Pathogenic hantaviruses are maintained world-wide within wild, asymptomatic rodent reservoir hosts, with increasingly frequent human spillover infections resulting in severe hemorrhagic fever or cardio-pulmonary disease. With no approved therapeutics or vaccines, research has, until recently, focused on understanding the drivers of immune-mediated pathogenesis. An emerging body of work is now investigating the mechanisms that allow for asymptomatic, persistent infections of mammalian reservoir hosts with highly pathogenic RNA viruses. Despite limited experimental data, several hypotheses have arisen to explain limited or absent disease pathology in reservoir hosts. In this study, we directly tested two leading hypotheses: 1) that reservoir host cells induce a generally muted response to viral insults, and 2) that these viruses employ host-specific mechanisms of innate antiviral antagonism to limit immune activation in reservoir cells. We demonstrate that, in contrast to human endothelial cells which mount a robust antiviral and inflammatory response to pathogenic hantaviruses, primary Norway rat endothelial cells do not induce antiviral gene expression in response to infection with their endemic hantavirus, Seoul orthohantavirus (SEOV). Reservoir rat cells do, however, induce strong innate immune responses to exogenous stimulatory RNAs, type I interferon, and infection with Hantaan virus, a closely related hantavirus for which the rat is not a natural reservoir. We also find that SEOV-infected rat endothelial cells remain competent for immune activation induced by exogenous stimuli or subsequent viral infection. Importantly, these findings support an alternative model for asymptomatic persistence within hantavirus reservoir hosts: that efficient viral replication within reservoir host cells may prevent the exposure of critical motifs for cellular antiviral recognition and thus limits immune activation that would otherwise result in viral clearance and/or immune-mediated disease. Defining the mechanisms that allow for infection tolerance and persistence within reservoir hosts will reveal novel strategies for viral countermeasures against these highly pathogenic zoonotic threats.

The data collected here were generated from in vitro (cell culture) studies of virus replication and innate immune activation in rat and human endothelial cells. We collected microscopy images (tif, jpeg), western blotting images (tif, png), and RT-PCR analyses (excel, prism). 

Viruses and in vitro infections

Seoul virus strain SR11 and Hantaan virus strain 76-118 were propagated on Vero E6 cells (ATCC, CRL-1586) for 12 days. Infectious virus was isolated by harvesting supernatant and centrifuging at 1000rcf for 10 minutes to remove cellular debris. Rat-passaged SEOV was generated using SR11 stock and propagated over three passages (MOI 0.01) in RLMVECs for ten days each. SEOV Baltimore strain, kindly provided by Dr. Steven Bradfute (UNM), was generated in Vero E6 cells and harvested 12 days post-infection. Sendai virus/52 was purchased from ATCC (Sendai/52, VR-105). For virus infections, cells were seeded in cell culture vessels 18-24 hours prior to infection at a target density of 70%. Virus stock was diluted to the target, cell-specific MOI/HAU using serum-free Dulbecco’s modified Eagle’s medium (DMEM, VWR 45000-304) supplemented with 1x pen/strep, 1% nonessential amino acids, 2.5% HEPES, and cells were infected for one hour at 37ºC. Cells were washed twice with sterile PBS solution (FisherScientific, SH30264FS), and appropriate culture medium was added for the duration of the experiment. For infected cell RNA experiments, Vero E6 cells were infected with SEOV for five days at MOI 0.5 prior to harvest in TRIzol reagent and subsequent RNA extraction. Percent SEOV N positive and infectious unit quantification experiment was performed using SEOV stock virus derived from Vero E6 cells, 1:2 dilution series in serum-free media was performed and cells were infected in a 96-well plate (Agilent, 204624-100) for 1 hour at 37ºC followed by addition of methylcellulose (Sigma M0512-500G) overlay with supplemented at 2x concentration DMEM (Gibco, 12100-046) supplemented with 2% FBS, 1x PenStrep, 1% HEPES. UV inactivated virus was treated with UV radiation (FisherBrand UV crosslinker, 13-245-221) until specific absorption of 103.8 mJ/cm².

Cell Culture

Vero E6 cells (ATCC, CRL-1586) and HEK293T (ATCC, CRL-3216) cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% heat-inactivated FBS, 1% pen/strep, 1% nonessential amino acids, 2.5% HEPES. Primary rat microvascular endothelial cells (RLMVEC, VEC Technologies) were cultured in MCDB-131 base medium (Corning) supplemented with EGMTM-2 Endothelial SingleQuots Bullet Kit (Lonza, CC-4176) and 10% heat-inactivated FBS. Human umbilical endothelial cells (HUVEC-C; ATCC, CRL-1730) were cultured in Lonza EGM-Plus (Lonza, CC-4542) supplemented with bullet kit (Lonza, CC-5036) and 10% heat-inactivated FBS. All endothelial cells were cultured in tissue culture-treated plastics coated with rat tail collagen (VWR, 47747-218). HUVEC RLR CRISPR knockout lines were generated and validated previously (69). Recombinant rat IFNβ was purchased from R&D Systems (13400-1) and used at 10-150U/milliliter. RLMVEC knockout cells were generated using LentiCrispr V2 plasmid system (88) with gRNAs targeting RIG-I, MDA5, or Cas9 Scramble as control (Table 1). Briefly, HEK293T cells were transfected using ProFection Mammalian Transfection System (Promega, PAE1200) with targeted LentiCrispr V2 plasmid (Addgene Plasmid #98290), pSPAX (Addgene #12260), and p-VSVG (Addgene #138479). One day post-transfection, fresh DMEM was added to HEK293T cells. Virus was harvested two days post-transfection from HEK293T supernatant, filtered through a 0.45μm filter, and added to RLMVECs for 24 hours. Single knockout cells were placed under puromycin (VWR, AAJ67236-8EQ) selection at 1μg/milliliter for one week. RLR knockout cells were generated using the opposite RLR gRNA (e.g., RLMVEC RIG-I^-/- with puromycin resistance, then treated with MDA5^-/- gRNA with blastocydin resistance) in a LectiCrispr V2 plasmid encoding blastocydin selection and placed under selection (10μg/milliliter) for 1 week. Knockout efficacy was verified via treatment with rat recombinant IFNβ and immunoblot (Supplemental Fig S5).  

Table 1. gDNA sequences for RLMVEC CRISPR knockout lines.

Focus Forming Unit Assay

Infectious virus was quantified using immunostaining as previously described (60). Briefly, Vero E6 cells were infected with 100μL of culture supernatants and incubated for two hours at 37ºC. After incubation, a 2% carboxymethylcellulose overlay containing supplemented DMEM (2% pen/strep, 2% nonessential amino acids, 2% HEPES, 4% heat-inactivated FBS). Cells were incubated for seven days at 37ºC. After incubation, cells were fixed with 95% EtOH:5% Acetic Acid for 10 minutes at -20ºC. Cells were then probed for SEOV N (α SEOV nucleocapsid, custom generated with Genscript) or HTNV N (anti-HTNV nucleocapsid 76–118, BEI resources NR-12152) at room temperature for 2 hours or 4ºC overnight. HRP conjugated secondary antibodies (donkey α rabbit for α HTNV, Jackson Immunoresearch 711-035-152; or goat α mouse for α SEOV, Jackson Immunoresearch 115-035-003) were incubated for 2 hours at room temperature. Foci were then stained using Vector VIP Substrate Kit (Vector Laboratories, SK-4600) and counted under a light microscope to calculate titer.

Immunofluorescence and Microscopy Analysis

To calculate SEOV N positivity and cell-specific titers, Vero E6, HUVEC, and RLMVEC were cultured in 96-well plates and infected with serial dilutions of SEOV as described in infection methods above. 24 hours post-infection, cells were fixed with 95% EtOH:5% Acetic Acid solution for ten minutes at -20ºC then blocked for one hour in 3% FCS in PBS. SEOV N positive RLMVECs at 0.05, 0.1, and 0.25 MOI timecourse were cultured in 6-well plates and fixed and blocked in the same manner as the 96-well plate assay. Cells were probed overnight at 4°C for SEOV N protein (custom, Genscript) at dilution 1:400 in 1xPBS and with secondary antibody AlexaFluor 555 goat a mouse (Thermo Fisher Scientific, A-31570) at dilution 1:400 for two hours at room temperature. Nuclear stain DAPI was used at 1:10,000 for ten minutes at room temperature (SeraCare, 5930-0006). High-content imaging was acquired on CellInsight CX7 High-content Analysis platform (Thermo Fisher Scienitific, CX7A1110) at 10x objective magnification using iDev software. SEOV N positive cells were quantified using the Spot Detector Applet in the iDev software. Regions of interest (ROI) were defined based on DAPI nuclear stain, and SEOV N positive cells were defined as having at least one spot within the ROI as thresholded relative to each cell type. Data were collected on a total of 25 fields per well for all cell types for 96-well plate, or 250 fields per well for 6-well plates. Representative images were acquired on the EVOS FL Auto imaging system (Thermo Fisher Scientific, AMF7000) at 10x magnification for 96-well plate, or 20x magnification for MOI series 6-well plate. RLMVECs assayed for Mx1/2/3 expression during SEOV infection were fixed using 4% paraformaldehyde (VWR, 97064-606) solution in sterile PBS for 30 minutes at room temperature, then permeabilized with 0.01% Triton X-100 (VWR, 97062-208) and blocked for one hour in 3% FCS in PBS. Cells were treated with DAPI nuclear stain 1:1000, SEOV N nucleocapsid 1:400, or Mx 1/2/3 1:400 in PBS (Santa Cruz Biotechnology, sc-166412 AF488). Representative images were acquired via the EVOS FL Auto imaging system at 20x magnification (Thermo Fisher Scientific).

Cell-specific Titer Calculations

Cell-specific titers were calculated by determining the number of SEOV N-expressing cells for each cell type using a serial dilution of SEOV stock. This method is adapted from a recent publication by Menke et al. (61). Briefly, total number of N expressing cells was quantified as described above. 

Titer was calculated for all replicate wells in each dilution for which the number of N-positive cells decreased two-fold, to match the inoculum dilutions. Replicates were averaged and then the calculated titers for each dilution condition were averaged to determine the titer on each cell type. Four replicates per dilution and >3 independent experiments were performed for each cell type.

RNA Methods

Cells were lysed for RNA analysis using TRIzol Reagent (ThermoFisher Scientific, 15596026). RNA used for infected cell RNA transfection was isolated using phenol:chloroform extraction following manufacturer protocol. RNA used for gene expression analysis was purified using Zymo Research Direct-zol RNA Miniprep Plus kit (VWR 76211-340) according to manufacturer’s instructions. RNA concentrations were quantified using a NanoDrop UV-Vis Spectrophotometer (ND-1000). cDNA was synthesized using Applied Biosystems High-Capacity cDNA Reverse Transcription Kit (FisherScientific, 43-688-14) by adding 400ng total RNA to the reaction mixture containing random primers following manufacturer guidelines. Real-time PCR and quantitative-Real-time PCR was performed using the SYBR Green PCR Master Mix (ThermoFisher Scientific, 4364346) and the primers described in Table 1. In-house designed primers were purchased from Integrated DNA Technologies and TaqMan primer assays were purchased from ThermoFisher. Host antiviral ISG expression and SeV gene expression was normalized to rat Chmp2a mRNA, and fold change calculated as ddCt over mock infection. Viral RNA copies for SEOV were calculated using a DNA plasmid standard encoding the N gene and the noted primers during qRT-PCR. Results were visualized and statistically analyzed using GraphPad Prism software (Prism 10). Synthetic polyinosinic:polycytidylic acid (poly(I:C)) was purchased from Sigma-Aldrich (MilliporeSigma, P0913-50MG). The hepatitis C virus (HCV) xRNA was in vitro transcribed from synthetic DNA oligonucleotide templates (Integrated DNA Technologies) using the T7 MEGAshortscript kit (Ambion) as previously described (89-91). Following manufacturer protocol, transfections were completed using JetPRIME transfection reagent (VWR, 89129-924). 

Protein Methods

Cell lysates to be analyzed for protein expression by immunoblot analysis were harvested in protein lysis buffer and prepared as previously described (92). Briefly, protein was harvested in RIPA buffer and clarified through 25,000rcf centrifugation for 15 minutes at 4ºC. Protein was quantified via Pierce BCA Protein Assay Kit (ThermoFisher Scientific, 23225). 15μg-30ug of protein was loaded per sample into a 10% polyacrylamide gel and, after denaturing electrophoresis, transferred to a 0.45µm nitrocellulose membrane (VWR, 10120-006). Membranes were blocked at room temperature in 10% FCS in PBS-T. Primary antibodies (table 2) were incubated at 4º overnight. HRP-conjugated secondary antibodies against primary antibody species were incubated for 1 hour at room temperature. Blots were imaged on BioRad Chemidoc MP Imaging System using chemiluminescence Pierce Substrate for Western Blotting (VWR, PI80196).

Table 3. Primary antibodies used for protein detection in immunoblot and immunofluorescence assays.

Densitometry Quantification of Immunoblots Using ImageJ
Densitometry analysis of immunoblots was performed using ImageJ (NIH) to quantify protein band intensities. Immunoblot images were first converted to 8-bit grayscale. The rectangular selection tool was used to draw a box around each band, ensuring consistent box sizes across all bands. The area under the curve (AUC) for each band was measured using the “Analyze → Gels → Plot Lanes” function, which generates intensity profiles of the bands. The peak areas corresponding to each band were selected using the wand tool, and background intensity was subtracted from each lane. The relative band intensities were normalized to the loading control (β-actin) to correct for variations in protein loading. Data from presented experiments were reported as relative protein expression levels compared to β-Actin.

Secreted cytokine analysis

Tissue culture supernatant from HUVEC (Cas9 and RLR^-/-) was collected four days post SEOV infection (MOI 0.01) and clarified for cellular debris by centrifugation (1000 rcf for 10 minutes). Clarified supernatants were UV-inactivated (specific absorption of 103.8 mJ/cm²) and shipped to Eve Technologies for interrogation on the Human Cytokine/Chemokine Panel A 48-Plex Discovery Assay Array (HD48A). One experiment was performed with three biological replicates and two technical assay replicates.

Statistical analyses

Statistical significance was assess using GraphPad Prism software version 10. Where noted in figure legends, ANOVA and student’s T-test analyses were applied, with * representing adjusted p<0.05. All comparisons reaching statistical significance are displayed on graphs.