A long-term stable cold-chain-friendly HIV mRNA vaccine encoding multi-epitope viral protease cleavage site immunogens inducing immunogen-specific protective T cell immunity

ABSTRACT

The lack of success in clinical trials for HIV vaccines highlights the need to explore novel strategies for vaccine development. Research on highly exposed seronegative (HESN) HIV-resistant Kenyan female sex workers revealed naturally protective immunity is correlated with a focused immune response mediated by virus-specific CD8 T cells. Further studies indicated that the immune response is unconventionally focused on highly conserved sequences around HIV viral protease cleavage sites (VPCS). Thus, taking an unconventional approach to HIV vaccine development, we designed lipid nanoparticles loaded with mRNA that encodes multi-epitopes of VPCS (MEVPCS-mRNA LNP), a strategic design to boost antigen presentation by dendritic cells, promoting effective cellular immunity. Furthermore, we developed a novel cold-chain compatible mRNA LNP formulation, ensuring long-term stability and compatibility with cold-chain storage/transport, widening accessibility of mRNA LNP vaccine in low-income countries. The in-vivo mouse study demonstrated that the vaccinated group generated VPCS-specific CD8 memory T cells, both systemically and at mucosal sites of viral entry. The MEVPCS-mRNA LNP vaccine-induced CD8 T cell immunity closely resembled that of the HESN group and displayed a polyfunctional profile. Notably, it induced minimal to no activation of CD4 T cells. This proof-of-concept study underscores the potential of the MEVPCS-mRNA LNP vaccine in eliciting CD8 T cell memory specific to the highly conserved multiple VPCS, consequently having a broad coverage in human populations and limiting viral escape mutation. The MEVPCS-mRNA LNP vaccine holds promise as a candidate for an effective prophylactic HIV vaccine.

Introduction

Safe, effective, and broadly accessible HIV vaccines are the most cost-effective solution to end the HIV pandemic, however, it remains unavailable after 40 years of intensive research and all the HIV vaccine efficacy clinical trials so far have primarily failed, emphasizing that novel vaccine strategies should be explored.

The elicitation of protective cellular immunity may be important for an effective HIV vaccine [1,2], while the elicitation of broadly neutralizing antibodies (bNabs) by HIV vaccines, which is particularly challenging [3], is highly desirable to prevent HIV infection. HIV highly exposed seronegative (HESN) female sex workers in Pumwani, Nairobi, in Kenya, remained uninfected despite repeated exposure to HIV-1 through high-risk sex work. Detailed studies of the protective immunity observed in the HESN group indicated that HIV resistance is associated with narrowly targeted CD8 T cell responses to functionally essential and conserved viral protease cleavage sites (VPCS) epitopes [4–8]. Based on the detailed study of the Pumwani HESN group [7,9], we developed a novel HIV vaccine approach targeting the sequence surrounding the 12 VPCS that elicit focused protective T cells. The study indicated that 12 HIV VPCS proteolytic reactions are critical for HIV to generate a mature and infectious progeny and a single failed proteolytic reaction will produce non-infectious virion. Also, VPCS sequences are extremely conserved among HIV subtypes [8,10], similar to other functional motifs identified in Gag and Pol, such as reverse-transcriptase and non-nucleoside analogue reverse-transcriptase inhibitors targets. All 12 VPCS regions are highly immunogenic [8], therefore, immunological defence to the highly conserved and tightly regulated VPCSs region would exert strong selection pressure affecting HIV’s reproductive fitness [7]. Hence, a prophylactic HIV vaccine targeting VPCSs would confer protection against multiple HIV subtypes by overcoming the major challenge of HIV diversity. In fact, our recent studies demonstrated vaccine targeting SIV sequences surrounding the 12 VPCS protected 80% of female Mauritian Cynomolgus macaques from pathogenic SIVmac251 [9], without inducing high mucosal inflammatory responses and CD4 T cell recruitment/activation [9], thus, minimizing HIV host-target T cell frequency at the HIV port of entry, responsible for initiating and amplifying HIV transmission.

To move the VPCS vaccine towards clinical applications, we designed a novel mRNA cassette expressing all 12 VPCS immunogens as a single polypeptide unit, termed as multi-epitope VPCS (MEVPCS), to promote effective multivalent antigen presentation and easy production, improving antigen receptor sensitivity [11]. Further, studies have demonstrated that modified mRNA-based lipid nanoparticle (LNP) vaccines against various infectious diseases such as ZIKA, Ebola, Influenza and SARS-CoV-2 are safe, highly immunogenic, and protective [12]. However, the major limitation of mRNA LNP-based vaccines is thermo-instability and ultra-cold storage requirement [13]. In line with our previous study, we hypothesized that a cold-chain friendly and long-term stable MEVPCS-mRNA LNP vaccine could generate VPCS-restricted CD8 T cell immunity while minimizing inflammation and CD4 T cell activation [9]. To test this hypothesis, we modified the LNP formulation method to synthesize stable MEVPCS-mRNA LNP vaccine that could be stored at 4°C for months, avoiding the current ultra-low temperature storage requirement of mRNA LNP vaccines [13]. By using modified LNP-mediated delivery, the MEVPCS-mRNA is delivered intracellularly, and translated into MEVPCS polypeptides (all 12 VPCS immunogens) in antigen-presenting cells (APCs). The intracellular expression of MEVPCS expression was confirmation by western blot analysis. The MEVPCS intracellular expression promotes endogenous proteolytic cleavage into individual VPCS epitopes typically specific to HLA class-I alleles [8] for antigen presentation and inducing VPCS-specific CD8 T cell immunity. We test the hypothesis as a proof-of-concept study by evaluating the immunogenicity of the MEVPCS-mRNA LNP vaccine in BALB/c mice. The study demonstrated that the MEVPCS-mRNA LNP vaccine generated potent VPCS-specific CD8 T cell immunity without inducing generalized inflammation and CD4 T cell activation. Therefore, this proof-of-concept study revealed that the cold-chain friendly and long-term stable MEVPCS-mRNA-LNP vaccine could be a potential candidate for an effective prophylactic HIV vaccine.

Materials and methods

Materials

The lipids: 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC, purity 99%), SM-102 (purity 99%), 1,2-dimyristoylrac-glycero-3-methoxypolyethylene glycol-2000 (DMG-PEG2000, purity 99%), and cholesterol (purity >99%) were obtained from ABP Biosciences (Beltsville, MD), DC chemicals (Shanghai, China), BroadPharma (San Diego, CA), and Sigma-Aldrich (St. Louis, MO), respectively.

Phosphate-buffered saline (PBS, pH 7.4), Fetal Bovine serum (FBS), Dulbecco’s Modified Eagle’s Medium (DMEM), RPMI-1640 medium (RPMI), and Dithiothreitol (DTT) were purchased from ThermoFisher Scientific (Waltham, MA). Whereas Triton-X100, Penicillin/Streptomycin (Pen/Strep), N-tert-butyl hydroxylamine (NtBHA), ethylenediaminetetraacetic acid (EDTA), Collagen IV, bovine serum albumin, DNase I, Formaldehyde solution Sigma-Aldrich (St. Louis, MO). All the materials were used as received.

Methods

Multi-epitope VPCS immunogens (MEVPCS) design

The 12 VPCS epitopes are immunogenic and highly conservative both within SIV and HIV-1 variants. Our previous study [8] conducted on human subjects, detailed the immunogenicity of the sequences around the VPCS and the population coverage of a vaccine targeting HIV-1 PCS. The sequence conservation was evaluated by comparing the entropy score of sequences around VPCS with Gag and Pol. Population coverage study based on HLA class I genotype frequencies indicated many HLA class I alleles have multiple epitopes in the 12 VPCS regions, inferring potential immunogenicity sequence around the VPCS region. Further, multiple regions (epitopes) around the 12 PCS region were found to aligned with multiple epitopes of many HLA class I alleles and are common to >95% of world populations, reflecting sequences around the VPCS are relatively conserved compared with other Gag and Pol regions. Further, the VPCS sequence conservation is comparable to the other identified functional motif sequences in Gag and the sequences targeted by reverse-transcriptase inhibitors NRTIs and NNRTIs. It suggests that VPCS sequence conservation are restricted to the sequence viability. Therefore, our previous study showed designing MEVPCS immunogen targeting these 12 VPCS sites is a rationale HIV vaccine approach.

As a linker unit, we selected GGS, a short three amino acid linker because GGS linker was known to give flexibility to multidomain proteins/peptides [14], necessary for appropriate proteolytic cleavage. Thus, GGS linker gives flexibility that is necessary to orientation the MEVPCS polypeptide chain for appropriate proteolytic cleavage and antigen processing to generate 9–11-mer VPCS peptides for MHC I presentation. Further, the presence of GGS in VPCS antigens is also known to promote preferential MHC-I presentation, as detailed in a published NMR-based study by McShan et al. [15]. This study revealed disulphide bond at the peptide-binding groove, selectively recognize discrete conformational states sampled by MHC-I alleles, towards editing the repertoire of displayed antigens. It has been established that editing the peptide cargo with proper confirmed is essential for chaperone interactions on MHC-I for folding and antigen repertoire selection by MHC-I molecules, which en route to the surface. Thus, the presence of a GGS linker promotes both appropriate VPCS antigen processing and MHC-I presentation.

MEVPCS-mRNA design and synthesis

In the present study, we designed MEVPCS-mRNA expressing SIV version of MEVPCS with the futuristic goal to comparatively evaluate the efficacies of MEVPCS-RNA-LNP vaccine and our previously reported vaccine (rVSV and nanoparticle) using an NHP model [8]. The codon-optimized MEVPCS DNA cassette encoding MEVPCS-mRNA sequence (between the T7 promoter and T7 terminator) was integrated between BamH1/HindIII cutting site of pBluescript KS(+) plasmid (Supplementary Figure 1(A,B)). The MEVPCS DNA cassette designed to synthesize MEVPCS-mRNA construct composed of the following base sequences: 5′ Cap followed by 5′ UTR (Ribosome binding site+ Kozak consensus sequence), the MEVPCS-coding ORF, a 3′ UTR sequence, followed by 105 poly(T) (Figure 1(C)). Prior to in-vitro transcription (IVT), the linearized MEVPCS-mRNA coding DNA cassette was digested out of the plasmid by BamH1/HindIII restriction enzyme digestion (Supplementary Figure 1(C)). The HiScribe™ T7 High Yield RNA Synthesis Kit (New England Biolabs®, Ipswich, MA) was used to synthesize highly efficacious IVT MEVPCS-mRNA from BamH1/HindIII digested linearized DNA. The MEVPCS-mRNA were synthesized in the presence of Anti-Reverse Cap Analog (ARCA), 1-methylpseudouridine-5′-triphosphate (at 1:2 ratio with UTP), 5-Methylcytidine-5′-Triphosphate (at 1:2 ratio with CTP), to obtain capped modified nucleoside-containing mRNA. The template DNA was removed by Turbo™ DNase (ThermoFisher, Waltham, MA), following the manufacturer’s instruction. Further, the unincorporated NTPs, enzymes, and buffer components were cleaned from MEVPCS-mRNA using Monarch® RNA Cleanup Kit (New England Biolabs®, Ipswich, MA). The cleaned mRNA quality and quantity were analysed by denaturing Agarose gel and absorbance at 260 nm (with 260/280 ratio ≥1.8), respectively (Supplementary Figure 1(D)). All the steps were carried out in the RNase-free environment and using RNase-free regents to avoid RNA degradation. The purified mRNA was kept frozen at −20°C until used.

Figure 1.

MEVPCS-mRNA synthesis and nanoencapsulation strategy. (A) The 12 VPCS sequences of Gag, Gag-Pol, and Nef precursor proteins from SIVmac239 strain. (B) The MEVPCS polypeptide sequence with a GGS spacers. (C) The codon-optimized MEVPCS-mRNA sequence as expressed from MEVPCS_GGS pBluescript II KS(+) clone. (D) Schematic diagram depicting MEVPCS-mRNA LNP synthesis method. (E) TEM image of MEVPCS-mRNA LNPs.

The MEVPCS-mRNA was quantified by UV-spectral analysis using DeNovix Inc. DS-11 Series Spectrophotometer (Wilmington, DE). The quality of the mRNA synthesized was evaluated by agarose gel electrophoresis (Supplementary Figure 1(D)). Briefly, 500 ng of MEVPCS-mRNA were mixed with 10 µL 2× RNA loading dye to a volume of 20 µL and heated for 15 min at 70°C in order to denature any secondary structures. After cooling down, samples were loaded in respective wells of the 1% bleach agarose gel for analysing RNA quality. The gel electrophoresis was conducted at 100 V for 1 h, and the gel was imaged under Bio-Rad ChemiDoc XRS+ station (Hercules, USA).

Rationale and design of cold-chain friendly LNP formulation

It has been established that mRNA-based lipid nanoparticle (LNP) vaccines are safe, highly immunogenic, and protective against various infectious agents such as influenza A viruses, rabies virus, and SARS-CoV-2 [16]. However, mRNA-LNP research has thus far mainly focused on enhancing mRNA-LNP vaccines’ in-vivo stability, with minimum attention on their stability during on-shelf storage. During the COVID-19 pandemics, when huge on-demand supply-chain pressure come out, the primary constrain of the mRNA LNP vaccine was its stability during worldwide distribution [17]. To be distributed worldwide and to keep it stable for more extended periods, the current mRNA LNP vaccines need to be frozen to avoid mRNA degradation and by-products generation leading to oxidative stress, (Moderna: −15°C to −25°C and BioNTech/Pfizer: −60°C to −90°C), which is a challenge for low- to middle-income country’s medical infrastructure, which works under limited resources (local pharmacy/clinics mostly have 4°C refrigerator, only). Infectious diseases like HIV are most prevalent in third-world countries, where medical facilities work under limited resources. Thus, to maximize the reach of the vaccine module in medical facilities with limited resources, a vaccine needs to render a sufficiently long shelf life, preferably at cold-chain temperatures (2–8°C) or above. Therefore, prolonged cold-chain temperatures (2–8°C) overcoming the above constraints is needed. We modified the LNP formulation methodology to promote its stability at cold-chain temperatures (2–8°C). Our modified LNP formulation method has been designed to tackle the multi-faceted stability challenge associated with mRNA LNP refrigerator storage, such as mRNA degradation and by-products production (e.g. RNA adducts and peroxides) during interaction between LNP excipients. Our modified mRNA LNP formulation improves mRNA stability both as a drug/cargo substance and a drug/cargo LNP product.

The LNPs were formulated following the previously described method [18] by rapid solvent exchange mechanism, with needful modifications to improve LNP stability and reproducibility, explained below (Figure 1(D)). Briefly, the four lipid components, i.e. SM-102: Cholesterol:DSPC: DMG-PEG2000 were dissolved at 50:38.5:10:1.5 molar% in ethanol with a total of 10 mM lipid concentration mix. The lipid mix (organic phase or oil) was added dropwise to the 10 mM sodium citrate buffer (pH 4) aqueous phase (water) at lipid: aqueous of 1:4 v/v ratio, under constant high-speed agitation. The above oil-in-water (o–w) emulsion was agitated overnight (O/N) under constant speed at 4°C. Further, the organic phase of the LNP suspension evaporated under sterile conditions. The LNP suspension was dialysed O/N at 4°C against modified PBS (pH 7.4, supplemented with 50 µM NtBHA and 10 µM EDTA; volume 1000-fold higher) using Slide-A-Lyzer™ G2 Dialysis Cassettes (10 K MWCO; ThermoFisher, Waltham, MA), with ever 6 h buffer exchange. The LNPs were always stored at 4°C until used.

MEVPCS-mRNA LNP formulation

Complete MEVPCS-mRNA LNPs formulation methodology was carried out in RNase-free environment and using RNase-free regents to avoid RNA degradation. The aqueous phase (AQ) consists of MEVPCS-mRNA (200 µg/mL) in 10 mM sodium citrate buffer (pH 4) to ionize the RNA molecule to a highly negatively charged polymeric molecule (Figure 1(D)). The organic phase (OP) consisting of lipid components (SM-102: Cholesterol: DSPC: DMG-PEG2000 at 50:38.5:10:1.5 molar%) in ethanol was added dropwise in the above MEVPCS-mRNA solution at 1:4 v/v OP to AP ratio. In the above oil-in-water (o–w) emulsion, the positively charged lipid (SM102, N) to negatively charged mRNA (MEVPCS-mRNA, P) was maintained at an 8 N/P ratio. It is evident from previous studies that lower positively charged to mRNA lipids ratio (N/P ratio at 1:1) mRNA LNPs during escape endocytosis leads to endo-extracellular vehicles (EVs) mediated loss of mRNA released from LNPs [19]. Thus, to avoid the loss EV-mediated mRNA loss, MEVPCS-mRNA was loaded at 8 (N/P) molar ratio with respect to the positively charged SM-102 lipid. The above (o–w) emulsion was mixed for 8 h under constant agitation at 4°C, followed by OP evaporation. The MEVPCS-mRNA LNPs were then purified from free unincorporated materials by O/N dialysis (10 kDa MWCO cassette; ThermoFisher Scientific; Waltham, MA) against modified PBS (pH 7.4). The MEVPCS-mRNA LNPs obtained were kept at 4°C and used within one week of preparation.

Physicochemical characterization studies

Size distribution, poly-dispersity index (PDI), and surface charge evaluation

The size and concentration of LNPs and MEVPCS-mRNA LNPs were evaluated by the NS300 Nanosight instrument (Malvern Panalytical, Malvern, Worcestershire, UK). The samples were diluted 200 times in PBS for analysis. However, the nanoparticle’s surface charge and PDI were evaluated by the Zetasizer Nano-Zs instrument (Malvern Panalytical, Malvern, Worcestershire, UK).

mRNA concentration evaluation and encapsulation efficiency (EE)

The concentration of MEVPCS-mRNA in MEVPCS-mRNA LNPs was determined by using the Quant-iT™ RiboGreen® RNA assay kit (Invitrogen; Carlsbad, USA). The mRNA concentration was determined based on the standard curve prepared from purified MEVPCS-mRNA (ranging from 0 to 4000 ng/mL). The MEVPCS-mRNA LNP samples were diluted 50- and 100 times to obtain the expected mRNA concentration within the standard curve range. Empty LNPs were diluted and considered as blank control for MEVPCS-mRNA LNP samples. Then, each standard diluent (150 µL), diluted empty LNPs (150 µL), and MEVPCS-mRNA LNP samples (150 µL) were treated with 140 µL 2% Triton X-100 and 10 µL heparin. The standards and samples were then incubated for 15 min at 70°C, causing the release of the mRNA from LNPs. After cooling down, in a 96-well light-protected plate, 100 µL of each standard and samples were added and treated with 100 µL of RiboGreen reagent (100-fold diluted in TE buffer). The sample mix was gently mixed and incubated for 5 min at RT, and the fluorescence intensity was measured at excitation and emission wavelengths 485 and 535 nm, respectively. The concentration of MEVPCS-mRNA in MEVPCS-mRNA LNPs was evaluated based on the standard curve fitting linear equation (r² = 0.99). The % EE was evaluated by the following equation.

$$\text{\%}\mathrm{EE}={\displaystyle \frac{\mathrm{Concentration}\mathrm{of}\mathrm{mRNA}\mathrm{in}\mathrm{LNPs}\left(\mu \mathrm{g}/\mathrm{mL}\right)}{\mathrm{Starting}\mathrm{concentration}\mathrm{of}\mathrm{mRNA}\left(\mu \mathrm{g}/\mathrm{mL}\right)}\times 100}$$

Morphology evolution by transmission electron microscopy (TEM)

The LNPs and MEVPCS-mRNA LNPs morphology was evaluated by TEM imaging. Briefly, the LNPs were diluted in ultrapure water at 1:10 v/v (pH 4) to maximize the number of particles in each image area and avoid LNPs coalescence. About 8 µL of sample solution was applied on the Carbon-formvar grid (300 mesh Cu grids) for 1 min. After removing excess liquid, the LNP on grids were negative-stained with 1% uranyl acetate solution for 1 h, and excess liquid was removed with a filter paper, and the LNP bearing grids were imaged with JEOL JEM 2010 TEM (Tokyo, Japan) at 80 kV magnification.

Physicochemical stability study

MEVPCS-mRNA LNPs stability at 4°C was evaluated based on size distribution, concentration, PDI, surface charge, and %EE studies. Briefly, six different batches (n = 6) of MEVPCS-mRNA LNPs were maintained at 4°C (each sample starting volume was 2 mL), and 200 µL of MEVPCS-mRNA LNPs were collected on days 0, 7, 15, 30, and 60 days. The samples were diluted 100 times on the respective day of collection. On the respective analysis day, the size distribution and concentration (LNPs/mL) of the samples were evaluated by the NS300 Nanosight instrument; the PDI and surface charge (ζ-potential) were evaluated by the Zetasizer Nano-Zs instrument. The % EE evaluated the MEVPCS-mRNA stability in the LNP based on the RiboGreen assay result and the following equation in the above section.

In vivo functional stability study

To affirm the long-term functional stability of mRNA within the modified mRNA LNP, in terms of mRNA cargo efficacy and the target mRNA expression ability, we formulated eGFP expressing mRNA LNP (n = 5 batches). The physicochemical stability of mRNA LNPs was analysed based on size distribution, LNP/RNA concentration and %EE (Supplementary Figure 3(A)), following the method as described in the above section.

To evaluate in vivo functional stability of modified eGFP expressing mRNA LNP, three BALBc mice were injected intraperitoneally (IP), with 10-month-old mRNA LNP (3 µg mRNA in mRNA LNPs/200 µL/mouse), respectively. After 7 days (mimicking the booster dose time point, as depicted in Figure 4(A)), mice were scarified and organs (liver and spleen), were collected. The tissue samples were fixed in SafeFix II and paraffin-embedded to evaluate the eGFP expression at the pre-systemic metabolism/clearance centre (liver) and at the HIV target site (spleen) by IHC analysis of eGFP expression (Supplementary Figure 3(B)). The SafeFix II and paraffin-embedded samples were cut at 6 μm and deparaffinized in xylene and ethanol washes. Antigen retrieval was performed with 98°C citrate buffer for 15 min, followed by 5% BSA blocking for 30 min. Endogenous non-specific IgG was blocked using Fab Fragment Goat Anti-Rabbit IgG (Jackson ImmunoResearch Laboratories-111-007-008) for 2 h, at room temperature. The conjugated eGFP Polyclonal Antibody (ThermoFisher Scientific-A10260) was added to slides at 1:250 and 1:500 dilutions for spleen and liver, respectively. Samples sections were incubated overnight at 4°C. Vina Green Chromogen kit (BioCare Medical-BRR807AH) was used to visualize the HRP conjugated antibody. After mounting with Cytoseal XYL and hardening, the slides were imaged under Olympus BX51 fluorescence microscope (PA, USA), using 40X and 60X Oil objectives, and Darkfield (DF) mode.

Figure 4.

MEVPCS-mRNA LNP in-vivo vaccine potency study. (A) Schematic diagram depicting the in-vivo study strategy (A). Graphical presentation showing the change in CD4+ population (B) and CD8+ population (C) subpopulation phenotype at two weeks post-prime and -boost immunization (day 28). The study was conducted on PBMCs isolated from whole blood collected from mice on the terminal day (day 28). The graph presents data as mean ± SD, and each data obtained was presented (solid dot: MEVPCS-mRNA LNP group; hollow dot: LNP (blank) group; and cross: untreated control (PBS) group). The MEVPCS-mRNA LNP group had eight mice, whereas, the LNP group and the untreated Control (PBS) had six mice each. The statistical difference between groups was analysed by one-way ANOVA followed by Dunnett’s multiple comparisons.

In-vitro VPCS expression studies by Western blot analysis

Human embryonic kidney 293 T cells (HEK293T) from ATCC (# CRL-3216) were cultured in complete Dulbecco’s Modified Eagle’s Medium (DMEM), supplemented with 10% FBS and 1X Pen/Strap solution. For the in-vitro VPCS expression, HEK293 T cells (10⁵ cells/mL) were seeded in 6-well plates. After O/N culture, the cells were treated with 0.1, 0.5, 1, and 2 µg of MEVPCS-mRNA, as Lipofectin MEVPCS-mRNA complex (free form) or as MEVPCS-mRNA LNPs (pre-incubated with 0.1 µg human recombinant Apolipoprotein E3 (Sigma-Aldrich; Burlington, MA)) in AIM V medium. After 72 h of culture, cells were lysed in the lysis reagent (CelLyticTM M, Sigma-Aldrich; Burlington, MA) supplemented with 1 mM PMSF on ice for 10 min. The solubilized cell content was centrifuged at 15,000g for 30 min at 4°C, and the supernatant was collected.

For WB analysis, the cell supernatants (15 µL cell lysate/well) mixed with SDS loading buffer plus DTT were boiled and run in SDS-PAGE (10%) gel (Bio-Rad). As a positive control, MEVPCS polypeptide in-vitro translated (IVT) from MEVPCS-mRNA and linearized MEVPCS pBluescript plasmid, respectively was synthesized using a 1-Step Human Coupled IVT kit (ThermoFisher Scientific; Waltham, MA). IVT MEVPCS polypeptide was considered in parallel, as positive control (at standard 15 µL IVT mix/well). Following SDS-PAGE separation, the proteins were transferred to PVDF membranes. The membrane was blocked with 2% BSA in Tris-buffered saline with 0.1% Tween® 20 detergent (TBST) for 4 h at RT. After a thorough wash with TBST, the member was incubated with the cocktail of mouse monoclonal anti-VPCS antibodies (anti-mouse VPCS 2, 3, and 5 antibodies), along with anti-β-actin antibody (Abcam; Woburn, MA), as the loading controls (targeting β-actin, a 42 kDa molecular weight (MW) housekeeping protein), O/N at 4°C. Whereas, during WB detection of MEVPCS polypeptide expression using an IVT kit, Vinculin anti-human antibody from Novus (Centennial, CO; MW124 kDa, a higher MW housekeeping protein) was used to clearly detect MEVPCS band (Figure 3(A)). After washing with TBST, the member was incubated with the anti-mouse-HRP antibody (ThermoFisher Scientific; Waltham, MA) in 1% BSA in TBST. The member was incubated with SuperSignal West Pico Chemiluminescent Substrate (ThermoFisher Scientific; Waltham, MA) to detect protein bands and imaged using the ChemiDoc MP Imaging System (BIO-RAD, Hercules, CA, USA). To rule out cross-reactivity of anti-β-actin antibodies and to evaluate the MEVPCS gene expression, the conjugated anti-β-actin antibodies were removed from the blot by using a mild stripping solution and rigorous washing, the membrane blot was incubated with anti-VPCS antibody cocktail, i.e. anti-PCS2, 3 and 5 antibodies mix and imaged for MEVPCS band (Supplementary Figure 1(E,F)). To further evaluate anti-VPCS binding specifically, a fresh blot was incubated with the anti-VPCS antibody cocktail only and imaged for the MEVPCS band (Supplementary Figure 1(E,F)).

Figure 3.

MEVPCS polypeptide expression study based on WB analysis. (A) In-vitro translation (IVT) of MEVPCS polypeptide (15 µL IVT product/well); Lane 2, “*” and Lane 3, “#” indicates MEVPCS expression from MEVPCS-mRNA and linearized MEVPCS_GGS (pBluescript plasmid, respectively) using 1-Step Human Coupled IVT Kit. The anti-Vinculin antibody (MW124 kDa), was used as a higher MW housekeeping protein. (B) Represents in-vitro expression of MEVPCS polypeptide in HEK293T cells at different concentrations and different time points along with anti-β-actin (MW: 46 kDa), as lower MW housekeeping protein. (C) Comparative graphical presentation of MEVPCS polypeptide expression based on band-intensity analysis. Each data point represents the mean of four independent experiments.

In-vivo immune-profiling studies

Mouse immunization

For the in-vivo immune response study, female BALB/c mice were purchased from Jackson Laboratory (Bar Harbor, ME). The mice were acclimatized in a pathogen-free environment, and the complete procedure was done according to the Institutional Animal Care and Use Committee (IACUC) approved protocol. After a week of acclimatization, the mice were immunized by intramuscular injection at the hind leg, with 0.05 mL of MEVPCS-mRNA LNPs (n = 8) and blank LNPs (n = 6) immunized on Days 0 (prime) and 14 (booster) (Figure 4(A)). The control mice (n = 6) received 0.05 mL intramuscular PBS injection. Blood was drawn on days 1, 14, and 28, post-immunization (Figure 4(A)). All mice were sacrificed on day 28, and tissues such as the spleen, cervicovaginal tissue, draining lymph nodes, and small intestine were collected for single-cell isolation.

Single-cell isolation

PBMCs were isolated from whole blood by following an optimized protocol. Briefly, after plasma removal, blood samples were subjected to RBC lysis using AKC lysis buffer (ThermoFisher Scientific, Waltham, MA), following manufacturer protocol. The PBMCs were washed thrice with PBA (0.1% BSA in PBS) by spinning at 600×g for 10 min at RT. Finally, PBMCs were resuspended in RPMI medium (supplemented with 10% FBS) for the VPCS activation study.

All the tissue mentioned above was collected in cold PBS. After washing several times in ice-cold PBS, single cells were harvested from the tissue following standardized protocols. Briefly, in the case of cervicovaginal tissue and small intestinal tissue (hard tissue), the minced tissues were first digested with pre-warmed collagenase type IV (2 mg/mL) for 1 h, followed by DNase I (1 mg/mL) incubation for 15 min at 37°C. The digested mix was then passed through a 70 µm nylon strainer (Thermofisher Scientific, Waltham, MA) with 0.6% sodium citrate containing PBA solution and centrifuged at 1000g for 5 min at 4°C. The cell pellet was subjected to RBC lysis by resuspending in RBC Lysis Buffer (BioLegend, San Diego, CA) for 15 min at 4°C. The single cells pellet was washed two times with PBA and resuspended in a complete RPMI medium (supplemented with 10% FBS) for the VPCS activation study.

In the case of the spleen and lymph nodes, the tissue was directly passed through a 70 µm nylon strainer with 0.6% sodium citrate containing PBA solution and centrifuged at 1000g for 5 min at 4°C. The cell pellet was subjected to DNase I (1 mg/mL) for 15 min at 37°C under rigorous agitation and re-filtered through a 70 µm nylon strainer as before. After centrifugation and washing with PBA, the RBC lysis step was carried out as described above. Finally, the splenocytes and lymphocytes in the complete RPMI medium for the VPCS activation study.

VPCS activation study

For the VPCS stimulation study, the single cells (from treated, MEVPCS-mRNA LNPs or Blank LNPs; and control, PBS/mock-treated mice, as described in the above section) from cervicovaginal and small intestinal tissue were seeded (at 10⁵ cells/200 µL/well), whereas splenocytes and lymphocytes were seeded (at 5 × 10⁴ cell/200 µL/well) in a 96-well plate (Nunc™ 96-Well Polystyrene Round Bottom Microwell Plates; Thermofisher Scientific, Waltham, MA) in triplicates for VPCS challenge study. The 96-well bearing respective tissue-specific single cells and PBMCs from treated and control mice (each sample seeded in triplicate well) were stimulated with 2.5 µg/mL of pool VPCS peptide mix (PCS1-4 and PCS5-8) for 18 h at 37°C. As the positive-controls, in the triplicate well with single cells and PBMCs (obtained from control mice) from control samples were treated with PMA (10 ng/mL) and Ionomycin (1 µg/mL) as T cell activation stimulants.

Flow cytometry

For MEVPCS-induced immune response study, the blood was drawn before priming (day −1), on the day before booster (day 14), and on the study termination day (day 28) to characterize the immune profiles of T cell subsets after immunizations (Figure 4(A)). Immediately, peripheral blood mononuclear cells (PBMCs) were isolated from whole blood for the immunophenotypes by flow cytometry (Figure 4(B,C) and Supplementary Figure 5) after staining them with different T cell subsets and activation markers (Supplementary Table 2 and gating strategy in Supplementary Figure 4), by following method described below.

For the VPCS stimulation study, after VPCS stimulation, the single cells (isolated from various tissues as described in the above VPCS activation study section) were centrifuged at 500g for 10 min (at 4°C), followed by supernatant collection and preserved at −20°C for cytokine analysis. The cell pellets obtained were washed and collected in flow buffer (PBS supplemented with 2% FBS) and surface stained for 30 min at RT (under dark), with T cell markers as depicted in Supplementary Table 2. Following surface staining, cells were washed thrice and fixed with 2% PFA (2% formaldehyde in PBS) for 30 min at RT. The cells were permeabilized using Cytoperm/Cytofix (BD Biosciences; Franklin Lakes, NJ) for 10 min at RT (under dark), and intracellular staining was performed by incubating them with anti-mouse FOXP3 antibody for 30 min at RT (under dark). The surface and intracellular stained cells were washed, fixed in 1% PFA (10 min at RT, under dark), and washed again with flow buffer for flow cytometry analysis. Flow cytometry analysis was performed on a multicolour CytoFLEX LX cytometer (Beckman Coulter Inc., Pasadena, CA) operating under CytExpert Software. Data obtained was analysed by FlowJo_v10.8.1 software (Ashland, OR) (Figures 4 and 5) and the gating strategy is detailed in Supplementary Figure 4.

Figure 5.

VPCS-specific immune response evaluation study. The graph presented the comparative immunophenotypic difference observed in CD8+ T-lymphocytes upon 18 h VPCS stimulation of PBMCs (A), single cells from draining lymph nodes (B), spleen (C), cervicovaginal tissue (D), and small intestine (E), isolated from respective tissues collected from two weeks post-prime and -boost immunization mice (day 28). The statistical difference was analysed by two-way ANOVA followed by Bonferroni’s multiple comparisons test (post-test). The graph presents all data obtained, with a median line. Data presented were obtained from eight mice for the MEVPCS-mRNA LNP group and six mice for the untreated Control (PBS). Each data point from the respective group was presented by their respective symbols (solid circle: MEVPCS-mRNA LNP group; hollow downward triangle: untreated control (PBS group)). The asterisk mark “*,” “**,” and “****” represent p-value “<0.05,” “<0.01,” and “<0.0001,” respectively.

Cytokine profiling

The relative expression of pro-inflammatory and immunomodulatory mediators up on VPCS activation was evaluated using the Proteome Profiler Mouse Cytokine Array Kit, Panel A (R&D Systems, Minneapolis, MN, USA) following the manufacturer’s protocol. Briefly, the supernatants collected from VPCS-activated PBMCs from treated (MEVPCS-mRNA LNPs or Blank LNPs) and control (PBS-treated mice), as described in the previous section, were incubated with the Mouse Cytokine Array Panel A detection antibody cocktail for 1 h at RT. The above mix was added to the nitrocellulose membrane (containing 40 different capture antibodies spots in duplicate) after blocking, and the membrane was incubated O/N at 4°C on a rocking platform. The membranes were washed thrice with the washing buffer and then incubated with Streptavidin-HRP for 30 min on a rocking platform shaker. After washing, the membrane was developed using Chemi Reagent Mix and imaged under ChemiDoc MP Imaging System. Each membrane was imaged with 5 min exposure time to maintain uniformity. The images obtained were analysed by ImageJ software (National Institutes of Health) and data was analysed using GraphPad Prism version 9.4.1 software (La Jolla, CA, USA) (Figure 6). The data obtained from the MEVPCS-mRNA LNP and LNP groups were normalized against untreated control (PBS) groups and were presented in the heat map (Figure 6(A)), and comparative of major pro-inflammatory and anti-inflammatory cytokine/chemokine profile is presented in a graph (Figure 6(B,C)).

Figure 6.

Comparative cytokine and chemokine profile of supernatant from VPCS-stimulated PBMCs. (A) Heatmap showing net fold change (relative to the untreated control group) in 40 tested cytokine/chemokine levels in MEVPCS-mRNA LNP group vs. LNP group. Graph demonstrating the relative level of secreted pro-inflammatory (B) and anti-inflammatory (C) cytokines/chemokines. The heat map data represents a geographical mean (A) whereas the graphical data (B–C) are presented as mean ± SD obtained from MEVPCS-mRNA LNP group (n = 6), LNP group (n = 3), and untreated Control (PBS) (n = 3). The two-way ANOVA statistical analysis followed by the post-test Bonferroni’s multiple comparisons test was used to predict the significant change. The asterisk marks “*” and “**” represent p-value “<0.05” and “<0.01,” respectively.

Statistical analysis

All statistical analyses presented were analysed by GraphPad Prism version 9.4.1 software (La Jolla, CA, USA). All study results presented are expressed as mean ± SEM of data from multiple sets of independent experiments/samples. The specific test used and the number of mice per group evaluated are indicated in the respective description section and in the figure legends.

Ethical statement

The animal experiments were conducted following the approved protocol by the Institutional Animal Care and Use Committee (IACUC) at the University of Nebraska-Lincoln (UNL).

Results

Design and characterization of MEVPCS-mRNA LNP vaccine

The MEVPCS-mRNA LNP-based HIV vaccine was designed to deliver MEVPCS-mRNA intracellularly to ensure all 12 conserved VPCS expression as a single polypeptide unit (MEVPCS) (Figure 1(A)). The novel MEVPCS polypeptide was constructed by connecting each of the 12 VPCS peptide sequences via the Gly-Gly-Ser (GGS) spacer sequence (Figure 1(B)). GGS linkers provide flexibility to the MEVPCS polypeptide chain, facilitating antigen processing through junction breaking [14]. The 12 highly conserved VPCS sequences were designed from the SIVmac239 genome (twelve 20-mer VPCS peptides, i.e. 10 amino acids flanking each side of the 12 VPCSs) [8,9] with the consideration of evaluation in the preclinical macaque model. The MEVPCS immunogenicity was further enhanced by designing MEVPCS-mRNA delivered via LNPs (Figure 1, Supplementary Figure 1).

The LNPs loaded with MEVPCS-mRNA molecules were formulated following the well-established lipid composition of mRNA-1273 vaccine. However, we modified the mRNA-LNP formulation method to prolong mRNA’s cold-chain (4–8°C) stability in mRNA-LNP. The standard microfluidic-mixture-based mRNA LNP production method does not include organic phase evaporation or any other modification to prevent adduct formation. In our modified mRNA LNP synthesis protocol, we introduced two modifications in the standard microfluidic-mixture-based LNP production method, i.e. organic phase removal/evaporation step and dialysis of the mRNA LNPs against modified PBS (Figure 1(D)). The first modification improves the molecular packaging (maturation) and compactness of lipid molecules in LNP (Supplementary Figure 2(A)), contributing to uniformity and stability, as evident from hydrodynamic diameter and poly-dispersity index (PDI) data (Supplementary Figure 2 and Table 1). Maturation of LNPs limits/reduces inner-core interactions with buffer ions, this minimizes RNA adducts, aldehydes, and peroxides formation during cold storage owing to buffering ions induced hydrolysis, oxidation, or reactions resulting of ionizable lipids, PEG-lipids, and RNAs [13]. The second, after organic phase removal was dialysis of LNPs against a modified PBS buffer (detailed in Method section). The modified PBS buffer with ions and aldehyde chelators, scavenges free ions responsible for impurity production during prolonged 4−8°C storage. Thus, as depicted in Figure 1(E), Supplementary Figure 2 and Table 1, with <5% deviations between batches of LNPs in size, PDI, surface charge, concentration, and morphology indicated that the modified LNP methodology is very efficient and reproducible.

To estimate the modified protocol’s stability potency, we synthesized mRNA LNPs by standard microfluidic- mixture-based mRNA LNP production protocol, i.e. without ethanol removal step (before dialysis) followed by PBS dialysis (unmodified) and compared the size, PDI, surface charge, and concentration of mRNA LNPs over time at 4–8°C (Supplementary Figure 2). The result demonstrated that the presence of ethanol in mRNA LNPs solution causes a broadening of particle distribution/area (50 –400 nm), compared to ethanol-removed mRNA LNP (50–100 nm), supporting the fact that ethanol removal promotes LNP maturation and compactness (Supplementary Figure 2(A–C)). In the presence of ethanol, over time (day 0, 7, and day 15 at 4–8°C), the trend of broadening of size distribution/area and increase in surface charge continues in mRNA LNP/w ethanol (Supplementary Figure 2(B,C)). The increase in surface charge (a negative to positive shift) indicates the unloading/degradation of loaded mRNA upon cold storage, as reported in various previous studies [13]. However, no significant difference in PDI and concentration were observed (Supplementary Figure 2(D)), indicating ethanol causes LNP swelling (promotes structural openness of LNPs). Consequently, the swelling effect increases the ionic interaction/buffer ion exchange with the LNPs inner RNA core, promoting RNA degradation probability. Whereas, stable MEVPCS-mRNA LNPs over time (0-, 30-, and 60 days between 2°C and 8°C) indicated a non-significant change in size, LNP concentration, and surface charge (Figure 2). We followed the physicochemical stability after 10 months at 2–8°C, we observed non-significant change in the physiochemical characteristic of mRNA LNP (w/o ethanol+in modified PBS, Supplementary Figure 3(A)) indicating that modified LNP formulation methodology improves compaction of LNPs, reducing ionic interaction/buffer ion exchangeability potentially prolonged mRNA stability in mRNA LNPs.

Figure 2.

MEVPCS-mRNA LNP stability study over 2 months at 4–8°C. The stability was analysed based on three variables, i.e. size (nm), concentration (LNP/mL), and surface charge (ζ-potential, mV). Each variable shows the mean ± standard deviation of means (SEM) of six MEVPCS-mRNA LNP batches. Each data point in the graph represents data from one MEVPCS-mRNA LNP batch. Statistical analysis (two-way ANOVA with post-Šídák’s multiple comparisons test) showed no significant differences between 0-, 30-, and 60-day study time in all tested variables.

Further, we evaluated in-vivo the functional efficacy of mRNA (GFP expressing mRNA) in mRNA LNP after 10 months at 2–8°C (Supplementary Figure 3(B)) based on GFP expression after 7 days of administration. Our study indicated, both liver (the sink organ) and spleen (HIV target site), still showed expression of GFP. The functional efficiency study further supported our observation that modified LNP formulation protects loaded mRNA from degradation.

The in-vitro efficiency of MEVPCS-mRNA LNP in translating MEVPCS polypeptides

The potency of MEVPCS-mRNA LNP as a vaccine candidate primarily depends upon the MEVPCS-mRNA expression efficiency, which in turn ascertains MEVPCS-derived VPCS’s presentation effectively by dendritic cells. First, we evaluated the in-vitro translation efficiency of nucleoside-modified MEVPCS-mRNA into MEVPCS peptides. Figure 3(A) shows MEVPCS-mRNA efficiently expressed MEVPCS polypeptide (26 kDa), whereas non-specific mRNA treated and untreated HEK293 T cells lysate did not show MEVPCS band. Next, the in-vitro MEVPCS expression efficacy of MEVPCS-mRNA LNP in the HEK293T cells was assessed based on western blot assay. After 24 h of treatment with different concentrations of MEVPCS-mRNA, MEVPCS-mRNA LNP-treated cells expressed a higher level of MEVPCS polypeptide only at the highest concentration (2 µg), compared to MEVPCS-mRNA transfected cells (Figure 3(B,C)). Interestingly, after 72 h treatment, MEVPCS-mRNA LNP-treated HEK293T cells showed a steady expression of MEVPCS polypeptide over the entire concentration (Figure 3(C)). Further, the MEVPCS-mRNA LNP treatment showed enhanced and sustained MEVPCS expression. Next, the possible cross-reactivity of anti-β-actin antibody with the MEVPCS peptides band was evaluated (Supplementary Figure 1(E,F)). No secondary band was observed at β-actin level in the blot-1 (upper panel) with anti-VPCS antibodies cocktail only, and no second band at the anti-β-actin antibody-treated blot-2 (lower panel) ruled out non-specific binding. Evidently, MEVPCS cellular expression (at all tested concentrations) was consistently lower in MEVPCS-mRNA LNP treated and MEVPCS-mRNA transfected HEK293T cells compared to MEVPCS in-vitro translation (1 µg MEVPCS-mRNA). It may be because MEVPCS undergoes endogenous antigenic processing for peptide-MHC presentation (pMHC). As expected, cellular antigen expression was lower compared to in-vitro translation from mRNA, which is very common to any vaccine antigen.

MEVPCS-mRNA-LNP vaccine and overall immune responses

The VPCS vaccine study on nonhuman primates (NHPs) demonstrated non-significant induction of generalized immune activation and inflammation, but induced VPCS-specific CD8+ T cell immunity, promoting a protective immune response against HIV infection [7,9]. Thus, we evaluated the validity of the above presumption for the MEVPCS-mRNA LNP as a prophylactic HIV vaccine candidate in the in-vivo mouse model. The result demonstrated that MEVPCS-mRNA-LNP compared to LNP (blank control), and untreated control group (PBS control) did not significantly alter the bulk T cell populations, i.e. CD4 and CD8 T cells (Supplementary Figure 5(A)). Further, immune phenotyping of T cell subsets revealed MEVPCS-mRNA-LNP vaccine significantly enhanced central memory (Tcm) CD8+ T cell population compared to LNP and untreated control group (Figure 4(C), and Supplementary Figure 5(B) right panel). However, no significant changes in CD4 T cell number, activation, and differentiation state were observed between MEVPCS-mRNA-LNP and control groups (Figure 4(B) and Supplementary Figure 5). The present study demonstrated that MEVPCS-mRNA LNP vaccine did not induce generalized immune activation, including CD4 T cell activation, however, CD8+ T cell memory was promoted.

MEVPCS-mRNA-LNP vaccine-induced VPCS-specific CD8+ T cell immunity

Previous studies have shown that mRNA LNP-based vaccines can induce protective memory CD8+ T cells against viral infection [20]. Our previous VPCS vaccine study in NHP also showed that VPCS induces memory CD8+ T cells against SIV infection [9]. We next evaluated the VPCS-specific T cell immunity induced by MEVPCS-mRNA-LNPs immunizations in peripheral blood mononuclear cells (PBMCs) and various tissues such as cervicovaginal tissue, small intestine, spleen, and lymph nodes (Figure 5).

The results demonstrated that VPCS peptide stimulation of cells from mucosa tissues (cervicovaginal and small intestines) and PBMCs of MEVPCS-mRNA-LNP group significantly increased CD8+ T cells activation, central memory (Tcm), and CTLs (degranulation) subsets (Figure 5(A,D,E)).

Further, VPCS peptide stimulation of single cells from secondary lymphoid organs (e.g. spleen and lymph nodes) also resulted in a similar T cell activation pattern in the MEVPCS-mRNA LNP group. Upon VPCS peptide stimulation, the lymph nodal cells from the MEVPCS-mRNA LNP group had a higher frequency of VPCS-specific CD8+ T cells activation. Among those CD8+ T cells, activated CD8+ T cell subsets (CD38+CD69+) frequencies were significantly higher, with no change in CD8+ T cell memory subset frequency (Figure 5(B)). However, VPCS-peptide-stimulated-isolated splenocytes demonstrated higher frequencies of CD8+ T cell activated, Tcm, and CTLs populations (Figure 5(C)). Further, the splenocytes upon VPCS peptide stimulation did not change overall CD4+ T cell frequency (Supplementary Figure 6(C)).

The above result indicates that MEVPCS-mRNA LNP vaccine immunizations-induced VPCS-specific CD8 T cell immune activation and memory responses in the MEVPCS-mRNA LNP group, similar to the VPCS vaccine study in the NHP model [9]. The high frequencies of VPCS peptide-specific CD8+ T cell activation, Tcm, and CTLs population immunophenotype at the tested HIV target sites (Figure 5) demonstrated that MEVPCS-mRNA LNP immunizations elicited strong VPCS-specific CD8+ memory T cells to prevent future SIV/SHIV infections.

MEVPCS-mRNA LNPs immunizations did not activate the CD4+ T cells

The VPCS vaccine NHP study showed that the PCS vaccine induced a low frequency of SIV-targeted CD4+ T cells [9]. The MEVPCS-mRNA LNPs group also induced low to no change in CD4+ T cell frequency in the PBMCs before (Supplementary Figure 5(A)) and after VPCS stimulation (Supplementary Figure 6(A)). The same observation was valid at all other studied HIV target tissue sites (Supplementary Figure 6(B–E)). The VPCS vaccine NHP study also reported a low frequency of naïve T cell population in systemic circulation upon VPCS stimulation [9]. Similarly, the VPCS stimulation of PBMCs from the MEVPCS-mRNA LNPs group showed a lower frequency of VPCS-peptide-specific naïve T cell population among both CD4+ and CD8+ T cell population (Supplementary Figure 7). Moreover, VPCS stimulation of cells from all the HIV major entry sites demonstrated a non-significant change in CD4+ T cell activation and in effector/effector-memory T cell frequency in all the HIV target organs (Supplementary Figure 8). Thus, altogether the result indicates that MEVPCS-mRNA LNPs immunization does not increase and activate HIV target CD4+ T cells, naïve and effector population. Of note, although comparatively VPCS stimulation of PBMCs and splenocytes resulted in no change in the CD4+ T cell population, the CD4+ T cell subset comprised of higher frequency of Tcm cells in PBMCs, as well as splenocytes had a higher frequency of activated CD4+ T cells (Supplementary Figure 6(C)), compared to the control group.

MEVPCS-mRNA LNP generates a VPCS-specific T-polyfunctional immune response

The concept of “Polyfunctionality,” the ability of antigen-specific T cells to express multiple cytokines and chemokines after target-peptide stimulation, is frequently employed in vaccine research. However, the correlation between “polyfunctionality” and the effectiveness of a vaccine, particularly in the case of HIV vaccines, is yet to be fully explored. Despite this, the secretion of cytokines following vaccine immunogen stimulation is regarded as a reliable marker of an antigen-specific immune cell response. Various studies have demonstrated that the potency of protective T cell immunity in controlling HIV infection is correlated with T cell polyfunctionality [21]. Pathogen-specific T cell polyfunctionality correlates with the frequency of effector molecules (cytokines/chemokines), a strong predictor of CTL efficiency [22]. Studies on the HESN population have also revealed that the elevated levels of HIV-specific polyfunctional CD8+ T cells and associated effector molecules are responsible for robust virus-inhibiting and keeping the plasma viral load undetectable [23]. Thus, we next evaluated the VPCS-induced T cell polyfunctionality based on the VPCS-stimulated T cell’s ability to secrete cytokines. The heat map depicted (Figure 6(A)), compared to the blank LNP group, the PBMCs from the MEVPCS-mRNA LNP group secreted more diverse cytokines, indicating MEVPCS-mRNA LNP indeed participated in inducing T cell polyfunctional response.

Further analysis of major cytokines demonstrated TNF-α, INF-γ, RANTES, MIP-1α, MIP-1β, IP-10, Interleukin-17 (IL-17), IL-1α, and IL-12p70 were secreted significantly higher in MEVPCS-mRNA LNP group compared to control groups (Figure 6(B)). Indeed, TNF-α, IFN-γ, and MIP-1β are key primary effector molecules that contribute to the T cell polyfunctionality, especially promoting activation of HIV-specific CD8+ T cells and associated killing [23]. RANTES, MIP-1α, and MIP-1β molecules (HIV-suppressive factors) are produced by activated CD8+ T cells and potentially suppress HIV infection by competitively binding to HIV coreceptor CCR5[23]. IL-12p70, another essential cytokine secreted by DCs, promotes expanding functional VPCS-specific CD8+ T cells [24]. The IP-10 [25] and IL-1α [26] help augment CTL effector and memory responses. The IL-17 expressed from subpopulations of CD8+ T cells promotes several beneficial effects in the intestinal tract to protect from mucosal injury [27]. Moreover, VPCS peptide-stimulated PBMCs produced significantly higher IL-7 and IL-13 cytokines compared to control groups (Figure 6(C)). High IL-7 is essential for T-lymphocyte proliferation/survival during adaptive immunity restoration in HIV patients under ART [28], and IL-13, unlike other anti-inflammatory cytokines, controls inhibition of TNF-α and IL-1β [29]. The cytokine profile data indicates that MEVPCS-mRNA LNPs immunization upon antigenic encounters stimulates and produces effector molecules associated with T-polyfunctional response essential for HIV-protective cellular immune responses.

Discussion

The failure of classical vaccine strategies in protecting against HIV indicated the need for unconventional approaches for an effective HIV vaccine. The natural immunity observed in protecting the HESN group from HIV infection demonstrated that an effective HIV vaccine should include protective cellular immunity to prevent HIV infection [1,2]. Thus, our rationale for selecting VPCS as the unconventional HIV-immunogens and mRNA-LNPs as the delivery system were (1) the VPCS immunogens are highly immunogenic, conserved and functionally essential to prevent HIV rapid escape mutability [8]; (2) the immunogen and delivery system ensures induction of focused protective anti-HIV immune response [30]; and (3) the vaccine delivery system also should guarantee mRNA intracellular delivery and long-term stability/cold-chain compatibility. Thus, MEVPCS-mRNA LNP design, ensures MEVPCS cytosolic expression, endogenously antigenic processing, and MHC-I/VPCS peptide presentation by APCs, inducing VPCS-specific focused immune response against HIV infection.

Potential escape mutants or resistance can arise due to HIV’s high mutation rate, which allows the virus to rapidly evolve and evade immune responses. However, the VPCS regions are essential for HIV protease to cleavage Gag, Gag-Pol, and Nef precursor polyproteins to produce infectious viral particles and are highly conserved among all the major subtypes of HIV-1 [8]. Additionally, all the chosen 12 VPCS regions are epitopes of HLA class I alleles (MHC-I expressing APCs), which are common to >95% of the world populations [8], endorsing the immunogenic potency of VPCSs. Hence, targeting VPCS immunogen for HIV vaccine strategy has several advantages, e.g. (i) focused immune response to the conserved VPCSs would lyse initial virally infected cells; (ii) HIV-1 could be forced to mutate at VPCSs in response to positive selection pressure, resulting in non-infectious virion production; (iii) With VPCS, immune response will be restricted to reduce distracting immune responses, i.e. fewer inflammatory responses and HIV-1 targeted CD4 T cells accumulation.

We design MEVPCS based on SIV/SHIV VPCSs (derived from SIVmac239 stain) [8,9] keeping our next step in mind, i.e. to evaluate the MEVPCS-mRNA LNP vaccine in an NHP/SIV infection model. To ensure the expression/differential presentation of all 12 VPCS immunogen within APCs, MEVPCS immunogen was designed by stringing all 12 conserved VPCS immunogen in a single unit (Figure 1(B)). Further, to strategically increase the immunogen payload and to ensure immune response towards all of the functionally conserved VPCSs, a step up in the central dogma strategy was considered, i.e. instead of delivering protein immunogen, delivering immunogen-expressing mRNA (MEVPCS-mRNA). The MEVPCS-mRNA design ensured multi-epitope polypeptide expression (all 12 individual PCSs-antigen), which upon host-proteolytic processing, produces VPCS antigens for MHC presentation. Another rationale behind designing MEVPCS-mRNA was that dendritic cells preferentially process long synthetic peptides better for antigen/MHC-I presentation and CD8 T cell activation [31]. The colocation of epitopes common to the HLA class I allele, with the cathepsin cleavage sites within the VPCS peptides, makes MEVPCSs more likely to generate MHC-I-specific epitopes [8].

MEVPCS-mRNA was loaded in LNP composition similar to mRNA-1273 vaccine, to achieve enhancing mRNA-LNP vaccines’ in-vivo stability and efficacy, however, our modified formulation methodology was able to overcome limited physical stability of mRNA-LNPs during cold-chain storage [17]. LNP vaccine stability was a primary concern during pandemics since COVID-19 vaccines must be frozen (Moderna: −15°C to −25°C, BioNTech/Pfizer: −60°C to −90°C) [17]. However, HIV is most prevalent in low-to-middle-income countries, where medical facilities work under limited resources. Thus, to maximize the reach of the vaccine module in medical facilities of limited resources, an HIV vaccine needs to render a sufficiently long shelf life preferably at cold-chain temperatures (2–8°C). Studies have shown that mRNA LNPs produce impurities (like peroxides and aldehydes adducts) upon 4°C storage due to the labile nature of mRNA and lipids, affecting the long-term stability of mRNA LNP-based vaccine, a limitation of nucleic acid-based medicines [13]. To improve and prolong the stability of mRNA LNP at 4–8°C, we modified the formulation methodology to achieve long-term stability. The stability study showed no significant changes in mRNA LNP’s physicochemical characteristics over 10 months of storage between 2°C and 8°C (Figure 2) and in vivo target protein expression from mRNA (GFP expression, Supplementary Figure 3), demonstrating modified protocol prolongs mRNA LNP physicochemical and in vivo functional stability. The major modification involved were the organic phase removal step (reverse-phase evaporation helps ease scale-up) and the use of modified PBS during buffer exchange step (Figure 1(D)). The organic phase (ethanol) removal promotes lipid compactness and maturation of LNPs (Figure 2 and Supplementary Figure 2). However, the presence of ethanol showed swelling effect (Supplementary Figure 2(B)), i.e. less compaction of the LNP layer which increases the chances of ionic exchange between inner core (mRNA) and outer buffer, and ionic interaction of mRNA with buffer ions, undergo RNA abducts and other impurities generation [13]. Whereas ionic composition of modified PBS (supplemented with low concentration of antioxidant and ion-chelator) further ensures the stability of mRNA in injectable physiological solution by reducing the possibility of accumulation of RNA adduct and impurities. Thus, compact mRNA LNP complex in a modified buffer reduces the chance of ionic interaction with the ionizable lipids and mRNA, which are responsible for the production of impurities, such as aldehydes, and peroxides due to hydrolysis, oxidation, or reaction of ionizable lipids, PEG-lipids and RNA adducts [13].

We found through serendipity that poly-U mRNA helps maintain a stable and minimal level of intracellular antigens over an extended period. It is known fact that the presence of poly(A) tail at 3′ end of mRNA is essential for efficient and high-level of mRNA to protein translation. Poly(A) tail essentially play an important role, i.e. nuclear to cytoplasmic translocation [32] and interacts with poly(A) binding proteins (PABPs) stabilizing mRNAs, which enhances mRNA to protein translation. At the same time, poly(A) tailed mRNA is also subjected to canonical de-adenylation-dependent decay promotes mRNA degradation [33]. Our aim was to mimic the translocation and stability aspect of poly(A) tail, as well as its mRNA degradation regulatory aspect to lower antigenic burden promoting MHC-I mediated CD8+ T cell showed response over CD4+ T cell-mediated response. We introduced poly(U) tail over poly(A) tail. The LNP-mediated cytoplasmic delivery of mRNA ensures slow but sustained cytoplasmic release of mRNA mimicking poly(A) tail’s cytoplasmic translocation property. Next, the prolonged stability and at the same time mRNA degradation property was achieved by LNP’s sustained release property (sustained polypeptide expression until 72 h, Figure 3 and Supplementary Figure 1(E,F)). Moreover, various investigations indicated that viral (exogenous) RNA with poly(U) sequence uses poly(U) tail to interact with poly(A) tails of host mRNA, inhibit the association of poly(A)-binding protein (protect viral RNA from exonucleases), thereby increasing the stability of the RNA within the host cell [34]. The mechanism involving the poly(U)/poly(A) tail engagement in double-stranded structures with other RNA elements in viral genes has proven to be associated with their increased stability [34–36]. Therefore, secondary doubled standard mRNA poly(U)/poly(A) tail (host) engagement may also have contributed to increased MEVPCS-mRNA stability upon introduction into ectopic transcripts, which may also have contributed to MEVPCS expression over 72 h. Recently discovered pathway indicated that the presence of poly(U) residues are interpreted as an RNA degradation signal by the cell, promoting endogenous (host) mRNA degradation pathway, regulating intracellular mRNA cellular level by non-canonical pathway [37]. We presume that all the above factors may have contributed to low-level but sustained expression of MEVPCS expression over 72 h. Based on reported studies as stated above, we presume that each modification step introduced would have contribution overall stability of mRNA LNPs.

Our next step was to investigate the MEVPCS-mRNA LNP’s intracellular MEVPCS polypeptide expression efficiency to ensure the generation of variable VPCS epitopes within APCs for MHC-I presentation [8]. The in-vitro MEVPCS expression kinetics study showed MEVPCS-mRNA LNPs achieved a sustained and prolonged MEVPCS expression (Figure 3), possibly due to LNP’s prolonged sustained cargo-delivery property. Another interesting observation was that regardless of the MEVPCS-mRNA treatment amount, the MEVPCS expression level never increased significantly higher when compared to IVT (Figure 3(B)). As discussed above most probable reason behind the low MEPVPCS level can be attributed to poly(U) tail and LNP-mediated sustained mRNA release could have established an equilibrium between the mRNA expression (translation capacity) and degradation pathway, that reached the maximum expression level, for that reason we see the low but steady level of MEVPCS expression (Figure 3 and Supplementary Figure 1(E,F)). In addition, as predicted, MEVPCS intracellular pools may have reduced due to cleavage by host proteasomal cathepsins D, S, and L of MEVPCS polypeptides, resulting in VPCS epitopes (epitopes of 8–10 amino acids as predicted by NetChOP IEDB Analysis Resources) for MHC-I presentation [8]. Moreover, low, slow, but steady antigenic polypeptide expression may be beneficial in promoting the MHC class I processing, presenting machinery for peptide loading, and providing substantial time to host proteasomal-cathepsin pathways for antigenic processing [38]. Additionally, other study has demonstrated single-injection vaccination that extended delivery of an antigen prolonged antigen presentation period, thereby have demonstrated generating a stronger immune response to the same dose [39]. Interestingly, as reported the prolonged but low antigen presentation also showed overall CD8+ active T cells skewed response [40], which aligns with our observations (Figures 4–6). All this observation justifies our result indicating CD8+ T cell skewed immune response. Prolonged low antigen exposure can favour the differentiation of CD8+ T cells into memory cells rather than effector cells. While memory T cells are important for long-term immunity, this might come at the cost of an immediate strong effector response. Our study strategic unconventional approach for prophylactic HIV vaccine could be an alternative effective way to establish long-term protective immunity against HIV.

The NHP model study has shown that the VPCS vaccine has been one of the best protective vaccines for the NHP/pathogenic SIVmac251 model [9]. The VPCS vaccine has been the only reported HIV vaccine strategy that has elicited narrowly focused immune responses to VPCSs, similar to HESNs [8]. Our strategy of MEVPCS-mRNA LNPs as a prophylactic HIV vaccine shows similar narrowly focused CD8+ T cell-mediated immune responses in the in-vivo mouse model (Figure 5). Evidently, MEVPCS-mRNA LNPs prime and boost immunizations-induced antigen-specific CD8+ central memory (Tcm) response (Figure 4(C)). Additionally, multiple epitopes generated from MEVPCS polypeptide unit predicted to be epitopes of HLA class I alleles [8], thus, could induce strong MHC-I directed CD8+ T cell-mediated immune response. A recent study on Merck AD5/HIV vaccine failure revealed that compared to chronically HIV-infected long-term nonprogressors/HECs and progressors, the T cell receptors of vaccinees showed insufficient sensitivity, impairing efficient degranulation in response to low pMHC levels on HIV-infected CD4+ T cells [11]. However, our strategy of promoting multi-epitope production and multimer VPCS-pMHC presentation on individual APCs will be able to overcome the low antigen receptor sensitivity limitation. Predictively, the reasons mentioned above could be responsible for the significantly elicitation of VPCS-peptide-CD8+ specific T cell cytotoxic (CTLs) and central memory (Tcm) phenotype at all studied HIV target tissue sites of MEVPCS-mRNA LNP vaccinated group (Figure 5). We previously studied the correlates of protection of live attenuated vaccine (LAV) SIVmac239Δnef (SIVΔnef) in NHP. The LAV vaccine confers the best protection among all the vaccine modalities tested in rhesus macaque model of HIV-1 infection. This vaccine has a unique feature of time-dependent protection: macaques are not protected at 3 ± 5 weeks post-vaccination (WPV), whereas immune protection emerges between 15 and 20 WPV. We observed that at 20 WPV, the most significant correlates of protection are antigen-specific central memory CD8 T cells [41]. Distinct transcriptome profiles of Gag-specific CD8+ T cells temporally correlated with the protection elicited by SIVΔnef live attenuated vaccine. Thus, a potent VPCS-specific CD8+ T cell memory for inducing a narrowed immune response to kill founder virus-infected cells is essential to promote prophylactic protection against HIV-I infection.

The MEVPCS-mRNA LNP immunization also significantly contributes to T-polyfunctionality, essential for the strong cytotoxic response. VPCS-specific T cells produce significantly elevated levels of TNF-α, IFN-γ, and MIP-1β (<5- to 60-fold higher) (Figure 6(A,B)), key cytokines associated with CD8+ T cell polyfunctionality [21–23]. Additionally, elevated levels of IP-10, IL-12p70, and IL-1α pro-inflammatory cytokines augment CTL effector and memory responses [24–26]. The enhanced primary and other supportive cytokines (IL-12p70, IL-1α, IP-10, and IL-17) establish T cell polyfunctionality, a cytokine profile similar to HESN/HEC group’s immune response during the HIV challenge [42]. Polyfunctional T cells play a crucial role in supporting the activation and killing capacity of HIV-specific CD8+ T cells. In line with findings in HESNs, MEVPCS-mRNA LNPs triggered an increased occurrence of polyfunctional HIV-1-specific T cell responses [43–45]. This has the potential to play a significant role in protecting against mucosal HIV-1 transmission during the establishment of initial infection of a single/few founder viruses [46]. Therefore, narrowed VPCS-specific CD8+ T cell polyfunctionality elicited by the MEVPCS-mRNA LNP vaccine might be enough to prevent HIV establishment and propagation of founder virus infection.

Increased availability of CD4+ T cells and their activation at mucosal sites could increase the risk of HIV acquisition [47]. However, we did not observe cytokine-induced “bystander-activated CD4+ T cells” in the neighbouring VPCS-stimulated cervicovaginal and small intestinal single cells (Figure 5). Additionally, elevated levels of IL-7 and IL-13 anti-inflammatory and IL-17 pro-inflammatory cytokines response indicate establishing a feedback mechanism to control the robust inflammatory response and protective immune response (Figure 6(A,C)). Further, with low or no increase in VPCS-specific CD4+ T cells along with naïve and Te-em population at the primary infection site (cervicovaginal cells, intestinal cells) and peripheral blood (PBMCs) (Supplementary Figures 5–7) indicates MEVPCS-mRNA LNP immunizations induces reduced immune activation and inflammatory response in CD4 T cells. Unlike classical HIV vaccine immunization, the MEVPCS-mRNA LNP group showed lower frequencies of Treg cells and activated CD4+ T cells, suggesting potential limitations in promoting robust inflammation or Th cell-mediated immune activation (Supplementary Figure 6(A)). Thus, the MEVPCS-mRNA LNPs as a vaccine candidate may restrict HIV target availability.

We would also like to point out the limitations of this study. The MEVPCS-mRNA LNPs in this proof-of-concept study were designed to produce SIV MEVPCS to evaluate the MEVPCS-mRNA LNPs vaccine effectiveness in the NHP model in preventing infection of a highly stringent NHP/SIVmac251 mucosal challenge. The NHP/SIVmac251 model will help understand the pharmacokinetics, prophylactic, and active vaccine potential of MEVPCS-mRNA LNP’s multi-epitope PCSs targeting strategy before designing HIV MEVPCS to target HIV for clinical evaluation in humans. Alongside, it will be interesting to evaluate MEVPCS-mRNA LNPs potential in inducing long-term immunity. Long-term immunity is a key requirement for an effective mRNA-based vaccines, given the known challenge of short-term immune memory in mRNA vaccines.

In conclusion, the design of novel MEVPCS-mRNA LNP in this study delivers all 12 VPCS immunogens intracellularly, ensuring a high expression and efficient endogenous process of MEVPCS polypeptides as an intracellular immunogen, enhancing the probability of multivariant MHC-I restricted VPCS peptides presented on APCs, but not on MHCII. Thus, MEVPCS immunogens increase the chances of the CD8+ T cell-mediated polyclonal immune response, not the CD4+ T cell response. The proof-of-concept study in a mouse model demonstrates that as an HIV vaccine candidate, MEVPCS-mRNA LNP expressing MEVPCS polypeptides promotes an immunogenic memory restricted to sequences surrounding the 12 viral VPCSs. As observed in previous reports VPCS-mediated immune response has significantly protected NHP from pathogenic SIVmac251 intravaginal acquisition, our study also demonstrated that MEVPCS-mRNA LNP immunization induced VPCS-specific CD8+ T cell immune memory and activated VPCS-specific CD8+ T cells with enhanced degranulation capacity (CTLs) and immunologic memory (Tcm), with low to no VPCS-specific CD4+ and Treg cells, during HIV challenge, thus potentially could overcome low antigen receptor sensitivity of HIV vaccine induce CD8+ T cells response [11]. Further, the MEVPCS-mRNA LNP VPCS-specific CD8+ T cells immune response will likely produce the T-polyfunctional response (cytokines) promoting protective anti-HIV immunity to defeat HIV-1 infection. Thus, as a prophylactic HIV vaccine candidate, MEVPCS-mRNA LNPs could induce VPCS-specific polyfunctional CD8+ T cell immunity to prevent/kill HIV-infected cells. Importantly, our modified mRNA LNP are long-term stable and cold-chain friendly. As observed in HESN, a narrowly focused, well-maintained virus-specific CD8+ T cell response induced by HLA class I specific VPCS multi-epitopes could be sufficient to destroy and eliminate a few founder viruses without inducing inflammatory responses and CD4+ T cells (targets for HIV-1). Hence, MEVPCS-mRNA LNP as an unconventional alternative HIV vaccine could be a feasible vaccine approach.

Funding Statement

This study was partly supported by the NIH/NIAID R01AI111805 (to M. L.) and a Task Order under HHSN272201800007C from the Contract Officer Representative (COR), Preclinical Research Development Branch, VRP, DAIDS (Q. L. and M. L.).

Disclosure statement

No potential conflict of interest was reported by the author(s).