Development and Characterization of a cDNA-Launch Recombinant Simian Hemorrhagic Fever Virus Expressing Enhanced Green Fluorescent Protein: ORF 2b’ Is Not Required for In Vitro Virus Replication

Simian hemorrhagic fever virus (SHFV) causes acute, lethal disease in macaques. We developed a single-plasmid cDNA-launch infectious clone of SHFV (rSHFV) and modified the clone to rescue an enhanced green fluorescent protein-expressing rSHFV-eGFP that can be used for rapid and quantitative detection of infection. SHFV has a narrow cell tropism in vitro, with only the grivet MA-104 cell line and a few other grivet cell lines being susceptible to virion entry and permissive to infection. Using rSHFV-eGFP, we demonstrate that one cricetid rodent cell line and three ape cell lines also fully support SHFV replication, whereas 55 human cell lines, 11 bat cell lines, and three rodent cells do not. Interestingly, some human and other mammalian cell lines apparently resistant to SHFV infection are permissive after transfection with the rSHFV-eGFP cDNA-launch plasmid. To further demonstrate the investigative potential of the infectious clone system, we introduced stop codons into eight viral open reading frames (ORFs). This approach suggested that at least one ORF, ORF 2b’, is dispensable for SHFV in vitro replication. Our proof-of-principle experiments indicated that rSHFV-eGFP is a useful tool for illuminating the understudied molecular biology of SHFV.


Introduction
Simian hemorrhagic fever (SHF) was first identified in 1964 as an acute, almost uniformly lethal disease of Asian macaques (Macaca spp.) in captive colonies in Sukhumi, Georgian Soviet Socialist Republic, USSR, and Bethesda, Maryland, USA [1][2][3]. Three is hypothesized to mediate virion entry, which, at least in the case of SHFV, occurs after interaction with a proteinaceous receptor, which may be CD163, via actin-independent, but dynamin-dependent and clathrin-mediated, endocytosis [28]. The functions of GP2 , GP3 , GP4 , and E' are unknown.
Because SHEV, SHFV, PBJV, and KRCV-1 are neglected animal pathogens, few tools and detection reagents are available for their characterization. In this study, we developed a recombinant cDNA-launch infectious clone of SHFV that expresses a reporter protein (enhanced green fluorescent protein [eGFP]). In proof-of-principle experiments, we used this rSHFV-eGFP to further characterize SHFV cellular tropism and to ascertain whether any SHFV minor structural proteins are dispensable for in vitro replication.  [36] (all generated with funds from the German Research Council (DR 772/10-2) and provided by Marcel A. Müller and Christian Drosten, Charité-Universitätsmedizin Berlin, Berlin, Germany) were grown in Dulbecco's modified Eagle's medium (DMEM; Thermo Fisher Scientific) supplemented with 10% heat-inactivated FBS.
The "NCI-60 panel" of highly characterized human cancer cell lines [39] was obtained from the National Cancer Institute's Developmental Therapeutics Program (NCI DTP; Frederick, MD, USA). For this study, 53 adherent cell lines of this panel were used and grown in Roswell Park Memorial Institute 1640 medium (RPMI-1640; Thermo Fisher Scientific) supplemented with 10% heat-inactivated FBS as previously described [40].
All cells were grown at 37 • C in a humidified 5% CO 2 atmosphere.

SHFV Growth Kinetics Comparison
MA-104 cells, seeded in collagen-coated 24-well plates (1 × 10 5 cells per well), were exposed to wild-type SHFV (ATCC), rSHFV, or rSHFV-eGFP at a multiplicity of infection (MOI) of 0.1 or 1. After 1 h of incubation at 37 • C, viral inocula were removed, and cells were washed twice with EMEM without FBS and then supplemented with EMEM with 2% FBS. At various times post-exposure (p.e.), TCSs were harvested. Viral titers were determined by plaque assay as previously described [28].

rSHFV-eGFP Genetic Stability Determination
To determine the stability of the inserted eGFP ORF, rSHFV-eGFP was passaged five times in MA-104 cells. Briefly, MA-104 cells were exposed to rSHFV-eGFP (defined as Passage 0) at an MOI of 1. At 48 h p.e., TCSs were collected (Passage 1), and virus titers were determined by plaque assay as previously described [28]. Fresh MA-104 cells were then exposed to Passage 1 TCSs at an MOI of 1 to generate Passage 2. This process was serially repeated until Passage 5.

SHFV Titration
SHFV titers were quantified by plaque assay as previously described [28]. Briefly, confluent monolayers of MA-104 cells in 6-well plates were exposed to serial dilutions of SHFV. After 1 h of incubation at 37 • C while gently rocking every 15 min, 2 mL of a 1:1 overlay consisting of 2X Gibco modified Eagle's medium (MEM; Thermo Fisher Scientific) and 1% Seakem ME Agarose (Lonza, Rockland, ME, USA) were added to each well. At 2 days p.e., agarose overlays were removed and the cells were stained with 0.2% crystal violet staining solution. Plaques were counted manually.

CD163 Antibody Inhibition Assay
MA-104 cells were incubated with different concentrations (0, 2.5, 5, 10, 20, or 40 µg/mL) of goat anti-human CD163 antibody (R&D Systems, Minneapolis, MN, USA) or control goat immunoglobulin G (IgG) at 37 • C for 1 h. Treated cells were exposed to rSHFV-eGFP at an MOI of 5 at 37 • C for 1 h in the presence of antibodies. The virusantibody inocula were then removed, the cells were washed, and EMEM plus 2% FBS was added. The cells were fixed with 10% neutral-buffered formalin (NBF; Thermo Fisher Scientific). Nuclei were stained with Hoechst 33342 (Thermo Fisher Scientific). The percentage of eGFP-positive cells was quantified with the Operetta High-Content Imaging System (PerkinElmer, Shelton, CT, USA) and analyzed with Harmony 3.1 analysis software (PerkinElmer) as previously described [28].

Rescue and Growth Characterization of Recombinant SHFVs
To construct a DNA-launch infectious clone of SHFV, we cloned the cDNA of the entire genome of SHFV variant NIH LVR42-0/M6941 (GenBank #AF180391; RefSeq #NC_003092), flanked by a CMV immediate early enhancer-containing promoter at the 5 end and the hepatitis delta virus 1 (HDV-1) ribozyme sequence at the 3 end to ensure generation of authentic 3 genome sequences, into plasmid pACYC177 ("pCMV-SHFV-NC_003092.2"). However, transfection of this plasmid into MA-104 cells never resulted in successful SHFV rescue. Suspecting that the GenBank sequence may be erroneous, we sequenced the genome of a commercially available SHFV variant NIH LVR42-0/M6941 (GenBank #KM373784.1) [16] and discovered several discrepancies (Table 1). We introduced most of these discrepancies into pCMV-SHFV-NC_003092.2, yielding pCMV-SHFV, and created a derived plasmid thereof, pCMV-SHFV-eGFP, encoding eGFP between ORF 1b and ORF 2a' (Figure 1). To rescue both wild-type (rSHFV) and eGFP-expressing SHFV (rSHFV-eGFP), we transfected MA-104 cells directly with pCMV-SHFV or pCMV-SHFV-eGFP. At 16 h p.t., we observed cell rounding and detachment in plates transfected with either plasmid reminiscent of the cytopathic effect (CPE) characteristic of SHFV infection [28] (Figure 2A). Under fluorescence microscopy, green fluorescent signal was observed in pCMV-SHFV-eGFP-transfected but not pCMV-SHFV-transfected or SHFV-exposed cells, indicating the expression of eGFP from the transfected plasmid (Figure 2A). Cell culture supernatants were harvested from these transfected cells at 48 h p.t., genomic sequences of rSHFV and rSHFV-eGFP in the harvested supernatants were validated by high-throughput sequencing, and standard plaque assays were performed after transferring the supernatants onto fresh MA-104 cells to demonstrate virus growth ( Figure 2B). These results indicate that viable viruses (rSHFV and rSHFV-eGFP) were recovered from the plasmids. As expected, the growth kinetics of SHFV and rSHFV were almost identical at both tested MOIs (0.1 and 1), reaching peak titers of 1 × 10 7 PFU/mL. The titer of rSHFV-eGFP virus was approximately 10-fold lower than that of SHFV or rSHFV, suggesting that the inserted eGFP-encoding cassette attenuated the virus ( Figure 2B).

Identification of New SHFV-Permissive Cell Lines
We performed several proof-of-principle experiments to demonstrate the usefulness rSHFV-eGFP (Passage 1) as a tool for the identification of novel SHFV-permissive cell lines. We exposed 53 adherent cell lines of the human NCI-60 cancer cell line panel and 11 bat cell lines to rSHFV-eGFP, using MA-104 cells as a control. None of these cell lines replicated rSHFV-eGFP (Figure 3).  11 bat cell lines were exposed to rSHFV-eGFP at a multiplicity of infection of 3. Shown are high-content images of mock-exposed controls and virus-exposed cells counterstained with Hoechst33342 at 72 h or 24 h post-exposure for positive control MA-104 cells.
We then repeated this experiment with various nonhuman primate cell lines (including MA-104, MARC-145, and BS-C-1 cells as positive controls), additional human cell lines, rodent cell lines, and primary ape fibroblasts (Figures 4 and 5). As expected, all control cell lines were permissive to rSHFV-eGFP. However, most other tested cell lines did not replicate rSHFV-eGFP. Surprisingly, common chimpanzee RP00226 and S008397 primary fibroblasts, western gorilla RPGor53 primary fibroblasts, and hispid cotton rat lung CRL cells were found to be permissive to the virus (Figures 4C and 5A). To further confirm the permissiveness of these cell lines to rSHFV-eGFP, we exposed them to wild-type SHFV at an MOI of 3 and measured virus growth kinetics over time ( Figure 5B). The results indicated that the CRL cells are the most permissive of the four cell lines, replicating rSHFV-eGFP to 10 7 -10 8 PFU/mL at 48 h p.e. rodent cells were exposed to rSHFV-eGFP at a multiplicity of infection of 3. Tissue culture supernatants were harvested at 72 h post-exposure, cells were counterstained with Hoechst 33342, and high-content images of exposed cells or mock-infected cells were taken. Supernatants of virusexposed cells were also subjected to plaque assay; obtained titers are printed beneath the images. Uninfected cells were also transfected with pCMV-SHFV-eGFP or a control plasmid expressing eGFP only. At 48 h post-transfection, cells were counterstained with Hoechst 33342, high-content images were taken, supernatants were harvested, and viral titers were measured by plaque assay; obtained titers are indicated beneath the images. Recombinant wild-type SHFV expressing enhanced green fluorescent protein-expressing (rSHFV-eGFP) infects hispid cotton rat lung (CRL) cells, western gorilla RPGor53 primary fibroblasts, and chimpanzee S008397 and RP00226 primary fibroblasts. (A) CRL cells, western gorilla primary fibroblasts, and common chimpanzee primary fibroblasts were exposed to rSHFV-eGFP at a multiplicity of infection (MOI) of 3. eGFP expression and cytopathic effect were observed at 24 h post-exposure. (B) Cells were exposed to wild-type SHFV at an MOI of 3. Tissue culture supernatants were harvested at the indicated time points post-exposure, and virus titers were determined by plaque assay. Graphs of viral titers represent the means ± the standard deviations of triplicate samples from one of two independent experiments.

Restriction of SHFV Likely Occurs at Multiple Steps in the Cellular Host
To examine whether resistance to SHFV infection is solely determined by the inability of SHFV to enter cells, we transfected pCMV-SHFV-eGFP directly into apparently resistant grivet (Vero E6), human (293T and HeLa), and rodent (CHO-K1, BHK-21, and NIH3T3) cells as well as permissive grivet (MA-104, MARC-145, and BS-C-1) and hispid cotton rat lung CRL cells ( Figure 4). As expected, we observed extensive eGFP expression in permissive cells. Only a few eGFP-expressing cells were observed in apparently resistant cell lines. Interestingly, plaque-assay data indicate that substantial viral replication occurred in 293T cells (~10 5 PFU/mL) and limited replication occurred in other apparently resistant cell lines of distinct species. These data indicate that cells permit or restrict against SHFV infection at virion entry (susceptibility), immediately after virion entry (permissibility), or both. For instance, grivet Vero E6 cells appear to restrict SHFV infection at the virion entry step because Vero E6 cells transfected with pCMV-SHFV-eGFP plasmid resulted in limited rSHFV-eGFP replication, whereas other grivet (MA-104, MARC-145, and BS-C-1) cells take up and replicate rSHFV-eGFP directly.

The eGFP Expression Cassette in rSHFV-eGFP Is Unstable
We investigated the stability of the eGFP reporter cassette in rSHFV-eGFP by serial passaging rSHFV-eGFP in MA-104 cells. Cells were exposed to rSHFV-eGFP at an MOI of 1. After 48 h p.e., TCSs were collected (Passage 1) and virus titers were measured by plaque assay. Fresh MA-104 cells were exposed to rSHFV-eGFP Passage 1 at an MOI of 1, and the process was serially repeated through Passage 5. The expression of eGFP in each passage, monitored by epifluorescent microscopy, dramatically decreased after Passage 2 ( Figure 6), whereas plaque-assay data were largely stable. These data indicate that the virus either lost parts of or the entire eGFP cassette or rapidly accumulated deleterious mutations within it during replication.

The Minor Structural Protein E' Is Dispensable for Infectious SHFV Particle Production
Vatter et al. previously reported that each of the eight SHFV minor structural proteins is functionally important [27]. To confirm these results, we generated pCMV-SHFV-eGFP mutants that contained stop codons likely to terminate their expression without altering the amino-acid residues encoded in overlapping ORFs ( Figure 7A). To evaluate whether infectious rSHFVs could be rescued from these plasmids, we transfected MA-104 cells. At 72 h p.t., TCSs were harvested and subjected to plaque assay. No CPE or eGFP expression was observed in cells transfected with plasmids preventing ORF 2b, 2a, 3, 4, 2a', 3 , or 4 expression and plaque assays were negative ( Figure 7B), suggesting that the minor structural proteins encoded by these ORFs (E, GP2, GP3, GP4, GP2 , GP3 , and GP4 , respectively) are likely essential for SHFV propagation. However, we observed SHFV-typical CPE, eGFP expression, and titers in cells transfected with the plasmid designed to prevent ORF 2b' expression, suggesting that the encoded protein (E') is not required for SHFV propagation. Since this result is contradictory to results reported by Vatter et al. [27] and because it is possible that the single stop codon we introduced was somehow circumvented by the viral polymerase, we created a second pCMV-SHFV-eGFP that included eight stop codons dispersed among the entire ORF 2b' ( Figure 7A). Consistently, SHFV-typical CPE and eGFP expression were observed from cells transfected with this plasmid, and viral titers were comparable to those measured for cells transfected with prototype pCMV-SHFV-eGFP ( Figure 7B). Supernatants harvested from cells transfected with pCMV-SHFV-eGFP and pCMV-SHFV-eGFP mutants containing stop codons to prevent expression of GP2 and E' were used to expose fresh MA-104 cells to evaluate infectious particle production. As expected, supernatants from cells transfected with the ORF 2a' mutant plasmid did not contain infectious particles, whereas both ORF 2b' mutants resulted in infectious progeny virions. These findings confirm that E' does not play a vital role in in vitro SHFV replication in vitro.  3.6. Replacement of SHFV ORFs Encoding Minor Proteins E, GP2, GP3, and GP4 of SHFV with EAV Orthologs Does Not Expand SHFV Cell Tropism PRSSV-1 and PRSSV-2 have highly restricted cell tropisms, similar to SHFV [44][45][46][47]. On the other hand, EAV is promiscuous, infecting numerous cell types [48]. The EAV minor structural proteins E, GP2b, GP3, and GP4, considered to be orthologs of PRRSV-1/2 and SHFV E, GP2, GP3, and GP4, respectively, appear to be major determinants of EAV cell tropism [49]. A chimeric recombinant PRRSV-2 encoding the four EAV minor envelope proteins instead of the PRRSV-1 orthologs lost the ability to infect PRRSV-1-permissive porcine alveolar macrophages. Instead, this chimeric virus gained the ability to infect PRRSV-resistant but EAV-permissive BHK-21 and Vero cells [49]. In contrast to EAV and PRRSV-1/2, SHFV and other simarterivirins encode a second set of presumed minor structural proteins: E', GP2 , GP3 , and GP4 . As a further proof-of-principle experiment to demonstrate the usefulness of an SHFV infectious cDNA clone, we replaced SHFV ORFs 2b, 2a, 3, and 4 in pCMV-SHFV-eGFP with the corresponding EAV ORFs, expecting that this replacement would change the tropism of the resulting rSHFV-eGFP-EAV-ORF2ab34 from that of SHFV to that of EAV. We successfully rescued the virus from MA-104 cells transfected with the chimeric plasmid ( Figure 8). Interestingly, rescued rSHFV-eGFP-EAV-ORF2ab34 continued to infect MA-104 cells but not Vero or BHK-21 cells (Figure 8).

CD163 Is a Crucial SHFV Cell Entry Factor
CD163 plays an important, although still relatively undefined, role in PRRSV-1/2 particle entry into host cells [28,50]. We previously demonstrated that CD163 plays a similar role in SHFV particle entry: As measured by plaque assays, SHFV-permissive cells pre-incubated with an anti-human CD163 (anti-hCD163) antibody and then exposed to wild-type SHFV produced significantly fewer progeny virions than cells treated with a mock antibody [28]. This observation suggested that CD163 may be an SHFV receptor. To confirm these results, MA-104 cells were incubated with increasing concentrations of a specific anti-hCD163 antibody and then exposed to rSHFV-eGFP. At 48h p.e., the percentage of eGFP-positive cells was measured by high-content imaging. Consistent with the previously published plaque-assay data, the percentage of rSHFV-eGFP-infected cells was dramatically reduced by anti-hCD163 antibody treatment (Figure 9).

Discussion
SHFV is a neglected pathogen. Consequently, knowledge of the molecular biology and virology of SHFV is extremely limited. The functions of most of SHFV's identified nonstructural and structural proteins are yet to be elucidated, and a relatively recent study indicated that the proteome of SHFV is much more complex than currently appreciated [24]. One reason for the lack of understanding of this important nidovirus [15] is the absence of specific reagents for SHFV manipulation and detection. There are no commercial (or otherwise easily available) antibodies that are suitable for standard immunofluorescence assays, and probes developed for SHFV detection using RNAscope in situ immunohistochemistry [22] are relatively expensive and labor-intensive.
One way of circumventing the current shortcomings in SHFV detection would be the use of a reporter protein-encoding SHFV. The development of such a recombinant virus requires the establishment of cDNA clones from which viruses can be rescued. Previously, Vatter et al. developed such a system for SHFV variant NIH LVR42-0/M6941; the SHFV genome was cloned into a large capacity plasmid vector (pACYC184 (V)) that, after linearization, served as a template for in vitro transcription of genomic RNA, which was then transfected into MA-104 target cells [27]. We aimed at simplifying this system by circumvention of the linearization and in vitro transcription step, in part because performing this step requires specialized reagents or kits and in part because working with RNA in general presents distinct challenges. Following a previous strategy used for PRRSV-1 [42,43], we established a DNA plasmid containing the SHFV genome that can be transfected directly into target cells without prior manipulation for successful rSHFV rescue. Further following the PRSSV-1 strategy, we modified the SHFV genome in the DNA plasmid to encode eGFP in a separate expression cassette immediately upstream of the SHFV structural ORFs and demonstrated successful rescue of rSHFV-eGFP.
Next, we used this virus to demonstrate that rSHFV-eGFP is a valuable, easy-touse tool to identify novel SHFV-permissive cell lines. Our initial results supported the notion that human cells are generally not permissive to SHFV multiplication. Given the fastidious cell tropism of SHFV, our discovery of primary common chimpanzee and gorilla fibroblast and a hispid cotton rat lung (CRL) cell lines as SHFV-permissive cells was surprising. These findings indicate that SHFV may not be as host-restricted as previously thought and infer that future experiments should include a broader panel of rodent and nonhuman primate cell lines to identify cell lines that are not only permissive to SHFV but also easy to maintain and manipulate and replicate SHFV to high(er) titers. Second, we demonstrated that some cell lines, including human, are not susceptible to rSHFV-eGFP infection after direct exposure but they support rSHFV-eGFP replication and particle egress if the infectious clone plasmid is transfected directly into them. Although thus far limited in scope, these data indicate that, for instance, human cells restrict SHFV replication in at least two points: the cell entry step (because SHFV replicates in some cells after entry restriction is circumvented) and the step post-entry (because SHFV does not replicate in some cells even after entry restriction is circumvented). Third, we used the same strategy previously published by Vatter et al. to evaluate the functional importance of the eight SHFV minor structural proteins [27]: We attempted to knock out each protein by introduction of stop codons into the respective ORFs. We confirmed the data from Vatter et al., i.e., that minor structural proteins E, 2, 3, 4, 2 , 3 , and 4 are likely required for the SHFV propagation. Unexpectedly, though, we successfully rescued rSHFV-eGFP containing one and eight stop codons in ORF 2b', indicating that expression of E' is dispensable for SHFV propagation. Fourth, in an already-published study, our cDNA clone was used to characterize −2/−1 programmed ribosomal frameshifting during SHFV ORF 1a translation [51].
Together, these data build confidence in our system being a useful tool for the study of SHFV molecular biology. Of course, the system will require further optimization. The current location of the eGFP-expressing cassette is not ideal as even a few passages of rSHFV-eGFP resulted in loss of fluorescence, likely due to a compromise of the eGFPencoding ORF. A study by Di et al. indicated that such compromise can be circumvented by shuffling the ORF to different locations in the SHFV genome, replacing the type of encoded reporter, or both [52]. In addition to the creation of a more stable reporter-encoding SHFV for advanced in vitro studies, it would be useful to determine the pathogenicity of such a virus in vivo in a macaque host. A virulent reporter virus would open the doors for sophisticated pathogenesis studies and, in the long run, simplify the evaluation of potential medical countermeasures against SHFV.

Conclusions
In conclusion, we developed a DNA-launch recombinant SHFV infectious clone system and, using proof-of-principle experiments, demonstrated the potential application of this system to study various aspects of the SHFV lifecycle.