Introduction

Arteriviruses (Family: Arteriviridae) are enveloped viruses with a positive-sense non-segmented single-stranded (+ss)RNA genome that are distantly related to Coronaviruses (both belong to the Nidovirales Order)1. The genetic diversity of known arteriviruses is vast, as is the diversity and geographic distribution of their many mammalian hosts2,3,4. Arteriviruses are noteworthy for their ability to evade host immune responses, enabling their persistence in individuals and populations over long periods of time5,6. Transmission of these viruses can occur via a variety of routes (respiratory droplet, sexual, vertical, blood-borne) and arterivirus infections can cause a wide variety of diseases ranging from long-term persistent viral shedding to acute pneumonia, abortion, encephalitis, hemorrhagic fever, and death7.

Host-vs-virus phylogenetic comparisons imply that arteriviruses have transmitted between hosts of different species on several occasions in the past2. Indeed, two arteriviruses—simian hemorrhagic fever virus (SHFV) and porcine reproductive and respiratory syndrome virus (PRRSV)—highlight the potential of these viruses to infect new host species and cause disease. SHFV was discovered in 1964 following an outbreak of highly-lethal hemorrhagic fever in Asian-origin macaque monkeys (Macaca mulatta, Macaca fascicularis, and Macaca arctoides) at a primate quarantine facility in Bethesda, MD8,9,10. Since this time, SHFV-like viruses have sporadically caused severe disease outbreaks in captive macaque colonies worldwide11. Although African monkeys were long suspected to be the source of these outbreaks, it was not until the 2010s that SHFV-like viruses were found in wild monkeys throughout Africa4,12.

In the 1980s, two distinct lineages of PRRSV emerged simultaneously in domesticated swine herds in Europe (PRRSV-1) and North America (PRRSV-2). Although most adult pigs exhibited mild-to-moderate signs of respiratory infection, these viruses transmit efficiently across the placenta with devastating effects on fetuses, resulting in herd-wide “abortion storms.” In recent years, several “highly pathogenic” PRRSV variants have emerged that kill ~20% of adult pigs that they infect13,14. Annually, PRRSV infections cause billions of U.S. dollars in global economic losses15. Of note, the origins of both SHFV and PRRSV remain unknown16.

The surface topology of the arterivirion is remarkably complex for an RNA virus. Arteriviruses encode approximately seven membrane glycoproteins that coalesce to form at least two discrete (“major” and “minor”) complexes on the virion surface17. Additionally, SHFV (and all simian-infecting arteriviruses discovered to date) encode a paralogous set of minor glycoprotein genes (called “prime”) and it appears that all eleven of these membrane proteins are required for virion infectivity18,19. Arteriviruses enter cells via clathrin-mediated endocytosis20, but none of the arterivirus proteins contain a structure that resembles a typical class I, II, III, or IV viral fusion protein, suggesting a highly novel mechanism of entry.

Multiple host factors can influence arterivirus entry17,21,22, but at least for PRRSV, infection of macrophages in vivo requires the presence of the hemoglobin-haptoglobin scavenger receptor CD163, as demonstrated by a complete lack of PRRSV-1 and PRRSV-2 replication in CD163−/− pigs23,24. Ectopic expression of porcine CD163 can also render non-permissive cell lines permissive to PRRSV infection in vitro25,26,27, and we have recently shown that ectopic expression of human CD163 on certain human cell lines can sensitize these cells to infection with SHFV28. Engagement of CD163 by the minor glycoprotein complex of PRRSV is thought to occur in the endosome and is presumed to mediate fusion of viral and host membranes25,28,29. The virus-host interactions that initiate endocytosis of the virion prior to CD163 binding are less clear. For PRRSV, interactions between Siglec-1 (a.k.a., sialoadhesin, CD169) and sialic acids on GP5 can initiate internalization in vitro30,31,32,33. Although Siglec-1 expression is largely restricted to macrophages, SIGLEC1−/− pigs and porcine alveolar macrophages are fully susceptible to PRRSV infection34. Furthermore, the MARC-145 cell line—a subpopulation of the grivet (Chlorocebus aethiops)-derived MA-104 cell line that is commonly used to propagate PRRSV—do not express Siglec-117,35,36. Given the large number of arterivirus envelope glycoproteins and the conspicuous gaps in our current knowledge of the arterivirus entry process, we hypothesized that interactions with additional pro-viral host factors remained to be discovered.

Results

FcRn is a pro-arteriviral host factor

The MA-104 cell line supports robust infection with SHFV and PRRSV-2, as demonstrated by production of infectious virus and the development of cytopathic effect following virus inoculation (Fig. 1A, B). We introduced a previously-described genome-wide pooled CRISPR-knockout library specific for the grivet genome into MA-104 cells, infected independent cultures with SHFV or PRRSV, and then performed deep sequencing to identify sgRNAs in surviving cells. Relative to the diversity of sgRNAs in the pre-infection library (Fig. 1C), a small number of single guide (sg)RNAs (and corresponding gene targets) were enriched post-infection, including CD163 (Fig. 1D, Supplementary Data 1). Two genes were identified as top-three hits in each screen: the Fc Gamma Receptor and Transporter Gene (FCGRT)—which encodes the neonatal Fc receptor (FcRn)—and beta-2-microglobulin (B2M)—which encodes the β2M protein that is required for proper expression of FcRn (Fig. 1D, E). Previously-reported pro-arterivirus host factors—including Siglec-1, CD151, Myosin Heavy Chain 9 (MYH9), Vimentin, DC-SIGN, or genes in the heparan sulfate synthesis pathway37—were not identified as pro-viral in our screen.

Fig. 1: FcRn is a pro-arteriviral host factor.
figure 1

A Growth of SHFV and PRRSV-2 in wild-type MA-104 cells, inoculated at MOI = 0.01 (n = 2 biological replicates). Data shows representative results from one of two independent experiments. B Cytopathic effect (CPE) of arterivirus infection of MA-104 cells at 4 days post-inoculation, taken at ×100 magnification. Data shows representative results from one of two independent experiments. C Frequency of sgRNAs, as determined via deep sequencing of genomic DNA from MA-104 cells, prior to and after infection with arteriviruses. D Robust rank aggregation scores (RRA) scores for gene knockouts enriched in surviving MA-104 cells compared to pre-infection; top 10 hits for each screen are shown as colored datapoints. E RRA scores of SHFV vs. PRRSV-2, with CD163, FCGRT (FcRn), and B2M circled and labeled. F Infection of cells transduced with single-guide RNA targeting grivet (for MA-104) or human (for ACHN) FCGRT (ΔFcRn), or empty vector control. Viruses tested on these cells are arranged in terms of relatedness to arteriviruses from left to right, with taxonomic relationships shown along the top. Data are presented as mean values ± SEM. Infections were performed with n = 3 biological replicates at an MOI of 0.01. Supernatant was titrated via plaque assay or focus-forming assay at the indicated time points; asterisks represent p-values (**p ≤ 0.01; ***p ≤ 0.001) derived from a two-tailed unpaired t-test with Welch correction and multiple comparisons testing. G Brightfield photographs of arterivirus infections from F taken at ×100 magnification at 3 days post-inoculation. Data shows representative results from one of two independent experiments. Source data are provided as a Source Data file.

To validate FcRn as a pro-viral host factor, we generated an FcRn-knockout MA-104 cell clone (MA-104ΔFcRn) using an FCGRT-targeting sgRNA. FCGRT editing was confirmed by deep sequencing and high-sensitivity mass spectrometry. Compared to cells transduced with an empty lentiviral vector, MA-104ΔFcRn cells did not produce virus (Fig. 1F) or display cytopathic effect (CPE) (Fig. 1G) when inoculated with SHFV, PRRSV, or equine arteritis virus (EAV)—a divergent arterivirus that does not use CD163 in its entry process. As expected, FcRn-knockout in MA-104 cells also conferred resistance to Echovirus-30, which is known to use FcRn as a cellular receptor38,39. However, infection with SARS-CoV-2 (another virus in the Nidovirales Order) was unaffected by FcRn knockout, indicating that among nidoviruses, FcRn-dependence is specific to viruses in the Arteriviridae family. FcRn-knockout had no effect on the infectivity of unrelated RNA viruses that enter cells via clathrin-mediated endocytosis (yellow fever virus, (YFV)-Asibi) or a clathrin-independent mechanism (YFV-17D)40. The human cell line ACHN is susceptible to SHFV infection when made to express human CD163 (hCD163)28. Thus, we created and validated an ACHN+hCD163ΔFcRn line and showed that these cells were also completely resistant to SHFV infection as determined by lack of CPE and production of infectious virus (Fig. 1G, F), extending this finding to an additional cell line in a non-natural host species.

Transcomplementation of FcRn restores arterivirus infectivity in FcRn-knockout cells

We next reintroduced grivet (g)FcRn into our MA-104ΔFcRn cell line, resulting in substantially higher levels of FcRn expression compared to MA-104 wildtype cells (Figs. 2A, S1). FcRn overexpression rendered these cells hyper-susceptible to SHFV infection, as demonstrated by enhanced cytopathic effect (Fig. 2B) and a significantly higher proportion of cells infected (~80%) compared with wild-type MA-104s (~20%) 24 h after inoculation with SHFV, as determined by highly-sensitive RNA-FISH-flow-cytometry (Fig. 2C, D)41. This sensitizing effect was observed at early timepoints, suggesting that FcRn expression is a rate-determining step in arterivirus infection of MA-104 cells (Fig. 2E).

Fig. 2: Transcomplementation of FcRn restores arterivirus infectivity in FcRn-knockout cells.
figure 2

A Western blot of wild-type (WT) MA-104 cells or MA-104ΔFcRn cells transcomplemented with grivet FcRn (gFcRn) or empty vector (EV). B Cytopathic effect in MA-104 cells from A, 24 hrs post-inoculation with SHFV at an MOI of 0.3. C Fluorescence in situ hybridization flow cytometry (FISH-Flow) targeting the SHFV ORF1a and hybridized to an ATTO-663 probe in the same cell types shown in A, 24 h after inoculation with SHFV at an MOI of 0.3. All flow-cytometric events were first gated on FSC x SSC properties, followed by singlet discrimination. All gates were drawn based on mock-infected cells. D Graphical representation of cells in C with data presented as mean values ± SEM, n = 3 biological replicates, one-way ANOVA,****p ≤ 0.0001. E Percent of cells positive for SHFV viral RNA (vRNA) by RNA FISH-Flow at early time points following inoculation with SHFV at an MOI of 10 (n = 3 biological replicates). Source data are provided as a Source Data file.

FcRn is involved in the entry step of the arterivirus life cycle

Lack of viral RNA accumulation in FcRn-knockout cells inoculated with SHFV suggested a role for FcRn early in the virus life cycle, prior to genome replication. Given the known localization of FcRn to the plasma membrane and early endosome42, and the known role of FcRn as a receptor for multiple unrelated viruses38,39,43,44, we hypothesized that FcRn functions as an arterivirus attachment factor and/or internalization receptor. To test this, we first transfected cells with a cDNA-launch SHFV rescue plasmid18 to bypass receptor requirements for cellular entry. Indeed, the introduction of viral genomes into MA-104ΔFcRn cells resulted in the production of infectious virus into the cell supernatant (Fig. 3A, B). However, viral titers in MA-104ΔFcRn cells were diminished relative to wild-type transfected cells given that released virions were unable to infect new cells. Next, we performed internalization assays in wild-type, ΔFcRn, and ΔFcRn+gFcRn MA-104 cells demonstrating a relationship between FcRn expression and virion internalization (Fig. 3C). To demonstrate a direct interaction between FcRn and arterivirus particles, we screened a panel of anti-FcRn monoclonal antibodies, identifying orilanolimab as a promising arterivirus-blocking candidate. Further testing demonstrated that pre-treatment with orilanolimab blocks SHFV infection of MA-104 cells in a dose-dependent manner (Fig. 3D, H). We then extended this finding to multiple cell lines. First, we showed that orilanolimab blocks SHFV infection of the human ACHN+hCD163 cell line (Fig. 3E, H). We then demonstrated orilanolimab’s ability to block SHFV infection of a more physiologically-relevant cell type45—induced-pluripotent stem cell (iPSC)-derived macrophages from macaques (mac iPSC-macs) (Fig. 3E, H, I). The use of mac iPSC-macs allowed us to expand orilanolimab blocking studies to include novel simian arteriviruses that do not infect immortalized cell lines (including MA-104) but can be cultivated on mac iPSC-macs. These viruses included Kibale Red Colobus Virus-1 (KRCV-1)—a virus discovered in a wild Ugandan red colobus monkey (Piliocolobus tephrosceles) in 201146,47—and Pebjah virus (PBJV), a virus that caused a highly-lethal outbreak at a primate research facility in 198911,48 that we recently isolated for the first time using mac iPSC-macs49. Again, pre-treatment of mac iPSC-macs with orilanolimab significantly reduced SHFV, KRCV-1, and PBJV replication, prevented the induction of cytopathic effect, and enhanced overall survivability of mac iPSC-macs in the presence of arteriviruses (Fig. 3F–I).

Fig. 3: FcRn is involved in the entry step of the arterivirus life cycle.
figure 3

A Transfection of a cDNA-launch SHFV rescue plasmid into wild-type (WT) and ΔFcRn MA-104 cells as a means of bypassing viral entry. Production of infectious virus in cell supernatant was quantified by plaque assay at various time points post-transfection (n = 3 biological replicates). All plots include error bars; no error bars are shown when the SEM was smaller than the size of the symbols. B Comparison of SHFV infectious virus production from cells either transfected with SHFV cDNA-launch clone or infected with virus (MOI = 0.01), 48 hrs after transfection/infection; n = 3 biological replicates per group with error bars showing SEM; statistical significance was determined using an two-tailed unpaired t-test with ** representing P < 0.01. C Wildtype, FcRn knockout (ΔFcRn), or ΔFcRn complemented with grivet FcRn (gFcRn) MA-104 cells were evaluated for SHFV internalization (MOI = 10) by SHFV-specific RT-qPCR-based internalization assay. The extent of internalization is shown as fold from wild-type control (mean ± SD, n = 3 biological replicates). Data shown is representative of three independent experiments. Significance determined by one-way ANOVA with multiple comparisons, **p ≤ 0.01; ***p ≤ 0.001. D Wild-type MA-104 cells were incubated with orilanolimab (open circles) or isotype control (IgG4,k, solid circles) at the indicated concentration (n = 2 biological replicates) in serum free media for 1 hr at 4 °C. Cells were then infected with SHFV (MOI 0.1) and incubated at 37 °C for an additional 24 hrs, followed by plaque assay. Error bars show SEM. E Production of SHFV RNA in cell-culture supernatant for wells containing no cells (grey) wild-type MA-104 cells, ACHN+hCD163 cells, or macaque iPSC-derived macrophages incubated with 80 μg/mL orilanolimab (open circles) or isotype control (closed circles). Wells were inoculated with an MOI = 0.1 and RNA extraction was performed 24 hrs after inoculation on n = 2 biological replicates. Significance determined by unpaired t-test, ***p ≤ 0.001; ****p ≤ 0.0001. F, G Production of KRCV-1 (purple) or PBJV (yellow) RNA in cell-culture supernatant for wells containing no cells (grey) or macaque iPSC-derived macrophages incubated with 80 μg/mL orilanolimab (open circles) or isotype control (closed circles). Wells were inoculated with KRCV-1 or PBJV and RNA extraction was performed 24 hrs after inoculation on n = 2 biological replicates. Significance determined by ANOVA with correction for multiple comparison testing, **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001. H Cytopathic effect and identification of infected cells via mCherry-expressing SHFV reporter-virus in uninfected cells, wild-type MA-104 cells, ACHN+hCD163 cells, or macaque iPSC-derived macrophages, incubated with 80 μg/mL orilanolimab or isotype control, imaged 24 hrs after inoculation with SHFV (MOI = 0.1). Data shows results from one experiment. I Viability, as determined by an ATP-based luminescence assay, of mac iPSC-macs inoculated 24 hrs prior with SHFV, KRCV-1, or PBJV in the presence of 80 μg/mL orilanolimab (open circles) or isotype control (closed circles), n = 2 biological replicates. Significance determined by unpaired t-test, *p ≤ 0.05; **p ≤ 0.01. Source data are provided as a Source Data file.

FcRn overexpression enhances arterivirus infection

To explore the relationship between FcRn and CD163 in arterivirus entry in greater detail, we introduced gFcRn, gCD163, or both, into WT MA-104s, which naturally express very low levels of these pro-viral molecules (Figs. 4A, S2). We then inoculated these cells with a high dose of SHFV and stained cells for viral RNA using single-molecule (sm) RNA FISH 4 hrs after inoculation to capture events related to viral entry. Viral RNA (vRNA) was visualized via confocal microscopy (Fig. 4B) and quantification was performed on multiple images from each condition (Fig. 4C). Compared to WT cells, overexpression of gCD163 had minimal effect on the percentage of vRNA+ cells, the number of puncta per infected cell, or the fluorescent intensity per infected cell. In contrast, introduction of gFcRn (or gFcRn+gCD163) increased each of these parameters. Interestingly, gFcRn+gCD163 dual-overexpressing MA-104 cells did not produce significantly more virus than WT MA-104 cells (Fig. 4D). However, these cells exhibited markedly more severe CPE (Fig. 4E), which translated into larger and more well-defined plaques when these cells were used as a substrate for plaque assay (Fig. 4F). When used to quantify virus in a plaque assay, the enhanced sensitivity of gFcRn (and gFcRn+gCD163) overexpressing cells resulted in detection of 5-10-fold more virus in the same sample (Fig. 4F, G).

Fig. 4: FcRn overexpression enhances arterivirus infection.
figure 4

A Western blot showing gCD163 and gFcRn expression in wild-type (WT) MA-104 cells and MA-104 cells with ectopic expression of gCD163, gFcRn, or both. B Cells from A were infected with SHFV (MOI = 30) and assayed for intracellular SHFV RNA 4 h later using single molecule RNA fluorescence in situ hybridization (smRNA FISH). Confocal microscopy (×600 magnification) was used to capture 16–18 fields of view taken at random (one representative shown). Blue = nuclei; red = smFISH detecting viral (v)RNA, Scale bar = 25 μm. C Quantification of microscopy performed in B. From each field of view, the percent vRNA positive cells, average number of puncta per infected cell, and average puncta FISH intensity per infected cell were determined and graphed as independent points. Error bars represent the mean ± SEM. Data shows results from one experiment. D Time course of SHFV production from wild-type (WT) MA-104 cells (red) compared to dual-overexpressing gCD163/gFcRn cells (black) at a low (0.03, left) and high (3, right) MOI, n = 3 biological replicates. E Cytopathic effect of SHFV infection of wild-type (WT) MA-104 cells (red) compared to dual-overexpressing gCD163/gFcRn cells (black) at a low MOI (0.03) 24 hrs after inoculation. Data shows results from one representative experiment in D. F Photographs of plaque assay wells (6-well plate) fixed and stained with crystal violet 2 days after inoculation. Note the larger and more well-circumscribed plaques in the CD163/FcRn-dual-expressing cells. G Quantification of SHFV via plaque assay using MA-104 cells (WT, +gCD163, +gFcRn, or +gCD163 & +gFcRn) as substrate for infection with data presented as mean values ± SEM, n = 3 biological replicates, normalized to WT cells and analyzed using one-way ANOVA with multiple comparisons relative to WT cells: ns, not significant; ***p ≤ 0.001; ****p ≤ 0.0001. Source data are provided as a Source Data file.

CD163 overexpression does not compensate for the absence of FcRn

Given these results, we wondered if FcRn overexpression could compensate for a complete lack of CD163, and vice versa. To evaluate this, we introduced gFcRn into MA-104ΔCD163 cells and gCD163 into MA-104ΔFcRn cells (Figs. 5A, S3). Compared to WT MA-104s, both of these cell lines were significantly impaired in their ability to produce infectious virus above background levels, with the MA-104ΔFcRn + gCD163 cells able to produce ~1 log of virus relative to background while the MA-104ΔCD163+gFcRn cells were completely incapable of supporting infection (Fig. 5B). Similar trends were also observed using smFISH to evaluate the percent of vRNA+ cells, the average puncta per cell, and the average intensity per puncta (Fig. 5C).

Fig. 5: CD163 overexpression does not compensate for the absence of FcRn.
figure 5

A Western blot showing gCD163 and gFcRn expression in wild-type (WT) MA-104 cells, overexpression of gCD163 in MA-104ΔFcRn cells, and gFcRn overexpression in MA-104ΔCD163 cells. B Kinetics of SHFV production from cells in (A) after inoculation with SHFV, MOI of 3. Graph shows mean ± SEM from three independent experiments. All plots include error bars; no error bars are shown when the SEM was smaller than the size of the symbols. C Cells from A were infected with SHFV (MOI = 30) and assayed for intracellular SHFV RNA 4 h later using single molecule RNA  fluorescence in situ hybridization (smRNA FISH). Confocal microscopy (×600 magnification) was used to capture 17–21 fields of view taken at random. From each field of view, the percent viral RNA positive cells, average number of puncta per infected cell, and average puncta FISH intensity per infected cell were determined and graphed as independent points. Error bars represent the mean ± SEM. Note that data for the WT condition is the same as that shown in Fig. 4C. Data shows results from one experiment. Source data are provided as a Source Data file.

FcRn and CD163 synergize to sensitize non-susceptible cells to arterivirus infection

Vero cells are another Chlorocebus-derived kidney epithelial cell line that are widely used to culture many different types of viruses, but unlike MA-104s Vero cells are not susceptible to arterivirus infection. We wondered whether this resistance could be due to a lack of FcRn, CD163, or both. To explore this idea, we first performed a Western blot and confirmed the absence of FcRn and CD163 in WT Vero cells; we then introduced gFcRn, gCD163, or both into Vero cells (Figs. 6A, S4). The addition of gFcRn alone was unable to sensitize Veros to SHFV infection. Addition of gCD163 did allow for a small but measurable amount of productive infection, but titers only achieved ~1 log greater than our limit of detection and no CPE was noted. However, addition of both gFcRn+gCD163 had a synergistic effect, resulting in a ~ 4.5 log increase in SHFV production and CPE typical of an SHFV permissive cell line (Fig. 6B, C).

Fig. 6: FcRn and CD163 synergize to sensitize non-susceptible cells to arterivirus infection.
figure 6

A Western blot showing gCD163 and gFcRn expression in wild-type (WT) Vero cells and Vero cells ectopically expressing gCD163, gFcRn, or both. B Kinetics of SHFV production from cells in A after inoculation with SHFV, MOI of 0.03, n = 3 biological replicates, graph show mean ± SEM. All plots include error bars; no error bars are shown when the SEM was smaller than the size of the symbols. C Cytopathic effect of cells in B at 72 hrs post-inoculation (colors match those in B). Source data are provided as a Source Data file.

FcRn is a molecular barrier to arterivirus cross-species infection

Emerging evidence suggests that the ability of divergent arteriviruses to utilize CD163 orthologs from different host species may be an important determinant of cross-species infection28. To examine whether FcRn represents a similar species-barrier, we cloned FCGRT cDNA from grivet, pig, human, and mouse tissue; we also synthesized the equine FCGRT ortholog. These cDNAs were ectopically expressed in MA-104ΔFcRn cells which were then inoculated with SHFV, PRRSV, and EAV (Fig. 7A). For comparison, we also performed a similar experiment using CD163 orthologs, introducing these genes into MA-104ΔCD163 cells (Fig. 7B). While SHFV could only use grivet and human CD163 orthologs for infection, PRRSV could utilize all CD163 orthologs tested and EAV infection was independent of CD163. A similar pattern emerged for FcRn, with SHFV productively utilizing grivet, human, and horse—but not mouse or pig FcRn. Interestingly, PRRSV could utilize all FcRn orthologs tested except for mouse, while EAV could utilize all orthologs. Thus, FcRn appears to represent a significant molecular barrier to arterivirus cross-species infection at the molecular level.

Fig. 7: FcRn is a molecular barrier to arterivirus cross-species infection.
figure 7

A FcRn orthologs from natural and potential arterivirus hosts—grivet (g), human (hu), mouse (m), pig (p), and horse (ho)–were introduced into MA-104ΔFcRn cells and inoculated with SHFV (red), PRRSV-2 (green), or EAV (blue) at an MOI of 0.01, n = 2 biological replicates. FcRn orthologs derived from the natural host of an arterivirus (i.e., positive control) are denoted by a black outline. Productive infection was then assessed via plaque assay of supernatant at 2 days post inoculation. B Similar experimental setup to A, except with CD163 orthologs in MA-104ΔCD163 cells. Source data are provided as a Source Data file.

FcRn/CD163-overexpressing cells can be used to isolate fastidious arteriviruses

Among simian arteriviruses, SHFV is unique in its ability to be cultivated on immortalized cell lines. We wondered if overexpression of FcRn and CD163 could render MA-104 cells permissive to infection with highly-divergent simian arteriviruses. To test this, we inoculated MA-104+gFcRn+gCD163 cells with KRCV-1 and PJBV, using SHFV as a control. MA-104+gFcRn+gCD163 cells were able to support high-titer replication of KRCV-1, as evidenced by increasing concentrations of KRCV-1 RNA in the supernatant (Fig. 8A) and the induction of cytopathology (Fig. 8B). However, the same was not true for PJBV. The supernatant positive for KRCV-1 RNA formed plaques on MA-104+gFcRn+gCD163 cells (Fig. 8C, D), substantiating the production of infectious virus and enabling PFU-based quantification of KRCV-1.

Fig. 8: FcRn/CD163-overexpressing cells can be used to isolate new arteriviruses.
figure 8

A Growth kinetics of three simian arteriviruses, SHFV (red), KRCV-1 (purple), and PBJV (yellow), on wild-type MA-104 cells (open circle) versus MA-104 cells ectopically expressing gFcRn and gCD163. Data are presented as mean values ± SEM, MOI = 1, n = 3 biological replicates. Ct values are shown on a reverse axis, with the SHFV standard curve (in N-gene copies per mL) shown on the right Y-axis of the SHFV plot for comparison, along with the limit of detection (dashed line). B Cytopathic effect of these viruses on these cells at 6 days post inoculation. Data shows representative results from one of two independent experiments. C Plaques of KRCV-1 on MA-104+gFcRn+gCD163 cells. D Production of infectious KRCV-1 on MA-104+gFcRn+gCD163 cells, plaqued on MA-104+gFcRn+gCD163 cells 4 days post inoculation Data are presented as mean values ± SEM, n = 3 biological replicates. Significance was determined using an two-tailed unpaired t test, ***p ≤ 0.001.

Discussion

FcRn plays several unique physiologic roles, the most prominent of these being the “recycling” of albumin and IgG molecules that are internalized by cells (mainly endothelial cells and macrophages) via nonspecific mechanisms like pinocytosis42. Upon acidification of the endosomal compartment, protonation of residues on IgG and albumin lead to enhanced affinity for FcRn, which “rescues” these molecules from proteasomal degradation. This common and highly orchestrated process clearly presents an attractive point of vulnerability that multiple unrelated groups of viruses are able to exploit to enter cells. This is particularly noteworthy given that non-enveloped viruses (i.e., echoviruses and astroviruses) and enveloped viruses (i.e., arteriviruses) have evolutionarily converged on this pathway as a means of cellular entry.

The specific way(s) in which each of these different viruses interacts with FcRn—and the unique and shared features amongst these interactions—is an important topic of ongoing investigation. Within the echoviruses, strains E2, E15, and E18 use FcRn for attachment and uncoating of the viral capsid43, whereas most other echoviruses employ a two-step entry process in which CD55 mediates virion attachment and FcRn mediates uncoating following endosomal acidification38,50. Our data imply that arterivirus entry is also a multistep process involving FcRn and CD163. Although CD163 has historically been referred to as a “receptor” for PRRSV51 and, more recently SHFV28, there is little evidence that CD163 functions as the fusogenic receptor in arterivirus entry (i.e., mediates fusion of the viral and host membranes). Both FcRn and CD163 cycle between the cell surface and early endosomes42,52, so the timing of FcRn and CD163 binding could plausibly happen in any order. Although binding of arteriviruses to CD163 and FcRn could occur simultaneously, multiple lines of evidence presented here (blocking SHFV infection of CD163-overexpressing cells with an anti-FcRn antibody; the requirement of both FcRn and CD163 to render non-permissive cells susceptible to infection) strongly suggests that interactions between arterivirus virions, FcRn, and CD163 occurs in a step-wise manner. Deciphering this process is made all the more complicated by the unusual preponderance of glycoproteins decorating arterivirions—none of which resemble a typical class I, II, III, or IV viral fusion protein53. Nevertheless, determining the timing of these events is an immediate priority for arterivirus research.

In normal physiology, once IgG and albumin bind to FcRn, they are rescued from degradation and released back into circulation, effectively extending the halflife of circulating IgG and albumin by 3–5 fold42. Consequently, antagonism of FcRn (e.g., by anti-FcRn antibodies) dramatically reduces the halflife of albumin and IgG resulting in hypoalbuminemia and hypogammaglobulinemia42,54,55. Thus, it is tempting to speculate that arteriviruses function as FcRn antagonists as part of their immune-evasion strategy, compromising humoral immunity and potentially contributing to arterivirus persistence in individual hosts. It is also possible that FcRn binding may traffic the virus into a non-degrading pathway, further promoting the virion’s survival upon endocytosis. In the absence of CD163 (i.e., in a non-permissive cell), it is theoretically plausible that arterivirus virions are treated simply as “cargo” by FcRn, which could have implications for arterivirus persistence and vertical transmission. For example, in FcRn+/CD163 endothelial cells, this could result in the intracellular storage of arterivirus virions along with IgG and albumin, allowing for the slow release of virions back into circulation; thus creating a semi-“latent” reservoir of virus that could play an important role in persistence. In FcRn+/CD163 syncytiotrophoblasts, FcRn may actively transport arterivirus virions across the blood-placental barrier in a manner similar to the transcytosis of maternal IgGs into fetal circulation; this could explain how PRRSV efficiently transmits across the placental barrier to infect fetal piglets56.

Aside from the biological implications of our discovering FcRn as a pan-arterivirus receptor, this finding has practical implications for all virologists working with arteriviruses. Ectopic expression of FcRn enhances the sensitivity of cells to arterivirus infection, resulting in faster-growing and more well-defined plaques. Plaque assays performed on these cells also yield significantly greater virus titers compared to wild-type MA-104 cells, implying that wild-type MA-104 cells and their MARC-145 subclones—which have served as the gold-standard for plaquing arteriviruses for decades—inaccurately under-quantify infectious arterivirus titers. FcRn/CD163-overexpressing cells will also be a useful tool for isolating fastidious arteriviruses that refuse to grow in MA-104s, which includes the majority of simian arteriviruses.

With viral metagenomic surveys continuing to identify new arteriviruses, it will be interesting to see how (and if) these different viruses utilize FcRn in their entry process. Our data suggests that at least five divergent arteriviruses (SHFV, KRCV-1, PBJV, PRRSV-2, and EAV) use FcRn in their entry process. Understanding the nuances of species-specific FcRn utilization will be important in anticipating the spillover of arteriviruses into new hosts, as our data indicates that FcRn can function as a molecular barrier to cross-species arterivirus infection. FcRn utilization also represents a potential point of vulnerability in the arterivirus life cycle that could be targeted for therapeutic and/or prophylactic purposes. As we show, clinical and pre-clinical anti-FcRn antibodies can block infection of cells in vitro with diverse arteriviruses; thus, it may be possible to prevent arterivirus infection or transmission in vivo via injection of an anti-FcRn antibody. Given that no direct-acting antivirals for arteriviruses exist, a host-directed pan-arterivirus countermeasure (e.g., anti-FcRn monoclonal antibody) may prove particularly useful in the event of a novel arterivirus emergence.

Methods

Ethics statement

The use of animals for this research was reviewed and approved by the UW-Madison IACUC.

Cell lines

MA-104 cells used in this study came from two different sources. Experiments performed in the lab of ALB, which included the CRISPR-knockout screen and generation of FCGRT-knockout cells, used MA-104 cells from Megan Baldridge, originally from the lab of Harry Greenberg (Stanford). Experiments performed in the laboratory of CJW used MA-104 cells purchased from ATCC (catalog # CRL-2378.1). Most experiments were performed independently by each lab, and major differences between the two lineages of MA-104 cells were not observed. The creation and maintenance of ACHN+hCD163 cells was described previously28. MA-104 and ACHN cells were maintained in DMEM (Gibco) supplemented with 10% FCS (Omega Scientific) + 1% L-Glutamine (Gibco), +1% Sodium Pyruvate (Gibco) + 1% Hepes (Gibco). iPSC macrophages were differentiated and maintained as previously described57. MA104 cells that were trans-complemented with Sleeping Beauty constructs were selected in either 10 µg/mL puromycin (Gibco) or 250 µg/mL Hygromycin B (Gibco). ACHN and MA-104 CRISPR-knockout lines were maintained in 10 µg/mL or 5 µg/mL Blasticidin (Gibco) respectively. Grivet kidney Vero E6 cells (ATCC Cat# CRL-1586) were maintained in DMEM supplemented with 10% FBS and 1× penn/strep. Vero cells engineered to stably express CD163 and/or FcRn were selected using 200 µg/mL hygromycin and/or 10 µg/mL puromycin, respectively.

Viruses

SHFV strain LVR and PRRSV-2 strain VR-2332 were obtained from Dr. Kay Faaberg (USDA–Ames) and propagated/titrated on MA-104 cells. Human echovirus 30 was obtained from ATCC (catalog # VR-1660) and propagated/titrated on MA-104 cells. SARS-CoV-2 strain USA-WA1/2020 was obtained from BEI Resources (catalog # NR-52281) and propagated/titrated on Vero E6-TMPRSS2-T2A-ACE2 (BEI catalog # NR-54970). YFV strains Asibi and 17D were obtained from WRCEVA and propagated/titrated on Vero E6 cells from ATCC (CRL-1586). KRCV-1 and PBJV stocks were obtained from Jens Kuhn (NIH, NIAID)58 and stocks prepared via inoculation of macaque iPSC cells described previously49. A recombinant SHFV cDNA-launch plasmid expressing eGFP was provided by Jens Kuhn. To avoid autofluorescence with macrophage cell lines, eGFP was removed by restriction enzyme digest (NotI and SpeI, NEB #R3189S, #R3133S), and the gene for mCherry was added by gibson assembly (NEB, #E2621S). A stock was generated by transfecting MA-104 cells with 2 µg of plasmid using TransIT-LT1 (Mirus, #MIR 2304) following the manufacturer’s instructions.

Plaque assays

Prior to plaquing, MA-104 cells were seeded at a density of 2 × 105 cells per well in a 24-well plate (TPP). Ten-fold serial dilutions were performed in DMEM + 2% FCS + 1%Hepes and 400 µL of diluted virus was added to each well. Plates were incubated for 1 hr at 37 °C with gentle rocking every 15 min. After 1 h, 400 µL of prewarmed 1% (w/v) methylcellulose (Sigma Aldrich) in 1× MEM (Gibco) was added to each well. Plates inoculated with SHFV, SARS-CoV-2, Echovirus 30 were cultured for 2 days. PRRSV and EAV were cultured for 3 and 4 days respectively. Prior to staining, cells were fixed in a final volume of 4% PFA in PBS then plaques were visualized using 0.05% crystal violet in methanol. Plaque counts were enumerated manually.

Focus forming assays

YFV focus-forming assay was performed as described previously59, using 2D12 hybridoma (ATCC, CRL-1689) supernatant (1:1000) as primary antibody.

CRISPR-knockout screen in MA-104 cells

2.4 × 107 MA-104 cells containing the grivet (Chlorocebus sabaeus) sgRNA Library (Addgene Pooled Library #178284) were inoculated with an MOI ≈ 1 for both SHFV and PRRSV. CPE was monitored daily. When CPE was nearly complete, all but 5 mL of media was removed and replenished with complete cell growth media. Cultures were observed to undergo multiple rounds of cytolysis followed by outgrowth of surviving cells; this ceased approximately three weeks after virus inoculation, at which time surviving cells were collected (~8 × 106 cells) and DNA was extracted using the NucleoSpin Blood XL Kit (Macherey-Nagel) following manufacturer’s instructions. PCR reactions to amplify libraries were carried out using Platinum II Taq Hot Start DNA Polymerase (Invitrogen) in accordance with the “PCR protocol for Illumina sequencing preparation” associated with the sgRNA library on AddGene’s website. Illumina deep sequencing was performed at the UW–Madison Biotechnology Center on a shared 150 × 150 NovaSeq run targeting 1 M reads per condition. sgRNA sequences targeting specific genes were pre-processed (removal of adaptor sequences) and analyzed through the published tool MAGeCK version 0.5.9.560.

CRISPR-knockout of FcRn in cell lines

Sequence for an sgRNA targeting the Chlorocebus sabaeus FCGRT gene (AGAAAGAGACCACAGATCTG) was pulled from the library file associated with the Chlorocebus sabaeus CRISPR-knockout library61. Alternatively, the CRISPick algorithm (GRCh38, SpyoCas9) was used to design a sgRNA targeting the Homo sapiens FCGRT gene (ACTGGGCCCTGACAACACCT)62,63. These sgRNA were cloned into the the LentiCRISPR_v2-blasticidin lentiviral vector (Addgene plasmid #83480) which was used to introduce sgRNAs into MA-104 and ACHN + CD163 cells, respectively. Knockout was confirmed by deep sequencing (Illumina NovaSeq) across the sgRNA-edited target region, high sensitivity mass spectrometry, and Western blot.

Genetic trans-complementation

RNA extracted from MA-104 cells (grivet), ACHN cells (human), primary peritoneal macrophages (mouse), and primary lung tissue (pig) was used to generate FCGRT cDNA for the amplification and cloning of FCGRT. The horse FCGRT gene was synthesized de novo (IDT). These DNA fragments were then cloned into the pSBbi Sleeping Beauty vector system (Addgene)64, which was used to stably express FcRn orthologs in cell lines.

Western blot

Cells were lysed in Nonidet P-40 (NP-40) lysis solution composed of the following: 150 mM NaCl (Promega #VA4221), 50 mM tris-HCl pH 7.4, 1% NP-40 (G-Biosciences, #DG001), 1 mM DTT (IBI Scientific, #IB2104), 1 l/mL Benzonase (Sigma, #E1014-25KU), and 1× EDTA-Free Protease Inhibitor mixture (Sigma # 11873580001). Briefly, cells were detached from tissue-culture plates by trypsin treatment, inactivated with FBS-containing media, and then pelleted (300 × g for 5 min) and washed once with PBS. PBS-washed cells were resuspended in the NP-40 lysis solution and incubated on ice for 30 min. Cell lysates were cleared by centrifugation at 16,000 × g for 15 min at 4 ºC. Whole-cell extracts were quantified using a BCA protein assay (ThermoFisher Scientific, #23227). 10 µg of total protein was resolved with polyacrylamide gel electrophoresis using gels prepared with the SureCast gel handcast system (ThermoFisher Scientific, #HC1000SR). Proteins were electrophoretically transferred to Immobilon-P PVDF membranes using a wet transfer apparatus set at 20 V for 1 h. Membranes were blocked for 1 hr at ambient temperature in blocking agent (Nestle Carnation 5% nonfat dried milk in TBST [1× tris-buffered saline + 0.1% Tween 20]). Primary antibodies, including CD163 (goat polyclonal anti-CD163 antibody [R&D Systems, #AF1607]), FcRn (rabbit polyclonal anti-FcRn [Abcam, #ab193148]), and beta-actin (mouse monoclonal 8H10D10 [Cell Signaling Technology, #3700 S]) were diluted in TBST + 5% nonfat dried milk and incubated either overnight at 4 °C (CD163 and FcRn) or for 1 h at ambient temperature (beta-actin). After primary incubation, cells were washed 3 times for 5 min in TBST. Secondary antibodies donkey anti-goat IgG HRP conjugate (Thermo-Fisher, #A16005), anti-mouse IgG HRP conjugate (Cell Signaling, #7076 S), anti-rabbit IgG HRP conjugate (Cell Signaling, #7074 S) were diluted in TBST + 5% nonfat dried milk and incubated with the membrane for 1 h at ambient temperature. After secondary incubation, membranes were washed 3 times for 5 min in TBST, developed using (ECL western blotting detection reagent Amersham, #RPN2232), and imaged on a ChemiDoc imaging System (Bio-Rad).

RNA FISH-flow

RNA FISH-Flow assays were performed as described previously28,41. Briefly, at the indicated times following virus exposure, the cell culture media was removed, and the cells were washed with PBS. Cells were detached with trypsin and then transferred to individual wells of a 96-well v-bottom plate. Cells were washed once with PBS and fixed with 4% PFA for 10 mins. After fixation, the cells were washed three times with PBS and then permeabilized with cold 70% EtOH for at least 1 h at 4 ºC. EtOH permeabilized cells were washed once with Wash Buffer A and resuspended in 50 µl of hybridization buffer containing probe (1:1000 dilution of SHFV-ORF1a-Atto633 labeled probes in hybridization buffer). The plate was incubated in the dark for 2 h at 37 ºC. After the incubation, 150 µl of wash buffer A was added and the cells were pelleted (500 × g for 5 min). Pelleted cells were resuspended in wash buffer A and incubated in the dark at 37 °C for 30 min. After incubation, the cells were pelleted, washed once in wash buffer B, and resuspended in FACS buffer (PBS + 2% FBS and 1 mM EDTA). Samples were analyzed on an Attune Flow Cytometer (ThermoFisher Scientific). At least 20,000 live cell events were collected following singlet discrimination. Data was analyzed with FlowJo v10.8.2.

Single-molecule FISH

Single-molecule FISH (smFISH) was performed as described previously28. Briefly, cells were seeded onto acid-etched coverslips in 6-well plates at ~80% confluence. Cells were infected with SHFV (MOI = 30) or mock exposed for 1 h at 37 °C 5% C02. Following exposure, the cells were washed with PBS (3 times), the media was replaced (EMEM 2% FBS), and then the cells were allowed to continue incubating for 3 additional hours. After the incubations (4 h total), the cell growth medium was removed and the cells were washed once with PBS. The cells were then fixed (4% PFA in PBS for 10 min at room temperature), permeabilized (70% ice-cold ethanol for 1 h at 4 °C), and washed with Wash Buffer A (2× SSC, 10% deionized formamide, diluted in nuclease-free water). The coverslips were removed from the 6-well dish and inverted (cell side down) onto hybridization buffer (10% dextran sulfate, 10% deionized formamide, 2× SSC, diluted in nuclease-free water) containing ATTO633 labeled SHFV ORF1a RNA FISH-Flow probes (diluted 1:200) overnight (<16 h; in the dark at 37 °C) in a humidified chamber. For more details on probe design and buffer preparation, see ref. 41. The next day, the coverslips were transferred to a new six-well dish containing Wash Buffer A and incubated in the dark at 37 °C for 30 min. After incubation, the Wash Buffer A was removed and replaced with Wash Buffer B (2× SSC diluted in nuclease-free water) for 5 min. The coverslips were then mounted onto slides using ProLong Gold Antifade Mountant with DNA stain DAPI (Invitrogen) and imaged after an overnight curing step.

Confocal microscopy and image analysis

All images were captured using a Nikon Ti2 confocal microscope with a 60× oil lens. The data was analyzed using NIS-Elements AR 5.30.06 software (Nikon.inc). The number of cells and virus particles in each sample field was determined using the segmentation function. The intensity of virus particles (smFISH puncta) was quantified through binary processing, measurement, and data management functions. For all samples, the data was collected with identical parameters, and imaging and analysis followed a double-blind procedure.

Receptor-bypass transfection experiments

Cells were seeded in 6-well dishes at 3 × 105 cells per well. The next day, cells were transfected with lipofectamine 3000 using the following conditions: Tube A (150 ul Opti-MEM, 4.5 μl lipofectamine 3000) and Tube B (150 μl Opti-MEM, 3 μg recombinant SHFV [rSHFV] expressing cDNA-launch plasmid, and 6 μl P3000). The contents of tube A were mixed into tube B, incubated for 10 min at ambient temperature, and then added dropwise to cells. To reduce the toxicity associated with residual transfection reagent, the media was changed to EMEM 2% FBS 4–6 h post transfection. The cells were allowed to incubate for the indicated timepoints, and then the cell supernatant was harvested for virus titrations by plaque assay.

Quantification of virus by RT-qPCR

For quantification of viral RNA in cell culture supernatants or sera, RNA was extracted from a 20- or 50-μL sample using a KingFisher Flex (Thermo Fisher Scientific) with the MagMAX Viral/Pathogen Nucleic Acid Isolation Kit (Thermo Fisher Scientific). RNA was eluted in 50 μL H2O, of which 8.5 μL was carried forward for quantitative reverse-transcription PCR (RT-qPCR). Virus-specific TaqMan assays were designed for detection of each virus using GenBank-deposited simarterivirus genome sequence data, Primer3 (https://primer3.org/) for primer and probe design, and IDT for primer and probe synthesis based on the FAM and Zen/Iowa-Black dual-quencher system—KRCV-1-F: ACACGGCTACCCTTACTCC; KRCV-1-R: TCGAGGTTAARCGGTTGAGA; KRCV-1-P: TTCTGGTCCTCTTGCGAAGGC; PBJV-F: GAGGATGGTCGCCTCAACTA; PBJV-R: AAGGACCCTCGTCAAATTCA; PBJV-P: TGCTGTCATCACACCAGATG; SHFV-F: CGACCTCCGAGTTGTTCTACCT; SHFV-R: GCCTCCGTTGTCGTAGTACCT; SHFV-P: CCCACCTCAGCACACATCAAACAGCT. The TaqMan RNA-to-CT 1-Step Kit (Thermo Fisher Scientific) was used for RT-qPCR setup. Thermocycling was performed on a Quantstudio 6 Pro (Thermo Fisher Scientific) using the following conditions: reverse transcription at 48 °C for 15 min; initial denaturation at 95 °C for 10 min; 50 cycles at 95 °C for 15 s followed by incubation at 60 °C for 1 min, during which signal acquisition was performed. Using a custom-designed in vitro-transcribed RNA standard curve for SHFV, we determined that cycle threshold (Ct) values of ≈38–40 corresponded to ≈10 RNA copies per reaction and thus represented a generalizable limit of detection for the purposes of this study.

Blocking assays

Wild type MA-104 cells, ACHN + hCD163, iPSC macrophages were plated at a seeding density of 4 × 104 cells in a 96-well plate (TPP Techno Plastic Products AG) in complete growth media. Prior to inoculation, plating media was removed and orilanolimab (Invitrogen, #MA5-42215) or IgG4,k isotype control (MedChemExpress, #HY-P99003) in serum free media was added. Cells were preincubated at 4 °C for 1 hr. After the initial incubation, SHFV and KRCV-1 (MOI 0.1) were added and the cells were incubated for 24 hrs at 37 C, 5% CO2 without removal of virus or antibody. PBJV was tested using a MOI 0.10-equivalent-volume. Cell supernatant was collected and stored at −80 °C prior to titration by plaque assay or RT-qPCR.

Viability assay

Viability was assessed using CellTiter-Glo 2.0 (Promega #G9241) following the manufacturer’s instructions. Luminescence was measured using a CLARIOstar plate reader (BMG Labtech).

Statistics and reproducibility

Based upon the relatively large effect size observed in many of the experimental systems, group sizes of 2–4 were chosen based primarily on logistical and workflow considerations; no formal statistical analysis was used to predetermine sample sizes. To enhance rigor and reproducibility, we attempted to show a statistically significant effect in as many different virus and host systems as possible. We considered the number of groups and distribution of data when selecting statistical tests; we also followed best practices regarding multiple comparison testing. No data were excluded from the analyses and experiments were not randomized. The Investigators were not blinded to allocation during experiments and outcome assessment.