Membrane-Associated and Secreted Forms of the Rhesus Macaque Rhadinovirus-Encoded CD200 Homologue and Cellular CD200 Demonstrate Differential Effects on Rhesus Macaque CD200 Receptor Signaling and Regulation of Myeloid Cell Activation

Viral CD200 homologues are encoded by Kaposi’s sarcoma-associated herpesvirus (KSHV) and the closely related rhesus macaque rhadinovirus (RRV). Though RRV viral CD200 (vCD200) has been examined, questions still exist in regard to the ability of this molecule to induce signaling via rhesus macaque CD200 receptor (CD200R) as well as the potential function of a secreted form of vCD200.

ABSTRACT

The CD200-CD200 receptor (CD200R) pathway is involved in inhibition of immune responses, and the importance of this pathway to infectious disease is highlighted by the fact that viral CD200 (vCD200) molecules have been found to be encoded by several DNA viruses, including the human gammaherpesvirus Kaposi’s sarcoma-associated herpesvirus (KSHV) and the closely related rhesus macaque rhadinovirus (RRV). KSHV vCD200 is the most extensively studied vCD200 molecule; however, the only herpesvirus vCD200 molecule to be examined in vivo is that encoded by RRV. Our prior studies have demonstrated that RRV vCD200 is a functional CD200 homologue that is capable of affecting immune responses in vivo, and furthermore, that RRV can express a secreted form of vCD200 (vCD200-Sec) during infection. Despite this information, RRV vCD200 has not been examined specifically for effects on RM CD200R signaling, and the functionality of vCD200-Sec has not been examined in any context. Thus, we developed an in vitro model system in which B cells expressing vCD200 were utilized to assess the effects of this molecule on the regulation of myeloid cells expressing RM CD200R, mimicking interactions that are predicted to occur in vivo. Our findings suggest that RRV vCD200 can bind and induce functional signals through RM CD200R, while vCD200-Sec represents a nonfunctional protein incapable of affecting CD200R signaling. We also provide the first demonstration of the function of RM CD200, which appears to possess more robust signaling capabilities than RRV vCD200, and show that KSHV vCD200 does not efficiently induce signaling via RM CD200R.

IMPORTANCE Viral CD200 homologues are encoded by Kaposi’s sarcoma-associated herpesvirus (KSHV) and the closely related rhesus macaque rhadinovirus (RRV). Though RRV viral CD200 (vCD200) has been examined, questions still exist in regard to the ability of this molecule to induce signaling via rhesus macaque CD200 receptor (CD200R) as well as the potential function of a secreted form of vCD200. Furthermore, all previous in vitro studies of RRV vCD200 have utilized an Fc fusion protein to examine functionality, which does not replicate the structural properties of the membrane-associated form of vCD200 that is naturally produced during RRV infection. In this study, we demonstrate for the first time that membrane-expressed RRV vCD200 is capable of inducing signal transduction via RM CD200R, while the secreted form of vCD200 appears to be nonfunctional. Furthermore, we also demonstrate that RM CD200 induces signaling via RM CD200R and is more robust than RRV vCD200, while KSHV vCD200 does not appear to induce efficient signaling via RM CD200R.

INTRODUCTION

CD200 is a membrane glycoprotein that is capable of inducing signaling in cells expressing its cognate cell surface receptor, CD200 receptor (CD200R). Both CD200 and CD200R are single-transmembrane type-1 proteins of the immunoglobulin (Ig) superfamily, with CD200 containing a short cytoplasmic tail not capable of signaling and CD200R possessing a cytoplasmic tail containing an NPXY signaling motif that becomes tyrosine phosphorylated upon CD200 binding, thus initiating downstream signaling events in cells expressing the receptor (1–3). The CD200-CD200R axis has been extensively studied in mice and humans (4, 5), and in general, it has been found that CD200 is broadly expressed on numerous cell types, while the distribution of CD200R is thought to be limited mainly to myeloid lineage cells (6–8) and some subsets of T cells and B cells (5, 8–10). The overall view of the CD200-CD200R interaction is that of an immune inhibitory mechanism, with signaling events induced in CD200R-expressing immune cells as a result of CD200 binding, leading to an inhibition of cellular activation, decreased cytokine production, and an overall diminishing of inflammatory responses (4, 5).

The importance of the CD200-CD200R axis to the regulation of immune responses during infection with various pathogens has been well documented (11). In particular, the significance of this signaling pathway to the regulation of viral infections is highlighted by the fact that some DNA viruses have been found to encode homologues of cellular CD200, including members of the poxvirus and herpesvirus families (10, 12–17). The most extensively studied viral CD200 (vCD200) molecule is that encoded by human herpesvirus-8 (HHV-8)/Kaposi’s sarcoma-associated herpesvirus (KSHV), a gamma-2 herpesvirus that is commonly associated with development of diseases in humans coinfected with HIV, including Kaposi’s sarcoma (KS), B cell lymphoproliferative disorders (LPD), primary effusion lymphoma (PEL), multicentric Castleman’s disease (MCD), and some non-Hodgkin’s lymphomas (18–21). In vitro studies have demonstrated that KSHV vCD200 is capable of inhibiting the activation of macrophages (14), basophils (10), neutrophils (22), and T cells (23), suggesting that this molecule plays a role in inhibiting the development of immune responses against KSHV in vivo. It has also been shown that KSHV vCD200 can directly inhibit antigen-specific T cells in vitro through binding of CD200R, suggesting that KSHV vCD200 may directly regulate T cells in vivo (23), although other mechanisms such as inhibition of CD200R⁺ antigen-presenting cells may also still be involved.

Due to the lack of an animal model that supports KSHV infection and disease development, KSHV vCD200 has only been examined alone in in vitro models of expression, limiting the ability to precisely define the role of KSHV vCD200 in regulating immune responses during KSHV infection in vivo. Importantly, however, a nonhuman primate (NHP) model of KSHV infection has been firmly established using rhesus macaque rhadinovirus (RRV), an NHP gamma-2 herpesvirus that is closely related to KSHV. RRV was isolated from a rhesus macaque monkey (RM) with lymphoproliferative disorder (LPD) and has been shown through in vivo infection studies to infect and establish latency in B cells and to induce diseases similar to those found in KSHV in infected humans, including an MCD-like B cell LPD, B cell lymphoma, and retroperitoneal fibromatosis (RF) (24–26). Furthermore, like KSHV, RRV encodes a variety of molecules involved in immune regulation within an infected host, including a functional CD200 homologue. Thus, the use of RRV as a model system provides a mechanism to define the precise functions of gammaherpesvirus-encoded vCD200 molecules both in vitro and in vivo and assess their contributions to the viral life cycle and viral pathogenesis.

Previous in vitro studies conducted in our laboratory utilizing an Fc fusion protein have shown that RRV vCD200 is a functional CD200 homologue that is capable of binding to RM CD200R (27) and inhibiting the activation of RM monocyte-derived macrophages (17). Using a bacterial artificial chromosome (BAC)-derived nonsense mutant form of RRV lacking expression of vCD200 (vCD200 N.S.), we have also previously examined the contributions of RRV vCD200 to infection, immune regulation, and pathogenesis during in vivo infection of RMs (27). These studies provided the first analysis of any gamma-2 herpesvirus vCD200 molecule during in vivo infection and demonstrated that RRV vCD200 inhibits T cell responses early after RRV infection and also negatively affects RRV replication. However, although implied by these findings, it has yet to be specifically demonstrated that RRV vCD200 is capable of inducing signaling via RM CD200R to produce inhibitory effects in cells expressing this receptor.

Open reading frame (ORF) R15 encodes RRV vCD200 and, similar to what has been demonstrated for the vCD200-encoding ORF K14 of KSHV, is transcribed as part of a bi-cistronic transcript with the downstream ORF74, which encodes the viral G protein-coupled receptor (vGPCR) (28). Unlike KSHV, in which splicing of the intergenic region of the bi-cistronic K14-ORF74 transcript does not appear to alter the coding sequence of either K14 or ORF74 (29), splicing of the RRV R15-ORF74 transcript results in the production of both a full-length membrane-associated form of vCD200 and a truncated and secreted form of vCD200 (vCD200-Sec) during RRV infection (28). Though secreted CD200-like proteins have been predicted to be produced by some yatapoxviruses and duck adenovirus (13, 30, 31), to date, no secreted forms of any viral CD200 homologues, other than that belonging to RRV, have been specifically identified. RRV vCD200-Sec has never been examined in any context for functionality; thus, assessing the properties of this molecule is important in order to ascertain the possible relevance of this form of the molecule to RRV infection in vivo. Interestingly, a soluble version of human cellular CD200, sCD200, was recently identified in humans and demonstrated to be elevated in plasma of chronic lymphocytic leukemia (CLL) patients (32), where it appears to promote the growth of CLL cells and may serve as a prognostic marker for disease development. Human sCD200 is produced via proteolytic cleavage of CD200 from the cell surface (33) and is capable of inducing human CD200R phosphorylation (34). Overall, this information suggests that the presence of soluble forms of CD200 or CD200-like molecules in vivo can have effects on CD200R signaling that ultimately affect host responses and disease development.

Although RRV vCD200 has been examined in vitro and in vivo, there are still gaps in understanding precisely how this molecule functions. For example, the ability of vCD200 to transduce signals via RM CD200R has not yet been specifically demonstrated. Of further importance, in all previous in vitro studies of RRV vCD200 functionality, a purified soluble vCD200-Fc fusion protein was utilized (17, 27). Although the use of purified vCD200 and cellular CD200 fusion proteins has provided a method with which to assess the properties of these molecules in vitro (8, 22, 31), they do not accurately represent the naturally occurring membrane-associated forms of these molecules that would normally interact with CD200R to induce receptor signaling in vivo. Furthermore, in the case of RRV, the properties of vCD200-Sec have yet to be examined in any manner. As this represents a naturally occurring soluble form of RRV vCD200, the use of a similarly soluble vCD200-Fc fusion protein in in vitro assays does not allow for the direct comparison of the functions of vCD200 and vCD200-Sec or allow for the assessment of the potential effects that vCD200-Sec may have on membrane-associated vCD200 interactions with CD200R. Lastly, the functionality of RRV vCD200 has also never been directly compared to that of its host homologue, RM CD200, which itself has yet to be examined for functionality in any manner.

To further define the properties of RRV vCD200 and the effects of this molecule on RM CD200R, and to provide an initial analysis of the functionality of RRV vCD200-Sec and RM CD200, we generated B cell lines that express these molecules as well as a monocytic cell line expressing RM CD200R. Using these cell lines in coincubation experiments, the ability of RRV vCD200 to induce functional signals via RM CD200R was determined, and the first analyses of the function of vCD200-Sec and cellular RM CD200 were conducted. This approach allows for the assessment of the effects of RRV vCD200 and host RM CD200 expressed in a natural state on B cells on the function of myeloid cells expressing RM CD200R, mimicking interactions that would occur in vivo between vCD200-expressing RRV-infected B cells and CD200R⁺ myeloid cells. Our findings demonstrate for the first time that RRV vCD200 is capable of inducing signaling via RM CD200R in myeloid cells and that this signaling ultimately results in inhibition of cellular activation, although vCD200 is less potent than its cellular counterpart. Furthermore, our analysis suggests that vCD200-Sec may be a nonfunctional molecule that does not possess the ability to transduce signals via RM CD200R and is also incapable of disrupting CD200R signaling induced by interactions with membrane-associated vCD200 or RM CD200. The ability of KSHV vCD200 to induce signaling via RM CD200R was also assessed utilizing a B cell line expressing this molecule.

RESULTS

Generation of B cell lines expressing vCD200 and monocytic cells expressing RM CD200R.

All previous assays performed to assess RRV vCD200 functionality have utilized a form of vCD200 produced as a soluble Fc fusion protein (vCD200-Fc), in which the extracellular domain of vCD200 is fused in frame with the Fc fragment of human IgG1 (17, 27). Although useful for analyzing some aspects of vCD200 function, this form of the molecule does not accurately mimic the membrane-bound form of vCD200 that is expressed on the surface of infected cells (17) and would thus interact in a cell-to-cell fashion with target cells expressing RM CD200R in vivo. Moreover, as a soluble protein, vCD200-Fc more closely resembles the secreted form of RRV vCD200 (vCD200-Sec) that was previously identified as being produced in RRV-infected cells due to an R15-ORF74 transcript splicing event that generates a form of vCD200 lacking a transmembrane domain (28). As RRV naturally infects and establishes latency in B cells in vivo, stable B cell lines were generated to express RRV-vCD200, vCD200-Sec, or RM CD200 for use in assays with monocytic cells engineered to express RM CD200R as targets for binding. This model system mimics interactions of vCD200 expressed on B cells with CD200R expressed on myeloid-lineage cells, as is envisioned to occur during RRV infection in vivo. Due to the lack of a suitable RM B cell line, human BJAB cells were used as a model cell line for generating stable B cells expressing RRV vCD200 and RM CD200 proteins. BJAB cells are an EBV- and KSHV-negative B cell line lacking CD200 expression which have been previously utilized for in vitro studies of RRV infection (35) as well as KSHV vCD200 and human CD200 function (23). Similarly, due to a lack of an RM myeloid cell line, human U937 cells were utilized as a model cell line to generate target myeloid cells to assess RM CD200R signaling induced by interactions with BJAB cells expressing vCD200 or RM CD200 molecules. U937 cells are a CD200R-negative monocytic cell line that have been extensively utilized in studies of human CD200R signaling and function (1, 36).

To generate stable cell lines, lentiviral vectors were designed to express RRV vCD200, RRV vCD200-Sec, RM CD200, and RM CD200R. Sequences for RRV vCD200, RRV vCD200-Sec, and RM CD200R have been previously described (17, 27, 28), while a cDNA clone encoding RM CD200 was isolated from RM peripheral blood mononuclear cell (PBMC) RNA by reverse transcription-PCR (RT-PCR) and found to be 100% identical to the predicted RM CD200 gene sequence (RefSeq accession NM_001257544.1). At the amino acid level, RM CD200 is 95.5% identical and 97.8% similar to human CD200, while RRV vCD200 is 32% identical and 48% similar to RM CD200 and 30.9% identical and 48% similar to human CD200 (Fig. 1A). Due to splicing of the full-length R15-ORF74 transcript during RRV infection, a truncated viral mRNA is also produced and encodes a form of vCD200 with a deletion of 23 amino acids at the C terminus, which are replaced with 6 different amino acids (28). As a result of this truncation, this form of vCD200 lacks the transmembrane domain present in the membrane-bound full-length form of vCD200 (Fig. 1B) and is thus produced as a secreted molecule (vCD200-Sec) in RRV-infected cells (17). In regards to CD200R, RM CD200R is 90% identical and 94.3% similar to human CD200R and, importantly, contains a conserved NPXY signaling motif that has been demonstrated to be necessary for receptor signaling (Fig. 1C) (3).

FIG 1.

Protein sequence alignments and overview of generated cell lines. Protein alignments were performed using ClustalW, with identities indicated by dark gray and similarities by light gray shading. (A) Alignment of RM CD200, human CD200, and RRV vCD200. RM CD200 displays 95.5% identity and 97.8% similarity to human CD200 and 32% identity and 48% similarity to RRV vCD200. (B) Alignment of RRV vCD200 with RRV vCD200-Sec, demonstrating the loss of the transmembrane (TM) domain of vCD200 in vCD200-Sec due to transcript splicing. (C) Alignment of RM CD200R and human CD200R. RM CD200R shares 90% identity and 94.3% similarity to human CD200R and maintains the conserved NPXY signaling motif in the C-terminal tail. (D) Cartoon depicting stable BJAB cell lines generated to express RM CD200, RRV vCD200, and RRV vCD200-Sec and their interaction with CD14⁺ U937 target cells engineered to express RM CD200R. Phosphorylation of the RM CD200R tail is indicated with a P.

Lentivirus was generated using individual vectors expressing RM CD200, full-length RRV vCD200, RRV vCD200-Sec, or empty vector as a control, and the resulting viruses were utilized to transduce BJAB cells (Fig. 1D). After transduction, BJAB cells were selected with hygromycin B and passaged until a stable population of resistant cells was obtained. The resulting stable cell lines expressing full-length RRV vCD200 or vCD200-Sec were analyzed by Western blotting, confirming expression of corresponding proteins in each cell type (Fig. 2A and B). While the full-length membrane-associated form of RRV vCD200 was only detectable in cell lysates from stable cells (Fig. 2A), vCD200-Sec was readily detected in unconcentrated supernatants from cells transduced to express this molecule (Fig. 2B), indicating that the truncated and secreted form of vCD200 is abundantly expressed in supernatants obtained from these cells. No secreted form of vCD200 was detectable in supernatants from cells expressing the full-length membrane-associated form of vCD200. Although the predicted size of vCD200 is 26.7 kDa and that of vCD200-Sec is 25.1 kDa, the observed molecular weights (MW) of these proteins is larger due to glycosylation (17). Western blot analysis of lysates from BJAB cells transduced with lentivirus encoding RM CD200 confirmed expression of this molecule in these cells (Fig. 2C) and also indicated that membrane-associated vCD200 and RM CD200 are similarly expressed in BJAB cells.

FIG 2.

Expression of RRV vCD200 and RM CD200 molecules in stable BJAB cell lines. RRV vCD200 or vCD200-Sec expression in lysates (A) or supernatants (B) of stable BJAB cells were assessed by Western blot analysis using an anti-RRV vCD200 monoclonal antibody. The full-length membrane-associated (F) and secreted (S) forms are indicated. (C) Western blot analysis of the expression of RM CD200 in lysates from stable BJAB cells using anti-CD200 antibody. For lysate samples, equivalent amounts of protein were loaded in each lane, while for supernatant samples, 50 µl of unconcentrated conditioned supernatants was loaded per lane. Blots in panels A and C were stripped and reprobed with anti-GAPDH antibody as a loading control.

To generate stable myeloid cell lines expressing RM CD200R, lentivirus was generated from RM CD200R-expressing vector and utilized to transduce U937 cells (Fig. 1D). Empty vector-transduced U937 cells were also generated as a negative control. To confirm expression of RM CD200R in stable U937 cells, cells were stained with anti-CD200R antibody and analyzed by flow cytometry (Fig. 3A). Compared to those with empty vector U937, >95% of RM CD200R lentivirus-transduced cells stained positive for CD200R, demonstrating that these cells abundantly express RM CD200R on their surface. To confirm that RRV vCD200 is capable of binding RM CD200R expressed on the surface of U937 cells, purified vCD200-Fc fusion protein was labeled with fluorescein isothiocyanate (FITC) and utilized to stain stable U937 cells (Fig. 3B). Compared to empty vector U937, vCD200-Fc protein displayed specific binding to CD200R⁺ U937 cells, with >98% of these cells staining positive for vCD200-Fc, while a FITC-labeled Fc control protein displayed no binding to either empty vector- or RM CD200R-transduced U937 cells (Fig. 3C and D). Taken together, this demonstrates the ability of RRV vCD200 to specifically bind to RM CD200R expressed on the surface of U937 cells.

FIG 3.

vCD200 binds to RM CD200R expressed on U937 cells. (A) Flow cytometry analysis of RM CD200R expression in stable U937 cells. U937 transduced to express RM CD200R or empty vector were stained with anti-CD200R antibody (dark gray) or Ig control (light gray) and gated to determine the percentage of CD200R⁺ cells. (B) Diagram depicting vCD200-Fc staining of CD200R U937 cells. Stable U937 cells transduced with empty vector (C) or RM CD200R-expressing lentivirus (D) were stained with FITC-labeled vCD200-Fc protein (black line with gray peaks) or Fc control protein (dashed line).

Membrane-associated RRV vCD200 induces signaling via RM CD200R but is less potent than RM CD200.

To assess the ability of vCD200 expressed in B cells to induce signaling via RM CD200R expressed in myeloid cells, flow cytometry was utilized to measure overall levels of phospho-Tyr (p-Tyr) in CD200R⁺ U937 cells after coincubation with vCD200-expressing BJAB cells. Induction of signaling via human CD200R has been shown to be initiated via phosphorylation of tyrosine residues in the C-terminal tail, with a tyrosine located in a conserved NPXY motif having been found to be critical for interaction with adaptor molecules such as Dok1 and Dok2 and the ultimate transduction of signals to other downstream molecules (1, 3). Importantly, this NPXY motif is fully conserved in RM CD200R (Fig. 1C). The measurement of p-Tyr levels in CD200R U937 cells thus allows for the indirect assessment of receptor tail phosphorylation resulting from vCD200 binding as well as accompanying elevations in overall cellular p-Tyr levels resulting from the propagation of downstream signaling.

To measure levels of p-Tyr induction in CD200R⁺ U937 cells, BJAB and U937 cells were first serum starved to reduce basal p-Tyr levels and then coincubated at 1:1 ratio for 20 min, conditions that were empirically determined to allow for optimal p-Tyr induction in these assays. No measurable p-Tyr signal was detected using lower (0.5:1) or higher (2:1) ratios of BJAB:U937 cells, due to under- and overstimulation of CD200R, respectively, and similar observations were made using shorter (10 min) or longer (40 min) incubation times with a 1:1 ratio (data not shown). After incubation, cells were immediately fixed and permeabilized and then stained with antibodies directed against CD14 and p-Tyr, before analysis by flow cytometry. As U937 cells are CD14⁺ and BJAB cells are CD14⁻, cells were first gated on CD14 and then gated on p-Tyr, allowing for the measurement of p-Tyr levels exclusively in U937 cells of the mixed BJAB:U937 cultures. In all experiments, the levels of p-Tyr measured in empty vector U937 control cells were subtracted from levels measured in CD200R⁺ U937 cells for each condition in order to discern levels of p-Tyr specifically induced due to CD200R activation. Examination of data from coincubation experiments indicates that little to no p-Tyr induction is observed when empty vector BJAB cells are incubated with CD200R⁺ U937 cells, while vCD200-expressing BJAB are capable of inducing p-Tyr levels to an average of ∼15% above basal levels (Fig. 4). Interestingly, no p-Tyr induction was detected when BJAB cells expressing vCD200-Sec were coincubated with CD200R⁺ U937, suggesting that cells expressing this form of vCD200 are not capable of inducing signaling in CD200R⁺ cells. Although it is possible that a relatively short incubation time of the mixed BJAB:U937 cultures could limit the amount of vCD200-Sec protein that is made available to interact with CD200R, these data suggest that cells expressing vCD200-Sec are not directly capable of inducing signals via RM CD200R. Importantly, RM CD200-expressing BJAB cells are capable of inducing p-Tyr levels in CD200R U937 cells, demonstrating that, as anticipated, RM CD200 is capable of inducing signaling via RM CD200R (Fig. 4). In addition, BJAB cells expressing the RM CD200 molecule induced higher levels of p-Tyr than those expressing vCD200, indicating that in the context of this expression system, RRV vCD200 appears to be less capable of inducing CD200R signaling than its cellular counterpart. These observations are similar to those made in studies of KSHV vCD200 and human CD200 signaling in vitro (23), suggesting that there are variations that may prevent herpesvirus vCD200 molecules from being as robust as their host homologues at inducing inhibitory signaling via CD200R.

FIG 4.

Membrane-associated vCD200 and RM CD200 induce RM CD200R signaling. (A) BJAB cells expressing vCD200, vCD200-Sec, RM CD200, and an empty vector control were coincubated with U937 cells expressing RM CD200R at a 1:1 ratio for 20 min, fixed, stained with antibodies directed against CD14 and p-Tyr, and analyzed by flow cytometry. Representative plots of triplicate RM CD200R U937 samples and the corresponding empty vector U937 control are shown for each BJAB cell type. (B) The percentage of p-Tyr⁺ CD14⁺ CD200R U937 cells detected in each sample after incubation with each BJAB cell type was subtracted from the percentage of p-Tyr⁺ CD14⁺ cells obtained with empty vector control U937, and the average from triplicate samples was then calculated. Error bars indicate standard deviations. The difference in levels of p-Tyr induction between vCD200 and RM CD200 BJAB cells was found to be statistically significant by unpaired t test. ***, P = 0.0002.

Although KSHV vCD200 is a human virus and thus would not normally interact with RM CD200R in vivo, we nevertheless generated BJAB cells expressing KSHV vCD200 to assess whether or not this molecule could induce RM CD200R p-Tyr signaling and to compare the signaling properties of this molecule to that of RRV vCD200. These cells were confirmed to express KSHV vCD200, though to levels lower than those achieved with RRV vCD200 expression in stably transduced BJAB cells (Fig. 5A). The reason for the apparent differences in expression of vCD200 molecules is unknown but could be attributed to factors such as variations in translation efficiencies, protein stability, or sensitivity of the different antibodies used in Western blotting for the detection of each protein. A direct comparison of the two BJAB cell lines expressing vCD200 indicates that KSHV vCD200-expressing cells are capable of inducing some level of signaling via RM CD200R, although levels of p-Tyr induction were significantly lower than those observed with cells expressing RRV vCD200 (Fig. 5B). In general, these data suggest that differences exist between RRV and KSHV vCD200 molecules in regard to their ability to induce RM CD200R signaling.

FIG 5.

Comparison of RRV vCD200 and KSHV vCD200 RM CD200R signaling. (A) BJAB cells expressing KSHV vCD200 were analyzed by Western blotting using a KSHV vCD200-specific antibody; the blots were stripped and reprobed for GAPDH as a loading control. (B) BJAB cells expressing KSHV vCD200, RRV vCD200, or an empty vector control were coincubated in triplicates with U937 cells expressing RM CD200R at a 1:1 ratio for 20 min, fixed, stained with antibodies directed against CD14 and p-Tyr, and analyzed by flow cytometry. The percentage of p-Tyr⁺ CD14⁺ CD200R U937 cells detected in each sample after incubation with each BJAB cell type was subtracted from the percentage of p-Tyr⁺ CD14⁺ cells obtained using empty vector control U937, and the average from triplicate samples was calculated. Error bars indicate standard deviations. The difference in levels of p-Tyr induction between RRV vCD200 and KSHV CD200 BJAB cells were found to be statistically significant by unpaired t test. ***, P = 0.0005.

Effects of vCD200-induced RM CD200R signaling on activation of and cytokine secretion from myeloid cells.

To determine if RRV vCD200 or RM CD200-induced CD200R signaling has measurable effects in regard to suppression of cellular activation, experiments were performed in which lipopolysaccharide (LPS)-stimulated U937 cells were coincubated with BJAB cells expressing vCD200, vCD200-Sec, or RM CD200 molecules. Specifically, U937 cells were treated with phorbol-12-myristate-13-acetate (PMA) for 72 h to induce differentiation into LPS-responsive cells, before coincubation with BJAB cells and stimulation with LPS for 24 h (Fig. 6A). LPS treatment resulted in the robust induction of tumor necrosis factor (TNF) secretion from PMA-treated U937 cells (Fig. 6B), while BJAB cells were unresponsive to LPS and did not produce TNF (data not shown). After BJAB:U937 coincubation and LPS stimulation, supernatants were collected and analyzed by enzyme-linked immunosorbent assay (ELISA) for TNF levels. Although TNF secretion in LPS-treated U937 cells lacking RM CD200R expression was not affected by coincubation with BJAB cells expressing vCD200, vCD200-Sec, or RM CD200, LPS-treated U937 cells expressing CD200R displayed a significant reduction in TNF secretion in the presence of BJAB cells expressing RRV vCD200 and RM CD200 (Fig. 6C). Specifically, full-length RRV vCD200-expressing BJAB cells induced a 27% reduction in TNF levels (P = 0.011, unpaired t test), while BJAB cells expressing RM CD200 induced a 51% reduction in TNF levels (P = 0.003, unpaired t test) compared to that in BJAB cells transduced with empty vector. Although BJAB cells expressing vCD200-Sec induced a slight reduction in TNF levels compared to that in control empty vector BJAB cells, this difference was not found to be significant. Interestingly, the observed difference in activity levels displayed between cells expressing RRV vCD200 and RM CD200 is statistically significant (P = 0.019, unpaired t test), indicating that although both RRV vCD200 and RM CD200 are capable of suppressing the activation of myeloid cells expressing RM CD200R, vCD200 is less potent than its cellular counterpart. The observed functional effects parallel the signaling patterns induced by both molecules in CD200R⁺ U937 cells (Fig. 4), indicating that CD200R signaling levels correlate with the observed levels of inhibition of cellular activation induced by LPS. Finally, as also suggested by p-Tyr staining, B cells expressing RRV vCD200-Sec appear unable to induce signals via CD200R that result in the inhibition of myeloid cell activation, as vCD200-Sec BJAB cells are unable to inhibit TNF production from LPS-stimulated U937 cells, even with a lengthened coincubation time of 24 h.

FIG 6.

Effects of RRV vCD200 on cellular activation. (A) Overview of LPS stimulation assays. U937 cells were treated with 250 ng/ml PMA for 72 h to induce differentiation into LPS-responsive macrophage-like cells, before coincubation with BJAB cells at a 1:1 ratio, and were stimulated with LPS at 1,000 ng/ml for 24 h. After 24 h, supernatants were collected from each well and analyzed for TNF levels by ELISA. (B) Control experiment demonstrating that LPS treatment induces TNF production in both empty vector and RM CD200R U937 cells. (C) Effects of RRV vCD200-, vCD200-Sec-, and RM CD200-expressing BJAB cells on TNF production in LPS-stimulated U937 cells. Averages from duplicate ELISA wells are shown, and error bars indicate standard deviations. RRV vCD200 BJAB (*, P = 0.011, unpaired t test) and RM CD200 BJAB (**, P = 0.003, unpaired t test) both induce significant reductions in TNF production from LPS-stimulated U937 cells expressing RM CD200R compared to that from BJAB cells transduced with empty vector, while vCD200-Sec BJAB cells have no significant effect on TNF production. U937 cells lacking expression of RM CD200R are unresponsive to BJAB cells expressing vCD200 or RM CD200, demonstrating the requirement of CD200R signaling for inhibition. The data presented are from a representative assay of three independent experiments with similar results.

Though vCD200-Sec-expressing BJAB cells were found to be unable to activate CD200R, it remained a possibility that exogenous vCD200-Sec might still be capable of regulating or disrupting vCD200 or RM CD200 signaling via CD200R. Thus, assays were performed to assess the ability of supernatants from vCD200-Sec cells to affect CD200R signaling. Specifically, conditioned supernatants were collected from vCD200-Sec-expressing BJAB cells and added to coincubated BJAB and U937 cells to determine the effects of this protein on p-Tyr levels in CD200R U937 cells (Fig. 7A). Conditioned supernatants were confirmed to contain vCD200-Sec by Western blot analysis, demonstrating that this protein is abundantly expressed in conditioned medium collected from these cells (Fig. 7B). Examination of p-Tyr levels indicated that the addition of conditioned vCD200-Sec supernatants had no measurable effect on the ability of vCD200- or RM CD200-expressing B cells to induce signaling in CD200R⁺ U937 cells and also did not promote the ability of empty vector or vCD200-Sec BJAB cells to induce CD200R signaling (Fig. 7C).

FIG 7.

Exogenous vCD200-Sec does not induce signaling via RM CD200R. (A) vCD200-Sec conditioned supernatants were incubated with BJAB and U937 cells to assess the effects of exogenous vCD200-Sec on CD200R signaling in U937 cells. (B) Unconcentrated conditioned supernatants collected from empty vector and vCD200-Sec BJAB cells were analyzed by Western blotting. Fifty microliters of each sample was loaded on a 10% acrylamide gel, transferred to membrane, and probed with antibody directed against RRV vCD200. (C) Each BJAB cell type was resuspended in 50 µl of conditioned supernatant from vCD200-Sec BJAB cells and then coincubated for 20 min with U937 RM CD200R cells in 50 µl of serum-free medium at a 1:1 ratio. Empty vector U937 cells were used as a control. After incubation, cells were fixed, stained with antibodies directed against CD14 and p-Tyr, and analyzed by flow cytometry. For each BJAB cell type, background p-Tyr levels obtained with empty vector U937 were subtracted from values obtained for triplicate samples incubated with CD200R U937, and averages were calculated. (D) CD200R U937 cells were treated with 250 ng/ml PMA for 72 h to induce differentiation into LPS-responsive macrophage-like cells, stimulated with LPS at a concentration of 1,000 ng/ml, and then incubated at a 1:1 ratio with BJAB cells expressing vCD200, vCD200-Sec, RM CD200, or empty vector control BJAB in the presence of 25% final volume of conditioned supernatants from empty vector or vCD200-Sec BJAB cells. After 24 h, supernatants were collected and analyzed by ELISA for levels of TNF. Each condition was assayed in duplicates, and average values are shown. Error bars indicate standard deviations. The level of inhibition of TNF production obtained with each BJAB cell type was directly compared to TNF levels obtained with empty vector control BJAB cells for the same condition, and lines with asterisks indicate a significant difference was detected between the indicated samples by unpaired t test (vCD200 BJAB plus empty vector supernatants, P = 0.0330; vCD200 BJAB plus vCD200-Sec supernatants, P = 0.0379; RM CD200 BJAB plus empty vector supernatants, P = 0.0132; RM CD200 BJAB plus vCD200-Sec supernatants, P = 0.0266). ns, not significant.

Despite an inability to induce CD200R signaling, the lack of any functional effects of vCD200-Sec on activated CD200R⁺ U937 cells was further assessed by performing LPS stimulation of coincubated BJAB:CD200R U937 cells in the presence of exogenous conditioned supernatants from vCD200-Sec or empty vector BJAB cells. These data demonstrate that the presence of vCD200-Sec in mixed cultures did not alter the ability of vCD200- or RM CD200-expressing BJAB cells to inhibit the activation of LPS-stimulated CD200R⁺ U937 cells and, as anticipated, did not promote or enhance the level of inhibition obtained with either cell type (Fig. 7D). Furthermore, a similar pattern of inhibition was obtained with vCD200 BJAB and RM CD200 BJAB cells in the presence of either empty vector- or vCD200-Sec-conditioned supernatants, as was observed in CD200R U937 cells in coincubation experiments performed in the absence of exogenous supernatants (Fig. 6C). Taken together, these data indicate that exogenous vCD200-Sec is not capable of inducing signals via CD200R nor does it affect the ability of RM CD200 or vCD200 to signal via CD200R, further suggesting that this form of RRV vCD200 is nonfunctional in these assays.

DISCUSSION

In this study, we directly compared the functionality of membrane-associated RRV vCD200 with that of vCD200-Sec as well as RM CD200, thus, better defining and comparing the biological properties of these viral and cellular CD200 proteins. We also provide the first direct analysis of the ability of RRV vCD200 to induce signaling via RM CD200R as well as the first assessment of the functionality of the RM CD200-CD200R signaling axis. Our findings demonstrate that membrane-associated RRV vCD200 expressed in B cells is capable of inducing signaling via RM CD200R expressed in myeloid target cells and, further, that the viral CD200 protein is less potent than cellular RM CD200 at inducing CD200R signaling. In addition, unlike vCD200 encoded by RRV, which naturally infects RM, vCD200 encoded by KSHV was found not to induce robust signaling via RM CD200R. Having established this system to analyze vCD200-CD200R signaling in the context of cell-cell interactions, we will now be able to further dissect the downstream pathways directly affected by RM CD200R signaling in myeloid cells and also assess the ability of RRV vCD200 to affect cell activation induced by a variety of stimuli.

The observation of functional differences between RRV vCD200 and RM CD200 is similar to what has been observed with KSHV vCD200 and human CD200 (23) and could be attributed to various factors, such as differences in protein expression levels or receptor binding efficiency. Regardless, this finding may indicate that RRV and KSHV have both evolved to encode vCD200 molecules with decreased activity compared to that of their cellular counterparts in their respective hosts, suggesting that expression of a CD200-like molecule with a high level of CD200R-inhibitory activity might be detrimental to these viruses in some fashion. For example, given that vCD200 expression has a negative effect on RRV viral loads in vivo, it is plausible that this molecule has evolved to remain capable of inhibiting immune responses in certain microenvironments while also being detuned to some degree to prevent major inhibitory effects on viral replication. Current work in our lab involves the development of a RM CD200 chimeric form of RRV to assess the effects of the viral expression of cellular CD200 on RRV replication and the development of anti-RRV immune responses in vivo.

RRV is capable of producing a secreted form of RRV vCD200 (vCD200-Sec) during infection. Examination of the function of vCD200-Sec indicates that this form of the molecule possesses no apparent ability to induce signals via RM CD200R in our cell assay system and that it is also incapable of disrupting or modulating the normal interaction of RRV vCD200 or cellular RM CD200 with CD200R. Taken together, this suggests that vCD200-Sec is likely a nonfunctional molecule in regard to CD200R regulation. Our results somewhat contradict findings from a study of KSHV vCD200, in which an engineered soluble monomeric version of this molecule was found to be capable of disrupting inhibition induced via human CD200R by both cellular CD200 and KSHV vCD200 (14), though it is important to note that the molecule used in these experiments was generated as a fusion protein with CD4. Furthermore, although we have previously found that a soluble RRV vCD200-Fc fusion protein is capable of binding to and inhibiting cellular activation via CD200R, this molecule exists as a dimer (17), and the ability of a naturally occurring monomeric vCD200-Sec molecule to bind and efficiently cross-link CD200R molecules to induce or disrupt signaling is less likely. In addition to this, the lack of apparent activity of vCD200-Sec could also be due to differential processing of this form of the molecule. Indeed, CD200 and CD200R molecules are heavily glycosylated (37), and though purified vCD200-Fc fusion protein has also been demonstrated to be glycosylated (17), the absence of a transmembrane domain in vCD200-Sec is likely to disrupt normal processing patterns that could ultimately affect the functionality of this molecule.

Though putative secreted CD200-like molecules have also been described as being encoded in the genomes of some poxviruses and an adenovirus (13, 30, 31), no secreted form of any viral CD200 molecule has been directly examined for functionality. However, a soluble version of human CD200, termed sCD200, was recently identified and has been found to be elevated in plasma of patients with CLL (32, 34), suggesting soluble CD200 molecules may play a role in disease development. Although our data suggest that vCD200-Sec produced by RRV is a nonfunctional protein, this finding could be attributable to the context in which vCD200-Sec function is being examined; thus, the possibility still exists that this form of the molecule may have functions in vivo that are currently unknown. Examining the role of vCD200-Sec during in vivo RRV infection utilizing a virus unable to generate this form of the molecule could help shed light onto any unknown contributions of this and other secreted vCD200 molecules to viral infection.

MATERIALS AND METHODS

Establishment of stable cell lines.

Human BJAB B cells and human U937 monocytic cells were maintained in RPMI medium (Corning, Manassas, VA) supplemented with 10% fetal bovine serum and l-glutamine, and the pLVX lentiviral vector system (Clontech, Mountain View, CA) was used to generate stable cell lines. Briefly, RRV ORF R15, which encodes full-length vCD200, was amplified by PCR from an RRV₁₇₅₇₇ R15 cDNA clone using a sense primer containing an SpeI site (5′-CTAACTAGTATTATGTCGGGAGGAATTACATT-3′) and an anti-sense primer containing an XbaI site (5′-ATCTAGATCACATAGACCTATACAAAA-3′), while the truncated form of R15 encoding vCD200-Sec was amplified from an RRV₁₇₅₇₇ cDNA clone derived from the spliced version of the R15 transcript (28) using the same R15 sense primer in conjunction with an anti-sense primer containing a NotI site (5′-TGCGGCCGCTCAAGGCGTCCATGTTGTTGT-3′). Restriction sites are underlined and start and stop codons are in boldface font. cDNA encoding RM CD200 was isolated from RNA obtained from RM peripheral blood mononuclear cells (RM 22933) by RT-PCR using a sense primer containing an XbaI site (5′-GCTCTAGAATGGAGAGGCTGGTGATCAG-3′) and an anti-sense primer containing a BamHI site (5′-TACGGATCCTTAGGGCTCTCGGTCCTGAT-3′). The sequence of RM CD200 cDNA was found to be 100% identical to the predicted sequence for Macaca mulatta CD200 (GenBank accession number AFE67082.1). KSHV ORF K14, which encodes KSHV vCD200, was isolated by PCR from KSHV BAC16 DNA using a sense primer containing an XbaI site (5′-CCTCTAGAACCATGTCTAGCCTCTTCATTTC-3′) and an anti-sense primer containing a BamHI site (5′-CAAGGATCCTCACTGGGTGGATAGGGGGT-3′) utilizing the second ATG start codon in the K14 ORF to express vCD200, as has been previously described (14). All resulting products were cloned into the lentiviral vector pLVX-IRES-Hyg, which was engineered to contain the human EF1α promoter in place of the original cytomegalovirus (CMV) promoter (pLVX-EF1α-IRES-Hyg). An RM CD200R-expressing cDNA clone was isolated as described previously (27) and was subcloned into pLVX-EF1α-IRES-Hyg as an XbaI/NotI fragment. All resulting plasmid clones were sequenced to confirm their correct identity before use in lentivirus production. Replication-defective recombinant lentivirus was produced in HEK 293T/17 cells and used to transduce BJAB or U937 cells. After transduction, cells were selected with hygromycin B at a concentration of 200 µg/ml, and resulting stable cell populations were maintained under selection.

Stable RRV vCD200 and vCD200-Sec BJAB cells were examined for expression by Western blot analysis using a previously described mouse monoclonal antibody directed against RRV vCD200 (17). RM CD200 expression in stable BJAB cells was examined via Western blot analysis using a cross-reactive anti-human CD200 antibody (SAB1401248; Sigma-Aldrich, St. Louis, MO). Expression of KSHV vCD200 in stable BJAB cells was assessed by Western blotting using an anti-KSHV vCD200 mouse monoclonal antibody (anti-V8 mAb59), which was generated by immunizing mice with purified vCD200-Fc protein containing the extracellular domain of KSHV vCD200 (amino acids [aa] 78 to 305) fused to the Fc fragment of human IgG1 (38). Lysates were prepared by pelleting and resuspending cells in radioimmunoprecipitation assay (RIPA) buffer (1× phosphate-buffered saline [PBS], 1% NP-40, 0.1% sodium dodecyl sulfate, 0.5% sodium deoxycholate) containing protease inhibitor, and equivalent amounts of protein were loaded and run on a 10% acrylamide gel and then transferred to a nitrocellulose membrane. For supernatants, cells were seeded at equal densities in a 25-cm flask, supernatants were collected at 48 h and cleared of cells and debris by centrifugation for 5 min at 900 × g, and protease inhibitor was added to each sample. Next, equivalent volumes of each sample were loaded and run on a 10% acrylamide gel and then transferred to a nitrocellulose membrane. As a loading control for lysate samples, membranes were stripped and reprobed with an antibody directed against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Sigma-Aldrich).

RM CD200R expression in U937 cells was assessed by flow cytometry using a cross-reactive anti-human CD200R antibody conjugated to phycoerythrin (PE) (clone OX-108; BioLegend, San Diego, CA). For each sample, 1 × 10⁶ cells were stained with anti-human CD200R PE antibody or a PE-conjugated mouse IgG1, k isotype control (BioLegend), and gating for positive staining was adjusted above background levels obtained using the control antibody. All samples were acquired using an LSRII instrument, and data were analyzed using FlowJo software (TreeStar, Ashland, OR).

vCD200-Fc staining of CD200R U937.

vCD200-Fc staining of RM CD200R-expressing cells was performed as described previously (27). Briefly, vCD200-Fc protein containing the extracellular domain of RRV vCD200 fused in frame to human IgG1 was produced in CHO cells, purified via high-performance liquid chromatography (HPLC) (17), and labeled with a FITC conjugation kit (Abcam). Human IgG Fc protein (EMD Millipore, Billerica, MA) was labeled with FITC for use as a control. U937 cells were collected, treated with TruStain FcX (BioLegend) to block nonspecific binding to Fc receptors, and then incubated with FITC-labeled vCD200-Fc or Fc control protein. Cells were then fixed, washed with PBS, and analyzed by flow cytometry using an LSRII instrument and FlowJo software.

Flow cytometry for phospho-tyrosine levels.

For phospho-tyrosine (p-Tyr) stimulation assays, stable BJAB and U937 cells were counted, centrifuged, resuspended in serum-free RPMI medium, and incubated for 2 h at 37°C prior to each assay to reduce background p-Tyr levels. After serum starvation, cells were resuspended to a final concentration of 0.5 × 10⁶ cells/50 µl serum-free RPMI medium, 50 µl of each U937 target cell type was seeded into a single well of a 96-well round-bottom plate, and 50 μl of each stimulator BJAB cell type expressing RRV vCD200, RRV vCD200-Sec, RM CD200, KSHV vCD200, or empty vector control was added to each well for a final ratio of 1:1. Cells were coincubated for 20 min at 37°C before fixation with paraformaldehyde (1.5% final concentration per well). Next, cells were centrifuged, washed twice in PBS plus 2% fetal bovine serum (FBS), and permeabilized by addition of 100% cold methanol for 20 min at 4°C. Cells were then collected, washed twice with PBS plus 2% FBS, stained with eFluor 450-conjugated antibody directed against p-Tyr (Thermo Fisher, Waltham, MA) and Alexa Fluor 700-conjugated antibody directed against human CD14 (BioLegend), and analyzed by flow cytometry. Resulting data were gated to determine the percentage of CD14⁺ and p-Tyr⁺ cells in each sample. All samples were acquired using an LSRII instrument, and data were analyzed using FlowJo software (TreeStar). In each assay, individual BJAB cell lines were assayed in triplicates with CD200R⁺ U937 target cells, background levels of p-Tyr staining obtained by incubating each corresponding BJAB cell line with an empty vector U937 control sample were subtracted from each triplicate sample, and the average level of p-Tyr⁺ staining for each cell type was then determined. For assays examining signaling of vCD200-Sec supernatants, conditioned supernatants from vCD200-Sec or empty vector BJAB cells were clarified of cells and cellular debris by centrifugation, and a maximum volume was added to each well to obtain a final concentration of 50% supernatant per well.

TNF ELISA.

U937 cells were treated with phorbol-12-myristate-13-acetate (PMA) (Millipore) at 250 ng/ml for 72 h to induce differentiation into LPS-responsive macrophage-like cells. After PMA treatment, U937 cells were collected and resuspended to 0.5 × 10⁶ cells per ml in RPMI medium plus 10% FBS and coincubated at a 1:1 ratio with an equal number of BJAB cells expressing RRV vCD200, vCD200-Sec, or empty vector control in a 96-well round-bottom plate. Mixed cultures were then stimulated with LPS (MilliporeSigma) at 1,000 ng/ml for 24 h, plates were spun at 740 × g to pellet cells, and supernatants were collected from each well for analysis using a TNF ELISA kit (BioLegend). For each assay, supernatant samples from individual wells were analyzed in duplicates by ELISA. For assays analyzing vCD200-Sec supernatants, 48-h conditioned supernatants from vCD200-Sec or empty vector BJAB cells were clarified by centrifugation at 900 × g for 5 min, and a maximum volume was added for a final concentration of 25% supernatant per well during LPS stimulation.

Protein alignments.

Protein alignments were performed with ClustalW using MacVector software version 14.5.3 (MacVector, Inc.).

Statistical analysis.

Unpaired t tests were performed using Prism software version 8.3.0 (GraphPad Software, LLC).