Elicit: Lipid Nanoparticles in mRNA Vaccine Delivery

Skip to main content

Lipid Nanoparticles in mRNA Vaccine Delivery

Research reportView only

Create alertChat

May 5, 2026

How do lipid nanoparticles deliver nucleoside-modified mRNA for SARS-CoV-2 vaccine antigens?

Lipid nanoparticles deliver nucleoside-modified mRNA for SARS-CoV-2 vaccines by encapsulating the mRNA through electrostatic and hydrogen bonding interactions, facilitating cellular uptake via endocytosis, and releasing the mRNA into the cytoplasm when ionizable lipids protonate in acidic endosomes and disrupt the endosomal membrane to enable translation.

Abstract

Lipid nanoparticles deliver nucleoside-modified mRNA for SARS-CoV-2 vaccines through a coordinated physicochemical and biological process. LNPs composed of ionizable lipids, helper lipids, cholesterol, and PEG-lipids form 50-85 nm particles that encapsulate mRNA via electrostatic interactions and hydrogen bonding. Following intramuscular administration, LNPs are taken up by muscle-resident immune cells through endocytosis, with >10% of myeloid and lymphoid cells successfully transduced. The critical delivery step occurs when ionizable lipids protonate in acidic endosomes (pH ~5-6), triggering ion pairing with endosomal phospholipids that disrupts membranes and releases mRNA into the cytoplasm. Nucleoside modifications such as m1ψ reduce innate immune recognition of the mRNA cargo, while optimized formulations achieve 93-95% encapsulation efficiency and mRNA detectability for 14-28 days post-vaccination in humans.

Beyond mRNA delivery, LNPs function as intrinsic adjuvants by triggering rapid neutrophil and dendritic cell recruitment within 24 hours and inducing IL-6 production critical for germinal center responses. This dual functionality produces potent immunogenicity, with optimized formulations achieving 10-fold higher neutralizing titers than standard LNPs and protective immunity at doses as low as 0.25 µg. Rational lipid design enables independent control of delivery efficiency, tissue targeting, immune activation, and stability: the C24 formulation reduces off-target liver expression 6-fold while maintaining potency, adjuvant lipidoids provide TLR7/8 activation, and specific formulations remain stable at 4°C for at least 19 days. These mechanisms collectively explain the exceptional clinical efficacy of LNP-mRNA SARS-CoV-2 vaccines.

Methods

We analyzed 10 sources from an initial pool of 200, using 6 screening criteria. Each paper was reviewed for 8 key aspects that mattered most to the research question. More on methods

Records from Elicit search

n = 200

Papers screened using: Lipid Nanoparticle Delivery System, Nucleoside-Modified mRNA, SARS-CoV-2 Antigen Target, Delivery Mechanism Data, Study Type, Delivery System Scope

n = 200

Papers screened out

n = 190

Papers included for extraction

n = 10

Press enter or space to select a node.You can then use the arrow keys to move the node around. Press delete to remove it and escape to cancel.

Press enter or space to select an edge. You can then press delete to remove it or escape to cancel.

Paper search

We performed a semantic search across over 138 million academic papers from the Elicit search engine, which includes all of Semantic Scholar and OpenAlex.

We ran this query: “How do lipid nanoparticles deliver nucleoside-modified mRNA for SARS-CoV-2 vaccine antigens?”

The search returned 200 total results from Elicit.

We retrieved 200 papers most relevant to the query for screening.

Screening

We screened in sources based on their abstracts that met these criteria:

We considered all screening questions together and made a holistic judgement about whether to screen in each paper.

Data extraction

We asked a large language model to extract each data column below from each paper. We gave the model the extraction instructions shown below for each column.

Extract complete lipid nanoparticle composition and design details for SARS-CoV-2 mRNA delivery, including:

Extract detailed characteristics of the nucleoside-modified mRNA for SARS-CoV-2 vaccine antigens, including:

Extract mechanistic details of how LNPs deliver mRNA into target cells, including:

Extract quantitative measures of mRNA delivery effectiveness specifically for SARS-CoV-2 vaccines, including:

Extract physical and structural characteristics that enable mRNA delivery, including:

Extract biodistribution data showing where LNPs deliver mRNA after administration, including:

Extract data on how LNPs contribute to immune activation beyond mRNA delivery, including:

Extract factors affecting LNP-mRNA stability and delivery performance over time, including:

Results

Characteristics of Included Studies

Study

Full text retrieved?

Study focus

LNP formulation

Study design

Suman Alishetty et al., 2021

Yes

Novel C24 ionizable lipid development

C24 lipid (48:13:37:2 ratio)

Mouse immunogenicity and challenge studies

Afshana Quadiri et al., 2024

Yes

Effect of helper lipid composition

LNP4 with DOPE phospholipid

Hamster protection studies comparing 4 LNP formulations

Afshana Quadiri et al., 2025

Yes

DOPE-containing LNP characterization

LNP4 with DOPE

Hamster immunogenicity and protection

Xuexiang Han et al., 2023

Yes

Adjuvant lipidoid substitution

C12-TLRa adjuvant lipidoid (35:16:46.5:2.5)

Mouse immunogenicity with TLR7/8 agonist lipid

J. Szebeni et al., 2023

Yes

Structural analysis of Comirnaty vaccine

BNT162b2 with ALC-0315

Physicochemical characterization using AFM, cryo-TEM

S. Kent et al., 2024

No

Human pharmacokinetics

Moderna SPIKEVAX

Clinical study in 19 vaccinated subjects

M. Alameh et al., 2021

Yes

Adjuvant activity mechanisms

Proprietary Acuitas formulation

Mouse studies with influenza and SARS-CoV-2 mRNA

Natalia Martinez et al., 2025

No

Monoolein as helper lipid

MO-based modified Moderna LNPs

In vivo delivery efficiency studies

Dorottya Laczkó et al., 2020

Yes

Nucleoside-modified mRNA immunogenicity

Standard LNP formulation

Mouse single-dose immunogenicity

Devdoot Majumdar et al., 2023

No

Mechanisms of muscle-based adjuvancy

Moderna sm-102 formulation

Mouse mechanistic studies of immune response

toof

Pageof

All 10 studies investigated lipid nanoparticle delivery systems for nucleoside-modified mRNA encoding SARS-CoV-2 spike protein antigens. Seven studies had full text available, while three were abstract-only. The studies employed diverse approaches including novel lipid development, structural characterization, immunogenicity testing in animal models, and clinical pharmacokinetic analysis in humans.

LNP Composition and Formulation Design

LNP formulations for SARS-CoV-2 mRNA delivery consist of four core lipid components: ionizable lipids, helper lipids, cholesterol, and PEG-lipids. The ionizable lipid component is critical for mRNA complexing and pH-dependent release.

Component

Example formulations

Molar ratios

Design rationale

Ionizable lipid

C24, ALC-0315, C12-TLRa

35-48%

Enhanced endosomal protonation; TLR7/8 activation

Helper lipid

DSPC, DOPC, DOPE, Monoolein

13-16%

Structural stability; pH-dependent mesophase transitions

Cholesterol

Standard component

37-46.5%

Structural support

PEG-lipid

DMG-PEG2000

2-2.5%

Immune evasion and circulation

toof

Pageof

The C24 LNP formulation demonstrated a specific composition of 48:13:37:2 for ionizable lipid/DSPC/cholesterol/DMG-PEG2000, formulated in PBS at pH 7.3-7.4. The adjuvant lipidoid formulation used a 35:16:46.5:2.5 ratio and achieved a particle size of approximately 52 nm with low polydispersity (PDI 0.127). Monoolein incorporation as a structural helper lipid induced pH-dependent mesophase transitions promoting inverse hexagonal structures to enhance mRNA release.

LNP morphology varied across formulations. Comirnaty exhibited a granular solid core enclosed by mono- and bilipid layers, with mechanical properties showing soft, compliant structures with stiffness of approximately 9 pN/nm. The particle size ranged from 52 nm to 80-85 nm with polydispersity indices of 0.02-0.06. Surface charge was generally neutral in PBS, though Comirnaty showed a less negative zeta potential of 8.6 ± 5.3. The C24 formulation displayed higher pKa (6.77) compared to MC3 (6.55).

mRNA Cargo Characteristics

The nucleoside-modified mRNA cargo encoded the SARS-CoV-2 spike protein with specific structural modifications. The target antigens included full-length spike protein with deleted furin cleavage site, diproline-stabilized spike protein (S2P), and prefusion-stabilized spike with two proline mutations.

mRNA feature

Specifications

Nucleoside modification

m1ψ (1-methylpseudouridine); Pseudo-U; m1J-5’-triphosphate

Sequence length

3,804 bp; 4,284 nucleotides

Cap structure

Trinucleotide cap1 analog (CleanCap)

Poly-A tail

80-101 adenine residues

Codon optimization

Applied to enhance translation

Encapsulation efficiency

93-95%

toof

Pageof

The nucleoside modifications served to reduce immune recognition and enhance stability. The mRNA sequences were codon-optimized for efficient translation in human cells, with coding sequences identical to FDA-approved vaccines mRNA-1273 and BNT162b2 in some formulations.

Cellular Uptake and Delivery Mechanisms

LNPs deliver mRNA through a multi-step process involving cellular uptake, endosomal escape, and cytoplasmic release. The ionizable lipids play a central role in pH-dependent delivery.

Uptake pathways and cell types:

Muscle-resident myeloid and lymphoid cells were directly transduced, with >10% of these cells successfully receiving mRNA. Lung macrophages, epithelial cells, and cancer cells also took up LNPs. Intramuscular administration of empty LNPs elicited strong neutrophil and dendritic responses within 24 hours, indicating rapid immune cell engagement. Skeletal muscle played a role in the early immune response.

Endosomal escape mechanism:

The cellular uptake occurred via endocytosis. Following endocytosis, ionizable lipids became protonated in the acidic endosomal environment. This protonation triggered ion pairing with endosomal phospholipids to open the endosomal membrane. The C24 formulation demonstrated greater endosomal protonation due to its multistage protonation behavior. Neutralization of LNP charge by anionic lipids inside cells disrupted nanoparticle structure, releasing mRNA.

The adjuvant lipidoid formulation enhanced endosomal escape through interaction with TLR7/8 receptors, with structural characterization showing LNPs initially trapped in endo/lysosomes before escaping into the cytoplasm. Monoolein incorporation promoted pH-dependent mesophase transitions that facilitated enhanced mRNA release.

Structural interactions:

The Comirnaty formulation revealed unique mRNA-lipid interaction patterns. Atomic force microscopy demonstrated that molecular strands corresponding to mRNA could be pulled from nanoparticles with stepwise rupture of mRNA-lipid bonds. The neutral intra-LNP core suggested hydrogen bonding rather than ionic interactions as the primary stabilization mechanism. The ionizable lipid ALC-0315 contained free =O and −OH groups capable of hydrogen bonding with nitrogenous bases in mRNA.

Delivery Efficiency and Protein Expression

Quantitative delivery efficiency varied across formulations, with several showing superior performance compared to standard formulations.

Protein expression levels:

The C24 LNP showed 2-fold higher luciferase expression at injection sites compared to MC3 at both high and low doses. Neutralizing antibody titers were 10-fold higher for C24 versus MC3. The adjuvant lipidoid formulation (C12-113/TLRa) outperformed standard C12-113 in mRNA transfection efficiency both in vitro and in vivo, with protein expression lasting over 14 days at injection sites. Monoolein-based LNPs achieved superior mRNA transfection efficiency compared to original Moderna LNPs across diverse cell types.

Dose-response relationships:

The C24 LNP was effective at very low doses, with protection against SARS-CoV-2 occurring at 0.25 µg prime/boost. The C12-113/TLRa formulation consistently outperformed control formulations at any mRNA dose tested.

Tissue specificity:

Enhanced mRNA transfection occurred at both injection sites and inguinal lymph nodes for the adjuvant formulation. The C24 LNP reduced off-target liver expression 6-fold compared to MC3. Monoolein-based LNPs achieved targeted pulmonary delivery by intranasal administration and spleen delivery by intravenous administration.

Biodistribution Patterns

Following intramuscular administration, LNPs demonstrated specific biodistribution kinetics with varying degrees of systemic exposure.

Distribution timeline in humans:

In a clinical study of 19 subjects receiving Moderna SPIKEVAX, both mRNA and ionizable lipid peaked in blood 1-2 days post-vaccination, with median peak levels of 0.19 ng/mL for mRNA and 3.22 ng/mL for ionizable lipid. The mRNA remained detectable for 14-28 days post-vaccination in most subjects. The decay kinetics of intact mRNA and ionizable lipid were identical, suggesting intact lipid nanoparticles recirculated in blood.

Tissue targeting:

The primary target tissues included skeletal muscle at the injection site and draining lymph nodes. Animal studies showed mRNA expression observed within 6-24 hours at injection sites and lymph nodes, with sustained expression over 14 days. Minimal accumulation occurred in major organs including liver, heart, spleen, lung, and kidney.

Factors controlling distribution:

Rapid increase in surface charge near neutral pH limited systemic biodistribution for C24 formulations. The magnitude of mRNA and ionizable lipid detected in blood correlated with boost in PEG antibodies. The ability of monocytes to phagocytose lipid nanoparticles showed an inverse relationship with PEG antibody rise.

Immune Activation and Adjuvant Properties

Beyond mRNA delivery, LNPs demonstrated intrinsic adjuvant activity contributing to robust immune responses.

Innate immune activation:

LNP formulations triggered strong innate immune responses through multiple mechanisms. The ionizable lipid component induced IL-6 cytokine production, critical for T follicular helper cell differentiation. Empty LNPs alone elicited strong neutrophil and dendritic responses within 24 hours of intramuscular administration. Enhanced dendritic cell maturation occurred with specific formulations, particularly LNP4.

Cytokine profiles:

LNPs induced proinflammatory cytokines including IL-6, GM-CSF, IL-1β, IL-5, IP-10, KC, LIF, Lix, and MIP-2. IL-6 production peaked at 4 hours post-injection and remained elevated for at least 24 hours. Increased secretion of IL-6, IL-21, and IFN-γ indicated robust inflammatory responses. The adjuvant lipidoid formulation increased production of TNF-α, IL-12p70, and IL-1β.

TLR activation:

The C12-TLRa adjuvant lipidoid conferred Toll-like receptor 7/8-agonistic activity to LNPs. This TLR7/8 activation enhanced dendritic cell maturation. MyD88-based signaling enhanced mRNA-LNP vaccine potency but was not required for LNP adjuvanticity. The adjuvant activity relied on the ionizable lipid component and IL-6 induction but not on MyD88- or MAVS-dependent sensing.

Reactogenicity:

Injection site inflammation was notably reduced for C24 compared to MC3. The inflammatory responses were primarily localized to injection sites and draining lymph nodes with minimal systemic inflammation for the adjuvant formulation.

Stability and Storage Considerations

LNP-mRNA stability depended critically on storage temperature and formulation design.

Temperature-dependent stability:

The C24 LNP was entirely stable in bioactivity and mRNA integrity when stored at 4°C for at least 19 days. Bioactivity remained stable for two weeks at 4°C but declined by 20-40% at room temperature. Storage at higher temperatures reduced both bioactivity and mRNA integrity, though less so for C24 than MC3. Standard formulations required frozen storage, typically at -80°C, though some could be stored at refrigerated temperatures for up to 30 days.

Mechanisms of degradation:

mRNA degradation occurred through phosphodiester transesterification reactions, which were temperature-dependent. The mechanism involved base-mediated phosphodiester transesterification. mRNA half-lives were significantly longer at 4°C compared to higher temperatures. The LNP environment affected mRNA degradation due to proton partitioning phenomena.

Formulation stability:

The adjuvant lipidoid formulation showed minimal changes in physicochemical parameters after one month of storage at 4°C. High encapsulation efficiency (>93%) contributed to stability. Nucleoside modifications and LNP formulation enhanced translational capacity and biological stability. The hydrogen bonding between mRNA and lipids in some formulations was weaker than ionic interactions, potentially affecting long-term stability.

Synthesis

The mechanisms by which lipid nanoparticles deliver nucleoside-modified mRNA for SARS-CoV-2 vaccines involve an integrated series of physicochemical and biological processes. The delivery cascade initiates with formulation design, where ionizable lipids complex with negatively charged mRNA through electrostatic and hydrogen bonding interactions. Following intramuscular administration, LNPs interact with muscle cells and immune cells through endocytic uptake, with >10% of muscle-resident myeloid and lymphoid cells successfully transduced.

The critical endosomal escape step relies on pH-dependent protonation of ionizable lipids in acidic endosomes, triggering membrane disruption through ion pairing with endosomal phospholipids. Different ionizable lipids achieve this through distinct mechanisms: the C24 formulation employs multistage protonation, while adjuvant lipidoids enhance escape through TLR7/8 receptor engagement. Upon cytoplasmic release, the nucleoside-modified mRNA is translated by cellular ribosomes to produce spike protein antigen.

Beyond simple mRNA delivery, LNPs function as adjuvants by triggering innate immune responses. Empty LNPs alone elicit neutrophil and dendritic cell recruitment within 24 hours, while the ionizable lipid component induces IL-6 production critical for germinal center formation. This dual functionality—mRNA delivery plus immune activation—underlies the exceptional efficacy of LNP-mRNA SARS-CoV-2 vaccines, with some formulations achieving protection at doses as low as 0.25 µg and neutralizing titers 10-fold higher than conventional formulations.

Human pharmacokinetic data reveals that intact LNPs recirculate in blood for weeks post-vaccination, with mRNA detectable for 14-28 days, supporting sustained antigen expression. The formation of PEG antibodies correlates with the magnitude of circulating mRNA/lipid and is inversely related to monocyte phagocytic clearance, suggesting immune-mediated clearance pathways influence delivery kinetics.

Structural innovations continue to optimize this delivery platform. Replacing phospholipids with monoolein induces pH-dependent mesophase transitions promoting inverse hexagonal structures that enhance mRNA release. The C24 formulation reduces off-target liver expression 6-fold while maintaining potency, and adjuvant lipidoids provide TLR7/8 activation to boost cellular immunity. These advances demonstrate how rational lipid design can independently modulate delivery efficiency, tissue targeting, immune activation, and stability—the four key parameters determining LNP-mRNA vaccine performance.

References

Xuexiang Han, M. Alameh, Kamila Butowska, James J Knox, Kendall A. Lundgreen, and 11 more\ (2023).Adjuvant lipidoid-substituted lipid nanoparticles augment the immunogenicity of SARS-CoV-2 mRNA vaccines. Nature Nanotechnology

J. Szebeni, B. Kiss, Tamás Bozó, K. Turjeman, Y. Levi-Kalisman, and 2 more\ (2023).Insights into the Structure of Comirnaty Covid-19 Vaccine: A Theory on Soft, Partially Bilayer-Covered Nanoparticles with Hydrogen Bond-Stabilized mRNA–Lipid Complexes. ACS Nano

Suman Alishetty, Manuel J. Carrasco, M. Alameh, M. Paige, Hooda Said, and 14 more\ (2021).Novel lipid nanoparticle provides potent SARS-CoV-2 mRNA vaccine at low dose with low local reactogenicity, high thermostability and limited systemic biodistribution

S. Kent, Shiyao Li, T. Amarasena, A. Reynaldi, W. Lee, and 9 more\ (2024).Blood Distribution of SARS-CoV-2 Lipid Nanoparticle mRNA Vaccine in Humans. medRxiv

Afshana Quadiri, S. Prakash, Latifa Zayou, N. Dhanushkodi, Amruth Chilukuri, and 4 more\ (2024).Spike mRNA Vaccine Encapsulated in a Lipid Nanoparticle Composed of Phospholipid 1,2-Dioleoyl-sn-Glycero-3-Phosphoethanolamine Induced Potent B- and T-cell Responses Associated with Protection against COVID-19 in Hamsters. bioRxiv

Afshana Quadiri, S. Prakash, Latifa Zayou, N. Dhanushkodi, Amruth Chilukuri, and 5 more\ (2025).A Spike-Based mRNA Vaccine Encapsulated in Phospholipid 1,2-Dioleoyl-sn-Glycero-3-PhosphoEthanolamine Containing Lipid Nanoparticles Induced Potent B- and T-Cell Responses Associated with Protection Against SARS-CoV-2 Infection and COVID-19-like Symptoms in Hamsters. Vaccines

M. Alameh, I. Tombácz, Emily Bettini, Katlyn Lederer, Sonia Ndeupen, and 29 more\ (2021).Lipid nanoparticles enhance the efficacy of mRNA and protein subunit vaccines by inducing robust T follicular helper cell and humoral responses. Immunity

Dorottya Laczkó, Michael J. Hogan, Sushila A. Toulmin, Philip D. Hicks, Katlyn Lederer, and 37 more\ (2020).A Single Immunization with Nucleoside-Modified mRNA Vaccines Elicits Strong Cellular and Humoral Immune Responses against SARS-CoV-2 in Mice. Immunity

Devdoot Majumdar, W. Dowell, Sylvester Languon, Zachary D. Miller, Jake Dearborn\ (2023).Understanding Mechanisms of Adjuvancy in Muscle by mRNA/Lipid Nanoparticles. Journal of Immunology

Natalia Martinez, Haitao Yu, Joshua Iscaro, Hao Wang, Brendan P. Dyett, and 5 more\ (2025).Structure–Function Correlation of Clinically Inspired Lipid Nanoparticles for Lung and Spleen Targeted mRNA Delivery. Advanced Healthcare Materials

Download BIBDownload RISDownload TXT

Report

Status

Gather sources

200 sources found

Details

Screen sources

10 sources included

Details

Extract data

80 data points extracted

Details

Generate report

Save PDF

BIBLaTeX, ZoteroRISZotero, MendeleyTXTAPA BibliographyPDFPDFDOCXMicrosoft Word

Chat

Got some follow-up questions?

Sign up or sign in to chat with this report.

Back

Adjuvant lipidoid-substituted lipid nanoparticles augment the immunogenicity of SARS-CoV-2 mRNA vaccines

Xuexiang Han, M. Alameh, Kamila Butowska, James J Knox, Kendall A. Lundgreen, Majed Ghattas, N. Gong, Lulu Xue, Ying Xu, M. Lavertu, Paul Bates, Junchao Xu, Guangjun Nie, Yi Zhong, D. Weissman, M. J. Mitchell

Nature Nanotechnology·

2023·

161 citations

SourceDOI

Plain textPDF

Searching for PDF

Unable to find PDF from source

Back

LNP Composition

- Ionizable lipid type: Adjuvant lipidoid C12-TLRa - Concentration: Molar ratio of 35 - Helper lipids: Phospholipid, cholesterol - PEG-lipid type and percentage: DMG-PEG, 2.5% - Novel or modified lipid components: C12-TLRa - Design rationale: Enhance mRNA delivery and adjuvanticity - Particle size and polydispersity: Approximately 52 nm, PDI of 0.127 - Formulation buffer and pH conditions: Phosphate-buffered saline (PBS), pH 7.4

mRNA Specifications

- Target antigen: SARS-CoV-2 spike protein (diproline-modified) - Specific nucleoside modifications: m1ψ (1-methylpseudouridine) - mRNA sequence length and structure: Not mentioned - Cap structure and poly-A tail details: Not mentioned - Codon optimization strategies: Codon-optimized - mRNA-to-lipid ratios in formulation: Not mentioned

Cellular Uptake Mechanism

- Cell types: Dendritic cells (DC2.4 cells) - Uptake pathways: Endocytosis - Endosomal escape mechanisms: Enhanced by adjuvant lipidoid interaction with TLR7/8 receptors - Role of ionizable lipids: pH-dependent release with pKa of 6.42 - Timeline: Rapid process with enhanced mRNA transfection efficiency - Subcellular trafficking patterns: Initial entrapment in endo/lysosomes followed by escape into cytoplasm - Membrane interaction studies: Structural characterization and interaction with cell membranes

Delivery Efficiency

- Protein expression levels: Enhanced mRNA delivery and protein expression with C12-113/TLRa LNP. - Transfection efficiency percentages: C12-113/TLRa LNP outperforms C12-113 LNP in mRNA transfection efficiency. - Duration of protein expression: Over 14 days at the injection site with C12-113/TLRa LNP. - Dose-response relationships: C12-113/TLRa LNP consistently outperforms C12-113 LNP at any mRNA dose. - Comparison with control formulations: C12-113/TLRa LNP vs. C12-113 LNP. - In vitro vs in vivo delivery efficiency: Enhanced mRNA transfection in both in vitro and in vivo settings. - Target tissue specificity: Enhanced mRNA transfection at injection sites and iLNs.

Structural Properties

- LNP morphology: Uniform, compact spherical structure - mRNA-lipid interaction patterns: High encapsulation efficiency (>93%) - Hydrogen bonding or ionic interactions: Neutral surface charge in PBS - Membrane organization: Not explicitly mentioned - Mechanical properties: Not explicitly mentioned - Surface charge and zeta potential: Neutral surface charge in PBS - Cryo-TEM or AFM structural observations: Uniform, compact spherical structure observed by TEM and cryo-EM

Biodistribution Pattern

- Primary target tissues: Injection site and draining lymph nodes - Distribution kinetics and timeline: mRNA expression observed within 6 to 24 hours, sustained over 14 days - Off-target expression levels: Minimal accumulation in major organs (liver, heart, spleen, lung, kidney) - Route of administration effects: Subcutaneous injection for localized delivery - Factors controlling tissue targeting: Route of administration (subcutaneous injection) - Methods used to track biodistribution: In vivo bioluminescence imaging, ex vivo fluorescence and luminescence imaging - Clearance mechanisms and timeline: Not explicitly detailed, but sustained expression suggests controlled release

Inflammatory Effects

- Innate immune responses triggered by LNPs: Enhanced by adjuvant lipidoid substitution with Toll-like receptor 7/8-agonistic activity. - Inflammatory cytokine induction: Increased production of TNF-α, IL-12p70, and IL-1β. - Neutrophil and dendritic cell recruitment: Dendritic cell maturation increased; no specific mention of neutrophil recruitment. - Local vs systemic inflammation: Localized to injection site and draining lymph nodes with minimal systemic inflammation. - Adjuvant effects of lipid components: Enhanced TLR activation pathways. - TLR activation pathways: Activation of TLR7/8. - Injection site reactogenicity: Not specifically mentioned; overall tolerability emphasized. - Timeline of inflammatory responses: Cytokine production analyzed at 6 and 24 hours post-vaccination.

Stability Factors

- Storage temperature requirements: 4 °C - mRNA integrity maintenance: High encapsulation efficiency (>93%) - LNP structure preservation: Minimal changes in physicochemical parameters after 1 month - Bioactivity retention timeline: Over 14 days - Mechanisms of degradation: Not mentioned - Formulation stability enhancers: Nucleoside modification and LNP formulation - Real-time vs accelerated stability data: Not mentioned - Impact on delivery efficiency: Enhanced translational capacity and biological stability

Lipid nanoparticle (LNP)-formulated messenger RNA (mRNA) vaccineare a promising platform to prevent infectious diseases as demonstrated by the recent success of SARS-CoV-2 mRNA vaccines. To avoid immune recognition and uncontrolled inflammation, nucleoside-modified mRNA is used. However, such modification largely abrogates the innate immune responses that are critical to orchestrating robust adaptive immunity. Here we develop an LNP component-an adjuvant lipidoid-that can enhance the adjuvanticity of mRNA-LNP vaccines. Our results show that partial substitution of ionizable lipidoid with adjuvant lipidoid not only enhanced mRNA delivery, but also endowed LNPs with Toll-like receptor 7/8-agonistic activity, which significantly increased the innate immunity of the SARS-CoV-2 mRNA-LNP vaccine with good tolerability in mice. Our optimized vaccine elicits potent neutralizing antibodies against multiple SARS-CoV-2 pseudovirus variants, strong Th1-biased cellular immunity, and robust B cell and long-lived plasma cell responses. Importantly, this adjuvant lipidoid substitution strategy works successfully in a clinically relevant mRNA-LNP vaccine, demonstrating its translational potential.

© 2023. The Author(s), under exclusive licence to Springer Nature Limited.

Lipid nanoparticle (LNP)-formulated messenger RNA (mRNA) vaccineare a promising platform to prevent infectious diseases as demonstrated by the recent success of SARS-CoV-2 mRNA vaccines. To avoid immune recognition and uncontrolled inflammation, nucleoside-modified mRNA is used. However, such modification largely abrogates the innate immune responses that are critical to orchestrating robust adaptive immunity.

Here we develop an LNP component-an adjuvant lipidoid-that can enhance the adjuvanticity of mRNA-LNP vaccines. Our results show that partial substitution of ionizable lipidoid with adjuvant lipidoid not only enhanced mRNA delivery, but also endowed LNPs with Toll-like receptor 7/8-agonistic activity, which significantly increased the innate immunity of the SARS-CoV-2 mRNA-LNP vaccine with good tolerability in mice. Our optimized vaccine elicits potent neutralizing antibodies against multiple SARS-CoV-2 pseudovirus variants, strong Th1-biased cellular immunity, and robust B cell and long-lived plasma cell responses. Importantly, this adjuvant lipidoid substitution strategy works successfully in a clinically relevant mRNA-LNP vaccine, demonstrating its translational potential.

The coronavirus disease 2019 (COVID- 19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has resulted in a global health crisis with over six million casualties thus far (https:// coronavirus.jhu.edu/map.html). Vaccines have proved to be efficacious at reducing the morbidity and mortality of this infectious disease. Currently, several vaccine modalities, including messenger RNA (mRNA) vaccines, antigen-expressing adenovirus, inactivated virus and subunit vaccines, are either approved or under clinical development 1 . Among the different platforms, mRNA vaccines not only demonstrate superior ability to elicit both neutralizing antibody (NAb) and T cell responses against multiple SARS-CoV-2 variants 2 , but also offer low production costs with short development and manufacturing times 3,4 .

Although mRNA was discovered in 1961 (ref. 5), due to its susceptibility to enzymatic degradation, inefficient in vivo delivery and high innate immunogenicity, its use in vaccine development has been hampered until recently 6 . The first two challenges have been overcome by the incorporation of mRNA into a delivery system that can protect it from degradation and escort it across various biological barriers [7][8][9] . Notably, four-component lipid nanoparticles (LNPs) comprising ionizable lipid (or lipidoid), phospholipid, PEGylated lipid and cholesterol are the most clinically advanced mRNA delivery platform as exemplified by Pfizer/BioNTech's BNT162b2 and Moderna's mRNA-1273 vaccines 8 . The last challenge is addressed through the incorporation of naturally occurring nucleosides such as 1-methylpseudouridine (m1ψ) into the Article https://doi.org/10.1038/s41565-023-01404-4 charged mRNA into LNPs. m1ψ-modified firefly luciferase mRNA (mLuc)-encapsulated C12-TLRa LNP was then used to treat DC2.4 dendritic cells. Although free mLuc was unable to transfect cells, C12-TLRa LNP demonstrated dose-dependent mRNA transfection (Fig. 2b ), showing that adjuvant lipidoid can successfully deliver mRNA into cells. To further evaluate the potency of C12-TLRa, it was compared with a series of polyamine-derived lipidoids synthesized using the same chemistry (Fig. 2c and Supplementary Fig. 5 ) 25 . It should be noted that some polyamine-derived lipidoids in this library exhibited potent RNA delivery in previous studies 25,31 . In vitro mLuc transfection results demonstrated that C12-TLRa had a relatively low transfection ability, especially when compared with the top performer C12-113 (Fig. 2d ).

Optimization and characterization of LNPs

Since adequate antigen expression is favourable for mRNA vaccines, we decided to partially replace the ionizable lipidoid in C12-113 LNP with C12-TLRa to endow the LNPs with TLR7/8-agonistic properties without compromising mRNA delivery. Therefore, a series of C12-113/TLRa LNP with an increased C12-TLRa substitution (1-17.5 mol%) was formulated and subjected to in vitro screening. Intererestingly, along with the increased substitution with C12-TLRa, the mRNA transfection efficiency of C12-113/TLRa LNP first increased and then decreased (Fig. 2e ). The highest transfection was achieved by C12-113/TLRa LNP with 5 mol% of C12-TLRa, and this was chosen for subsequent studies.

The adjuvant activity of mRNA-loaded C12-113/TLRa LNP was then verified in HEK-Blue reporter cells. Whereas C12-113 LNP possessed no TLR7-agonistic activity, C12-113/TLRa LNP exhibited dose-dependent TLR7-agonistic activity (Fig. 2f ). Consequently, C12-113/TLRa LNP stimulated proinflammatory cytokine TNF-α production in DC2.4 cells in a dose-dependent manner, which significantly outperformed C12-113 LNP (Supplementary Fig. 6 ). To further confirm the general applicability of this strategy, DLin-MC3-DMA (MC3) LNPs and adjuvant lipidoid-substituted MC3 LNP (that is, MC3/TLRa LNP), and SM-102 LNP and adjuvant lipidoid-substituted SM-102 LNP (that is, SM-102/TLRa LNP) were formulated and tested (Supplementary Table 1 ). Consistently, MC3/TLRa LNP and SM-102/TLRa LNP, but not MC3 LNP or SM-102 LNP, showed obvious TLR7-agonistic activity (Supplementary Fig. 7 ). These results suggest that the substitution of partial lipidoid with C12-TLRa can endow LNPs with adjuvant activity irrespective of LNP formulation.

Next, the physicochemical properties of the C12-113/TLRa LNP were characterized. This LNP had a hydrodynamic diameter of approximately 52 nm and a polydispersity index (PDI) of 0.127 (Supplementary Table 1 ). Additionally, it had a neutral surface charge in phosphate-buffered saline (PBS, pH 7.4), and its apparent pK a was determined to be 6.42 by conducting a 6-(p-toluidinyl) naphthalene-2-sulfonic acid (TNS) assay over the range of pH 2-12 (ref. 33), which meet the criteria for potent in vivo mRNA delivery 34,35 . The encapsulation efficiency of mRNA in C12-113/TLRa LNP was >93%. After 1 month of storage at 4 °C, we observed minimal changes in all major LNP physicochemical parameters (Supplementary Fig. 8 ), indicating the good stability of this LNP. Representative transmission electron microscopy (TEM) and cryogenic electron microscopy (cryo-EM) images showed that C12-113/TLRa LNP had a uniform, compact spherical structure (Fig. 2g and Supplementary Fig. 9a ). C12-113 LNP without C12-TLRa substitution had very similar parameters and morphology (Supplementary Table 1 , and Supplementary Figs. 8 and 9 ), suggesting that substitution with 5% C12-TLRa does not alter the physicochemical properties or affect the stability of these LNPs.

In vitro mRNA delivery and DC activation

We next systematically investigated the in vitro mRNA delivery and adjuvant activity of C12-113/TLRa LNP. Both C12-113 LNP and C12-113/ TLRa LNP exhibited dose-dependent mLuc transfection in DC2.4 cells (Fig. 3a ); however, in line with Fig. 2e , C12-113/TLRa LNP consistently outperformed C12-113 LNP at any mRNA dose tested. Similarly, in in vitro transcribed mRNA sequence to avoid its recognition by pattern recognition receptors 10,11 . Despite its improved tolerability and translation, the use of nucleoside-modified mRNA largely compromises the innate immune responses and weakens the activation of dendritic cells (DCs, the major recipient cells of mRNA-LNP vaccines 12 ), a process that is critical to mounting a strong adaptive immunity to vaccination 3,13 .

Interestingly, we and others have found that certain LNPs possess intrinsic adjuvant activity that can promote the induction of cellular and humoral immune responses of mRNA and subunit vaccines [14][15][16] . However, the adjuvanticity of LNPs is variable based on the lipidoid used and is generally weak and non-specific 14 . Toll-like receptor agonists (TLRas) are potent adjuvants, which activate DCs to present antigen, express co-stimulatory molecules and produce selective cytokines (for example, tumour necrosis factor-α (TNF-α)) that effectively stimulate the transition from innate to adaptive immunity 17 . Several small-molecule TLRas with high specificity have been approved for immunomodulation 18 , including imidazoquinoline-based TLR7/8 agonists (for example, imiquimod) 19 . The incorporation of fatty acid-modified TLRas in antigen-loaded liposomes or NPs has been demonstrated to enhance the magnitude and duration of adaptive immune responses 20,21 . We hypothesized that formulation of a rationally designed TLRa-derived ionizable lipidoid into LNPs could further augment the immunogenicity of mRNA-LNP vaccines due to the accessorial adjuvant properties.

In this study, we chemically synthesized an LNP component-an adjuvant lipidoid-based on a TLR7/8 agonist. Unlike previous studies in which the TLRa was typically conjugated to macromolecules (for example, lipids, peptides and polymers) through acylation [22][23][24] , we applied the ring-opening reaction between an amine-bearing TLRa and an alkyl epoxide to yield a tertiary amine-containing ionizable lipidoid that can complex with and deliver negatively charged mRNA into cells 25 . We optimized the formulation of this adjuvant lipidoid-incorporated SARS-CoV-2 mRNA-LNP vaccine and systematically evaluated its innate and adaptive immune responses (Fig. 1 ).

Synthesis of adjuvant lipidoid and polyamine-based lipidoids

Adjuvant lipidoid C12-TLRa was synthesized by the ring-opening reaction between amine-containing TLR7/8 agonist 1 and C12 epoxide (Fig. 2a ) 25 . The purity and structure of C12-TLRa were verified by liquid chromatography-mass spectrometry and proton NMR (Supplementary Figs. 1 and 2 ). Structural simulation via computational analysis demonstrated that C12-TLRa could form multiple interactions with TLR7 receptor (for example, hydrogen bonding with Asp 555 and Thr 586 and π-π stacking with Phe 408 ) at its first binding site that were similar to TLR7/8 agonist 1 (Supplementary Fig. 3 ) 26,27 , suggesting that such modification barely interrupts the agonist-receptor interaction. The TLR7-agonistic activity of C12-TLRa was further verified on HEK-Blue reporter cells stably expressing the TLR7 receptor. As a positive control, TLR7/8 agonist 1 exhibited a biphasic dose-response profile of TLR7-agonistic activity characterized by an initial dose-dependent increase followed by a decrease of analytes (Supplementary Fig. 4 ), which was also observed by others 28,29 . However, monotonically dose-dependent TLR7-agonistic activity of C12-TLRa was observed at the same dose range, indicating that C12-TLRa has a more predictable pharmaceutical property despite its reduced potency.

Next, adjuvant lipidoid, phospholipid, cholesterol and 1,2-dimyristoyl-rac-glycero-3-methoxypolyethylene glycol-2000 (DMG-PEG) at a molar ratio of 35:16:46.5:2.5 were formulated into C12-TLRa LNP along with mRNA using microfluidic mixing (Supplementary Table 1 ) 30 . In this formulation, 1,2-dioleoyl-sn-glycero-3-p hosphoethanolamine was used as it has generally exhibited greater mRNA delivery than other phospholipids such as 1,2-distearoyl-sn-glycero-3-phosphocholine 31,32 . The mRNA encapsulation efficiency was determined to be ~75%, suggesting that this adjuvant lipidoid is able to complex with and encapsulate negatively Article https://doi.org/10.1038/s41565-023-01404-4 primary murine bone marrow-derived dendritic cells (BMDCs) and human monocyte-derived dendritic cells (MoDCs), C12-113/TLRa LNP consistently exhibited higher mRNA transfection efficiency than C12-113 LNP (Fig. 3b, c ). No obvious cytotoxicity towards these cells was observed at the dose tested for both LNPs (Supplementary Fig. 10 ). Moreover, MC3/TLRa LNP and SM-102/TLRa LNP significantly outperformed their respective counterpart in terms of mLuc delivery (Supplementary Fig. 11 ). These results suggest that C12-TLRa substitution can increase LNP-mediated mRNA delivery in vitro irrespective of LNP formulation.

To investigate the mechanism of enhanced mRNA transfection following C12-TLRa incorporation, we studied the endosomal escape of LNPs. Lipid-like fluorescent dye 3,3′-dioctadecyloxacarbocyanine perchlorate (DiO)-labelled LNPs were used to treat DC2.4 cells and their subcellular distribution was visualized after endo/lysosomes and nuclei were stained (Fig. 3d ). In C12-113 LNP-treated cells, there were large yellow spots, indicating severe entrapment of C12-113 LNP (green) inside endo/lysosomes (red). In contrast, C12-113/TLRa LNP-treated cells exhibited more diffuse green signal across the cytoplasm, suggesting its more efficient escape from endo/lysosomes. This observation was further supported by the quantification of LNP endosomal escape. Additionally, similar results were obtained when LNPs encapsulating Cy5-tagged mRNA were used for this endosomal escape study (Supplementary Fig. 12 ). These results suggest that C12-TLRa substitution enhances endosomal escape of LNPs, which could explain the enhanced mRNA transfection (Fig. 3a-c ). Since multiple interactions are predicted between C12-TLRa and TLR7 based on the molecular simulation result (Supplementary Fig. 3 ), it is reasonable to assume that the strong affinity between the incorporated C12-TLRa and endosomal TLR7/8 could reinforce the physical interaction between LNPs and the endosomal membrane, which leads to enhanced endosomal escape (Fig. 3e ).

Next, we studied the immunostimulatory effect of C12-113/ TLRa LNP carrying m1ψ-modified SARS-CoV-2 mRNA encoding the diproline-modified spike glycoprotein. Notably, the coding sequence of this nucleoside-modified mRNA is identical to the mRNA used in two US Food and Drug Administration-approved vaccines (mRNA-1273 and BNT162b2) 36 . Since TLR7/8 agonist can stimulate DC maturation and proinflammatory cytokine release, we analysed matured DCs (CD80 + CD86 + ) using flow cytometry and measured TNF-α, interleukin (IL)-12p70 and IL-1β secretion using enzyme-linked immunosorbent assay (ELISA) at 24 h post-transfection. As expected, C12-113/TLRa LNP markedly increased the percentage of matured DCs in DC2.4 cells, BMDCs and MoDCs (Fig. 3f -h and Supplementary Fig. 13 ). In contrast, C12-113 LNP only modestly induced DC maturation. Moreover, C12-113/ TLRa LNP stimulated a significant increase of TNF-α, IL-12p70 and IL-1β levels compared to C12-113 LNP in DC2.4 cells, BMDCs and MoDCs (Fig. 3i-k ). Altogether, these results suggest that C12-TLRa substitution can greatly enhance the adjuvanticity of LNPs and DC activation.

In vivo mRNA transfection and innate immune responses

We next explored whether increased mRNA transfection and innate immune responses by adjuvant lipidoid-substituted LNPs could be reproduced in vivo. C57BL/6 mice were subcutaneously (s.c.) immunized at the tail bases of mice with mLuc-loaded C12-113 LNP or C12-113/ TLRa LNP, and luciferase expression was visualized by in vivo bioluminescence imaging at 6 and 24 h post-treatment (Fig. 4a ). Both LNPs mainly transfected injection sites, and could also transfect inguinal lymph nodes (iLNs). In line with in vitro transfection results, C12-113/ TLRa LNP achieved greater mRNA transfection than C12-113 LNP at both injection sites and iLNs (Fig. 4a ).

To avoid systemic toxicity, it is important to spatially restrict adjuvant activity and mRNA transfection to the sites of vaccine administration and draining lymph nodes 37 . Therefore, we studied the distribution and transfection of LNPs in major organs and iLNs at 24 h post-injection of lipid-like fluorescent dye 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindot ricarbocyanine iodide (DiR)-labelled, mLuc-loaded LNPs. Ex vivo fluorescence and luminescence imaging confirmed that no obvious accumulation and transfection in major organs (liver, heart, spleen, lung and kidney) were observed for either LNP (Fig. 4b ). Interestingly, although these two LNPs achieved comparable accumulation amounts in iLNs based on the fluorescence quantification (Fig. 4c ), C12-113/TLRa LNP enabled stronger mRNA transfection in iLNs than C12-113 LNP, which was in line with in vivo ventral luminescence imaging (Fig. 4a ). These results also correlate well with in vitro mRNA transfection (Fig. 3a-c ). Moreover, C12-113/TLRa LNP-mediated enhanced mRNA expression at both the injection site and iLNs was highly reproducible from different LNP batches (Supplementary Fig. 14 ). Finally, the duration and translational kinetics of LNP-formulated mRNA was investigated. C12-113/ TLRa LNP enabled more durable (over 14 days) and consistently stronger mRNA expression at the injection site than C12-113 LNP (Supplementary Fig. 15 ). Such long-lasting expression is attributed to the enhanced translational capacity and biological stability of mRNA as a result of nucleoside modification and LNP formulation 38 . Together, these results suggest that C12-TLRa substitution can greatly enhance the expression of LNP-formulated mRNA at both injection sites and draining lymph nodes with minimal systemic off-target distribution or expression.

To assess in vivo innate immune responses, iLNs were harvested from mice at 24 h post-immunization with SARS-CoV-2 mRNA-loaded LNPs, and intralymphatic DC maturation was analysed by flow cytometry. C12-113/TLRa LNP significantly increased the percentage of matured DCs compared with C12-113 LNP (Fig. 4d ,e and Supplementary Fig. 16 ). Next, iLNs and serum cytokines (TNF-α, IL-12p70 and IL-1β) were analysed by ELISA at 6 and 24 h post-vaccination. While (RLU per viability) C12-113 LNP moderately and transiently induced intralymphatic cytokine production, C12-113/TLRa LNP elicited higher-magnitude and more persistent cytokine responses (Fig. 4f-h ). As expected, due to the minimal systemic exposure of LNPs (Fig. 4b, c ), these serum cytokine levels did not increase for either LNP (Supplementary Fig. 17 ). To further assess and compare the tolerability of these two mRNA-LNP vaccines, we evaluated the serum levels of acute phase protein, haptoglobin, and two inflammatory chemokines, interferon-γ-induced protein 10 (IP-10) and monocyte chemoattractant protein 1 (MCP-1), which have been demonstrated to be sensitive to s.c. injected mRNA-LNPs 39 . Both LNPs induced transient and comparable elevations of haptoglobin, IP-10 and MCP-1 levels (Supplementary Fig. 18 ). These results demonstrate that C12-TLRa substitution can greatly enhance the magnitude and duration of local innate immune responses without exacerbating systemic inflammation triggered by the pristine SARS-CoV-2 mRNA-LNP vaccine.

Adjuvant lipidoid substitution enhances cellular responses

We further investigated the adaptive immune responses elicited by original and adjuvant lipidoid-substituted SARS-CoV-2 mRNA-LNP vaccines. Mice were vaccinated twice, 3 weeks apart, using a prime-boost strategy (Fig. 5a ). Of note, neither skin abnormalities at the injection sites (Supplementary Fig. 19 ) nor body weight loss was observed after immunization (Supplementary Fig. 20 ). Two weeks after the boost dose, spleens from immunized mice were collected, and splenocytes were stimulated with SARS-CoV-2 receptor binding domain (RBD) peptide pools. Antigen-specific CD4 + and CD8 + T cells were measured by flow cytometry (Fig. 5b ,c and Supplementary Fig. 21 ). Both LNPs elicited RBD-specific CD4 + T cells expressing Th1 cytokines (interferon (IFN)-γ, IL-2 and TNF-α); however, the effect elicited by C12-113/TLRa LNP was significantly higher than that elicited by C12-113 LNP (Fig. 5b ). Moreover, neither vaccine formulation induced CD4 + T cells expressing type 2 (Th2) cytokines (IL-4, IL-5 and IL-17), supporting a Th1-biased T cell immune response. C12-113/TLRa LNP also elicited more RBD-specific CD8 + T cells expressing Th1 immune response cytokines (IFN-γ, IL-2 and TNF-α) and cytotoxic marker (CD107α) (Fig. 5c ). These results suggest that C12-TLRa-substituted SARS-CoV-2 mRNA-LNP vaccines can induce stronger RBD-specific Th1 and CD8 + T cell immune responses. Polyfunctional T cells are generally considered immune correlates for protection against pathogens [40][41][42] . Next we examined the polyfunctionality of these RBD-specific T cell responses. C12-113/ TLRa LNP induced a significantly higher proportion of double-and triple-positive CD4 + and CD8 + T cells compared with the C12-113 LNP (Fig. 5d ). The pattern of cytokine secretion of the double-positive cells was different between CD4 + (IL-2 + TNF-α + ) and CD8 + (INF-γ + TNF-α + ) cells and consistent with the role of the helper and cytotoxic activities of these two different lymphocyte populations.

Adjuvant lipidoid substitution enhances humoral responses

Next, SARS-CoV-2 mRNA-LNP vaccine-induced humoral immune responses were analysed (Fig. 6a ). Serum from vaccinated mice was subjected to endpoint dilution ELISA to determine the total RBD-specific IgG titres. Both mRNA-LNP vaccines triggered high levels of RBD-specific IgG (Fig. 6b ); however, the IgG titre for C12-113/ TLRa LNP was almost one order of magnitude higher than that for C12-113 LNP (1.22 × 10 7 versus 1.35 × 10 6 ). Moreover, the ELISA analysis of RBD-specific IgG isotypes showed that C12-113/TLRa LNP elicited similar IgG1 titre and higher IgG2c titre with a higher IgG2c/IgG1 ratio compared with C12-113 LNP (Supplementary Fig. 22 ), further highlighting that adjuvant lipidoid-substituted LNPs tend to trigger stronger Th1-biased immune responses. Next, NAb levels were measured using a vesicular stomatitis virus (VSV)-based pseudovirus neutralization assay (Fig. 6c ). While both mRNA-LNP vaccines induced high levels of NAbs against multiple SARS-CoV-2 pseudovirus variants, including D614G-mutated ancestor strain, Beta (B.1.351) variant and Delta (B.1.617.2) variant, C12-113/TLRa LNP elicited more NAbs against all three pseudoviruses as demonstrated by the higher 50% foci reduction neutralization titres (FRNT 50 ).

To investigate the induction of B cell memory, we assessed splenic B cell responses using fluorescent RBD probes (Supplementary Fig. 23 ) 14,43 . C12-113/TLRa LNP vaccination significantly increased the number of RBD-specific B cells compared with C12-113 LNP (Fig. 6d ), and these cells displayed a memory phenotype (CD38 + GL7 -; Fig. 6e ). The majority of RBD-specific B cells in C12-113/TLRa LNP-vaccinated mice co-expressed the memory markers PD-L2 and CD80 (Fig. 6e ), consistent with a potential for rapid differentiation into antibody-secreting cells (ASCs) upon rechallenge 44 . Thus, C12-113/TLRa LNP vaccination generates a larger but phenotypically similar RBD-specific memory B cell pool.

Finally, we analysed bone marrow-resident long-lived plasma cells (LLPCs) 3 months after the second vaccination, which can mediate durable protection from infection by persistently producing antigen-specific antibodies. Bone marrow from immunized mice was collected, and various subsets of RBD-specific ASCs were characterized by enzyme-linked immunospot assay (ELISPOT). Whereas C12-113 LNP modestly induced the generation of RBD-specific IgG1-, IgG2a-and IgG2b-expressing ASCs (Fig. 6f-h ), C12-113/TLRa LNP induced 6.7-, 2.4-and 6.5-fold more IgG1-, IgG2a-and IgG2b-expressing ASCs, respectively. Altogether, these results suggest that adjuvant lipidoid-substituted mRNA-LNP vaccine elicits stronger humoral immune responses and LLPC responses.

Enhanced immunogenicity of clinically relevant mRNA vaccines

To verify if adjuvant lipidoid substitution is a generally applicable strategy to enhance the adaptive immune responses of mRNA-LNP vaccines, we further chose the approved SM-102 LNP formulation for investigation (Supplementary Table 1 ). In a pilot study, we noticed that SM-102/ TLRa LNP achieved greater mRNA transfection than SM-102 LNP after local injection as well, and both of them significantly outperformed C12-113 LNP or C12-113/TLRa LNP (Supplementary Fig. 24 ). To avoid the excessive expression of antigen, mice were vaccinated twice with SM-102 LNP or SM-102/TLRa LNP at a lower dose that achieved comparable protein expression as that in the previous study (Extended Data Fig. 1a and Supplementary Fig. 24 ). No skin abnormalities or body weight loss was observed during the experiment (Supplementary Figs. 25 and 26 ). Flow cytometry analysis of splenocytes from immunized mice showed that SM-102/TLRa LNP elicited a significantly larger fraction of RBD-specific CD4 + and CD8 + T cells expressing Th1 cytokines (IFN-γ, IL-2 and TNF-α) compared with SM-102 LNP (Extended Data Fig. 1b ,c and Supplementary Fig. 27 ). Moreover, more RBD-specific cytotoxic CD8 + T cells were induced by SM-102/TLRa LNP (Extended Data Fig. 1c ). Further examination of these RBD-specific T cells showed that SM-102/TLRa LNP elicited a much higher proportion of doubleand triple-positive CD4 + and CD8 + T cells in comparison with SM-102 LNP (Extended Data Fig. 1d ). Together, these results suggest that C12-TLRa substitution can enhance the Th1-skewed, antigen-specific T cell immune responses of clinically relevant mRNA-LNP vaccines. Next, humoral immune responses in immunized mice were studied (Extended Data Fig. 2a ). SM-102/TLRa LNP achieved 5.4-fold higher anti-RBD IgG titre than SM-102 LNP (2.33 × 10 6 versus 4.30 × 10 5 ; Extended Data Fig. 2b ). Moreover, SM-102/TLRa LNP-vaccinated mice generated more NAbs against all three SARS-CoV-2 pseudoviruses (D614G-mutated ancestor strain, Beta variant and Delta variant; Extended Data Fig. 2c ). Flow cytometry analysis of splenocytes showed that SM-102/TLRa LNP elicited more RBD-specific B cells than SM-102 LNP, and these cells typically exhibited a memory phenotype (CD38 + GL7 -; Extended Data Fig. 2d, e ). Moreover, most RBD-specific B cells in SM-102/TLRa LNP-vaccinated mice were PD-L2 and CD80 double positive, and the proportion of these cells was larger than that in SM-102 LNP-vaccinated mice (Extended Data Fig. 2e ). These results demonstrate that C12-TLRa substitution can increase antigen-specific antibody responses and B cell responses of clinically relevant mRNA-LNP vaccines.

Together, all the above results confirm that adjuvant lipidoid substitution can enhance the immunogenicity of clinically relevant SARS-CoV-2 mRNA-LNP vaccines, which holds translational potential.

Conclusions

We developed a type of ionizable lipidoid-an adjuvant lipidoid-which serves not only as a structural component of LNPs to enhance mRNA delivery, but also as a functional moiety to increase the adjuvanticity of the LNPs. Adjuvant lipidoid-substituted SARS-CoV-2 mRNA-LNP vaccine significantly enhanced intralymphatic DC activation and intranodal proinflammatory cytokine production with good tolerability in mice due to its exclusive distribution and transfection at the injection site and draining lymph nodes. Due to the improved innate immunity, a more potent adaptive immunity against SARS-CoV-2 was established after immunization with adjuvant lipidoid-substituted mRNA-LNP vaccine, including strong Th1-biased cellular immune responses, antibody responses, B cell responses and LLPC responses. Moreover, we confirmed that adjuvant lipidoid substitution could enhance the immunogenicity of clinically relevant mRNA-LNP vaccines. Our study provides a LNP formulation basis for introducing adjuvant lipidoid to improve the quality and magnitude of adaptive immune responses for mRNA vaccines.

Online content

Any methods, additional references, Nature Portfolio reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at https://doi.org/10.1038/s41565-023-01404-4.

https://doi.org/10.1038/s41565-023-01404-4

Materials

TLR7/8 agonist 1 dihydrochloride was purchased from Cayman Chemical. 1,2-Epoxydodecane (C12) was obtained from Sigma-Aldrich. Core 200 was customized from Enamine, and other polyamine cores were purchased from Sigma-Aldrich and TCI. Anti-mouse CD16/32 antibody, APC anti-mouse CD11c antibody, FITC anti-mouse CD80 antibody, PE anti-mouse CD86 antibody, APC anti-human CD11c antibody, FITC anti-human CD80 antibody and PE anti-human CD86 antibody were purchased from Biolegend. Mouse IL-1β uncoated ELISA, mouse IL-12p70 uncoated ELISA, mouse TNF-α uncoated ELISA, mouse MCP-1 uncoated ELISA, human IL-1β uncoated ELISA, human IL-12p70 uncoated ELISA, human TNF-α uncoated ELISA, LysoTracker Deep Red, LysoTracker Green, DiO DiR were bought from Invitrogen. Mouse haptoglobin ELISA and mouse IP-10 ELISA were obtained from Abcam. 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine, 1,2-distearoyl-sn-glycero-3-phosphocholine, DMG-PEG and cholesterol were obtained from Avanti Polar Lipids. DLin-MC3-DMA and SM-102 were purchased from MedChem Express. Codon-optimized m1ψ-modified luciferase mRNA and SARS-CoV-2 diproline-modified spike (S2P) mRNA were produced by in vitro transcription 14 . Cy5-tagged luciferase mRNA was produced in house by incorporating Cy5-UTP (TriLink) into the in vitro transcription reaction.

Synthesis of adjuvant lipidoid

Adjuvant lipidoid C12-TLRa was synthesized by reacting epoxydodecane (C12) with TLR7/8 agonist 1 dihydrochloride using the ring-opening reaction 25 . Briefly, 10 mg of TLR7/8 agonist 1 dihydrochloride was dissolved in 0.8 ml of ethanol in a glass vial with a magnetic stir bar. Then, 8 μl of triethylamine was added to neutralize the hydrochloride before adding 20 mg of C12. The vial was sealed, and the mixture was stirred for 48 h at 80 °C. The crude product was purified by a CombiFlash NextGen 300+ chromatography system (Teledyne ISCO) with gradient elution from CH 2 Cl 2 to 75:22:3 CH 2 Cl 2 /MeOH/ NH 4 OH (aq.). The desired fraction was collected (yield, 44%). C12-TLRa was characterized by mass spectrometry (calculated MS, 728.12; found [M + 2H] 2+ , 365.25) and NMR spectroscopy. 1 H NMR (400 MHz, DMSO-d 6 ) δ (ppm): 7.78 (d, J = 8.3 Hz, 1H), 7.57 (doublet of doublets, J = 8.4, 1.3 Hz, 1H), 7.35-7.29 (m, 1H), 7.27 (d, J = 7.9 Hz, 2H), 7.05-7.00 (m, 1H), 6.98 (d, J = 8.0 Hz, 2H), 5.84 (s, 2H), 3.61-3.47 (m, 2H), 3.44 (s, 2H), 2.90 (t, J = 7.7 Hz, 2H), 2.68 (q, J = 1.9 Hz, 2H), 2.34 (t, J = 2.8 Hz, 2H), 1.69 (quintet, J = 7.6 Hz, 2H), 1.37 (doublet of triplets, J = 14.9, 7.5 Hz, 2H), 1.22 (s, 36H), 0.90-0.81 (m, 9H).

General method for the synthesis of polyamine-derived lipidoids

The polyamine cores were reacted with excess moles of C12 as needed to saturate the amines 25,33 . Taking C12-113 as an example, 113 core (1 equiv.) was mixed with C12 (4.8 equiv.) for 48 h at 80 °C in a neat condition. The crude product was used for the initial library screening. To purify the top-performing C12-113 lipidoid, the crude product was separated as described above, and the fully saturated product was collected and identified by mass spectrometry (calculated MS, 854.49; found [M + 2H] 2+ , 429.13) and used for subsequent experiments.

Structural simulation of agonist-TLR7 interaction

The structures of TLR7/8 agonist 1 and C12-TLRa were first optimized by molecular dynamics simulation with the CHARMm force field 45 . The exact TLR7 protein crystal structure was derived from the structure of the TLR7/R848 complex (PDB ID, 5GMH), removing any ligands or solvent molecules 26 . Structural simulation between TLR7 dimer and agonists was carried out by CDocker docking simulation 46 and in situ structural superimposition. Potential non-covalent interactions, binding pockets and overviews of the binding sites between TLR7 dimer and the corresponding agonists were generated using BIOVIA Discovery Studio 2018.

LNP formulation

LNPs were formulated by microfluidic mixing 30 . Briefly, an ethanol phase containing lipidoid (with or without C12-TLRa substitution), phospholipid, cholesterol and DMG-PEG at a designated molar ratio (Supplementary Table 1 ) was mixed with an aqueous phase (10 mM citrate buffer, pH 3) containing mRNA at a flow rate ratio of 1:3 and at a lipidoid/RNA weight ratio of 10:1 in a microfluidic chip device. LNPs were dialysed against 1× PBS in a 20 kDa molecular weight cut-off cassette for 2 h, sterilized through a 0.22 μm filter and stored at 4 °C. DiO-or DIR-labelled LNPs were obtained by mixing DiO or DiR (1 mol% of total lipids) with LNPs before dialysis.

LNP characterization

The hydrodynamic size, PDI and zeta potential of LNPs were measured using a Zetasizer Nano ZS90 (Malvern Instruments). The morphology of LNPs was characterized by TEM ( JEOL 1010) and cryo-EM (Titan Krios, Thermo Fisher) with a K3 Bioquantum (Gatan). The mRNA encapsulation efficiency and the pK a of LNPs were determined using a modified Quant-iT RiboGreen RNA assay (Invitrogen) and a 6-(p-toluidinyl)naphthalene-2-sulfonic acid assay, respectively 30,33 . LNP formulations were routinely examined by the Limulus amebocyte lysate (LAL) test, and endotoxin levels were consistently found to be <1 endotoxin unit per ml.

Cell culture and animal studies

The HEK-Blue mTLR7 cell line was kindly provided by J. Shi at Harvard Medical School, who obtained it from InvivoGen (#hkb-mtlr7). These cells were maintained according to vendor's instruction. Murine macrophage DC2.4 cell line was obtained from American Type Culture Collection and maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% foetal bovine serum, 100 U ml -1 penicillin and 100 μg ml -1 streptomycin. All cells were cultured at 37 °C in a humidified incubator of 5% CO 2 , and routinely tested for mycoplasma contamination.

BMDCs were generated from C57BL/6 mice. Briefly, bone marrow cells were flushed from mouse femurs and tibias, lysed by ammonium-chloride-potassium (ACK) buffer to remove red blood cells and then cultured in RPMI 1640 medium supplemented with 10% foetal bovine serum, 100 U ml -1 penicillin and 100 μg ml -1 streptomycin, 1% HEPES, 0.1 mM β-mercaptoethanol, 20 ng ml -1 murine IL-4 (#214-14, PeproTech) and 20 ng ml -1 murine granulocyte-macrophage colony stimulating factor (#315-03, PeproTech). On day 6, non-adherent and loosely adherent cells were collected for studies.

MoDCs were generated from a 38-year-old healthy male volunteer donor. Monocytes were isolated from donated apheresis blood using the RosetteSep Human Monocyte Enrichment Cocktail (#15068, Stemcell Technologies), and provided by the Human Immunology Core at the University of Pennsylvania. These cells were induced into MoDCs by culturing in complete RPMI medium supplemented with 20 ng ml -1 human IL-4 (#574002, Biolegend) and 20 ng ml -1 human granulocyte-macrophage colony stimulating factor (#572902, Biolegend) for 6 days. This study was approved by the Institutional Review Board of the University of Pennsylvania (#705906). Informed consent was obtained from the donor, who was compensated for this blood donation.

All animal protocols were approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania (#806540), and animal procedures were performed in accordance with the Guidelines for Care and Use of Laboratory Animals at the University of Pennsylvania. C57BL/6 female mice (6-8 weeks of age, 18-20 g body weight) were purchased from Jackson Laboratory.

In vitro mLuc delivery

DC cells, BMDCs or MoDCs were seeded onto a 96-well plate at a density of 10,000 per well overnight and then mLuc-loaded LNPs were used to treat cells at the indicated doses for 24 h. Luciferase expression was https://doi.org/10.1038/s41565-023-01404-4 evaluated by Luciferase Reporter 1000 Assay System (E4550, Promega), and cell viability was measured using a CellTiter-Glo Luminescent Cell Viability Assay (G7572, Promega) according to the manufacturer's protocols. The relative luciferase expression was reported as relative light units normalized to cell viability. Free mRNA was used as a control.

TLR7 reporter assay

The TLR7-agonistic activity of C12-TLRa was tested on HEK-Blue mTLR7 reporter cells using a HEK-Blue Detection Kit (#hb-det2, InvivoGen) according to the manufacturer's instructions. Briefly, HEK-Blue mTLR7 reporter cells were seeded into a 96-well plate at a density of 40,000 cells per well in HEK-Blue Detection medium containing different concentrations of C12-TLRa. After incubation for 24 h, the absorbance at 650 nm was measured using a plate reader (Infinite M200, Tecan) and data were normalized to untreated cells. TLR7/8 agonist 1 was used as a positive control. Similarly, the TLR7-agonistic activity of LNPs was measured.

Cellular uptake

DC2.4 cells were seeded into 35 mm glass-bottom dishes for 24 h and then treated with DiO-labelled C12-113 LNP or DiO-labelled C12-113/ TLRa LNP at an mRNA concentration of 500 ng ml -1 for 2 h. Cells were sequentially stained with LysoTracker Deep Red (100 nM) for 30 min and Hoechst 33342 (10 μg ml -1 ) for 5 min. Images were taken immediately using a confocal laser scanning microscope (LSM 710, Zeiss).

Analysis of DC maturation and cytokine production in vitro

DC2.4 cells or BMDCs were seeded into a 12-well plate at a density of 1 × 10 6 cells per well overnight and then treated with SARS-CoV-2 mRNA-loaded LNPs (500 ng ml -1 ) for 24 h. Cell cultures were collected for ELISA of TNF-α, IL-12p70 and IL-1β. Cells were collected, blocked with anti-mouse CD16/32 antibody and then stained with APC anti-mouse CD11c antibody, FITC anti-mouse CD80 antibody and PE anti-mouse CD86 antibody for 30 min at 4 °C before being analysed by flow cytometry (BD, LSR II). Similarly, MoDCs were treated. Cell cultures were collected for ELISA of human TNF-α, IL-12p70 and IL-1β. MoDCs were collected and stained with APC anti-human CD11c antibody, FITC anti-human CD80 antibody and PE anti-human CD86 antibody before analysis. Antibodies were used according the manufacturer's instruction with a typical dilution at 1:100.

Analysis of DC maturation and cytokine production in vivo

Two iLNs from each mouse were harvested at 24 h post-injection of SARS-CoV-2 mRNA-loaded LNPs (5 μg mRNA per mouse) at the tail base and were gently mechanically disrupted using sterile pestles in 0.1 ml of RPMI complete medium in a 1.5 ml tube. The resulting cell suspensions were collected, blocked with anti-mouse CD16/32 antibody and then stained with APC anti-mouse CD11c antibody, FITC anti-mouse-CD80 antibody and PE anti-mouse-CD86 antibody before being analysed by flow cytometry.

Blood was collected into serum separator tubes (BD #365967) through the retro-orbital route at 6 and 24 h post-immunization. Serum was separated from blood following an incubation period of 30 min at room temperature (r.t.), and samples were centrifuged at 10,000g for 5 min. The serum was stored at -20 °C until use. To analyse the intralymphatic cytokine production, the resulting cell suspensions from iLNs were placed to a 96-well plate at a density of 10,000 cells per 100 μl per well and cultured for 8 h. Supernatant was collected for ELISA of TNF-α, IL-12p70 and IL-1β together with serum samples. Additionally, serum at 6, 24 and 48 h post-immunization was collected for ELISA of haptoglobin, IP-10 and MCP-1.

Distribution and transfection of LNPs in vivo

Mice were s.c. injected at the tail base with mLuc-loaded LNPs at a dose of 5 μg mRNA per mouse. At 6 or 24 h post-injection, mice were intraperitoneally (i.p.) injected with d-luciferin potassium salt (150 mg per kg (body weight)), and bioluminescence imaging was performed on an IVIS imaging system (PerkinElmer). To enable concurrent bioluminescence and fluorescence imaging, DiR-labelled, mLuc-loaded LNPs were s.c. injected into mice. At 24 h post-injection, mice were i.p. injected with d-luciferin potassium salt, and major organs and iLNs were collected for bioluminescence and fluorescence imaging.

In vivo immunization

Mice were s.c. immunized with SARS-CoV-2 mRNA-loaded LNPs at a dose of 1 or 5 μg mRNA per mouse twice using a prime-boost strategy at a 3 week interval. Body weight was recorded twice a week during the experiment. Serum was collected using serum separator tubes as described above, stored at -20 °C and used for ELISA and virus neutralization assay. Two weeks after the boost vaccination, mice were anaesthetized and spleens were collected for flow cytometry analysis.

Determination of anti-RBD antibody titres using ELISA

Purified SARS-CoV-2 His tagged RBD (1 μg ml -1 ) (Sino Biological, #40592-V08H) was used to coat High Bind Stripwell Corning 96-well clear polystyrene microplates overnight. Plates were washed with wash buffer (0.05% Tween 20/PBS) once, and blocked for 2 h at r.t. using a solution of heat-inactivated, IgG-depleted, protease-free bovine serum albumin (2% w/v BSA/PBS). Afterwards, plates were washed three times, and mouse sera were serially diluted in the blocking solution and incubated for 2 h at r.t. Plates were washed three times before adding horseradish-peroxidase-conjugated anti-mouse secondary antibody specific to total IgG (1:10,000, Abcam #ab97040) or subclasses (IgG1, 1:10,000, Abcam, #ab98693; IgG2c, 10,000, Abcam, #ab98722) in blocking buffer. Plates were incubated for 1.5 h and washed three times before the addition of 100 μl KPL 3,3′,5,5′-tetramethylbenzidine substrate per well for 8 min. The reaction was stopped by adding 50 μl of 2 N sulfuric acid, and the absorbance was measured at 450 nm using a SpectraMax 190 microplate reader. RBD-specific IgG endpoint dilution titre was defined as the highest dilution of serum to give an optical density greater than the cut-off optical density value determined using the Frey method 47 .

Pseudovirus neutralization assay

A VSV pseudotype with SARS-CoV-2 S was first produced 36 . We performed an antibody neutralization assay using VSVΔG-RFP SARS-CoV-2. Vero E6 cells stably expressing TMPRSS2 were seeded in 100 μl DMEM at 2.5 × 10 4 cells per well in a 96-well collagen-coated plate. After 12 h, twofold serially diluted serum samples were mixed with VSVΔG-RFP SARS-CoV-2 pseudotype virus (50-200 focus-forming units per well) encoding the spike of D614G, Beta or Delta variant and incubated for 1 h at 37 °C. A mouse anti-VSV Indiana G, 8G5F11 (#Ab01401-2.0, Absolute Antibody), was also included in this mixture to neutralize any potential VSV-G carryover virus at a concentration of 100 ng ml -1 . The antibody-virus mixture was then used to replace the media on Vero E6 TMPRSS2 cells. At 20 h post-infection, the cells were washed and fixed with 4% PFA before visualization on an S6 FluoroSpot Analyzer (CTL). Individual infected foci were enumerated, and the values were compared with control wells without antibody. The focus reduction neutralization titre 50% (FRNT 50 ) was measured as the greatest serum dilution at which focus count was reduced by at least 50% relative to control cells that were infected with pseudotype virus in the absence of mouse serum. FRNT 50 titres for each sample were measured in two technical replicates performed on separate days.

Flow cytometry analysis of T and B cells

T cell. Spleens were collected, processed as single cells, filtered using a 70 μm cell strainer in complete RPMI 1640 and centrifuged, and red blood cells lysed in ACK lysis buffer to obtain a clear single-cell suspension. To measure antigen-specific T cells, two million splenocytes were stimulated with 2.5 μg ml -1 of SARS-CoV-2 RBD peptide pools https://doi.org/10.1038/s41565-023-01404-4 (#PM-WCPV-S-RBD-1, JPT) in a FACS tube for 6 h at 37 °C, 5% CO 2 with 2 mg ml -1 anti-CD28 (Tonbo #40-0281-M001) providing co-stimulation. Stimulations proceeded for 1 h before adding 5 mg ml -1 brefeldin A (Biolegend #420601), 2 mM monensin (Biolegend #420701) and 5 mg ml -1 anti-CD107a Alexa Fluor 647 (Biolegend #121610) for 5 h. DMSO served as a negative control, and the combination of 50 mg ml -1 phorbol 12-myristate 13-acetate and 1 mg ml -1 ionomycin served as a positive control. After a total of 6 h, samples were washed with PBS, stained with Live/Dead Aqua for 5 min, blocked using anti-mouse CD16/32 antibody for 20 min and stained extracellularly for 30 min using antibodies (Supplementary Figs. 21d and 27d ). Cells were washed in FACS buffer, fixed and permeabilized using the Cytofix/Cytoperm kit (BD Biosciences #554714), and stained intracellularly using antibodies for 30 min (Supplementary Figs. 21d and 27d ). After intracellular staining, cells were washed twice and fixed with 300 μl (1% PFA), and samples were acquired on a BD LSR II equipped with four laser lines and 18 photomultiplier tubes. The gating strategy, and the antibody list and catalogue numbers are provided in Supplementary Figs. 21 and 27 .

Memory B cell. Spleens were collected, processed as single cells, filtered using a 40 μm cell strainer in complete RPMI 1640 and centrifuged at 300g for 5 min; red blood cells were lysed with ACK (1 min), washed twice and counted, and two million cells per sample were incubated with anti-mouse CD16/32 antibody for 20 min at 4 °C. Cells were then washed with FACS buffer (1% BSA/PBS) and stained for 1 h using antibodies (Supplementary Fig. 23d ). Following staining, cells were washed twice and fixed with 300 μl (1% PFA), and samples were acquired on a BD LSR II equipped with four laser lines and 18 photomultiplier tubes. The gating strategy, and the antibody list, fluorescent RBD probes 14 and catalogue numbers are provided in Supplementary Fig. 23 .

ELISpot assay

Bone marrow was flushed from femurs and tibia into FACS buffer and filtered through a 63 μm Nitex mesh. Red blood cells were lysed in ACK buffer for 5 min on ice, and washed twice with FACS buffer. The resulting cells were counted using a Beckman Coulter ViCell. MultiScreenHTS IP filter plates, 0.45 μm (Millipore Sigma, MSIPS4W10), were coated with RBD protein antigen at 10 μg ml -1 in sodium carbonate/sodium bicarbonate buffer pH 9.6 (35 mM NaHCO 3 and 15 mM Na 2 CO 3 ) for 1 h at 37 °C. Plates were then washed with 200 μl PBS per well three times and blocked at 37 °C in complete RPMI for 30 min. Bone marrow cells were plated in six halving dilutions beginning with one million total bone marrow cells per well and incubated overnight in complete RPMI. Plates were then washed with wash buffer (1× PBS + 0.1% Tween 20) five times, and biotinylated anti-IgG detection antibody (goat anti-mouse IgG human ads-BIOT; Southern Biotech, 1030-08) was added at a final dilution of 3 μg ml -1 in 2% BSA/PBS and incubated at r.t. for 1 h. Plates were once again washed five times, and streptavidin-alkaline phosphatase (1:20,000 dilution in 2% BSA/PBS) was added prior to incubation at r.t. for 30 min. Plates were then washed five times with wash buffer, and 50 μl per well 5-bromo-4-chloro-3-indolyl-phosphate/ nitro blue tetrazolium chloride solution (Sigma, #B1911, 100 ml) was added for ~10 min or until spots developed at which time the reaction was quenched with 100 μl 1 M sodium phosphate monobasic solution. After plates were rinsed with deionized H 2 O and dried overnight, they were scanned and counted using an S6 FluoroSpot Analyzer.

Statistics and reproducibility

All data are presented as mean ± s.d. Student's t-test or one-way analysis of variance (ANOVA) followed by Tukey's test was applied for comparison between two groups or among multiple groups using Graphpad Prism 7.0, respectively. P < 0.05 was considered to be statistically significant. Each experiment is repeated at least three times independently with similar results, and the representative dataset is presented.

Data availability

All data that support the findings of this study are provided within the paper and its Supplementary Information. The raw datasets generated during the study are provided within the source data. Source data are available for Figs. 2 3 4 5 6 , Extended Data Figs. 1 and 2 , and Supplementary Figs. 4, 6-8, 10-12, 15, 17, 18, 20, 22, 24 and 26. The crystal structure data of TLR7 are available in the Protein Data Bank (PDB) under PDB ID: 5GMH. Source data are provided with this paper.

Reporting summary

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

Author contributions

X.H., M.-G.A and M.J.M conceptualized, conceived and designed experiments. X.H., M.-G.A., K.B., J.J.K., K.L. and M.G. performed experiments. M.-G.A. and P.B. provided reagents and protocols. X.H., M.-G.A., J.J.K., N.G. and L.X. analysed data. X.H., M.-G.A., D.W. and M.J.M. wrote the manuscript. All authors discussed and edited the manuscript content.

Competing interests

X.H. and M.J.M. have filed a patent application based on this work. In accordance with the University of Pennsylvania policies and procedures and our ethical obligations as researchers, we report that D.W. is named on patents that describe the use of nucleoside-modified mRNA as a platform to deliver therapeutic proteins and vaccines. D.W. and M.G.A are also named on patents describing the use of lipid nanoparticles, and lipid compositions for nucleic acid delivery and vaccination. We have disclosed those interests fully to the University of Pennsylvania, and we have in place an approved plan for managing any potential conflicts arising from licensing of our patents. The remaining authors declare no competing interests.

Additional information

Extended data is available for this paper at https://doi.org/10.1038/s41565-023-01404-4.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41565-023-01404-4.

Correspondence and requests for materials should be addressed to Drew Weissman or Michael J. Mitchell.

Peer review information Nature Nanotechnology thanks Francis Szoka, Anna Blakney and Donata Medaglini for their contribution to the peer review of this work.

Reprints and permissions information is available at www.nature.com/reprints.

Reporting Summary

Nature Portfolio wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency in reporting. For further information on Nature Portfolio policies, see our Editorial Policies and the Editorial Policy Checklist.

Statistics

For all statistical analyses, confirm that the following items are present in the figure legend, table legend, main text, or Methods section.

n/a Confirmed

The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement A statement on whether measurements were taken from distinct samples or whether the same sample was measured repeatedly

The statistical test(s) used AND whether they are one-or two-sided

Only common tests should be described solely by name; describe more complex techniques in the Methods section.

A description of all covariates tested

A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals) For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated Our web collection on statistics for biologists contains articles on many of the points above.

Software and code

Policy information about availability of computer code Data collection Zeiss LSM 710, BD LSR II Flow Cytometer, PerkinElmer IVIS Spectrum, S6 FluoroSpot Analyzer, Infinite M200 plate reader, Bruker Avance III, 400 MHz

Data analysis

Statistical analysis was performed on Graphpad Prism 7.0, flowcytometry data were analyzed on FlowJo software package (Flowjo V10) and molecular simulation was performed on BIOVIA Discovery Studio 2018;1H NMR spectrum was analyzed by MestReNova x64 (V14.2.0); Confocol images were analyzed by Zen 2011 software.

For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Portfolio guidelines for submitting code & software for further information.

Data Policy information about availability of data

All manuscripts must include a data availability statement. This statement should provide the following information, where applicable:

-Accession codes, unique identifiers, or web links for publicly available datasets -A description of any restrictions on data availability -For clinical datasets or third party data, please ensure that the statement adheres to our policy All data that support the findings of this study are provided within the paper and its Supplementary Information. The raw datasets generated during the study are provided within source data. Source data are provided with this paper. The crystal structure data of TLR7 is accessible in Protein Data Bank (PDB) under the PDB ID: 5GMH.

nature portfolio | reporting summary

March 2021

Field-specific reporting

Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences

For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Life sciences study design

All studies must disclose on these points even when the disclosure is negative.

Sample size

No statistical methods were used to determine the sample size in the study. The sample sizes are clearly described in each figure legend. The sample size were determined by allowable error size, accuracy, resources, and need for statistical analysis (generally n>=3 throughout all the studies) according to the standards of the field.

Data exclusions No animals and/or data were excluded.

Replication

All experiments were repeated for at least three times and experimental findings were reproducible.

Randomization The dosing groups were filled by random selection from the same pool of animals for in vivo experiments. Groups in all the in vitro and in vivo experiments were selected randomly.

Blinding

All the investigators were blinded to group allocation during data collection and analysis.

Animals and other organisms

Policy information about studies involving animals; ARRIVE guidelines recommended for reporting animal research Laboratory animals C57BL/6 mice (female, 6-8 weeks) were ordered from Jackson laboratory and housed in a specific-pathogen-free animal facility at ambient temperature (22 ± 2 °C), air humidity 40%-70% and 12-h dark/12-h light cycle.

Wild animals

No wild animal was used in this study.

Field-collected samples The study did not involve samples collected from field.

Ethics oversight

All animal experiment protocols were reviewed and approved by the institutional animal care and use committee of the University of Pennsylvania. Animal procedures were performed in accordance with the Guidelines for Care and Use of Laboratory Animals at the University of Pennsylvania.

Note that full information on the approval of the study protocol must also be provided in the manuscript.

Human research participants

Policy information about studies involving human research participants Population characteristics 38-year-old healthy male. Other characteristics were not disclosed due to the privacy.

Recruitment

The participant for blood donation was recruited by volunteering.

Ethics oversight

Institutional Review Board of University of Pennsylvania approved pre-clinical studies of the human immune system.

Note that full information on the approval of the study protocol must also be provided in the manuscript.

Flow Cytometry Plots

Confirm that:

The axis labels state the marker and fluorochrome used (e.g. CD4-FITC).

The axis scales are clearly visible. Include numbers along axes only for bottom left plot of group (a 'group' is an analysis of identical markers).

All plots are contour plots with outliers or pseudocolor plots.

A numerical value for number of cells or percentage (with statistics) is provided.

Methodology Sample preparation

Spleens were collected, processed, filtered using a 70 μm cell strainers in complete RPMI-1640, centrifuged, and red blood cells lysed in ACK lysis buffer. Single-cell suspensions were obtained and stained with antibodies according to the manufacturer's protocols, and then analyzed by flow cytometry.

Software

FlowJo software package (Flowjo V10)

Cell population abundance

The absolute cells >= 500,000 were analyzed for fluorescent intensity in the defined gate.

Gating strategy

In general, cells were first gated on FSC/SSC. Singlet cells were gated using FSC-H and FSC-A. Dead cells were then excluded and further surface and intracellular antigen gating was performed on the live cell population (See supplementary infomation).

Tick this box to confirm that a figure exemplifying the gating strategy is provided in the Supplementary Information.

Acknowledgements

Acknowledgements M.J.M. acknowledges support from a US National Institutes of Health (NIH) Director's New Innovator Award (DP2 TR002776), a US National Science Foundation CAREER Award (CBET-2145491), an American Cancer Society Research Scholar Grant (RSG-22-122-01-ET) and a Burroughs Wellcome Fund Career Award at the Scientific Interface (CASI).K.B. acknowledges support from the Polish National Agency for Academic Exchange (number PNN/IWA/2019/00057, Poland).The authors thank S. Steimle from the Beckman Center for Cryo Electron Microscopy at the University of Pennsylvania for help in characterizing the morphology of LNPs.The authors acknowledge the Cell Center Services Facility at the University of Pennsylvania for assistance with the LAL test.The authors thank L. Chen, M. A. Eldabbas and E. Maddox of the Human Immunology Core at the University of Pennsylvania for providing human monocytes, who are supported by NIH P30 AI045008 and P30 CA016520.Elements of Figs. 1, 3, 5 and 6 created with BioRender.com.

Failed to load PDF:

StripeM-Inner

Paper sources

Abstract screening pilot

Abstract screening results

Extraction pilot

Extraction results

Research report

Modify setup