Genetically engineered bacterial‐like particles induced specific cellular and humoral immunity as effective tick‐borne encephalitis virus vaccine

Tick‐borne encephalitis (TBE) is a natural focal disease with fatal encephalitis induced by tick‐borne encephalitis virus (TBEV), seriously threatening human and public health. Protection of TBE depends on vaccination with inactivated vaccine, which requires high cost and multiple immunizations. Here, we construct genetically engineered bacterial‐like particles (BLPs) as an effective TBEV vaccine with simplified immunizations and improved immune efficacy. The TBEV BLPs involve the combination of the gram‐positive enhancer matrix from Lactococcus lactis, and TBEV envelope (E) protein expressed by genetically engineered recombinant baculovirus. The prepared TBEV BLPs can effectively stimulate the activation of dendritic cells to present the TBEV E proteins to T and B cells, leading to strong and durable cellular and humoral immune responses in mice. Surprisingly, the serum levels of specific IgG antibodies in mice remain about 106 at 6 months after the secondary immunization. Overall, the TBEV BLPs can be used as a potent vaccine candidate, laying the foundation for developing novel TBEV genetically engineered vaccines.


INTRODUCTION
Tick-borne encephalitis (TBE), also known as forest encephalitis, is a natural focal disease caused by tick-borne encephalitis virus (TBEV) with predominantly neurological lesions that can cause fatal encephalitis with long-term sequelae in humans and animals, posing a severe threat to human and public health. [1] Recently, TBE has taken on new epidemiological characteristics with global climate change, since the distribution of the virus-transmitting vector ticks expanding. Therefore, threats to humanity caused by TBEV are on the rise, manifested by renewed outbreaks and epidemics in old epidemic foci, the continued emergence of new epidemic foci and even the emergence of new virus subtypes. [2,3] Until now, vaccination is essential and the only effective means for preventing and controlling TBE, as there is no specific therapeutic drug for the disease. [4] # Mengyao Zhang and Hongli Jin contributed equally to this work.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Currently, almost all approved TBEV vaccines are inactivated, simple to make, and easy to transport. Still, large immune dose, short immune protection period, and single immunization route led to high cost and multiple immunizations. [4] Alternatively, live attenuated vaccines chimeric with flaviviruses provide an elevated level of immune protection. However, the pathogenicity is not sure to disappear completely, carrying the risk of virulence reversion, [5][6][7] and a certain level of neurotoxicity persists. [8][9][10] There is an urgent need to develop a TBE vaccine with favorable safety and immunogenicity profiles.
Subunit vaccines process the pathogen's major protective antigen and stimulate the body to produce antibodies. This led to novel vaccine research due to the reduced side effects, high safety profile, and minimal cost. [11] Based on the findings of other members of the Flavivirus, the TBEV envelope (E) protein contains virus-specific epitopes recognized by neutralizing antibodies and is expected to play a role as an antigen for subunit vaccines. The vast majority of F I G U R E 1 Structure of tick-borne encephalitis virus (TBEV) bacterial-like particles (BLPs) and the schematic diagram of TBEV BLPs exerting immune effects neutralizing antibody responses to flaviviruses are directed against complex antigenic sites, and such antigenic sites rely on the interaction of E protein oligomers on the surface of viral particles, while the free E protein is unable to form these antigenic sites. [12] It was demonstrated that the recombinant TBEV E protein was weakly immunogenic on its own; [13] Cimica et al. [14] displayed a large E-protein domain of Zika virus on virus-like particles and discovered that the vaccine candidate had no significant immunogenicity. Consequently, it was the difficulty of the construction to display the E protein as an oligomer onto the surface of the subunit vaccine and generate good immunogenicity.
Recently, bacterium-like particles (BLPs), a nonrecombinant Lactococcus lactis-based antigen delivery platform that anchors exogenous proteins to the surface of the grampositive enhancer matrix (GEM) composed of cell wall peptidoglycan backbone, have attracted our interest due to their excellent biosafety, high density of antigen presentation and high dendritic cells (DC) stimulation capacity. [15] Lactococcus lactis (L. lactis) is a harmless gram-positive bacteria, widely used in the food industry and granted generally recognised as safe status by the US Food and Drug Administration, the excellent biosafety of which makes it an ideal candidate for use in vaccines. As a conserved microbe-associated molecular pattern of gram-positive bacteria, peptidoglycan is recognized by innate immune cells, and the triggered signaling cascade causes the production of pro-inflammatory cytokines and activated intrinsic immune cells capable of forming adaptive B-cell and Tcell-mediated immune responses, [16] all of which have great potential in vaccine development. Vaccines prepared using BLPs induce a robust local and systemic immune response, and this immune response is protective against infection by specific pathogens, including viruses, bacteria, and parasites. [17][18][19][20][21][22] Herein, we constructed a TBEV BLPs vaccine ( Figure 1) by displaying the TBEV E protein onto the surface of GEM via the anchoring region PA3 (3 represents the number of repeat motifs LysM) of peptidoglycan hydrolase. [23] The BLPs exhibited good dispersibility and high protein capacity, which are simple and inexpensive to prepare. The resultant TBEV BLPs significantly increased the immunogenicity of TBEV E protein, and the enriched TBEV E protein was more readily captured by DC in mice and activated B cells with the assistance of T cells to generate a specific humoral immune response. Meanwhile, the TBEV BLPs vaccine prepared based on the surface display system of GEM can effectively stimulate DC maturation, promote T-cell differentiation, and generate a specific cellular immune response in mice. Overall, the TBEV BLPs vaccine can effectively stimulate the production of specific humoral and cellular immune responses in mice. As a result, it can be used as e TBEV vaccine candidates and hold great promise for developing novel TBEV vaccines.

Preparation and characterization of TBEV BLPs
Three essential components are involved in constructing the TBEV BLPs: a GEM display platform, an anchor protein PA3, and the antigenic protein TBEV E ( Figure 1). First, GEM is a gram-positive bacteria enhanced matrix sourcing from food-grade L. lactic. The intra-and original extracellular macromolecules such as proteins, nucleic acids, and lipoteichoic acids were removed by thermal acid treatment, acquiring the hollow particles of the cell wall peptidoglycan backbone as GEM. Second, the anchor protein PA3 is the C-terminal structural domain (anchoring region) of the peptidoglycan hydrolase AcmA expressed by L. lactis itself, acting as an anchor in the GEM surface and can specifically bind to peptidoglycan with non-covalent bonds (van der Waals forces and hydrogen bonds). [24] Last but important, the TBEV E protein with a deletion of the stem-transmembrane domains was selected as the antigen, which contains virusspecific epitopes recognized by neutralizing antibodies. To enhance protein expression, the transmembrane domain of the JEV capsid (C) protein was inserted as the signal peptide, which was more conducive to the expression and secretion of flavivirus proteins. [25][26][27][28][29] Meanwhile, the PA3 protein was bound to the antigenic protein by a segment of the linker. The constructed TBEV-E-PA3 sequence is shown in Figure 1.
To obtain highly and stably expressed TBEV-E-PA3 protein, we integrated the constructed TBEV-E-PA3 gene into a bacmid containing the baculovirus genome. The resultant recombinant bacmid was transfected into Sf9 cells for TBEV-E-PA3 expression. As shown in Figure S1, Sf9 cells transfected with recombinant bacmid exhibited obvious deformation, including expanded, more rounded, and heavily shed, suggesting that recombinant bacmid could infect Sf9 cells to generate recombinant baculovirus (rBV). To confirm the expression of TBEV-E-PA3 protein in Sf9 cells infected with the rescued rBV, the indirect immunofluorescence assay (IFA) and western blot (WB) was conducted ( Figure 2A). A noticeable green fluorescence can be detected in infected cells while non-infected cells showed only red fluorescence, indicating that the rBV infected cells could successfully express the TBEV E protein. Meanwhile, the TBEV-E-PA3 protein could secretorily express in suspension-cultured Sf9 cells by analyzing different components of Sf9 cells through WB assay ( Figure 2B).
The GEM was prepared according to the previous method and displayed a spherical structure with a smooth surface and negligible contents ( Figure 2C). To obtain TBEV BLPs, the prepared GEM particles were shake-bound to the supernatant of suspension-cultured Sf9 cells infected with rBV at room temperature, which contained the secretory expressed TBEV-E-PA3 protein. According to the transmission electron microscopy (TEM) images, TBEV BLPs were successfully prepared with a quantity of flocculent materials on the surface of GEM ( Figure 2C). To further demonstrate that the flocculent materials contain antigenic protein TBEV E, GEM particles and TBEV BLPs were evaluated by IFA. Compared to GEM particles, TBEV BLPs appeared as bright green particles under fluorescence microscopy ( Figure 2D). Meanwhile, TBEV BLPs showed a specific band of TBEV-E-PA3 protein at 70 kDa by WB assay ( Figure 2E). In summary, results indicated that the fusion protein TBEV-E-PA3 were successfully anchored on the GEM particles, and TBEV BLPs were successfully prepared.
To examine the loading capacity of GEM-bound fusion protein TBEV-E-PA3, 0.5 U GEM particles (1 U = 2.5 × 10 9 GEM particles) were bound to different volumes of rBV culture supernatant for sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis ( Figure 2D,E). The highest amount of protein bound to GEM particles was found when the rBV culture supernatant equaled 10 mL. Quantitative protein analysis ( Figure S2) yielded a maximum binding of 40.32 μg/U of TBEV-E-PA3 to GEM particles, which is approximately equal to that previously reported for the protein loading of the Alzheimer's disease Aβ epitope vaccine based on BLPs, [30] indicating that the constructed TBEV BLPs have the conditions to function as vaccines.

TBEV BLPs elicit a specific antibody response in mice
Humoral immunity plays an essential role in the prevention of flavivirus infection, and virus-neutralizing antibody-mediated humoral immunity is the primary protective mechanism,  7). The data are expressed as the means ± SDs from each group. ***p < 0.001; ****p < 0.0001 making eliciting a robust antibody response a key objective in the development of flavivirus vaccines. [31] General laboratories are not qualified to detect TBEV-neutralizing antibodies, since TBEV necessitates operation in a biosafety level 3 laboratory or higher. Specific antibodies against TBEV E-D III dominate the humoral immune response in mice, [32,33] and we measuerd the levels of IgG antibodies specific for TBEV E-D III to evaluate the effect of humoral immunity produced by the immunized mice. TBEV BLPs supplemented with Poly (I:C) & Montanide ISA 201VG complex adjuvant were used as immunogens to immunize BALB/c mice according to a prime-boost immunization scheme ( Figure 3A). Serum samples were collected at weeks 1 (1 week after the first immunization), 3, 4 (1 week after the second immunization), and 6, 8, 10, 12, 16, 20, 24, and 28 after immunization, respectively. The levels of specific IgG antibodies in the sera of mice after immunization were analyzed by indirect enzyme-linked immunosorbent assay (ELISA) using recombinant TBEV E-D III protein as the coated antigen ( Figure 3B). The serum IgG antibodies in mice increased significantly to 1:10 6.9 (mean value 1:10 6.3 ) after the second immunization. Although the serum TBEV IgG antibodies in mice began to decrease slowly at week 8, it was maintained at about 1:10 6 6 months after the second immunization. Overall, TBEV BLPs stimulated mice to produce robust and persistent humoral immunity.
Based on the strong humoral immunity generated by TBEV BLPs, we were intrigued to assess the levels of IgG subtypes in the sera of TBEV BLPs immunized mice. [34] Th2-type immune response in mice is related to IgG1, whereas Th1-type is associated with IgG2a. Th1-type immune response is a cellular immune response mediated by Th-cells against intracellular pathogens, while Th2-type immune response is a humoral immune response mediated by Th-cells against targeting extracellular pathogens. [35,36] The levels of IgG1 and IgG2a subtype antibodies in the sera of mice at week 8 after immunization were analyzed by indirect ELISA. The results indicated that the ratios of positive/negative (P/N) of OD 450 in mice in the experimental (BLPs-Ad) group were more significant than 2.1 and were much higher than that in the Control and Adjuvant groups ( Figure 3C), suggesting that mice produced both Th1type cellular and Th2-type humoral immune responses after immunization with TBEV BLPs. The ratios of IgG2a/IgG1 in the serum of the BLPs-Ad group were less than 1, much lower than the Control and Adjuvant groups ( Figure 3D), indicating that the immune response stimulated by TBEV BLPs in mice was more biased towards Th2-type humoral immune response.

TBEV BLPs stimulate cellular immunity in mice
Cellular immunity is the most effective defence system to clear intracellular viruses. [35,36] To examine the cellular immune response induced by TBEV BLPs in mice, we stimulated mouse splenocytes with TBEV-specific antigen in vitro after the second immunization and evaluated the cellular immune response by detecting their proliferation and secretion. The Cell Counting Kit-8 (CCK-8) assay examined the proliferation of mouse splenocytes, and the results showed that the splenocyte proliferation index was significantly higher in the BLPs-Ad group than in the Control The Th2-Type cytokine levels in splenocyte culture supernatants were measured by MSD multifactor assay. The data are expressed as the means ± SDs from each group. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 and Adjuvant groups ( Figure 4A), suggesting that the spleen could produce strong cellular immune responses. Enzymelinked immunospot assay (ELISpot) detected a significant increase in the number of interferon (IFN)-γ and interleukin (IL)-4 secreting splenocytes in mice, among which the number of IL-4 secreting splenocytes was more significantly different from Control and Adjuvant groups ( Figure 4B). Since the Th1-type immune response mainly produced IFNγ, while the Th2-type immune response produced cytokines such as IL-4, IL-5, and IL-10. Thus, TBEV BLPs induced Th1-type cellular and Th2-type humoral immune responses simultaneously in mice, and the Th2-type humoral immune response showed a higher level. The finding was consistent with the IgG subtype antibody level assay results. The Meso Scale Discovery (MSD) multifactor assay measured the secretion levels of cytokines in mouse splenocytes under TBEV-specific antigen stimulation and found that the Th1-type cytokines (IL-12p70 and tumour necrosis factor (TNF)-α) and the Th2-type cytokines (IL-4, IL-5, IL-6, IL-10, and IL-1β) were significantly higher than those in the Control and Adjuvant groups ( Figure 4C). In summary, TBEV BLPs could stimulate the cellular immune response in mice, with a Th2-biased response.

TBEV BLPs stimulate recruitment/activation of immune cells in mice
To further elucidate how a nanoparticle vaccine is recognized and processed by the host immune system, the activation of DC cells (as in Figure 5), T cells (as in Figure  S3), and B cells (as in Figure 6) were analyzed by flow cytometry in the inguinal lymph nodes (the lymph nodes directly downstream of the immunisation site) 1 week after the first immunization. [37][38][39][40] The proportions of CD11c + CD80 + , CD11c + MHC I + , and CD11c + MHC II + double positive cells were significantly higher than those in the Control and Adjuvant groups ( Figure 5), indicating that TBEV BLPs immunized mice can be captured by inguinal lymph nodes, stimulate DC activation, process the antigen, and then present it to lymphocytes. DC showed MHC class I and MHC class II molecules to T cells, which were The data are expressed as the means ± SDs from each group. **p < 0.01; ***p < 0.001; ****p < 0.0001 activated by synergistic stimulation with CD80. However, the proportions of CD4 + CD69 + and CD8 + CD69 + doublepositive cells in Figure S3 was not significantly different from the Control and Adjuvant groups, indicating that early activation of T cells was not detected. This phenomenon may be due to the low amount of antigen: when the initial stages of immunization, DC have just started to secrete CD80, MHC I, and MHC II class molecules, which have not yet caused the activation of T cells. Although B cell response to most antigens requires T cell assistance, the proportions of CD19 + CD40 + and CD19 + CD69 + doublepositive cells detected in the absence of T cell activation was significantly higher than in the Control and Adjuvant groups ( Figure 6). It indicates that B cells were activated, and this T-cell-independent B cell activation stimulated mainly the production of low-affinity IgM antibodies, which precisely explains the weak levels of TBEV-specific IgG antibodies in the serum of mice after the first immunization ( Figure 3B). According to the findings, TBEV BLPs stimulate DC and B cells activation after the first immunization, which facilitates antigen presentation and immune response generation.
Antibody production for most antigens requires the involvement of T cells, and Th cells induce B cell proliferation, differentiation, antibody production, class switch, and affinity maturation of antibodies. T cell activation was not detected after the first immunization with TBEV BLPs ( Figure S3). However, the proportions of CD4 + CD69 + and CD8 + CD69 + double-positive cells in mouse spleen cells stimulated by TBEV-specific antigen were significantly higher than in Control and Adjuvant groups after the second immunization as detected by flow cytometry (Figure 7), indicating that splenic T cells were recruited/activated by TBEV-specific antigen stimulation in mice after secondary immunization. At that point, the mice had developed a systemic immune response. The elevated level of TBEV-specific IgG antibodies detected in the serum of the mice after the second immunization also indicated that the activation of B cells in the mice after the second immunization is done with the involvement of T cells ( Figure 3B).

2.5
Secondary immunization with TBEV BLPs stimulates a long-term specific immune response in mice Memory T cells have a long survival time in vivo and can effectively mediate protective immune responses, which are classified into central memory T cells (T CM ) and effector memory T cells (T EM ). [41] Central memory T cells, mainly present in the CD4 + cell subpopulation, migrate in response to antigen stimulation, secrete IL-2 and CD62L in large numbers, divide and proliferate rapidly to replenish effector T cells in surrounding organs, and secrete IFN-γ and IL-4 in large numbers after differentiation into effector cells, which play an essential role in secondary responses. Effective F I G U R E 8 Memory T cells analysis. Flow cytometry analysis of mouse splenocytes taken at day 7 (week 4) after the 2 nd immunization (n = 3). The proportion of T CM (B) or T EM (D) among CD4 + T or CD8 + T cells was evaluated by the presence of CD44 and CD62L surface markers. [43] (A and C) Representative flow cytometric plots of lymphocytes from each group. (B and D) Percentages of the indicated lymphocytes. The data are expressed as the means ± SDs from each group. ****p < 0.0001 memory T cells have low IL-2 and CD62L secretion and weak proliferative capacity, and T EM in an activated state is similar to effector T cells, secreting IFN-γ and IL-4. [42,43] We assayed the productions of memory T cells in the CD4 + cell subpopulation and CD8 + cell subpopulation of mouse splenocytes in response to TBEV-specific antigen stimulation by flow cytometry after secondary immunization. The proportion of CD44 + CD62L + and CD44 + CD62L − cells in the CD4 + cell subpopulation and the proportion of CD44 + CD62L + and CD44 + CD62L − cells in the CD8 + cell subpopulation were found to be significantly different compared to Control and Adjuvant groups ( Figure 7B). It indicates that a large number of T CM (CD44 + CD62L + ) and T EM (CD44 + CD62L − ) were activated by TBEV-specific antigens in the splenic CD4 + cell subpopulation and CD8 + cell subpopulation of the immunized mice compared to the Control and Adjuvant groups. At this time, significant proliferation of splenocytes ( Figure 4A) and recruitment/activation of T cells (Figure 8) occurred in mice. This series of responses confirmed that memory T cells could rapidly proliferate to produce effector T cells upon TBEV-specific antigen stimulation. Thus, TBEV BLPs only require secondary immunization to stimulate the production of memory T cells in mice, which can provide effective long-term protection to the body.

CONCLUSIONS
For this study, TBEV BLPs were prepared using GEM particles, which displayed the TBEV E protein at high density and effectively stimulated the activation of DC to present the TBEV E protein to T and B cells. TBEV BLPs stimulate stronger cellular and humoral immunity in mice; the immunization schedule is simpler, involving only two immunizations to produce elevated levels of IgG antibodies for a longer duration, and also stimulate specific immune memory in mice. Above all, the TBEV BLPs vaccine is an effective TBEV vaccine candidate.

C O N F L I C T O F I N T E R E S T
The authors declare no conflict of interest.

D ATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.