Colloidal Manganese Salt Improves the Efficacy of Rabies Vaccines in Mice, Cats, and Dogs

ABSTRACT

Rabies, caused by rabies virus (RABV), remains a serious threat to public health in most countries worldwide. At present, the administration of rabies vaccines has been the most effective strategy to control rabies. Herein, we evaluate the effect of colloidal manganese salt (Mn jelly [MnJ]) as an adjuvant of rabies vaccine in mice, cats, and dogs. The results showed that MnJ promoted type I interferon (IFN-I) and cytokine production in vitro and the maturation of dendritic cells (DCs) in vitro and in vivo. Besides, MnJ serving as an adjuvant for rabies vaccines could significantly facilitate the generation of T follicular helper (Tfh) cells, germinal center (GC) B cells, plasma cells (PCs), and RABV-specific antibody-secreting cells (ASCs), consequently improve the immunogenicity of rabies vaccines, and provide better protection against virulent RABV challenge. Similarly, MnJ enhanced the humoral immune response in cats and dogs as well. Collectively, our results suggest that MnJ can facilitate the maturation of DCs during rabies vaccination, which can be a promising adjuvant candidate for rabies vaccines.

IMPORTANCE Extending the humoral immune response by using adjuvants is an important strategy for vaccine development. In this study, a novel adjuvant, MnJ, supplemented in rabies vaccines was evaluated in mice, cats, and dogs. Our results in the mouse model revealed that MnJ increased the numbers of mature DCs, Tfh cells, GC B cells, PCs, and RABV-specific ASCs, resulting in enhanced immunogenicity and protection rate of rabies vaccines. We further found that MnJ had the same stimulative effect in cats and dogs. Our study provides the first evidence that MnJ serving as a novel adjuvant of rabies vaccines can boost the immune response in both a mouse and pet model.

INTRODUCTION

Rabies is an ancient zoonosis caused by rabies virus (RABV), a nonsegmented, negative-sense RNA virus that belongs to the Lyssavirus genus within the Rhabdoviridae family. Its genome encodes five structural proteins in the following order: nucleoprotein (N), phosphoprotein (P), matrix protein (M), glycoprotein (G), and RNA-dependent RNA polymerase (L) (1, 2). The G protein is the only protein on the virion surface and induces virus-neutralizing antibodies (VNAs) (3, 4). Globally, it causes an estimated 59,000 human deaths annually, and most deaths occur in less-developed countries in Asia and Africa (5, 6). In most cases, rabies is transmitted through bites by rabid dogs (7). From the entry site, RABV from the saliva moves along the peripheral nervous system and eventually reaches the central nervous system (CNS). The mortality rate is almost 100% once clinical signs are apparent, and patients that survive RABV infection show extensive neuron damage (8).

Although deadly, rabies can be prevented effectively by appropriate vaccination in humans and animals (9, 10). It is estimated that the rabies vaccine saves more than 250,000 people from dying of rabies every year (11). The World Health Organization recommends using inactivated rabies vaccines based on their safety. However, inactivated vaccines have exhibited relatively poor immunogenicity. Generally, they require repeated administration: three times for preexposure vaccination and four or five times for postexposure prophylaxis (PEP), which significantly increases the cost of vaccination. Thus, adjuvants are generally introduced to boost the immunogenicity of inactivated rabies vaccines. Aluminum-containing adjuvants have been used in animals and humans for over 90 years. To date, aluminum hydroxide has been incorporated widely as an adjuvant in billions of doses of commercial vaccines and administered annually by millions of people (12, 13). Aluminum hydroxide adjuvant could promote an immune response to DNA vaccine (14) and inactivated rabies vaccines (15, 16). However, some drawbacks related to aluminum hydroxide-based adjuvants have been found. It may cause some adverse effects, such as increased IgE production, devastating effect on local tissues, prolonged inflammation, and neurotoxicity (17). Therefore, the discovery of a highly effective adjuvant is still urgent to protect animals from rabies.

Manganese (Mn) is an essential micronutrient for diverse physiological processes, including development, reproduction, neuronal function, and antioxidant defenses (18, 19). But its functions in innate and adaptive immunity remain elusive. Recently, it was reported that Mn²⁺ was required for the host defense against DNA viruses by increasing the sensitivity of the DNA sensor cyclic GMP-AMP synthase (cGAS) and its downstream adaptor protein stimulator of interferon genes protein (STING) (20, 21). Besides, manganese is critical for the antitumor immune response via cGAS-STING and improves clinical immunotherapy efficacy (22, 23). In addition to its antiviral and antitumor effects, studies demonstrated that MnJ displayed adjuvant effects on inactivated viruses, including vesicular stomatitis virus (VSV), herpes simplex virus 1 (HSV-1), and vaccinia virus (VACV) (24). However, the immune effects of MnJ as an adjuvant of rabies vaccines have not been reported yet.

In this study, we evaluated the effect of MnJ as an adjuvant of rabies vaccines in mice, cats, and dogs. Our results demonstrated that MnJ enhanced VNA production and protected vaccinated mice from lethal RABV challenge by facilitating the maturation of dendritic cells (DCs). It enhanced the efficacy of rabies vaccines in cats and dogs as well. Therefore, MnJ has great potential to be an adjuvant candidate for rabies vaccines.

RESULTS

MnJ induces the production of IFN-I and cytokines in BMDCs.

A recent study showed that Mn²⁺ could facilitate the production of type I interferon (IFN-I) in bone marrow-derived DCs (BMDCs) (20, 24). To evaluate if MnJ has a similar effect in BMDCs, we treated BMDCs with MnJ and performed transcriptome sequencing (RNA-seq) analysis to analyze genes upregulated in MnJ-stimulated BMDCs. RNA-seq analysis of MnJ- or DMEM-treated BMDCs revealed that MnJ induced robust production of both beta interferon (IFN-β) and various IFN-α’s, IFN-stimulated genes (ISGs), and some chemokines and proinflammatory cytokines (Fig. 1A). Next, quantitative real-time PCR (qRT-PCR) analysis was performed to verify the results of RNA-seq in BMDCs treated with MnJ (200, 400, and 800 μM in a 100-μl volume) for 12 and 24 h, respectively (n = 3). As shown in Fig. 1B, MnJ enhanced the expression of IFN-I and ISGs, including Ifna1, Ifna2, Ifna4, Ifnb1, Irf7, Rsad2, Ifit1, Ifit2, and Ifit3, as well as some chemokine (Cxcl10) and proinflammatory cytokine (Il-6 and Tnf) genes. Moreover, the expression of these genes was stimulated in a dose-dependent manner when treated with MnJ. Together, these results demonstrate that MnJ can facilitate the production of type I IFN and cytokines in BMDCs in a dose-dependent manner.

FIG 1.

MnJ enhances the production of IFN-I and cytokines in BMDCs. BM cells were collected from C57BL/6 mice, and DC precursors were cultured with GM-CSF and IL-4. (A) BMDCs were harvested on day 6 and treated with MnJ (800 μM) or DMEM for 24 h. Total RNA was isolated for RNA-seq analysis, and representative upregulated genes are listed. (B) BMDCs were treated with MnJ (200, 400, and 800 μM) or DMEM for 12 and 24 h (n = 3). Total RNA was isolated, and quantitative RT-PCR (qRT-PCR) analysis was performed. Error bars represent standard deviations (SD).

MnJ facilitates the maturation of DCs in vitro and in vivo.

Previous studies have established that mature DCs could enhance the humoral immune response induced by rabies vaccines (25–27). To evaluate the ability of MnJ to activate DCs in vitro, BMDCs were prepared and incubated with various concentrations of MnJ or aluminum hydroxide gel (Alum) (200, 400, and 800 μM) (n = 3). At 12 and 24 h posttreatment, BMDCs were collected, washed, stained, and analyzed by flow cytometry. A representative gating strategy of mature BMDCs (CD11c⁺ and CD86⁺) and representative BMDCs from different groups are presented in Fig. 2A and B, respectively. As expected, a significantly higher percentage of CD86⁺ cells in CD11c⁺ cells was observed in BMDCs incubated with MnJ than in those treated with Alum and Dulbecco’s modified Eagle’s medium (DMEM) (mock) (Fig. 2C).

FIG 2.

MnJ promotes DC maturation both in vitro and vivo. BMDCs were treated with MnJ or Alum (200, 400, and 800 μM) or DMEM (n = 3). At 12 and 24 h, cells were harvested and subjected to flow cytometric analysis. (A) Representative gating strategy of CD11c⁺ CD86⁺ cells in BMDCs. (B) Representative flow cytometric plots of mature BMDCs from different groups. (C) Statistical results of CD11c⁺ CD86⁺ BMDCs. Error bars represent SD. C57BL/6 mice were immunized with RABV, MnJ-adjuvanted RABV, Alum-adjuvanted RABV, or DMEM via the i.m. route (n = 5). At 3 dpi, the inguinal LNs were collected and single cells were stained with antibodies representing the markers of CD11c⁺ CD80⁺, CD11c⁺ CD86⁺, and CD11c⁺ MHC-II⁺ cells. (D) Representative gating strategy for analysis of CD11c⁺ CD80⁺ and CD11c⁺ CD86⁺ cells. (E) Representative flow cytometric plots of CD11c⁺ CD80⁺ cDCs and CD11c⁺ CD86⁺ cDCs in LNs from four groups. (F) Statistical results of mature cDCs in LNs. Error bars represent standard errors of the mean (SEM). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, no significant difference.

Mature conventional dendritic cells (cDCs) upregulate CD80 and/or CD86 (costimulatory molecules) or major histocompatibility complex class II (MHC-II). This process is essential to facilitate the interaction between cDCs and T cells (28–30). Next, the effect of MnJ on cDC maturation in mice was examined. Four groups of C57BL/6 mice (n = 5) were immunized with RABV or RABV supplemented with adjuvants (MnJ-adjuvanted RABV and Alum-adjuvanted RABV) via the intramuscular (i.m) route, respectively. Inguinal lymph nodes (LNs) were collected at 3 days postimmunization (dpi). The single-cell suspension was prepared to measure CD11c, CD80, CD86, and MHC-II via flow cytometry. The representative gating strategy for mature DCs (CD11c⁺ CD80⁺ or CD11c⁺ CD86⁺ cDCs) is shown in Fig. 2D. The representative plots of mature cDCs in LNs from 4 groups are presented and analyzed in Fig. 2E and F. As shown in Fig. 2E and F, significantly more mature cDCs were observed in LNs of mice immunized with MnJ-adjuvanted RABV than in mice immunized with RABV and Alum-adjuvanted RABV. Together, these results indicate that MnJ can facilitate the maturation of BMDCs in vitro and cDCs in vivo more effectively than Alum.

MnJ promotes the proliferation of Tfh and GC B cells in LNs post-RABV immunization.

The costimulatory molecules CD80/CD86 expressed on cDCs could regulate the activation and recruitment of CD4⁺ T cells (31). Afterwards, T follicular (Tfh) cells facilitate the generation of germinal center (GC) B cells with high affinity for the antigen (32, 33). Thus, the role of MnJ in Tfh and GC B cell induction was further investigated. Four groups of C57BL/6 mice (n = 5) were immunized with RABV, MnJ-adjuvanted RABV, or Alum-adjuvanted RABV. Then, flow cytometry was performed to quantify the Tfh cells (CD4⁺ PD1⁺ CXCR5⁺) and GC B cells (B220⁺ GL7⁺ CD95⁺) in draining LNs at 7 and 14 dpi. The gating strategy for measuring Tfh and GC B cells is shown in Fig. 3A, and representative flow cytometric plots from the four groups are shown in Fig. 3B and D. Significantly more Tfh (Fig. 3C) and GC B cells (Fig. 3E) among the draining LNs were observed in mice immunized with MnJ-adjuvanted RABV than in mice immunized with RABV alone or Alum-adjuvanted RABV at 7 and 14 dpi (Fig. 4C and E). Together, our results indicate that MnJ can promote the generation of Tfh and GC B cells more than Alum post-RABV immunization in mice.

FIG 3.

MnJ enhances the recruitment of Tfh and GC B cells in mice immunized with rabies vaccines. C57BL/6 mice were immunized with RABV, MnJ-adjuvanted RABV, Alum-adjuvanted RABV, or DMEM (n = 5). At 7 and 14 dpi, 10⁵ single inguinal LN cells from four groups of mice were stained by antibodies identifying Tfh cells (CD4⁺ CXCR5⁺ PD1⁺) and GC B cells (B220⁺ GL7⁺ CD95⁺) by flow cytometry. (A) Gating strategies of Tfh or GC B cells in LN cells. (B and D) Representative flow cytometric plots of Tfh or GC B cells from four groups at 7 and 14 dpi. (C and E) Statistical results of CD4⁺ CXCR5⁺ PD1⁺ Tfh cells or B220⁺ GL7⁺ CD95⁺ GC B cells. Error bars represent the SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, no significant difference.

FIG 4.

MnJ increases the generation of PC and ASC formation. C57BL/6 mice were immunized with RABV, MnJ-adjuvanted RABV, Alum-adjuvanted RABV, or DMEM (n = 5). At 7 and 14 dpi, single BM cells were stained by antibodies for analyzing the PCs (B220^low CD138⁺) by flow cytometry, and single inguinal LN cells were prepared for counting RABV-specific ASCs by the ELISpot assay. (A) Gating strategies of PCs (B220^low CD138⁺). (B) Representative flow cytometric plots of results from four groups at 7 and 14 dpi. (C) Statistical results of B220^low CD138⁺ PCs. (D) Representative sections from the ELISpot assay. (E) Statistical results of the number of RABV-specific ASCs. Error bars represent the SEM. *, P < 0.05; **, P < 0.01; ****, P < 0.0001; ns, no significant difference.

MnJ facilitates the generation of PCs and ASCs post-RABV immunization.

In the GC reaction, B cells gain an affinity for the cognate antigen and expand preferentially. In this process, GC B cells can differentiate into either memory B cells or PCs secreting large quantities of antibodies in response to antigens (34, 35). To examine the effects of MnJ on PCs formation, the percentage of PCs was assessed by flow cytometry after vaccination with RABV, RABV with MnJ, or RABV with Alum (n = 5). The gating strategy of PCs (B220^low CD138⁺) in bone marrow (BM) cells and representative flow cytometric plots in BM cells at 7 and 14 dpi are shown in Fig. 4A and B. As can be seen in Fig. 4C, the mice immunized with MnJ-supplemented RABV generated significantly more PCs in the BM cells than the mice immunized with RABV alone or Alum-supplemented RABV. Besides, the generation of RABV-specific antibody-secreting cells (ASCs) in LNs was evaluated by an enzyme-linked immunosorbent spot (ELISpot) assay at 7 and 14 dpi. Representative sections and the numbers of RABV-specific ASCs are shown in Fig. 4D and E. As expected, compared with the mice immunized with RABV and MnJ-supplemented RABV, the mice immunized with MnJ-supplemented RABV possessed significantly more RABV-specific ASCs. Together, these results demonstrate that MnJ can better facilitate the generation of PCs and ASCs than Alum after RABV immunization.

MnJ induces higher antibody production than Alum post-RABV immunization in mice.

Since MnJ significantly augmented the percentage of PCs and ASCs, the role of MnJ in antibody production was further evaluated. ICR mice (n = 10) were i.m. inoculated with 10⁷ focus-forming units (FFU) of RABV alone, RABV with Alum (50 μg, 100 μg, and 200 μg), or RABV with MnJ (50 μg, 100 μg, and 200 μg). Serum samples were collected weekly. RABV-specific IgM and total IgG in the serum samples of immunized mice were determined by ELISA. As shown in Fig. 5A and B, MnJ but not Alum induced significantly higher levels of IgM at day 7 postimmunization (p.i.), and MnJ induced higher total IgG than Alum post-RABV immunization in mice. Next, the RABV-specific VNAs, which play a pivotal role in blocking viral fusion with cells, were measured by a fluorescent-antibody virus neutralization (FAVN) assay. As shown in Fig. 5C, Alum enhanced VNA titers induced by rabies vaccines at only 3 and 4 weeks p.i., but MnJ worked at all the time points. Besides, compared with Alum, the same dose of MnJ induced a significantly higher VNA level than Alum in mice post-RABV immunization. The dynamics of geometric mean titers (GMTs) of VNA are also presented. As shown in Fig. 5D, at 3 weeks p.i., the highest GMTs of VNA in RABV, Alum-supplemented RABV, and MnJ-supplemented RABV immunized mice were 2.76, 7.37, and 21.45 IU/ml, respectively. All together, MnJ induces more rapid and higher antibody production than Alum post-RABV immunization in mice.

FIG 5.

MnJ enhances antibody production in mice post-RABV immunization. ICR mice were immunized with 10⁷ FFU (100 μl) of RABV alone and with a mixture of RABV and different doses of Alum or MnJ (50, 100, and 200 μg) (n =10). RABV-specific IgM, total IgG, and VNAs were measured. (A and B) RABV-specific IgM (day 7 p.i.) (A) and total-IgG (1 to 4 weeks p.i.) (B) were measured by indirect ELISA. (C) RABV-specific VNAs were measured by the FAVN assay. (D) Geometric mean titers (GMT) of VNA were calculated and presented. Error bars represent the SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, no significant difference.

MnJ facilitates Th1-biased immune response and protects mice against virulent RABV challenge.

The IgG isotypes were then measured by ELISA. In line with the results of total IgG and VNAs, mice immunized with MnJ-supplemented RABV showed significantly higher levels of RABV-specific IgG2a and IgG2b (beneficial for Th1-biased immune response) at 1 to 4 weeks p.i. (Fig. 6A and B). However, Alum induced more IgG1 (beneficial for Th2-biased immune response) than MnJ at 2 to 4 weeks post-RABV immunization (Fig. 6C). Next, we evaluated the role of MnJ in protecting mice against a virulent RABV challenge. Four groups of ICR mice (n = 10) were immunized with 10⁷ FFU of RABV alone, RABV with Alum, or RABV with MnJ, respectively. At 3 weeks p.i., the immunized mice were challenged via the intracerebral (i.c.) route with 100 50% lethal doses (100× LD₅₀) of challenge virus standard 24 (CVS 24), and then the survival ratio was monitored for another 21 days. As shown in Fig. 6D, 100% of the DMEM-immunized mice succumbed to rabies within 12 days, while 100% of mice in the MnJ-supplemented RABV group survived, compared with only 50% survival in the RABV group and 80% survival in the Alum-supplemented RABV group. Our results indicate that MnJ facilitates a Th1-biased immune response post-RABV immunization and significantly enhances the protective effect of rabies vaccines.

FIG 6.

MnJ induces a Th1-biased immune response and provides enhanced protection for mice. ICR mice were immunized with 10⁷ FFU (100 μl) of RABV, RABV with Alum (200 μg), or RABV with MnJ (200 μg) (n =5). (A to C) IgG1 (A), IgG2a (B), and IgG2b (C) were measured by indirect ELISA. (D) ICR mice were immunized with 10⁷ FFU of RABV, RABV with Alum (200 μg), or RABV with MnJ (200 μg) (n =10). At 3 weeks p.i., mice were challenged with 50× LD₅₀ of CVS-24, and the percent survival was recorded for 21 days. Error bars represent the SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, no significant difference.

MnJ enhances rabies vaccine efficacy in cats and dogs.

Most human rabies cases are associated with cats and dogs (36, 37). Thus, we evaluated the adjuvant effect of MnJ in cats and dogs. Three groups of cats (n = 3 or 4) or dogs (n = 5) received a single dose of 10⁸ FFU of inactivated RABV with Alum, MnJ, or DMEM subcutaneously in the back neck area. Animals were monitored intermittently for 1 h postvaccination for any systemic adverse events. Similarly, blood samples were collected weekly and processed into serum for VNA measurement and ELISA. As shown in Fig. 7A and D, VNA titers of animals immunized with RABV alone were all below 0.5 IU/ml in both cats and dogs, while the peak VNA titers of MnJ-supplemented RABV groups reached 1.79 IU/ml (Fig. 7A) in cats and 3.39 IU/ml in dogs (Fig. 7D). The Alum has the same effect, but the peak VNA titers (1.50 IU/ml in cats and 2.88 IU/ml in dogs) were lower than those of MnJ in both cats and dogs, although the difference was not significant. Consistently, the geometric means of VNA titers and RABV-specific total IgG in cats (Fig. 7B and C) and dogs (Fig. 7E and F) elicited by MnJ were higher than those elicited by Alum. Besides, RABV-specific IgG1, IgG2, and IgM were measured by ELISA. Similarly, MnJ induced more IgG2 (Th1 biased) than Alum, while Alum induced more IgG1 (Th2 biased) than MnJ (Fig. 7G). Furthermore, MnJ induced significantly more IgM than DMEM and Alum (Fig. 7H). Overall, MnJ improves the efficacy of rabies vaccines in cats and dogs.

FIG 7.

MnJ promotes antibody production in cats and dogs post-RABV immunization. Cats (n = 3 or 4) or dogs (n = 5) were immunized with RABV supplemented with MnJ, Alum, or DMEM. Serum samples were collected weekly. (A to C) RABV-specific VNA titers (A), geometric mean titers (GMT) of VNA (B), and RABV-specific total IgG (C) in serum samples of cats (1 to 8 weeks) are shown. (D to H) RABV-specific VNA titers (D), geometric mean titers (GMT) of VNA (E), and RABV-specific total IgG (F), IgG1 and IgG2 (G), and IgM (H) in serum samples of canines (1 to 6 weeks) are shown. Error bars represent the SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, no significant difference.

DISCUSSION

Rabies is a devastating and fatal infectious disease and occurs throughout the world. Nowadays, commercial inactivated rabies vaccines are widely used to prevent RABV infection and transmission, although their efficacy is relatively poor and multiple doses are required to achieve protective immunity. Over the past few years, many types of adjuvants have been tested for rabies vaccines, e.g., bacterium-derived substances like monophosphoryl-lipid A (MPLA) (38), plant-derived substances such as saponins (39), synthetic organic compounds like squalene (40), and silver nanoparticles (41). Recently, Shi et al. (42) conducted a comparison of different adjuvants (including bacteria-like particles [BLPs], AS02, AS03, MF59, poly(I:C), and two types of Alum) for rabies vaccines. Among all adjuvants evaluated in the study, regular and nano-sized Alum afforded stronger and more durable protection (42). However, Alum was argued to delay the early production of antibodies and induce an inflammatory response (43). Therefore, exploring a safe and effective adjuvant for rabies vaccines is still required.

The activation of innate immunity exerts a distinct impact on the induction of adaptive immunity (8, 44). Molecules that can activate the innate immune system may become promising vaccine adjuvants, such as a complex containing imiquimod (a TLR7 agonist) (45) and polyinosinic-poly(C)-based adjuvant (PIKA), a TLR-3 agonist that has completed phase II clinical trials in Singapore (46, 47). Notably, Ifnar^−/− mice were shown to develop lower levels of VNAs than wild-type mice (48, 49), which implies the importance of IFN-I in antibody development. In another way, IFN-β expressed by a RABV-based HIV-1 vaccine vector served as a molecular adjuvant, and recombinant IFN-α1 had also been expressed by a recombinant RABV to enhance the immune response (50, 51). Mn²⁺ was found to be a strong type I IFN stimulator by activating the cGAS-STING pathway both in vitro and in vivo. Mn²⁺ is released from organelles and Golgi apparatus upon virus infection, subsequently facilitating the activation of DNA sensor cGAS and its downstream adaptor protein STING in the cytosol (20). Afterward, cGAS-STING agonists like cyclic GMP-AMP (cGAMP) and chitosan also showed adjuvant effects (52, 53). In our study, MnJ-treated BMDCs exhibited significantly higher levels of IFN-I (including Ifna1, Ifna2, Ifna4, and Ifnb1), as well as some chemokine and proinflammatory factor genes such as Cxcl10, Il-6, and Tnf. Previous research has established that CXCL10 could facilitate the interaction of T cells with DCs or B cells and optimize the Th1-mediated immune response (54). In addition, interleukin 6 (IL-6) and tumor necrosis factor alpha (TNF-α) were reported to play essential roles in the final differentiation of B cells into immunoglobulin-secreting cells (55, 56).

Our previous studies have demonstrated that recombinant RABV expressing chemokines or cytokines recruiting and/or activating DCs stimulates an enhanced humoral immune response (57). For example, granulocyte-macrophage colony-stimulating factor (GM-CSF) has been successfully expressed to enhance the maturation and recruitment of DCs and B cells (58), and a live vaccine containing the gene encoding canine GM-CSF was inoculated orally in dogs (27). Similar results were also observed with the addition of the macrophage inflammatory protein 1α (MIP-1α or CCL3) (26, 57), high-mobility group box1 (HMGB1) (25), and DC-binding peptide (59) to a live-attenuated or inactivated rabies vaccine. MnJ-adjuvanted RABV could significantly upregulate costimulatory molecules CD80/CD86 on DCs and promote the maturation of DCs in comparison to Alum-adjuvanted RABV and RABV alone at 3 and 6 dpi in mice (Fig. 2). Consequently, mature DCs could present antigens to CD4⁺ T cells and subsequently stimulate B cells to generate antigen-specific antibodies (60).

Massive vaccination of dogs is the most efficient strategy to interrupt RABV transmission to humans (61, 62). Hence, we explored whether MnJ as an adjuvant was effective in the cat and dog models. In the present study, the rabies vaccines alone elicited a relatively weak immune response and low VNA titers (almost below 0.5 IU/ml) in cats and dogs, while MnJ- and Alum-adjuvanted rabies vaccines induced much higher VNA titers, which were generally more than 0.5 IU/ml (Fig. 6). These results indicate that MnJ-adjuvanted rabies vaccines are effective in both cats and dogs and also imply the indispensability of adjuvants in combination with rabies vaccines used in pets.

Unlike most preexposure vaccinations, rabies vaccines are usually administered after exposure. Therefore, rabies vaccines must be capable of inducing a rapid production of VNAs to prevent the invasion of RABV into the CNS. When the antigen is introduced into an individual who has not encountered it before, a primary immune response occurs, initially with IgM production. Additionally, the pentameric structure of IgM results in high valency. A previous study suggested that one IgM antibody is needed to cover 9 or 10 glycoprotein spikes on the surface of RABV particles for neutralization, compared with one or two IgG antibodies to cover only 3 glycoprotein spikes (63, 64). In addition, Dorfmeier et al. reported that early vaccine-induced IgM could limit the dissemination of pathogenic RABV to the CNS and mediate protection against pathogenic RABV challenge (65). In this study, MnJ induced significantly more IgM in mice and dogs than Alum did (Fig. 5A and 7H) at 7 dpi. In contrast, Alum did not improve the early immune response induced by rabies vaccines. RABV supplemented with Alum has little effect on the proliferation of Tfh cells, GC B cells, PCs, and ASCs at 7 dpi. Similarly, Alum-adjuvanted RABV elicited the same levels of VNA titers and IgM as RABV alone at 7 dpi. This finding is consistent with previous observations, which showed that aluminum hydroxide might even delay early antibody production (43).

A previous study demonstrated that the Th1 immune response is superior in eliminating viruses, such as lactate dehydrogenase-elevating virus, mouse cytomegalovirus, lymphocytic choriomeningitis virus, and vaccinia virus (66, 67). Similar findings on rabies were also reported by Hooper et al. They found that Th1 cells are critical in the clearance of RABV from the CNS (68, 69). Likewise, McGettigan’s group reported that a viral vaccine vector that could elicit an IgG2a-biased antibody response was more effective against RABV infection (70). There is evidence that IFN-γ and IL-2 (produced mainly by Th1 cells) were highly induced in splenocytes from ovalbumin (OVA)-MnJ, but not OVA-Alum, while IL-4 and IL-10 (produced mainly by Th2 cells) were preferably produced via OVA-Alum immunization, which indicated that MnJ potently stimulated a Th1-biased response in mice (24). In our study, IgG subclass results suggested that coadministration of the rabies vaccine with both MnJ and Alum resulted in increased levels of RABV-specific IgG1, IgG2a, and IgG2b in comparison with those of mice immunized with vaccine alone. Consistent with previous literature, our data confirmed that more IgG2a was observed in mice immunized with MnJ-adjuvanted RABV than Alum-adjuvanted RABV at each time point, while Alum induced significantly higher levels of IgG1 than MnJ. The results were the same in dogs (71). Our data suggest that MnJ is beneficial for Th1 polarization and IgG2a production, resulting in better protection against RABV.

MnJ was identified as the second metal element that functions as an adjuvant nearly 100 years after Alum was found. First, in comparison with Alum, MnJ prefers to stimulate a faster humoral immune response and a more effective antibody subclass to fight the rabies virus. Mechanically, Alum adjuvant exerts its beneficial effects by activating the proinflammatory nucleotide-binding oligomerization domain (NOD)-like receptor protein 3 (NLRP3) pathway, along with high levels of proinflammatory factors such as IL-1β and IL-18 (72). In contrast, Mn²⁺ promoted DC maturation via cGAS-STING activation. Mn²⁺ also activated the NLRP3 inflammasome. Interestingly, IL-1β/IL-18 induction and release by Mn²⁺-activated inflammasomes were not observed (20, 24). In terms of stability, Alum-adjuvanted vaccines should avoid freezing and thus are hard to store or transport. In contrast, MnJ showed excellent stability against repeated freeze-thaw treatment and thus is convenient to store or transport (24). There is evidence that too much aluminum accumulation is associated with long-lasting macrophagic myofasciitis (73), nervous disorders (74), and bone disease (75). Otherwise, mammals maintain the Mn level in tissue via tight control of both absorption and excretion (76). Excessive Mn causes reduced Mn absorption and increased Mn metabolism and excretion (77, 78).

Given the confirmed efficacy and wide usage, aluminum adjuvant should be considered the “gold standard” against which all new adjuvant candidates are compared. Herein, in comparison with traditional Alum, MnJ serving as an adjuvant for rabies vaccines exhibited superior adjuvant effects, with the following advantages: (i) induces an earlier and better immune response, (ii) promotes Th1 immune response, (iii) has component simplicity and steadiness, and (iv) produces lower inflammation. A previous study has proved that MnJ exhibits excellent adjuvant effects to several tested antigens, including viruses, recombinant proteins, and peptides, by either intramuscular or intranasal immunization (24). At present, vaccine development for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causing coronavirus disease 2019 (COVID-19) is the highest priority of the global medical research community. Keeping in view the robust immune response induced by MnJ, it is expected to be an effective adjuvant for novel coronaviruses, which is essential to be further explored.

In summary, this study demonstrates that MnJ serving as an adjuvant of rabies vaccines facilitates the maturation of DCs and further effectively enhances Tfh cell, GC B cell, PC, and ASC proliferation, thus significantly promoting VNA production and providing better protection against virulent RABV challenge in mice. More importantly, MnJ as an adjuvant can also improve the efficacy of rabies vaccines in cats and dogs, which supports MnJ as a potential adjuvant candidate for rabies vaccines.

MATERIALS AND METHODS

Cells, viruses, adjuvants, and animals.

BSR cells (kindly provided by Bernhard Dietzschold at Thomas Jefferson University, Philadelphia, PA, USA), a cloned cell line derived from BHK-21 cells, were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Mediatech, Herndon, VA) supplemented with 10% fetal bovine serum (FBS) (Gibco, Grand Island, NY). RABV vaccine strain LBNSE carrying two mutations at amino acid positions 194 and 333 of the G protein was constructed from the SAD L16 cDNA clone, as described previously (57, 79). DRV-Mexico, a dog-derived RABV wild-type strain, was isolated from a human patient and propagated in suckling mouse brains (80, 81). An MnJ adjuvant described previously (24) was provided by MnStarer Biotechnology Co., Ltd. (Jiangsu, China). Aluminum hydroxide gel was purchased from Invivogen (CA, USA). Female C57BL/6 and ICR mice at the age of 6 to 8 weeks were purchased from the Center for Disease Control and Prevention of Hubei Province, China. The mice were housed in the animal facility at Huazhong Agricultural University. Cats were purchased from Jiaxiang Huarong Breeding Professional Cooperative. Beagles between the ages of 4 and 13 months were purchased from and housed at Yizhicheng Biological Technology Co., Ltd., Hubei, China.

Antibodies.

The antibodies with directly labeled fluorescein for flow cytometric analyses were purchased from BioLegend (CA, USA). Fluorescein isothiocyanate (FITC) anti-mouse CD11c antibody (catalog no. 117306), allophycocyanin (APC) anti-mouse CD80 antibody (catalog no. 104714), phycoerythrin (PE) anti-mouse CD86 antibody (catalog no. 105008), and PE/Cy7 anti-mouse I-A/I-E (MHC-II) antibody (catalog no. 107630) were used to analyze the maturation of DCs, including BMDCs in vitro and cDCs in vivo; FITC anti-mouse CD4 antibody (catalog no. 100510), APC anti-mouse CD185 (CXCR5) antibody (catalog no. 145506), and PE anti-mouse CD279 (PD1) antibody (catalog no. 135206) were utilized to mark T follicular helper (Tfh) cells in inguinal lymph nodes (LNs) (82); FITC anti-mouse CD45R/B220 antibody (catalog no. 103206), 647 anti-mouse GL7 antibody (catalog no. 144606), and PE anti-mouse CD95 antibody (catalog no. 554295) were used to mark GC B cells in LNs (83); FITC anti-mouse CD45R/B220 antibody (catalog no. 103206) and APC anti-mouse CD138 (Syndecan-1) antibody (catalog no. 142506) were used to analyze the number of PCs in BM cells. FITC-conjugated antibodies against the RABV N protein were purchased from Fujirebio Diagnostics, Inc. (Malvern, PA). Horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG, IgG1, IgG2a, IgG2b, and IgM for ELISA were purchased from Boster (Wuhan, China). HRP-conjugated goat anti-dog IgG, IgG1, IgG2, and IgM for ELISA were purchased from Bethyl Laboratories (TX, USA).

Preparation of BMDCs.

BMDCs were isolated and cultivated as described previously (80, 81). Briefly, 6-week-old C57BL/6 mice were euthanized, the femurs were then separated, and the BM was collected. The BM cells were filtered through a plastic 40-μm mesh and cultured in RPMI 1640 (Mediatech, Herndon, VA) supplemented with glutamine, penicillin/streptomycin, 10% heat-inactivated FBS, 20 ng/ml GM-CSF, and 10 ng/ml IL-4 (R&D Systems, MN, USA) at a density of 2 × 10⁵ cells/ml. Half of the medium was removed, and fresh DC medium was replenished on days 2 and 4. On day 6, the nonadherent cells in the culture supernatant and semiadherent cells were harvested and washed once with RPMI 1640 medium. The cells were collected in 12-well plates (10⁶ cells/ml) and ready for study.

RNA-seq analysis.

Each sample was a mixture of BMDCs from three replicates. According to the manual instructions, total RNA was extracted from cells using TRIzol (Invitrogen, Carlsbad, CA, USA) and quantified using a NanoDrop and Agilent 2100 bioanalyzer (Thermo Fisher Scientific, MA, USA). Purified mRNA was fragmented into small pieces with fragment buffer at the appropriate temperature. Then, RNA-Seq library construction was performed, and single-end 50-base reads were generated on the BGISEQ500 platform (BGI-Shenzhen, China). The sequencing data were filtered with SOAPnuke (v1.5.2), and the clean reads were mapped to the reference genome using HISAT2 (v2.0.4). Bowtie 2 (v2.2.5) was applied to align the clean reads to the reference coding gene set, and then the expression level of the gene was calculated by RSEM (v1.2.12). Differential expression analysis was performed using PossionDis with a false discovery rate (FDR) of ≤0.001 and a |Log2Ratio| of ≥1. To gain insight into the change of phenotype, GO (http://www.geneontology.org/) and KEGG (https://www.kegg.jp/) enrichment analysis of annotated differently expressed genes was performed by Phyper (https://en.wikipedia.org/wiki/Hypergeometric_distribution) based on a hypergeometric test. The significant levels of terms and pathways were corrected by q value with a rigorous threshold (q value ≤ 0.05) by Bonferroni. The heat map was drawn by GraphPad Prism software (GraphPad Software, Inc., CA) according to the gene expression significantly upregulated in the experimental group.

Quantitative real-time PCR (qRT-PCR).

Total RNA was isolated by TRIzol reagent (Invitrogen, Karlsruhe, Germany) and treated with DNase. RNA was then converted to cDNA by reverse transcription using FSQ-201 ReverTra Ace (Toyobo, Osaka, Japan). qRT-PCR analysis using SYBR green (Bio-Rad, CA, USA) was performed on an Applied Biosystems 7300 real-time PCR system (Applied Biosystems, CA, USA). Primer sets used in this study are listed in Table 1.

TABLE 1.

Primer pairs used for qRT-PCR in this study

Primer	Sequence (5′ to 3′)
Ifna1-F	CACAGCCCTCTCCATCAACT
Ifna1-R	TCCCACGTCAATCTTTCCTC
Ifna2-F	TCTGTGCTTTCCTCGTGATG
Ifna2-R	TTGAGCCTTCTGGATCTGCT
Ifna4-F	GAAGGACAGGAAGGATTTTGGA
Ifna4-R	TGAGCCTTCTGGATCTGTTGGT
Ifnb1-F	CACAGCCCTCTCCATCAAC
Ifnb1-R	GCATCTTCTCCGTCATCTCC
Irf7-F	CTGGAGCCATGGGTATGCA
Irf7-R	AAGCACAAGCCGAGACTGCT
Rsad2-F	ATAGTGAGCAATGGCAGCCT
Rsad2-R	AACCTGCTCATCGAAGCTGT
Ifit1-F	CTCTGAAAGTGGAGCCAGAAAAC
Ifit1-R	AAATCTTGGCGATAGGCTACGA
Ifit2-F	CTGAAGCTTGACGCGGTACA
Ifit2-R	ACTTGGGTCTTTCTTTAAGGCTTCT
Ifit3-F	TTCCCAGCAGCACAGAAAC
Ifit3-R	AAATTCCAGGTGAAATGGCA
Cxcl10-F	CCTGCTGGGTCTGAGTGGGA
Cxcl10-R	GATAGGCTCGCAGGGATGAT
Il6-F	ACAGAAGGAGTGGCTAAGGA
Il6-R	CGCACTAGGTTTGCCGAGTA
Tnf-F	GGGTGATCGGTCCCCAAAGG
Tnf-R	CTCCACTTGGTGGTTTGCTACGA
β-actin-F	AGGTGACAGCATTGCTTCTG
β-actin-R	GCTGCCTCAACACCTCAAC

Flow cytometry.

Flow cytometry was conducted to quantify BMDCs cultivated in vitro and immune cells in the draining LNs and BMs. Briefly, the mice were anesthetized, the draining LNs and BM cells were collected and homogenized into cell suspensions through a 40-μm nylon filter, and BMDCs were collected from the plate directly. Red blood cells were removed using lysis buffer (catalog no. 555899; BD Biosciences Inc., Franklin Lakes, NJ, USA). After washing twice, the single suspended cells in phosphate-buffered saline (PBS; containing 0.2% BSA) (1 × 10⁶ cells/sample) were incubated for 30 min at 4°C with fluorescence-conjugated antibodies in the dark. After incubation, the cells were washed 2 times with PBS (containing 0.2% BSA) and then fixed in 1% paraformaldehyde–PBS for 30 min. Finally, the data collection and analysis were performed using a BD FACSVerse flow cytometer (BD Biosciences, CA, USA) and FlowJo software (TreeStar, CA, USA).

ELISpot assay.

An ELISpot assay was conducted to analyze the RABV-specific ASCs. Inguinal LNs were isolated at 7 and 14 dpi and homogenized into cell suspensions. Multiscreen HA ELISpot plates (Millipore, MA, USA) were coated with 500 ng/well purified RABV virions and incubated for 16 h at 4°C. Coated plates were washed and blocked with RPMI 1640 supplemented with 10% FBS for 2 h at 37°C. Cell suspensions were 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 spots were quantitated. The plates were then scanned and analyzed by the Mabtech IRIS FluoroSpot/ELISpot reader, using RAWspot technology for multiplexing at the single-cell level.

Virus titration.

Virus titers were determined using a direct fluorescent-antibody assay (dFA) in BSR cells as previously described (84). Briefly, a serial 10-fold dilution of the virus was inoculated into BSR cells in 96-well microplates in procedures performed in quadruplicate, and the cells were then incubated at 37°C for 48 h. After incubation, the culture medium was discarded, and the adherent cells were fixed with 80% ice-cold acetone at –20°C for 1 h. After three washes with PBS, the cells were stained with FITC-conjugated anti-RABV N antibodies for 1 h at 37°C. Finally, fluorescent foci in the cells were counted under an Olympus IX51 fluorescence microscope (Olympus, Tokyo, JPN), and virus titers were calculated and are presented as numbers of focus-forming units (FFU) per milliliter.

Immunization and challenge experiment in mice.

RABV vaccine strain LBNSE was inactivated with millesimal formaldehyde at 37°C. Six-week-old female ICR mice were randomly divided into four groups (n = 10) and immunized with 100 μl of solution containing 10⁷ FFU of RABV with DMEM, MnJ, or Alum (Invivogen; Alhydrogel adjuvant, 2%) and 100 μl of DMEM as a control group via the intramuscular (i.m.) route. Each of the groups was then further subdivided into two subgroups. On day 21 dpi, mice were intracerebrally (i.c.) challenged with 50× LD₅₀ of CVS-24 in a volume of 30 μl and observed daily for 3 weeks.

Immunization in cats and dogs.

No history of rabies vaccination was recorded for the cats and dogs, and they were confirmed negative by VNA measurement before the study. Three groups of cats (n = 3 or 4) or dogs (n = 5) received a single dose of 10⁸ FFU of RABV with DMEM, Alum, or MnJ subcutaneously in the back neck area. Animals were monitored intermittently for 1 h postvaccination for any systemic adverse events. Similarly, blood samples were collected for VNA measurement and ELISA.

VNA measurement.

VNA titers were measured by using the fluorescent-antibody virus neutralization (FAVN) assay as previously described (82). Briefly, serum samples from mice were separated and inactivated for 30 min at 56°C. One hundred microliters of DMEM was added to a 96-well plate, and 50 μl of test serum or standard serum was added to the first column and serially diluted 3-fold. Each sample was added to four adjacent wells. A rabies challenge virus (CVS-11) suspension was added to each well, and the plates were incubated at 37°C for 1 h. Following incubation, 2 × 10⁴ BSR cells were added to each well, and the solutions were incubated at 37°C for 72 h. The samples were then fixed with 80% ice-cold acetone for 30 min and stained with FITC-conjugated antibodies against the RABV N protein. Fluorescence was observed under an Olympus IX51 fluorescence microscope (Olympus, Tokyo, Japan). The values for fluorescence were compared with the values 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 milliliter.

ELISA.

RABV-specific ELISAs were conducted to determine antibody isotypes as previously described (85). Briefly, serum samples from mice were separated and inactivated for 30 min at 56°C. ELISA plates were coated with 500 ng/well purified RABV virion diluted with protein coating buffer (5 mM Na₂CO₃, pH 9.6) overnight at 4°C. The next day, the plate was washed three times with phosphate-buffered saline (PBS)-Tween (PBST) (0.5% [wt/vol] Tween 80) and blocked in 5% low-fat milk–PBS for 2 h at 37°C. Afterwards, 100 μl of the diluted serum (1:10,000 for IgG, 1:500 for IgG1, 1:1,000 for IgG2a, 1:2,000 for IgG2b, 1:100 for IgM) was added to the plates and incubated for 1.5 h at 37°C. After incubation, the plates were washed three times in PBST, and then 100 μl of horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG, IgG1, IgG2a, IgG2b, or IgM was added to each well for 45 min at 37°C. Postincubation, the color was developed using tetramethylbenzidine (TMB) substrate (Biotime Biotechnology, Shanghai, China), and reactions were stopped with 2 M sulfuric acid. Optical densities were recorded at 450 nm using a SpectraMax 190 spectrophotometer (Molecular Devices, CA, USA).

Statistical analysis.

All data were analyzed using GraphPad Prism software (GraphPad Software, Inc., CA). For the percent survival tests, survival curves were analyzed using the log rank (Mantel-Cox) test. For the other data, one-way analysis of variance (ANOVA) was used to determine statistical significance. Error bars represent the standard deviation (SD) or the standard error of the mean (SEM). The following notations are used to indicate significant differences between groups: *, P < 0.05; **, P < 0.01; ***, P < 0.001, ****, P < 0.0001; “ns” indicates no significant difference.

Ethics statement.

All the animals used in this study were maintained in the animal facility of Huazhong Agricultural University and in compliance with the Regulations for the Administration of Affairs Concerning Experimental Animals made by the Ministry of Science and Technology of China. The experiments were carried out in accordance with the protocols (permit number HZAUMO-2019-063) approved by the Scientific Ethics Committee of Huazhong Agricultural University.

Data availability.

The RNA-seq data were deposited in the NCBI Sequence Read Archive (SRA) (http://www.ncbi.nlm.nih.gov/Traces/sra) under BioProject accession number PRJNA759407.