Comparison of gE/gI- and TK/gE/gI-Gene-Deleted Pseudorabies Virus Vaccines Mediated by CRISPR/Cas9 and Cre/Lox Systems.

Pseudorabies (PR), caused by pseudorabies virus (PRV), is an acute and febrile infectious disease in swine. To eradicate PR, a more efficacious vaccine needs to be developed. Here, the gE/gI- and TK/gE/gI-gene-deleted recombinant PRV (rGXΔgE/gI and rGXΔTK/gE/gI) are constructed through CRISPR/Cas9 and Cre/Lox systems. We found that the rGXΔTK/gE/gI was safer than rGXΔgE/gI in mice. Additionally, the effects of rGXΔgE/gI and rGXΔTK/gE/gI were further evaluated in swine. The rGXΔgE/gI and rGXΔTK/gE/gI significantly increased numbers of IFN-γ-producing CD4+ and CD8+ T-cells in swine, whereas there was no difference between rGXΔgE/gI and rGXΔTK/gE/gI. Moreover, rGXΔgE/gI and rGXΔTK/gE/gI promoted a PRV-specific humoral immune response. The PRV-specific humoral immune response induced by rGXΔgE/gI was consistent with that caused by rGXΔTK/gE/gI. After the challenge, swine vaccinated with rGXΔgE/gI and rGXΔTK/gE/gI showed no clinical signs and viral shedding. However, histopathological detection revealed that rGXΔgE/gI, not rGXΔTK/gE/gI, caused pathological lesions in brain and lung tissues. In summary, these results demonstrate that the TK/gE/gI-gene-deleted recombinant PRV was safer compared with rGXΔgE/gI in swine. The data imply that the TK/gE/gI-gene-deleted recombinant PRV may be a more efficacious vaccine candidate for the prevention of PR.


Introduction
Pseudorabies (PR), also known as Aujeszky's disease (AD), is an acute and febrile infectious disease in swine, and it is caused by pseudorabies virus (PRV) [1,2]. PR is characterized by reproductive failure, neurological disorders, and respiratory ailments in swine [3][4][5]. The disease can lead to substantial economic costs because of reproductive losses in sows and weight loss in PRV-infected adults in swine farmers [2]. Currently, some strategies based on the DIVA (differentiating infected from vaccinated individuals) vaccination program are applied to the eradication of PR in commercial swine populations [6]. A few developed countries, such as United States [3], several European countries [7],

Cre-Mediated Recombination In Vitro
Conditions for Cre-mediated recombination in vitro were described in a previous study [31]. Briefly, 2 µg of rGX∆gE/gI or rGX∆TK/gE/gI genome, 50 mM Tris-HCI, pH 7.5/33 mM NaCI/10 mM MgCl2 and 8 units of Cre recombinase (New England Biolabs, Beijing, China) were mixed and incubated at 37 • C for 30 min. The reactions were stopped by heating the samples at 70 • C for 10 min to inactivate Cre, and then HEK293T cells were transfected with DNA. Fluorescent gene excision was developed by three rounds of plaque purification, as previously described [24].

In Vitro Growth Properties
One-step growth kinetics and plaque sizes of all the viruses were detected in this study, as previously described [20].

Animal Experiments
The 6-week-old female Balb/c mice (20-25 g) were purchased from the Laboratory Animal Research Center of Huazhong Agricultural University (Wuhan, China). The 5-week-old crossbred weaning piglets were purchased from the experimental farm of Huazhong Agricultural University. All experimental protocols were conducted according to the Research Ethics Committee of College of Veterinary Medicine, Huazhong Agricultural University, Hubei, China (No. 42000600035617, 17 September 2019). Experiment 1: The BALB/c mice were randomly divided into 9 groups (n = 5/group). The mice in Groups A, B, C, or D were intraperitoneally inoculated with 100 µL of different doses (10 2 ,10 3 , 10 4 , or 10 5 TCID50) of rGX∆gE/gI. The mice in Groups E, F, G, or H were intraperitoneally inoculated with 100 µL of different doses (10 2 ,10 3 , 10 4 , or 10 5 TCID50) of rGX∆TK/gE/gI. Mice in Group I were injected with PBS serving as control. Then, clinical signs were monitored daily. At 14 days post-inoculation, all surviving mice were euthanatized, and the brain and lung tissues were collected. Experiment 2: Pigs were randomly divided into three groups (n = 4/group). The piglets were seronegative for PRV, which were identified using a commercially available PRV-gB antibody detection kit (Combined Biotech Co., Ltd., Shenzhen, China). The pigs in Group A or B were intramuscularly immunized with 10 6 TCID50 rGX∆gE/gI or rGX∆TK/gE/gI. Pigs in Group C were treated with 1 mL of PBS serving as control. The pigs were boosted with the same dose at 21 days post-immunization (dpi) and intranasally challenged with 10 7 TCID50 virulent PRV GX strain at three weeks after the booster vaccination. Clinical signs were recorded daily for up to 16 days.

Flow Cytometry
The frequencies of IFN-γ-producing CD3 + CD4 + and CD3 + CD8 + T-cells from the CD3 + lymphocytes in the blood were analyzed using flow cytometry. Samples were collected at 14 days after the booster immunization. Then, sample processing and flow cytometry analysis were carried out as described in a previous report [32]. Mononuclear cells from the blood were isolated based on the previous description [33]. The mononuclear cells were stimulated in vitro with inactivated PRV (MOI = 1) for 17 h before intracellular staining, as previously described [32].

PRV-gD Specific Antibodies Measurement
The levels of PRV-gD specific antibodies in the serum were measured using an indirect enzyme-linked immunosorbent assay (ELISA). Briefly, the 96-well flat-bottomed microtiter plates were incubated with 0.5 µg per well of purified gD protein in coating buffer (pH 9.5) at 4 • C overnight, and blocked with 1% bovine serum albumin at 37 • C for 1 h. Then, the serially diluted serum samples were added into the plate and incubated for 1 h at 37 • C. Horseradish peroxidase (HRP)-conjugated goat anti-swine IgG (1:5000) (AntGene Bio Co., Ltd.,Wuhan, China) was used to cover the plate for 1 h at 37 • C. The serum PRV-gD specific antibody titers were measured at wavelength 450 nm. The antibody endpoint titer was calculated based on the highest dilution, which gave an OD450 twice that of the naïve group without dilution.

Serum Neutralisation Test
Serum neutralizing antibody titer was detected as previously described [34,35]. Briefly, the serum samples were diluted using PBS. The 96-well flat-bottomed tissue culture plates (Thermofisher, Waltham, MA, USA) were covered with the diluted serum samples and supplemented with a viral suspension with a titer of 200 TCID50 PRV GX strain in 50 µL. After incubation for 1 h at 37 • C, the plates were incubated with 50 µL of the PK-15 cell suspension for 3 days. Finally, the PRV-specific neutralizing antibody titer was analyzed and expressed as the reciprocal of the highest dilution while PK-15 cell infection was inhibited.

Virus Isolation
Rectal and nasal swabs were collected everyday post-immunization or post-challenge, and then the virus was isolated according to the previous description [18].

Hematoxylin and Eosin (HE) Staining
Brain and lung tissues were fixed in 10% neutral-buffered formalin, subjected to paraffin embedding, and then cut into 4-µm thick slices. The slices were deparaffinized, rehydrated and stained with hematoxylin and eosin. The photograph was taken under a light microscope.

Statistical Analysis
All data were presented as the mean ± standard deviation (SD) and analyzed using Graphpad Prism 6.0 software. Comparisons were conducted via one-way ANOVA, followed by Tukey's test. A p value less than 0.05 was considered as statistically significant.

Generation and Identification of gE/gI (or TK/gE/gI)-Deleted Recombinant PRVs via a CRISPR/Cas9-and Cre-lox-Based System
In order to obtain gE/gI (or TK/gE/gI)-deleted recombinant PRV (rGX∆gE/gI or rGX∆TK/gE/gI), CRISPR/Cas9 and Cre-lox-based systems were applied into producing new PRV virulent variants. The amino acid sequences of major antigens, gB and gD of PRV GX, are shown in Figure 1A,B, and they exhibited some variations and deletion compared with the Bartha strain and previous pandemic Ea strain by sequence alignment.
We then established an express vaccine development strategy using two highly efficient gene edit systems, the CRISPR/Cas9 and Cre/Lox systems. HEK293T cells were co-transfected with the PRV genome, cas9 plasmid sgRNA-gE and sgRNA-gI (or sgRNA-gE, sgRNA-gI and sgRNA-TK), fragment gIhm-loxN-CMV-mCherry-SV40polyA-loxN-gEhm (or TKhm1-loxP-CMV-GFP-SV40polyA-loxP-TKhm2 and gIhm-loxN-CMV-mCherry-SV40polyA-loxN-gEhm). Then, the cells were collected and subjected to single-cell FACS technique to purify recombinant PRV. Further, the Cre/Lox system was used to facilitate fluorescent marker genes excision. GFP (or GFP and mCherry) genes were flanked with LoxP and Lox N pairs, respectively ( Figure 1C, Figure 2A). Fluorescence detection revealed that the recombinant viruses (rGX∆gE/gI or rGX∆TK/gE/gI) expressing red or/and green fluorescence were successfully visualized ( Figure 1D, Figure 2B). Subsequently, the recombinant PRV (rGX∆gE/gI or rGX∆TK/gE/gI) were collected and infected HEK293T cells. HEK293T cells with fluorescence were analyzed using the single-cell FACS technique and plated one cell per well to a 96-well plate pre-cultured with HEK293T cells. Fluorescence detection was conducted ( Figure 1E, Figure 2C). Then, one round of plaque purification was performed to obtain the pure recombinant viruses when the wells displayed maximum fluorescence overlapping signals ( Figure 1F, Figure 2D). In addition, PCR amplification confirmed the purity of the recombinant virus. The results demonstrated that recombinant PRV (rGX∆gE/gI) showed TK amplification, whereas gE/gI amplification was negative ( Figure 1H). Similarily, TK and gE/gI gene amplification exhibited negative in recombinant PRV (rGX∆TK/gE/gI; Figure 2F). Finally, the fluorescence markers were removed in vaccine candidates due to vaccine safety concerns and regulation. Cre-treated recombinant viruses were obtained and infected to HEK293T cells. The fluorescent gene was excised through four rounds of plaque purification ( Figure 1G, Figure 2E). Further, one-step growth kinetics demonstrated that the rGX∆gE/gI and rGX∆TK/gE/gI strains propagated slightly slower than the parental strain GX ( Figure 2G). These results indicated that the gE/gI (or TK/gE/gI)-deleted recombinant virus was successfully generated using the CRISPR/Cas9-and Cre-lox-based systems.

Safety of rGX∆gE/gI and rGX∆TK/gE/gI in Mice
We then evaluated the safety of rGX∆gE/gI and rGX∆TK/gE/gI in mice. Results revealed that morbidity and mortality of mice immunized with rGX∆TK/gE/gI or PBS were not affected. However, the rGX∆gE/gI-inoculated mice displayed high morbidity and mortality (Table 2). Additionally, HE staining demonstrated that no histopathological lesions were observed in mice immunized with rGX∆TK/gE/gI or PBS. In contrast, purkinje neuron injury in the brains, and slight hemorrhages and congestion in the lungs were observed and obvious in mice immunized with the rGX∆gE/gI group ( Figure 3A,B). These findings suggest that rGX∆TK/gE/gI is safer than rGX∆gE/gI in mice.

Safety of rGXΔgE/gI and rGXΔTK/gE/gI in Mice
We then evaluated the safety of rGXΔgE/gI and rGXΔTK/gE/gI in mice. Results revealed that morbidity and mortality of mice immunized with rGXΔTK/gE/gI or PBS were not affected. However, the rGXΔgE/gI-inoculated mice displayed high morbidity and mortality (Table 2). Additionally, HE staining demonstrated that no histopathological lesions were observed in mice immunized with rGXΔTK/gE/gI or PBS. In contrast, purkinje neuron injury in the brains, and slight hemorrhages and congestion in the lungs were observed and obvious in mice immunized with the rGXΔgE/gI group ( Figure 3A,B). These findings suggest that rGXΔTK/gE/gI is safer than rGXΔgE/gI in mice.

The Production of IFN-γ-Producing CD4 + and CD8 + T-Cells
To examine the amount of IFN-γ-producing CD4 + and CD8 + T-cells, which are critical for virus elimination, flow cytometry analysis was performed. We found that rGXΔTK/gE/gI and rGXΔgE/gI significantly elevated the IFN-γ-producing CD4 + and CD8 + T-cells in pigs. However, the numbers of

The Production of IFN-γ-Producing CD4 + and CD8 + T-Cells
To examine the amount of IFN-γ-producing CD4 + and CD8 + T-cells, which are critical for virus elimination, flow cytometry analysis was performed. We found that rGX∆TK/gE/gI and rGX∆gE/gI significantly elevated the IFN-γ-producing CD4 + and CD8 + T-cells in pigs. However, the numbers of IFN-γ-producing CD4 + and CD8 + T-cells were not affected between the rGX∆TK/gE/gI group and the rGX∆gE/gI group. (Figure 4A,B). The data imply that rGX∆TK/gE/gI and rGX∆gE/gI immunizations induce equivalent PRV-specific T-cell immune responses. IFN-γ-producing CD4 + and CD8 + T-cells were not affected between the rGXΔTK/gE/gI group and the rGXΔgE/gI group. (Figure 4A,B). The data imply that rGXΔTK/gE/gI and rGXΔgE/gI immunizations induce equivalent PRV-specific T-cell immune responses.

Immunogenicity of rGXΔgE/gI and rGXΔTK/gE/gI in Pigs
In order to monitor PRV gD-specific antibody responses and NAbs against the PRV GX strains, the serum samples were collected at 0, 14, 28, 42, 56, and 63 after vaccination. ELISA analysis demonstrated that there was no difference shown in PRV gD-specific antibody from rGXΔTK/gE/gI and rGXΔgE/gI groups at 0, 14, 28, 42, 56, and 63 days post-immunization ( Figure 5A). Moreover,

Immunogenicity of rGX∆gE/gI and rGX∆TK/gE/gI in Pigs
In order to monitor PRV gD-specific antibody responses and NAbs against the PRV GX strains, the serum samples were collected at 0, 14, 28, 42, 56, and 63 after vaccination. ELISA analysis demonstrated that there was no difference shown in PRV gD-specific antibody from rGX∆TK/gE/gI and rGX∆gE/gI groups at 0, 14, 28, 42, 56, and 63 days post-immunization ( Figure 5A). Moreover, neutralizing antibody assay revealed that the levels of NAbs against the GX strain were similar in the two vaccinated groups ( Figure 5B). These results indicate that rGX∆TK/gE/gI and rGX∆gE/gI immunizations induce equivalent PRV-specific humoral immune responses.
Viruses 2020, 12, x FOR PEER REVIEW 10 of 16 neutralizing antibody assay revealed that the levels of NAbs against the GX strain were similar in the two vaccinated groups ( Figure 5B). These results indicate that rGXΔTK/gE/gI and rGXΔgE/gI immunizations induce equivalent PRV-specific humoral immune responses.

Protection of Pigs Immunized with rGXΔgE/gI and rGXΔTK/gE/gI from Virulent Challenge
In order to evaluate the protection efficacy of the two genetically deleted PRV strains against lethal GX challenge, pigs were challenged intranasally with a highly virulent PRV GX strain. No clinical signs were observed in pigs immunized with rGXΔgE/gI and rGXΔTK/gE/gI after virulent challenge. However, pigs in the PBS group displayed typical PR signs (depression, anorexia, cough, diarrhea, and systemic neurological signs) with high fever (40.5-42 °C) from 2 dpc till death ( Figure  6A). Meanwhile, the fever frequencies were the highest in the PBS group (21/25; Table 3). In addition, Figure 5. PRV-gD-specific IgG titre and neutralizing antibody titre were evaluated. (A) The indirect ELISA was used to detect gD-specific IgG titer from serum samples at 0, 14, 28, 42, 56, and 63 dpi. (B) Neutralizing antibody titer was examined in the serum of the immunized pigs. The neutralizing ability of antisera generated against PRV-GX strain was calculated and presented as the log2 of the reciprocal of the highest serum dilution when PK-15 cell infection was inhibited. n = 4. ns represents not significant.

Protection of Pigs Immunized with rGX∆gE/gI and rGX∆TK/gE/gI from Virulent Challenge
In order to evaluate the protection efficacy of the two genetically deleted PRV strains against lethal GX challenge, pigs were challenged intranasally with a highly virulent PRV GX strain. No clinical signs were observed in pigs immunized with rGX∆gE/gI and rGX∆TK/gE/gI after virulent challenge. However, pigs in the PBS group displayed typical PR signs (depression, anorexia, cough, diarrhea, and systemic neurological signs) with high fever (40.5-42 • C) from 2 dpc till death ( Figure 6A). Meanwhile, the fever frequencies were the highest in the PBS group (21/25; Table 3). In addition, all pigs died within 10 dpc in the PBS group, while pigs were survived in other groups ( Figure 6B). The challenge virus was isolated from the nasal and rectal swabs of unvaccinated pigs at 1-16 dpc. In contrast, no viral shedding was detected in pigs vaccinated with rGX∆TK/gE/gI and rGX∆gE/gI (Table 3). all pigs died within 10 dpc in the PBS group, while pigs were survived in other groups ( Figure 6B). The challenge virus was isolated from the nasal and rectal swabs of unvaccinated pigs at 1-16 dpc. In contrast, no viral shedding was detected in pigs vaccinated with rGXΔTK/gE/gI and rGXΔgE/gI (Table 3).   6. The effect of rGX∆gE/gI and rGX∆TK/gE/gI on rectal temperature and survival rates in pigs after challenge. Pigs were challenged with virulent PRV GX strain (10 7 TCID50) at 49 dpi, then (A) rectal temperature was recorded and (B) survival rates were calculated. Rectal temperature ≥40.0 • C was fever. n = 4. Groups of pigs (n = 4) were inoculated with 10 6 TCID50 of rGX-TK/gE/gI or rGX-gE/gI or PBS and then challenged with 1 mL 10 7 TCID50/100 µL of the PRV-GX strain at 49 days post-inoculation (dpi). Following challenge, fever, days to fever onset, fever frequency, and survival were recorded. Fever is defined as rectal temperature ≥40.5 • C. a , Days with fever/total days observed.
Further, after challenge, the HE staining revealed that the pigs in the PBS group suffered severe microscopic pathological lesions in multiple organs, such as obvious non-suppurative meningoencephalitis and hemorrhages in the brains, and congestion in the lungs. Moreover, the pigs in the rGX∆gE/gI-vaccinated group still had slight histopathological lesions, including meningoencephalitis, hemorrhages, and congestion in the lungs. In contrast, the pigs immunized with rGX∆TK/gE/gI did not show pathological lesions ( Figure 7A,B).
then challenged with 1 mL 10 7 TCID50/100 μL of the PRV-GX strain at 49 days post-inoculation (dpi). Following challenge, fever, days to fever onset, fever frequency, and survival were recorded. Fever is defined as rectal temperature ≥40.5 °C. a, Days with fever/total days observed Further, after challenge, the HE staining revealed that the pigs in the PBS group suffered severe microscopic pathological lesions in multiple organs, such as obvious non-suppurative meningoencephalitis and hemorrhages in the brains, and congestion in the lungs. Moreover, the pigs in the rGXΔgE/gI-vaccinated group still had slight histopathological lesions, including meningoencephalitis, hemorrhages, and congestion in the lungs. In contrast, the pigs immunized with rGXΔTK/gE/gI did not show pathological lesions ( Figure 7A,B).

Discussion
In the current study, we constructed the gE/gI-and TK/gE/gI-gene-deleted recombinant PRVs (rGXΔgE/gI and rGXΔTK/gE/gI) using CRISPR/Cas9 and Cre/Lox systems. The comparison of rGXΔTK/gE/gI and rGXΔgE/gI mediated by CRISPR/Cas9 and Cre/Lox systems was evaluated. rGXΔTK/gE/gI was demonstrated to be safer than rGXΔgE/gI in mice. The rGXΔTK/gE/gI and rGXΔgE/gI significantly promoted PRV-specific T-cell immune response and humoral immune response in swine. Further analysis revealed that rGXΔTK/gE/gI was safer compared with rGXΔgE/gI in swine.
Although there were significant efforts to control and eliminate PR, the disease has been endemic in swine in many places [36]. More effective PRV vaccines need to be explored to control the disease. Interestingly, it has been reported that piglets immunized with rPRVTJ-delgE were protected, whereas incomplete protection was provided by the Bartha-K61 vaccine [18], suggesting that rPRVTJ-delgE can update Bartha-K61 for the control of the currently epidemic PR. Notably, in our study, we isolated the PRV-GX strain and developed a gene-deleted vaccine for this variant. Several steps of single gene recombination and marker gene excision were performed to delete multiple genes to develop PRV vaccines. A PRV double gene deletion vaccine candidate was obtained, which involved

Discussion
In the current study, we constructed the gE/gI-and TK/gE/gI-gene-deleted recombinant PRVs (rGX∆gE/gI and rGX∆TK/gE/gI) using CRISPR/Cas9 and Cre/Lox systems. The comparison of rGX∆TK/gE/gI and rGX∆gE/gI mediated by CRISPR/Cas9 and Cre/Lox systems was evaluated. rGX∆TK/gE/gI was demonstrated to be safer than rGX∆gE/gI in mice. The rGX∆TK/gE/gI and rGX∆gE/gI significantly promoted PRV-specific T-cell immune response and humoral immune response in swine. Further analysis revealed that rGX∆TK/gE/gI was safer compared with rGX∆gE/gI in swine.
Although there were significant efforts to control and eliminate PR, the disease has been endemic in swine in many places [36]. More effective PRV vaccines need to be explored to control the disease. Interestingly, it has been reported that piglets immunized with rPRVTJ-delgE were protected, whereas incomplete protection was provided by the Bartha-K61 vaccine [18], suggesting that rPRVTJ-delgE can update Bartha-K61 for the control of the currently epidemic PR. Notably, in our study, we isolated the PRV-GX strain and developed a gene-deleted vaccine for this variant. Several steps of single gene recombination and marker gene excision were performed to delete multiple genes to develop PRV vaccines. A PRV double gene deletion vaccine candidate was obtained, which involved approximately twenty rounds of time-consuming plaque purification [37]. In this study, to promote the multi-gene editing efficiency in viral genomes, two highly efficient gene edit systems, the CRISPR/Cas9 system and the Cre/Lox system, were combined. Further single-cell FACS technology was used to elevate virus purification efficiency. The gE/gI/TK and gE/gI genes were deleted, and marker genes from recombinant viruses rGX∆TK/gE/gI (GFP and mCherry) and rGX∆gE/gI (mCherry) were also excised using the Cre/lox site-specific recombination system. The procedure was simple and easy to carry out, requiring only the readily harvested Cre protein [38]. Subsequently, the gE/gI-and gE/gI/TK-gene-deleted PRVs (rGX∆gE/gI or rGX∆TK/gE/gI) were developed via the CRISPR/Cas9 and Cre/Lox systems.
Previously, gE-deleted PR vaccines were found to be safe and efficacious for the control and eradication of PR [39]. The TK gene, as one of the first PRV genes, was responsible for virulence [40]. Accumulating evidence has shown that TK, gI, and gE genes are considered as major virulence determinants of PRV genome [41][42][43][44]. It has been reported that deleting the TK gene decreases virus replication in the nervous system and the ability to cause encephalitis [45,46]. In addition, the deletion of TK and gE/gI genes caused the reduction of virulence and attenuation of PRV, but did not affect the immunogenicity of PRV [18,47,48]. Interestingly, in this study, we demonstrate that the LD50 of rGX∆TK/gE/gI was higher than 10 5 TCID50 in mice, whereas the LD50 of rGX∆gE/gI in mice was (10 3.68 TCID50) less than 10 5 TCID50, suggesting that rGX∆TK/gE/gI may be safer than rGX∆gE/gI in mice. Notably, Wang et al. discovered that viral shedding was not showed in all pigs vaccinated with rPRVTJ-∆gE [18]. Consistently, this study verified that no other clinical signs associated with PRV infection and viral shedding were observed in pigs vaccinated with rGX∆TK/gE/gI and rGX∆gE/gI after challenge. However, after challenge, the pigs in rGX∆gE/gI-vaccinated group had slight histopathological lesions compared with the pigs immunized with rGX∆TK/gE/gI. We speculate that virulence of rGX∆gE/gI may be stronger than that of rGX∆TK/gE/gI, leading to the histopathological lesions in the rGX∆gE/gI-immunized group. It is important that a gene-deleted PRV can retain immunogenicity following the gene deletion [47]. In the current study, we demonstrate that the gE/gI-and gE/gI/TK-gene-deletions do not impair the immunogenicity of the virus in pigs. The PRV-specific cellular immune response and humoral immune response induced by rGX∆TK/gE/gI were consistent with those caused by rGX∆gE/gI in pigs. Taken together, these results imply that rGX∆TK/gE/gI may be a better candidate for the protection of PRV.

Conclusions
In conclusion, we combined CRISPR/Cas9 system and Cre/Lox system to generate gE/gI-and gE/gI/TK-gene-deleted variants. The pathogenicity and immunogenicity were evaluated in susceptible animals. The findings indicate that the recombinant virus protects pigs against PR, and rGX∆TK/gE/gI may be a promising vaccine vector and better control of prevalent PR.