Production of soluble human and mouse FcRn

Soluble truncated human FcRn and mouse FcRn were produced using a baculovirus expression vector system^70,71,72. Briefly, semi-adherent High Five cells (Cat. B85502, Invitrogen) were grown until confluency in flat culture flasks with Express Five SFM Medium (Cat. 10486, Gibco) at 27 °C before suspension and adjustment to 1×10⁶ cells/mL and transfer to Erlenmeyer flasks, which were then placed in an orbital shaker kept at 27 °C and 160 rpm. Next, the cells were transfected using a viral stock of baculovirus containing a vector encoding a soluble, truncated form of either mouse or human FcRn^70,72. The viral stocks were kindly gifted by Prof. Sally Ward (University of Southampton, UK). Cells were kept under orbital shaking for 72 h, and at 23-24 °C if transfected for hFcRn or 27–28 °C if transfected for mFcRn. After 72 h, supernatant was harvested and filtrated though a 0.2 µm filter before affinity chromatography on HisTrap, as described below.

Vector design and production of HA and HA fusion proteins

The cDNA fragment encoding aa 18-541 of HA from influenza A H1N1 (A/Puerto Rico/8/1934; PR8) was used to design mouse serum albumin (MSA) with HA fused to its C-terminal end. Briefly, HA with an upstream cDNA sequence encoding a GGSGGSGGSGGSGG-linker was cloned into a naked pcDNA3 (Cat. V79020, Invitrogen) vector (pcDNA3-HA), using the restriction sites XhoI and ApaI. A silence mutation of the HA-encoding cDNA fragment was done to remove an additional XhoI restriction site before subcloning. Next, the cDNA fragment encoding MSA (wild-type or K500A/H510Q (AQ)) was cloned on the restriction sites HindIII and XhoI (pcDNA3-MSA-HA) for the production of MSA-HA or AQ-HA. Two HindIII restriction sites were removed in the MSA encoding cDNA sequence by silence mutation to enable subcloning of fragments encoding MSA. A pLNOH2-vector with the cDNA for HA with a C-terminal 6xHis-tag was used for the production of His-tagged HA. The vectors were transiently transfected into adherent HEK293E cells (Cat. CRL-1573, ATCC)⁷³. Briefly, cells were kept at 37 °C and 5% CO₂ and transfected using Lipofectamine 2000 Transfection reagent (Cat. 11668500, Invitrogen) according to the manufacturer’s protocol. Supernatant was harvested every other day for two weeks.

The HA-Tf and HA-QMP fusions were made by synthesizing and cloning (GenScript) the cDNA encoding 18-519 of HA from influenza A H1N1 (A/Puerto Rico/8/1934; PR8) followed by either full-length mouse Tf and a 6xHis-tag or full-length human serum albumin (HSA) containing the amino acid substitutions E505Q/T527M/K573P (QMP)³³ into pFUSE2ss-CLIg-hk (Cat. pfuse2ss-hclk, InvivoGen) using the restriction sites EcoRI and NheI. The resulting vectors were transiently transfected into Expi293F cells (Cat. A14527, Gibco) using Expifectamine 293 Transfection Kit (Cat. A14525, Gibco) according to the manufacturer’s instructions.

Vector design and production of RBD and RBD fusion proteins

cDNA encoding 6xHis-tagged RBD of ancestral SARS-CoV-2 (Wuhan) was expressed in the plasmid pCAGGS⁷⁴, whereas cDNA encoding 10xHis- and Avi-tagged RBD were synthesized and cloned (GenScript) into pFUSE2ss-CLIg-hk using the restriction sites EcoRI and NheI. Mutagenesis was performed in the vector of 6xHis-RBD to generate a panel of SARS-CoV-2 variants (GenScript). To generate RBD fusion proteins, the RBD (Wuhan) was N-terminally fused to MSA, HSA (WT or QMP) or Tf. RBD fused to Tf was designed with a C-terminal 6xHis-tag. The vectors were transiently transfected into Expi293F cells using the ExpiFectamine 293 Transfection Kit according to the manufacturer’s instructions. 6xHis-tagged RBD was used for in vitro applications, whereas 10xHis- and Avi-tagged RBD was used for in vivo vaccination.

Protein purification and verification

Purification of MSA-HA variants and His-tagged HA was conducted using a CNBr-activated Sepharose 4 Fast Flow (Cat. C5338, Invitrogen) coupled to anti-HA mAb H36-4-52⁷⁵, packed in a 5 mL column (Repligen) coupled to a BioLogic workstation and recorder (BIO-RAD). Briefly, approximately 10 column volumes (CV) of 1x Phosphate Buffered Saline (PBS) (Cat. D8537, Sigma-Aldrich)/0.05% sodium azide (Cat. S2002, Sigma-Aldrich) (pH 7.2) was used to pre-equilibrate the column before supernatant (sterile filtrated with a 0.22 µm vacuum filter (Corning) containing 0.05% sodium azide) was applied with a flow rate of 1-2 mL/min. Subsequently, equilibration was done using 10-30 CV of 1xPBS, and bound MSA-HA variants or His-tagged HA were eluted with 5-10 CV of 0.1 M glycine-HCl (Cat. G7126, Sigma-Aldrich) (pH 2.7). Similarly, the His-tagged FcRn receptors, RBD-variants and transferrin fusions were purified using HisTrap HP 5 mL columns (Cat. 17524801, Cytiva). The column was pre-equilibrated with 1xPBS, and for equilibration 1xPBS was followed by 25 mM imidazole (Cat. 56748 Sigma-Aldrich) in PBS (pH 7.2) was used for equilibration, before elution with 250 mM imidazole in 1xPBS (pH 7.3). HSA fusions were purified using columns packed with CaptureSelect Human Albumin Affinity Matrix (Cat. 1912970, Thermo Scientific) by Repligen and were pre-equilibrated and equilibrated with 1xPBS before elution with buffer consisting of 2 M MgCl₂ (Cat. 63064, Sigma-Aldrich) and 20 mM Tris (Cat. T6066, Sigma-Aldrich) (pH 7.0). The RBD-MSA-fusions were purified using columns packed with AlbuPure selective affinity chromatography adsorbent (Cat. 3151, Astrea), pre-equilibrated and washed with 50 mM sodium phosphate (Cat. 342483, Sigma-Aldrich) (pH 6-8) and eluted with buffer containing 50 mM ammonium acetate (Cat. 09689, Sigma-Aldrich) and 20 mM sodium octanoate (Cat. C5038, Sigma-Aldrich) (pH 7.0). Eluted fractions were collected, up-concentrated and buffer-changed to 1xPBS using Amicon Ultra Centrifugal Filter Units (Merck Millipore) with appropriate cutoff (10-100 K) and volume (0.5 mL, 4 mL or 15 mL). To ensure monomeric fractions of purified proteins, size exclusion chromatography was performed using a Superdex 200 Increase 10/300 GL Column (Cat. 28-9909-44, Cytiva) coupled to an ÄKTA Avant 150 Chromatography System (Cytiva) run using Unicorn Software (Cytiva) prior to up-concentration by Amicon Ultra Centrifugal Filter Units.

Protein concentrations were determined using a DS-11+ (M/C) Spectrophotometer (DeNovix) programmed with the protein’s extinction coefficients and molecular weights, while protein purity was assessed by non-reduced SDS-PAGE on a Bolt 12% Bis-Tris Plus Gel (Cat. NW00125, Invitrogen) and compared to Spectra Multicolor Broad Range Protein Ladder (Cat. 26623, Thermo Scientific). Briefly, 2 µg of the protein samples were prepared in 4X Bolt LDS Sample Buffer (Cat. B0007, Invitrogen) and Milli-Q water, according to the manufacturer’s instructions. The gel was run in 1X Bolt MES SDS Running Buffer (Cat. B0002, Invitrogen) for 22 min at 200 V with a PowerPac HC power supply (Bio-Rad) and stained with Bio-Safe Coomassie Stain (Cat. 1610786, Bio-Rad).

Direct conjugation of adjuvant to albumin fusion

A purified fraction of RBD-QMP was directly conjugated to the adjuvant. First, CpG-linkers were made. In brief, 5’ amine-modified CpG 1826 (Ordered from Integrated DNA Technologies (/5AmMC6/TCCATGACGTTCCTGACGTT), all PS backbone) was conjugated to an NHS-ester-modified linker by mixing DMSO (Cat. 34869, Sigma-Aldrich), 1 mM CpG 1826, 0.1 M 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) (Cat. H4034, Sigma-Aldrich), and 100 mM Maleimide-(PEG)8-NHS-ester (SM(PEG)8) linker (Cat. 746207, Sigma-Aldrich) at a volume ratio of 3:2:10:1, respectively. After overnight incubation at room temperature (RT) with 650 rpm orbital shaking, excess linker was removed by ethanol precipitation, which entailed mixing CpG, absolute ethanol (EtOH) (Cat. 20821.310, VWR), and 3 M sodium acetate (Cat. S2889, Sigma-Aldrich) in a 5:31:4 volume ratio and incubate at −18 °C for 3–4 h followed by centrifugation at 17,000 x g for 45 min. The pellet was then washed with EtOH and centrifuged again at 17,000 x g for 30 min. The supernatant was aspirated, and the pellet was dried for 10 min and subsequently dissolved in nuclease-free water (Cat. AM9937, Invitrogen).

Subsequently, the CpG-linkers were conjugated to RBD-QMP. RBD-QMP was mixed with a 1.3 molar excess of CpG-maleimide in 0.1 M HEPES (pH 7) and incubated overnight at RT with 650 rpm orbital shaking.

The CpG-RBD-QMP was purified with HPLC using IEX chromatography with a Mono Q 5/50 GL anion exchange column (Cytiva) running on Chromeleon (Thermo Scientific), and included an UltiMate AFC-3000 automated fraction collector connected to an LPG-3400RS pump and a VWD-3400RS detector (Thermo Scientific). The sample was filtered through a 0.2 μm polypropylene filter (Kinesis) before injection in the HPLC, and purification was run at 0.3 mL/min using the following program: 5 min buffer A (20 mM Tris and 10 mM NaCl (pH 7.6)), a 70 min gradient to buffer B (20 mM Tris and 3 M NaCl (pH 7.6)), 5 min of buffer B, a 5 min gradient to buffer A and 5 min of buffer A. Absorbance was measured at 224 and 260 nm. The collected fractions (min 29 to 35) were pooled, desalted and up-concentrated using 10 kDa cut-off Amicon Ultra Centrifugal Filter Units.

Protein purity after direct conjugation was assessed using native PAGE. A 10% native PAGE gel was made by mixing 3.4 mL ProtoGel (30%) (Cat. EC-890, National Diagnostics), 5.6 mL Milli-Q water, 1 mL 10X UltraPure TBE Buffer (Cat. 15581-044, Invitrogen), 100 μL 10% Ammonium Persulfate (Cat. 17874, Thermo Scientific), and 10 μL TEMED (Cat. 17919, Thermo Scientific). The mixture was poured into 1 mm cassettes and let to polymerize for 45 min with a 10-well comb (Invitrogen). 10 pmol sample was mixed with 10% glycerol (Cat. 24388.295, VWR) and 1 g/L Orange G (Cat. O3756, Sigma-Aldrich), loaded on a gel, and run in 1X UltraPure TBE buffer at 150 V for ca 75 min at RT with an EPS 601 electrophoresis power supply (Amersham Biosciences). The SYBR Gold (Cat. S11494, Invitrogen) staining was performed by submerging the gel in 1X SYBR Gold in Milli-Q water for 15 min. After SYBR Gold image acquisition, the gel was stained with Coomassie by submerging the gel in 0.3 g/L Coomassie Brilliant Blue R (Cat. B7920, Sigma-Aldrich), 4.5% methanol (Cat. 34860, Sigma-Aldrich) and 1% acetic acid (Cat. 1.01830, Supelco) for 15 min with orbital shaking. The gel destaining was performed in 10% methanol and 7.5% acetic acid overnight. Imaging of SYBR Gold and Coomassie stained gels was performed using Image Lab (Bio-Rad) on a Gel Doc EZ Imager (Bio-Rad).

Human endothelial cell-based recycling assay (HERA)

HERA was performed on RBD-HSA fusions according to a published protocol^76,77. Briefly, the human microvascular endothelial cell line-1 (HMEC-1) stably expressing HA-human FcRn-EGFP (HMEC-1-hFcRn, kind gift from Dr. Wayne I. Lencer (Boston Children’s Hospital, Harvard Medical School and Harvard Digestive Diseases Center, USA)) (7.5×10⁴ viable cells/well) were seeded into Costar 48-well TC-treated Multiple Well Plates (Corning) and cultured in cell culture medium of MCDB 131 Medium (Cat. 10372019, Gibco) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Cat. F7524, Sigma-Aldrich), 1xPenicillin-Streptomycin (Cat. P4458, Gibco), 2 mM L-Glutamine (Cat. 25030081, Gibco), 10 ng/mL Recombinant Mouse Epidermal Growth Factor (Cat. PMG8041, Gibco), 1 µg/mL hydrocortisone (Sigma-Aldrich), 100 µg/mL Geneticin Selective Antibiotic (Cat. 10131035, Gibco), and 5 µg/mL Blasticidin S HCl (Cat. A1113903, Gibco) for 24 h at 37 °C in a humidified atmosphere of 5% CO₂. The cells were then washed three times and starved in Hank’s Balanced Salt Solution (HBSS) (Cat. 14025100, Gibco) for 1 h. After starvation, 800 nM of RBD-HSA fusions diluted in 125 µL of HBSS was added and incubated for 3 h followed by washing five times with cold HBSS. Subsequently, warm assay medium (complete cell culture medium without FBS, Blasticidin S HCl and Geneticin, but supplemented with 1xEagle’s Minimum Essential Medium Non-Essential Amino Acids Solution (Cat. 11140050, Gibco)) was added for 3 h. The assay media were collected to determine the recycled amounts of fusions by ELISA, as described below.

Transwell assay

A previously described transwell assay²⁹ was adapted to the Madin-Darby Canine Kidney II cell line that overexpresses human FcRn (MDCKII-hFcRn), and received as a kind gift from Dr. Alexander Haas and Jens Fischer (pRED Roche Innovation Center Munich, Germany). Briefly, 12 mm Transwell-COL Collagen-coated 0.4 µm Pore PTFE Membrane Inserts (Cat. 10457031, Corning) in Costar 12-well TC-treated Multiple Well Plates (Cat. 07-200-83, Corning) were initially equilibrated by adding complete cell culture medium (Dulbecco’s Modified Eagle Medium (DMEM) with GlutaMAX Supplement (Cat. 31966, Gibco), added 10% FBS and 300 µg/mL Geneticin Selective Antibiotic), and incubating overnight. MDCKII-hFcRn cells (1.25×10⁶ viable cells/well) were then seeded into the inserts and incubated for 24–26 h at 37 °C in a humidified atmosphere of 5% CO₂. The cells were then washed three times with pre-warmed HBSS, before transepithelial electrical resistance ((600-1100) Ωxcm²) was measured using a Millicell ERS-2 Voltohmmeter (Millipore). Then, the cells were starved for 1 h in HBSS. Subsequently, 800 nM of RBD-HSA fusions diluted in 200 µL of HBSS was added to the transwell inserts (apical side) followed by sampling of 900 µL of incubation solution (HBSS) from the wells (basolateral side) after 4 h. The transcytosed amounts of the RBD-HSA fusions were determined by ELISA, as described below.

In vivo vaccine studies

All mice included in this study were housed under controlled environments including ventilated cages with regularly change of bedding and nesting material, ad libitum access to water and food, light/dark cycle of 12 h and regulated ambient temperature (21 °C ± 2 °C) and relative humidity (30-70%). BALB/c mice (BALB/cAnNRj, strain #0003 Janvier-Labs) (female, 6-8 weeks old; 5-12 mice per group (HA, AQ-HA and MSA-HA), or female, 8-9 weeks; 5 (PBS and BioNTech-Pfizer) or 6 (RBD and RBD-MSA) mice per group), Tg32-hFc mice (B6.Cg-Tg(FCGRT)32Dcr Fcgrt^tm1Dcr Ighg1^{em2(IGHG1)Mvw}/MvwJ, strain #029686, The Jackson Laboratory) (female and male, 6-13 or 31 weeks; 5-7 mice per group (except RBD group in intranasal vs intramuscular experiment – 3 and 4 mice, respectively) and HSA/hFcRn mice (C57BL/6N-Fcrgt^{tm1.1(FCRGT)Geno};Alb^{tm1.1(ALB)Geno}, genOway) (female, 13-14 weeks, 6 mice per group) were used for SARS-CoV-2 and Influenza HA vaccine studies.

The mice were anesthetized by intraperitoneally administering 5–7.5 mL/kg of ZRF cocktail containing 250 mg/mL of Zoletil Forte, 20 mg/mL of Rompun, and 50 µg/mL of Fentanyl in saline solution (Section for Comparative Medicine, Rikshospitalet, Oslo University Hospital). After sedation, 20 µL of vaccination solutions containing 0.22 nM 10xHis-RBD (6.2 µg), RBD-MSA (19.9 µg), RBD-HSA (20.0 µg), or RBD-Tf (22.0 µg) fusion, or 0.16 nM His-tagged HA (13.1 µg), MSA-HA (27.5 µg), AQ-HA (27.5 µg), HA-QMP (26.9 µg) or HA-Tf (28.9 µg) mixed with 20 µg CpG ODN 1826 VacciGrade (Cat. Vac-1826-1, InvivoGen), or 0.22 nM RBD-QMP (20.0 µg) or CpG-RBD-QMP (21.4 μg), or 1xPBS or saline (NaCl) alone were intranasally administered to each mouse (10 µL/nostril). For intramuscular delivery in the quadriceps muscle, 0.22 nM RBD-QMP (20.0 µg) with 20 µg CpG, 3 µg of Comirnaty (BioNTech-Pfizer) or 1xPBS were given (15 µL/leg). A subsequent dose with 10% of the RBD, RBD-albumin, HA or MSA-HA fusion (0.022 nM) mixed with 20 µg CpG, 0.3 µg of Comirnaty or PBS/NaCl was given 3–4 weeks after the initial dose.

Blood samples were collected from the saphenous vein. For mice vaccinated with an HA-fusion, blood was collected at weeks 1, 3, 6 and 7.5 after vaccination. For mice vaccinated with an RBD-fusion, blood was collected at weeks 1, 2, 4 and/or 5 after vaccination. Sera were isolated by centrifugation at 17,000 x g for 10-25 min at 4 °C. In addition, mucosal flushes and bronchioalveolar lavage fluid (BALF) were collected at endpoint for selected experiments. All sites were flushed with sterile PBS, and except BALF all samples were directly transferred to protease inhibitor to a final concentration of x1-x2. The following volume of PBS was used: Nose – 15 µL in each nostril, saliva – 25 µL for each check with 3 aspirations per side, vaginal and rectal – 25 µL x 2 with 3 aspirations, BALF – 1000 µL. Samples with protease inhibitors were centrifuged at 17,000 x g for 10–25 min at 4 °C to remove debris. All samples were stored at −20 °C until analyses of antigen-specific immune responses, using ELISA or FCBA as described below. At endpoint mediastinal lymph nodes (also inguinal for PBS groups) were harvested for flow cytometry analyses. The animal experiments were carried out at the Department for Comparative Medicine, Oslo University Hospital (KPM Rikshospitalet; Oslo, Norway), with approval from the Norwegian Food Safety Authority, and performed in accordance with the Guide for the Care and Use of Laboratory Animals of the Norwegian National Institute of Health. The animal facility is classified as specific pathogen-free as defined by Federation of European Laboratory Animal Science Associations. Experimental and control animals were bred in the same facility and co-housed.

Vaccination of K18-hACE2 mice (B6.Cg-Tg(K18-ACE2)2Prlmn/J, strain #034860; The Jackson Laboratory) female, 10–12 weeks; 10 (PBS) or 12 (RBD and RBD-MSA) mice per group) were performed at the Association for the Assessment and Accreditation of Laboratory Animal Care (AAALAC) accredited The Scripps Research Institute (TSRI; La Jolla, CA, USA), and the experiment was approved by The Institutional Animal Care and Use Committee (IACUC) at TSRI. Vaccine mixtures and administration methods were as above, except anesthesia was achieved using inhaled isoflurane and blood samples were collected through retroorbital bleed at days 14 and 28. Experimental and control animals were bred in the same facility and co-housed.

Live virus challenge

The 5x Lethal Dose 50 (5xLD50) of mouse-adapted Influenza A H1N1 A/PR/8/34 virus (Cat. VR-95, ATCC) for BALB/c mice and Tg32-hFc mice were determined according to the Reed and Muench method^78,79. Vaccinated mice were transferred to the BSL2 approved facility at KPM Rikshospitalet prior to challenge with a deadly dose (5xLD50) of PR8 either 5 weeks, 8 weeks or 4.5 months after initial vaccination. Specifically, the mice were anesthetized and intranasally administered with the virus (10 µL/nostril). Weight loss was monitored daily or every second or third day after infection, in which the endpoint was set at 20% weight reduction. If the endpoint was reached prior to the end of the experiment, mice were terminated by cervical dislocation or using a CO₂ gas chamber. PR8 challenge was performed with approval from the Norwegian Food Safety Authority.

Vaccinated K18-hACE2 mice were transferred to the BSL3 approved vivarium at TSRI and challenged with 2×10⁴ pfu SARS-COV-2 WA1/2020 (kind gift from Dr. John Teijaro (TSRI, La Jolla, CA, USA)) as above, anesthetized by isoflurane. The mice were monitored and weighed daily for seven days, after which they were euthanized and the lungs harvested for analyses. The SARS-CoV-2 challenge was approved by the IACUC at TSRI.

Histopathology

The left lung lobes were rinsed with PBS and fixed with 4% formalin at endpoint. At The Unit for Research Support at the Department of Pathology, Oslo University Hospital the fixed lungs were embedded in paraffin and cut at 3 µm thick slices before they were manually stained with ready-to-use hematoxylin and eosin (Harris HTX Histolab 1 L, cat. 01800, Histolab). Blinded microscopic analysis was performed by an educated pathologist. Pictures of a selection of slides were taken with an Olympus BX53 microscope connected to a XC50 camera using cellSens Entry.

Viral RNA quantification

Viral RNA was isolated from half of each right lung lobe, and subsequently amplified and quantified by reverse transcription qPCR. The lung tissue was homogenized in 1 mL TRIzol reagent (Cat. 15596026, Invitrogen) using a Bead Ruptor 12 (Omni International). After centrifugation at 12,000 x g for 10 min at 4 °C, 750 µL supernatant was added to 200 µL chloroform, thoroughly mixed and spun again. The RNA containing aqueous phase was mixed with 350 µL of 70% EtOH for a 1:1 ratio, loaded onto the spin columns of RNeasy Mini kit (Cat. 74104, Qiagen) and purified according to the manufacturer’s instructions. RNA was eluted using 25 µL of UltraPure RNase-free water (Cat. 10977035, Invitrogen). After normalization, 1.2 µg RNA was mixed with 2.5 µL UltraPlex 1-Step ToughMix (Cat. 95166, QuantaBio) and 0.75 µL of Centers for Disease Control and Prevention’s N1 (nucleocapsid) primer sets (forward, 5′-GACCCCAAAATCAGCGAAAT-3′; reverse, 5′-TCTGGTTACTGCCAGTTGAATCTG-3′) (Cat. 10006821 and 10006822, Integrated DNA Technologies) and a fluorescently labeled (FAM) probe (5′-FAM-ACCCCGCATTACGTTTGGTGGACC-BHQ1-3′) (Cat. 10006823, Integrated DNA Technologies) in a total reaction volume of 10 µL. A standard consisting of 1.5×10⁶ PFU equivalents of SARS-CoV-2 RNA was serially diluted 10-fold and a no template control was included in each run. PCR was performed using CFX96 Real-Time PCR Detection Systems (Bio-Rad), running on CFX Manager (Bio-Rad).

ELISA

All consecutive ELISAs were performed according to the following protocol, unless otherwise stated. Corning 96-well EIA/RIA Polystyrene High Bind Microplates (Corning) were coated with 100 µL/well of the desired protein diluted in 1xPBS overnight at 4 °C or 2 h at RT. Wells were blocked with 200 µL/well of PBS containing 4% (w/v) skimmed milk powder (S) (Cat. 115363, Millipore) (PBS/S) for 1 h at RT before being washed with 250 µL/well of PBS supplemented with 0.05% (v/v) Tween 20 (Cat. P1379, Sigma-Aldrich) (PBS/T), three times consecutively. Next, 100 µL/well of samples or proteins (50 µL/well for sera or BALF) diluted in PBS/T supplemented with 4% (w/v) S (PBS/T/S) were added, incubated for 1–2 h at RT, and washed as above and 100 µL/well of a secondary antibody was added for 1 h at RT, followed by another wash cycle. Development of horseradish peroxidase (HRP)-conjugated antibodies was done by adding 100 µL/well 3,3’,5,5’-Tetramethylbenzidine (TMB) solution (Cat. CL07, Calbiochem), and the reaction was stopped by adding 100 µL/well of 1 M HCl. Assays using alkaline phosphatase (ALP)-conjugated antibodies were developed by adding 100 µL/well p-nitrophenyl phosphate substrate tablets (Cat. S0942, Sigma-Aldrich) dissolved in diethanolamine buffer (pH 9.8) (made in-house) to a final concentration of 1 mg/mL (pNpp). Absorbance measurements were performed using a Sunrise spectrophotometer (TECAN) at 405 nm or 450 nm for pNpp or TMB substrates, respectively, with a reference wavelength of 620 nm.

For detection of His-tagged HA and HA-fusions Influenza A H1N1 (A/Puerto Rico/8/1934) HA ELISA Pair Set (Cat. SEK11684, Sino Biological) was used.

Quantification of RBD-HSA fusions from HERA and Transwell assay was done in a two-way anti-albumin ELISA, detecting present albumin from serial dilution of a standard curve or samples (HERA, 1:2-1:8 and Transwell, 1:1-1:450) between a goat anti-HSA (Cat. A1151, Sigma-Aldrich) (1:2,000) and goat anti-HSA-ALP (Cat. A80-229AP, Bethyl Laboratories, Inc.) (1:4,000).

To determine binding between FcRn and albumin fusions, 10 µg/mL of His-tagged mouse FcRn or human FcRn (made in-house) diluted in PBS/S/T (pH 5.5) was added to plates coated with 8 µg/mL of a human IgG1 mutant (M252Y/S254T/T256E/H433K/N434F)⁸⁰ (made in-house) with specificity for 4-hydroxy-3-iodo-5-nitrophenylactic acid, followed by addition of serial dilutions of albumin fusions (225.6 nM-0.004 nM) in PBS/S/T (pH 5.5). For MSA-HA and AQ-HA, 39.4 nM-0.48 nM was used. MSA-HA and AQ-HA were detected with an anti-HA mAb H36-4-52 from mouse⁷⁵ (2 µg/mL) (made in-house), pre-incubated with anti-mIgG(Fc)-ALP (Cat. A2429, Sigma-Aldrich) (1:3,000). Albumin fusions were detected with either anti-MSA-HRP (Cat. ab19195, Abcam) (1,5000) or anti-HSA-ALP as above.

To investigate binding between ACE2 and RBD-albumin fusions, plates were coated with 2 µg/mL of His-tagged recombinant human ACE2 (Cat. 10108-H08H, Sino Biological) before serial dilutions of RBD-MSA and RBD-HSA fusions (150 nM-0.023 nM) were added, followed by detection with anti-MSA-HRP or anti-HSA-ALP.

Binding between ACE2 and RBD-Tf was confirmed by coating plates with 35 µg/mL RBD-Tf diluted 4-fold (349.60-0.08 nM) to an end volume of 100 uL. After blocking, plates were incubated with 1 µg/mL of a recombinant ACE2-HSA fusion protein (made in-house)⁴⁹. Finally, goat anti-HSA-ALP was used for detection as described above.

Binding to monoclonal antibodies with known epitope specificity was used as a measure for epitope availability on the fusions. In short, HA and HA fusion proteins were tested by adding 15 nM samples to plates coated with anti-His (Cat. ab18184, Abcam) (1:2000), anti-MSA (Cat. ab19194, Abcam) (1:1000), or anti-HSA. A 3-fold dilution series of the following monoclonal anti-HA with known antigenic site⁸¹ were added (mAbs from supernatant starting at 1:30 dilution and purified mAbs starting at 10 µg/mL, performed with 8 dilution steps); Y8-2C6, H28-E23, H17-L2, H36-11, H18-S413 and H9-A15 and detected using a biotinylated anti-mouse kappa light chain (made in-house, clone 187) (1:1000) and SA-ALP (Cat. 7105-04, SouthernBiotech) (1:3000). RBD and RBD fusion proteins (except RBD-Tf) were tested by adding a 4-fold serial dilution from 4 µg/mL (equimolar to RBD-HSA; 43.7-0.003 nM) to plates coated with 1 µg/mL of the monoclonal SARS-CoV-2 antibodies sotrovimab (GSK), cilgavimab or tixagevimab (AstraZeneca), before detection using anti-His-ALP (Cat. Ab49746, Abcam) (1:8000), anti-MSA-ALP or anti-HSA-ALP as above. For the RBD-Tf fusion, a 4-fold titration from 17.6 µg/mL (174.80–0.01 nM) was used for coating, followed by the monoclonal SARS-CoV-2 antibodies as above, before detection by anti-hIgG Fc specific-ALP (A9544, Sigma-Aldrich) (1:5000).

For the detection of RBD- or HA-specific antibodies from sera and BALF, titrated amounts of sera (diluted 1:50-1:109,350) or BALF (diluted 1:2-1:2,048) were added to plates coated with 1 µg/mL recombinant 6xHis-RBD (made in-house) or 0.5 µg/mL HA from Influenza A H1N1 (A/Puerto Rico/8/1934) (Cat. 11684-V08H, Sino Biological), while for detection of MSA- or HSA-specific antibodies from sera, titrated amounts (diluted 1:50-1:109,350) were added to plates coated with 0.5 μg/mL of recombinant MSA (Cat. A3559, Sigma-Aldrich) or HSA (made in-house). Anti-mIgG(Fc)-ALP or anti-hIgG(Fc)-ALP (1:5,000) were used to detect total IgG. For detection of IgG subclasses, anti-mIgG1-bi, anti-mIgG2a-bi or anti-mIgG2b-bi (Cat. 553500, 553502, 553393, BD Pharmingen), all pre-incubated with Streptavidin-ALP (1:3000) were used, while IgA was detected using anti-mIgA-ALP (Cat. 3865-9 A, Mabtech) (1:1,000). Data were presented as OD values, or as antibody titers which equal to the highest dilution factor for each mouse with a higher OD value than the background, where the background is the mean absorbance of mice given PBS/NaCl plus 5x the standard error of the mean of the same observations.

Serum samples were tested for their ability to inhibit human ACE2-RBD binding in ELISA. Plates were coated with 0.25 µg/mL of 6xHis-RBD. Serial dilutions of serum samples were added (1:30 dilution presented). Next, equimolar amounts of human ACE2 fused to HSA WT (1.364 µg/mL) and anti-HSA-ALP were used to determine the amount of RBD-bound ACE2.

Surface plasmon resonance

A Biacore T200 instrument (Cytiva) was used with 1xHBS-P+ (Cat. BR100671, Cytiva) as running buffer. 10 µg/mL of recombinant His-tagged mouse transferrin receptor (TFR1/CD71) (Cat. 50741-M07H, Sino Biological) was diluted in Acetate 4.5 (Cat. BR100350, Cytiva) and immobilized with the Amine Coupling Kit (Cat. BR10050, Cytiva) on a Series S CM5 Sensor chip (Cat. 29149603, Cytiva) to reach about 350 RU. Triplicates of 8,000 nM of either RBD-Tf or HA-Tf were injected using 177 mM phosphate, 85 mM NaCl, 0.005% Tween 20, pH 5.5 (made in-house) as running buffer, and 1x PBS supplemented with 0.005% Tween 20, pH 7.4 (made in-house) as regeneration buffer between every run. All runs were performed at 25 °C using a 40 µL/min flow rate, 120 seconds association phase and 680 seconds dissociation phase, with a 30 seconds regeneration phase at 30 µL/min in between each run. The binding response from a blank control flow cell was subtracted from all response curves to correct for background.

Quantification of RBD-specific antibodies in FCBA

Previously, a bead-based flow cytometric assay was adapted for the detection of antibodies against RBD from ancestral SARS-CoV-2 (Wuhan) as well as SARS-CoV-2 variants^41,42. Bead-based arrays with content of virus proteins were incubated for 1 h or overnight at RT, for measurements of IgGs and IgA, respectively, in sera and BALF, or overnight for measurements in other mucosal flushes, diluted in an assay buffer consisting of PBS supplemented with 1% Tween 20 (PBT), 1% Bovine serum albumin (BSA) (Cat. A7906, Sigma-Aldrich), 10 µg/mL neutravidin (NA) (Cat. 31000, Thermo Scientific), 10 µg/mL D-biotin (Cat. 2031, Sigma-Aldrich) and 0.1% sodium azide. The dilutions were as follows: serum – 1:100, or 1:300 for IgG in Tg32-hFc from experiment comparing intranasal and intramuscular vaccination from Fig. 4, or 1:900 for IgG in BALB/c from Fig. 1 and Supplementary Fig. 2, BALF – 1:10, or 1:30 for IgG in BALB/c from Fig. 1 and Supplementary Fig. 2, and mucosal flush –1:4 (nose; BALB/c and HSA/hFcRn), 1:6 (nose; Tg32-hFc), 1:3 (saliva; BALB/c and Tg32-hFc, and rectal; Tg32-hFc), 1:2 (saliva; HSA/hFcRn, and vaginal; all) or 1:1.5 (rectal; BALB/c and HSA/hFcRn)). The beads were then washed three times with PBT to remove unbound immunoglobulins and labeled for 30 min at RT with 10 µL of R-Phycoerythrin (PE)-conjugated goat anti-mIgG (Cat. 31861, Invitrogen), goat anti-hIgG (Cat. 109-115-098, Jackson Immunoresearch), rat anti-mIgG1, IgG2a, IgG2b, IgG3 (Cat. 1144-09, 1155-09, 1186-09, 1191-09, SouthernBiotech) or digoxigenin (DIG)-conjugated rat anti-mIgA (Cat. 3865-3, Mabtech). Stocks were diluted 1:200, 1:200, 1:100 and 1:100, respectively, in PBT containing 1% BSA and 0.1% sodium azide. After incubation with anti-mIgA-DIG, beads were washed three times before labeling for 30 min with 30 µL mouse anti-DIG-PE (Cat. 200-002-156, Jackson Immunoresearch) (0.5 µg/mL). Following a final wash, the beads were resuspended in PBT containing 0.1% BSA, and run on Attune NxT Flow Cytometer (Invitrogen). Specific antibody binding was measured as the ratio of PE median fluorescence intensity (MFI) of antigen-coupled beads to beads coupled with NA only, referred to as relative MFI (rMFI). The anti-mIgA-DIG conjugate was prepared by first changing the buffer of the antibody to PBS without sodium azide using 100 K cutoff Amicon Ultra Centrifugal Filter Units. Then, 1 mg/mL anti-mIgA was mixed with 200 µg/mL DIG-N-hydroxysuccinimidyl-ester (DIG-NHS) (Cat. 55865. Sigma-Aldrich) and incubated overnight. Excess DIG-NHS was removed by centrifugation through Amicon Ultra Centrifugal Filter Units with 100 K cutoff twice, before the solution was passed over a G50-Sephadex column (Cat. 17004202, Cytiva) equilibrated with PBS. Anti-DIG was conjugated to PE using standard thiol-maleimide chemistry.

FCBA for detection of neutralizing antibodies

A published method for measuring antibody-mediated inhibition of ACE2-RBD interactions was adapted to a microsphere format⁴². Briefly, an aliquot of the bead-based arrays used for antibody measurement (see above) were incubated for 1 h at RT with serum (diluted 1:100, or 1:300 for BALB/c from Fig. 1 and Supplementary Fig. 2) or BALF (diluted 1:10, or 1:30 for BALB/c from Fig. 1 and Supplementary Fig. 2) in assay buffer. The supernatant was removed, and the beads were incubated with 20 µL of an ACE2-HSA-DIG (0.8 µg/mL) for 30 min at RT. Following two washes with PBT, the beads were labeled with 30 µL anti-DIG-PE (0.5 µg/mL) for 30 min at RT, washed 4 times with PBT before resuspension in PBT containing 0.1% BSA and 10 µg/mL D-biotin, and run on Attune NxT Flow Cytometer (Invitrogen). Percentage inhibition of the ACE2-RBD interaction was calculated by measuring the ratio of PE MFI of antigen-coupled beads to beads coupled with NA only (rMFI), followed by normalization to relative MFI-values from PBS vaccinated mice, which was defined as 0% inhibition. The ACE2-HSA conjugate was prepared in-house as described^42,49. To confer readability and direct comparison with the pseudovirus neutralization assays, no detection of neutralizing antibodies in FCBA (0% inhibition) was defined as 100% binding.

Preparation of Spike (S) pseudotyped HIV-1 virions

Replication deficient SARS-CoV-2 pseudotyped HIV-1 virions were prepared as described previously⁴³. Briefly, virions were produced in 293T WT (293T-WT) cells by transfection with 1 μg of the plasmid encoding SARS CoV-2 Spike protein (pCAGGS-Spike Δc19)⁴³, 1 μg pCRV Gag-Pol⁸² and 1.5 μg GFP-encoding plasmid (CSGW)⁸³. Viral supernatants were filtered through a 0.45 μm syringe filter at 48 and 72 h post-transfection and pelleted for 2 h at 28,000 x g. Pelleted virions were drained and then resuspended in DMEM.

Spike pseudotyped neutralization with mouse sera or BALF

293T-hACE2-TMPRSS2 cells⁴³ were plated into 96-well plates at a density of 0.75 x 10³ cells per well and allowed to attach overnight. 20 µL pseudovirus was mixed with 2 µL dilutions of heat-inactivated mouse sera (made in-house) (1:100) or BALF (1:7 or 1:8.4) and incubated for 40 min at RT. 10 μL of this mixture was added to the cells. 72 h later, infection was detected through GFP expression by visualization on an Incucyte S3 live cell imaging system (Sartorius). The percentage of infection was quantified as GFP-positive cell area over total cell area. Relative infection was calculated as percentage virus infection in the presence of immune sera/BALF relative to virus only control.

Pseudovirus neutralization was performed on sera from mice⁸⁴. The serum was diluted 1:10 in DMEM (Gibco) supplemented with 10% heat-inactivated FBS and serially diluted 3-fold to a total of 8 dilutions. 20 μL of diluted serum was mixed with 20 μL of lentivirus-based SARS-CoV-2 pseudovirus in a 384-well white tissue culture plate (Greiner Bio-One), and the plate was incubated for 60 min at 37 °C humidified atmosphere of 5% CO₂. Subsequently, 20 μL containing 5,000 HeLa-ACE2 cells were added per well, following 48 h of incubation, before the cell culture media was discarded. The cells were lysed with 25 µL lysis buffer (25 mM Gly-Gly pH 7.8 (Cat. G1002, Merck), 15 mM MgSO₄ (Cat. 83266, Sigma-Aldrich), 4 mM EGTA (Cat. E3889, Merck), 1% Triton X-100 (Cat. X100, Merck)) with added BrightGlo substrate (Cat. E2620, Promega) (1:10 dilution) and luciferase activity/luminescence was measured on a BioTek Synergy 2 microplate reader running Gen5. Percentage neutralization was calculated accordingly: 100 x (1 – (RLU of sample – Average RLU of CC) / (Average RLU of VC-Average RLU of CC)), RLU referring to relative light unit. Values were normalized so negative neutralization percentages were set to equal 0%.

Flow cytometry on mediastinal lymph nodes

Collected mediastinal lymph nodes from mice were pooled according to treatment group and kept on ice or at 4 °C continuously from harvest to analysis and protected from light after addition of viability dye. Between steps, cells were centrifuged at 400 x g for 7 min at 4 °C before transfer to a 96-well plate, and 500 x g for 5 min at 4 °C after plating. Lymph nodes were mashed through a 70 µm cell strainer (VWR) to a single-cell suspension, before treatment with 3 mL RBC lysis buffer (150 mM NH₄Cl, 10 mM Na₂CO, 0.1 mM Na₂EDTA, made in-house) for 5 min. Cells were then washed with FACS buffer (2% FBS in PBS), and subsequently divided into triplicates of approximately 2 x 10⁶ cells. Next, the cells were stained with GhostDyeV510 (Cat. 13-0870-T100, Tonbo Bioscience) (1:400) for 20 min followed by blocking with 100 µL 25% normal rat serum (made in-house) in PBS for another 20 min.

The following surface markers were used: TCRβ-BB700 (Cat. 745846, BD Bioscience) (1:200), CD19-APC/Cy7 (Cat. 557655, BD Bioscience) (1:200), CD38-BUV395 (Cat. 740245, BD Bioscience) (1:200), CD45R(B220)-BUV805 (Cat. 748867, BD Bioscience) (1:200), GL7-Alexa Fluor 488 (Cat. 144612, Biolegend) (1:200) and PE-Klickmer-coupled RBD. The staining buffer also included BD Horizon Brilliant Stain Buffer (Cat. 563794, BD Bioscience) (1:50). 2 µL PE-Klickmer (Cat. DX01K PE 1000, Immudex) was used per sample, and it was coupled prior to use in a 1:15 molar ratio to biotinylated 6xHis- and Avi-tagged Wuhan RBD (made in-house)⁸⁵ according to the manufacturer’s protocol, yielding approximately 3.5 µL total volume per sample. The PE-Klickmer-coupled RBD was added to the cells first along with 5 µL FACS buffer and was allowed to incubate for 10 min before addition of 45 µL of the premixed surface antibody staining solution, followed by incubation for 30 min. Following washing, the cells were fixated with 100 µL 4% paraformaldehyde (Cat. 47608, Sigma-Aldrich) for 10 min, washed, and finally resuspended in 350 µL FACS buffer before analysis on a FACSymphony A5 (BD Bioscience) running FACSDiva (BD Bioscience) or an Attune NxT Flow Cytometer (Invitrogen) with the Attune NxT software (Invitrogen).

Western Blot on mucosal flushes

IgA-specific Western Blot was run on mucosal flushes to verify the presence of oligomeric forms of IgA at mucosal surfaces. Briefly, samples within a group were merged and amounts were reflected on IgA levels from FCBA. The samples were mixed in Milli-Q water to a total volume of 20 µL, containing 1X Bolt LDS Sample Buffer. For reducing gels 1 µL of DL-Dithiothreitol solution (Cat. 43816, Sigma-Aldrich) was added prior to exposure at 90 °C for 10 min. Samples were run on Bolt 4-12% Bis-Tris Plus Gel (Cat. NW04125, Invitrogen) in ice-cold 1X Bolt MES SDS Running Buffer, using Spectra Multicolor Broad Range or High Range Protein Ladders (Cat. 26623 and 26625, Thermo Scientific) as size standard. The gel was run at 80 V for 15 min followed by 110 min at 140 V while surrounded by ice blocks. After 6 min in 20% EtOH, the gel was transferred to iBlot 2 PVDF Mini Stacks (Cat. IB24002, Invitrogen) and run according to the manufacturer’s instructions on an iBlot 2 Gel Transfer Device (Invitrogen). After blocking the membrane in PBS/T containing 5% BSA, IgA was detected using goat anti-mouse IgA-HRP (Cat. A90-103P, Bethyl) (1:5,000 in PBS/T 1% BSA) and SuperSignal West Pico PLUS substrate (Cat. 34580, Thermo Scientific), and subsequently imaged using GeneSys on a Syngene G:Box Chemi-XX9.

Data analysis, graphing and illustration

HPLC data were obtained using Chromeleon Chromatography Data System Software (Version 7.2.4.0) before being graphed using GraphPad Prism software (Version 8.4.6). FCBA data were processed using Winlist 3D (Version 9.0.1, Verity Software House), and further analyses were performed in Microsoft Excel. Flow cytometry on lymph nodes were processed using FlowJo (Version 10.8.1, BD Bioscience). ELISA data were processed using Microsoft Excel. Spike pseudotyped neutralization data were processed using IncuCyte S3 (Version 2021 C, Sartorius), or BioTek Gen5 (Version 2.09, Agilent) and Microsoft Excel. SPR curves were processed using Biacore T200 Evaluation Software (Version 3.0, Cytiva). Western Blots were imaged using GeneSys (Version 1.4.1.0, Syngene). GraphPad Prism software (Version 9.0.0 or later) was used for data analysis, statistics, graphing and figure assembly. Biorender.com was used to create illustrations.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.