Plasmid construction

All plasmids for soluble SOSIP Env expression were codon-optimized and included the tissue plasminogen activator (TPA) signal peptide instead of the natural signal peptide. A SpyTag SOSIP Env-expressing plasmid was generated by adding downstream to the 1059-SOSIP gene (after residue 664) the DNA sequences encoding for the following peptides: an 8-reside linker sequence (GSGSGGSG), an 8-residue polyhistidine tag (HHHHHHHH), and a 13-residue SpyTag peptide (AHIVMVDAYKPTK). DNA was cloned by Gene Universal (Newark, DE). BG505 SOSIP.v6 expression plasmid was a kind gift from R.W. Sanders and I. Del Moral Sanchez (University of Amsterdam) and supports the expression of soluble, stabilized SOSIP.v6 trimer that contains additional 8 amino acid changes in comparison with BG505 SOSIP28. SpyCatcher003-mi3 plasmid was from Addgene (Cat# 159995). In vitro transcription (IVT) cassettes for mRNA synthesis were designed in-house and included bacteriophage T7 RNA polymerase promoter, stable untranslated regions (UTRs) based on previously published reports29,30,31, gene-of-interest, and a 120–125 base polyA tail (Supplementary Fig. 4). Designed DNA was synthesized and cloned into a pUC19 plasmid by Gene Universal. All plasmids were fully sequenced to verify correct sequence of the IVT cassettes. IVT plasmid for mRNA-mediated expression of HIV-1AD8 ΔCT Envs was generated by digesting the parental HIV-1AD8 WT env plasmid with BsiWI restriction enzyme (2 sites) to remove the CT DNA-encoding fragment and self-ligating the digested vector. The resulting vector contained only a 27-base polyA tail due to bacteria-mediated deletion during generation of this variant.

Protein expression and purification

293 F cells were co-transfected with a SOSIP- or SpyTag SOSIP-expressing plasmid and a human furin-expressing plasmid at a ratio of 4:1 using Turbo293 transfection reagent (Speed Biosystems; Gaithersburg, MD). Transfected cells were grown on a shaker in a tissue culture incubator at 37 °C, 8% CO2 for 3–5 days, and culture supernatants were then harvested, clarified by centrifugation at 4000 × g for 20 min, and filtered through 0.2 μm filter (VWR). Supernatant-containing SOSIP glycoproteins was loaded on a Galanthus nivalis lectin (GNL) column (Vector Laboratories) at 4–8 °C, the column was washed with 500 mM NaCl in phosphate-buffered saline pH 8 and proteins were eluted with 1 M Methyl-α-D-mannopyranoside/PBS solution, filtered through 0.2 μm filter and concentrated using Vivaspin 6 centrifugal concentrators (30 kDa; Cytiva). Purified SOSIP glycoproteins were then separated on a HiLoad 16/600 Superdex 200 pg size-exclusion chromatography column (Cytiva) and SOSIP trimer fractions were pooled, concentrated, and stored in aliquots at −80 °C until use. Full and un-cropped scans of the SDS-PAGE gels and western blots are provided as Supplementary Fig. 8.

SpyCatcher003-mi3 protein expression and purification

Synthetic nanocages displaying SpyCatcher on their surface were expressed and purified following protocols adapted from Rahikainen et al.25 Briefly, SpyCatcher003-mi3 plasmid was transformed to E. coli BL21(DE3) RIPL bacteria (Agilent, Santa Clara, CA) and transformed bacteria were grown on LB-Agar plates supplemented with 50 µg/mL kanamycin for 16 h at 37 °C. Subsequently, a single colony was picked and grown in a 10 mL LB culture containing 50 µg/mL kanamycin at 37 °C for 16 h with continuous shaking. The culture was diluted 1:100 into LB containing 50 µg/mL kanamycin and 0.8% (w/v) glucose and incubated at 37 °C with continuous shaking until OD600 reached 0.6. The bacteria were then induced with 0.42 M Isopropyl ß-D-1-thiogalactopyranoside (IPTG) (Fisher Scientific) at 22 °C for 16 h, centrifuged, and suspended in 25 mM Tris–HCl, 300 mM NaCl, pH 8.5. The bacteria were lysed by sonication (4 times for 60 s, at a 50% duty-cycle, using an ultrasonic processor (Qsonica) equipped with a microtip) in the presence of 0.1 mg/mL lysozyme (Sigma), 1 mg/mL cOmplete mini EDTA-free protease inhibitor (Merck), and 1 mM phenylmethanesulfonyl fluoride (PMSF) (Thermo Scientific). The lysate was clarified by centrifugation at 35,000 × g for 45 min at 4 °C, filtered through a 0.22 µm filter, and treated with 170 mg ammonium sulfate per mL of lysate followed by incubation at 4 °C for 1 h with continues mixing using with a magnetic stirrer. The precipitated nanocages were collected by centrifugation at 30,000 × g for 30 min at 4 °C, resuspended in 25 mM Tris–HCl, 150 mM NaCl, pH 8.5, and then dialyzed against 500-fold excess of the same buffer to remove residual ammonium sulfate. The SpyCatcher003-mi3 nanocages were further clarified by centrifuging at 17,000 × g for 30 mins at 4 °C, filtered to remove any insoluble material, and purified by size-exclusion chromatography (SEC) using a HiPrep Sephacryl S-400 HR 16–600 SEC column (GE Healthcare) equilibrated with 25 mM Tris–HCl, 150 mM NaCl, pH 8.5. Fractions containing the SpyCatcher003-mi3 nanocages were identified and concentrated using a Vivaspin 20 100 kDa MWCO concentrator. Endotoxin was removed from the sample using Triton X-114 (Sigma) phase separation as previously described32. Briefly, 1% (v/v) Triton X-114 was mixed with SpyCatcher003-mi3 by gentle pipetting and incubated on ice for 15 min to dissolve all the Triton X-114. The sample was then incubated in a 37 °C water bath for 5 min, centrifuged at 16,900 × g for 5 min at 30 °C, and the top phase was carefully transferred into a new microcentrifuge tube and the method was repeated twice. The final concentration of the endotoxin-depleted purified particles was measured by bicinchoninic acid (BCA) assay (Pierce, ThermoFisher Scientific) and nanocages purity was assessed by SDS-PAGE.

1059-SOSIP display on viral-like particles (synthetic nanocages)

1059-SOSIP-SpyTag and SpyCatcher nanocages were conjugated by incubation at different ratio (between 1:1 and 6:1) in 25 mM Tris–HCl, 150 mM NaCl, pH 8.0 at 4 °C for 16 h. Aggregates were removed by centrifugation at 16,900 × g for 30 mins at 4 °C. To confirm conjugation, samples were separated on an 8–16% Mini PROTEAN TGX stain-free gradient gel (Bio-Rad) under reducing conditions and imaged using a ChemiDoc XRS+ imager (Bio-Rad). At 6:1 ratio (SOSIP: synthetic nanocages) most of the nanocages were saturated and the unbound SOSIP Env was removed by SEC.

ELISA

We used enzyme-linked immunosorbent assay (ELISA) to analyze rabbit sera binding to SOSIP trimers. GNL was immobilized in a 96-well plate (Greiner Bio-One, NC) by adding 0.2 μg of GNL in 100 μl PBS in each well and incubating the plates overnight at room temperature (RT). Next, the wells were washed 3 times with PBS containing 0.2% Tween-20 (wash solution) using an in-house vacuum system and blocked with PBS containing 3% bovine serum albumin (blocking solution) for 2 h at RT. The wells were then washed 3 times, 0.2–0.25 μg of purified SOSIP trimers in blocking solution were added to test wells, and the plate was incubated for 1–2 h(s) at RT. Wells were washed 6 times and different dilutions of rabbit sera in a blocking solution were added. After 1 h and 30 min incubation, wells were washed 6 times, and 1:50,000 dilution of Peroxidase AffiniPure F(ab’)2 Fragment Goat Anti-Rabbit IgG (H + L) (cat# 111–036–045; Jackson ImmunoResearch, PA) was added in blocking solution to each well and the plate was incubated for 1 h at RT. Wells were then washed 6 times and 100 μl of TMB solution (1 ml of 1 mg/ml 3,3,5,5-tetramethylbenzidine (Sigma) in DMSO, 9 ml of 0.1 M sodium acetate, pH 5.0, and 2 μl of fresh 30% hydrogen peroxide) was added to each well. After ~18-min incubation, the HRP reaction was stopped by adding 50 μl of 0.5 M H2SO4, and optical density at 450 nm was measured using a spectrophotometer. We used the same procedure to measure the binding of human antibodies to the 1059-SOSIP. The following primary human antibodies were obtained from the NIH HIV Regent Program and used at 0.625 µg/ml: PGT121 (cat# ARP-12343), PG9 (cat# ARP-12149), VRC03 (cat# ARP-12032), F105 (cat# ARP-857), 2G12(cat# ARP-1476), VRC01 (cat# ARP-12033), 39 F (cat# ARP-11437), PGT145 (cat# ARP-12703), 10-1074 (cat# ARP-12477), and 17b (cat# ARP-4091). We used HRP-conjugated donkey anti-human IgG secondary antibodies at 1:5,000 dilution for detection (Peroxidase AffiniPure F(ab’)2 Fragment Donkey Anti-Human IgG, Fcγ fragment specific; cat# 709-036-098; Jackson ImmunoResearch Inc.).

In vitro transcription, mRNA purification & mRNA-LNP preparation

mRNA was simultaneously in vitro transcribed and capped using the T7-FlashScribe transcription kit (CellScript, Madison, WI) and CleanCap reagent AG (TriLink Biotechnologies, San Diego, CA) according to the manufacturer’s instructions. In some cases, we used modified pseudouridine (N1-methyl-pseudouridine-5′-triphosphate) instead of uridine-5′-triphosphate to stabilize the mRNA molecules. Transcribed mRNA was purified using a MEGAclear transcription clean-up kit (ThermoFisher Scientific) according to the manufacturer’s instructions and impurities were further removed on cellulose columns as previously described33. We measured mRNA concentration by optical density at 260 nm and froze the mRNA preparation at −80 °C until it was encapsulated by LNPs at Acuitas Therapeutics Inc (Vancouver, British Columbia, Canada). All mRNAs used for mRNA-LNP preparation were transcribed in vitro using modified pseudouridine and IF4-based plasmids (pIF4; Supplementary Fig. 4) except for second immunization of rabbits group 1, in which mRNA-LNPs contained mRNAs encoding for HIV-1AD8 ΔCT that were separately transcribed from two plasmids: IF4 and TEV (50% from pIF4 and 50% from pTEV; Supplementary Fig. 4). For co-expression and VLP generation, mRNA encoding HIV-1 Envs was mixed with mRNA encoding for HIV-1NL4-3 Gag at 1:1 ratio (based on experiments shown in Fig. 1d) and the total mRNA molecules were encapsulated in LNPs. All mRNA-LNP were stored in single-use tubes at −80 °C.

Animal care

Experiments involving New Zealand White rabbits were carried out according to NIH guidelines for the housing and care of laboratory animals. Protocols were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Minnesota.

Rabbit immunizations

Rabbits were intramuscularly immunized in a single site in the quadriceps muscle with either mRNA-LNPs (35 μg/rabbit), synVLP-1059-SOSIP (21.5 μg/rabbit) or soluble SOSIP trimers (50 μg/rabbit). Protein immunizations were administered after mixing the protein solution (PBS) with AddaVax adjuvant (InvivoGen, San Diego, CA) at a 1:1 volume ratio. Due to technical error, during the first immunization with mRNA-LNPs, rabbit 6 was immunized with only 11 μg and an additional immunization with 24 μg was administrated one week later. Experiment end points were week 37 (preliminary experiment 1) and week 36 (experiment 2) post the first immunizations in each experiment.

Blood collection

Blood was collected from the marginal ear vein/central ear artery according to an approved IACUC protocol. No anesthesia was used during blood collection. Rabbits’ ears were rubbed with gauze containing wintergreen oil before blood collection.

Rabbit anesthesia and euthanasia

Rabbits were euthanized by professional team at the animal facilities of the University of Minnesota under the supervision of a certified veterinarian and according to the approved IACUC protocol. Pentobarbital was used for euthanasia of the rabbits, and the following drugs were used for sedation prior to euthanasia: Ketamine 35 mg/kg IM; Xylazine 5 mg/kg IM; and Acepromazine 1 mg/kg IM. After sedation, 1.5 ml/5 kg Euthasol (pentobarbital sodium and phenytoin sodium) was administered IV to euthanize the rabbits. Euthanized rabbits were subjected to exsanguination according to the approved IACUC protocol.

On one occasion, due to ketamine shortage, the veterinarians recommended the temporary use of additional drug options for sedation prior to euthanasia for a subset of rabbits. These included Torbugesic (butorphanol tartrate) 0.5 mg/kg IM, midazolam 1.0 mg/kg IM, and dexmedetomidine 0.01 mg/kg IM.

Production of single-round pseudoviruses

We produced pseudoviruses as we previously described9,15,34,35,36. Briefly, a packaging plasmid (psPAX2), a reporter plasmid (pHIVec2.luc), and an Env-expressing plasmid were co-transfected into 293 T cells using effectene transfection reagent (Qiagen). After a 48-h incubation, the cell supernatant was collected and centrifuged for 5 min at 600–900 × g at 4 °C. The amount of p24 in the supernatant was measured using the HIV-1 p24 antigen capture assay (Cat# 5421, Advanced BioScience Laboratories), and the virus-containing supernatant was frozen in single-use aliquots at −80 °C. In some cases, the Env-expressing plasmid was replaced by Env-expressing mRNA that was transfected using Trans-IT (Mirus Bio LLC, Madison, WI) or by the direct addition of mRNA-LNPs, both 24 h post-transfection of the packaging and reporter DNA plasmids.

Viral infection assay

A single-round infection assay was performed in 96-well white plates (Greiner Bio-One, NC) using TZM-bl target cells (NIH AIDS Reagent Program) as previously described37. Thirty μl of diluted serum was added to each well followed by the addition of 30 μl of specified pseudoviruses and the plate was incubated for 1-h at 37 °C in a tissue culture incubator with 5% CO2. Then, approximately 7000 TZM-bl cells in 30 µl DMEM were added to each well, the plate was incubated for 48 h, cell lysed, and luciferase activity was measured by Centro XS³ LB 960 microplate luminometer (Berthold Technologies GmbH & Co. KG, Germany). Dose–response curves were non-linearly fitted to the logistic equation using Prism 9 (GraphPad Software, Boston, MA), after importing the logistic equation to the analysis software, as previously described34,38,39,40,41 but doses were expressed as dilution instead of absolute concentration and reported parameters were half-maximal inhibitory dilution (ID50). Lowest dilution tested was 1:20 for experiment 1 or 1:25 for experiment 2 (Supplementary Table 1). Due to weak neutralization response, ID50 values in some cases were below the lowest dilution tested (non-specific effects of the serum prevented testing lower than 1:25 dilution for experiment 2) and they were based on extrapolation of the fitted curves.

Peripheral blood mononuclear cells (PBMCs) isolation

Blood was collected from the ear vein of rabbits in 6 ml heparinized tubes and used immediately after collection. PBMCs were purified by density centrifugation (800 × g for 30 min) on Lymphocyte separation medium 1077 (PromoCell, Heidelberg, Germany), isolated from the gradient interface, washed twice in Dulbecco’s phosphate-buffered saline (DPBS, ThermoFisher Scientific, Waltham, MA), and resuspended in Roswell Park Memorial Institute (RPMI) 1640 medium (ThermoFisher Scientific, Waltham, MA) supplemented with 10% heat-inactivated fetal bovine serum (FBS).

ELISpot assay

ELISpot assay was based on the ELISpot Flex: Rabbit IFN-γ (HRP) kit (Mabtech, Cincinnati, OH). On day 1, a 96-well PVDF membrane white plate (Mabtech, Cincinnati, OH) was pre-treated with 35% ethanol for 1 min to activate the membrane and obtain maximal antibody binding capacity. The plates were then coated with 100 µl of 15 µg/ml unconjugated anti-rabbit IFN-γ mAb (MT327) in PBS and incubated overnight at 4 °C. On day 2, the plate was washed 5 times with sterile PBS and then blocked with RPMI 1640 medium at room temperature for at least 30 min. Freshly isolated PBMCs were added to the plate together with 2 µg/ml of the Peptide Pool, HIV-1 Subtype B (Consensus) Env Region (NIH-ARP Cat# 12540). Medium-only samples were added as controls to assess background level of lymphokine secretion. PMA/ionomycin was added to positive control wells. The plate was incubated for 18 h at 37 °C in 5% CO2 tissue culture incubator. On day 3, the plate was washed with PBS and then incubated for 2 h at room temperature with 0.1 µg/ml of biotinylated anti-rabbit IFN-γ mAb (MT318). The plate was then washed and incubated with 100ul of Streptavidin-HRP (1:1000 dilution) for 1 h at room temperature. ELISpot substrate (TMB; Mabtech, Cincinnati, OH) was added until distinct spots appeared. The plate was dried at room temperature overnight. On day 4, the number of spots in each well was measured using the CTL Immunospot analyzer (Cleveland, OH). We used 2 different cell densities in duplicate (either 2*105 or 5*105 cells/well), normalized the results to the frequency of cells in 1*106 PBMCs and averaged the 4 replicates (2 from each cell density).

ELISpot IgG assay

ELISpot IgG assay was based on the ELISpot Flex: Rabbit IgG (HRP) kit (Mabtech, Cincinnati, OH). On day 1, a 96-well PVDF membrane white plate (Mabtech, Cincinnati, OH) was pre-treated with 35% ethanol for 1 min to activate the membrane and allow maximal antibody binding capacity. The plate was washed 5 times with sterile water and coated with 100 µl of 2.5 µg/ml unconjugated antigen (1059-SOSIP trimer) in PBS and incubated overnight at 4 °C. On day 2, the plate was washed 5 times with sterile PBS and then blocked with RPMI 1640 medium with 10% FBS at room temperature for at least 30 min. Freshly isolated PBMCs suspended in RPMI 1640 10% FBS were added to the plate and medium-only samples were added as controls to assess the background level of IgG secretion. The plate was incubated for 20 h at 37 °C in 5% CO2 tissue culture incubator. On day 3, the plate was washed with PBS and then incubated for 2 h at room temperature with 0.5 µg/ml of biotinylated detection mAb (MT536; Mabtech, Cincinnati, OH) in PBS-0.5% FBS. The plate was then washed and incubated with 100 μl of Streptavidin-HRP (1:1000 dilution) for 1 h at room temperature. ELISpot substrate (TMB) was added until spots were visible. The plate was dried at room temperature overnight. On day 4, the number of spots in each well was measured using the CTL Immunospot analyzer (Cleveland, OH). We used 5*105 cells/well in duplicate, subtracted background measurements of multiple wells (medium only) and calculated the number of antibody-secreting cells in 1*106 PBMCs.

Reporting summary

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