Upregulation of the Chemokine Receptor CCR2B in Epstein‒Barr Virus-Positive Burkitt Lymphoma Cell Lines with the Latency III Program

CCR2 is the cognate receptor to the chemokine CCL2. CCR2–CCL2 signaling mediates cancer progression and metastasis dissemination. However, the role of CCR2–CCL2 signaling in pathogenesis of B-cell malignancies is not clear. Previously, we showed that CCR2B was upregulated in ex vivo peripheral blood B cells upon Epstein‒Barr virus (EBV) infection and in established lymphoblastoid cell lines with the EBV latency III program. EBV latency III is associated with B-cell lymphomas in immunosuppressed patients. The majority of EBV-positive Burkitt lymphoma (BL) tumors are characterized by latency I, but the BL cell lines drift towards latency III during in vitro culture. In this study, the CCR2A and CCR2B expression was assessed in the isogenic EBV-positive BL cell lines with latency I and III using RT-PCR, immunoblotting, and immunostaining analyses. We found that CCR2B is upregulated in the EBV-positive BL cells with latency III. Consequently, we detected the migration of latency III cells toward CCL2. Notably, the G190A mutation, corresponding to SNP CCR2-V64I, was found in one latency III cell line with a reduced migratory response to CCL2. The upregulation of CCR2B may contribute to the enhanced migration of malignant B cells into CCL2-rich compartments.


Introduction
Epstein-Barr virus (EBV) is a human gammaherpesvirus 4, which establishes a life-long latent infection in memory B cells and can be reactivated under immunosuppression [1]. The virus is associated with various cancers, including B-cell malignancies such as Burkitt lymphoma (BL), a set of diffuse large B cell lymphomas (DLBCL), and post-transplant and immunodeficiency-related lymphomas (reviewed in [2][3][4][5][6]). EBV encodes several latent proteins and can affect the expression of different cellular genes depending on its latency program. Four EBV infection latency types are known (I, IIA, IIB, and III) [7][8][9][10]. EBV-encoded nuclear protein 1 (EBNA1), which is expressed in all EBV latency programs, is responsible for episome replication and the maintenance of latent infection. EBNA1 can upregulate the STAT1 (signal transducer and activator of transcription 1) protein and

Cell Lines
Two sets of isogenic BL cell lines from the cell line collection at MTC, Karolinska Institute (Stockholm, Sweden) were studied. The Mutu cell lines were generated from an EBV-carrying early passage BL cell line by in vitro culture and clone selection. Mutu cl.148 with EBV latency I had a group I phenotype, while Mutu cl.99 with EBV latency III had a group III phenotype [16]. Mutu III was derived from the latency I Mutu clone cell line that acquired EBV latency III in vitro [28]. Mutu cl. 30, the EBV-negative BL line, was also derived from the parental latency I Mutu cells [29]. The EBV-positive Mutu cell lines carry EBV strain type 1. The BL cell lines, Jijoye M13 (latency I) and Jijoye P79 (latency III), carry EBV strain type 2 [29,30]. Cells were cultured in Iscove's modification of Dulbecco's medium (IMDM) with 10% fetal bovine serum (FBS) and harvested for analyses on the following day after cell splitting.

Detection of the CCR2 and EBV Gene Expression Using Real-Time RT-PCR
Total RNA was isolated from the cell lines using TRI Reagent (Sigma-Aldrich, Steinhem, Germany). A cDNA was transcribed with the random (N) 6 primer using a cDNA synthesis kit (Thermo Fisher Scientific, Waltham, MA, USA). The resulting cDNA (corresponding to 100 ng of total RNA equivalent) was amplified in triplicate, using PerfeCTa SYBR Green FastMix (Quanta BioSciences Inc., Beverly, MA, USA), with the CCR2-specific primers overlapping the first and second exons and also with the specific primers for the house-keeping gene TBP (encoding TATA box-binding protein), as described previously [27]. The expression of two EBV-encoded latent genes (EBNA2 and LMP1) was examined as described above for CCR2. The real-time PCR analyses were performed using the CFX96 Touch Real-Time PCR Detection System (Bio-Rad Laboratories Inc., Richmond, CA, USA). The primer sequences for EBNA2 type 1 and LMP1 were described previously [31]. The EBNA2 transcript in the Jijoye M13 and Jijoye P79 cell lines was determined using the EBV type 2-specific primers [32]. Melting curve analysis was performed to confirm the specific products. The CCR2A and CCR2B ORF-transcripts were determined using RT-duplex-PCR with two pairs of primers, flanking the CCR2A or CCR2B ORF and corresponding to the housekeeping gene GAPDH. We described the RT-duplex-PCR protocol and primer sequences previously [27]. We also published the primers used for the amplification of the CCR3 (exon 1-4 in both variants 1 and 2) and CCR5 (exon 2-3) transcripts previously [33]. For RT-duplex-PCR, purified total RNA was reverse transcribed using RevertAid First Strand cDNA Synthesis Kit with the oligo(dT) 18 primer (Thermo Fisher Scientific).
The CCR2B ORF PCR-products from BL cell lines were gel-purified and cloned using the TOPO TA cloning kit (Thermo Fisher Scientific). To prevent clonal amplification of sequences, the transformed competent cells were plated immediately after the heat shock, excluding shaking bacteria for 1 hour. Five clones for each PCR product were sequenced using the BigDye Terminator sequencing kit (v. 3.1; Thermo Fisher Scientific). The mutation analysis of CCR2B ORF was performed by alignment of the obtained sequences against the CCR2B mRNA RefSeq (NM_001123396.2). Nucleotide substitutions were accepted when they were present in the same position in all five clones.

Detection of CCR2 Using Western Blot Analysis
Total cell lysates in Laemmli buffer were prepared: each well was loaded with the lysate made of 2 × 10 5 cells. Proteins were separated using SDS polyacrylamide gel electrophoresis and transferred onto a nitrocellulose membrane. The CCR2 protein was detected using immunoblotting with the rabbit monoclonal anti-CCR2 antibody (clone E68; Novus Biologicals, Littleton, CO, USA). The LCL that was established by infection of ex vivo PB B cells with the EBV strain B95-8, and the human myeloid histiocytic lymphoma cell line U937 was used as the positive control. For detection of the EBV antigens EBNA2 and LMP1, mouse monoclonal antibody PE-2 (anti-EBNA2; DAKO, Copenhagen, Denmark) and NCL-EBV-CS1-4 (anti-LMP1; Novocastra Laboratories, Newcastle upon Tyne, UK) were used.

Detection of CCR2B Using Fluorescent Immunostaining Analysis
The cells were fixed in the methanol-acetone mixture (1:1) at −20 • C. Immunostaining with the mouse monoclonal anti-CCR2B antibody (CKR-2B, sc-74490; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and the FITC-conjugated rabbit anti-mouse antibody was performed as described previously [17]. Hoechst 33258 (Sigma-Aldrich) was added at a concentration of 0.4 µg/mL in the secondary antibody solution for DNA nuclear counterstaining. The Eclipse 80i Nikon microscope (Nikon Instruments Europe, Amsterdam, The Netherlands) was used to capture the images.

Cell Migration Assay
A cell migration assay was performed using the HTS Transwell-96 Well Permeable Support System (Corning Incorporated Life Sciences, Cambridge, MA, USA), with pore size 8 or 5 µm. Before the chemotactic migration experiment, cells were washed and cultured in a serum-free medium at 37 • C in 5% CO 2 for 3 h. 10 5 cells in 100 µL of the chemotaxis medium (CM, phenol-red, and serum free IMDM medium supplemented with 0.5% BSA) were loaded into each upper well to account for the chemotactic migration of cells toward different concentrations of human recombinant MCP1/CCL2 (BioLegend, San Diego, CA, USA): 50, 100, 200, and 1000 ng/mL of CM (100 µL). CM and 7% FBS in CM were used as negative and positive controls. After incubation at 37 • C in 5% CO 2 for 4 h, the cells were collected in the lower wells were assessed using the MTT test (Vybrant MTT Cell Proliferation Assay Kit; Thermo Fisher Scientific). GraphPad Prism software (v.7; GraphPad Software, La Jolla, CA, USA) was used for multiple comparisons of nonparametric criteria for the experimental data. Two-way ANOVA was applied to analyze the mean differences between the relative migration of the cells and two independent variables, i.e., the CCL2 concentration and cell type.

Results
The CCR2 mRNA expression was detected using real-time RT-PCR with the specific primers for both isoforms in the EBV-positive BL cell lines with latency III, while the latency I BL cell lines showed very low expression levels ( Figure 1). The CCR2 mRNA expression (the CCR2/TBP mRNA expression ratio) in the Mutu cell lines with latency III, Mutu III and Mutu cl.99, was 806-and 392-fold higher, respectively, compared to the latency I Mutu cl.148 cell line. The CCR2/TBP expression ratio in the Jijoye cells with latency III, Jijoye P79, was 1495-fold higher in comparison with the latency I Jijoye M13 cells. At the same time, the EBNA2/TBP mRNA ratio in Mutu III and Mutu cl.99 cells were 132and 74-fold higher than the ratio in Mutu cl.148 with latency I, respectively; in Jijoye P79, a 4.8-fold increase only compared to that in latency I Jijoye M13. The LMP1/TBP mRNA ratio was 8.7-fold higher in Mutu III cells compared to Mutu cl.99 cells. The LMP1 transcript was not found in Mutu cl.148 cells (latency I). The LMP1 mRNA expression was determined in both Jijoye cell lines, although the LMP1/TBP mRNA ratio was 11.7-fold higher in the latency III Jijoye P79 cells. We should note that the BL cell line Jijoye M13 was previously characterized as an EBV-positive BL cell line with the latency I program [30]. In the present study, the EBNA2 and LMP1 transcripts were detected at low level in this cell line, indicating a shift towards latency III. Nonetheless, CCR2 mRNA expression was correlated with EBNA2 expression (Figure 1).
The CCR2B ORF-transcript was determined in all BL cell lines with latency III using RT-duplex-PCR (Figure 2A). The CCR2A ORF-transcript was not detected in any of the cell lines in this study. At very low levels, the CCR3 and CCR5 transcripts were found by RT-duplex-PCR in only two latency III BL cell lines, Mutu cl.99 and Jijoye P79, but not in Mutu III ( Figure 2B). Thus, this excludes their possible involvement in the chemotactic response to CCL2. The analysis of sequences of the CCR2B ORF-transcripts revealed a single-nucleotide nonsynonymous mutation (G190A) resulting in substitution of valine to isoleucine (V64I) in one of the Mutu cell lines, Mutu cl.99. This mutation corresponds to the known single-nucleotide polymorphism (SNP) CCR2-V64I.
Viruses 2018, 10, x 5 of 11 The analysis of sequences of the CCR2B ORF-transcripts revealed a single-nucleotide nonsynonymous mutation (G190A) resulting in substitution of valine to isoleucine (V64I) in one of the Mutu cell lines, Mutu cl.99. This mutation corresponds to the known single-nucleotide polymorphism (SNP) CCR2-V64I.  The immunoblotting analysis showed that all BL cell lines with latency III expressed the CCR2 protein ( Figure 3A). A very weak signal can be also seen in the cells with latency I. Immunostaining demonstrated the membrane localization of the CCR2B protein in the latency III cells ( Figure 3B). The analysis of sequences of the CCR2B ORF-transcripts revealed a single-nucleotide nonsynonymous mutation (G190A) resulting in substitution of valine to isoleucine (V64I) in one of the Mutu cell lines, Mutu cl.99. This mutation corresponds to the known single-nucleotide polymorphism (SNP) CCR2-V64I.  The immunoblotting analysis showed that all BL cell lines with latency III expressed the CCR2 protein ( Figure 3A). A very weak signal can be also seen in the cells with latency I. Immunostaining demonstrated the membrane localization of the CCR2B protein in the latency III cells ( Figure 3B). The immunoblotting analysis showed that all BL cell lines with latency III expressed the CCR2 protein ( Figure 3A). A very weak signal can be also seen in the cells with latency I. Immunostaining demonstrated the membrane localization of the CCR2B protein in the latency III cells ( Figure 3B).  To find out whether CCR2 is functional in BL cell lines with latency III, the chemotactic migration (chemotaxis) assay, using the human recombinant chemokine CCL2, was performed ( Figure 4). Migration of the cells with latency I and EBV-negative in response to CCL2 was not determined. Cells from the latency III BL cell lines migrated toward the CCL2, although the Mutu cl.99 cells showed significantly lower, if at all, migration that can be putatively connected with the mutation G190A found within the CCR2B ORF transcript. The reduced migration of the CCR2-positive cells was observed in the presence of CCL2 at high concentrations, which was likely due to the fast saturation and consequent internalization of the receptor [21,22]. To find out whether CCR2 is functional in BL cell lines with latency III, the chemotactic migration (chemotaxis) assay, using the human recombinant chemokine CCL2, was performed ( Figure 4). Migration of the cells with latency I and EBV-negative in response to CCL2 was not determined. Cells from the latency III BL cell lines migrated toward the CCL2, although the Mutu cl.99 cells showed significantly lower, if at all, migration that can be putatively connected with the mutation G190A found within the CCR2B ORF transcript. The reduced migration of the CCR2-positive cells was observed in the presence of CCL2 at high concentrations, which was likely due to the fast saturation and consequent internalization of the receptor [21,22].

Discussion
Our results have shown that only the EBV-positive BL cells with latency III expressed the functional CCR2B protein. Moreover, the CCR2 mRNA level correlated with the EBNA2 mRNA expression. We may speculate that EBNA2 can be directly involved in the CCR2B upregulation in cells with EBV latency III infection. Recently, we have reported that CCR2B, but not CCR2A, was induced in PB B cells upon EBV infection in vitro and maintained its upregulation in the established LCLs with EBV latency III [27]. The EBV-infected latency III B cells can be found in vivo. However, in immunocompetent persons, such cells are eliminated by cytotoxic T cells because of the high immunogenicity of the infected cells [1]. CCR2 is not only the cognate (dominant) receptor for CCL2/MCP1, it binds and responds to several inflammatory chemokines, such as CCL7/MCP3, CCL8/MCP2, CCL13/MCP4, and CCL16/MTN-1, which are secreted by inflammatory macrophages, monocytes, and cytotoxic T cells [34]. We suggested earlier that the expression of CCR2B on EBV-infected B cells with latency III in immunocompetent persons might be involved in recognition and elimination of those cells by cytotoxic immune cells [27]. Indeed, B-cell lymphoproliferative disorders (LPDs), which are associated with the EBV latency III program, are preferentially found in patients with immunodeficiency, e.g., HIV-related and post-transplant lymphoproliferative disease (reviewed in [1,5,6]).
Here, we also report the single-nucleotide mutation corresponding to the known CCR2-V64I SNP that we detected in one BL cell line with latency III (Mutu cl.99). These cells demonstrated significantly lower chemotactic migration toward the CCL2. The correlation between the mutation G190A (CCR2-V61I) and the response to CCL2 remains to be verified. The CCR2-V64I variant in the transmembrane region can limit responses of other receptors, e.g., the CCR2-V64I variant had no influence on the CCR2B expression in T cells while the CXCR4 expression in these cells was slightly lower [35]. The heterodimerization between CCR2 and CCR5 was described previously [35]. Transinhibition of ligand binding for CCR2/CXCR4 was demonstrated in transfected and primary leukocytes. For the CCR2/CCR5/CXCR4 multimer, this effect was shown in transfected cells, primary T cells, and monocytes [36]. The synergistic effect, after the simultaneous activation of two receptors, was observed in embryonic kidney cells that were transfected with plasmids encoding CCR2B and CCR5. Upon the simultaneous stimulation of the transfected cells with CCL2/MCP1 and CCL5/RANTES, Ca 2+ flux was induced by much lower concentrations of these chemokines than were necessary for induction by one chemokine [37]. Therefore, the CCR2-V64I variant might affect responses of other receptors expressed on B cells. Notably, the CCR2-V64I polymorphism, found in DNA isolated from the PB of patients with various cancers, was associated with the increased risk of cancer development [38][39][40].
The CCR2-CCL2 axis has been implicated in the pathogenesis of several diseases, including cancer, allergy, inflammatory, and autoimmune diseases. Enhanced levels of CCL2 in a variety of tumors were associated with cancer progression and increased tumor growth. Not only tumor cells, but also the surrounding non-neoplastic stroma express high levels of CCL2, which mediates the recruitment of T lymphocytes, fully differentiated M2 macrophages, and circulating monocytic cell subsets into the tumor environment (reviewed in [21,25,26]). Macrophages and circulating monocytes constitutively express CCR2 while its expression in T cells is inducible (reviewed in [21,22]). In transient immunosuppression, the EBV-infected CCR2B-expressing latency III B cells may be attracted into the tumor-surrounding stroma, where the virus can spread and infect the neighboring cells.
EBV is transmitted in humans by saliva. The secretion of the virus into saliva is well documented [41,42] (reviewed in [1]), which suggests an involvement of the oral epithelium. Indeed, cell-free EBV enters polarized oral epithelial cells, which can lead to a lytic productive infection [43,44]. Furthermore, EBV can transmigrate across oral epithelial cells by transcytosis; i.e., the transport of virions from one membrane to another through the cytoplasm without infecting cells [45]. Notably, the expression of CCL2 (MCP1) and CCL5 (RANTES) in EBV-negative epithelial keratinocyte cells in vitro can be induced by the CD40 ligation or by ectopic expression of LMP1 [46]. The ligand of the Viruses 2018, 10, 239 9 of 11 constitutively expressed CD40 is presented by T cells. The transmission of EBV to monocytes was demonstrated in vitro by the cocultivation of EBV-infected B cells with monocytes [47], which suggests that EBV-infected B lymphocytes can be the source of a viral infection of intraepithelial macrophages and dendritic cells. The EBV infection of macrophages, monocytes, and dendritic cells has been shown in vitro and in vivo [48,49]. Thus, a possible mechanism of EBV spread is that EBV-infected CCR2-expressing B lymphocytes migrate into the CCL2-expressing oral epithelium where they reproduce the cell-free virus via lytic replication. The virus infects the intraepithelial macrophages and dendritic cells, and, stimulated by factors released by the intraepithelial immune cells, transmigrates through the uninfected oral epithelial cells into the saliva.
We showed that EBV upregulates CCR2B expression in BL cells with latency III. The upregulation of CCR2B in these malignant B cells enhances their migration toward CCL2. Since immunodeficiency-related B-cell LPDs, including the post-transplant lymphoma and subsets of DLBCL, are characterized by EBV latency III, we suggest that the EBV-encoded latency III proteins may be involved in the regulation of CCL2-CCR2 signaling, which is a part of the immune-response machinery, thus affecting the pathogenesis of these malignancies.