A modified recombinant adenovirus vector containing dual rabies virus G expression cassettes confers robust and long-lasting humoral immunity in mice, cats, and dogs

ABSTRACT

During the COVID-19 epidemic, the incidence of rabies has increased in several countries, especially in remote and disadvantaged areas, due to inadequate surveillance and declining immunization coverage. Multiple vaccinations with inactivated rabies virus vaccines for pre- or post-exposure prophylaxis are considered inefficient, expensive and impractical in developing countries. Herein, three modified human recombinant adenoviruses type 5 designated Adv-RVG, Adv-E1-RVG, and Adv-RVDG, carrying rabies virus G (RVG) expression cassettes in various combinations within E1 or E3 genomic regions, were constructed to serve as rabies vaccine candidates. Adv-RVDG mediated greater RVG expression both in vitro and in vivo and induced a more robust and durable humoral immune response than the rabies vaccine strain SAD-L16, Adv-RVG, and Adv-E1-RVG by more effectively activating the dendritic cells (DCs) – follicular helper T (Tfh) cells – germinal centre (GC) / memory B cells (MBCs) – long-lived plasma cells (LLPCs) axis with 100% survival after a lethal RABV challenge in mice during the 24-week study period. Similarly, dogs and cats immunized with Adv-RVDG showed stronger and longer-lasting antibody responses than those vaccinated with a commercial inactivated rabies vaccine and showed good tolerance to Adv-RVDG. In conclusion, our study demonstrated that simultaneous insertion of protective antigens into the E1 and E3 genomic regions of adenovirus vector can significantly enhance the immunogenicity of adenoviral-vectored vaccines, providing a theoretical and practical basis for the subsequent development of multivalent and multi-conjugated vaccines using recombinant adenovirus platform. Meanwhile, our data suggest Adv-RVDG is a safe, efficient, and economical vaccine for mass-coverage immunization.

KEYWORDS:

• Rabies
• adenovirus vector
• long-lasting humoral immune
• dogs
• cats
• vaccine

Introduction

Rabies, caused by rabies virus (RABV), claims nearly 60,000 lives annually and causes the loss of approximately 2.2 and 1.34 million disability-adjusted life years (DALYs) in Asia and Africa respectively [1]. Globally, the economic cost of canine-mediated rabies is $8.6 billion, which mainly includes productivity losses due to mortality (55% of total costs), the expenses of post-exposure prophylaxis (PEP), and direct costs to the medical sector and bite victims (40%) [1,2]. However, these data may need to be more accurate in terms of actual costs for high-risk populations. In 2018, WHO proposed the Zero by 2030 initiative to eliminate canine-mediated human rabies [3]. The most effective preventive measure against rabies is vaccination. Canine-mediated human rabies can be interrupted when 70% of dogs are effectively vaccinated [4].

Currently rabies vaccines are either inactivated, modified-live, or recombinant products in domestic mammals and wildlife. Of these, inactivated vaccines are the mainstay of rabies vaccines for pets and often require multiple immunizations and the addition of adjuvants to enhance their immunogenicity, which undoubtedly increases the cost of production and raises a number of safety concerns that make them unsuitable for mass-coverage immunization [5]; Modified-live rabies virus has been administered to wildlife to control rabies prevalence in Europe and North America [6]. However, rabies caused by live-vaccine has been a common occurrence [7–9]. Live vaccines are banned in China. In 1980s, a number of recombinant live-vectored rabies vaccines had emerged, mainly consisting of poxvirus and adenovirus vectors. V-RG, a recombinant vaccinia vectored vaccine expressing the RABV glycoprotein, has been used for rabies control in multiple target species [10]. However, two V-RG transmitted pox cases were reported and patients develop symptoms characteristic of poxvirus infection, suggesting there is a potential risk of V-RG leakage [11,12]. AdRG1.3, a replicate-competent adenovirus expressing ERA G protein, was distributed in rural West Virginia to eliminate the rabies in wildlife by oral route [13]. However, researchers have demonstrated replicate-competent adenovirus have the potential to cause horizontal transmission [14,15]. Besides, messenger RNA (mRNA)-based vaccines have shown great promise as a nontraditional vaccine platform against rabies [16–18]. Nevertheless, high frequencies of adverse reactions to vaccinations, had been associated with mRNA side effects [19,20]. Besides, complex production processes and specific production plants undoubtedly increase costs of mRNA vaccine. Considering the high lethality of rabies and severe economic losses in developing countries, development of safe, efficient and economical vaccines is warranted.

The replication-defective adenovirus type 5 vector, in which E1 and E3 genes have been deleted, is widely recognized as a safe, inexpensive, and efficient vector for gene delivery [21–24]. An rAd5-based human vaccine expressing Ebola G protein was approved for registration as a new drug by the Chinese Food and Drug Administration (FDA) in 2017 (NCT02326194) [25]. In 2021, the recombinant COVID-19 vaccine (Adenovirus Type 5 vector) was approved by the National Medical Products Administration (NMPA) for SARS-COV-2 prevention [24]. However, pre-existing Ad5-neutralizing antibodies in the population have somewhat hampered its prospects as a human vaccine vector [26]. Adenoviruses are strictly species-specific, and animals do not naturally become infected with human adenoviruses. Heterologous adenoviral vectors provide an excellent platform for developing animal vaccines. A human adenovirus-vectored African swine fever vaccine elicited systemic and mucosal immune responses against AFSV in pigs [27].

Increasing the antigen expression level of the vector is an effective way to produce cheaper and more efficient vaccines. Researchers have attempted to increase the expression of the exogenous protein by inserting multiple antigens into the E1-deleted region via 2A peptide ligation, but have often failed to achieve the desired effect due to incorrect cleavage [28]. In previous studies, the antigen was inserted only into the E1 genomic region of the adenovirus vector and the E3 genomic region was not efficiently utilized [13,24,25,27,29]. We hypothesized that insertion of multiple exogenous genes into the E1 and E3 regions of adenoviral vectors could significantly improve their immunogenicity as vaccines. In our study, three recombinant adenovirus-vectored rabies vaccines (Adv-RABV) were constructed. We have creatively inserted RVG into E1 and E3 genomic regions of adenovirus. The safety and immunogenicity of Adv-RABV were compared in mice. We found that a replication-defective adenovirus carrying two copies of G genes, named Adv-RVDG, elicited a significantly more robust and durable humoral immune response and provided long-lasting protection against a lethal RABV challenge in mice. Meanwhile, mechanisms of the stable antibody response were elucidated by monitoring the dynamics of immune cells in lymphoid organs after immunization. Furthermore, we investigated the immunogenicity of a single vaccination with Adv-RVDG versus a commercially inactivated vaccine (IRV) in dogs and cats. Our results indicated that Adv-RVDG induced significantly more and longer-lasting RABV-specific neutralizing antibody (VNA) than IRV in dogs and cats.

Materials and methods

Cells, viruses, antibody and reagents

HEK 293 and BSR cells, a cloned cell line derived from BHK-21 cell, were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Gibco, USA) supplemented with 10% bovine serum (Gibco) at 37°C with 5% CO₂. The rabies virus (RABV) vaccine strain SAD-L16 was obtained by reverse inheritance of the weakly virulent strain SAD-B19 (GenBank: M31046.1). The pathogenic RABV strain CVS-24 was propagated in the brains of suckling ICR mice. RABV glycoprotein (RVG) was prepared as previously described [30]. Fluorescein isothiocyanate (FITC)-conjugated antibodies against RABV nucleoprotein (RVN) were purchased from Fujirebio Diagnostics, Inc. Adenovirus hexon protein antibody (clone 3G0) was purchased from Santa Cruz Biotechnology. Horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG, IgG1, IgG2a, and IgG2b for ELISA were purchased from Boster (Wuhan, China). HRP-conjugated goat anti-dog IgG, IgG1, and IgM and sheep anti-dog IgG2 for ELISA were purchased from Bethyl Laboratories (TX, USA). Fluorescein directly labelled antibodies for flow cytometry analysis were purchased from BioLegend (CA, USA), including FITC anti-mouse CD11c antibody (clone N418), allophycocyanin (APC) anti-mouse CD80 antibody (clone 16-10A1), PE/Cy7 anti-mouse I-A/I-E (MHC-II) antibody (clone M5/114.15.2) and phycoerythrin (PE) anti-mouse CD86 antibody (clone GL-1) for analysis of mature dendritic cells (DCs); T follicular helper (Tfh) cells in inguinal lymph nodes (LNs) were marked by FITC anti-mouse CD4 antibody (clone RM4-5), APC anti-mouse CD185 (CXCR5) antibody (clone L138D7), and PE anti-mouse CD279 (PD1) antibody (clone J43); FITC anti-mouse CD45R/B220 antibody (clone RA3-6B2), 647 anti-mouse GL7 antibody (clone GL7), and PE anti-mouse CD95 antibody (clone SA367H8) were used to mark germinal centre B (GC B) cells in LNs; PE/Cy7 anti-mouse CD45R/B220 antibody (eBioscience, RA3-6B2), FITC anti-mouse IgD (clone 11-26C.2a), APC anti-mouse CD138 (Syndecan-1) antibody (clone 281-2) and PE anti-mouse CD273 (clone TY25) were used to symbolize memory B cells (MBCs) in spleen; FITC anti-mouse CD45R/B220 antibody (clone RA3-6B2) and APC anti-mouse CD138 (Syndecan-1) antibody (clone 281-2) were used to quantify the number of long-lived plasma cells (LLPCs) in bone marrow cells. In addition, biotin-labelled antihuman/mouse GL7, Alexa Fluor 647-conjugated anti-mouse/human CD45R/B220, Alexa Fluor 488-conjugated goat anti-mouse IgG, and the Alexa Fluor 594-conjugated streptavidin were purchased for immunofluorescence staining of lymph node germinal centre. The plasmids were transfected with Jetprime (Polyplus; number 23Y2307M4). RIPA buffer (P0013c), enhanced BCA protein assay kit (P0010), and enhanced chemiluminescence (ECL) detection kit (P0018M) were purchased from Beyotime Biotechnology. Commercial inactivated rabies vaccine (IRV) was purchased from Intervert International R.V. (Boxmeer, Netherlands)

Animals

Six-week-female ICR and C57BL/6 mice were purchased from the Centre for Disease Control and Prevention of Hubei Province, and bred at the Huazhong Agricultural University (HZAU) animal facility in Wuhan City, Hubei Province, People’s Republic of China. Three-month-old cats were purchased from the Jiaxiang Huarong Breeding Professional Cooperative. Three- to five-month-old beagles were purchased from and housed at Yizhicheng Biological Technology Co., Ltd., Hubei, China.

Construction of recombinant adenovirus vectored rabies vaccine

Admax system (Microbix Biosystems Inc.) was used to rescue recombinant adenoviruses and consisted of pBHGcre/loxp and pDC315 plasmid. The RABV antigens were derived from SAD-L16, obtained by reverse inheritance of the weakly virulent strain SAD-B19 as previously described [31]. PCR amplified the RABV glycoprotein (RVG) expression cassette, subcloned into the E3 region of the backbone plasmid pBHGcre/loxp linearized by Cas9 Nuclease, S. pyogenes (NEB #M0386) based on Crispr Cas9 technology using NEBuilder HiFi DNA Assembly Cloning Kit (NEB # E5520) based on Gibson seamless cloning technology to generate pBHGcre/loxp-RVG. sgRNAs were generated by in vitro transcription using the EnGen® sgRNA Synthesis Kit, S. pyogenes (NEB #E3322) and listed in Table 1. In addition, RVG, eGFP, and Adv-E1 genes were amplified by PCR and subcloned into multiple cloning sites of the shuttle plasmid pDC315 to generate pDC315-RVG, pDC315-eGFP, and pDC315-E1 respectively. To generate recombinant adenovirus, plasmids were co-transfected into HEK 293 cells, as described in Table 2. At ten days post-transfection, cell supernatants were collected and inoculated into HEK 293 cells, which were observed daily until cytopathic effect (CPE) appeared. The recombinant adenovirus construction strategy is shown in Figure S1. Four recombinant adenoviruses were rescued, designated as Adv-eGFP, Adv-RVG, Adv-E1-RVG and Adv-RVDG. Recombinant adenoviruses containing rabies virus antigen are uniformly named Adv-RABV.

Table 1. SgRNA used for Crispr Cas9.

Primer name	Sequence (5′ to 3′)
SgRNAE3-F	GATCACTAATACGACTCACTATAgcaactgcgcctgaaacaccGTTTTAGAGCTAGAAA
SgRNAE3-F1	GATCACTAATACGACTCACTATAgcgggcaaagcacttgtggGTTTTAGAGCTAGAAA
sgRNA-R	AAAAGCACCGACTCGGTGCCACTTTTTCAAGTTGATAAC GGACTAGCCTTATTTTAACTTGCTATTTCTAGCTCTAAAAC

Table 2. Plasmids for constructing recombinant adenoviruses.

Skeleton plasmids	Shuttle plasmids	Recombinant adenovirus
pBHGcre/loxp	pDC315-eGFP	Adv-eGFP
pBHGcre/loxp	pDC315-RVG	Adv-RVG
pBHGcre/loxp-RVG	pDC315-E1	Adv-E1-RVG
pBHGcre/loxp-RVG	pDC315-RVG	Adv-RVDG

Growth curves of recombinant adenovirus in HEK 293 cells

HEK 293 cells were seeded into 12-well plates and infected with the recombinant adenoviruses at a multiplicity of infection (MOI) of 0.1 (MOI = 0.1) when cell confluence reached 90%. Cell supernatants were collected at successive 12-hours intervals post-infection, and the viral titres were determined. Briefly, viral supernatants were diluted 10-fold, and 100 μl per well was inoculated into 96-well plates, with eight replicates per dilution. The supernatants were discarded and cells were washed gently twice with PBS after 72 h post-infection (hpi). The infected cells were fixed for 30 min at room temperature by 4% paraformaldehyde and permeabilized for 15 min in PBS containing 0.1% Triton X-100. After washing three times with PBS, cells were blocked with PBS containing 2% BSA for 2 h at room temperature and incubated with the adenovirus hexon monoclonal antibody (clone 3G0) for 2 h at 37°C. After washing twice with PBS, the cells were incubated with Alexa Fluor™ 488 Goat anti-Mouse IgG (H + L) Cross-Adsorbed Secondary Antibody for 45 min and washed twice with PBS. Finally, antigen-positive foci were observed under an Olympus IX51 fluorescence microscope, and the virus titre was calculated as the 50% tissue culture infective dose per millilitre (TCID₅₀/ml).

Western blotting

Western blotting was performed for the detection of RVG expression at 48 hpi. The cell samples were lysed, collected, and centrifuged. The supernatants were boiled for 10 min with SDS loading buffer. The prepared samples were separated on SDS-PAGE gels and transferred to PVDF membranes (Bio-Rad). The membranes were blocked with TBST containing 5% (wt/vol) nonfat dry milk for 3 h and probed with primary antibodies overnight at 4°C. The proteins were visualized using the BeyoECL Star kit (Beyotime; P0018A), and images were captured with an Amersham Imager 600 (GE Healthcare) imaging system.

Indirect immunofluorescence assay

HEK 293 cells were seeded into 48-well plate and infected with Adv-RABV (MOI = 0.1). Cell supernatants were discarded at 48 hpi. The cells were fixed with 4% paraformaldehyde for 30 min at room temperature and then permeabilized for 15 min in PBS solution containing 0.1% Triton X-100. After washing three times with PBS, cells were blocked with PBS containing 2% BSA for 2 h at room temperature and incubated with the anti-RVG mouse monoclonal antibody (clone 1E11) for 2 h at 37°C. After washing with PBS, the cells were incubated with Alexa Fluor™ 488-conjugated Goat anti-Mouse IgG (H + L) Cross-Adsorbed Secondary Antibody for 45 min. Washing twice with PBS, cells were incubated with 4′,6-diamidino-2-phenylindole (DAPI; 1:500 dilution in PBS) at room temperature for 5 min and then washed with PBS. The coverslips were imaged with EVOS FL Auto Cell Imaging System (Thermo Fisher Scientific).

Quantitative real-time PCR (qRT-PCR)

The tissue samples were collected at indicated time points and were ground with liquid nitrogen pre-cooled mortar. The total RNA was extracted by Trizol reagent (Invitrogen, Karlsruhe, Germany). RNA was then converted to cDNA by reverse transcription using HiScript II 1st Strand cDNA Synthesis Kit (Vazyme, R211-01). Primers for analysis of viral gene and host gene were designed as shown in Table 3, and then the qRT-PCR was performed with the following programme: 95°C for 2 min for one cycle followed by 40 cycles at 95°C for 5 s and 60°C for 30 s.

Table 3. Primers used for qPCR in this study.

Primer name	Sequence (5′ to 3′)
β-actin-F	GTACTCTGTGTGGATCGGTGG
β-actin-R	GGTGTAAAACGCAGCTCAGTAAC
RABV-G-F	GATACCAGACAAGCTTGGTC
RABV-G-R	GTCACAACGCCTGTGCAAGTG
Adv-Hexon-F	ACCTTTGACTCTTCTGTCAGC
Adv-Hexon-R	GTAGCCAATGTTATAGTTAG

Safety, immunization and protection experiment in animals

To determine the immunizing dose of Adv-RABV, six groups of six-week-female ICR mice (n = 10) were immunized intramuscularly (i.m.) with 10⁸, 10⁷, 10⁶ TCID₅₀ (in a volume of 100 μl) SAD-L16 and Adv-RVG respectively. The body weight was recorded for 21 days, and the blood samples were harvested and subjected to FAVN assay.

To evaluate the safety of Adv-RABV, C57BL/6 (n = 3) were immunized intramuscularly with 10⁷ TCID₅₀ SAD-L16, Adv-RABV and DMEM, respectively. Parenchymal organs, injection site muscle and inguinal lymph nodes were collected at different time points and transcript levels of RVG and Hexon were detected by qRT-PCR.

To assess the immunogenicity and protective capacity of Adv-RABV, six-week-female ICR mice (n = 10) were immunized intramuscularly with 10⁷ TCID₅₀ SAD-L16, Adv-RABV, and DMEM, respectively and were challenged via the intracerebral (i.c.) route with 50× LD₅₀ of standard challenge virus 24 (CVS-24) at 24 weeks post-immunization (wpi). Mice that were moribund or had lost more than 25% of their initial body weight were humanely euthanized. Brain tissues of sacrificed mice were collected and rabies virus distribution was detected by immunohistochemical techniques. In addition, C57BL/6 mice (n = 5) were immunized with 10⁷ TCID₅₀ SAD-L16, Adv-RABV and DMEM. Subsequently, inguinal lymph nodes, spleens or bone marrow cells were harvested at different time points, subjected to flow cytometry assay, ELISpot or histology and immunofluorescence assays.

Four groups of three-month-old cats and dogs (n = 5) were immunized with inactivated rabies vaccine, Adv-RVG, Adv-RVDG (10⁸ TCID₅₀) or DMEM in a volume of 1000 μl subcutaneously in the neck. Anal temperature and weight changes were monitored continuously for 8 weeks. Serum samples were collected at different time points and subjected to FAVN or ELISA assay.

Flow cytometry assay (FACS) and enzyme-linked immunosorbent spot (ELISpot)

Flow cytometry assay was performed to quantify the number of immune cells. For this purpose, mice were anesthetized, and the inguinal lymph nodes, spleens or bone marrow cells were collected and prepared into single-cell suspension. Red blood cells were removed using lysis buffer (catalog no. BL503A; Biosharp). For blocking of Fc receptors in flow cytometric analysis, pre-incubate the cells with TruStain FcX™ (catalog no.101319; BioLegend) at 1.0 µg per 10⁶ cells in 100 µl volume for 5–10 min on ice prior to immunostaining. After washing twice with PBS, the cell suspensions (10⁶ cells/ sample) were stained by fluorescence-conjugated antibodies in the dark in 4°C. After washing twice, the samples were analysed using a BD FACSVerse flow cytometer (BD Biosciences, CA, USA) and FlowJo software (TreeStar, CA, USA).

Multiscreen HA ELISpot plates (Millipore, MA, USA) were coated with 300 ng/well purified RVG and incubated at 4°C for 12 h according to the instructions. After washing twice, the coated plates were blocked by RPMI 1640 medium containing 10% FBS for 2 h at 37°C. Single-cell suspensions (10⁶ cells/well) of inguinal lymph nodes and bone marrow cells collected at different time points were then transferred to the blocked ELISpot plates, and assays were conducted by using biotin-conjugated mouse IgG antibody (Bethyl Laboratories, TX, USA), streptavidin-alkaline phosphatase (Mabtech, Stockholm, Sweden), and BCIP/NBT-plus (Mabtech, Stockholm, Sweden). The plates were then scanned and analysed using Mabtech IRIS FluoroSpot/ELISpot reader, which uses RAWspot technology for multiplexing at the single-cell level.

Fluorescent-antibody virus neutralization test (FAVN) assay and enzyme-linked immunosorbent assay (ELISA)

Rabies virus neutralizing antibody (VNA) titre was measured by fluorescent-antibody virus neutralization (FAVN) assay, as previously described [32]. In brief, 100 μl of DMEM was added to a 96-well plate, and 50 μl of serum or standard serum was added to the first column in quadruplicate and serially diluted 3-fold. A 50 μl suspension containing 100 FFU of a rabies challenge virus, CVS-11, was added to each well. After incubation at 37°C for 1 h, 2 × 10⁴ BSR cells were added to each well and incubated at 37°C for 72 h. Samples were then fixed with 80% ice-cold acetone for 30 min and stained with FITC-conjugated antibody against RABV N protein. Fluorescence was observed under an Olympus IX51 fluorescence microscope (Olympus, Tokyo, Japan). Fluorescence values were compared with those of a reference serum (obtained from the National Institute for Biological Standards and Control, Hertfordshire, UK), and the results were normalized and quantified in international units per millilitre.

RVG-specific ELISA were preformed to determine antibody isotypes. In brief, ELISA plates were coated with 500 ng/well of RVG diluted in coating buffer overnight at 4°C. Plates were then washed three times in PBST and blocked for 2 h at 37°C in PBS containing 5% skim milk. Afterwards, 100 μl of the diluted serum (1:5000 for IgG, 1:500 for IgG1, 1:500 for IgG2a, 1:1000 for IgG2b, 1:500 for IgG2, 1:100 for IgM) was added to the plates and incubated for 1.5 h at 37°C. After washing twice, the HRP-conjugated goat anti-mouse IgG, IgG1, IgG2a and IgG2b and HRP-conjugated goat anti-dog IgG, IgG1and IgM and sheep anti-dog IgG2 were added to each well for 45 min at 37°C. After washing twice, 50 μl tetramethylbenzidine (TMB) substrate (Biotime Biotechnology) was added to each well for 20 min and reaction was stopped with 2 M sulphuric acid. Optical densities were recorded at 450 nm using a SpectraMax 190 spectrophotometer (Molecular Devices, CA, USA).

Histology and immunofluorescence assays

To detect the number of germinal centres in immunized lymph nodes. The inguinal lymph nodes were harvested and fixed with 4% paraformaldehyde for 72 h at 4°C. The tissues were dehydrated with 30% sucrose and cut into 25 μm-thick slices at maximum cross-section. Slices were blocked with 10% goat serum and fluorescently stained with Alexa Fluor 647-conjugated anti-mouse B220, Alexa Fluor 488-conjugated goat anti-mouse IgG, and biotin-labelled antihuman/mouse GL7 followed by staining with Alexa Fluor 594-conjugated streptavidin [33]. Images were captured under an Olympus IX51 fluorescence microscope.

Statistical analysis

All data were analysed using Prism 8.0 software (GraphPad, USA). For the percent survival, Kaplan-Meier survival curves were analysed using the log-rank test. Error bars represent the standard deviation (SD) or the standard error of the mean (SEM). Analysis of unpaired Mann–Whitney test (two-tailed) was used to determine the statistical significance of differences among different groups (ns, no significance; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).

Ethical statement

All animal experiments performed in this study strictly followed the recommendations of the Guide for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of the People’s Republic of China. All animal experiments were supervised and approved by the Scientific Ethics Committee of Huazhong Agricultural University (approval number: HZAUMO-2023-0201; HZAUCA-2023-0026; HZAUDO-2023-0013).

Results

Construction of recombinant adenoviruses and evaluation of biological characterization in vitro

Four recombinant adenoviruses were constructed, designated Adv-eGFP, Adv-RVG, Adv-E1-RVG, and Adv-RVDG and the RVG gene was derived from the RABV vaccine strain SAD-L16 (Figure 1(A)). To verify whether the expression of multiple antigens affected the growth characteristics of Adv-RABV, the growth curves of four recombinant adenoviruses were performed in HEK 293 cells. As shown in Figure 1(B), no significant difference was observed in growth of recombinant adenoviruses, indicating that the insertion of two RVG expression cassettes (Adv-RVDG) into the E1 and E3 regions, respectively, did not affect the replication of recombinant adenoviruses. Moreover, the expression of RVG was detected by indirect immunofluorescence (IFA) on HEK 293 at 48 hpi (MOI = 0.1). Our results showed that HEK 293 cells infected with Adv-RABV were able to successfully express RVG (Figure 1(C)). Moreover, RVG expression in HEK 293 infected with Adv-RABV (MOI = 0.1) were quantified at 48 hpi by western blotting. (Figure 1(D)). Notably, the expression of RVG is significantly higher in HEK 293 cells infected with Adv-RVDG than those infected with Adv-RVG and Adv-E1-RVG (Figure 1(E)).

Figure 1. Construction and characterization of the Adv-RABV. (A) Schematic diagram of the construction of recombinant adenoviruses. (B) The one step growth curves of four recombinant adenoviruses on HEK 293 cells (n = 3) at a multiplicity of infection (MOI) of 0.1 were shown. (C) The expression of RVG in Adv-RABV infected HEK 293 cells was determined by IFA. (D and E) The expression of RVG in HEK 293 cells (n = 3) infected with Adv-RABV was detected and quantified by western blotting. Error bars represent standard deviation (SD). The data are representative of results from two independent experiments (***, P < 0.001; ns, no significance).

Safety assessment and monitoring of the biodistribution pattern of Adv-RABV in vivo

Safety is of paramount importance in vaccine development. To eliminate the influence of other components of commercial inactivated rabies vaccines on the study, the RABV vaccine strain SAD-L16 was used as an immune control. In a preliminary experiment, mice were immunized intramuscularly with graded doses (from 10⁶ to 10⁸ TCID₅₀ in a 100 μl volume) of Adv-RVG and SAD-L16. By comparing the neutralizing antibodies and weight changes of the two groups, a moderate immunization dose (10⁷ TCID₅₀) was selected for follow-up experiments (Fig. S2). In addition, ICR mice (n = 10) were immunized intramuscularly with SAD-L16, Adv-RABV (at a dose of 10⁷ TCID₅₀) or DMEM in a volume of 100 μl. Then, the body weight of the mice was continuously recorded for 21 days to evaluate their safety. SAD-L16 immunized mice showed mild weight loss and they regained original body weight at 11 days post-immunization (dpi), whereas the Adv-RABV immunized group did not show a significant reduction in body weight (Figure 2(A)). This result suggests that Adv-RVG has better safety than SAD-L16.

Figure 2. Safety assessment and monitoring of the biodistribution pattern of Adv-RABV in vivo. Groups of six-week-female ICR mice (n = 10) were immunized intramuscularly with 10⁷ TCID₅₀ SAD-L16, Adv-RABV or DMEM respectively. The body weight change was recorded for 21 days (A). Groups of C57BL/6 mice (n = 3) were immunized intramuscularly with 10⁷ TCID₅₀ SAD-L16, Adv-RABV and DMEM as negative control respectively. The tissues were harvested at indicated time points and the total RNA were extracted Trizol reagent. RNA was converted to cDNA by reverse transcription. (B–I) The transcription level of Hexon were quantified in different tissues were shown by qRT-PCR. Error bars represent SD (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, no significant difference).

In order to investigate the biodistribution pattern of Adv-RABV in vivo after immunization via i.m route. C57BL/6 mice (n = 3) were immunized intramuscularly with Adv-RABV (at a dose of 10⁷ TCID₅₀) or DMEM in a volume of 100 μl, respectively. Subsequently, inguinal lymph nodes (LN), quadriceps (injection site) and parenchymal organs were harvested at the indicated time points. RNA in the above tissues was extracted with Trizol reagent. The quantitative real-time PCR (qRT-PCR) analysis was performed to determine the relative expression of Hexon mRNA. As expected, Hexon mRNA was not detected in parenchymal organs (Figure 2(B–H)), suggesting that the recombinant adenoviruses did not replicate efficiently in these tissues. Adv-E1-RVG is a replication-competent recombinant adenovirus and should replicate in vivo, little hexon mRNA was detected at the injection site at 1 and 3 dpi in the Adv-E1-RVG group, which may be attributed to the fact that mice are an adenovirus semi-permissive replicating species (Figure 2(I)).

Adv-RVDG elicits higher and more sustained VNA production and provides superior protection than other Adv-RABV and SAD-L16 in mice

To determine whether our construction strategy significantly improved the immunogenicity of Adv-RABV. ICR mice (n = 10) were immunized intramuscularly with SAD-L16 and Adv-RABV (a dose of 10⁷ TCID₅₀) or DMEM in a volume of 100 μl. Serums were collected and subjected to measure VNA titres at the indicated time points (Figure 3(A)). As shown in Figure 3(B), the mice immunized with Adv-RVDG induced significantly higher VNAs level than the groups vaccinated other Adv-RABV and SAD-L16 at each time point. The dynamics of geometric mean titres (GMTs) of VNA were also shown (Figure 3(C)). The highest antibody GMT in SAD-L16, Adv-RVG, Adv-E1-RVG, Adv-RVNG, and Adv-RVDG groups were 13.05, 16.71, 20.38, 19.86, and 35.09 IU/ml respectively. Consistent with the results shown in Fig. S2B, the level of VNA in SAD-L16 immunized mice decreased sharply from 4 to 14 wpi, while the level in Adv-RABV immunized mice attenuated more slowly. The GMT of VNA in Adv-RVDG immunized mice was 2.2-fold higher than those in Adv-RVG and Adv-E1-RVG and 33-fold higher than those in SAD-L16 immunized mice at 24 wpi.

Figure 3. Comparison of immunogenicity and protective capacity of Adv-RABV in mice. (A) Groups of 6-week-female ICR mice (n = 10) were immunized intramuscularly with 10⁷ TCID₅₀ SAD-L16, Adv-RABV and DMEM respectively and challenged via the i.c. route with 50× LD₅₀ of CVS-24 at 24 weeks post immunization. Mice that were seen to be moribund or that had lost more than 25% of their initial body weight were humanely euthanized. (B) Serum samples were harvested at the indicated time points and subjected to quantify VNA titres. (C) Geometric mean titres (GMT) of VNA were presented. (D and E) The transcription level of RVG were quantified by qRT-PCR at the injection sites and inguinal lymph nodes. (F and G) The body weight change was monitored and the percent survival was recorded for 21 days. (H) Detection of rabies virus in the brain after the immunized groups was challenged by immunohistochemistry techniques, the arrow indicates positive signal for RABV. Error bars represent SD (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, no significant difference).

To confirm whether VNA levels in each group correlate with the expression of RVG in vivo. C57BL/6 mice (n = 3) were immunized intramuscularly with Adv-RABV (at a dose of 10⁷ TCID₅₀) or DMEM in a volume of 100 μl. Inguinal lymph nodes (LNs) and quadriceps (injection site) were collected at the indicated time points. RNA in the above tissues was extracted by Trizol reagent. The quantitative real-time PCR (qRT-PCR) analysis was performed to detect the relative transcription level of RVG mRNA. RVG in Adv-RVDG was rapidly expressed in the quadriceps at 2 days post-infection (dpi), reaching a maximum at 5 dpi and persisting for 15 days (Figure 3(D)). Notably, the transcription level of RVG in Adv-RVDG was 4-fold higher than in other Adv-RABV and 10-fold higher than SAD-L16 at 5 dpi. A similar expression tendency of RVG was detected in the LNs (Figure 3(E)), but no RVG was detected in the SAD-L16 group, indicating that the Adv-RABV can infect lymphoid tissues and deliver the exogenous gene to immune cells. The present results confirmed that Adv-RVDG expresses more RVG over a longer period than other Adv-RABV.

To compare the long-term protective effect of Adv-RABV against pathogenic RABV challenge, the immunized mice were infected intracerebrally (i.c.) with 50× 50% lethal doses (LD₅₀) of challenge virus standard 24 (CVS-24) at 24 wpi, and then the survival rate was calculated. As shown in Figure 3(F), all DMEM-immunized mice succumbed to rabies within 13 days, whereas over 90% of mice immunized with Adv-RABV survived the lethal challenge, and only 20% of mice survived in the SAD-L16 immunized group. Large numbers of rabies-infected neurons were detected in the brain tissue of the sacrificed mice (Figure 3(H)). Meanwhile, the changes in body weight were recorded, and significant weight loss was observed in the DMEM and SAD-L16 immunized groups. Although all mice in the Adv-RVDG and Adv-E1-RVG-immunized groups survived, the weight loss in the ADV-E1-RVG mice was more pronounced than that in the Adv-RVDG-immunized group within 1 week after challenge (Figure 3(G)). Overall, this result indicated that Adv-RABV, particularly Adv-RVDG, stimulated higher and more durable VNA production and provided the long-term protective efficacy of Adv-RABV against pathogenic RABV challenge.

Adv-RVDG effectively promotes maturation of cDCs in vivo

Previous researches have verified that activation and maturation of resident dendritic cells (DCs) enhances the humoral immune response elicited by vaccination [33,34]. To verify whether Adv-RVDG activates more DCs in vivo, C57BL/6 mice (n = 5) were immunized intramuscularly with DMEM, SAD-L16, Adv-RVG, and Adv-RVDG (at a dose of 10⁷ TCID₅₀ in 100 μl). Inguinal lymph nodes were collected at 6 dpi and prepared into single-cell suspensions that were subjected to FACS analysis. Mature conventional dendritic cells (cDCs) upregulate CD80 and CD86 or major histocompatibility complex class II (MHC-II) [35,36]. Up-regulation of these surface markers is essential to displaying portions of the antigens for recognition by T helper cells. The gating strategy and representative flow cytometric plots for mature cDCs (CD11c⁺ CD80⁺, CD11c⁺ CD86⁺, or CD11c⁺ MHCII⁺ cDCs) were shown in Figure 4(A,B). Adv-RVG facilitated the mice to produce more mature cDCs than SAD-L16. In addition, significantly more mature cDCs were detected in the mice immunized with Adv-RVDG than in the group immunized with Adv-RVG. Overall, these results indicated that Adv-RVDG can promote the maturation of cDCs in vivo more effectively than Adv-RVG and SAD-L16.

Figure 4. Adv-RVDG more effectively promotes the activation and maturation of cDCs in vivo. Groups of 6-week-female C57BL/6 mice (n = 5) were immunized with 10⁷ TCID₅₀ SAD-L16, Adv-RVG, Adv-RVDG and DMEM, respectively. The inguinal lymph nodes were collected at 6 dpi and prepared into single-cell suspensions, subjected to FACS analysis. (A) Representative gating strategy for analysis of CD11c⁺ CD80⁺, CD11c⁺ CD86⁺ and CD11c⁺ MHC II⁺ cells. (B) Representative flow cytometric plots for CD11c⁺ CD80⁺, CD11c⁺ CD86⁺ and CD11c⁺ MHC II⁺ cDCs in LNs. (C–E) Statistical results of mature cDCs in LNs. Error bars represent SD (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, no significant difference).

Adv-RVDG elicits an intense germinal centre response

The germinal centre reaction is considered an important part of the humoral immune response, mainly including the activated T follicular (Tfh) and germinal centre B cells (GC B) that migrate back into follicles to form germinal centres, where specialized antibody responses are induced [37]. To verify whether Adv-RVDG facilitates the Tfh cells differentiation and GC B cells generation, C57BL/6 mice (n = 5) were immunized intramuscularly with SAD-L16, Adv-RVG, and Adv-RVDG (at a dose of 10⁷ TCID₅₀) or DMEM in a volume of 100 μl. Inguinal lymph nodes were collected at 7 and 14 dpi. Then, the number of Tfh cells (CD4⁺ PD1⁺ CXCR5⁺) and GC B cells (B220⁺ GL7⁺ CD95⁺) were quantified by flow cytometry. The gating strategy for measuring Tfh and GC B cells is shown in Figure 5(A), and representative flow cytometric plots were shown in Figure 5(B). Significantly more Tfh and GC B cells were measured in the inguinal lymph nodes of mice immunized with Adv-RVDG than in mice immunized with SAD-L16 or Adv-RVG at 7 and 14 dpi (Figure 5(C,D)). For a more visual representation of the post-immunization lymph node germinal centre (GCs), the inguinal lymph nodes were harvested at 14 dpi, and the number of GCs was measured by histology and immunofluorescence assays. Representative results for GC formation were shown in Figure 5(E), more GCs were counted at LNs in Adv-RVDG immunized mice than those in Adv-RVG or SAD-L16 immunized mice (Figure 5(F)). Moreover, RVG-specific antibody-secreting cells (ASCs) in the inguinal lymph nodes were quantified by enzyme-linked immunosorbent spot (ELISpot) assay at 14 dpi. Representative sections and the number of RVG-specific ASCs were shown in Figure 5(G,H). As expected, the number of RVG-specific ASCs was significantly higher in Adv-RVDG vaccinated mice than in Adv-RVG and SAD-L16 immunized mice. In summary, Adv-RVDG elicits more effectively differentiation of Tfh cells and proliferation of GC B cells and facilitate the formation of GCs better than SAD-L16 and Adv-RVG.

Figure 5. Adv-RVDG elicits an intense germinal centre response. C57BL/6 mice (n = 5) were immunized with 10⁷ TCID₅₀ SAD-L16, Adv-RVG, Adv-RVDG and DMEM, respectively. The inguinal lymph nodes were collected at 7 and 14 dpi and prepared into single-cell suspensions, subjected to FACS analysis. (A) Representative gating strategy for analysis of Tfh cells (CD4⁺ PD1⁺ CXCR5⁺) and GC B cells (B220⁺ GL7⁺ CD95⁺). (B) Representative flow cytometric plots for Tfh cells (CD4⁺ PD1⁺ CXCR5⁺) and GC B cells (B220⁺ GL7⁺ CD95⁺) in LNs. (C and D) Statistical results of Tfh and GC B cells. Besides, C57BL/6 mice (n = 3) were immunized with 10⁷ TCID₅₀ SAD-L16, Adv-RVG, Adv-RVDG and DMEM respectively. The inguinal lymph nodes were harvested at 14 dpi subjected to quantify ASCs and GCs. Tissue sections were prepared to detect GCs (B220⁺, blue; IgG⁺, green; GL7⁺, red). Scale bars, 1000 or 500 μm (rightmost column only). The representative results of GC formation in inguinal lymph nodes are shown (E). The numbers of GCs formed at 14 dpi were calculated (F). Representative sections and the number of RVG-specific ASCs were shown (G and H). Error bars represent SD (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, no significant difference).

Adv-RVDG induces significantly more MBCs and LLPCs production and elicits a better secondary immune response

In T_fh cell dependent responses, some GC B cells experience heavy chain isotype (class) switching and affinity maturation. Meanwhile, helper T cells stimulate the production of long-lived plasma cells (LLPCs) and the generation of memory B cells (MBCs). To determine whether Adv-RVDG can increase the number of MBCs (B220⁺ CD38⁺ CD138^- or B220⁺ IgD^- CD38⁺ CD273⁺) in spleen and LLPCs (B220^low CD138⁺) in bone marrow and induce a good secondary immune response. In brief, four groups of C57BL/6 mice (n = 5) were intramuscularly boosted with the same vaccines at 24 wpi (Figure 6(A)). Spleen and bone marrow samples (n = 5) were harvested one week after boosting and subjected to FACS analysis. The gating strategy for quantification of MBCs cells was shown in Figure 6(B), and representative flow cytometric plots were shown in Figure 6(D). Interestingly, Adv-RVDG induced significantly more MBCs in the spleen than did SAD -L16 and Adv-RVG (Figure 6(G,H)). Moreover, the number of MBCs increased significantly more in mice immunized with Adv-RVDG than in mice immunized with SAD-L16 and Adv-RVG. The long-lasting antibody response is mainly attribute to the quantify of LLPCs, the gating strategy for quantify LLPCs is shown in Figure 6(C). As expected, Adv-RVDG induced significantly more LLPCs production in bone marrow at 24 wpi (Figure 6(E,I)). The greatest increase in the number of LLPCs in Adv-RVDG immunized mice was observed among the three immunization groups. To count the number of RVG-specific long-lived ACSs induced by SAD-L16, Adv-RVG and Adv-RVDG immunization, bone marrow samples (n = 3) were harvested at 24 wpi, subjected to ELISpot assay. Representative sections and the number of RVG-specific ASCs are shown in Figure 6(F,K). Consistent with the above experimental results, significantly more long-lived ACSs were induced in Adv-RVDG immunized mice than in SAD-L16 and Adv-RVG immunized groups. In addition, the number of long-lived ASCs was 4, 213, 361 and 630 per 10⁶ bone marrow cells in DMEM, SAD-L16, Adv-RVG and Adv-RVDG after boosting, respectively. To determine whether the augmented LLPCs and ASCs affect the production of VNAs. Serum samples were harvested one week after boosting, the RABV-specific VNAs were measured by FAVN. As shown in Figure 6(L), the level of VNA significantly increased in all three immunized groups and the highest value was 44.62 IU/ml in Adv-RVDG groups one week after boosting, which was 2.47- and 7.89-fold higher than in the Adv-RVG and SAD-L16 immunized groups, respectively. In conclusion, Adv-RVDG induced significantly more MBCs and LLPCs production and elicited a better secondary immune response.

Figure 6. Adv-RVDG induces significantly more MBCs and LLPCs production and elicits better secondary immune response. C57BL/6 mice (n = 10) were immunized with 10⁷ TCID₅₀ SAD-L16, Adv-RVG, Adv-RVDG and DMEM, respectively. Each of the groups was then further subdivided into two subgroups. One subgroup was boosted at 24 wpi. The spleens and bone marrow cells were harvested at a week post-boost and subjected to FACS and ELISpot to quantify MBCs (B220⁺ CD38⁺ CD138^– or B220⁺ IgD^– CD38⁺ CD273⁺) in spleen, LLPCs (B220^low CD138⁺) and long-lived RVG-specific ASCs in bone marrow. Meanwhile, the serum samples were collected at a week post-boost and subjected to measure VNA titres. (A) The timeline shows the key immunization and immune analysis steps. (B and C) The gating strategy for quantify MBCs and LLPCs cells is shown. (D and E) Representative flow cytometric plots for MBCs cells and LLPCs in spleens and bone marrow. (F) Representative sections for long-lived RVG-specific ASCs in bone marrow. (G–K) Statistical results of MBCs, LLPCs and long-lived RVG-specific ASCs. (L) The VNA titres were quantify by FAVN. Error bars represent SD (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, no significant difference).

Adv-RVDG elicits a robust and sustained humoral immune response in cats and dogs

Although a variety of animal species can be infected with RABV, dogs are responsible for 99% of human cases. Cats, the largest group of pets in China, are also important hosts for rabies. Commercially available rabies vaccines are inactivated. To test the commercial value of Adv-RVDG, four groups of cats and dogs (n = 5) were immunized subcutaneously in the neck with one dose of inactivated rabies vaccine (IRV), Adv-RVG, Adv-RVDG (a dose of 10⁸ TCID₅₀) or DMEM (in a volume of 1000 μl), respectively (Figure 7(A)). Adv-RVDG was well tolerated by cats and dogs, with neither clinical signs nor local reactions at the injection site recorded during the 24 weeks of follow up. Consistent with mouse immunization results, significantly more VNAs were observed in dogs immunized with Adv-RVDG (Figure 7(B,C)). The GMT of VNA titre induced by Adv-RVDG reached a maximum of 40.5 IU/ml in canine, which was 2.38- and 7.5-fold higher than that of Adv-RVG and IRV, respectively, and remained at a high level for half a year. The VNA titre in the group immunized with Adv-RVDG was 29.7 IU/ml, which was significantly higher than the protective value of 0.5 IU/mL at 24 wpi. Meanwhile, the GMT of VNA titre in the cats immunized with Adv-RVDG reached a maximum of 12.53 IU/mL, which was 1.68- and 2-fold higher than that of Adv-RVG and IRV, respectively. Consistently, the GMT of VNA titre and RVG-specific total IgG in dogs (Figure 7(C,F)) elicited by Adv-RVDG were higher than those elicited by Adv-RVG and IRV and remained at a high level for 24 weeks post-immunization. In addition, VNA titres declined significantly faster in animals receiving a single dose of IRV respect to those immunized with Adv-RVDG. Furthermore, Adv-RVDG induced significantly more IgM than Adv-RVG at 2 weeks post-immunization (Figure 7(G)). Meanwhile, antibody subtypes against RVG were measured in canines by ELISA. Similarly, dogs immunized with Adv-RVDG induced more IgG1 and IgG2 than those immunized with IRV and Adv-RVG (Fig. S3A and B). As shown in Fig. S3C, the IgG2/IgG1 ratio result demonstrated that Adv-RVDG induced a Th1-biased cellular immune response which is critical in the clearance of RABV from the CNS.

Figure 7. Adv-RVDG elicits robust and sustained humoral immune response in cats and dogs. Four groups of RABV antibody-negative cats and dogs (n = 5) were immunized with 10⁸ TCID₅₀ Adv-RVG, Adv-RVDG, a dose of commercial inactivated rabies vaccine (IRV) and DMEM subcutaneously in the neck, respectively. The serums were harvested at indicated time points and subjected to quantify VNA titres by FAVN and measure RVG-specific total IgG and IgM by ELISA. (A) The timeline shows the key immunization and immune analysis steps. (B and C) The VNA titres and its Geometric mean titres (GMT) were quantified in immunized dog. (D and E) The VNA titres and its Geometric mean titres (GMT) were quantified in immunized felines. (F and G) Optical density (OD) values of total IgG and IgM against RABV G were determined by ELISA in immunized canines. Error bars represent SEM (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, no significant difference).

In summary, these results demonstrated that Adv-RVDG, a modified adenoviral vector carrying two RVG expression cassettes in the E1 and E3 regions, significantly elicits VNA production and provides long-term protection against virulent RABV challenge in mice by facilitating the maturation of DCs and further effectively enhances Tfh cell, GC B cell, LLPCs, and ASCs proliferation. More importantly, Adv-RVDG can also induce more robust and durable antibody response in cats and dogs.

Discussion

Rabies is an ancient zoonotic disease with a 100% mortality rate once it occurs [38]. Given on the current serious epidemic situation of rabies and the fact that commercially available vaccines are expensive, ineffective, and require multiple vaccinations [39,40], the development of a new rabies vaccine with a single dose, low price, high efficacy, and long-lasting immunity is urgently needed to achieve large-scale immunization. Recently, adenovirus vectors have been successfully used for vaccine platforms due to their excellent characteristics: i. Wide spectrum of infection, infecting a variety of dividing and non-dividing cells; ii. High yield and low production cost; iii. Stable physicochemical properties, which can be stored at low temperatures without loss of activity; iv. Low biocontainment requirements for production; v. Ability to promote durable and robust humoral and cellular immune response [41]. Researchers have developed several novel adenoviral vectors for gene delivery, such as simian adenoviruses [26], human adenovirus type 4 (Ad4) [42], human adenovirus type 35 [43] (Ad35) vectors, and chimpanzee adenovirus type 68 (cAd68) [44]. Previous studies have demonstrated that these novel adenovirus vectors mediate the diverse immune responses in hosts [45]. The Ad49L vectored COVID-19 vaccine induces a strong cellular immune response but low antibody response, while the Sad23L-nCoV-S and Ad5-nCoV-S vaccines induced a low level of interferon-α (IFN-α) and a high level of antigen-specific antibody responses. Rabies prophylaxis relies heavily on the production of virus-specific neutralizing antibodies following immunization. In our study, human adenovirus type 5 was selected for the development of a new rabies vaccine.

In the first generation of the adenovirus 5 vector, the E1 and E3 regions were removed to improve safety and the ability to incorporate exogenous genes [46]. Most of the current studies introduced exogenous genes into the E1 region and the E3 region could not be used efficiently [27,29,47]. To increase the expression of protective antigens in adenovirus vectors, researchers introduced RVG and RVN genes into the adenovirus E1 region by linking them through the P2A sequence. However, when RVG was fused to RVN, the expression of RVG was lower and a similar uncut band was detected by western blotting in vitro, suggesting that the strategy of introducing multiple antigens into the adenovirus E1 region by linking them through 2A sequences seems to be infeasible [34]. Besides, a study has shown that the expression of RVG was improved using AdHu5-tRVG-hrGFP, carrying three copies of RVG gene in the E1 region by 2A sequence. However, immunogenicity of AdHu5-tRVG-hrGFP has not been further investigated in vivo [48]. In our study, we creatively inserted RVG genes into both E1 and E3 genomic regions and we confirmed that Adv-RVDG not only improved the expression of RVG in vitro but also in vivo. Moreover, our study demonstrated that the simultaneous insertion of exogenous genes in the E1 and E3 genomic regions did not affect the replication efficiency of recombinant adenoviruses.

Safety is the most important evaluation metric for vaccine development. The E1 gene is an early first-initiated gene required for transcription of other early genes of adenoviruses. Therefore, adenoviral vectors lacking E1 cannot replicate efficiently and produce various viral proteins, thus failing to complete the viral life cycle [49]. We confirmed that Adv-RVDG did not replicate efficiently in vivo by examining the transcription levels of adenoviral structural protein Hexon in parenchymal organs of Adv-RVDG-immunized mice. Surprisingly, little expression level of Hexon gene were detected at the injection site of Adv-E1-RVG immunized mice for 3 days after immunization but rapidly disappeared after 5 days. This phenomenon is similar to previous findings suggesting that replication-competent recombinant adenoviruses (RCA) do not replicate efficiently in ICR mice [50]. However, it has been reported that wild-type human adenoviruses can replicate efficiently in Syrian hamsters and non-human primates [51], suggesting that RCA may pose potential safety risk. Meanwhile, no body weight changes and clinical symptom were observed in mice immunized with Adv-RVDG.

Compared with previous studies, we more comprehensively evaluated the immunological efficacy and protective effect of Ad5 vectors used as rabies vaccines. In agreement with previous findings, replication-competent adenoviruses (Adv-E1-RVG) are comparable to replication-deficient adenoviruses (Adv-RVG) in their ability to induce immunized mice to produce neutralizing antibodies to exogenous proteins, probably because mice are a semi-permissive replicating species of human adenoviruses [52]. Our results showed that Adv-RVDG immunized mice had the highest VNA titres and were 1.5–2 times higher than other Adv-RABV immunized groups, which is consistent with the RVG mRNA detection results. In another way, five weeks after inoculation with recombinant adenovirus expressing green fluorescent protein, green fluorescence still was clearly visible in the leg muscles of mice [53]. Surprisingly, the RVG mRNA also was instead consistently detected in inguinal lymph nodes, which decreased over time but was still detectable at 15 dpi in Adv-RVDG immunized group. Likewise, a previous study has shown that the DNA of a replication-defective chimpanzee adenovirus vector still can be detectable in injection sites and in draining lymph nodes at 49 dpi [54]. These clues revealed two phenomena: i. Adenoviruses at the injection site may enter nearby lymphoid tissues or organs through blood vessels or lymphatic vessels; ii. Replication-deficient adenoviral genomes continue to maintain transcriptional activity for an extended period of time. In addition, compared to SAD-L16, Adv-RVDG produced a long-lasting antibody response and provided better protection against virulent rabies challenge.

Vaccination creates robust, persistent humoral immunity by inducing the germinal centre (GC) reaction, an intricate immune response that produces memory B cells (MBC), and long-lived plasma cells (LLPC) that provide protection against (re)infection. Dendritic cells are responsible for recognizing and processing antigens, and the maturation of dendritic cells is an important indicator for evaluating the immunogenicity of vaccines. A previous study has shown that the ecto-domains of RVG can bind and activate DCs [55]. In our study, significantly more mature cDCs were detected in inguinal lymph nodes in Adv-RVDG than Adv-RVG vaccinated mice, which may be attributed to the increased level of RVG expression. Similar results were observed in AAV9-RABVG vaccinated mice [56]. Logically, more robust germinal centre responses were produced the inguinal lymph nodes of Adv-RVDG immunized mice including Tfh cells differentiation, GC B cells proliferation, and more antibody-secreting cells production. Consistent with the antibody results, mice in the Adv-RVDG immunized group produced more MBCs and LLPCs. And the number of MBCs and LLPCs increased rapidly after a boost immunization, which is responsible for the significant increase of VNA titre. These results indicated that Adv-RVDG elicits robust and durable humoral immune response through activating the DCs-Tfh cell-GC B cell/ MBCs-LLPCs axis.

The currently licensed rabies vaccines based on the whole inactivated virus are amongst the most expensive vaccines on the market and require higher number of doses. Besides, commercial inactivated rabies vaccines elicited a relatively weak immune response and lower VNA titres in cats and dogs, while Adv-RVDG induced much higher VNA titres. Consistent with our previous studies, antibody levels peaked 3–4 weeks after IRV immunization and then declined sharply [34]. In our study, antibody levels decreased more slowly after Adv-RVDG immunization during the 24-week study period, demonstrating that it can also provide long-lasting protection for dogs and cats. Hooper et al. found that Th1 cells are critical in the clearance of RABV from the CNS [57]. Besides, McGettigan’s group reported that IgG2-biased antibody response was more effective against RABV infection [58]. In our study, significantly more IgG2 were produced in Adv-RVDG immunized canines, suggesting Adv-RVDG better promotes Th1-type polarization of helper T cells. In addition, the pentameric structure of IgM allows it to have the highest potency. Previous studies have shown that IgM can cover more glycoprotein spikes on the RABV surface for neutralization compared to IgG, and can limit the spread of pathogenic RABV to the central nervous system early after post-exposure prophylaxis (PEP) and mediate protection against pathogenic RABV challenges [59,60]. In our study, Adv-RVDG immunized dogs induced more IgM than IRV and Adv-RVG at 14 dpi.

In summary, our study demonstrated that the simultaneous insertion of protective antigens into the E1 and E3 genomic regions of adenoviral vectors can significantly enhance the immunogenicity of adenoviral-vectored vaccines, which provides a theoretical and practical basis for the subsequent development of multivalent and multi-conjugate vaccines using adenoviral vectors. Adv-RVDG induces a more robust and durable humoral immune response in mice, cats, and dogs, providing greater protection against RABV as a single-dose immunization, low price, safety, high-efficiency, and long-lasting rabies vaccine candidate. Meanwhile, mucosal immunization by oral and inhalation is an effective way to achieve mass immunization of dogs, especially stray dogs [33]. In future studies, we will further evaluate whether Adv-RVDG equipped with dual RVG expression cassettes can improve its immunological success and immunological efficacy as an oral vaccine.

Acknowledgements

The authors thank all the members of the Rabies group at HZAU. We would like to thank the National Key Laboratory of Agricultural Microbiology Core Facility for assistance in Flow cytometry, and we would be grateful to Fangkui Wang for his support of sample preparation, data acquisition and analysis. The authors acknowledge the use of BioRender.com that is used to create the panel A in Figures 1, 3, 6, 7 and S1.

Disclosure statement

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

Additional information

Funding

This study was partially supported by supported by the National Key Research and Development Program of China [No. 2022YFD1800100] and the Fundamental Research Funds for the Central Universities [2662023PY005].