Combinatorial design of ionizable lipid nanoparticles for muscle-selective mRNA delivery with minimized off-target effects

Significance

Lipid nanoparticles (LNPs) crucial for delivering mRNA (messenger RNA)-based therapies including COVID-19 mRNA vaccines can oftentimes inadvertently transfect off-target tissues, leading to potential safety issues. To facilitate tissue-specific mRNA delivery with improved precision, our study presents a platform to quickly create chemically diverse lipids for building LNPs. This method successfully identifies iso-A11B5C1, a lipid that, when incorporated into LNPs, enables efficient muscle-focused mRNA delivery while minimizing off-target delivery to other tissues. Intriguingly, despite limited transfection in lymph nodes, intramuscular administration of mRNA delivered by these LNPs can trigger potent cellular immune responses, of which the efficacy is further validated in a melanoma vaccine model. This work advances methods for muscle-specific mRNA delivery and prompts rethinking of mRNA vaccine designs.

Abstract

Ionizable lipid nanoparticles (LNPs) pivotal to the success of COVID-19 mRNA (messenger RNA) vaccines hold substantial promise for expanding the landscape of mRNA-based therapies. Nevertheless, the risk of mRNA delivery to off-target tissues highlights the necessity for LNPs with enhanced tissue selectivity. The intricate nature of biological systems and inadequate knowledge of lipid structure–activity relationships emphasize the significance of high-throughput methods to produce chemically diverse lipid libraries for mRNA delivery screening. Here, we introduce a streamlined approach for the rapid design and synthesis of combinatorial libraries of biodegradable ionizable lipids. This led to the identification of iso-A11B5C1, an ionizable lipid uniquely apt for muscle-specific mRNA delivery. It manifested high transfection efficiencies in muscle tissues, while significantly diminishing off-targeting in organs like the liver and spleen. Moreover, iso-A11B5C1 also exhibited reduced mRNA transfection potency in lymph nodes and antigen-presenting cells, prompting investigation into the influence of direct immune cell transfection via LNPs on mRNA vaccine effectiveness. In comparison with SM-102, while iso-A11B5C1’s limited immune transfection attenuated its ability to elicit humoral immunity, it remained highly effective in triggering cellular immune responses after intramuscular administration, which is further corroborated by its strong therapeutic performance as cancer vaccine in a melanoma model. Collectively, our study not only enriches the high-throughput toolkit for generating tissue-specific ionizable lipids but also encourages a reassessment of prevailing paradigms in mRNA vaccine design. This study encourages rethinking of mRNA vaccine design principles, suggesting that achieving high immune cell transfection might not be the sole criterion for developing effective mRNA vaccines.

The success of messenger RNA (mRNA) vaccines developed to combat the severe acute respiratory syndrome coronavirus 2 (SARS-COV-2) pandemic has catalyzed a surge of interest in mRNA-based therapies (1, 2). Many applications, spanning prophylactic and therapeutic vaccines, protein replacement therapies, cancer immunotherapies, cellular reprogramming, and gene editing, are under clinical investigation (3). However, an overarching challenge persists for the effective and safe delivery of mRNA to the cytoplasm of target cells. Three major hurdles must be addressed: rapid degradation and lack of localization of naked RNA; impeded passive uptake due to the RNA’s large, polyanionic nature (4); and the necessity for safe, well-tolerated, and body-clearable delivery vehicles (5, 6). Currently, ionizable lipid nanoparticles (LNPs) are widely employed as the delivery vehicle for mRNA with demonstrated efficacy and safety profiles in clinical practice. Thus far, LNPs have enabled three FDA-approved RNA-based therapies, including an siRNA drug (Onpattro) (7) and two mRNA vaccines against SARS-COV-2 (Comirnaty and Spikevax) (8, 9). It is noteworthy that numerous mRNA-LNP therapies are progressing through various clinical trial stages (10), including the first-ever CRISPR/Cas9-based therapy approved for in-human testing (ClinicalTrials.gov Identifier: NCT05120830) (11). While LNPs have been instrumental in clinical translation of mRNA, many designs can result in substantial mRNA expression in the liver, potentially causing liver damage. This effect has been documented in the case of Pfizer’s COVID-19 vaccine (BNT162b2), which upon intramuscular administration, predominantly localizes to the liver and the injection site, leading to undesirable hepatic injury (12). On the other hand, studies have shown that mRNA-free LNPs do not inflict liver damage (13), suggesting that the safety issues associated with mRNA vaccines might arise from mRNA expression in nontarget tissues. Hence, there is an urgent need for LNPs designed for highly muscle-specific mRNA delivery, minimizing off-target effects and addressing safety issues associated with current LNPs.

The classical LNP formulation consists of four components: ionizable lipids, helper phospholipids, cholesterol, and lipid-conjugated polyethylene glycol (PEGylated lipids) (14). Of these, the ionizable lipid is pivotal in this complex as it not only stabilizes mRNA via electrostatic interactions but also promotes endosome escape, facilitating efficient cytosolic delivery of mRNA (15, 16). During formulation, the amine headgroups of ionizable lipids acquire positive charge under acidic conditions, thus favoring the encapsulation of negatively charged mRNA molecules into LNPs. The remaining, unbound ionizable lipids then revert to neutrality during buffer exchange at physiological pH, preventing potential toxicity observed upon nonionizable cationic lipids (16–19). Post endocytosis by target cells, ionizable lipids are protonated again in the acidic endosomal environment, allowing them to interact with the negatively charged endosomal membrane. This facilitates the disruption of both LNP and endosome, ultimately leading to the release of mRNA cargos into the cytoplasm. The importance of ionizable lipids is underscored by the fact that all three FDA-approved RNA-LNP drugs utilize unique ionizable lipids in their formulations. Therefore, innovating ionizable lipid design is crucial for the development of LNPs with enhanced performance in muscle-specific mRNA delivery.

While previous research has shed light on the rational design of ionizable lipids, their synthesis remains a complex, time-consuming, and costly process. This necessitates rapid, streamlined synthetic approaches. Toward this goal, combinatorial chemistry, employing multicomponent reactions, has emerged as an effective tool for high-throughput synthesis (HTS) of expansive and chemically diverse lipid libraries. Typical examples include a Ugi-based three-component reaction (3-CR), which has been demonstrated to expedite the synthesis of a combinatorial library of ionizable lipids within 24 h at room temperature, aiding in the discovery of a STING-activating ionizable lipid beneficial for mRNA vaccine delivery (20). Despite the potential, ionizable lipids synthesized using this approach suffer from inconsistent and poor yield (3.9 to 34%). Such low yields complicate high-throughput analysis as the presence of impurities and low-yielding potent ionizable lipids in LNP formulations can obscure their genuine potential. Additionally, ionizable lipids generated by this method lack biodegradable bonds in the lipid tails, which are crucial for improving the safety profiles of ionizable lipids, increasing the therapeutic window, and enabling repeated dosing of LNPs (20, 21). Rodent studies have shown that LNPs containing nonbiodegradable ionizable lipids can cause liver damage (22–24), underlining the imperative for biodegradable lipid design. To address this issue, we recently reported a Michael Addition-based 3-CR method that generates biodegradable ionizable lipids. This led to the identification of a lipid uniquely suited for efficient mRNA delivery to the lung epithelium (25). Nevertheless, the applicability of this approach is hampered by its protracted 72-h duration and the need for heat. Furthermore, ionizable lipids created by both 3-CR methods were restricted to possessing symmetric tails. Notably, the asymmetry of ionizable lipid tails has been increasingly recognized for facilitating endosomal release and cytosolic payload delivery (26). This is attributed to the pronounced difference in their stretch and bending moduli of asymmetric ionizable lipids significantly compared to naturally occurring symmetric phospholipids, enhancing their efficacy in disrupting the endosomal membranes (26). Consequently, there is a prominent need, yet unmet, for a streamlined HTS platform capable of swiftly synthesizing biodegradable, asymmetrical lipids with a high yield.

Here, we introduce an advanced Ugi-based 3-CR platform that enables the one-pot, high-yield fabrication of biodegradable, asymmetrical ionizable lipids at room temperature. By strategically reconfiguring functional groups and optimizing component design, we have enhanced the chemical diversity of the resulting lipids, particularly for their lipid tails. Notably, this platform’s yield has been markedly improved, consistently achieving a reaction efficiency exceeding 70% through the incorporation of a nontoxic catalyst. Leveraging this platform, we identified an ionizable lipid, iso-A11B5C1, demonstrating an mRNA delivery potency comparable to SM-102. Importantly, LNPs with iso-A11B5C1 enable precise mRNA transfection in muscle tissues, while considerably reducing inadvertent mRNA expression in main organs such as the liver and spleen. Moreover, we demonstrated the potential of iso-A11B5C1 LNPs in muscle-specific gene editing and its promise as a safe and effective delivery vehicle for cancer mRNA vaccines. Intriguingly, despite its low potency to transfect antigen-presenting cells (APCs) and lymph nodes, iso-A11B5C1-mediated mRNA vaccination through intramuscular injection could still induce robust cellular immune responses and demonstrate potent antitumor effect in a melanoma model, providing key insights into the mechanisms of mRNA vaccines and establishing its potential for future therapeutic mRNA vaccine development.

Results

HTS and Screening of a Combinatorial Ionizable Lipid Library for mRNA Delivery.

Within our conceptual framework, we deconstructed ionizable lipid structures into three distinct components: an amine head group, lipid tail A, and lipid tail B, aligning them with the three reactants in a standard 3CR-Ugi reaction (amine, isocyanide, and aldehyde), respectively. An ionizable lipid can thus be readily synthesized by simultaneously coupling the amine headgroup to the isocyanide tail and the aldehyde tail (Fig. 1A). This design differentiates itself from conventional 3CR methodologies, which typically categorize ionizable lipids into three segments: amine head group, linker, and lipid tail. Our 3CR synthetic approach places a stronger emphasis on the significance of lipid tails by introducing an additional dimension to their design. This enables us to explore a unique chemical space for ionizable lipids with diverse tails, accommodating both symmetrical and asymmetrical structures.

Fig. 1.

Ionizable lipids synthesis via combinational chemistry for high-throughput screening. (A) Reaction scheme showing the 3-CR of headgroup A (amine), tail B (aldehyde), and tail C (isocyanide) for the HTS of lipids. (B) Cartoon illustrating the highlights of the 3-CR platform. (C) Structures of the three components of the ionizable lipids used in the synthesis library. (D) High-throughput screening of the lipid library. The relative luciferase expression (relative luminescence unit) after incubating with mLuc LNPs overnight is shown in a heat map. (E) Illustration image of LNP components. (F) The structure of the top-performing lipid is shown. Fig. 1 (A–C) was created with Biorender.com. Fig. 1D balloon plots were created with bioinformatics.com.cn.

To address the limitation of low yield encountered in the previous method of lipid generation using the 3-CR-Ugi reaction, we conducted an investigation into various catalysts, including p-toluenesulfonic acid, phosphoric acid, boric acid, and phenylphosphinic acid for boosting the reaction efficiency (27–31). Remarkably, phenylphosphinic acid (20% mol) emerged as the most effective catalyst, increasing the reaction yield from 10 to 70% (SI Appendix, Fig. S1 and Table S1). To validate this finding, multiple lipids were synthesized and purified, consistently showing high yields ranging from 75 to 90% (SI Appendix, Fig. S2). This improvement allows us to establish a reliable platform for HTS and screening of ionizable lipid libraries (Fig. 1B).

Next, we devised a combinatorial lipid library harnessing the improved 3-CR-Ugi reaction. This lipid library comprises sixteen head groups and five lipid tail A (aldehyde) and three lipid tail B (isocyanide), as shown in Fig. 1C. The amine head groups were selected to include both linear and cyclic structures bearing tertiary amines to provide an ionizable moiety with an appropriate pKa (32). The three isocyanide tails C1 to C3 exhibit variations in length and saturation. The six unique tails B1 to B6 vary in length, branching, saturation, and the presence of ester bonds to explore tail asymmetry. These 288 ionizable lipids were formulated with phospholipids [1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE)], cholesterol, and PEGylated lipids (C14-PEG2000) along with firefly luciferase mRNA (mLuc) according to a classic LNP formulation ratio (33) (Fig. 1E). The mRNA transfection potency of these LNPs was then evaluated in HeLa cells via the bioluminescence signal from luciferase expression (20, 34, 35). Overall, lipids with linear headgroups outperformed those with cyclic headgroups: average normalized transfection potency of lipids with headgroups (A1 to A7) ranged from 0.36 to 0.61, contrasting with cyclic headgroups (A8 to A16) ranging from 0.28 to 0.46 (Fig. 1D). Specifically, lipids with head group A6 displayed superior performance. Additionally, increased carbon chain length in the ionizable amine headgroup corresponded with diminished luciferase signal, as showcased by the groups A6 (0.61), A2 (0.53), to A4 (0.42), respectively. This is probably because steric hindrance might inhibit optimal mRNA encapsulation by restricting RNA-lipid electrostatic interactions.

Of the six variable tails, B2, B3, and B5 provided the highest average relative transfection efficiencies. Despite A6 and B2 being the overall most potent headgroup and tail, the most potent lipid was the product of the ester-bearing branched tail B5, cyclic headgroup A11, and isocyanide tail C1, designated henceforth as iso-A11B5C1. As previously reported, the branched nature of the tail chain likely contributes to its heightened transfection efficiency due to the endosomal escape (36). This can be explained by the significant difference in the cross-section, which plays a crucial role in shaping the overall structure of the lipid. Thus, the asymmetry resulting from the nonlinear tail and small head group contributes to the formation of the cone-shaped lipid structure that disrupts bilayer stability (16), leading to the increased potency of mRNA delivery. Lastly, we further investigated the 10 top-performing lipids tested in vivo. The highest in vivo muscle transfection efficiency was also observed from lipid iso-A11B5C1 (SI Appendix, Fig. S3), which was unexpected as in vitro and in vivo results are often poorly correlated (37). This poor correlation may be partly attributed to the poor yield reactions associated with previous synthetic ionizable lipid libraries, thereby clouding the measurement of the genuine activity of ionizable lipids.

Optimization of Iso-A11B5C1 LNP for Muscle-Specific mRNA Delivery.

Following the identification of the top-performing ionizable lipid iso-A11B5C1, we sought to optimize the transfection potency through formulation Design of Experiment (DOE) studies. Following previously described methods (33), we simultaneously investigated six independent parameters known to influence the transfection efficiency of LNPs: I) ratio of ionizable lipid to mRNA mass, II) ionizable lipid molar composition, III) phospholipid type, IV) phospholipid molar composition, V) PEGylated lipid molar composition, and VI) aqueous-phase pH. The levels tested for these variables are displayed in Fig. 2A, and detailed formulations are listed (SI Appendix, Table S2). The optimized LNP formulation was obtained by employing a desirability function, defined by an encapsulation efficiency (EE) exceeding 80% and maximized transfection potency. All formulations contained mLuc to measure their transfection efficiency in HeLa cells. Based on the transfection potency results of all formulations (Fig. 2B), lipids that passed EE gating were optimized within the complete design space with identical weighting (Fig. 2C). The optimal formulation tailored for iso-A11B5C1 was defined by the following parameters: 10:1 ionizable lipid:mRNA mass ratio with 60% iso-A11B5C1, 10% DOPE, 29% cholesterol, and 1% PEGylated lipid molar composition, as well as pH 4.5 aqueous phase. LNP characterizations indicated particle size and dispersity similar to SM-102 formulated LNPs (SI Appendix, Fig. S4). This optimized formulation was used for all the following experiments.

Fig. 2.

LNP formulation optimization and biodistribution. (A) Library of formulations with varied parameters. (B) Box plot of the transfection potency of 18 LNP formulations (F1 to F18), indicated by relative luminescence unit (RLU log₂ fold change). (C) Multiple response surface plots based on desirability function optimization for iso-A11B5C1 formulation. (D) Representative IVIS images of C57BL/6 J mice at 6 h following intramuscular administration of SM-102 or iso-A11B5C1 mLuc-LNPs (0.3 mg/kg). (E) Quantification of Fluc protein expression in muscle, liver, and spleen. (n = 3) (F) Representative IVIS images of C57BL/6 J mice at 6 h following intramuscular administration of SM-102 or iso-A11B5C1 Cy5-mLuc-LNPs (0.3 mg/kg). (G) Quantification of Cy5 signal in muscle, liver, and spleen. (n = 3) Statistical significance was analyzed by the two-tailed Student’s t test. *P-value < 0.05, **P-value < 0.01, ***P-value < 0.005, and ****P-value < 0.001. Data are presented as mean ± SD. Fig. 3C was generated from JMP 13 software.

We next evaluated the in vivo muscle transfection potency of our optimized iso-A11B5C1 LNP and compared it to Moderna’s SM-102 LNP formulation used in their COVID-19 vaccine, Spikevax. Both LNPs were formulated with mLuc and delivered to C57BL/6 mice via intramuscular (I.M.) injection (0.3 mg/kg, n = 3). As shown in Fig. 2 D and E, iso-A11B5C1 and SM-102 LNP showed comparable transfection efficiency at the injection site (no statistical difference). Remarkably, iso-A11B5C1 showed no measurable transfection in off-target organs, whereas SM-102 showed high transfection in the liver and spleen (Fig. 2E), indicating exceptional muscle specificity for iso-A11B5C1. We then performed a biodistribution study of iso-A11B5C1 LNP utilizing Cyanine5 (Cy5)-mLuc. Both iso-A11B5C1 and SM-102 LNPs were formulated with Cy5-mLuc and delivered to C57BL/6 mice via I.M. injection (0.3 mg/kg, n = 3). As depicted in Fig. 2 F and G, the fluorescent signal distribution analysis revealed that iso-A11B5C1 LNP primarily localized at the injection site, with minimal transfer to organs like the liver or spleen. In contrast, SM-102 exhibited a broader distribution pattern, being detected not only at the injection site but also within the liver and spleen. Upon intravenous injection of iso-A11B5C1 or SM-102 LNP/mLuc into mice, it was evident that the expression levels in the liver and spleen remained significantly lower in the iso-A11B5C1 group compared to the SM-102 group (SI Appendix, Fig. S5), demonstrating that inherent cell-type specificity of the lipid significantly contributes to its tissue specificity. Furthermore, we also explored the administration of iso-A11B5C1 LNPs through intradermal and subcutaneous routes, where a reduction in muscle signal was observed compared to that of the intramuscular injection group (SI Appendix, Fig. S6).

To explore the mechanisms behind the muscle-specific delivery by iso-A11B5C1 LNP, we investigated the in vitro transfection efficiency of cell lines representing organs of interest: HepG2 (Liver), and splenocyte (Spleen) (Fig. 3A). Compared to the SM-102 LNP, iso-A11B5C1 LNP had 42-fold (5.4 log2-fold) higher transfection efficiency in HeLa but 28-fold (4.8 log2-fold) and 24-fold (4.6 log2-fold) lower in HepG2 and splenocytes respectively, suggesting that both biodistribution and cellular factors likely play a role in tissue specificity of iso-A11B5C1 LNP.

Fig. 3.

Characterization of iso-A11B5C1 LNP muscle selectivity. (A) iso-A11B5C1 LNP transfection efficiency indicated by bioluminescence signal in different cell lines: HeLa, HepG2, and splenocyte. (B) Apparent pKa of iso-A11B5C1 LNP and SM-102 LNP calculated from TNS assay. (C) Representative image of the mechanism for subcellular analysis of mRNA expression in mTmG reporter mice. (D) Quantification of GFP⁻ to GFP⁺ converted cells by SM-102 or iso-A11B5C1 LNPs, measured from confocal microscopy images. n = 3, 5 sections from each independent biological replicate were averaged for quantification. (E) Representative confocal microscopy images of tdTomato and GFP expression in cryosection of muscle, liver, and spleen of mTmG mice postinjection of LNP/mCre by I.M. injection. Scale bar: 50 µm for liver sections; 20 µm for muscle and spleen sections. Statistical significance was analyzed by the two-tailed Student’s t test. *P-value < 0.05, **P-value < 0.01, and ***P-value < 0.005. Data are presented as mean ± SD. Fig. 3C was created with Biorender.com.

Next, we measured the apparent pKa of LNPs, which can influence tissue specificity. Through a TNS assay, the apparent pKa of iso-A11B5C1 LNP was approximately 7.2 (Fig. 3B), compared to the apparent pKa of 6.7 of SM-102. The lower pKa of SM-102 could potentially explain why SM-102 LNP showed higher transfection efficiency in both HepG2 cells and splenocytes, as previous reports state that lower pKa is associated with higher transfection in the liver and spleen (38). As physiological pH is around 7.35 to 7.45, the lower pKa of SM-102 LNP means it remains more net-neutral, possibly increasing LNP stability and dispersion from the injection site which contains negatively charged membranes and extracellular matrices. Hence, the SM-102 LNP that entered circulation after I.M. injection can navigate to and transfect organs such as the liver and spleen. Consistent with previous findings (39), LNPs with high pKa, such as iso-A11B5C1, show localized transfection after I.M. injection.

Efficacy and Safety Study of Iso-A11B5C1 LNPs for Muscle-Specific Gene Editing.

To investigate the potential for iso-A11B5C1 in gene editing applications, we I.M. injected iso-A11B5C1 and SM-102 LNPs carrying Cre-recombinase mRNA (mCre) into gene-engineered mTmG reporter mice (40) at 0.5 mg/kg. In this mTmG reporter animal, tdTomato is ubiquitously expressed on the cell membranes. The functional delivery of Cre mRNA facilitates the genetic recombination of cells such that tdTomato is replaced with membrane-bound EGFP. We observed high EGFP signal from the administration sites (quadricep muscle) for both iso-A11B5C1 and SM-102 LNP, quantified by the conversion rate of cells (41). Iso-A11B5C1 and SM-102 treated groups showed comparable levels of editing at the administration sites. However, the SM-102 treated group showed high off-target editing in the liver (83.83%) and spleen (72.67%), while editing in these tissues by iso-A11B5C1 was undetectable (Fig. 3 D and E). In gene therapy applications, unlike COVID-19 vaccines, immunogenic LNPs could pose a considerable challenge. Addressing this concern, we delved deeper into the immunogenic characteristics of the lipid iso-A11B5C1. Our in vitro tests utilized a reporter human leukemia monocytic (THP-1) cell line, which facilitates the study of NF-kB pathways via the examination of inducible reporter construct activity. Treatment of these reporter cells with iso-A11B5C1 LNPs showed no marked elevation in NF-kB activation in comparison with the untreated control, suggesting a negligible innate immune response elicited by the iso-A11B5C1 LNP (SI Appendix, Fig. S7). Additionally, an in vivo study involved the intramuscular injection of iso-A11B5C1 LNP into BALB/c mice, followed by an assessment of the acute inflammatory cytokine profile in the blood 4 h (42) postadministration. It’s worth noting that there were no discernible differences in acute inflammatory cytokine levels, including GM-CSF, IFN-γ, IL-1a, IL-5, IL-6, and TNF-α, between the iso-A11B5C1-treated mice and those treated with PBS (SI Appendix, Fig. S8), underscoring the low immunogenic characteristic of our lipid. Collectively, our findings indicate that iso-A11B5C1 LNPs are potent and safe mRNA delivery vehicles with high selectivity for muscle tissues.

Investigation of Iso-A11B5C1 LNPs for mRNA Vaccination.

While iso-A11B5C1 LNPs exhibit robust mRNA transfection potency in muscle tissue, their transfection potency is markedly reduced in bone marrow dendritic cells (BMDCs) and inguinal lymph nodes relative to the SM-102 LNP (Fig. 4 B and C). This observation aligns with our aforementioned finding that iso-A11B5C1 LNPs predominantly transfect muscle tissues over others. To elucidate the underlying cause of this reduced transfection in immune cells, we assessed the internalization of iso-A11B5C1 LNPs versus SM-102 LNPs in BMDCs using Cy5-tagged mLuc as a marker. Results showed that the internalization of iso-A11B5C1 LNP was limited (SI Appendix, Fig. S9), with uptake rates around 12%, in stark contrast to SM-102 LNP’s uptake at approximately 56%. Further in vitro analysis indicated the diminished capacity of iso-A11B5C1 LNP to instigate DC maturation compared to that of SM-102 LNP, evident from the decreased expression of CD86 or CD40 compared to those treated with SM-102 LNP. This trend was also observed in vivo, as DCs from inguinal lymph nodes of mice treated with iso-A11B5C1 LNP also exhibited reduced maturation, (SI Appendix, Fig. S10) in comparison with the SM-102 LNP-treated cohort.

Fig. 4.

Immune responses after mOVA vaccination. (A) Timeline for vaccination and sample collection. (B) iso-A11B5C1 and SM-102 LNP transfection efficiency in BMDCs. (C) iso-A11B5C1 and SM-102 LNP distribution in the ipsilateral inguinal lymph node after I.M. injection. (0.3 mg/kg, n = 3) (D and E) OVA-specific antibody titers in the mice treated with iso-A11B5C1/mOVA and SM-102/mOVA on day 14 and day 28. Control: unvaccinated animals. (n = 4). (F) Representative flow cytometry diagrams and quantification of percentages of IFN-γ+ CD8+ cells in splenocytes. (G) Heatmap of cytokines and chemokines released from splenocytes upon MHC II OVA peptide stimulation. (H) Schematic illustrating DCs capturing OVA antigens that are released from the transfected muscle cells, followed by DC antigen presentation to CD4 T cells. Statistical significance was analyzed by using one-way ANOVA with Tukey’s multiple comparisons test. *P-value < 0.05, **P-value < 0.01, ***P-value < 0.005, and ****P-value < 0.001. Data are presented as mean ± SD. Fig. 4 A, B, and H was created with Biorender.com.

Moderna’s Spikevax COVID-19 vaccine effectively transfects APCs after I.M. delivery. Yet, the imperative of direct transfection by LNPs for potent mRNA vaccines remains a topic of discussion.

The muscle-specific affinity of iso-A11B5C1, combined with its diminished APC transfection, presents a unique opportunity probe the significance of direct APC transfection in mRNA vaccine efficacy. To deepen our understanding of muscle-specific LNPs, like iso-A11B5C1, in mRNA vaccination, we delivered full-length ovalbumin mRNA (mOVA) as a model antigen, encapsulated in iso-A11B5C1 LNPs or SM-102 LNPs. BALB/c animals were primed with LNP/mOVA on day 0, followed by a booster on day 14. The antigen-specific IgG antibody levels induced by iso-A11B5C1/mOVA at days 14 and 28 were significantly lower than SM-102/mOVA after the prime and after the booster (Fig. 4 D and E), indicating a reduced humoral response for iso-A11B5C1 vaccination. We then conducted ex vivo stimulation of splenocytes from both vaccinated and unvaccinated animals using MHC-I OVA epitopes. We quantified the responses through flow cytometry, specifically measuring the percentage of IFN‐γ+ and CD8+ cells among the stimulated splenocytes. Following MHC I OVA peptide stimulation, both the SM-102 LNP-treated and iso-A11B5C1-treated groups exhibited a significant and comparable increase in IFN‐γ+ cells compared to the control group (Fig. 4F). This observation suggests a comparable cellular immune response between the iso-A11B5C1 and SM-102 treatments. In addition to MHC I, we also investigated MHC II OVA peptide stimulation. The results showed splenocytes from both iso-A11B5C1/mOVA and SM-102/mOVA groups displayed comparable proinflammatory cytokines patterns and profiles (Fig. 4G), especially for IFN-γ, IL-1b, and IL-6. As illustrated in Fig. 4H, the observed positive response upon the stimulation of MHC II OVA is likely due to DCs capturing the OVA antigen that is released from the transfected muscle cells. As these muscle cells undergo lysis by CD8 T cells, it facilitates the presentation of the antigen via the MHC II pathway. Consequently, even in the absence of direct transfection, DCs can facilitate antigen presentation through this mechanism. Therefore, despite the lower humoral immune response induced by iso-A11B5C1/mOVA, there was a comparable cellular immune response between iso-A11B5C1 and SM-102.

Efficacy Study of mRNA Cancer Vaccines Delivered by Iso-A11B5C1 LNPs.

To assess the potential efficacy of iso-A11B5C1 LNP for therapeutic vaccines, we conducted experiments using a B16F10 melanoma model. We chose to encapsulate tyrosinase-related protein 2 mRNA (mTrp2) in the LNPs for our mRNA cancer vaccine, as previous research had proved Trp2 as a validated tumor-associated antigen for melanoma (20, 43). C57BL/6 mice were subcutaneously injected with B16-F10-Luc melanoma cells in the right flank. These B16-F10-Luc melanoma cells were engineered to express firefly luciferase, allowing us to monitor tumor volume. Subsequently, we administered full-length mTrp2 encapsulated in iso-A11B5C1 LNPs, as well as two other LNP formulations, SM-102 and MC3. Animals received two doses of LNP/mTrp2 or a control PBS solution on day 7 and day 12 after tumor inoculation. Our findings revealed that iso-A11B5C1 LNPs significantly slowed tumor growth and outperformed MC3 LNPs, as shown in Fig. 5E. Moreover, the analysis of stimulated splenocytes from the iso-A11B5C1 LNP-treated group demonstrated a notably higher percentage of IFN-γ+ CD8+ cells compared to the MC3 LNP-treated group (Fig. 5 B and C). Additionally, tumors extracted from the iso-A11B5C1-treated group exhibited significantly greater CD8 T cell infiltration compared to those from the MC3 LNP-treated group (Fig. 5D and SI Appendix, Fig. S11).

Fig. 5.

The therapeutic effect of iso-A11B5C1 LNP cancer vaccine on B16-F10-Luc melanoma model. (A) Timeline for tumor inoculation and vaccination. (B) Representative flow cytometry diagrams of IFN-γ+ CD8+ cells in splenocytes. (C) Quantification of percentages of IFN-γ+ CD8+ cells in splenocytes. (D) Representative immunofluorescence staining of CD8/IFN-γ killer T cells within tumor regions of PBS or mRNA cancer vaccine-treated mice. CD8 staining is shown as red; IFN-γ is shown as green; counterstain is DAPI (blue) in the merged image shown at right (n = 5 biologically independent mice per group). (E) Tumor volumes of B16-F10-Luc model represented by luminescence signal. Statistical significance was analyzed by using one-way ANOVA with Tukey’s multiple comparisons test. *P-value < 0.05, **P-value < 0.01, ***P-value < 0.005, and ****P-value < 0.001. Data are presented as mean ± SD. Fig. 5A was created with Biorender.com.

Discussion

The potential for mRNA drugs was recently showcased by the development of the COVID-19 mRNA vaccines; however, challenges such as increasing delivery to on-target tissues, limiting off-target delivery, and improving lipid clearance remain as barriers to the development of next-generation mRNA medicines. For vaccines, this task is further complicated by a limited understanding of which cell types are necessary targets to generate effective protection. Underlying these challenges is the necessity for methods that can rapidly synthesize diverse lipids for effective evaluation.

Here, we report an innovative approach for the HTS of chemically diverse lipids. This provides the capability to generate and identify novel lipids via combinational chemistry and gain a deeper understanding of structure–activity relationships. There are multiple advantages to our system. The rapid product synthesis through a convenient one-pot method eliminates the need for multiple reactions and purification steps. Additionally, by incorporating a nontoxic catalyst into the reaction, we can achieve high yields exceeding 70%. This not only saves time but also reduces resource consumption, thereby affording three key desirables of large-scale manufacturing: less time, greener chemistry, and lower resource costs. Beyond synthetic gains, incorporating biodegradable ester moieties improves lipid clearance, expanding the therapeutic window of drug formulations by limiting toxicity (5). Lastly, tail asymmetry increases the chemical space and plays a role in forming a conical cross-sectional, rather than cylindrical, shape of lipids. This attribute helps to disrupt the endosomal bilayer structure, improving transfection efficiency through endosomal escape (16). Together, this 3-CR provides a major improvement over previous 3-CR lipid strategies (20, 25, 44) to meet the rising need for novel ionizable lipids in the rapidly growing field of RNA-centric treatments.

Through screening our diverse lipid library, we identified the lipid iso-A11B5C1 for its exceptional transfection efficiency. Upon optimization of our iso-A11B5C1 LNP formulations, we achieved comparable in vivo muscle transfection to SM-102. Remarkably, biodistribution studies of iso-A11B5C1 LNP indicate transfection only in muscle tissues, whereas SM-102 LNP showed considerable off-target transfection in the liver and spleen. We propose two nonexclusive mechanisms for this muscle selectivity property. One possibility is that iso-A11B5C1 LNP can only transfect muscles due to muscle-exclusive receptors or pathways. This is supported by our in vitro studies in which iso-A11B5C1 LNP was far less effective at transfecting HepG2 cells or splenocytes compared with SM-102 LNP, as well as the observation of low liver/spleen transfection of iso-A11B5C1 LNP upon intravenous injection. Alternatively, iso-A11B5C1 LNP may not reach solid organs, such as the liver and spleen, after I.M. injection. This is proved by our in vivo physical biodistribution study of the iso-A11B5C1 LNP. The apparent pKa of iso-A11B5C1 LNP is 7.2, while SM-102 LNP has the apparent pKa of 6.7. The high pKa of iso-A11B5C1 LNP is close to the physiological pH (7.35 to 7.45), therefore, there is overall a more positive nature of iso-A11B5C1 LNP compared to SM-102 LNP. Upon I.M. injection, iso-A11B5C1 LNP the partial positive surface charge may mediate affinity for the negatively charged cell surfaces and extracellular matrix found in the muscle tissues, providing a possible explanation for retention and uptake of LNPs at the injection site (39). Although muscle cell targeting and the role of pKa in guiding tissue specificity of LNP have been investigated (39, 45), earlier studies on ionizable lipids have rarely achieved the pronounced muscle transfection efficacy we observed for iso-A11B5C1 (comparable to SM-102), especially when simultaneously curbing transfection in areas such as the liver and spleen. Thereby, unlike SM-102, which shows lymph node drainage and entry into circulation via the lymphatic system, iso-A11B5C1 is likely unable to effectively travel beyond the site of injection. This finding is consistent with a previous report that claimed lipids with higher pKa showed high transfection efficiency after I.M. administration, but showed low to no transfection after IV injection (15). In contrast, a different group suggested that an apparent pK_a between 6.6 and 6.9 maximizes intramuscular mRNA delivery (46). Clearly, transfection specificity and efficiency are multifaceted, complex challenges in which a single factor is insufficient to predict LNP performance. This poses a great challenge for the rational design of lipids, necessitating the synthesis and evaluation of diverse lipid structures.

Considering the exceptional muscle specificity of iso-A11B5C1, we investigated its potential for either muscular gene therapy or vaccination. For a proof of concept for muscle gene therapy, we used the Cre recombinase-mTmG reporter animal. Similar to our in vivo biodistribution study, we report exceptional muscle specificity of our iso-A11B5C1 LNPs for gene editing. While no editing in liver or spleen cells was detected by iso-A11B5C1 LNP, SM-102 LNP edited the majority of these cells. The remarkable potential for high-specificity myocyte gene editing by iso-A11B5C1 warrants further study.

Inspired by the low off-organ targeting of iso-A11B5C1, we next explored its potential for vaccination. The low potency in transfecting nonmuscle cells, such as APCs, renders iso-A11B5C1 less effective at inducing a humoral immune response than SM-102, as indicated by a 10-fold lower antigen-specific IgG titer. However, the splenocyte cytokine profiles from iso-A11B5C1 and SM-102 vaccinated animals showed a similar IFN-γ expression from CD8+ T cells upon peptide stimulation ex vivo, suggesting comparable Th1-based cellular immune responses for iso-A11B5C1 and SM-102. Both vaccinated groups showed comparable levels of proinflammatory cytokines, such as IFN-γ, IL-1b, and IL-6, which are important in promoting Th1-biased immune responses to generate robust cellular immunity. The Th1-based immune response may provide an ideal platform for approaches such as therapeutic vaccines which prioritize CD8+ cytotoxic T lymphocyte (CTL) activation. Thereby, we then tested the therapeutic potential of the LNP as cancer mRNA vaccine. In the B16-F10 melanoma model, iso-A11B5C1 LNP demonstrated a potent antitumor effect that outperformed MC3 LNP, as indicated by markedly reduced tumor progression, elevated levels of antigen-specific CD8+ T cells, and a substantial infiltration of CD8+ T cells into the tumor. It is noticed that although iso-A11B5C1 LNP demonstrated potent antitumor effect, the efficacy of iso-A11B5C1 LNP is lower than that of the SM102 LNP group. This observed difference likely suggests that direct APC transfection can augment the immune response, but it is not an absolute requirement for eliciting a robust immune response.

The discrepancy of antibody titer induced by the two LNPs might be associated with the APC transfection efficiency. Typically, I.M. injected mRNA vaccines generate antigen-specific immune responses after the transfection of muscle cells or APCs (such as dendritic cells). When muscle cells are transfected with the antigen-coding mRNA, the antigen will be translated, and peptide fragments will be presented to the cell surface through MHC I. These muscle cells will then be recognized by CTLs, leading to CTL activation and initiating CTL-mediated killing (47). Muscle cells will then be killed by the activated CTLs, causing antigen release into the extracellular space. The local inflammation will recruit APCs to the site, where these released antigens can be engulfed by APCs for further immune activation (48). Alternatively, APCs can be directly transfected with the antigen-coding mRNA. Here, the antigen protein fragments will be presented on the APC surface via MHC I and MHC II molecules to assist CTL activation and induce antigen-specific antibody responses (49). As iso-A11B5C1 does not transfect APCs, it results in a lower chance for the cytosolic OVA antigens to be presented by MHC II. Whereas SM-102 not only has the capability to transfect muscle, it also directly transfects APCs, allowing for a greater presentation of cytosolic OVA antigens through MHC II in APCs. A potential improvement could involve replacing the nonsecreted OVA with a secreted equivalent to study its impact on antibody responses when APC direct transfection is absent. Nevertheless, the CD8+ mediated cellular response was largely similar between SM-102 and iso-A11B5C1. Therefore, transfection of APCs by mRNA-LNPs is not necessary to mediate the cellular immune response, a key consideration for the development of future mRNA vaccines, especially therapeutic vaccines that prioritize CTL activation. Additionally, iso-A11B5C1 LNP may be of interest for vaccinating at-risk populations that may be prohibited from receiving the current mRNA vaccines due to potential liver damage.

In summary, we developed a muscle-targeted mRNA delivery system, iso-A11B5C1 LNP by using our optimized combinatory chemistry approach. The Iso-A11B5C1 LNP demonstrated exceptional muscle specificity in mRNA delivery, exhibiting transfection efficiency comparable to the commercially available lipid SM-102 in the muscle. Importantly, our investigations underscored the capacity of iso-A11B5C1 for efficient muscle-specific gene editing with undetectable off-target effects, and thus introducing a promising approach for safe and effective gene therapy targeting muscle diseases. Additionally, our exploration using a model antigen (OVA) revealed that APC transfection is crucial in affecting the ability of mRNA vaccines to elicit humoral immune responses but not cellular immune responses. This finding substantially advances our fundamental understanding of the roles of muscle cells and immune cells in shaping the immune response to mRNA vaccines, thereby offering valuable insights for the development of future mRNA vaccines. Furthermore, we substantiated the exceptional therapeutic efficacy of iso-A11B5C1 LNP, as demonstrated by its potent antitumor effects in a melanoma model, thereby solidifying its promise for the development of mRNA-based therapeutic vaccines.

Materials and Methods

Ionizable Lipid Library Synthesis.

For ionizable lipid library synthesis, 3-CR-Ugi was utilized to synthesize ionizable cationic lipids between amine groups (−NH₂), aldehyde groups (−CHO), and isocyanide groups (-NC). Briefly, amine, aldehyde, and isocyanide with the molar ratio (1:1:1) were dissolved in methanol using a single-neck round-bottom flask, and then, phenylphosphinic acid as catalyst was dissolved in 50 µL of methanol and added into the flask. The reaction lasted for 12 h under nitrogen atmosphere at room temperature. All reactions were performed in 96-well plates. For the in vitro high-throughput transfection study or in vivo analysis assay, the lipid mixtures were used without purification. Otherwise, the lipid was purified by flash column chromatography on a C-815 preparative chromatography system (BUCHI). The structure was confirmed by ¹H NMR spectrometry (400 MHz, Bruker spectrometer) and Q Exactive HF-Orbitrap mass spectrometer (Thermo). Detailed chemical synthesis route is available in SI Appendix.

LNP Synthesis.

LNPs were synthesized by mixing an aqueous phase containing the mRNA with an organic phase containing the lipids at a 3:1 volume ratio and a lipid/mRNA weight ratio of 10:1. LNPs were formulated by pipetting for all in vitro work without dialysis. LNPs for in vivo work were formulated using a T junction device via syringe pumps and buffer exchanged by dialysis against 1× PBS in a 10,000 MWCO cassette (Thermo Fisher) at 4 °C for 12 h before animal injection. The organic phase was prepared by dissolving a mixture of the synthesized ionizable lipid, helper phospholipid DSPC (Avanti) or DOPE (Avanti), cholesterol (Avanti), and C14-PEG2000 (Avanti) in ethanol at a predetermined molar ratio: 50:10:38.5:1.5 for DSPC formulation; 35:16:46.5:2.5 for DOPE formulation (33). SM-102 was purchased from Echelon Bioscience. The aqueous phase was prepared with mRNA in 10 mM citrate buffer (pH 4.0, Fisher). mRNA included in the study: mLuc (firefly luciferase mRNA, TriLink), mCre (Cre-recombinase mRNA, 5-methoxyuridine (5moU), TriLink), and mOVA (full-length ovalbumin mRNA, 5moU, TriLink). Plasmid information on TRP2 is available in SI Appendix.

In Vitro High-Throughput Screening.

LNPs for in vitro screening were prepared in a 96-well plate by direct pipette mixing of the ethanol and the aqueous phase. LNPs/mLuc complex was added to 96-well plates preseeded with HeLa cells at the dosage level of 100 ng mLuc/10,000 cells. After overnight incubation, mLuc transfection efficiency was measured using the Bright-Glo Luciferase Assay System (Promega) according to the manufacturer’s instructions. The relative luminescence unit (RLU) was quantified by the Cytation 1 Cell Imaging Multimode Reader (BioTek).

Animals and Cells.

HeLa and HepG2 cells were purchased from ATCC. THP-1 monocyte cell line (THP-1 Dual cells) was purchased from InvivoGen. B16-F10-Luc cells were kindly provided by Gang Zheng from Princess Margaret Cancer Center, University Health Network, Toronto. Cells were grown in RPMI 1640 media containing 10% fetal bovine serum (FBS) and penicillin–streptomycin (Gibco) according to manufacturer protocol in a humidified incubator containing 5% CO₂ at 37 °C. BMDCs were prepared with the previously reported protocol (50). Briefly, bone marrow cells, flushed from the femurs of C57BL/6J mice, were cultured in the dendritic cell medium: RPMI 1640 supplemented with 10% FBS, penicillin–streptomycin (Gibco), and 20 ng mL⁻¹ GM-CSF (PeproTech). Then, medium replacement was performed every 2 d. On day 7, immature dendritic cells that are nonadherent or loosely adhering were collected and seeded at 10,000 cells per well in a 96-well plate. All animal studies were approved and conducted in compliance with the University Health Network Animal Resources Centre guidelines. Female C57BL/6 and BALB/c mice (4 to 8 wk) were obtained from the Jackson Laboratory. mTmG [Gt(ROSA)26Sor ^{tm4(ACTB-tdTomato,-EGFP)Luo}/J] mice were kindly provided by Agostino Pierro from the Hospital for Sick Children, Toronto. For tumor model establishment, B16-F10-Luc cells (10⁶ cells/mouse) were implanted subcutaneously in the right flank of C57BL/6 mice (female, 6 wk old). Tumor volumes were measured by IVIS Spectrum Imaging System (PerkinElmer). Luminescence was quantified using the Living Image software (PerkinElmer).

Bioluminescence.

At 6 h after the injection of the mRNA LNPs, mice were intraperitoneally (IP) injected with 0.2 mL D-luciferin (10 mg/mL in PBS). Mice were anesthetized in a ventilated anesthesia chamber with 1.5% isoflurane in oxygen and imaged 10 min after the injection using the IVIS Spectrum Imaging System (PerkinElmer). Luminescence was quantified using the Living Image software (PerkinElmer).

LNP Formulation Optimization.

DOE methodology was used to optimize the six independent parameters known to influence the transfection and EE of LNPs: I) ratio of ionizable lipid to mRNA mass, II) ionizable lipid molar composition, III) phospholipid type, IV) phospholipid molar composition, V) PEGylated lipid molar composition, and VI) aqueous-phase pH. DOE was carried out using JMP 13 software (SAS Institute).

LNP Characterization.

The size, polydispersity index, and zeta potentials of LNPs were measured using dynamic light scattering (ZetaPALS, Brookhaven Instruments). Diameters are reported as the intensity mean peak average. To calculate the nucleic acid EE, a modified Quant-iT RiboGreen RNA assay (Invitrogen) was used as previously described (33).

TNS Binding Assay for pKa.

TNS binding assay was used to determine the LNP apparent pKa. LNPs were diluted and mixed with TNS in a total volume of 150 μL of buffered solution with the final concentrations: (LNP, 75 μM) and (TNS, 6 μM). The buffered solutions with pH ranging from 3 to 10 were made containing 20 mM boric acid, 10 mM imidazole, 10 mM sodium acetate, 10 mM glycylglycine, and 25 mM NaCl. The Cytation 5 Cell Imaging Multi-Mode Reader (BioTek) was used to read the fluorescence (Ex321/Em445). The pH was measured in each well after TNS addition. Mathematica (Wolfram Research) was used to fit the fluorescence data to the Henderson–Hasselbalch equation to provide the pKa.

Immunohistochemistry.

Gt(ROSA)26Sor^{tm4(ACTB-tdTomato,-EGFP)Luo}/J mTmG mice were intramuscularly injected with LNPs/mCre containing 10 μg mRNA formulated with iso-A111B5 or SM-102. Thirty-six hours after the injection, the mice were killed for tissue collection: muscle (quadricep), liver, and spleen. After collection, tissues were fixed in 4% buffered paraformaldehyde overnight at 4 °C and then dehydrated in 30% sucrose overnight at 4 °C. Next, cryosection was performed on tissues after being frozen in OCT. Tissue sections were mounted with DAPI (1:1,000). For the tumor model, mice were killed for tumor collection. For tumor sections, slides were stained with DAPI (1:1,000), anti-CD8 (APC, BioLegend, 1:200), and anti-IFN- γ (AF488, BioLegend, 1:200). Zeiss LSM900 was used to obtain and process the images.

Enzyme-Linked Immunosorbent Assay (ELISA) for Antibody Titer.

Indirect ELISA was performed to measure the antibody titer. The high-binding 96-well plates (Corning) were covered with 100 μL of ovalbumin protein (InvivoGen) with a concentration of 20 μg/mL at 4 °C overnight. The plates were then washed four times with PBS-T and blocked by 1% bovine serum albumin solution (Sigma-Aldrich). The sera collected from immunized and control mice were initially diluted at 1:100. Then, a twofold serial dilution was performed on all serum samples. 100 μL of samples were added to the plates for 1 h at 37 °C. Then, the plates were washed and then incubated with goat anti-mouse IgG (H+L) secondary antibody with HRP at 1:10,000 (InvitroGen) for 1 h at 37 °C. Next, the plates were washed and 50 μL of TMB ELISA substrate solution was added for 15 min incubation at room temperature. Then, 50 μL of stop solution (0.16M sulfuric acid) was added. Cytation 1 Cell Imaging Multimode Reader (BioTek) was used to measure the Optical Density (OD₄₅₀).

Flow Cytometry.

Antibodies purchased from BioLegend and Thermo Scientific for flow cytometry are listed in SI Appendix, Table S3. For the flow cytometry analysis of surface markers, cells from mouse BMDCs, tumors, splenocytes, and lymph nodes were preincubated with anti-CD16/32 antibody (BioLegend) and stained on ice with fluorophore-conjugated antibody. For the staining of intracellular markers (IFN-γ), cells were fixed and permeabilized using the Foxp3/Transcription Factor Staining Buffer Set (eBioscience). Then, cells were stained with anti-IFN-γ. Flow data were acquired on a CytoFlex S flow cytometer and analyzed using FlowJo software.

Splenocyte Stimulation and Cytokine Study.

BALB/c mice were immunized with full-length OVA mRNA (5moU, TriLink) encapsulated in either SM-102 LNP or iso-A11B5C1 LNP (n = 4). Both vaccinated groups, as well as the unvaccinated group (PBS injection) (n = 4), were killed at D28 for spleen collection. Splenocytes were then disassociated as previously described (51). In brief, spleens were homogenized and passed through the 70 μm cell strainer (Fisherbrand). The cells were washed twice with PBS, followed by RBC cell lysis treatment. Next, splenocytes were plated at a density of 10 million cells per milliliter in 48-well culture plates with culture medium RPMI 1640 supplemented with 10% FBS and penicillin–streptomycin (Gibco). Then, splenocytes were stimulated with either one of the OVA-derived peptides: H-2K^b–restricted MHC class I epitope (InvivoGen) or I-Ad-restricted OVA MHC class II epitope (InvivoGen) at the concentration of 10 μg/mL. Supernatants of splenocytes were collected for cytokine study after 72 h. The supernatants were diluted 4 times for 20 different cytokines measurements using the Quantibody^® Mouse Cytokine Array 1 Kit (RayBiotech, USA) according to the manufacturer’s protocol. The assay was analyzed using the GenePix^® Professional 4200 Microarray Scanner at 530 nm.

Statistics.

GraphPad Prism version 9.4.1 (GraphPad Software) was used for all statistical analysis and for generating the plots. Data are reported as mean ± SD. Significant differences between groups were analyzed by using Student’s t test or Tukey’s multiple comparisons test based on test groups. *P value < 0.05, **P value < 0.01, ***P value < 0.005, and ****P value < 0.001.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix.