A self-amplifying mRNA vaccine expressing PRV gD induces robust immunity against virulent mutants

Pseudorabies virus (PRV) belongs to the alphaherpesvirus subfamily within the Herpesviridae family and the Herpesvirales order. It can infect various livestock and wild animals, including pigs, dogs, cats, foxes, rabbits, cattle, and sheep. Only pigs serve as the natural and reservoir host for PRV. Infection with PRV can lead to fatal encephalitis in newborn piglets, respiratory symptoms and growth retardation in fattening pigs as well as reproductive disorders in sows, resulting in significant porcine health and welfare issues. While PRV infection in the natural host is associated with robust virus replication in many tissues, non-productive infection is the hallmark in other hosts that regularly succumb to often-lethal brain infection¹.

The genome of PRV is approximately 143 kb in length and consists of double-stranded linear DNA. It encodes 16 envelope proteins, including 11 glycoproteins and 5 non-glycosylated type 1, type 2 and type 3 transmembrane proteins. These proteins play crucial roles in viral entry and egress, as well as in immune regulation, and they facilitate the formation of syncytia. Among the envelope proteins, gB, gC, gD, gH, and gL are especially significant, as they mediate the attachment of virions to the surface of host cells and promote the fusion of viral and host membranes. They also are primary targets recognized by host immune defenses¹. Notably, PRV gD is the receptor-binding protein and forms complexes with glycoproteins on the cell surface, which ultimately determines the ability of the virus to infect target cells, even though gD is not essential for PRV cell-to-cell spread¹. Furthermore, gD can induce the production of high levels of neutralizing antibodies and may also stimulate cellular immunity during cross-protection^2,3,4,5. Consequently, gD plays a pivotal role in eliciting protective immune responses.

The gE-deleted Bartha-K61 strain is the most widely used modified live PRV vaccine in China and plays a crucial role in the prevention and control of the virus. However, since 2011, new epidemic variants have emerged in numerous vaccinated pig herds across northern China^6,7. These variants result in relatively high pathogenicity and fatality rates, rendering the traditional Bartha-K61 vaccine insufficient for providing complete protection against the disease. Recent research has shown that PRV remains widespread in China, with genotype 2 strains dominating. Even though belonging to the same serotype, the genotype of mutant viruses differs from that of commercial PRV Bartha-K61vaccines^8,9.

While it is generally understood that humans are resistant to PRV infection, Liu et al. isolated the PRV hSD-1/2019 strain from the cerebrospinal fluid of a human patient in 2020, suggesting that PRV may indeed be capable of infecting humans¹⁰. These findings underscore the potential public health implications associated with PRV infection in the natural host. Currently, traditional inactivated vaccines, genetically engineered attenuated live vaccines, and subunit vaccines developed for PRV do not have satisfactory safety profiles nor to they exhibit sufficient immunogenicity⁶. Therefore, the key to effective prevention and control of PRV in China and elsewhere lies in the development of more effective vaccines that are based on currently circulating PRV variants. Such advancements will be crucial in reducing cross-species transmissions and potentially achieving eradication within pig herds.

mRNA vaccines have the advantages of simplicity in design and production, strong immunogenicity, rapid development, and limited safety issues^11,12. With the COVID-19 epidemic, research on and application of mRNA vaccines have increased exponentially. mRNA vaccines have considerable potential in many circumstances and are in use for both prevention and treatment^13,14. Currently, in the veterinary field, there is a focus on establishing mRNA vaccines for foot-and-mouth disease¹⁵, rabies¹⁶, influenza¹⁷, mosquito-borne flaviviruses¹⁸, porcine epidemic diarrhea virus¹⁹, porcine delta coronavirus²⁰ and monkeypox virus^21,22, all of which show great potential.

While the overall safety and immunogenicity of mRNA vaccines is proven, they are more expensive than traditional live, subunit or inactivated vaccines. To develop economical mRNA vaccines, self-amplifying RNA (saRNA) vaccines have been established²³. In most cases, the RNA replicon system of alphaviruses, including Venezuelan Equine Encephalomyelitis virus (VEEV), Semliki Forest virus (SFV) or Sindbis virus (SIN), is frequently used, which results in mRNA amplification upon delivery and production of large amounts of target antigens that in turn effectively stimulate the immune response at lower nucleic acid doses^17,24. Two saRNA vaccines for COVID-19 have recently been approved in Japan and India²³, indicating a promising strategy for application of saRNA for infectious diseases of humans and animals.

In this study, we developed an LNP-packaged saRNA vaccine based on the gD gene of a PRV variant that relies on the replicase of VEEV. In a mouse model, the PRV saRNA-gD vaccine induced significantly more robust humoral immunity than did the inactivated vaccine. Of note, repeated immunization with 5 μg of PRV saRNA gD resulted in complete clinical protection of piglets against lethal challenge infection with a currently circulating and highly virulent PRV isolate.

Generation of saRNA vaccines based on PRV gD

An in vitro transcription system was designed with the firefly luciferase gene (GenBank: M15077.1) as a marker to verify the in vitro mRNA synthesis system. The gD gene, the protective antigen of PRV, was selected as the target gene. The gD gene was optimized based on pig-origin codons. A HiBiT tag was added to detect the expression of the target protein. A modified saRNA was designed to add cap1, 5′UTR and 3′UTR and optimized gD sequences (Fig. 1a), and the in vitro transcription template was cloned and inserted into the JFV01 expression vector, which contains a T7 promoter and self-amplifying-elements derived from VEEV.

The transcription template was then subjected to in vitro transcription, and fragment integrity was tested via capillary gel electrophoresis. The saRNA-gD fragment integrity was 87%, and the percentage of saRNA-Luc fragments was 84.4%. Twenty hours after transfection of HEK-293 cells, a chemiluminescence test revealed that the control saRNA harboring Luc resulted in robust expression of firefly luciferase (Fig. 1b). The codon-optimized gD genes with the added HiBiT tag were transfected into ST cells and PK-15 cells for fusion protein detection. gD2 was expressed at high levels in both cell lines (Fig. 1c, d) and was used for the subsequent construction of the PRV saRNA gD vaccine without a HiBiT tag (Supplementary Fig. 1). gD sequences were codon-optimized, resulting more than 80% sequence identity with the wild-type nucleotide sequence of gD.

The mRNA transcripts were then encapsulated in LNPs. No free RNA was detected in the prepared LNPs, and the total RNA concentration reached more than 50 ng/μL. The average particle size of the LNPs containing saRNA-gD was 73.24 nm, the polydispersity coefficient (PdI) was 0.120, and the zeta potential was −1.72 mV. The average particle size of LNPs containing saRNA- Luc was 75.15 nm, the PdI was 0.145, and the zeta potential was −2.62 mV (Fig. 1e, f).

Immune efficacy in mice

We tested the immunogenicity of the PRV saRNA gD in a mouse model after intramuscular immunization, as shown in Fig. 2a. Mice in groups 1 to 4 were vaccinated with 10 ng/mouse (one dose), 100 ng/mouse (one dose), 10 ng/mouse (two doses of prime-boost) or 100 ng/mouse (two doses of prime-boost) PRV saRNA-gD, respectively. Mice in Group 5 were vaccinated with the inactivated PRV LA-A vaccine (two doses in a 28-day interval with each dose containing 10^6.50 TCID₅₀ PRV LA-A virus before inactivation), and those in Group 6 were vaccinated with saRNA-NC as a control. All animals in groups 3, 4 and 5 were vaccinated twice, 28 dpi (days post first inoculation) apart. Serum was collected from mice at 21 dpi and 42 dpi for the detection of PRV-neutralizing antibodies. After vaccination, the mice did not experience any adverse reactions irrespective of the formulation or dose, and no significant differences in body weights were observed after injection (Fig. 2b). Throughout the study period (0–72 h post first inoculation), no erythema, swelling or induration was observed at injection sites in mice. Histopathological analysis of intramuscular injection sites revealed healthy tissue architecture without inflammatory infiltrates, hemorrhage or necrosis in the mice that had received 100 ng of the saRNA (Fig. 2f). Furthermore, systematic examination of major organs (heart, liver, spleen, lung, brain) showed no vaccine-related pathological alterations compared to PBS controls (Fig. 2f). These findings confirmed the favorable safety profile of the saRNA-LNP platform.

At 21 dpi, animals vaccinated with the PRV saRNA-gD formulation had developed neutralizing antibody levels equivalent to those in the inactivated vaccine group. No significant difference was found between the 100 ng/mouse and 10 ng/mouse groups. At 42 dpi, the neutralizing antibody titers in mice vaccinated with 100 ng of the PRV saRNA gD reached 1:218.5, which was significantly higher than those in the LA-A inactivated vaccine booster group (Fig. 2c). However, the antibody titers in the 100 ng/mouse (one dose), 10 ng/mouse (one dose) or 10 ng/mouse groups (two doses of vaccine) were not significantly higher when compared to the LA-A inactivated vaccine booster group.

To further characterize the early innate immune activation induced by the PRV saRNA-gD vaccine, we used ELISA to detect IL-12 and IFN-α levels in the serum. There were no significant differences in the levels of the IL-12 and IFN-α produced between any of the groups, including the saRNA-NC group, on 21 dpi and 42 dpi (Fig. 2d, e).

Neutralizing antibody levels in piglets after immunization

The immunogenicity and overall efficacy of the PRV saRNA-gD vaccine in weaned piglets were evaluated. Groups 1 to 3 were vaccinated with 50 μg/piglet (one dose), 5 μg/piglet (one dose), or 5 μg/piglet (two doses of prime-boost), respectively (Fig. 3). Group 4 was vaccinated with the PRV LA2017 live vaccine (one dose), Group 5 was vaccinated with the inactivated PRV LA-A vaccine (two doses), and Group 6 was vaccinated with the saRNA-NC as a control. All piglets from Groups 3 and 5 were vaccinated twice, 28 dpi apart. After vaccination, no adverse events occurred in any of the piglets. Blood samples were collected weekly to detect PRV-specific neutralizing antibodies in serum. On 42 dpi, all animals were challenge-infected by intranasal instillation with the virulent PRV AH02LA strain at a dose of 10^6.5 TCID₅₀. At 14, 21 and 28 dpi, the average neutralizing antibody level of the animals in the three groups that had been vaccinated with the saRNA-gD vaccine was lower than that in animals immunized with the modified live virus vaccine or the inactivated vaccine, but the differences were not signicant. The levels of PRV-specific neutralizing antibodies in the 50 μg saRNA-gD group were slightly higher than those in the 5 μg group, but the difference was not significant. On 35 dpi, the level of neutralizing antibodies in the two-dose 5 μg saRNA-gD group significantly increased, and the mean neutralizing titer reached 1:200, which was significantly higher than that in the live vaccine 1:23 or inactivated vaccine 1:23 immunization groups (P < 0.0001). On 42 dpi, the mean neutralizing titer in the saRNA-gD group increased to 1:272 (Fig. 3b).

IL-4 and IFN-γ cytokine levels in the serum of piglets after immunization

To characterize the early innate immune activation induced by the saRNA-gD vaccine in weaned piglets, we used ELISA to measure the levels of IL-4 and IFN-γ. IL-4 cytokine levels were not significantly different between the groups on 21 dpi. At 7 dpi, IL-4 levels in animals that had received 5 μg of saRNA-gD was significantly greater than those in animals of the control group (P = 0.0144) or the PRV LA2017 live vaccine group (P = 0.0369) (Fig. 3c). Similarly, IFN-γ levels at 21 dpi were significantly higher in the saRNA-gD group than those of the control group (P = 0.0018). Moreover, there was no significant difference between the group receiving the PRV LA2017 live vaccine and the control group (P = 0.1173). In addition, IFN-γ in the PRV LA-A inactivated vaccine group was significantly greater than that in the control group (P = 0.0196) (Fig. 3c). On the 35 dpi, IL-4 levels in the PRV saRNA gD 5 μg group was significantly greater than that in the control group (P < 0.0001), PRV LA2017 live vaccine group (P = 0.0003) and the prime-boost group receiving the inactivated PRV LA-A vaccine (P = 0.0010). In addition, the IFN-γ in the PRV LA-A inactivated vaccine group was significantly greater than that in the control group (P = 0.0196) (Fig. 3d).

Clinical symptoms of piglets post-challenge

At 42 dpi, all piglets were challenged intranasally with 10^6.5 TCID₅₀ PRV AH02LA. On the 3 dpc (days post challenge), the body temperature of all the piglets in the control group reached above 41 °C, and all the piglets began to show typical symptoms such as breathing difficulty, purulent nasal discharge and ataxia. One piglet in the control group died on the 7dpc (Fig. 4c). One piglet in the PRV saRNA gD 50 μg primary immunization group had a body temperature exceeding 41 °C for 4 days. One piglet in the one-dose PRV saRNA gD 5 μg group presented a body temperature exceeding 41 °C for 3 days during the observation period (Fig. 4a). All piglets in the two one-dose saRNA groups manifested a loss of appetite and depression for several days. Notably, all the piglets in the two-dose prime-boost PRV saRNA gD 5 μg group did not experience fever and had no symptoms or only transient slight symptoms after challenge for no longer than one day (Fig. 4b). One piglet in the two-dose prime-boost group receiving the PRV LA-A inactivated vaccine presented a body temperature exceeding 41 °C for 3 days after the challenge but presented only mild symptoms, such as loss of appetite and a lack of spirit piglets, whereas the piglets in the PRV LA2017 live vaccine group presented no symptoms or only transient mild symptoms for 1 or 2 days (Fig. 4b, e).

Virus shedding in piglets post-challenge

Analysis of daily nasal swabs revealed virus shedding in all piglets from 3 days post-challenge (dpc), with a duration ranging from 3-10 days across groups (Table 1). Vaccination of mice with the 5 μg two-dose saRNA vaccine resulted in a vaccination efficacy that was comparable to that of the LA2017 live vaccine (P = 0.7698). Both vaccines significantly outperformed the inactivated LA-A vaccine (P = 0.8562). Notably, the 50 μg single-dose saRNA group (10^5.833±0.577 TCID₅₀/mL) and 5 μg single-dose saRNA group (10^6.325±0.967 TCID₅₀/mL) exhibited higher titers than the 5 μg two-dose regimen (P = 0.339, P = 0.0305 respectively), demonstrating dose-dependent effects. Control animals maintained consistently high viral loads (10^6.003±0.505 TCID₅₀/mL). Taken together, the the two-dose 5 μg saRNA vaccine regimen induced protection equivalent to that of the modified live virus vaccine (Fig. 4d).

Pathological and histopathological changes in lungs of piglets post-challenge

All piglets were necropsied upon death or at the end of the observation period after challenge to record gross pathological and histopathological lesions (Fig. 5a). Lungs of piglets in the negative control group developed lesions typical for PRV infection that included bleeding and swelling. There were no pathological changes in the lung tissues or in other organs of piglets in the two-dose 5 μg saRNA-gD group or the one-dose PRV LA2017 live vaccine group. Only mild swelling of the lungs was observed in the one-dose 50 μg saRNA-gD group. Similarly, there were no obvious pathological change in the one-dose 5 μg saRNA-gD group or the two-dose PRV LA-A inactivated vaccine group although we documented bleeding into the lungs of one piglet in each group (Fig. 5a, b).

Histopathological analysis revealed distinct lesions across experimental groups. In control group subjects, severe dilation and blood stasis were observed in alveolar wall capillaries, accompanied by alveolar exudates, narrowed alveolar spaces, and bronchiolar epithelial cell detachment. Lesion severity showed notable mitigation in the PRV LA-A, PRV LA2017 and one-dose saRNA-gD groups (both 50 μg and 5 μg). Specifically, the PRV LA-A group exhibited hemorrhagic manifestations characterized by vascular bleeding into the small bronchial lumen as well as alveolar capillary blood stasis adjacent to bronchioles. In the PRV LA2017 and 1-dose 5 μg saRNA-gD groups, we observed persistent capillary stasis in alveolar walls with compensatory alveolar cavity dilation. Animals in the one-dose 50 μg saRNA-gD group presented mild alveolar wall capillary dilation accompanied by localized alveolar stenosis. Notably, piglets receiving a two-dose 5 μg saRNA-gD regimen displayed essentially normal pulmonary histoarchitecture, with only minimal capillary stasis observed near small bronchi (Fig. 5a, c). These microscopic findings correlated well with corresponding gross pathological observations across all experimental groups. Importantly, qPCR analysis of liver, spleen, lung, tonsil, and lymph node tissues confirmed complete viral clearance in piglets immunized with two-dose 5 μg saRNA-gD, PRV LA-A or PRV LA2017 vaccines, with no detectable PRV DNA in any samples (Fig. 5d).

Since 2011, PRV variants, which exhibit increased pathogenicity and lethality, have emerged on pig farms in China⁷. Traditional vaccines are unable to provide adequate protection against these variants²⁵, although many candidate vaccines have been engineered utilizing various strains and have included inactivated vaccines^26,27, subunit vaccines²⁸ and live attenuated vaccines^29,30,31. Problems such as insufficient safety and immune efficacy remain. Since the outbreak of the COVID-19 pandemic, research on mRNA vaccine applications has significantly advanced^32,33. Compared with traditional vaccines, mRNA vaccines offer numerous advantages³⁴, such as ensuring good safety profiles without risks of infection, insertional mutagenesis, or integration into the host genome, and strong stimulation of both cellular and humoral immunity. The transcription template can be tailored to facilitate efficient expression of the target antigen, and mRNA vaccines possess the potential for rapid, cost-effective mass production through suitable in vitro transcription templates³⁵. Existing research on mRNA vaccine development emphasizes several critical areas, including reducing the inherent immunogenicity of the mRNA, selecting and modifying genes, optimizing the use of 5′ and 3′ UTRs of in vitro transcription templates, employing various capping methods, ensuring appropriate lengths of poly(A) tails, and choosing highly efficient delivery systems^36,37,38. In contrast to conventional mRNA vaccines, self-amplifying RNA (saRNA) vaccines incorporate the replicons of positive strand RNA viruses, particularly of alphaviruses, which enables the simulation of not only translation but RNA replication³⁹. Consequently, a smaller dose of saRNA vaccines can, in theory, elicit much higher levels of protein expression, which can significantly reduce the immunogenicity of the mRNA itself as well as reduce adverse effects of the immune responses⁴⁰.

In this study, we engineered an saRNA vaccine that incorporates self-amplifying elements of VEEV and the gD gene of PRV variant strains. Our detailed evaluation demonstrated that the saRNA-gD vaccine has substantial potential for vaccine development against pseudorabies in pigs. It is well known that gD promotes the binding of virus particles to its cognate receptors present in target cell membranes¹. It also is an important inducer of the cellular and particularly humoral (neutralizing) immune responses. High titers of neutralizing antibodies were documented after immunization with gD subunit vaccines^3,41 and gD-specific monoclonal antibodies can effectively inhibit the attachment of PRV in a non-cell-type-specific manner and prevent infection³. In addition, gD-induced cellular immunity may cooperate with neutralizing antibodies to play a synergistic role in reducing viral loads in the brain, which further suggests the possibility that gD-induced cellular immunity is cross-protective². We focused on the PRV gD protein to study the efficacy of saRNA vaccines.

Neutralizing antibodies are critical indicators of the humoral immune response. Immunization tests in mice demonstrated that the saRNA-gD vaccine did not elicit high levels of neutralizing antibodies after a single immunization, indicating that saRNA vaccines require a booster as inactivated vaccines regularly do. After booster immunization, neutralizing antibodies produced by mice immunized with as low as 10 ng saRNA-gD were comparable to those produced by the inactivated vaccine. Notably, the neutralizing antibody titers in mice in the two-dose saRNA-gD group (100 ng) were as high as 1:218.5, significantly surpassing the levels induced by the inactivated vaccine. These findings suggested that the saRNA vaccine may be more effective than the inactivated vaccine in inducing humoral immunity.

We therefore went on and performed a study in the natural host of PRV. The results of the piglet protection experiment demonstrated that immunization with as low as 5 μg of the saRNA-gD vaccine provided complete protection against challenge infection with a highly virulent, contemporaneous strain. The protective ability was similar to that induced by the modified live virus vaccine PRV 2017 (Fig. 4e). Compared with single-dose immunization, booster immunization significantly reduced clinical symptoms and histopathological lesions of the lung after challenge infection (Fig. 4b and Fig. 5c). Neutralizing antibody assays in serum of piglets revealed that neutralizing antibody levels in the saRNA-gD groups (both 50 μg and 5 μg) at 14 and 21 dpi were lower than those in the one-dose PRV LA2017 live vaccine group and the PRV LA-A inactivated vaccine group. Notably, 28 dpi, levels of neutralizing antibodies in the live vaccine and inactivated vaccine groups were significantly higher than those in both the 50 μg and 5 μg saRNA-gD immunization groups. These findings suggest that the neutralizing antibody response elicited by different doses of the saRNA-gD vaccine upon single immunization is limited. This finding may also indicate that the amount of antigen needed to effectively stimulate hosts immune responses after a single immunization using only gD are insufficient, potentially necessitating the co-stimulation of multivalent antigens. In contrast, both live and inactivated vaccines may stimulate the production of antibodies against antigens other than gD, which could lead to increased neutralizing antibody titers⁴². Surprisingly, only at 7 days following booster immunization, the neutralizing antibody level in the two-dose prime-boost 5 μg PRV saRNA gD group significantly increased and was significantly greater than not only the live vaccine group but also the inactivated vaccine group. Furthermore, on the 14 days after booster immunization, an increasing level of neutralizing antibodies was detected in the prime-boost 5 μg PRV saRNA gD group, whereas the inactivated vaccine group and live vaccine group showed a slight downward trend. The findings suggest that the PRV saRNA gD vaccine may have induced strong immune memory and that booster vaccination could overcome the interference from pre-existing immunity and stimulate memory cells. Additionally, our results show that the PRV saRNA gD vaccine can stimulate piglets to produce high level cytokines (IFN-γ, IL-4) in piglets, which were superior to those induced by the live vaccine and inactivated vaccine, indicating that the saRNA vaccine can promote a more balanced and stronger immune response (Fig. 3c, d).

Taken together, our study demonstrates that the PRV saRNA-gD vaccine elicits robust humoral immunity in piglets, with neutralizing antibody titers and cytokine levels significantly higher than those induced by conventional LA2017 live and LA-A inactivated vaccines. However, several limitations deserve consideration. First, the current analysis focused on systemic immune responses without direct quantification of gD-specific T-cell immunity, leaving the cellular immune responses incompletely characterized. In addition, insufficient murine serum volumes precluded IgG subclass analysis (IgG1/IgG2a) for determining Th1/Th2 polarization. Second, although the single gD antigen application proved to work, it resulted in limited impact on virus shedding compared to live vaccines. This may suggest a need for multivalent formulations incorporating additional structural or non-structural antigens to achiever broader immune responses. Third, while the LNP delivery system demonstrated excellent safety, optimization for mucosal delivery through adjuvants or alternative administration routes could further improve protection. Despite these limitations, the vaccine’s ability to match the protective efficacy of live vaccines while maintaining an exceptional safety profile provide evidence for its potential as a next-generation PRV vaccine platform. Future studies should prioritize comprehensive cellular immunity analysis, multivalent antigen design, and mucosal delivery optimization to fully realize this platform’s potential.

Virus and cells

ST cells (ATCC® CRL-1746) and PK-15 cells (ATCC® CCL-33™) were purchased from the American Type Culture Collection (ATCC, USA) and cultured in Dulbecco’s Modified Eagle Medium (DMEM, Thermo Fischer Scientific, USA) containing 10% fetal bovine serum (FBS, Gibco, USA) with 100 IU/mL penicillin and 100 g/mL streptomycin (both from Sigma-Aldrich, USA) in an environment of 5% CO₂ at 37 °C. The PRV variant AH02LA (CGMCC No. 10891, isolated in 2016 from the brain tissue of an aborted fetus delivered by a sow during a pseudorabies outbreak on a swine farm in Anhui Province, China), attenuated LA2017 (generated by in vitro 120 sequential passages of of parental AH02LA) and inactivated LA-A (constructed via gE gene deletion from AH02LA backbone) were identified and preserved in our laboratory.

mRNA construction and LNP encapsulation

The gD protein (GenBank accession number: KR605321.1) of the PRV variant AH02LA was used as a reference sequence and two gD sequences (gD1 and gD2 with different suspected spatial structure) were codon optimized from pig origin for mRNA vaccine design. The JFV01 expression vector was designed and constructed by Jinfa Pharmaceutical (Nanjing) Co., Ltd.; it contains self-amplifying elements from the VEEV and the T7 promoter to construct an in vitro transcription template for the target gene, and the plasmid was linearized via BspQ I for in vitro transcription.

LNPs were synthesized via microfluidic mixing (iNanoE, Ignite Biotech) using an organic phase containing ALC-0315, DSPC, cholesterol and ALC-0159 (46.3:9.4:42.7:1.6 molar ratio; MedChemExpress, USA) in ethanol, and an aqueous phase of mRNA in 100 mM citrate buffer (pH 4.0). Precursors were mixed at 1:3 ethanol:aqueous ratio (N:P 6:1) with 12 mL/min flow rate. Post-synthesis, LNPs were dialyzed against 10 mM Tris buffer (pH 7.4) supplemented with 8% (w/v) sucrose using Amicon® ultracentrifuge filters (30 kDa MWCO; 4 °C, 3000 × g, 25 min) and stored at −80 °C.

mRNA expression of genes encoding Luc and PRV-gD

Luc-mRNA was transfected into HEK293 cells via the HieffTrans® mRNA transfection reagent (Yeasen, China) and the expression of firefly luciferase was detected via a chemiluminescent instrument. gD1-HiBiT-mRNA and gD2-HiBiT-mRNA were transfected into PK-15 cells and ST cells, respectively, and the expression of the fusion protein was detected via the Nano-Glo® HiBiTLytic Detection System (Promega, USA).

Encapsulation efficiency and effective nucleic acid concentration

RNA encapsulation efficiency and concentration were determined by the Equalbit RNA BR Assay Kit (EQ212-01, Vazyme, China). Both standards and samples were diluted with 1× Tris-EDTA (TE) buffer, pH 8.0. Fluorescence was measured using Qubit3.0 (Thermo Fisher Scientific, USA). RNA encapsulation of LNP samples was determined by comparing the signal of the RNA-binding fluorescent dye in the absence and presence of a detergent (0.1% Triton X-100). In the absence of a detergent, the signal comes only from accessible (unencapsulated) RNA. In the presence of a detergent, the LNP is disrupted so that the measured signal comes from the total RNA (both encapsulated and non-encapsulated). The encapsulation percentage is calculated using the following equation: Encapsulation efficiency (%) = ([Fluorescence]total − [fluorescence]unencapsulated)/(fluorescence)total × 100%.

Particle size and zeta potential

Dynamic light scattering (Malvern Panalytical, England) was performed to determine the hydrodynamic size, polydispersity index, and zeta-potential of the LNPs. For hydrodynamic size measurements, 10 μL of particles were diluted in 800 μL of deionized water and placed into the 1.5 mL cuvette (Thermo Fisher Scientific, USA) for measurement. The same sample was transferred to the folded capillary zeta cell (Malvern Panalytical, England) to measure the zeta-potential. Electron microscopy was conducted to qualitatively assess the LNP size and polydispersity.

gD protein expression in saRNA-transfected ST cells

To confirm gD protein expression from saRNA-transfected ST cells, Western blot (WB) and immunofluorescence assay (IFA) were performed using convalescent sera collected from PRV AH02LA-infected piglets. For WB analysis, cell lysates were separated by 10% SDS-PAGE and transferred to PVDF membranes, followed by incubation with the mouse anti-PRV gD monoclonal antibody (1:1000, Boehringer Ingelheim, Germany) and HRP-conjugated anti-mouse IgG secondary antibody (1:5000, Sigma-Aldrich, USA). For IFA, transfected cells were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and probed with anti-PRV gD monoclonal primary antibody (1:1000, Boehringer Ingelheim) and FITC anti-mouse IgG (1:1000, Abcam, England). Fluorescence signals were captured using a Zeiss LSM 880 confocal microscope. Mock-transfected cells served as negative controls and all assays were conducted in triplicate to ensure reproducibility.

Mouse vaccination test design

The 4- to 6-week-old healthy Institute of Cancer Research (ICR) female mice (18–22 g) were purchased from the Comparative Medicine Center of Yangzhou University (Yangzhou, China). Thirty-six ICR mice were randomly divided into 6 groups, with 6 mice in each cage. The plants were immunized via intramuscular injection, booster immunized at 28 dpi, and weighed once before vaccination and at 3, 6 and 9 dpi. Orbital blood samples were collected at 21 and 42 dpi, and the serum was separated to test neutralizing antibody and cytokine levels.

Test design for piglet challenge protection

Eighteen PRV/PRRSV/PCV2-seronegative weaned piglets (28–35 days old, Henan Minwang Agriculture & Animal Husbandry Co., Ltd., China) were randomly allocated to six groups (n = 3/group) receiving: saRNA-LNP vaccine, live attenuated PRV LA2017 (positive control), inactivated PRV LA-A (positive control), or corresponding placebo controls. All vaccines were administered intramuscularly (prime at day 0; booster at 28 days post-immunization, dpi). Serum samples collected via anterior vena cava puncture at 14, 21, 28, 35, and 42 dpi were analyzed for neutralizing antibodies and cytokines.

At 42 dpi, animals were challenged intranasally with 10^6.5 TCID₅₀ PRV AH02LA. Post-challenge clinical signs were evaluated daily for 10 days by two blinded observers using a 10-point scoring system assessing respiratory distress (0: <40 breaths/min; 1: 40–50; 2: 51–60; 3: >60 with abdominal effort), neurological signs (0: normal; 1: ataxia; 2: head tilt/tremor; 3: paralysis), systemic status (0: active; 1: appetite ↓ <30%; 2: depression; 3: recumbency) and fever (0: <41.0 °C; 1: ≥ 41.0 °C), any mortality resulted in a fixed score of 10 points on the day of death, superseding other symptom scores. Viral shedding quantification via nasal swab titration on ST cells, and terminal necropsy at study conclusion for gross/histopathological evaluation.

Serum neutralization titer test

To evaluate the neutralizing antibody level in piglets in the vaccinated group, ST cells were plated into a 96-well cell plate and cultured for 24 hours. The serum samples to be tested and the PRV-positive and PRV-negative sera were deactivated in a water bath at 56 °C for 30 min. Serum samples were diluted appropriately on the basis of sample size and predicted antibody titers. The mixture was mixed with an equal volume of PRV AH02LA (containing 200 TCID₅₀) virus mixture and incubated at 37 °C for 1 h. The premixed samples were then transferred into a 96-well cell plate prepared in advance, incubated at 37 °C, incubated with 5% CO₂ for 1 h, and 100 μL of maintenance solution was added to each well. After being cultured at 37 °C and 5% CO_2, the cytopathic effects were observed daily, the number of CPEs at each dilution was recorded, and the results were read 5 days after infection. The tissue culture infective dose of 50% (TCID₅₀) was calculated according to the Reed‒Muench method, after which the neutralizing antibody titer of the serum was calculated. All assays were conducted in triplicate to ensure reproducibility.

Cytokine testing

Serum cytokine levels were quantified using commercial ELISA kits according to standardized protocols. For mouse samples, IL-12 were determined using the Mouse IL-12 p70 ELISA Kit (R&D Systems, USA), while IFN-α levels were measured with the Mouse IFN-α ELISA Kit (PBL Assay Science, USA). Porcine serum analysis employed the Porcine IL-4 ELISA Kit (R&D Systems, USA) and Porcine IFN-γ ELISA Kit (Thermo Fisher Scientific, USA). All assays were performed in duplicate using 100 μL serum samples per well. Following manufacturer instructions, plates were incubated with detection antibodies and streptavidin-HRP conjugates, developed with TMB substrate solution (Sigma-Aldrich, USA), and reactions stopped with 2N sulfuric acid. Absorbance was measured at 450 nm using a SpectraMax i3x microplate reader (Molecular Devices, USA). Cytokine concentrations were calculated from standard curves (15.6–1000 pg/mL range) included in each kit, with appropriate positive and negative controls run in parallel.

Virus shedding testing

To evaluate whether the piglets in the vaccinated group could stop or reduce virus shedding, nasal swabs were collected post-challenge. The nasal swab samples were placed on a vortex instrument and shaken fully for approximately 3 min to allow the virus attached to the nasal swabs to be shed and released into DMEM. The samples were then centrifuged at 10,000 rpm/min and 4 °C for 10 min and filtered with a 0.22 μm filter. 100 μL of the supernatant was diluted to 10⁻⁶ via a 10-fold dilution method, the dilution was inoculated into a monolayer ST 96-well cell culture plate, which was incubated in a cell culture incubator for 2 h, the old culture medium was discarded, 100 μL of DMEM culture medium containing 1% cyanine‒streptomycin resistance and 10% FBS was added to each well, and the mixture was placed in a 5% CO₂, 37 °C, constant-temperature incubator to continue culture. The cell status was observed continuously for 3 to 5 days, the occurrence of CPE was recorded and the virus shedding status was analyzed. 100 μL of the supernatant of the samples at 3 and 4 days post-challenge was taken and diluted to 10⁻⁶ via a 10-fold dilution method and then inoculated into a monolayer ST 96-well cell culture plate to detect the titers as described above.

Viral DNA quantification in tissues

Post-euthanasia tissues (liver, lung, spleen, tonsil, lymph nodes) were homogenized and genomic DNA extracted using the MiniBEST Universal Genomic DNA Extraction Kit (Takara, Japan). qPCR was performed on a LightCycler® 480 II (Roche, Switzerland) with TB Green® Premix Ex Taq™ (Takara, Japan) under cycling: 95°C/30 s; 45 cycles of 95°C/5 s, 60°C/30 s. PRV gD gene detection used primers (F: 5′-CTGGTGCTGGTGCTGTTG-3′, R: 5′-CACGATGTTGTCGTCGTAGTT-3′). Viral copies were quantified against a serially diluted gD plasmid standard (10⁴–10⁸ copies/μL). Data expressed as log₁₀ copies/mg tissues (sensitivity: 100 copies/reaction).

Evaluation of gross and histological lesions

Following challenge infection, lung tissue specimens were systematically collected from all experimental groups and immediately fixed in 10% neutral buffered formalin (Sigma-Aldrich, St. Louis, MO, USA) for 48 h at room temperature. For gross pathology evaluation, two independent veterinary pathologists blinded to treatment groups scored lesions using a standardized 0–4 scale (0 = no lesions; 1 = mild; 2 = moderate; 3 = severe; 4 = very severe) assessing three key parameters: (1) pulmonary consolidation area, (2) hemorrhage severity, and (3) pleural involvement.

For histopathological analysis, fixed tissues were processed through graded ethanol, xylene, and paraffin embedding using standard protocols³¹. Sections (5 μm) were cut, mounted on charged slides and H&E stained (Leica ST5020 stainer, Germany). Two blinded pathologists evaluated lesions by light microscopy (Olympus BX53, Japan) at 100–400× magnification (5 fields/section), using a 12-point scoring system (0–3 points each for: (1) inflammatory infiltration, (2) alveolar septal thickening, (3) hemorrhage/edema and (4) bronchiolar epithelial damage). Digital images were captured using the PANNORAMIC DESK/MIDI/250/1000 whole-slide scanner (3DHISTECH, Hungary) and analyzed with CaseViewer 2.4 virtual microscopy software (3DHISTECH, Hungary).

Animal welfare and ethics statement

During all treatments, animals were conscious and not anaesthetized. Animals meeting predefined humane endpoints (>20% body weight loss within 24 h or severe neurological impairment) and all surviving individuals at study termination were euthanized. Mice were euthanized by cervical dislocation performed by trained personnel, with death confirmed by absent corneal reflex and respiratory arrest. Piglets received intramuscular xylazine sedation (2 mg/kg) followed by intravenous pentobarbital sodium (150 mg/kg), ensuring irreversible unconsciousness prior to cardiac arrest.

All animal experiments were approved by the Institutional Animal Care and Ethics Committee at the Jiangsu Academy of Agriculture Sciences (authorization number SYXK (Su) 2020-0023) and performed strictly with the guidelines provided by the Institutional Biosafety Committee. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent was obtained from the owners for the participation of their animals in this study.

Statistical analysis

GraphPad Prism version 10.0 was used to perform the statistical analyses and analyzed using a one-way analysis of variance (ANOVA) with a Tukey’s post-hoc test or two-way ANOVA with Šidák’s multiple comparisons test. Asterisks indicate statistical significance: *P < 0.05, **0.001 < P < 0.01, ***P < 0.001, ****P < 0.0001, ns no statistical difference. Error bars indicate ± standard deviation (SD).

1. Pomeranz, L. E., Reynolds, A. E. & Hengartner, C. J. Molecular biology of pseudorabies virus: impact on neurovirology and veterinary medicine. Microbiol. Mol. Biol. Rev. 69, 462–500 (2005).

   CAS  PubMed  PubMed Central  Google Scholar 

2. Ren, J. et al. Glycoproteins C and D of PRV Strain HB1201 contribute individually to the escape from Bartha-K61 vaccine-induced immunity. Front Microbiol 11, 323 (2020).

   PubMed  PubMed Central  Google Scholar 

3. Zhang, T. et al. A single dose glycoprotein D-based subunit vaccine against pseudorabies virus infection. Vaccine 38, 6153–6161 (2020).

   CAS  PubMed  Google Scholar 

4. Jiang, Z. et al. Immunogenicity and protective efficacy induced by an mRNA vaccine encoding gD antigen against pseudorabies virus infection. Vet. Microbiol. 251, 108886 (2020).

   CAS  PubMed  Google Scholar 

5. Cao, Z. et al. Efficacy of a gB + gD-based subunit vaccine and the adjuvant granulocyte-macrophage colony stimulating factor for pseudorabies virus in rabbits. Front. Microbiol. 13, 965997 (2022).

   PubMed  PubMed Central  Google Scholar 

6. Bo, Z. & Li, X. A review of pseudorabies virus variants: genomics, vaccination, transmission, and zoonotic potential. Viruses https://doi.org/10.3390/v14051003 (2022).

7. Wu, R., Bai, C., Sun, J., Chang, S. & Zhang, X. Emergence of virulent pseudorabies virus infection in northern China. J. Vet. Sci. 14, 363–365 (2013).

   PubMed  PubMed Central  Google Scholar 

8. Liu, Q. et al. The epidemiology and variation in pseudorabies virus: a continuing challenge to pigs and humans. Viruses https://doi.org/10.3390/v14071463 (2022).

9. Zhao, P. et al. Epidemiological investigation, risk factors, spatial-temporal cluster, and epidemic trend analysis of pseudorabies virus seroprevalence in China (2017 to 2021). Microbiol. Spectr. 11, e0529722 (2023).

   PubMed  Google Scholar 

10. Liu, Q. et al. A novel human acute encephalitis caused by pseudorabies virus variant strain. Clin. Infect. Dis. 73, e3690–e3700 (2021).

    CAS  PubMed  Google Scholar 

11. Zhang, G., Tang, T., Chen, Y., Huang, X. & Liang, T. mRNA vaccines in disease prevention and treatment. Signal Transduct. Target Ther. 8, 365 (2023).

    CAS  PubMed  PubMed Central  Google Scholar 

12. Gote, V. et al. A comprehensive review of mRNA vaccines. Int. J. Mol. Sci. https://doi.org/10.3390/ijms24032700 (2023).

13. Szabó, G. T., Mahiny, A. J. & Vlatkovic, I. COVID-19 mRNA vaccines: platforms and current developments. Mol. Ther. 30, 1850–1868 (2022).

    PubMed  PubMed Central  Google Scholar 

14. Haq, H. N. et al. Pfizer-BioNTech (BNT162b2), Moderna (mRNA-1273) COVID-19 mRNA vaccines and hypersensitivity reactions. J. Natl Med. Assoc. 114, 601–612 (2022).

    PubMed  Google Scholar 

15. Pulido, M. R., Sobrino, F., Borrego, B. & Sáiz, M. RNA immunization can protect mice against foot-and-mouth disease virus. Antivir. Res. 85, 556–558 (2010).

    CAS  PubMed  Google Scholar 

16. Liu, J. et al. A nucleoside-modified mRNA vaccine forming rabies virus-like particle elicits strong cellular and humoral immune responses against rabies virus infection in mice. Emerg. Microbes Infect. 13, 2389115 (2024).

    PubMed  PubMed Central  Google Scholar 

17. Vogel, A. B. et al. Self-amplifying RNA vaccines give equivalent protection against influenza to mRNA vaccines but at much lower doses. Mol. Ther. 26, 446–455 (2018).

    CAS  PubMed  Google Scholar 

18. Kofler, R. M. et al. Mimicking live flavivirus immunization with a noninfectious RNA vaccine. Proc. Natl Acad. Sci. USA 101, 1951–1956 (2004).

    CAS  PubMed  PubMed Central  Google Scholar 

19. Zhao, Y. et al. PEDV-spike-protein-expressing mRNA vaccine protects piglets against PEDV challenge. mBio 15, e0295823 (2024).

    PubMed  Google Scholar 

20. Li, J. et al. A spike-based mRNA vaccine that induces durable and broad protection against porcine deltacoronavirus in piglets. J. Virol. 98, e0053524 (2024).

    PubMed  Google Scholar 

21. Zeng, J. et al. Mpox multi-antigen mRNA vaccine candidates by a simplified manufacturing strategy afford efficient protection against lethal orthopoxvirus challenge. Emerg. Microbes Infect. 12, 2204151 (2023).

    PubMed  PubMed Central  Google Scholar 

22. Zuiani, A. et al. A multivalent mRNA monkeypox virus vaccine (BNT166) protects mice and macaques from orthopoxvirus disease. Cell 187, 1363–1373.e1312 (2024).

    CAS  PubMed  Google Scholar 

23. Silva-Pilipich, N., Beloki, U., Salaberry, L. & Smerdou, C. Self-amplifying RNA: a second revolution of mRNA vaccines against COVID-19. Vaccines https://doi.org/10.3390/vaccines12030318 (2024).

24. Pourseif, M. M. et al. Self-amplifying mRNA vaccines: Mode of action, design, development and optimization. Drug Discov. Today 27, 103341 (2022).

    CAS  PubMed  Google Scholar 

25. Zhou, J., Li, S., Wang, X., Zou, M. & Gao, S. Bartha-k61 vaccine protects growing pigs against challenge with an emerging variant pseudorabies virus. Vaccine 35, 1161–1166 (2017).

    CAS  PubMed  Google Scholar 

26. Hua, T. et al. Protective efficacy of intranasal inactivated pseudorabies vaccine is improved by combination adjuvant in mice. Front. Microbiol. 13, 976220 (2022).

    PubMed  PubMed Central  Google Scholar 

27. Gu, Z. et al. A novel inactivated gE/gI deleted pseudorabies virus (PRV) vaccine completely protects pigs from an emerged variant PRV challenge. Virus Res. 195, 57–63 (2015).

    CAS  PubMed  Google Scholar 

28. Ren, X. et al. A self-assembled nanoparticle vaccine based on pseudorabies virus glycoprotein D induces potent protective immunity against pseudorabies virus infection. Vet. Microbiol. 284, 109799 (2023).

    CAS  PubMed  Google Scholar 

29. Sun, L., Tang, Y., Yan, K. & Zhang, H. Construction of a quadruple gene-deleted vaccine confers complete protective immunity against emerging PRV variant challenge in piglets. Virol. J. 19, 19 (2022).

    CAS  PubMed  PubMed Central  Google Scholar 

30. Ma, Z. et al. Pathogenicity and immunogenicity of a quadruple gene-deleted pseudorabies virus variant as a vaccine candidate. Vet. Microbiol. 288, 109931 (2024).

    CAS  PubMed  Google Scholar 

31. Xu, M. et al. Construction of pseudorabies virus variant attenuated vaccine: codon deoptimization of US3 and UL56 genes based on PRV gE/TK deletion strain. Front. Microbiol. 14, 1248573 (2023).

    PubMed  PubMed Central  Google Scholar 

32. Fang, E. et al. Advances in COVID-19 mRNA vaccine development. Signal Transduct. Target Ther. 7, 94 (2022).

    CAS  PubMed  PubMed Central  Google Scholar 

33. Teo, S. P. Review of COVID-19 mRNA vaccines: BNT162b2 and mRNA-1273. J. Pharm. Pract. 35, 947–951 (2022).

    PubMed  Google Scholar 

34. Fiolet, T., Kherabi, Y., MacDonald, C. J., Ghosn, J. & Peiffer-Smadja, N. Comparing COVID-19 vaccines for their characteristics, efficacy and effectiveness against SARS-CoV-2 and variants of concern: a narrative review. Clin. Microbiol. Infect. 28, 202–221 (2022).

    CAS  PubMed  Google Scholar 

35. Xu, S., Yang, K., Li, R. & Zhang, L. mRNA vaccine era-mechanisms, drug platform and clinical prospection. Int. J. Mol. Sci. https://doi.org/10.3390/ijms21186582 (2020).

36. Linares-Fernández, S., Lacroix, C., Exposito, J. Y. & Verrier, B. Tailoring mRNA vaccine to balance innate/adaptive immune response. Trends Mol. Med. 26, 311–323 (2020).

    PubMed  Google Scholar 

37. Chen, H. et al. Branched chemically modified poly(A) tails enhance the translation capacity of mRNA. Nat. Biotechnol. https://doi.org/10.1038/s41587-024-02174-7 (2024).

    Article  PubMed  PubMed Central  Google Scholar 

38. Lui, K. H., Geisberg, J. V., Moqtaderi, Z. & Struhl, K. 3’ Untranslated regions are modular entities that determine polyadenylation profiles. Mol. Cell Biol. 42, e0024422 (2022).

    PubMed  Google Scholar 

39. McCafferty, S. et al. A dual-antigen self-amplifying RNA SARS-CoV-2 vaccine induces potent humoral and cellular immune responses and protects against SARS-CoV-2 variants through T cell-mediated immunity. Mol. Ther. 30, 2968–2983 (2022).

    CAS  PubMed  PubMed Central  Google Scholar 

40. Wayne, C. J. & Blakney, A. K. Self-amplifying RNA COVID-19 vaccine. Cell 187, 1822–1822.e1821 (2024).

    CAS  PubMed  Google Scholar 

41. Zhao, M. et al. Pseudorabies gD protein protects mice and piglets against lethal doses of pseudorabies virus. Front. Microbiol. 14, 1288458 (2023).

    PubMed  PubMed Central  Google Scholar 

42. Hou, F. et al. mRNA vaccines encoding fusion proteins of monkeypox virus antigens protect mice from vaccinia virus challenge. Nat. Commun. 14, 5925 (2023).

    CAS  PubMed  PubMed Central  Google Scholar 

Download references

Author notes

1. These authors contributed equally: Tong Ling, Zhang Xin.

2. These authors jointly supervised this work: Nikolaus Osterrieder, Xia Shu-hua, Wang Ji-chun.

Authors and Affiliations

1. Institute of Veterinary Immunology & Engineering, Jiangsu Academy of Agricultural Sciences, Nanjing, 210014, China

   Ling Tong, Xin Zhang, Huan-huan Li, Ya-ting Zheng, Ya-mei Liu, Sai-sai Chen, Chuan-jian Zhang & Ji-chun Wang

2. National Research Center of Engineering and Technology for Veterinary Biology, Jiangsu Academy of Agricultural Sciences, Nanjing, 210014, China

   Ling Tong, Ya-ting Zheng, Ya-mei Liu, Sai-sai Chen, Chuan-jian Zhang & Ji-chun Wang

3. Jiangsu Coinnovation Center for Prevention and Control of Important Animal Infectious Diseases and Zoonoses, Yangzhou, 225009, China

   Ling Tong, Ya-ting Zheng, Ya-mei Liu, Sai-sai Chen, Chuan-jian Zhang & Ji-chun Wang

4. GuoTai (Taizhou) Center of Technology Innovation for Veterinary Biology, Taizhou, 225300, China

   Ling Tong, Ya-ting Zheng, Ya-mei Liu, Sai-sai Chen, Chuan-jian Zhang & Ji-chun Wang

5. College of Veterinary Medicine, Nanjing Agricultural University, Nanjing, 210095, China

   Xin Zhang

6. College of Animal Science and Technology, Guangxi University, Nanning, 530004, China

   Huan-huan Li

7. Institute of Veterinary Medicine, Jiangsu Academy of Agricultural Science, Nanjing, 210014, China

   Rong-li Guo

8. Faculty of Medical Sciences, National University of Cuyo, Mendoza, 5500, Argentina

   Armando Mario Damiani

9. Jinfa Pharmaceutical (Nanjing) Co, Ltd, Nanjing, 210032, China

   Rui Duan, Shu-yu Mao, Jia-li Yu, Qian-qian Zhang, Ruo-nan Tao & Shu-hua Xia

10. Tierärztliche Hochschule Hannover, Bünteweg 2, 30559, Hannover, Germany

    Nikolaus Osterrieder

Contributions

W.J. and S.X. conceived the project; T.L., Z.X., L.H., Z.Y., L.Y., G.R. and C.S. and performed experiments; D.R, M.S., Y.J., Z.Q., T.R. and X.S. contributed to methodology; W.J., T.L., Z.X., Z.C. and D.A. collected and analyzed the data; Z.X. and T.L. drafted the manuscript; W.J. and O.N. revised the manuscript; T.L. and Z.X. are co-first authors of this manuscript. All authors have read and agreed to the published version of the manuscript.

Corresponding authors

Correspondence to Nikolaus Osterrieder, Shu-hua Xia or Ji-chun Wang.

Competing interests

The authors declare no competing interests.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions