Introduction
Multiple proteins encoded in herpesvirus genomes are required for origin dependent DNA synthesis, including an origin binding protein, a helicase, a primase, a primase-associated factor, a single-stranded DNA binding protein, a DNA polymerase, and a DNA polymerase processivity factor, see for example [
1]. In order for the DNA polymerase to copy the viral genome during virus replication, it requires the processivity factor, which binds the polymerase and enhances its ability to synthesize full-length products [
2,
3]. There is considerable interest in understanding the biology of the polymerase-processivity factor interaction due to the potential for developing herpesvirus-specific drugs that could inhibit genome replication and the production of progeny virus [
4,
5]. A concerted effort has been made to develop serological reagents that specifically react with each of the different herpesvirus processivity factors to study their cellular localization and trafficking and interactions with other cellular proteins that are critical for virus replication[
6‐
8]. Because of its role in viral DNA synthesis, the processivity factor has also become an important protein marker for virus replication.
Kaposi's sarcoma-associated herpesvirus (KSHV)/human herpesvirus 8, the recently discovered human rhadinovirus belonging to the gammaherpesvirus subfamily, encodes a highly conserved DNA polymerase in open reading frame (ORF) 9 and a DNA polymerase processivity factor in ORF59 [
9]. KSHV ORF59 binds DNA as a homodimer, interacts with the ORF9 DNA polymerase to strongly enhance its ability to synthesize full-length DNA [
6,
10,
11] and is necessary for origin-dependent viral DNA replication [
12]. In comparison to other human herpesviruses, the KSHV processivity factor displays the strongest similarity (29% amino acid identity) with the processivity factor (BMRF1) of Epstein-Barr virus (EBV), the only other known human gammaherpesvirus. More closely-related processivity factors are present in the related New World and Old World primate rhadinoviruses, herpesvirus saimiri (HVS) of the squirrel monkey (30% identity)[
13]) and rhesus rhadinovirus of the rhesus macaque (49% identity) [
14,
15], respectively. The specificity of the interaction between KSHV ORF59 and its DNA polymerase was demonstrated by the inability of the related processivity factors of herpes simplex type 1 (UL42)(20% identity) and p41 of human herpesvirus 6 (19% identity) to functionally replace KSHV ORF59 [
10].
KSHV is now considered to be the cause of Kaposi's sarcoma (KS), pleural effusion lymphoma (PEL) and multicentric Castleman's disease [
16]. Essentially all affected cells in these diseases are latently infected with KSHV and only rare cells have active virus replication [
17]. Similarly, infection of cells
in vitro by KSHV results in the rapid establishment of latency with only rare cells showing productive viral replication. Treatment of latently-infected cells with the phorbol ester, TPA, induces activation of virus replication and production of progeny virus. Using an antibody that recognizes KSHV ORF59, the processivity factor was found to be highly expressed in the nuclei of infected cells that harbor actively replicating virus [
18,
19].
Immunofluorescence studies have localized KSHV ORF59 and other core replication proteins within the viral DNA replication compartments in the nuclei of infected cells [
20]. Deletion analysis of ORF59 has revealed regions critical for binding to the DNA polymerase and double-stranded DNA, as well as a region that enhances the processivity of the DNA polymerase [
6]. Transport of the viral DNA polymerase into the nucleus is dependent on ORF59 [
20], which contains a nuclear localization signal, and binding domains within ORF59 have been identified that are required for nuclear transport of the polymerase [
21]. The reactivity of a monoclonal antibody to KSHV ORF59 with the cognate antigen in the nucleus is an important marker of activation of KSHV replication [
6].
Rhadinoviruses closely-related to KSHV have been identified in a variety of Old World primates, including macaques, gorillas and chimpanzees. Cloning, sequencing and functional characterization of these viruses is of interest due to the structural and functional similarities with KSHV and the possibility of developing an animal model of KSHV pathology. These Old World primate rhadinoviruses segregate into two distinct lineages. The RV1 lineage consists of KSHV in humans, retroperitoneal fibromatosis herpesviruses (RFHV) in different species of macaque [
22], and closely-related viruses in chimpanzees [
23,
24], gorillas [
24] and African Green monkeys [
25]. The RV2 lineage consists of rhesus rhadinovirus (RRV) [
26] and related viruses in different species of macaques [
22,
27,
28], chimpanzee [
29], baboon [
30], gibbons [
31] and drills [
32]. These data suggest that every Old World primate species, including humans, are host to rhadinoviruses of both RV1 and RV2 lineages, although a human RV2 virus has yet to be identified.
Previous studies have shown that RRV, the RV2 rhadinovirus prototype in the rhesus macaque, produces a permissive, replicative infection in cultured rhesus monkey fibroblast cells with obvious cytopathic effects, first evident between 4-7 days, and the production of infectious virus [
26]. A genome-wide transcription profile and a more restricted profile of several key RRV genes have been determined during the time course of infection [
33,
34]. In general, these studies have shown that the transcription of RRV genes after
de novo permissive infection parallels the transcription profile of KSHV genes after activation of latently-infected cells by treatment with phorbol esters or sodium butyrate.
In order to examine the conservation of ORF59 within the RV1 and RV2 rhadinovirus lineages and develop reagents to detect and differentiate RV1 and RV2 permissive infections, we cloned and sequenced the ORF59 homologs of the RV1 and RV2 rhadinoviruses from chimpanzee and three species of macaque and compared these with the KSHV and RRV ORF59 sequences. Sequence comparisons revealed strong conservation between the RV1 and RV2 rhadinoviruses within the majority of the ORF59 coding sequences. However, lineage-specific sequences were identified in the C-terminal domains, and a polyclonal rabbit antiserum was developed against this domain in the RV2 ORF59 proteins. Specificity of this antiserum was demonstrated by Western blot and immunofluorescence analysis. Using this antiserum and ORF59-specific RT-qPCR assays, we demonstrate that RRV and the related RV2 rhadinovirus, MneRV2, from M. nemestrina, both undergo a permissive infection in rhesus primary fetal fibroblast (RPFF) cell cultures and in Vero African green monkey kidney epithelial cells with abundant expression and nuclear localization of ORF59. We also show that RV2 ORF59 expression is coupled with active replication of the viral genome and production of infectious virions. Finally, we demonstrate reactivity of the anti-RV2 ORF59 antiserum within the nuclei of epithelial cells present in the differentiated layer of skin epithelium of a pig-tailed macaque that had been naturally infected with MneRV2. This in vivo data with a macaque RV2 rhadinovirus is congruent with previous in vitro studies, in which expression of ORF59 and other markers of KSHV replication correlated with epithelial cell differentiation, suggesting that differentiated epithelial cells are a specific source of infectious virions in hosts naturally infected with Old World primate rhadinoviruses.
Discussion
We have compared the ORF59 homologs of members of the RV1 and RV2 lineages of KSHV-like rhadinoviruses from chimpanzee and three species of macaques. We obtained the complete sequences of the ORF59 homologs of the RV1 and RV2 rhadinoviruses from chimpanzee (PtrRV1 and PtrRV2) and the cynomolgus macaque (RFHVMf and MfaRV2). These are the only complete gene sequences presently determined for these rhadinovirus species. We also obtained the complete ORF59 sequences of the RV2 rhadinovirus from pig-tailed macaques (MneRV2) and the RV1 homologs from pig-tailed and rhesus macaques (RFHVMn and RFHVMm). We compared the sequences obtained in the current study with the published sequences of ORF59 from KSHV [
9] and the rhesus macaque RRV [
14]. Sequence alignment and phylogenetic analysis of the ORF59 sequences from these viruses confirmed the existence of separate RV1 and RV2 lineages of KSHV-like rhadinoviruses in Old World primates, which was originally demonstrated using fragments of the viral DNA polymerase gene [
22‐
25]. The ORF59 sequence comparisons and phylogenetic analyses strongly support a close evolutionary relationship between KSHV and the RV1 rhadinoviruses from the chimpanzee (PtrRV1) and macaque (RFHVMn, RFHVMm, and RFHVMf), while the rhesus macaque rhadinovirus, RRV, and the other RV2 rhadinoviruses constitute a separate and distinct lineage of KSHV-like rhadinoviruses.
Within the RV1 lineage, the ORF59 sequences of the human and chimpanzee rhadinoviruses, KSHV and PtrRV1, respectively, were closely related, reflecting the close evolutionary relationship of human and chimpanzee hosts. No RV2 rhadinovirus has yet been identified in humans to date, so the relationship between the human and chimpanzee RV2 rhadinoviruses is unknown. In both the RV1 and RV2 lineages, the ORF59 sequences from the
M. mulatta and
M. fascicularis rhadinoviruses clustered together, with the
M. nemestrina viral sequences as a distinct outgroup. Phylogenetic studies have shown that the macaque species
M. mulatta and
M. fascicularis are more closely related to each other than they are to the
M. nemestrina species which evolved in a geographically distinct area [
38]. Thus, the phylogenetic clustering of the viral ORF59 sequences from the different RV1 and RV2 rhadinoviruses mirrors the phylogenetic relationship of their primate hosts, providing strong evidence for co-speciation of the virus and host.
A comparison of the ORF59 sequences revealed a higher conservation among the RV2 rhadinoviruses than was observed for the RV1 rhadinoviruses from the same primate species, even though the RV1 and RV2 viruses appear to have co-evolved within the same host species over the same period of time. The amino acid conservation between the chimpanzee and macaque RV2 rhadinoviruses was 64-65% whereas the conservation between the chimpanzee and macaque RV1 rhadinoviruses was only 56-58%. Similarly, conservation between rhadinoviruses from M. nemestrina and M. mulatta/M. fascicularis macaque species was 88-89% for the RV2 lineage and 82% for the RV1 lineage. Finally, conservation between the M. mulatta and M. fascicularis macaque species was 94.9% for the RV2 lineage and 93.8% for the RV1 lineage. This suggests that the two viral lineages have experienced different evolutionary pressures within the same host that has directly affected the rate of divergence of the ORF59 gene. Functionally, the ORF59 gene family encodes a DNA polymerase processivity factor which is required for copying the viral genome during the replicative cycle of the virus for the production of new virions. Our data suggests that an inherent difference in the biology of the RV1 and RV2 lineages has exerted an effect on the divergence of the ORF59 gene.
Nevertheless, our studies show a strong sequence similarity between the ORF59 sequences of the RV1 and RV2 rhadinoviruses within the N-terminal 300 amino acids with 45% of the amino acids conserved in all of RV1 and RV2 species examined. Our truncation study indicated that the N-terminal domains of the macaque RV2 rhadinovirus ORF59 proteins contain one or more antigenic epitopes recognized in naturally infected macaques. This contrasts with the epitopes of the ORF59 homologs of KSHV and EBV (BMRF1) recognized by mice during development of murine monoclonal antibody reagents, which cluster in the C-terminal ORF59 domain. Although only 6% of amino acid residues in the ~100 amino acid C-terminal domain were conserved across all of the RV1 and RV2 species, a strongly lineage-specific conservation was detected, especially within the macaque RV2 sequences. We developed a rabbit anti-RV2 ORF59 polyclonal antiserum against the conserved C-terminal domain of the RRV and MneRV2 ORF59 proteins. This antiserum reacts specifically with the ORF59 homologs of different macaque RV2 rhadinoviruses and not with ORF59 homologs of RV1 rhadinoviruses or with the related ORF59 homolog (BMRF1) of EBV or macaque lymphocryptoviruses. The antiserum does not react with the ORF59 homolog of the chimpanzee RV2 rhadinovirus due to significant differences in the amino acid sequence of the chimpanzee virus in the targeted C-terminal region.
We tested the immunoreactivity of the commercial monoclonal antibody developed against KSHV ORF59 with the different macaque ORF59 proteins. Recombinant ORF59 protein from the macaque RV1 rhadinovirus, RFHVMn, reacted with the anti-KSHV ORF59 monoclonal antibody. Surprisingly, this antibody also showed reactivity with the MneRV2 ORF59, but not the RRV ORF59. Truncation studies have indicated that residues 279-301 in KSHV ORF59 constitute part of the epitope recognized by anti-ORF59 monoclonal antibody 11D1[
6]. Examination of this region in the ORF59 alignment in Figure
1, revealed an amino acid substitution in aa292 from a Lys in KSHV ORF59 to an Arg in RRV ORF59. All other RV1 and RV2 ORF59 sequences contained the Lys residue in this region of high sequence conservation. These sequence comparisons suggest that the Lys at residue 299 may compose part of the monoclonal epitope that is disrupted by the Arg in the RRV ORF59 sequence.
In this study, we used the rabbit anti-RV2 ORF59 antiserum to analyze RV2 rhadinovirus infections of cultured cells in vitro. Our results demonstrate that significant numbers of fibroblast (RPFF) and epithelial (Vero) cells express ORF59 after infection with RRV or the closely related RV2 rhadinovirus from pig-tailed macaques, MneRV2. ORF59 expression is localized to the nucleus of infected cells in a non-uniform pattern. Our studies indicate that the RV2 infection in these cells spreads from an initial focus of infection, detected as a single ORF59-positive cell, to adjacent cells that over time also become ORF59 positive. Infection of adjacent cells could be due to cell-cell transmission by passage of cell-associated virus through physical contact, or due to a strong concentration gradient of cell-free virus immediately surrounding the infected ORF59-positive cell that results in preferential infection of immediately adjacent cells. However, numerous cases were identified in our micrographs in which an ORF59-negative cell was detected in close proximity to one or more ORF59 positive cells (see cells immediately adjacent to the labeled cells in panels 5F and I). This suggests that direct physical contact between infected and non-infected cells is more important for virus transmission than proximity. We also continued to detect single ORF59-positive cells during the 8 day incubation, indicative of de novo infections in the cell culture by cell-free virus.
KSHV ORF59 is critical for replication of the viral genome within the nucleus [
6,
10‐
12]. This processivity factor has an arginine/lysine rich nuclear localization signal in its C-terminal domain that is responsible for targeting the protein to centers of replication in the nucleus [
20,
21]. We identified homologous arginine/lysine rich domains in both the chimpanzee and macaque RV1 and RV2 ORF59 proteins that could function to target the proteins to nuclear replication centers.
Our data show a strong correlation between the expression and nuclear localization of the RV2 ORF59 proteins and the replication of viral genomes and production of infectious virus during RRV infection of RPFF cells. This confirms previous studies that RRV infection of rhesus fibroblasts cells is permissive (Desrosiers et al., 1997; DeWire et al., 2003). Our studies demonstrate that Vero cells are similar to RPFF cells in their ability to be infected by and support the replication of RRV. The RRV-infected Vero cell cultures showed high levels of nuclear ORF59-positive cells, replicated viral genomes and infectious virus released into the culture medium. The level of cell-associated viral genomes at 24 and 48 hours was comparable to that seen with RPFFs, but the level did not increase as significantly over time. The amount of cell-free virus in the culture medium of the infected Vero cells mirrored the results seen with infected RPFF cells. Interestingly, the ORF59-positive RRV-infected Vero cells displayed evidence of aggregation and syncytia formation that was not evident with the RRV-infected RPFF cells. Further analysis of this phenomenon is ongoing. Our data do not agree with a previous study, which indicated that Vero cells did not support RRV replication (DeWire et al., 2003). The reason for this discrepancy is not clear, however, we utilized the 17577 isolate of RRV in our study, while the DeWire study utilized the H26-95 isolate [
26].
Previous microarray analysis has shown that RRV ORF59 mRNA expression initiates 12-24 hours post infection and the levels increase significantly between 48 and 72 hours. RRV ORF59 mRNA expression during this time course was significantly inhibited by treatment of cells with phosphonoacetic acid (PAA) suggesting that the
orf59 gene could be classified as an early-late viral gene whose expression is dependent upon viral replication[
34]. KSHV
orf59 gene expression is induced 10-24 hours after TPA induction of latently-infected PEL cells [
18,
39], and mRNA abundance was maximal 48 hours post-induction and with subsequent decline thereafter [
18]. PAA treatment blocked the majority of KSHV ORF59 protein synthesized at 24 hours, indicating that it also belongs to the early-late class of viral proteins [
18]. We analyzed the expression of RRV ORF59 mRNA after RRV infection of RPFF cells using real-time quantitative RT-qPCR and compared it to the expression of several other RRV genes. RRV ORF59 mRNA expression was first detected 8 hours post infection, similar to the time frame for KSHV ORF59 mRNA expression after TPA induction of latently infected cells. We found that RRV ORF59 mRNA expression was concomitant with the expression of the mRNA for the ORF50 viral activator, the ORF9 DNA polymerase, and the ORF8 glycoprotein B. The co-expression of these genes at early time points after RRV infection of RPFF cells is consistent with a permissive RRV infection and the onset of viral replication and production of infectious virions.
Using our rabbit RV2 ORF59 antiserum, we detected RV2 infections in keratinocytes within the differentiated epithelium of the skin of an MneRV2-infected pig-tailed macaque. This is the first evidence for an RV2 rhadinovirus infection of a non-lymphoid cell
in vivo. The anti-RV2 ORF59 antiserum gave low level and intermittent staining of the columnar keratinocytes in the basal layer of the skin epithelium in this animal. Cells in this layer consist of stem cells and transit-amplifying cells, which are continuously dividing. These cells are the source of the suprabasal cells that have migrated into the upper epithelial layers and undergone terminal differentiation [
40]. In our study, we observed consistent and strong nuclear ORF59 staining in the nuclei of keratinocytes present in the more differentiated spinous and granular layers of the epithelium. Whereas cells in the upper epithelium of uninfected animals typically lose their nuclei [
40], we observed RV2 ORF59-positive nuclei throughout all layers of the infected epithelia. This is reminiscent of papillomavirus-infected epithelia where the infected cells leave the basal layer and remain active in the cell cycle due to the action of the E7 oncoprotein [
41]. As the papillomavirus-infected cells differentiate during their migration from the basal layer, the replicative cycle of the virus is activated with high-level expression of transcripts encoding viral replication-associated proteins and late gene products. The viral genomes are replicated and the virions are assembled with the release of the mature virions from nucleated cells present in the uppermost layers of the epithelium. Our results suggest that a similar phenomenon is occurring in the epithelial cells infected with the RV2 rhadinoviruses. As in the case of the RPFF and Vero cells undergoing permissive
in vitro infection with the RV2 rhadinoviruses, the suprabasal epithelial cells in the skin of the MneRV2 infected macaque were strongly reactive with the anti-RV2 ORF59 antiserum in nuclei. In KSHV infected cells, nuclear localization of ORF59 drives nuclear import of the KSHV DNA polymerase during activation of the replicative cycle of the virus [
20]. Our
in vitro results demonstrate a strong correlation of the expression and nuclear localization of RV2 ORF59 with the replication of the viral genome and production of infectious virions. Our
in vivo ORF59 staining results demonstrate that suprabasal keratinocytes in differentiated skin epithelium are infected with the RV2 rhadinovirus and express nuclear ORF59 suggesting that they are actively replicating the viral genome and producing infectious virions. Verification of this is ongoing.
Infectious KSHV virions are detected in the saliva of infected individuals, and saliva is thought to be the major route of transmission for
de novo infections [
42,
43].
In vitro tissue culture models support a role for differentiation of latently infected oral epithelial cells in the activation of KSHV replication and production of infectious virions in saliva [
44]. We have also detected RV2 rhadinovirus genomes in macaque saliva suggesting a similar mode of transmission (unpublished results). Our immunohistochemical results indicate that epithelial differentiation is also an activating event to initiate virus replication and production of infectious RV2 virions in latently infected cells. The activation of virus replication detected in our study would likely lead to the production and release of infectious RV2 virions from skin epithelium in addition to virus released into saliva from the oral epithelium. Current studies are ongoing to determine whether skin epithelia plays a role in RV2 transmission, as occurs with papillomavirus infections.
Materials and methods
Tissue
Retroperitoneal fibromatosis (RF) tumor tissue from M. nemestrina (MneM78114), naturally infected with RFHVMn and MneRV2, was provided by C.-C. Tsai, Washington National Primate Research Center (WaNPRC). Spleen tissue from Mne442N, an M. nemestria naturally infected with RFHVMn and MneRV2, was obtained from R. Shibata, National Institutes of Health, Bethesda, MD. RF tumor tissue from M. mulatta (MmuYN91-224), naturally infected with RFHVMm, was kindly provided by the late H. McClure (Yerkes National Primate Research Center). Spleen tissue from an M. fascicularis (Mfa95044) naturally infected with RFHVMf and MfaRV2 was obtained through the Tissue Distribution Program at the WaNPRC. Tissue samples from a chimpanzee naturally infected with PtrRV1 and PtrRV2 were a gift from S. Shapiro at M.D. Anderson. Formalin-fixed paraffin-embedded tissue blocks were obtained from a female M. nemestrina (MneF94292) (WaNPRC) that was naturally infected with simian retrovirus-2 and MneRV2.
Mammalian Cell Culture
Rhesus primary fetal fibroblasts (RPFF) were kindly provided by M. Axthelm, Oregon National Primate Research Center (ONPRC) and grown in DMEM complete (Invitrogen) at 37°C, 5% CO2. African green monkey kidney epithelial cells (Vero) were obtained from M.E. Thouless, University of Washington and grown in DMEM complete at 37°C, 5% CO2 .
Rhadinovirus isolates
An isolate of the rhesus macaque RV2 rhadinovirus, RRV strain 17577, was kindly provided by M. Axthelm and S. Wong (ONPRC). The MneRV2 isolate J97167 obtained from an
M. nemestrina at the WaNPRC was described previously [
30]. Virus stocks were prepared by infecting cultures of RPFF cells with either MneRV2 or RRV. Viral particles were harvested from culture supernatant by high speed centrifugation. or by concentration on a 50% Opti-Prep (Iodixanol) cushion [
45]. Virus stock was passed through a 0.45 μm filter prior to use.
OFR59 cloning and sequence analysis
The protein sequences of the ORF58, ORF59 and ORF60 genes from KSHV and RRV (NCBI) were aligned using ClustalW. The consensus-degenerate hybrid oligonucleotide primer (CODEHOP) technique [
46‐
48] was used to design CODEHOP PCR primers from conserved amino acid motifs within these genes to enable the amplification and sequence analysis of this region within other Old World primate RV1 and RV2 rhadinoviruses. Primers NFFEa (ORF60) and PQFVb (ORF59) (see Table
1) were designed based on the homology between KSHV and RRV and used to amplify a portion of the
orf59-60 intergenic region of RFHVMn from spleen DNA from Mne442N. Amplification reactions were performed on an iCycler (BioRad) with Platinum TAQ (Invitrogen) using the supplied buffer, 2 mM MgCl
2 and a 55-70°C temperature gradient. From the RFHVMn
orf59 sequence obtained with these primers, a specific primer, YGVRb, was identified (Table
1). A CODEHOP PCR primer, WCFIb (Table
1), designed from conserved sequence motifs identified in the
orf58 homologs of KSHV and RRV, was used with the YGVRb primer to amplify the remainder of the
orf59 allowing the determination of the complete RFHVMn ORF59 sequence. A similar approach was used to amplify the complete ORF59 genes from RFHVMm, RFHVMf, PtrRV1, MneRV2, MfaRV2, PtrRV2 and human EBV (BMRF1) using the primers shown in Table
1.
Sequence alignment and phylogenetic analysis
Protein sequences were aligned using ClustalW and analyzed using the protein maximum-likelihood program from the Phylip package, version 3.62 (University of Washington, Seattle). Phylogenetic tree output was produced using TreeView (R.D.M. Page).
RT-qPCR assay for RRV mRNA expression
RPFF cells were plated at 2 × 10
5 in a 6-well plate and grown overnight at 37°C, 5% CO
2. The cells were incubated with purified RRV (MOI of ~0.01) for 2 hours, washed several times and then further incubated at 37°C for various times. The cells were harvested with trypsin and pelleted by centrifugation. RNA was isolated using an RNeasy kit (Qiagen) following the manufacturers instructions, treated with DNAase to remove contaminating viral genomic DNA and converted to cDNA (250 ng of DNAase-treated RNA, 50 pmoles of random decamer primers (IDT), 0.5 mM dNTPs, M-MLV buffer (Promega) and 1 unit M-MLV reverse transcriptase - 50 minutes at 42°C followed by inactivation at 70°C for 15 minutes). SYBR green-based qPCR assays were developed to quantitate RRV mRNA levels for ORF8 (glycoprotein B), ORF9 (DNA Polymerase), ORF50 (viral transactivator), ORF73 (latency-associated nuclear antigen) and ORF59 (DNA polymerase processivity factor). Amplification primers for each gene were derived from the published RRV17577 genome sequence (NC_003401) (Table
4) [
14]. PCR amplification of cDNA was performed with IQ SYBR Supermix (BioRad) with a 95°C denaturation step for 30 seconds, a 62°C annealing step for 30 seconds and elongation at 72°C for 30 seconds for 50 cycles. The presence of contaminating viral DNA in the DNAase-treated RNA was controlled for by eliminating reverse transcriptase in the cDNA conversion. All of the assays were found to be > 95% efficient based on standard curves using 4-fold dilutions of DNA purified from RRV infected RPFF cells. Viral mRNA expression was normalized to the cellular ribosomal phosphoprotein (RPO) mRNA levels detected by RT-qPCR (see Table
4 for primers) using the delta cycle threshold (ΔCT) method.
Table 4
Oligonucleotide primers for RT-qPCR amplification of RRV (17577) genes
ORF8 (Glycoprotein B) | Forward | IQTTa | CAGGGCGATACAGACGAC |
| Reverse | ATIKb | TCTTGATGGTGGCGTTGA |
ORF9 (DNA Polymerase) | Forward | HGFTa | GGATCACGGGTTGACCAC |
| Reverse | TREGb | GCCTTCCCTGGTCGTGT |
ORF50 (Replication and Transcription Activator) | Forward | ATCNa | CCGCCACCTGTAACGTC |
| Reverse | SFPRb | CCTTCATTTCCGCGTAG |
ORF59 (DNA Polymerase Processivity Factor) | Forward | HKSVa | GTGCACAAGAGCGTCGTG |
| Reverse | CKIPb | ATCCCGGAATCTTACATG |
ORF73 (Latency-associated nuclear antigen) | Forward | RGGTa | CGCGGCGGCACTAGA |
| Reverse | MATLb | CCTGATGGCGACCCTT |
Cellular Ribosomal Phosphoprotein (RPO) | Forward | RP01a | AGCAGGTGTTCGACAATGGCA |
| Reverse | RP01b | ACTCTTCCTTGGCTTCAACC |
Bacterial expression of ORF59 proteins
pGEX-2T and pQE30 vectors were constructed to express ORF59 homologs from different RV1 and RV2 rhadinoviruses fused to either GST or 6X-His, respectively. A sense primer containing a BamHI site followed by the 5' coding sequence downstream of the initiation codon ATG was designed for each ORF 59 homolog except for PtrRV1 and PtrRV2 for which the primers contained BglII sites. Anti-sense primers were derived from the 3' end of the individual ORF59 genes, and contained a stop codon and a HindIII site with the exception of the PtrRV1 primer which had a BglII site. The primers were used to amplify the
orf59 genes from the DNA templates shown in Table
1 with a temperature gradient as described above. The PCR products were digested with the appropriate restriction enzymes and inserted into BamHI digested pGEX-2T(PtrRV1) or BamHI and HindIII digested pGEX-2T to produce GST-ORF59 fusion expression clones. The same PCR fragments were inserted into digested pQE30 vector to produce 6X-His ORF59 fusion expression clones. The resulting plasmids were transfected into Rosetta 2 cells (pGEX-2T) or M15 cells (pQE30). Bacteria were grown to an OD
600 of 0.7-1.0, induced with 0.2 mM IPTG and incubated for 5 hours. GST-fusion proteins were obtained by sonication of the cell pellets, adsorption onto s-hexylglutathione-agarose (Sigma) and elution with 20 mM glutathione. 6XHis fusion proteins were obtained by lysis of the bacterial cell pellets in 6 M guanidine, adsorption onto Ni-NTA-agarose (Qiagen), washing with 1 M urea pH 6.3 and elution in 1 M urea pH 4.5.
Preparation of rabbit polyclonal anti-RV2 ORF59 antisera
The coding sequences for amino acids 300-388 of RRV(17577) and MneRV2(442N, this study) ORF59 were each inserted into the pQE30 expression vector to produce N-terminal 6XHis-tagged ORF59 fusion polypeptides. Recombinant proteins were prepared in M15 bacteria, as described above. The two ORF59 polypeptides were purified on Ni-NTA-agarose and reverse-phase HPLC, combined as a single immunogen, and injected into two rabbits, 424 and 425.
Western analysis
The ORF59 recombinant proteins were analyzed on Invitrogen 4-12% Bis-Tris SDS-PAGE gels in MES buffer and transferred to PVDF membranes. The membranes were blocked with tris-buffered saline with 0.1% Tween-20 and 5% non-fat milk prior to incubation with either a 1:1000 dilution of macaque plasma, a 1:10,000 dilution of 425 anti-RV2 ORF59 rabbit polyclonal serum, a 1:5,000 dilution of mouse monoclonal anti-GST (Sigma), or a 1:10,000 dilution of mouse monoclonal anti-RGS 6X His (Qiagen). The membranes were then incubated with a 1:10,000 dilution of anti-human IgG-Dylight 700 for macaque serum, anti-rabbit IgG-DyLight 800 or anti-mouse IgG-Dylight 800 (Thermo Scientific). The immunoreactive proteins were visualized on an Odyssey Infrared Imager (Li-Cor). The Pageruler Plus prestained protein ladder (Fermentas) was used for molecular weight estimation.
ORF59 immunofluorescence assays
Approximately 1.5-2.0 × 104 cells were plated onto 17 mm spots in 60 mm Petri dishes in 100 μL of complete DMEM and were incubated overnight at 37°C. The cells were infected at low MOI (100 μL of purified RRV or MneRV2 at a dilution of 1:10) for 2 hours at 37°C. Virus was removed and 150 μL fresh media was added to each spot. Cells were incubated for various times prior to fixation in 8% paraformaldehyde. Cells were stained with a 1:1000 dilution of HHV8-ORF59 mouse monoclonal (ABI) or a 1:650 dilution of polyclonal rabbit 425 anti-RV2 ORF59 antiserum. For visualization of ORF59, cells were stained with a 1:500 dilution of goat anti-mouse IgG-Alexa 594, or goat anti-rabbit IgG-Alexa 488 (Invitrogen). To identify nuclei, cells were stained with a 1:500 dilution Topro-3 (Invitrogen) before the last fixation. Images were viewed using a Zeiss LSM 5 Pascal confocal microscope.
RRV replication and production of infectious virions
RPFF or Vero cells were plated at 2 × 10
5 in a 6-well plate and grown overnight at 37°C, 5% CO
2. The cells were incubated with purified RRV (MOI of ~0.01) for 2 hours, washed several times and then further incubated at 37°C. At 1, 2, 3, 5 and 8 days post infection, the culture medium was collected and the cells were harvested by trypsinization and low-speed sedimentation. DNA was extracted from each cell pellet and from the combined culture medium obtained from each well during the total incubation period. The levels of RRV DNA in the cultured cells and medium were determined by real-time qPCR as described [
30]. To demonstrate the production of infectious virus in the cell cultures, culture medium was removed from the RRV-infected RPFF and Vero cell cultures 5 days post-infection and incubated with uninfected RPFF cell cultures for 2 hours. The cell cultures were washed several times and further incubated. Five days post infection, the level of RRV DNA in the culture medium was determined by qPCR.
Immunohistochemistry
Tissue slides were deparaffinized and heated in Target Retrieval, pH 9 (DAKO) for 25 min followed by 20 min at room temperature. This was followed by blocking steps (0.3% H2O2 for 10 min, 0.15 M glycine/PBS for 15 min, 1% BSA and 10% normal horse serum in TBST (Tris-buffered saline with 0.2% Tween-20) for 30 min). The slides were then incubated with the 425 rabbit anti-RV2 ORF59 antiserum (1:300 in TBST/BSA/NHS) for 2 h at room temperature. Bound antibody was visualized with the Envision kit from DAKO. All incubation steps were interspersed with washing-steps using TBST (3 × 5 min).
Acknowledgements
We would like to thank R. Shibata of the Laboratory of Molecular Microbiology, National Institute of Allergy and Infectious Disease, NIH (currently at Gilead Sciences), the late H. McClure, the Yerkes National Primate Research Center, S. Schapiro, M.D. The University of Texas M.D. Anderson Cancer Center, and C.-C. Tsai, M.E. Thouless and S.-L. Hu, WaNPRC, for their generous gifts of tissue. We would also like to thank M. Axthelm and S. Wong, Oregon National Primate Research Center for the RRV strain 17577 and the RPFF cells and M.E. Thouless, University of Washington, for the MneRV2 strain J97167.
This work was supported by Awards RR13154, RR023343 and RR15090 from the National Center for Research Resources, NIH. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
Design and conception of the study (AGB, TMR); development of the methods and amplification of the ORF59 homologs (AGB, AMB); Development of the RT-qPCR assays and quantitative analysis (AGB, CAG); Virus infection studies (AGB, CAG, KLB); Sequence analysis, alignment and phylogeny (AGB, TMR); Immunofluorescence studies (CAG, KLB, LKD); Immunohistochemical analysis (HBO); Antiserum development (AGB, AMB, TMR); Protein purification and Western analysis (AGB; AMB; CAG); Manuscript preparation (AGB, CAG, LKD, HBO, KLB TMR). All authors read and approved the final manuscript.