Ethics

All in-life procedures were performed at the Wageningen Bioveterinary Research Institute (WBVR, Lelystad, The Netherlands). The Institute is holder of the authorization (ref. letter TRC/NVWA/2016/10456) to perform animal experiments according to the Dutch Law on Animal Experiments (WoD) and in accordance to European legislations and guidelines (2010/63/EG and ETS 123). The study described here was performed under legislation (license no. AVD401002015194) of the Dutch Central Authority for Scientific procedures on Animals (CCD) with approval of the experimental plans (2016.D-0067.007) by the Animal Welfare Body of Wageningen University and Research.

Calves were under veterinary supervision from the day of arrival and any animal exhibiting evidence of disease, distress or discomfort was examined by a veterinarian and appropriate treatment was applied when deemed necessary. To limit severe discomfort after bovine respiratory syncytial virus (bRSV) challenge, humane endpoints were defined prior to study start as follows: persistent severe depression e.g. inability to stand on its own (max. 24 h) or anorexia (max. 48 h); persistent severe dyspnea e.g. continuous respiratory rate of >100 breathings/minute (max. 48 h) or heavily forced breathing with open mouth and/or frothing (immediately).

Permission for working with the genetically modified vaccine Ad26.RSV.preF was obtained by the Dutch GMO permit with code IG 18-093. In this permit, Ad26.RSV.preF is classified as GMO level II organism. Because of the risk of aerosol transmission, vaccination was performed in GMO level III facilities (DM-III). Preparation of the vaccine and processing of samples of vaccinated animals were performed in GMO level II laboratories (ML-II).

Animals & Husbandry

Newborn dairy (Holstein) or crossbreed calves (n = 10 male & n = 7 female) were obtained from local, commercial dairy farms from herds of origin that were free of bovine viral diarrhea virus (BVDV). Calves were transported to WBVR within 2 h after birth to prevent an unintended infection with bRSV or other endemic respiratory pathogens. Animals did not receive colostrum on the farm of origin but were administered a single feeding of preselected pasteurized bovine colostrum, tested free of bRSV antibodies by commercial ELISA (Svanovir® BRSV Ab ELISA, Svanova) and by in-house virus neutralization assay, upon arrival at WBVR. Colostrum was followed by commercial milk replacer, provided three times a day during the first 2–3 weeks of life and twice daily until weaning at 8–9 weeks of age ( ~ 3 weeks post first immunization). Starting from 1 to 2 weeks of age, calves additionally received roughage (pelleted and dried grass) and concentrates and potable water was supplied. During the first 5 weeks, calves were housed in individual pens (2.2 m2 per calf) bedded with straw. From 5 weeks of age, calves were group-housed with n = 8 or 9 calves of the same treatment in one pen on concrete floor partly covered rubber cow matrasses with a floor space of >3.40 m2 per calf. Pens were separated so that there was no physical contact between treatment groups.

Vaccines & vaccine administration

Ad26.RSV.preF (Janssen Vaccines and Prevention (JVP), Leiden, the Netherlands) is a recombinant, replication-incompetent, serotype 26 adenovirus (Ad26) containing a DNA transgene that expresses the full-length conformation-stabilized preF of the RSV A2 strain. The RSV F protein has been stabilized in the preF conformation using unique animo acid substitutions focused on the stabilization of helix α4 in the disorder apex and in the area of charge repulsion in the base45, which has been demonstrated to increase the stability of RSV F protein in the preF conformation when expressed on the cell surface29. The vector has been used in various preclinical and clinical studies and is described in those publications27,28,29.

Calves were randomized to two treatment groups of n = 8 (Ad26.RSV.preF group) and n = 9 calves (placebo group), considering a balanced distribution of sex, breed, age and clinical history. The veterinarian performing the clinical scoring post-challenge and the pathologist involved in the post-mortem examinations were blinded to the treatment groups.

At 6 weeks of age (42 ± 2 days), animals were immunized intramuscularly with 5 × 1010 viral particles (vp) Ad26.RSV.preF in a two-dose regimen, with a 4-week interval. A control group received 2 placebo immunizations with formulation buffer. Vaccines were administered in the left (first vaccination) and right (second vaccination) pre-scapular area, with an injection volume of 2 mL. One animal of the placebo group (#2481, indicated in light blue symbols in the figures) did not receive any placebo immunizations due to transient fever at the time of immunization.

bRSV challenge material and administration

bRSV challenge was performed with in vivo passaged bRSV (strain Odijk, subtype A, passage 6), in BAL material obtained from caesarian-derived specific-pathogen-free calves at a dose of 6.3 × 103 TCID50 per calf as determined by back titration of challenge inoculum. At 56 days after first immunization animals were inoculated by aerosol with 2 mL bRSV via an airjet nebulizer (Harder & Steenbeck)18, administering the inoculum to both nostrils. One animal of the placebo group (#2476, indicated with grey symbols in Fig. 1) did not receive bRSV challenge due to insufficient recovery of lameness, requiring the need for continued administration of pain medication, which could potentially interfere with study outcome. As a consequence, this animal was withdrawn and not available in all post-challenge data analysis and visualizations.

After the observation of an unusual severe outcome of the experimental bRSV challenge, BAL samples collected 5 days prior and post bRSV challenge were tested for a viral, bacterial or mycoplasma respiratory co-infection by testing on Mycoplasmata (M. bovis, M. dispar and M. bovirhinis) and Pasteurella multocida, Histophilus somni, Mannheimia heamolytica and Trueperella pyogenes by an in-house RT-PCR. In addition, serum samples collected on Day 28 and upon completion of the study were tested for antibodies against Bovine Viral Diarrhoea Virus, Bovine Herpes Virus type 1, Bovine Parainfluenza Virus type 3, Mycoplasma Bovis, Bovine Adenovirus type 3 and Bovine Coronavirus by commercial ELISA (BIO-X).

Sampling & processing

Serology and heparinized blood samples, bronchoalveolar lavages and nasopharyngeal brushes were collected at pre-defined timepoints after vaccination and challenge (See Fig. 2A for collection events).

Blood samples for serology were collected in vacutainer tubes and allowed to clot at room temperature and then centrifuged at 1250 g for 10 min at room temperature. Serum was harvested and stored at −20 °C prior to analysis by ELISA and VNA. Heparinized blood samples were kept at ambient temperature and shipped directly to Janssen Vaccines & Prevention with same day delivery, where they were stored overnight at room temperature. PBMCs were isolated the following day by density separation in Leucosep® tubes (Greiner Bio-One) with prefilled separation medium. PBMCs were resuspended in 10%FCS/RPMI, and cell count and viability were determined by Vi-Cell cell counter (Beckman Coulter). PBMCs were immediately used in an ELISpot assay.

Nasopharyngeal brush samples were collected from alternate nostrils using sterile nylon bristle brushes (MW126, Medical Wire and Equipment Co. Ltd). Following sampling, the brushes were directly agitated in approximately 3 mL of EMEM supplemented with 2% antibiotics and 2% fetal bovine serum and kept on melting ice. Subsequently, the brushes were removed from the medium and samples were centrifuged at 1300 g for 10 min at 4 °C. The supernatants were stored at −70 °C before determination of bRSV viral load by qPCR and TCID50 assays. See Fig. 2A for collection events.

Bronchoalveolar lavages were performed in non-sedated animals according to the method described by Fogarty et al. Briefly, BAL fluid was obtained after instillation of approximately 100 mL D-Phosphate Buffered Saline (Gibco) via intubation of the ventral nasal meatus. The fluid recovered from the lung (50–80%) was directly supplemented with 5% fetal bovine serum and kept on melting ice until processing as follows: a small volume (approximately 1 mL) of the BAL samples was inactivated by incubation with 0.03 M formaldehyde for 1 h at 37 °C to allow processing outside the ML-II laboratory for the purpose of total and differential cell count. The remaining BAL sample was centrifuged at 250Gg for 10 min at 4 °C and its supernatant was stored at −70 °C before determination of bRSV viral load by qPCR and TCID50 assays. See Fig. 2A for collection events.

Clinical Observations & Scoring

Clinical Scoring was performed once daily by one and the same examining veterinarian from two days prior to challenge virus inoculation until and including twelve days post challenge virus inoculation. On the day of challenge virus inoculation, observations were performed prior to inoculation. In short, signs of general illness, upper respiratory tract disease (URTD) and lower respiratory tract disease (LRTD) were scored per animal on a scale from 0 to 3 (Supplemental Table 1). In addition, the respiratory rate in breaths/minute, as part of the LRTD score, was counted by the veterinarian and analyzed separately. Rectal temperatures were measured once daily during morning feeding sessions. Body weight of the calves was measured on two days prior to challenge virus inoculation and on day 2, 5, 7, 9 and 12 post challenge. Oxygen saturation levels were measured one day prior to challenge virus inoculation and on day 1, 5 to 9 and 12 post challenge. Measurement was performed non-invasively by pulse oximetry using a telemetric system (TMS International) consisting of a Nonin 8000 J flexwrap sensor attached to the tail (with skin shaved) and connected to a Mobi-6-4b2 recording device enabling a wireless visualization and transcription of the data on a laptop (Polybench® software).

ELISA

hRSV prefusion F binding antibodies in bovine serum were determined by ELISA. In brief, white 96-well plates were coated with Streptavidin and incubated at 4 °C overnight. After washing, the wells were blocked with bovine serum albumin (BSA) for 30 min at RT, washed again, followed by addition of biotinylated hRSV preF protein and incubation for 1 h at RT. After washing, serially diluted serum was added and incubated for 1 h at RT. hRSV preF binding antibodies were detected by HRP-labeled anti-bovine IgG (clone MT391, Mabtech, incubated 1 h at RT) and after washing the wells were developed with LumiGlo substrate. The luminescence signal was measured with a Synergy Neo plate reader. Endpoint titers were determined using a threshold of 2 times the highest signal of the naïve serum samples. The LOD was defined as the log10 of one dilution step below the first serum dilution tested (1.08 log10).

RSV F IFNy ELISpot

ELISpot assays were performed on freshly isolated PBMC using the bovine MabTech kit (Catalog number 3119-4APW), according to the manufacturer’s recommendations, using 500,000 PBMC/well. Antigen-specific stimulation was performed in triplicate, using a peptide pool representing the complete F protein from human RSV A2 protein (final concentration 2 µg/ml). Controls included unstimulated cells (duplicate), or cells stimulated with PMA/ionomycin (final concentrations 50 µg/ml, and 2 µg/ml, in duplicate). Data presented represent spot forming units (SFU) per 106 RSV F stimulated PBMC, subtracted by SFU of the unstimulated cells. Data was excluded from graphing and statistical analysis if background was too high ( >100 SFU/106 PBMC in both duplicate measurement of unstimulated cells), variation between the triplicates was too high (square root of SD > 2.5), or stimulation with PMA/ionomycin was too low ( < 300 SFU/106 PBMC). For statistical comparisons, a LOD of 50 SFU/106 PBMC was used.

Virus neutralization assays

Serum samples collected during the study were analyzed for bRSV and hRSV neutralizing antibodies.

hRSV neutralizing antibody titers in bovine serum were determined by an automated microneutralization assay using Fire Fly Luciferase (FFL) labeled RSV CL57V224. In brief, heat inactivated, serial diluted sera were mixed with 2.5 × 104 plaque-forming units of RSV CL57V224 FFL in ½ area white tissue culture plates and incubated for one hour at RT. Subsequently 5 × 103 A549 cells (Multiplicity of Infection (MOI) = 5) per well were added and plates were incubated for 20 h at 37 °C, 10% CO2. After the 20 h Neolite substrate (PerkinElmer, Waltham, MA, USA) was added. The luminescence signal was determined with an EnVision plate reader (PerkinElmer). VNA titers were calculated as the antibody concentration that caused a 90% reduction in luminescence, expressed as IC90 titers. For statistical comparisons, a LOD was defined based on historical data (5.28). All samples with a value lower than this LOD were set at 5.28 for visualization purposes.

To measure bRSV neutralizing antibodies, sera were serially diluted and incubated with 30–300 TCID50 bRSV (strain RB94), after which a cell suspension of Embryonic Bovine Tracheal cells was added. Plates were incubated for 7 days, after which infected cells were stained in an Immuno Peroxidase Monolayer Assay (IPMA). For each sample dilution, duplicate monolayers were read by microscopic evaluation for the absence (IPMA score -) or presence (IPMA score +) of bRSV-positive cells. bRSV was detected by using mouse monoclonal antibody MAB3 directed against the F-protein of bRSV (produced at WBVR)47 followed by incubation with peroxidase-conjugated rabbit anti-mouse Ig (Dako, catalog number P0260), and AEC as substrate (Sigma, catalog number A6926). The virus neutralization titer was expressed as the log2 value of the highest serum dilution that showed complete neutralization (IPMA score -) in 50% of the inoculated wells. Samples below the limit of detection (log2 IC50 = 2) were set at 1 log beyond the detection limit (log2 IC50 = 1).

bRSV qPCR

All nasopharyngeal brush and BAL samples were tested for the presence of bRSV by qPCR. RNA isolations were done on 200 µL sample by automated easyMAG® isolation. Isolated RNA was eluted in a volume of 50 µL. bRSV primers (forward 5′-GCA-ATG-CTG-CAG-GAC-TAG-GTA-TAA-T-3′ and reverse 5′-ACA-CTG-TAA-TTG-ATG-ACC-CCATTC-T-3′) and a Taqman probe (5′-FAM-ACC-AAG-ACT-TGT-ATG-ATG-CTGCCA-AAG-CA-TAMRA-3′) for one-step real-time PCR (QuantiFast Multiplex RT-PCR + R Kit, Qiagen) were based on nucleocapsid gene, and detect both replicating and non-replicating virus. Each PCR run (Applied Biosystems 7500 RT PCR system) on 5 µL RNA eluate included a positive, weak positive and negative control sample and RNA samples of 3-fold serial dilution of a bRSV stock (RB94) with a predefined infectious virus titer, serving as a reference standard and used to transform Ct values into PCR equivalent virus titers.

bRSV Titration Assay (TCID50)

Based on the results of the qPCR, a selection of the nasopharyngeal brush samples and all BAL samples were titrated in a virus titration assay to determine levels of infectious virus. Briefly, four replicates (50 µL sample) of ten-fold dilutions were added to a suspension of Embryonic Bovine Tracheal cells (WBVR origin) on duplicate 96-wells plates. Duplicate plates were incubated for 7 days at 37 °C and 5% CO2. An IPMA was performed to detect virus infected cells by means of an anti-F bRSV monoclonal antibody (MAB3, produced at WBVR)47 followed by incubation with peroxidase-conjugated rabbit anti-mouse Ig (Dako, catalog number P0260) and AEC as substrate (Sigma, catalog number A6926). Virus titers were expressed as endpoint titers at which 50% cell cultures were infected, estimated according to Reed and Munch, resulting in a detection limit of 1.3 log10 TCID50. Samples that were below the limit of detection were set at 0.3 for visualization purposes. Samples that were at the limit of detection on plate 1 but below on plate 2 were set at 0.8.

Post-mortem pathology assessment

A post-mortem examination was performed whenever an animal reached a humane endpoint, succumbed to bRSV infection, or at study end 13 days post-challenge. For euthanasia, calves were administered an overdose of pentobarbital injected intravenously followed by exsanguination. Macroscopic findings were summarized and lungs were removed for processing.

Dorsal and ventral photographs of the lungs were taken to determine the extent of macroscopic lesions by calculating the percentage of consolidated lung tissue area over the total lung surface area using computer software (Image Pro Premier 64-bit). Lung tissue samples of predetermined sites were collected and stored in 10% neutral buffered formalin. Formalin-fixed samples of 3 sampled sites, identical between all animals, were embedded in paraffin, sectioned, and stained with hematoxylin and eosin for histological examination. Each tissue slide was scored for the presence of endo-bronchi(oli)tis, peri-bronchitis/vasculitis, interstitial cell infiltration and alveolitis, with scores ranging from 0 (absent) to 4 (severe), according to Supplemental Table 2.

Data analysis & statistics

Immunogenicity (humoral responses (VNA and ELISA), and log10 transformed ELISpot responses) measured during the pre-challenge period, were compared between the placebo and Ad26.RSV.preF group using a Tobit regression model with a likelihood ratio test (LRT). To study the effect of the second immunization, the immunogenicity readouts were compared post-prime (Day 28) and post-boost immunization (Day 42) using a Tobit model. The anamnestic response was statistically analysed by comparing the immunogenicity readouts measured pre-challenge at day 0, and 12 days post-challenge using a Tobit model. The survival outcome was compared using a log-rank test, whereas scores for general illness, URTD and LRTD were compared using a cumulative logistic regression model. The Area Under the Curve (AUC) of the continuous parameters body temperature, body weight, respiratory rate, oxygen saturation, as well as the viral load as determined by virus titration (TCID50) and qRT-PCR, were compared between the two groups using an ANOVA model. Missing values due to pre-termination of an animal were imputed using last observation carried forward. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Inc., Cary, NC, USA).