initial 188 aa of GST) [24]. Goat anti-rabbit and mouse anti-goat IgG conjugated to horseradish peroxidase (HRP) were purchased from Beijing TransGen Biotech, Co., Ltd. (China).
Mapping strategy and biosynthesis of overlapping 16mer and 8mer peptides. To map epitopes on the NP1 and NP3 segments, we used the feasible strategy shown in Fig 1. A total of 48 16mer peptides numbered P1-P48 were bio-expressed for the two segments NP1 and NP3. The 16mers all had an overlap of 8 aa residues between each two adjacent peptides. For fine epitope motif mapping, eight sets of a total of 58 8mer peptides (named P49-P106) with an overlap of 7 aa residues were bio-expressed based on the reactive 16mer peptides mapped in the first round of antigenic peptide mapping. The aa sequences of the biosynthesized 16/8mer peptides and their positions on NP are shown in S1 and S2 Tables, respectively.

Construction of recombinant plasmids
All plus and minus strands of DNA fragments that encoded target 16/8mer peptides and had cohesive end nucleotides of BamH I and TAA-Sal I sites at the 5'-and 3'-ends were synthesized by SBS Genetech Co., Ltd (Shanghai, China). Each r-plasmid expressing a 16/8mer peptide was constructed according to the GST188-BSP method [21], in which the major steps were as follows: i) conducting an annealing reaction involving paired plus and minus strands; ii) conducting a ligation reaction involving annealed DNA fragment and pXXGST-3 plasmid cut by BamH I and Sal I; iii) transforming E. coli BL21 (DE3) competent cells with the ligation mixture; iv) screening of the r-clones by carrying out sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) using total proteins from each induced clone and observing whether there is a specific 16/8mer peptide on the gel; and v) sequencing of inserted DNA fragment encoding each 16/8mer peptide for each determined r-clone to ensure that all synthesized DNA sequences are accurate.

Expression of target short peptide
Each above-determined r-clone was used to express a 16/8mer peptide in E. coli BL21 (DE3) cells, which was fused with the GST188 protein [21], that is, each r-clone was cultivated in 2 mL luria bertani (LB) medium with 100 μg/mL ampicillin at 220 rpm overnight. The following day, 30 μL of bacteria culture was added to 3 mL of fresh LB medium, grown at 30˚C for 4 h to increase the bacterial density until reaching an optical density at 600 nm (OD 600 ) of 0.5-0.7, and then grown at 42˚C for 4 h to induce the expression of the target short peptide. All collected cell pellets containing the expressed 16/8mer peptide fusion proteins were stored at -20˚C.

SDS-PAGE and western blotting
The cell pellets obtained from 2 mL culture of expressed 16/8mer peptide were boiled at 95˚C in 200 μL of 1×SDS-PAGE loading buffer for 10 min, and the proteins were resolved by 15% SDS-PAGE under reducing conditions. Two gels were obtained for each16/8mer. One gel was stained with Coomassie brilliant blue G-250 for analyzing the bands corresponding to the target 16/8mer peptide. The other gel was used for western blotting by electrotransferring the proteins onto a 0.2 μm nitrocellulose (NC) membrane [25,26]. Regarding the specific antigenantibody reaction, the NC membrane was blocked with 5% (w/v) skimmed milk powder in Tris-buffered saline-Tween 20 (TBS-T), treated with rabbit pAbs (1:1000 dilution) or sheep serum (1: 100 dilution) as the primary antibody, and then reacted with goat anti-rabbit IgG or mouse anti-goat IgG conjugated to HRP (1:5000 dilution) as the secondary antibody. Finally, the blot was performed using the ECL plus western blotting detection reagent according to the manufacturer's instructions, and it was then imaged by GE-Image Quant LAS 4000 (GE Healthcare, Buckinghamshire, UK).

Sequence alignment of homologous CCHFVs
To assess the conservation of each identified epitope among CCHFV homologous proteins, 12 NP aa sequences from different countries and genetic lineages were obtained from GenBank based on the phylogenetic tree of CCHFV strains [27]. As far as we know, 11 complete NP sequences of CCHFV strains isolated in China have been registered in the GenBank database, and these were all compared to the NP sequence investigated in this study The aa sequences of the NP1 and NP3 segments from the YL04057 strain (GenBank code: ACM78470.1) and other CCHFV-NP homologous proteins were aligned using the ClustalW program [28] and visualized using Genedoc [29].

Antigenicity identification of NP1 and NP3 segments
In a previous study, the truncated NP1 and NP3 segments of CCHFV-NP were expressed using the prokaryotic expression vector pET-32a, and each predicted protein was fused with Trix tag, S tag and His tag (with a size about 18 kDa) [22]. However, the fusion proteins failed to be recognized by rabbit pAbs against CCHFV-NP (YL04057 strain) because of an unknown reason [22]. Numerous data on epitope mapping show that mapped epitopes do not only exist in the central area of a protein macromolecule. Therefore, to reveal all the epitopes on the NP protein after complete epitope mapping of the NP2 segment (which was cloned into a pGEX-KG vector), the predicted protein was fused with GST with a size about 26 kDa [19], the original rabbit pAbs against NP were used to re-identify the antigenicity of the NP1 and NP3 segments by immunoblotting. As shown in Fig 2, both r-segments NP1 and NP3 were recognized by the rabbit pAbs, like NP2 was, suggesting that there are also antigenic sites or epitopes on NP1 and NP3.

Mapping of epitopes on the NP1 and NP3 segments
To determine the existence of epitopes on the NP1 and NP3 segments, both NP1 and NP3 were truncated into 24 16mer BSPs (P1-P24 and P25-P48, respectively). The 48 overlapping 16mer BSP sequences and the corresponding sites on NP are shown in S1 Table. The 48 overlapping 16mer peptides were fusion expressed with GST188 in E. coli [20], and the fusion proteins were observed as approximately 24 kDa bands on the SDS-PAGE gel (Fig 3a and 3c). In the western blotting analysis, NC membranes were blocked with 5% (w/v) skimmed milk powder, incubated sequentially with rabbit pAbs against CCHFV-NP (1:1000 dilution) and goat anti-rabbit IgG (1:5000 dilution), and then visualized by ECL. Among the 48 expression products, the following eight 16mer peptides were identified as positive by western blotting: P23, P33, P34, P38, P39, P45, P47 and P48 (Fig 3b and 3d). Of these, one 16mer peptide (P23) was located in the NP1 segment (Fig 3b), while the remaining seven 16mer peptides were located in the NP3 segment (Fig 3d). None of the 16mer peptides reacted with the CCHFV-negative rabbit sera (data not shown).

Reactivity of the identified epitope motifs with anti-CCHFV serum
To determine whether the minimal epitopes are rabbit specific or also recognizable by other host species, nine selected 8mer peptides, each of which contained one of the nine pAbs-identified epitopes, were subjected to western blotting using sera from sheep with or without CCHFV infection. NC membranes were blocked with 5% (w/v) skimmed milk powder, incubated sequentially with anti-CCHFV sheep serum (1:100 dilution) and mouse anti-goat IgG (1:5000 dilution), and then visualized by ECL. All the selected 8mers were recognized by the serum of sheep with a confirmed history of CCHFV infection as the primary antibody (Fig 6), while none reacted with the CCHFV antibody-negative sheep sera. Among the nine epitopes, P58 (containing E2) and P79 (containing E4) showed the strongest reaction with CCHFV antibody-positive sheep sera, whereas P49 (containing E1a) and P54 (containing E1b) displayed weaker reactions.

Discussion
CCHFV-NP is the major viral protein that can be detected in the early stages of CCHFV infection. Serological studies have shown that patients have an early and high titer of antibody response to NP [32]. Due to its highly conserved features, CCHFV-NP can be used to develop diagnostic antigens and vaccines [19]. For rational vaccine design or development of diagnostic tools, it is necessary to map the specific and conserved BCEs of target proteins [21]. Also, multi-epitope-based vaccines offer numerous advantages compared to the complete protein, including cost-effective production, stability under different conditions and multivalency [24]. Recently, Zhao et al. [33] constructed a recombinant epitope vaccine that contains six immunodominant epitopes, and they demonstrated its capacity to induce a robust IgG specific response. However, problems remain regarding incorporating more selectable BCEs from a given antigen. Therefore, it is crucial to reveal all BCEs on a target antigen for the development of an effective multivalent peptide vaccine in future.
Saijo et al. [15] previously reported that high titer antisera from CCHF patients can react with NP 201-306 intermediate segments, which are highly conserved in many CCHFV strains. Previously, Wei et al. [22] also found that NP 237-305 was a dominant antigenic region. Burt et al. [34] identified the antigenicity of the region N 182-195 , and Liu et al. [19] performed a fine IgG-epitope analysis of the CCHFV-NP2 segment (NP 237-305 ). Although these studies confirm that the central region of CCHFV-NP is highly antigenic, the N-(NP  ) and C-terminals (NP 286-482 ) of CCHFV-NP also have antigenicity (Fig 2). However, there are no previous reports of epitope identification in the 1-181 and 306-486 regions of CCHFV-NP. Carter et al. [10] found that the N-and C-terminals of the head region of CCHFV-NP were involved in RNA binding. This region may play an important role in the replication of the virus or the activation of the host's early humoral immunity [12].
Epitope mapping is the key to basic and applied research in the field of virology [35]. Compared to other techniques such as r-DNA [36], peptide synthesis [37], and peptide/protein display technology [38], the BSP method is feasible and has been successfully used in the identification of certain target antigen epitopes, such as the epitopes of human egg zona pellucida protein [20], the nucleoprotein of Peste des petits ruminants virus (PPRV) [35], and the E6, E7 and L1 proteins of human papillomavirus type 58 [30]. The BSP method is commonly used to expressmultiple polypeptide segments of 8-20 aa residues, which cover all of a target protein [21]. This further confirms that the BSP method is a reliable epitope scanning mapping strategy.
In this paper, to carry out a comprehensive linear epitope scan of CCHFV-NP, the modified epitopes of N-terminal NP1 and C-terminal NP3 segments were identified using a modified BSP method. First, we successfully synthesized 48 overlapping 16mer peptides and 58 overlapping 8mer peptides, and we identified nine minimal BCE linear motifs by western blotting with pAbs. The sequences of 12 strains of CCHFV from different countries and lineages were analyzed. The results showed that six epitopes, E3, E4, E5a, E5b, E6 and E7, were 100% conserved among the 12 CCHFV strains, while the epitopes of E1a, E1b and E2 showed homologies of 93.75%, 96.67% and 97.14%, respectively. Burt and colleagues [34] predicted the antigen epitopes of CCHFV-NP and synthesized more than 60 peptides according to the results of a software analysis. Thereafter, the antigenicity of these peptides was tested with patient sera using enzyme-linked immunosorbent assay (ELISA). Their results showed that the "NRGGDENPRGPVSR" polypeptide at the aa residue position of NP 182-195 reacted with 13 of 16 human serum samples and had high antigenicity. Our study identified the epitope E1b ( 184 GGDENP 189 ), which is consistent with results from Burt et al [34].
The NP 3D structure and the exact sites of the nine smallest epitope motifs of CCHFV-NP were analyzed using PyMOL™ software. The results showed that only E1b ( 184 GGDENP 189 ) is located on unordered loop with no electron density, according to the 3D structural simulation analysis. The other eight motifs were exposed on the surface of the 3D structure, in the helixangle-helix region (Fig 7b), indicating that they could easily bind to antibodies. The epitopes located on the surface of the target protein play an important role in the future development of drugs that interact with target antigens. Wadood and colleagues [39] first designed epitopebased drugs related to the dengue virus envelope protein and then successfully screened for drugs that interact with viral epitopes. Their results provide novel and potential drugs for the treatment of dengue fever.
Epitopes identified in this study are likely to be highly species-specific. Epitopes E3, E4, E5a, E5b, E6 and E7 in the NP3 segment of 12 strains from different lineages (Fig 8) and 11 strains from China were highly conserved, with a similarity of 100% (S3 Fig), indicating that the six epitopes on the NP3 segment may be conserved epitope sequences of CCHFV. Whether the minimal antigenic epitopes identified in this study can induce the long-term production of neutralizing antibodies requires further studies, as CCHFV is classified as a BSL4 pathogen. This indicates that CCHFV may cause serious public health threats. Studies on the identification of neutralizing antibodies and other experimental studies of CCHFV are limited [40]. Therefore, future work also needs to establish suitable cell and animal models to further identify the neutralizing activity related to these epitopes.
In conclusion, nine fine epitope motifs were obtained and analyzed for antigenicity, which provided new data for the development of a CCHFV multi-epitope detection kit and a multiepitope peptide vaccine. The results also provide a theoretical basis for the elucidation of CCHFV pathogenesis and immune defense mechanisms.
Supporting information S1 Table. 16mer  To determine the sensitivity of the antigen-antibody reaction involving the 8mer peptides, quantitative analyses were performed using the same quantity of peptides for detection. The relative grayscale level of each 8mer peptide compared to the positive control, CP (16mer peptide P38, identified as positive by pAbs), was analyzed according to the results in Fig 4. Statistical analysis of data was performed using one-way analysis of variance (ANOVA) to determine the significant differences using SPSS software. Letters (a, b, c) indicate the significant differences (P<0.05).