Virus
The SARS-CoV-2 Delta (B.1.617.2) variant was purchased from the BEI Resources Repository (Ref NR-55612, Batch 70045240, National Instituted of Health, USA) and was stored at −135 °C until administration. The Delta strain SARS-CoV-2, hCoV-19/USA/PHC658/2021 (B.1.617.2) was produced in Calu-3 cells using two passages: titer = 6.45 × 105 TCID50/mL, volume = 1 mL.
Description, evaluation, and radiochemistry of the radiotracer
Production of COVA1-27
The COVA1-27 human IgG1 monoclonal antibody was previously described51 and generated as follows. A suspension of HEK293F cells (Invitrogen, cat no. R79007) was cultured in FreeStyle medium (Gibco) and co-transfected at a density of 0.8–1.2 million cells/mL with a 1:1 ratio of the IgG heavy and light chain plasmids of COVA1-27 together with 1 mg/L PEImax (Polysciences) at a 1:3 ratio. After five days, the COVA1-27 antibody was purified by centrifugation of the cell supernatant for 25 min at 4000 rpm, filtered using 0.22-µm pore size SteriTop filters (Millipore), and run over a 10 mL protein A/G column (Pierce), followed by two column volumes of PBS wash. COVA1-27 was eluted from the column using 0.1 M glycine pH 2.5 into neutralization buffer (1 M TRIS pH 8.7) at a 1:9 ratio. Finally, the buffer was changed to PBS using 100 kDa VivaSpin20 columns (Sartorius). The IgG concentration was determined using a NanoDrop 2000 and the antibodies were stored at 4 °C.
Chemicals and molecules
A nonspecific whole human IgG1 kappa molecule control was purchased from Thermo Fisher Scientific (No. 31154, 11.6 mg/mL, purified in PBS). [89Zr] (185–370 MBq, 1.0 M oxalic acid solution) was produced by BV Cyclotron (The Netherlands) and purchased from Reevity (France).
General method for size-exclusion chromatography (SEC)
SEC was performed using an Alliance e2695 system equipped with a 2489 UV/Vis detector (Waters, USA) and a HERM LB 500 with an fLumo (Berthold, France) gamma detector. The system was operated using Empower® 3 (Waters, USA) software. SEC was performed on a Superdex 200 Increase 10/300GL (Cytivia, USA) column using PBS (0.1 M, 0.5 mL/min) as eluent. UV detection was performed at 280 nm. The chemical identification was carried out by comparing the retention time of the radiolabeled antibody with that of the non-radioactive antibody without chemical modification (tRref). Radiochemical purity was calculated as the ratio of the area under the curve (AUC) of the radiotracer peak over the sum of the AUCs of all other peaks on gamma chromatograms. The radiochemical yield was calculated as the ratio of the activity of the radiolabeled antibody at the end of the synthesis, measured in an ionization chamber (Capintec®, Berthold, France), over the starting activity afforded to the Zr-89. Specific activity was calculated as the ratio of the activity of the collected peak of [89Zr]Ab measured in an ionization chamber (Capintec®, Berthold) over the mass of the antibody determined using a calibration curve.
Antibody functionalization with p-SCN-Bn-DFO
COVA1-27 and non-specific IgG1 kappa antibodies were functionalized with 1-(4-isothiocyanatophenyl)-3-[6,17-dihydroxy-7,10,18,21-tetraoxo-27-(N-acetylhydroxylamino)-6,11,17, 22-tetraazaheptaeicosine] thiourea (p-SCN-Bn-DFO, simplified as DFO in this study). COVA1-27 and non-specific IgG1 kappa antibody (6–10 mg) solutions (in 0.1 M PBS) were concentrated by centrifugation (Vivaspin, 50 kDa) until the final volume was <1 mL. The final concentration was determined using a Nanodrop and was typically between 5 and 10 mg/mL. The volume was adjusted to 1 mL using a 0.9% NaCl solution. The pH was measured using a 780 pH-meter (Metrohm, France) and adjusted to 9.1–9.3 using a 0.1 M Na2CO3 solution. A 10 mM p-SCN-Bn-DFO solution in DMSO (20 µL, 5.0 equiv. of p-SCN-Bn-DFO relative to the antibody) was added stepwise (5 µL steps) to the solution of antibody while shaking. The resulting mixture was stirred at 500 rpm using a Thermoshaker (ThermoFischer Scientific, USA) at 37 °C for 1 h. At the same time, a PD-10 column (GE Healthcare, USA) was rinsed with 20 mL 0.1 M PBS. The reaction mixture was loaded onto the column and the flow-through discarded. An additional 1.5 mL of PBS 0.1 M was loaded and the flow-through again discarded. Finally, 2 mL 0.1 M PBS was added and the eluted solution was collected. The resulting functionalized antibody was analyzed by SEC following the general method.
Radiolabeling of functionalized antibodies with Zr-89
The functionalized COVA1-27 and non-specific IgG1 kappa antibodies were radiolabeled with Zr-89 according to the protocol of Vosjan et al. 63. First 200 µL (150-200 MBq) of the [89Zr]oxalate solution in 1 M oxalic acid buffer was transferred to an Eppendorf tube and Na2CO3 (2 M, 90 µL) was added. The solution was shaken for 3 min and the pH measured and had to be approximately 8. An aqueous solution of HEPES (0.5 M, 710 µL) was added, followed by the addition of 1 mL of the functionalized antibody solution. The resulting mixture was stirred at room temperature for 1 h. At the same time, a PD-10 column (GE Healthcare, USA) was conditioned with 20 mL of a solution of gentisic acid (5 mg/mL) in 0.25 M sodium acetate buffer. The crude reaction mixture was loaded onto the column and the gentisic acid solution used as eluent. The flow-through was collected in 500-µL fractions. The radioactivity of the collected fractions was measured using a Capintec® device (Berthold, France) and the fractions showing the highest activity (typically fractions 4–7) were pooled and analyzed by SEC following the general method.
In vitro assessment of COVA1-27 binding and stability
The binding affinity and stability of the functionalized COVA1-27-DFO antibody to the wildtype (ancestral lineage A) and Delta (lineage B.1.617.2) spike proteins were assessed in vitro using the V-PLEX SARS-CoV-2 Panel 13 (IgG) kit (Meso Scale Discovery, Rockville, USA) according to the manufacturer’s instructions. The assay protocol involved blocking the plates with MSD Blocker A, followed by the addition of reference standards, controls, and serial dilutions of the COVA1-27 and COVA1-27-DFO antibodies: seven threefold dilutions, starting at 1.33 µg/mL, were prepared to assess binding. After incubation, a detection antibody (MSD SULFO-TAGTM Anti-Human IgG Antibody) was added, followed by the addition of MSD GOLDTM Read Buffer B. The plates were subsequently analyzed using a MESO QuickPlex SQ 120MM Reader. The data were processed using Discovery Workbench software and recorded as electrochemiluminescent (ECL) signals. Binding efficiency was quantified as the ratio of the ECL signal of the COVA1-27-DFO to that of the COVA1-27 antibody across the antibody dilutions. The stability assay consisted of incubating the COVA1-27-DFO antibody at three different dilutions (0.05 µg/mL, 0.02 µg/mL, and 0.005 µg/mL) in macaque serum at 37 °C for up to two weeks. The sera were analyzed after 0, 1, 3, 7, and 13 days of incubation. Results are expressed as the percentage of the ratio of the ECL signal (or AU/mL) at each time point relative to the signal at day 0.
Ethics and biosafety statement
Young adult male and female CMs (Macaca fascicularis, 6 males and 7 females, aged 3.19 ± 0.47 years), F1 and F2 generations originating from Mauritian AAALAC-certified breeding centers, were used in this study. All animals were exempt from previous SARS-CoV-2 infection and did not exhibit lung lesions as observed by CT before the beginning of this study.
All CMs were housed at the IDMIT infrastructure facilities (CEA, Fontenay-aux-roses) under BSL-2 and BSL-3 containment, (Animal facility authorization #D92-032-02, Préfecture des Hauts de Seine, France) and in compliance with European Directive 2010/63/EU, French regulations, and the Standards for Humane Care and Use of Laboratory Animals of the Office for Laboratory Animal Welfare (OLAW, assurance number #A5826-01, US). The protocols were approved by the institutional ethical committee ‘Comité d’Ethique en Expérimentation Animale du Commissariat à l’Energie Atomique et aux Energies Alternatives’ (CEtEA number 44) under statement number A20-063. The study was authorized by the “Research, Innovation and Education Ministry” under registration number APAFIS #28752-2020121710032163v1. Animals from the different experimental groups were housed in separated BSL-2/BSL-3 rooms in which several cages modules were positioned to house one single animal per module to avoid viral & radioactive contamination between individuals but to preserve eye & sound contacts between animals.
Animals and study design
Eight CMs (4 males & 4 females) were exposed to a 1 × 105 TCID50 dose of the SARS-CoV-2 Delta variant (lineage B.1.617.2—isolate hCoV-19/USA/PHC658/2021, BEI NR-55612) and three CMs (1 male & 2 female) to PBS. Two additional CMs (1 male & 1 female) were convalescent animals infected by the SARS-CoV-2 Delta variant (same infectious dose) three months before the study. For exposure, animals were premedicated using atropine (0.04 mg.kg−1) and anesthetized with ketamine (5 mg.kg−1) and medetomidine (0.05 mg.kg−1). For all other procedures, the same anesthesia was used without atropine. Both mock and viral exposure were performed using the intranasal (250 µL in each nostril) and intratracheal routes (4.5 mL) at day 0. Eleven macaques (4 males, 7 females) were additionally injected intravenously with either 2.5 mg (n = 8) or 150 µg (n = 3) IgG1 monoclonal human antibody COVA1-27. Two macaques (2 males) were injected intravenously with 2.5 mg human IgG1 whole molecule as a control. Both antibodies were coupled with [89Zr] chelated by deferoxamine (DFO) as described in the previous section. The groups are described in Table 1. The intravenous injection of the radiotracers was performed one day prior to the first imaging session to allow whole-body biodistribution of the molecule.
Animals of the high-dose group, followed for 7 or 14 days (n = 8, cf. Table 1), were also injected subcutaneously with 25 µg (100 µL) of purified spike protein (SARS-CoV-2 wildtype, S_2P-Foldon-His, Mitch 20211122, diluted in DPBS) in the left thigh and 100 µL DPBS (ref 14190-094, Gibco) in the right thigh to evaluate recognition of the spike protein in vivo. The body weight and temperature of the macaques were recorded at each sampling time point.
PET/CT imaging was performed on days 0, 1, 2, 3, 7, 10, and 14 post-exposure (pe), corresponding to the presence of the virus during the acute phase of infection25. Convalescent animals were imaged with the same timing for one week after injection. During acquisition, the macaques were maintained under 0.5–2% isoflurane. CT was used to evaluate the lung lesions and locate the PET signal. Nasopharyngeal, tracheal, and rectal swabs (collected in Viral Transport Medium, 3 mL, CDC, DSR-052-01) were performed for each animal on days 1, 2, 3, 4, 8, 11, and 15 pi to assess viral titers, as well as local radioactivity. Blood was collected (1 mL, EDTA) on days 1, 2, 3, 4, 8, 11, and 15 pi to assess blood radioactivity. All sampling and imaging experiments were also performed one month before exposure to assess baseline levels. Convalescent animals were followed for one week with the same sampling and schedule.
Quantification of radioactivity
The radioactivity was quantified for nasal, tracheal and rectal swabs, as well as whole blood and associated plasma for every sampling time described above. Briefly, samples (3 mL for swabs in VTM, 700 µL whole blood and associated plasma) were loaded into a gamma counter (Hidex Automatic Gamma Counter, Finland), weighted, and counted for 1 min (energy window 480–558 keV). Of note, plasma was obtained from the supernatants of EDTA-treated blood samples after centrifugation (1800 × g, 10 min). All specimens were stored at −80 °C for one month after the injection date to allow radioactive decay before establishing the viral load.
Quantification of the virus in fluids by RT-qPCR
Genomic RNA (gRNA) and subgenomic RNA (sgRNA) from nasal and tracheal swabs (diluted in VTM) were extracted using the NucleoSpin™ 95 virus core kit (Macherey-Nagel) according to the manufacturer’s protocol. Reverse transcription (RT) and quantitative polymerase chain reaction (qPCR) of gRNA and sgRNA were performed to assess the viral titers using the SuperScript III platinium one-step quantitative RT-qPCR System and RNase out (Life Technologies, Invitrogen) and CFX96 Touch Real-Time PCR Detection System (Bio-Rad). The following primers and probes were used: primers IP4 gene (SARS-CoV-2) F GGTAACTGGTATGATTTCG, R CTGGTCAAGGTTAATATAGG and IP4 gene probe FAM-TCATACAAACCACGCCAGG-BHQ1 for gRNA, primers E gene (SARS-CoV-2) F CGATCTCTTGTAGATCTGTTCTC, R ATATTGCAGCAGTACGCACACA and E gene probe HEX-ACACTAGCCATCCTTACTGCGCTTCG-BHQ1 for sgRNA (Eurofins Genomics). The estimated lower limit of detection (LoD) was 5.79 × 102 copies/mL and 7.49 × 102 copies/mL of SARS-CoV-2 gRNA and sgRNA respectively. The estimated lower limit of quantification (LoQ) was 5.79 × 103 copies/mL and 7.49 × 103 copies/mL of SARS-CoV-2 gRNA and sgRNA, respectively.
Quantification of the virus in tissues by RT-qPCR
All tissue samples were collected at euthanasia and stored dry in cryotubes at −80 °C after gamma counting. Tissue fragments were obtained from the lungs, trachea, heart, spleen, kidney, liver, and axillary, inguinal, and tracheobronchial lymph nodes. A fragment of 50 mg of frozen sample was lysed in NucleoZOL (Macherey-Nagel, Duren, Germany) using a Precellys Evolution/Cryolys Device (Bertin Technology, Montigny-le-Bretonneux, France). RNA isolation was performed using a NucleoSpin™ RNA Core Kit (Macherey-Nagel) according to the manufacturer’s instructions. Reverse transcription and qPCR were performed using the SuperScript III Platinium One-step Quantitative RT-qPCR System (Life Technologies). The protocol used to quantify SARS-CoV-2 IP4 genomic mRNA (gRNA) was as previously described for liquid samples, with an estimated lower LoD of 2.24 × 102 copies/mL of SARS-CoV-2 gRNA and an estimated lower LoQ of 2.24 × 103 copies/mL.
CT and PET-CT acquisition
All image acquisition was performed using the same clinical Digital Photon Counting PET-CT system (Vereos-Ingenuity, Philips) implemented in a BSL-3 laboratory64.
All sessions were performed using the same experimental conditions (acquisition time and animal order) to limit experimental bias. Animals were anesthetized with ketamine and medetomidine as already described, intubated, and then, maintained under 0.5–2% isoflurane. They were placed in a supine position on a warming blanket (Bear Hugger, 3 M) on the machine bed, with monitoring of the cardiac rate, oxygen saturation, and temperature.
For PET-CT acquisitions, CT was performed twice, once under breath-hold (for CT anatomical segmentation) and again prior to PET acquisition for attenuation correction and anatomical localization.
All technical parameters, detector collimation (64 × 0.6 mm), tube voltage (120 kV), and intensity (150 mA) were identical for all acquisitions. Chest-CT images were reconstructed with a slice thickness of 1.25 mm and an interval of 0.625 mm. Whole-body CT images were reconstructed with a slice thickness of 1.5 mm and an interval of 0.75 mm.
A whole-body PET scan (3–4 bed positions, 3 min/bed position) was performed 24 h and on days 1, 2, 3, 4, 8, 11, and 15 post-injection of 150 µg/animal (low dose) or 2.5 mg/animal (high dose) of [89Zr]COVA1-27-DFO or [89Zr]IgG1-DFO via the saphenous vein. PET images were reconstructed onto a 256 × 256 matrix using OSEM (3 iterations, 15 subsets).
Image analysis
Image analysis was carried out on whole organs: brain, nasal cavity, lungs, trachea, heart, liver, spleen, kidneys, tracheobronchial, axillary, and inguinal lymph nodes using the free software 3Dslicer (v.5.0.3.). Standard uptake deviation (SUV) takes into account the radioactive decay of [89Zr]. It was used to analyze the PET signal in the abovementioned compartments. The regions of interest (ROI), brain, nasal cavity, lungs, trachea, heart, spleen, liver, and kidneys, were defined using the CT signal and corrected using the PET signal, when necessary, using the same threshold/time point. The following 3D slicer functions were used to generate the segmentations: chest imaging platform, fast marching, paint, draw, and fill between slices. All segmentations were normalized against the PET signal of the segmentation at D1 post-tracer injection. On the contrary, subcutaneous injection sites and lymph nodes were defined using the PET signal with the help of the CT signal for anatomical localization. For all SARS-CoV-2 exposed animals, lesional areas in the lungs were identified using the CT signal; 3 to 4 identical spheres per lesion (volume = 0.07 ± 0.01 mL) were drawn and compared to the same ROI at the same location in the lungs of control NHPs. Similarly, the same spheres were drawn in CT-lesion-exempt areas of the lungs of all animals for the analysis of non-lesional areas of the lung. Pulmonary lesions were defined as ground-glass opacity, crazy-paving pattern, or consolidations, as previously described65. Two to three individuals assessed the lesion features (type (scored from 0 to 3) and extension (scored from 0 to 3)) independently for each lung lobe and the final CT score results (sum of each lung lobe score) were determined by consensus as described previously47. Pre-existing background lesions were scored 0.
Statistical analysis
Data are presented as individual values with the mean and standard deviation. Linear regression parameters and Mann–Whitney (p < 0.05) unpaired two-sided non-parametric t-test results (without adjusting for multiple testing) were calculated using GraphPad® Prism 8 software. Main individual data were also reported in Supplementary table 2.
Euthanasia at the acute phase of the infection
Two CMs (CM9 & CM10) were injected with 2.5 mg [89Zr]COVA1-27-DFO and exposed to SARS-CoV-2 24 h after tracer injection. The radioactivity of the specimens from nasal and tracheal swabs, as well as blood and plasma, were evaluated using the same gamma counter (Hidex Automatic Gamma Counter, Finland). The biodistribution of the radiotracer was followed by PET/CT imaging in the nasal cavity, lungs, trachea, heart, liver, spleen, kidneys, inguinal, axillary, and tracheobronchial lymph nodes at 0, 1, 2, and 3 dpe. At 3 dpe, the animals were anesthetized and euthanized by intravenous injection pentobarbital (180 mg.kg−1). Organs of interest were placed on the PET/CT bench and further imaged to more precisely locate the signal. The PET signal was then compared to RT-qPCR of the tissues in the ROI and gamma counts of the same tissues. The organs were fixed in 10% formalin for 24 h at room temperature and then transferred to PBS at 4 °C before being analyzed by in situ hybridization (RNAscope®) after radioactive decay.
In situ hybridization of SARS-CoV-2 RNA in tissues
After 24 h of formalin fixation by immersion and PBS transfer, organs were included (Excelsior ES, Thermo Scientific) and embedded in paraffin. Four micron slices for each organ of interest were generated using a microtome (Leica RM2255) and mounted on SuperFrost Plus™ slides (ThermoFisher Scientific). In-situ hybridization (RNAscope®) experiments were performed using a Ventana® Ultra medical system (Roche Diagnostics) with the following probes: Probe V-nCoV2019-S-C1, S gene encoding the spike protein (Catalog No. 848561), positive control UBC (Catalog No. 310041), and negative control bacterial gene diaminopimelate (DapB) (Catalog No. 310043), as described in the literature53. The slides were counterstained using 0.02% ammonia water in a Leica st5020 coloring tray and mounted using Eukitt® mounting medium.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
