mRNA vaccine immunity is enhanced by hepatocyte detargeting and not dependent on dendritic cell expression – Nature Biotechnology

Materials

SM-102 was purchased from MedKoo Biosciences. Lipids 1,2-disteraroyl-sn-glycero-3-phosphocholine (DSPC), cholesterol and 1,2-dimyristoyl-sn-glycerol methoxypolyethylene glycol (DMG-PEG₂₀₀₀) were purchased from Avanti Polar Lipids. Buffered formaldehyde (4%, J60401-AK) was purchased from Thermo Fisher Scientific. All mRNAs and modified RNAs used were produced using IVT⁶⁶. All antibodies for flow cytometry were purchased from Thermo Fisher or Biolegend, as specified below. All antibodies for tissue histology were purchased from Thermo Fisher Scientific, Cell Signaling, Abcam or Takara, as specified below. Tetramer reagents were kindly provided by the National Institutes of Health (NIH) Tetramer Core Facility (Emory University Vaccine Center), as detailed below.

Cell lines

HEK293T, Phoenix-Eco, RAW264.7, MEL, AML12, 3T3, BCL1, C2C12 and HuH7 were obtained from the American Type Culture Collection (ATCC). GFP-A20 is a previously generated murine lymphoma cell line on the BALB/c background (H-2K^d) expressing eGFPP³⁶. All cell lines were cultured at 37 °C with 5% CO₂ in the ATCC-recommended growth media. Cell lines were not independently authenticated aside from the identity provided by the ATCC. Cell lines were cultivated in a humidified incubator at 5% CO₂ and 37 °C. Cells were split at 70% confluency and tested for Mycoplasma quarterly.

Cloning of RNA template for IVT

Template plasmid with a T7 promoter site and codon-optimized eGFP were constructed by Twist Biosciences. Open reading frames (ORFs) contained restriction sites for the exchange of reporter gene constructs. Forward and reverse primers were used to amplify all reporter constructs. ORFs, UTRs and primer sequences are summarized in Supplementary Table 1. Templates for IVT were generated by restriction cloning using a T7 promoter sequence containing 5′UTR, an ORF and 3′UTR. Our construct used NASAR UTRs⁶⁷. To construct miRNA-binding sites, we annealed two pairs of reverse and complementary oligonucleotides encoding 1–4 perfectly complementary target sites for a specific miRNA⁹. We then ligated the annealed oligonucleotide duplex into the IVT plasmid by cutting with BsrgI-HF (New England Biolabs (NEB), R3575) and SalI-HF (NEB, R3138) and ligated with T4 DNA ligase (NEB, 0202). Our design allowed us to encode four tandem miRTs and we could screen colonies for correct oligonucleotide insertion by inclusion of an MluI (NEB, R3198) restriction site, which only forms if oligos successfully anneal. The addition of a 120-nt poly(A) tail to the mRNAs was encoded on the PCR template with a reverse primer containing a 120-nt poly(T) sequence. To generate IVT template from plasmid, we amplified DNA with Q5 high-fidelity master mix (NEB, M0492). Following PCR, RNA was incubated at 37 °C for 30 min with DpnI (NEB, R0176) to remove plasmid contaminants and PCR product was cleaned up with the QIAquick PCR purification kit (Qiagen). Concentration was measured using a NanoDrop 2000 (Thermo Scientific) for application in IVT.

Synthesis and purification of mRNAs

All mRNAs were synthesized using the HiScribe T7 high-yield RNA synthesis kit (NEB) according to the manufacturer’s instructions. Native mRNAs were synthesized with kit-supplied ATP, CTP, UTP and GTP. Modified nucleotides including N1m, ΨU and m5C were purchased from TriLink Biotechnologies and used as full substitutions of UTP or CTP in the reaction to synthesize modified mRNAs. Both mRNAs and modified mRNAs were capped by inclusion of a 1:4 premix of CleanCap reagent (TriLink) and capping occurred cotranscriptionally. The reaction mixtures were incubated at 37 °C for 5 h and further incubated at 37 °C for 30 min in the presence of TURBO DNase I (Thermo Scientific). RNA products were purified with the Monarch RNA cleanup kit (NEB, T2050). The RNA concentration was measured using a NanoDrop 2000 (Thermo Scientific) and species purity and size were assessed on a denaturing RNA gel.

Nanoparticle formation

RNA-loaded LNP formulations were formed using a microfluidic based mixing device (Ignite, Precision NanoSystems) for in vitro and in vivo studies. SM-102 particles²⁴ were formulated with the helper lipids (DSPC, cholesterol and DMG-PEG₂₀₀₀; molar ratio 50:10:38.5:1.5) and mRNA or modified RNA dissolved in a citrate buffer. After formulation, the freshly formed RNA-LNPs were dialyzed overnight against PBS buffer using Slide-A-Lyzer dialysis cassettes (3.5-kDa molecular weight cutoff (MWCO); Life Technologies) and subsequently concentrated in Amicon Ultra centrifugal filters (10-kDa MWCO; Sigma) to the desired concentration. The particle size and zeta potential of LNPs were measured using a Zetasizer Advance (Malvern Panalytical) at a scattering angle of 173° and a temperature of 25 °C. The encapsulation efficiency of LNPs was determined using a Quant-it RiboGreen RNA assay kit according to manufacturer protocols. To quantify the RiboGreen assay, we used the Cytation 3 cell imaging reader (BioTek). Empty LNPs were formulated without mRNA in the aqueous phase. To label LNPs, DiI stain was added to the lipid mix at a 1:1,000 dilution before mRNA encapsulation. All formulations were within the following parameters: average size of 80–100 nm, encapsulation > 90% and polydispersity < 0.2.

Cloning of CAR constructs

An SFG-h1928z vector (hCD19 CAR) was constructed by stepwise Gibson assembly using the cDNA of a previously described anti-hCD19 scFv Myc-tag sequence (EQKLISEEDL), the murine CD28 transmembrane and intracellular domain and the murine CD3z intracellular domain into an SFG retroviral vector⁶⁸. A GFP moiety, separated from the CAR by a T2A self-cleaving peptide, was cloned downstream of the CAR coding region⁶⁹. CAR expression was verified by flow cytometry identification of GFP⁺ cells.

Retroviral vector production

Phoenix-Eco cells were seeded into 15-cm tissue culture plates (Thermo Fisher Scientific, Nuclon, 168381) 24 h beforehand to achieve an approximate cell density of 70% at the moment of transfection. Transfection was carried out using the calcium phosphate method, as described previously⁶⁹. Briefly, hCD19 CAR or control plasmid constructs were suspended in 0.1× TE (Maxi Qiagen kit, 12362) and 0.25 M CaCl₂ (Sigma-Aldrich, C7902-1KG) and one volume of 2× HBS (for 500 ml: 1 M HEPES (50 ml; Corning, 25-060-Cl), 2 M NaCl (70.25 ml; Fisher Bioreagents, BP358–1) and 0.5 mol l⁻¹ Na₂HPO₄ (1.5 ml; Fisher Bioreagents, BP332–500)). Then, 378.25 ml of tissue-culture-tested water (Corning, 46–000-CV), supplemented with 5 M NaOH (Thermo Fisher, 134070010) to the desired pH, was added in dropwise fashion while continually vortexing and the resulting solution was immediately added to Phoenix-Eco cells and allowed to sit overnight. DMEM (Gibco, 11-965-118) was replaced the next morning; supernatants were collected and passed through a 0.22-μm filter 24–30 h later. Supernatant aliquots were stored at −80 °C until use.

hCD19 CAR T cell generation

CAR T cells were generated as previously⁶⁹. Briefly, naive T cells were isolated from the spleens of CD45.1 C57BL/6 mice using the EasySep mouse T cell isolation kit (StemCell Technologies, 19851). Activation was carried out at a cell density of 1 × 10⁶ cells per ml in RPMI medium + 100 U per ml recombinant murine IL-2 (Peprotech 212-12). Mouse T-activator CD3/CD28 Dynabeads (Thermo Fisher Scientific, 11453D) were used to activate cells at a 1:4 bead-to-cell ratio for 24 h before transduction. Nontreated culture plates (Nunc, Thermo Fisher Scientific) were coated overnight at 4 °C with 20 μg ml⁻¹ RetroNectin (Takara, T100B). hCD19 CAR vector supernatant (see above) was spun for 90 min at 2,000g and 30 °C onto the plated RetroNectin and half the supernatant volume was removed carefully after spinning. T cells were resuspended in fresh RPMI medium with IL-2, added to the supernatant-containing wells (final IL-2 concentration of 50 U per ml) and allowed to sit for 24 h. After this, Dynabeads were magnetically removed and T cells were resuspended in fresh RPMI medium at 50 U per ml IL-2. New RPMI medium containing 50 U per ml IL-2 was added daily to keep T cells at a concentration between 1 × 10⁶ and 2.5 × 10⁶ cells per ml until use, within 5 days of isolation. Detection of GFP by flow cytometry was used to quantify the efficiency of transduction.

In vitro cell transfection and fluorescence detection

Cells were plated at 5 × 10⁵ cells per well in a clear-bottom tissue culture plate. After 24 h, the medium in each well was replaced with 100–500 µl of medium containing modified RNA-LNPs. After 18 h, the cells were collected in flow tubes and cell fluorescence was measured using a LSRFortessa-X20 flow cytometer (BD Biosciences). To test the effect of miRT inclusion into Cas9-mRNA, Cas9-mRNA (original sequence purchased from Addgene, plasmid 42230) was coencapsulated into LNPs with β2-microglobulin single guide RNA (Synthego, 5′-GAGTAGCGCGAGCACAGCTAAGG-3′) at a mass ratio of 2.5:1. The 293T cells were plated at a concentration of 1.2 × 10⁵ cells per well of a 24-well plate and transfected with 1 μg ml⁻¹ of RNA-LNPs after 24 h. After 4 days of culture, cells were detached and stained for 30 min on ice with APC anti-human β2-microglobulin antibody (Biolegend, 395712) diluted 1:200 in FACS buffer (PBS with 2% FBS and 0.5 mM EDTA). Cells were then acquired with BD FACSDiva software on an LSRFortessa-X20 (BD Biosciences). The effect of miR-122 antagomiR on the expression of GFP mRNA was tested on HuH7 cells. Cells were plated at a concentration of 1 × 10⁵ cells per well in a 24-well plate. After 24 h, each well was transfected with different amount of miR-122 antagomiR using Lipofectamine 3000 transfection reagent (Thermo Fisher, L3000015), according to manufacturer instructions. Then, 6 h after antagomiR transfection, cells were treated with 1 μg ml⁻¹ of LNP-encapsulated GFP RNA.WT or RNA.122T. Cells were acquired 18 h after LNP transfection with BD FACSDiva software on an LSRFortessa-X20 (BD Biosciences). Bone-marrow-derived DCs were differentiated from the bone marrow of BALB/c mice using GM-CSF and plated at a concentration of 6.25 × 10⁵ cells per ml. After 3 h, each well was transfected with 1 μg ml⁻¹ mRNA encapsulated in DiI-stained LNP. Supernatant was collected 24 h after LNP transfection for Olink analysis. Cells were sorted for expression of DiI on a FACSymphony S6 (BD Biosciences).

RT–qPCR

Total RNA was extracted from each sample using QIAzol lysis reagent (Qiagen) and cDNA was synthesized from the isolated RNA with the iScript cDNA synthesis kit (Bio-Rad). Transcript levels of Il10, Il6, Il12, Tnf and Il1b were measured by qPCR using the SsoAdvanced Universal SYBR green supermix (Bio-Rad). Gapdh was used as the endogenous reference gene for normalization across all samples and the relative expression of the target genes in treatment groups compared to the control group was calculated using the ΔΔC_t method. All qPCR primers were purchased from Integrated DNA Technologies.

Mice

In vivo studies were performed using C57BL/6, BALB/c, Ai14 and B6.SJL-Ptprc^aPepc^b/BoyJ (B6 CD45.1) mice obtained from the Jackson Laboratory (JAX, 000664, 000651, 007914 and 028062, respectively) and housed in the Mount Sinai vivarium during use. Jedi mice on a BALB/c background, derived from B10.D2 Jedi mice generated in the B.D.B. lab⁴⁰ by backcrossing. All mouse experiments were carried out under Institutional Animal Care and Use Committee (IACUC) approval. All mice were randomized before experimentation. For bone marrow transplantation experiments, lethally irradiated (11 Gy) Ai14 mice were reconstituted with 2 × 10⁶ bone marrow cells from B6 CD45.1 or Ai14 mice. Mice were allowed to reconstitute for at least 8 weeks before use. Age-matched (6–12 weeks) female animals were used throughout all experiments. For studies of the CD8 T cell response to GFP, OVA or Spike, the strain choice was dictated by the availability of MHC-I-restricted tetramers, as described below. All experiments carried out at the Icahn School of Medicine at Mount Sinai were performed in compliance with federal laws and institutional guidelines and were approved by the IACUC (protocol no. IACUC-2018-0070).

Mouse injections and sample collections

For biodistribution experiments, C57BL/6 or Ai14 mice with weights of 18–20 g were injected with GFP RNA-LNPs or Cre RNA-LNPs. For i.v. administration, mice were injected with 20 μg (0.1 mg ml⁻¹) RNA-LNPs through the lateral tail vein, unless otherwise stated. For i.m. administration, mice were injected with 5 μg (0.1 mg ml⁻¹) RNA-LNPs in the left tibialis anterior muscle. At experimental endpoints, mice were killed and tissue was collected. In Ai14 experiments, mice were first imaged using whole-organ fluorescence with IVIS. In all experiments, organs were subsequently divided in two, with half processed as a cell suspension for flow cytometry and the other half fixed overnight in 4% PFA and paraffin-embedded for immunohistochemistry (IHC).

For vaccination experiments, mice were i.m. or i.v. injected as described above. At midpoint intervals between doses, blood was collected by retroorbital sampling using a sterile hematocrit capillary tube. For flow cytometry analysis of immune cells, blood was collected at 4 °C and processed as described below. For collection of serum, blood was incubated at 25 °C for 30 min to allow clotting and pelleted for 10 min at 2,000g. At experimental endpoints, mice were killed; blood and tissue were collected and processed for immune assays, as described below.

For tumor inoculation, GFP-A20 lymphoma cells³⁶ were passed through a 70-μm cell strainer and resuspended in ice-cold sterile PBS at a concentration of 10⁷ cells per ml. A total of 10⁶ cells were injected subcutaneously in a volume of 100 μl into the right back flank using a 27G needle. Mice were vaccinated with 20 μg of RNA-LNPs injected i.v. according to the schedule shown in respective figures. Tumor sizes were measured with a caliper every 2 or 3 days for calculating tumor volumes using the equation (width² × length)/2. Animals were killed when exhibiting signs of impaired health, when the tumor ulcerated or when the tumor volume exceeded 1,500 mm³.

Adoptive cell transfer

Jedi T cell experiments were performed as previously described⁴⁶. Briefly, donor spleen was collected from Jedi mice and processed into a single-cell suspension by mincing using a sterile blade, homogenizing by pipetting and filtering through a 70-µm cell strainer. The spleen solution was washed once with R10 medium and pelleted by centrifuging for 5 min at 400g. The supernatant was removed and the cell pellet was resuspended in 1 ml of 1× red blood cell (RBC) lysis buffer. After incubation, 10 ml of PBS was added to stop RBC lysis. The solution was centrifuged again at 400g for 5 min to obtain a cell pellet. A single-cell suspension was processed for isolation of CD8⁺ T cells using the EasySep mouse CD8⁺ T cell isolation kit (StemCell Technologies, 19853) as per the manufacturer’s protocol. Isolated T cells were counted and suspended in sterile PBS to a concentration of 5 × 10⁶ per ml. A total of 5 × 10⁴ Jedi T cells were transplanted into Balb/c mice through i.v. injection into the retrobulbar venous plexus. After another 24 h, recipient mice were i.v. injected with 20 μg of GFP-encoding RNA-LNPs. Mice were killed at specified time points (24, 72 and 120 h) and blood and tissue were collected for immune analyses and tissue imaging.

For adoptive cell transfer of hCD19 CAR T cells, CAR-expressing CD45.1⁺ T cells were incubated with CellTrace violet (CTV) according to the manufacturer’s protocol and injected i.v. into the retrobulbar venous plexus into total-body-irradiated C57BL/6 recipient mice (4 Gy). Then, 24 h later, mice received 20 μg of RNA-LNPs through i.v. injection. Mice treated with irrelevant antigen RNA-LNPs served as controls.

PD1 blockade

For immune checkpoint blockade, BALB/c mice were injected i.v. with a priming dose of 20 μg of GFP RNA-LNPs. Then, 5 days later, mice were injected intraperitoneally with 200 μg per dose of InVivoMAb anti-mouse PD1 (CD279) (clone RMP1-14; BioXCell, BE0146) or InVivoMAb rat IgG2a isotype control (BioXCell, BE0089) antibodies biweekly until they received a booster dose of 20 μg of GFP RNA-LNPs on day 21. The experiment endpoint was 5 days after the booster dose, at which point mice were killed and tissues samples were collected.

Whole-organ imaging

To measure whole-organ fluorescence of organs from Ai14 mice following Cre RNA-LNP injection, mice were first killed and then transcardiac perfused with 20 ml of PBS to remove blood, which would confound measurement of organ fluorescence. Indicated organs were collected on ice and fluorescence was measured (excitation: 550 nm and emission: 580 nm) using an IVIS imaging system (PerkinElmer) and quantified using LivingImage software (PerkinElmer). Tissue was subsequently processed for use in other analysis modalities as described below.

Tissue preparation

Peripheral blood was collected from the orbital sinus. A20 lymphoma tumors, spleens and LNs were collected in RPMI with 2% FBS on ice, homogenized by smashing using the plunger of a 3-ml syringe (BD Biosciences) and filtered through a 70-µm cell strainer. Erythrocytes for peripheral blood, spleens and tumors were removed using 1× RBC lysis buffer (Thermo, 00-4333-57). Bone marrow cells were flushed from femur and tibia bones, homogenized and filtered through a 70-μm cell strainer, before erythrocytes were removed by hypotonic lysis. For tissue preparation of livers, transcardiac perfusion with 10 ml of ice-cold PBS was first performed. Livers were collected in RPMI with 2% FBS on ice, minced using a sterile blade and digested with 1 mg ml⁻¹ collagenase IV (Gibco, 17104019) and 40 μg ml⁻¹ DNase I (Signma-Aldrich, DN25) dissolved in PBS with 5% FBS for 30 min at 37 °C with constant shaking. Liver homogenate was filtered through a 70-μm cell strainer. Cells were resuspended in 40% Percoll (Cytiva), previously adjusted with 10× PBS and layered onto 70% Percoll. The Percoll gradient was centrifuged at 400g for 25 min at 24 °C with no acceleration and no brake. The layer containing immune cells was collected and washed with FACS buffer (PBS with 2% FBS and 0.5 mM EDTA).

Flow cytometry

To assess expression of tdTomato⁺ or GFP⁺ cells in the different cell types of each organ, cell isolation and staining were performed after either day 1 or day 3 of treatment with RNA-LNPs followed by flow cytometry analysis. In all organs analyzed, after generating single-cell suspensions, samples were divided in two and stained with a lymphoid-specific panel or myeloid-specific panel of antibodies. Lineage-negative gates were included to exclude myeloid cells from lymphoid analysis and vice versa. All flow panels included a viability stain to discriminate live cells (live/dead fixable blue dead cell stain kit; L23105, Thermo Fisher). Single-cell suspensions were incubated in FACS buffer (PBS, 2% FBS and 0.5 mM EDTA) containing CD16/CD32 (Mouse Fc Block, BD) for 10 min before and during staining with extracellular antibodies. GFP-specific CD8⁺ T cells were detected with H-2K^d/GFP_200–208 (HYLSTQSAL) tetramer, OVA-specific CD8⁺ T cells with H-2Kb/Ova_257–264 (SIINFEKL) tetramer and SARS-CoV-2 Spike-specific CD8⁺ T cells with SARS-CoV-2 Spike_539–546 (VNFNFNGL) tetramer, all kindly provided by the NIH Tetramer Core Facility (Emory University Vaccine Center). For tetramer staining, splenocytes or tumor cell suspensions were stained for 1 h at room temperature in the dark with tetramer and 50 nM dasitinib (StemCell Technologies, 73084). Viability stain with live/dead fixable blue stain (Thermo Fisher) was performed for 20 min at 4 °C, followed by staining for cell surface markers for 20 min at 4 °C.

Antibodies for extracellular staining included the following: from Biolegend, CD11b (M1/70, FITC and APC), FOLR2 (10/FR2, APC), XCR1 (ZET, APC–Cy7 and APC), MHC-II (M5/114.15.2, BV510), F4/80 (BM8, BV605), CD11c (N418, BV650), CD64 (X54-5/7.1, BV711 and PE), CD45 (30-F11, BV510, BV785 and Spark NIR 685), CD3 (17A2, Spark UV 387), CD4 (RM4-5, PE, BV605 and BV785), CD25 (PC61, PE/Dazzle594), CD44 (IM7, BV510), CD62L (MEL-14, BV570), TIM3 (RMT3-23, BV711), CD8a (53-6.7, BV785), CD69 (H1.2F3, APC–Cy7); from eBioscience, NKp46 (29A1.4, PerCP–eFluor 710), CD45.1 (A20, PE–Cy7), Ly6G (1A8–Ly6g, Alexa Fluor 700), Ly6C (HK1.4, eFluor450), PD1 (J43, PerCP–eFluor 710), CD3e (145-2C11, FITC), NK1.1 (PK136, APC–eFluor780), CD19 (eBio1D3, eFluor450), CD64 (10.1, PE–Cy7); from BD Bioscience, B220 (RA3-6B2, BUV563), CD8a (53-6.7, APC), Siglec-F (E50-2440, APC–Cy7) and CD25 (PC61, PE). Intracellular cytokine staining was performed with monoclonal antibodies to IFNγ (XMG1.2, PE–Cy7), and granzyme B (QA16A02, APC) from Biolegend, using the eBioscience intracellular fixation and permeabilization buffer set (Invitrogen, 88-8824-00) after stimulation of 2 × 10⁶ splenocytes or tumor cells with 1 μM GFP_200–208 peptide, 1 μM SIINFEKL or Spike peptide pool (2 μg ml⁻¹ per peptide) in the presence of brefeldin A (eBioscience 1000X solution) for 5 h at 37 °C. Transcription factor staining was performed with FoxP3 antibody from eBioscience (FJK-16s, PE–Cy7) using the FoxP3, transcription factor fixation and permeabilization concentrate and diluent kit (eBioscience). All intracellular staining was performed overnight at 4 °C. tdTomato fluorescence was detected in the PE channel and GFP fluorescence was detected in the FITC channel. Samples were acquired with BD FACSDiva software on a LSRFortessa-X20 (BD Biosciences) or Cytek SpectroFlo software on an Aurora 5L (Cytek) and analyzed with FlowJo (version 10.10.0).

Ex vivo uptake studies

For the ex vivo splenocyte culture, spleens from BALB/c mice were processed to a single-cell suspension as above. A total of 5 × 10⁵ splenocytes were resuspended in R10 medium and transfected with 1 μg of RNA-LNPs. After 18 h, splenocytes were collected for flow cytometry to quantify GFP. For primary monocyte transfection, monocytes were isolated from the spleen using the Miltenyi Biotec monocyte isolation kit for mice (Miltenyi, 130-100-629) and cultured in nontissue-culture-treated plates at 5 × 10⁵ cells per ml R10 medium supplemented with 40 ng ml⁻¹ of M-CSF (PeproTech). The following day, cells were transfected with GFP-encoding RNA-LNPs. After 18 h, GFP expression was quantified by flow cytometry.

Multiplexed IHC consecutive staining on single slide

Iterative cycles of immunostaining on 5-µm-thick formalin-fixed paraffin-embedded (FFPE) tissue sections were performed as previously described⁷⁰. Briefly, slides were baked overnight at 60 °C, then deparaffinized in xylene and rehydrated in descending series of 100%, 90%, 70% and 50% ethanol. Slides were incubated at 95 °C for 30 min in antigen retrieval solution (pH 9, Dako), cooled at room temperature and rinsed in Tris-buffered saline. Tissue endogenous peroxidase activity was quenched by a 15-min incubation in 3% H₂O₂ and slides were subsequently blocked with serum-free protein block (Agilent) for 30 min at room temperature. Tissue was stained with primary antibody diluted in background reducing antibody diluent (Agilent) for 1 h at room temperature, then washed three times and incubated with the horseradish-peroxidase-conjugated secondary antibody for 30 min at room temperature. Antigen detection was performed using the AEC peroxidase substrate kit (Vector Laboratories) and slides were counterstained with Harris-modified hematoxylin solution (Sigma-Aldrich). The slides were mounted in Glycergel mounting medium (Agilent) and imaged on an Aperio AT2 slide scanner (Leica) at ×40 magnification. To perform the subsequent staining, the coverslip was removed by incubating the slides in 60 °C water and AEC and hematoxylin were removed in ascending series of 50%, 70% (with 1% HCl 12 N) and 100% ethanol. Sections were then rehydrated in descending series of 70% and 50% ethanol. From that point, the staining process continued iteratively, with a shortened antigen retrieval step (15 min at 95 °C). If two primary antibodies used consecutively were raised in the same species, an extra species-specific blocking step was performed with AffiniPure Fab fragment donkey anti-‘species’ IgG (Jackson Immuno Research). The following primary antibodies were used: anti-GFP (Takara, 632381), anti-DsRed (Takara, 632496), anti-F4/80 (Cell Signaling, 70076S), anti-CD31 (Abcam, ab182981), anti-CD8a (Cell Signaling, 98941S), anti-B220 (Thermo Fisher Scientific, 14-052-82), anti-LY6G (BioLegend, 127601), anti-αSMA (Abcam, ab5694), anti-MHC-I (Cell Signaling, 76828) and anti-cleaved caspase 3 (Cell Signaling, 9661).

Immunofluorescence tissue imaging

Flash-frozen unfixed mouse livers were embedded in OCT. Tissue was cryosectioned and mounted on slides with Fluoromount-G containing DAPI (Southern Biotech, 0100-20). The slides were blinded and immediately imaged with a Zeiss LSM780 fluorescence microscope.

Cyclic immunofluorescence

Cyclic immunofluorescence (CyCIF) was performed as previously described in detail^71,72. Briefly, 5-μm-thick FFPE tissue sections were baked at 60 °C overnight, deparaffinized in xylene and rehydrated in decreasing concentrations of ethanol solutions (100%, 90%, 70% and 50%). Antigen retrieval was performed by incubating the slides in antigen retrieval solution (pH 9, Dako) at 95 °C for 30 min. Slides were then cooled at room temperature for 30 min and washed with PBS. Slides were photobleached by immersing them in a bleaching solution (4.5% H₂O₂ and 20 mM NaOH in PBS) with LED light exposure twice for 45 min to reduce autofluorescence. To mitigate nonspecific antibody binding, slides were washed three times for 5 min with 1× PBS and then incubated overnight with secondary antibodies (anti-rat, anti-mouse and anti-rabbit) diluted 1 in 1,000 in 150 μl of Odyssey blocking buffer at 4 °C in the dark. Slides were subsequently washed three times with 1× PBS before photobleaching them again twice for 45 min. For each round of CyCIF, samples were incubated overnight at 4 °C in the dark with Hoechst 33342 (1:10,000 dilution; Thermo Fisher Scientific) for nuclear staining along with either primary conjugated or primary unconjugated antibodies diluted in 150 μl of Odyssey blocking buffer (LI-Cor). Incubation with primary unconjugated antibodies was followed by secondary antibody incubation at room temperature for 2 h in the dark. Then, the slides were washed three times for 5 min and mounted with 200 μl of 70% glycerol. Slides were automatically imaged on the RareCyte Cytefinder II HT using the following channels: ultraviolet, Cy3, Cy5 and Cy7. Imaging was performed with the following parameters: binning, 1 × 1; objective, ×20; numerical aperture, 0.75; resolution, 0.325 μm per pixel. Image exposures were optimized for each channel to avoid signal saturation but kept constant across samples. After imaging, slides were placed in containers of 1× PBS and heated in a water bath for 1 h to remove coverslips. Between each cycle, slides were photobleached twice for 45 min and washed three times for 5 min in 1× PBS. The antibodies used in this panel were anti-MHC-I (Cell Signaling, 76828), anti-F4-80 (Cell Signaling, 70076S) and anti-CD8a (Cell Signaling, 98941s).

ELISA assays

To coat plates, 96-well polystyrene high-bind microplates (Corning, 3361) were incubated with recombinant Aequorea victoria GFP protein (Abcam, ab84191) at a concentration of 1 μg ml⁻¹ at 4 °C overnight. Plates were washed and incubated in blocking buffer containing PBS with 5% milk and incubated for 1 h at 25 °C. Plates were incubated with serial dilutions of sera for 1 h at room temperature. Next, plates were washed and incubated with goat anti-mouse IgG(H + L) AP antibody (Southern Biotech, 1031-04) at a concentration of 1:2,000 in blocking buffer and incubated for 30 min at room temperature. Plates were washed one last time and the signal was detected with the AttoPhos AP fluorescent substrate system (Promega, PR-S1000). The plates were allowed to develop for 10 min and stopped with 3 N NaOH (Sigma). Plate fluorescence (excitation: 450 nm, emission: 580 nm and gain = 55) was measured with the Cytation 3 cell imaging reader (BioTek). To measure anti-Spike serum antibody titers, 96-well plates were coated with SARS-CoV-2 spike protein S1 (residues 14–683; Thermo Fisher, RP-87681) at a concentration of 2.5 μg ml⁻¹ in PBS and incubated at room temperature overnight. Blocking was performed with 1% BSA, 1 mM EDTA and 0.05% Tween in PBS at room temperature for 2 h. The following steps were performed as previously described, with the only difference being the secondary antibody incubation that was extended to 1 h. In between each step, plates were washed three times with PBS and 0.05% Tween. Endpoint titers were calculated as the dilution that emitted fluorescence exceeding the background tenfold from negative control mice and extrapolated from a linear regression using R.

Olink proteomics

Cell culture supernatant and mouse serum samples were analyzed using the Olink proximity extension assay technology (Olink Proteomics). Samples were processed by the Human Immune Monitoring Center at Mount Sinai using the Target 96 mouse exploratory panel according to the manufacturer’s instructions. Protein abundance is reported as normalized protein expression values on a log₂ scale.

Quantification and statistical analysis

Statistical values including the number of replicates and statistical significance are reported in the figures or figure legends when appropriate. For the majority of in vivo experiments, the experiments were repeated at least two separate times with different cohorts of mice, synthesized RNAs and encapsulated and quantified RNA-LNPs. Statistical analysis was performed using Microsoft Excel, GraphPad Prism 10 (GraphPad Software) or R. Flow cytometry analysis was performed using FlowJo (version 10.10.0). The histology was processed, images were rendered and signal was quantified using QuPath and ImageJ software. The levels of significance (unpaired two-tailed Student’s t-test) are denoted as *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001. Mechanistic schema and experimental schema were made using BioRender and Microsoft PowerPoint.

Reporting summary

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