An adenovirus serotype 2-vectored ebolavirus vaccine generates robust antibody and cell-mediated immune responses in mice and rhesus macaques

An adenovirus serotype 2-vectored ebolavirus vaccine generates robust antibody and cell-mediated... Ebolavirus vaccines based on several adenoviral vectors have been investigated in preclinical studies and clinical trials. The use of adenovirus serotype 2 as a vector for ebolavirus vaccine has not been reported. Herein, we generated rAd2- ZGP, a recombinant replication-incompetent adenovirus serotype 2 expressing codon-optimized Zaire ebolavirus glycoprotein, and evaluated its immunogenicity in mice and rhesus macaques. rAd2-ZGP induced significant antibody and cell-mediated immune responses at 2 weeks after a single immunization. The glycoprotein (GP)-specific immune responses could be further enhanced with a booster immunization. Compared to protein antigens, Zaire ebolavirus GP and Zaire ebolavirus-like particles, rAd2-ZGP could induce stronger cross-reactive antibody and cell-mediated immune responses to heterologous Sudan ebolavirus in mice and rhesus macaques. In rAd2-ZGP-immunized macaques, GP- specific CD8 T cells could secret IFN-γ and IL-2, indicating a Th1-biased response. In adenovirus serotype 5 seropositive macaques, rAd2-ZGP could induce robust antibody and cell-mediated immune responses, suggesting that the efficacy of rAd2-ZGP is not affected by pre-existing immunity to adenovirus serotype 5. These results demonstrated that rAd2-ZGP can be considered an alternative ebolavirus vaccine for use in adenovirus serotype 5 seropositive subjects or as a sequential booster vaccine after the subjects have been immunized with a recombinant adenovirus serotype 5-based vaccine. Introduction diseases with high fatality . The 2013–2016 epidemic Ebolavirus (EBOV) belongs to a genus of the Filoviridae caused by ZEBOV in West Africa has become the largest family and consists of five species, including Zaire, Sudan, outbreak recorded, with more than 28,000 people affected 1 3 Bundibugyo, Reston, and Taï Forest EBOVs . Among and a death toll of at least 12,000 . The economic and these species, Zaire EBOV (ZEBOV) and Sudan EBOV health burdens posed by EBOV highlighted the need to (SEBOV) have the highest pathogenicity in humans . develop safe and effective prophylactic vaccines. EBOV infection usually leads to severe hemorrhagic fever Recombinant adenoviruses (rAd) have been extensively explored as vaccine vectors for many pathogens, including HIV, mycobacterium tuberculosis, malaria, influenza Correspondence: Liqiang Feng (feng_liqiang@gibh.ac.cn)or 4–7 virus, and EBOV . rAd vectors expressing EBOV gly- Ling Chen (chen_ling@gibh.ac.cn) State Key Laboratory of Respiratory Disease, Guangzhou Institutes of coprotein (GP), the major surface protein mediating the Biomedicine and Health, Chinese Academy of Sciences, Guangzhou 510530, attachment and entry of EBOV , effectively protected China mice and non-human primates (NHPs) against lethal University of Chinese Academy of Sciences, Beijing 100049, China Full list of author information is available at the end of the article. © The Author(s) 2018 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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 theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license 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 license, visit http://creativecommons.org/licenses/by/4.0/. 1234567890():,; 1234567890():,; Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 2 of 12 6, 9 EBOV infection . At least three rAd-vectored vaccines, evaluated the immunogenicity of purified GP of ZEBOV including recombinant human adenovirus serotype 5 (ZGP) and virus-like particles of the ZEBOV Makona (rAd5), human adenovirus serotype 26 (rAd26), human strain (ZVLPs). adenovirus serotype 35 (rAd35), and chimpanzee adeno- 10–13 virus type 3 (ChAd3), have been investigated . Results Recently, the rAd5-vectored vaccine expressing GP of the Characterization of rAd2-ZGP, ZGP, and ZVLPs ZEBOV Makona strain (GenBank accession number We generated rAd2-ZGP, a recombinant replication- KJ660346) underwent a phase II clinical trial in Sierra incompetent adenovirus serotype 2 carrying the GP of the Leone and showed good safety and immunogenicity . contemporary ZEBOV Makona strain (GenBank acces- The rAd5-GP-induced antibody response to EBOV sion number KJ660346) (Supplementary Figure S1A). The peaked at day 28 after immunization but declined by 85% antigen components in three EBOV vaccine candidates 14, 15 during the following 6 months , suggesting that a were confirmed by SDS-PAGE followed by western blot booster immunization later might be necessary. Of par- analysis as described in the Methods (Supplementary ticular note, rAd5-GP vaccine-induced GP-specific anti- Figure S1B). We estimated that the accumulated level of body and T cell responses were weakened by the presence GP expressed by 5 × 10 vp rAd2-ZGP on cultured Vero of pre-existing anti-Ad5 antibody in a phase I trial con- cells was ~1 μg (between 0.64 and 1.28 μg) by 72 h post ducted in China and a phase II trial conducted in Sierra infection (Supplementary Figure S1C). The GP of ZEBOV Leone, which suggested that pre-existing anti-Ad5 2014 (ZGP) with a truncated mucin-like domain and a immunity might dampen the immunogenicity of rAd5- transmembrane domain (~75 kDa) was produced in Sf9 14–16 vectored EBOV vaccines . Therefore, there is a need cells using the baculovirus expression system. The ZVLPs to evaluate more EBOV vaccines based on other rAd were generated by expressing full-length GP and VP40 of serotypes for immunization in Ad5-seropositive people the ZEBOV Makona strain. Under electron microscopy, and for immunization in people who previously received these filamentous ZVLPs were ~70 nm in diameter and rAd5-vectored vaccines. 500–1000 nm in length, resembling the filovirus particles Both neutralizing antibody and cell-mediated immune in size and morphology (Supplementary Figure S1D). (CMI) responses play important roles in the protection against EBOV infection by rAd-vectored vaccines in rAd2-ZGP elicited specific antibody responses to 17–19 rodent and NHP models . A strong association has homologous ZEBOV and heterologous SEBOV in mice been observed between the titers of GP-specific neu- To evaluate whether rAd2-ZGP can elicit GP-specific tralizing antibodies and the protective efficacy of rAd5-GP antibody responses, female Balb/c mice were intra- 18–20 in rodent and NHP challenge models . It was also muscularly immunized with either a lower dosage of 5 × + 8 9 demonstrated that rAd5-GP-induced CD8 T cell 10 vp rAd2-ZGP or a higher dosage of 5 × 10 vp rAd2- responses are critical for preventing the establishment of ZGP. Mice injected with either 2 µg ZGP, 20 µg ZGP, 2 µg infection through clearing of EBOV-infected host cells in ZVLPs, or 20 µg ZVLPs were evaluated in parallel. Mice 13 8 9 NHPs . Although rAd5 appeared to be a safe and effec- injected with either 5 × 10 vp rAd2-Empty, 5 × 10 vp tive vector for an EBOV vaccine, there is a high pre- rAd2-Empty, or PBS were used as mock controls (Fig. 1a). valence of pre-existing anti-Ad5 neutralizing antibodies in The antibody response was evaluated at 3 weeks after the 16, 21, 22 8 human populations, especially in Asia and Africa . single immunization. Although 5 × 10 vp rAd2-ZGP is We therefore investigated whether recombinant human estimated to produce ~1 µg GP in cultured Vero cells, it adenovirus serotype 2 (rAd2) could be used as another generated a much higher level of GP-binding IgG anti- adenovirus vector for EBOV vaccines. Ad2 and Ad5 both bodies to autologous ZEBOV GP than both the 2 and 20 belong to human adenovirus subgroup C but are distinct µg ZGP and ZVLPs. We also assessed if the GP-specific in sero-reactivity . rAd2 has been shown to be a safe and antibody response has any cross-reactivity to heterologous efficient vector in humans in many gene therapy clinical SEBOV GP. Although ZEBOV GP shares 67% amino-acid trials . Given that pre-existing anti-Ad5 neutralizing identity with SEBOV GP (Supplementary Figure S1E), antibodies have minimal cross-neutralizing activity to rAd2-ZGP could elicit significant cross-reactive anti- Ad2 and people who are Ad5 seropositive might not be bodies to heterologous SEBOV GP (Fig. 1b). To further Ad2 seropositive , we proposed that Ad2 could be assess the quality of GP-specific antibodies, we measured exploited as another vector for an EBOV vaccine. the neutralizing antibodies using a neutralization assay In this study, we generated an rAd2 vector carrying a based on reporter lentiviruses pseudo-typed by ZEBOV codon-optimized GP gene encoding the contemporary GP or SEBOV GP. A single immunization with rAd2-ZGP ZEBOV Makona strain (rAd2-ZGP). We investigated the elicited a significantly higher level of neutralizing anti- immunogenicity of rAd2-ZGP by assessing GP-specific bodies to ZEBOV than did the ZGP and ZVLP vaccines antibody and CMI responses in mice and NHPs. We also (Fig. 1c, Supplementary Figure S2). A single immunization Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 3 of 12 Fig. 1 Antibody and cell-mediated immune responses to homologous ZEBOV and heterologous SEBOV after a single immunization in 8 9 mice. a Seven-week-old Balb/c female mice were intramuscularly immunized with 5 × 10 vp rAd2-ZGP, 5 × 10 vp rAd2-ZGP, 2 µg ZGP, 20 µg ZGP, 2 8 9 µg ZVLPs, or 20 µg ZVLPs. Mice injected with PBS, 5 × 10 vp or 5 × 10 vp rAd2-Empty were used as control groups. b Three weeks after immunization, serum samples were collected and subjected to ELISA analysis of IgG antibodies that bind to ZEBOV GP and SEBOV GP. The titers were calculated as reciprocal endpoints. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 SDs. c Three weeks after immunization, the serum samples were measured for the neutralizing activities to ZEBOV GP pseudo- typed lentivirus or SEBOV GP pseudo-typed lentivirus. Pseudo-typed lentiviruses at 100 TCID were incubated with 8 serial twofold dilutions of serum samples from each group and infected into Huh-7 cells. The neutralizing activity was measured as the decrease of luciferase expression relative to negative control sera. The IC50 was calculated by the dose–response inhibition function in GraphPad Prism 7.00. Data are presented as the mean ± SD (n = 5). d Mice immunized with higher dosage groups were sacrificed 3 weeks after immunization. Splenocytes were isolated and stimulated with a peptide pool derived from ZEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay and imaged with an ELISpot reader. Data are shown as the number of SFCs in one million splenocytes. e CD8 T cells secreting IFN-γ were determined with an ICS assay after stimulation with a peptide pool derived from ZEBOV GP. Data are shown as the mean ± SD (n = 5). f Splenocytes were stimulated with either ZEBOV GP or SEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million splenocytes. Comparisons between groups were performed by one-way ANOVA, comparisons of the neutralizing antibody titers between low- and high-dose immunizations in the rAd2-ZGP group were performed by Student’s t test, and a P-value < 0.05 was considered statistically significant. *, P < 0.05; **, P < 0.01; ***, P < 0.001 with rAd2-ZGP, but not ZGP or ZVLPs, induced apparent antibodies to heterologous SEBOV (Fig. 2b, c, Supple- cross-reactive neutralizing antibodies to heterologous mentary Figure S2F). SEBOV (Fig. 1c, Supplementary Figure S2). Both GP- specific binding antibodies and neutralizing antibodies rAd2-ZGP elicited cell-mediated immune responses to increased with a higher dosage of immunization (Fig. 1b, homologous ZEBOV and heterologous SEBOV in mice c, Supplementary Figure S2). The rAd5-vectored EBOV vaccine has been shown to We also tested a prime-boost immunization regime generate CMI responses, which also contribute to the 8 13, 19 usingeither5×10 vp rAd2-ZGP, 2 µg ZGP, or 2 µg protection against EBOV infection . We therefore ZVLPs with a booster immunization of the same vaccine examined whether rAd2-ZGP induces CMI responses. 6 weeks after the first immunization. Mice injected with Mice immunized with higher dosages (5 × 10 vp rAd2- PBS and 5 × 10 vp rAd2-Empty were used as mock ZGP, 20 µg ZGP, or 20 µg ZVLPs) were sacrificed at controls (Fig. 2a). Although all vaccines generated bind- 3 weeks after a single immunization. Mice immunized ing antibodies to ZEBOV GP after the two immuniza- with lower dosages (5 × 10 vp rAd2-ZGP, 2 µg ZGP, or 2 tions, rAd2-ZGP appeared to elicit the strongest IgG µg ZVLPs) were sacrificed at 3 weeks after the second antibodies that bind to homologous ZEBOV GP (Fig. 2b). immunization. Splenocytes were stimulated with a rAd2-ZGP stimulated the strongest neutralizing anti- ZEBOV GP peptide pool. rAd2-ZGP induced significant bodies to homologous ZEBOV, especially after the ZEBOV GP-specific IFN-γ spot forming cells (SFCs). In booster immunization (Fig. 2c, Supplementary Fig- contrast, ZVLPs only induced weak ZEBOV GP-specific ure S2C). A booster immunization of rAd2-ZGP induced IFN-γ SFCs, and ZGP induced little ZEBOV GP-specific + + cross-reactive IgG binding antibodies and neutralizing IFN-γ SFCs (Figs. 1d and 2d). We also measured IFN-γ Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 4 of 12 Fig. 2 Antibody and cell-mediated immune responses to homologous ZEBOV and heterologous SEBOV after prime-boost immunization in mice. a Seven-week-old Balb/c female mice were intramuscularly injected with 5 × 10 vp rAd2-ZGP, 2 µg ZGP, or 2 µg ZVLPs. Six weeks later, mice received a booster immunization of the same vaccine. Mice injected with PBS or 5 × 10 vp rAd2-Empty were used as control groups. b Three weeks after each immunization, serum samples were collected and subjected to ELISA analysis of IgG antibodies that bind to ZEBOV GP and SEBOV GP. The titers were calculated as reciprocal endpoints. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 SDs. c Three weeks after each immunization, serum samples were measured for neutralizing activities to the ZEBOV GP pseudo-typed lentivirus or SEBOV GP pseudo-typed lentivirus. Pseudo-typed lentiviruses at 100 TCID were incubated with 8 serial twofold dilutions of serum samples from each group and infected into Huh-7 cells. The neutralizing activity was measured as the decrease of luciferase expression relative to negative sera. The IC50 was calculated by the dose–response inhibition function in GraphPad Prism 7.00. Data are presented as the mean ± SD (n = 5). d Mice were sacrificed 3 weeks after the booster immunization. Splenocytes were isolated and stimulated with a peptide pool derived from ZEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay and imaged with an ELISpot reader. Data are shown as the number of SFCs in one million splenocytes. e CD8 T cells secreting IFN-γ were determined using an ICS assay after stimulation with a peptide pool derived from ZEBOV GP. Data are shown as the mean ± SD (n = 5). f Splenocytes were stimulated with either ZEBOV GP or SEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million splenocytes. Comparisons between groups were performed by one-way ANOVA; comparisons of the neutralizing antibody titers between prime and booster immunizations in the rAd2-ZGP group were performed by Student’s t test. A P-value < 0.05 was considered statistically significant. *, P < 0.05; **, P < 0.01; ***, P < 0.001 CD8 T cells in mice that received either one immuni- rAd2-ZGP induced robust antibody responses to zation of the higher dosage or two immunizations of the homologous ZEBOV and heterologous SEBOV in rhesus lower dosages. rAd2-ZGP induced the most robust GP- macaques + + specific IFN-γ CD8 T cell responses compared with To evaluate the antibody response elicited by rAd2-ZGP ZGP and ZVLPs (Figs. 1e and 2e, and Supplementary in NHPs, Chinese rhesus macaques were injected intra- Figure S3). muscularly with 1 × 10 vp rAd2-ZGP, 400 µg ZGP, 400 To assess whether the CMI response is cross-reactive to µg ZVLPs, or PBS. The same macaque received a booster heterologous SEBOV, mouse splenocytes were stimulated immunization of the same vaccine 4 weeks after the first with purified ZEBOV GP or SEBOV GP. Although the use immunization (Fig. 3a). rAd2-ZGP rapidly induced a high of GP protein is not as efficient as that of GP peptides for level of IgG antibodies that bind to ZEBOV GP 2 weeks measuring GP-specific IFN-γ SFCs, rAd2-ZGP appeared after one immunization, whereas ZGP and ZVLPs required to induce IFN-γ SFCs responsive to ZEBOV GP, and two immunizations to elicit comparable antibody respon- rAd2-ZGP also elicited, to a less extent, IFN-γ SFCs ses (Fig. 3b, Table 1). rAd2-ZGP generated a high titer of responsive to SEBOV GP (Figs. 1f and 2f). In contrast, neutralizing antibodies against autologous ZEBOV 2 weeks ZGP and ZVLPs induced no significant IFN-γ SFCs after one immunization (Fig. 3c, Supplementary Fig- responsive to SEBOV GP. These results suggested that ure S4A, Table 1). Although no significant enhancement in rAd2-ZGP could elicit strong CMI responses, including ZEBOV GP-binding antibodies was observed, a booster + + IFN-γ CD8 T cell responses to autologous ZEBOV. immunization with rAd2-ZGP further increased the titer of rAd2-ZGP can also elicit cross-reactive CMI responses to neutralizing antibodies against ZEBOV (Fig. 3c, Supple- heterologous SEBOV in immunized mice. mentary Figure S4B, Table 1). In contrast, ZGP and ZVLPs Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 5 of 12 Fig. 3 Antibody and cell-mediated immune responses to homologous ZEBOV and heterologous SEBOV after immunization in rhesus macaques. a Chinese rhesus macaques were divided into four groups and immunized intramuscularly with either 1 × 10 vp rAd2-ZGP, 400 µg ZGP, 400 µg ZVLPs, or PBS. After 4 weeks, macaques received a booster immunization of the same vaccine. b At 0, 2, 4, and 6 weeks after the first immunization, serum samples were collected and subjected to ELISA analysis of IgG antibodies to ZEBOV GP and SEBOV GP. The titers were calculated as reciprocal endpoints. Control serum samples were run each time the assay was performed. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 SDs. c Serum samples were measured for the neutralizing activities to ZEBOV GP pseudo-typed lentivirus or SEBOV GP pseudo-typed lentivirus. Pseudo-typed lentiviruses at 100 TCID were incubated with 8 serial twofold dilutions of serum samples from each group and infected into Huh-7 cells. The neutralizing activity was measured as the decrease of luciferase expression relative to negative sera. The IC50 was calculated by the dose–response inhibition function in GraphPad Prism 7.00. Data are presented as the mean ± SD (n = 4). d PBMCs were collected 2 weeks after the first and second immunizations and were stimulated with a peptide pool derived from ZEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million PBMCs. e PBMCs were collected at 2 weeks after each immunization. PBMCs were stimulated with either ZEBOV GP or SEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million PBMCs. f Qualitative profiles of ZEBOV GP-specific CD8 T cell + + responses in PBMCs at 2 weeks after a booster immunization. GP-specific CD8 T cells secreting IFN-γ , IL-2, and TNF-α were determined with ICS analysis. Data are presented as the mean ± SD (n = 4). Comparisons between groups were performed by one-way ANOVA; comparisons of the neutralizing antibody titers between prime and booster immunizations in the rAd2-ZGP group were performed by Student’s t test. A P-value < 0.05 was considered statistically significant. *, P < 0.05; **, P < 0.01; ***, P < 0.001 generated weaker GP-binding antibodies and neutralizing increased further after a booster immunization, but at a antibodies than did rAd2-ZGP after one immunization. much lower magnitude than to ZEBOV (Fig. 3b, c, Sup- Even after a booster immunization, the neutralizing anti- plementary Figure S4C-D, Table 1). In contrast, ZGP and bodies elicited by ZGP and ZVLPs were still lower than ZVLPs generated little neutralizing antibodies to SEBOV those by rAd2-ZGP (Fig. 3b, c, Supplementary Figure S4A- even after two immunizations (Fig. 3c, Supplementary B, Table 1). Consistent with the results in mice, rAd2-ZGP Figure S4C-D, Table 1). These results demonstrated that also induced a cross-reactive antibody response to SEBOV rAd2-ZGP can effectively and rapidly generate antibody in rhesus macaques. Binding antibodies to SEBOV GP and responses to autologous ZEBOV. rAd2-ZGP can elicit neutralizing antibodies to SEBOV were detected in maca- some cross-reactive GP-binding antibodies and neutraliz- ques 2 weeks after one immunization with rAd2-ZGP and ing antibodies to heterologous SEBOV. Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 6 of 12 Table 1 Reciprocal ELISA titers and neutralizing antibody titers 2 weeks after each immunization in rhesus macaques Group Day 14 Day 42 a b ELISA IgG titer nAb titer (IC50) (range) ELISA IgG titer nAb titer (IC50) ZEBOV SEBOV ZEBOV SEBOV ZEBOV SEBOV ZEBOV SEBOV PBS <100 <100 <10 <10 <100 <100 <10 <10 rAd2-ZGP 45,050 ± 5577 2575 ± 790 276 (236–293) 147 (72–253) 68,452 ± 7597 6100 ± 3189 485 (365–560) 278 (187–390) ZGP 3550 ± 1700 152 ± 36 67 (42–88) 56 (38–66) 22,000 ± 5530 1642 ± 300 110 (89–127) 90 (28–139) ZVLP 8300 ± 1400 1050 ± 320 64 (48–83) 47 (26–71) 60,800 ± 19,800 3000 ± 2144 193 (84–288) 83 (45–118) rAd2-ZGP 57,650 ± 26,000 3700 ± 717 241 (196–281) 114 (98–124) 80,290 ± 18,000 5675 ± 3220 390 (298–457) 180 (155–215) (Ad5 seropositive) The ELISA titers were calculated as reciprocal endpoints. Control serum samples were run each time the assay was performed. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 standard deviations The neutralizing activity was measured as the decrease of luciferase expression relative to negative sera. IC50 was calculated by dose–response inhibition function in GraphPad Prism 7.00 rAd2-ZGP elicited cell-mediated immune responses to and SEBOV GP in macaques immunized with rAd2 vector autologous ZEBOV and heterologous SEBOV in rhesus carrying an unrelated Zika virus antigen, rAd2-zika, and macaques we found no significant GP-specific CMI response in To evaluate CMI responses elicited by rAd2-ZGP in macaques immunized with rAd2-zika (Supplementary NHPs, GP-specific enzyme-linked immunospot (ELISpot) Figure S6). assay and intracellular cytokine staining (ICS) assays were Taken together, these results demonstrate that rAd2- performed using freshly isolated peripheral blood mono- ZGP could elicit significant CMI responses against auto- nuclear cells (PBMCs) from these rhesus macaques logous ZEBOV and generate cross-reactive CMI respon- (Fig. 3a). After the first immunization, macaques immu- ses to heterologous SEBOV. nized with rAd2-ZGP developed the strongest IFN-γ ELISpot response to ZEBOV GP peptides, which were rAd2-ZGP could induce antibody and cell-mediated further enhanced after a booster immunization, whereas immune responses in Ad5-seropositive rhesus macaques macaques immunized with ZGP and ZVLPs only gener- To investigate whether rAd2-ZGP is effective in indu- ated a weak IFN-γ ELISpot response after a booster cing GP-specific immune responses in subjects who have immunization (Fig. 3d). To assess whether the CMI pre-existing anti-Ad5 neutralizing antibodies, rhesus response was cross-reactive to heterologous SEBOV, macaques that had been previously immunized with rAd5 macaque PBMCs were stimulated with either autologous were immunized with 1 × 10 vp rAd2-ZGP followed by a ZEBOV GP or heterologous SEBOV GP protein. rAd2- booster immunization 4 weeks later. These Ad5- ZGP generated IFN-γ SFCs responsive to ZEBOV GP, seropositive macaques were injected with Ad5 over 1 especially after a booster immunization (Fig. 3e). rAd2- year ago and had Ad5 neutralizing antibody (mean titer ZGP also induced, but to a less extent, IFN-γ SFCs 1:2630) but little cross-neutralization to Ad2 (mean titer responsive to SEBOV GP. In contrast, ZGP and ZVLPs 1:173) before immunization with rAd2-ZGP (Supple- induced no significant cross-reactive CMI response to mentary Table S1). rAd2-ZGP-induced GP-specific anti- SEBOV GP. Given that antigen-specific CD8 T cells body and CMI responses were compared with Ad5- secreting cytokines play important roles in the clearance seronegative macaques immunized with the same 13, 18, 26 of EBOV-infected cells , we further analyzed immunization regimen. rAd2-ZGP induced similar levels ZEBOV GP-specific CD8 T cells for the production of of GP-specific IgG antibodies and neutralizing antibodies Th1 cytokines, including IFN-γ, IL-2, and TNF-α,by an to ZEBOV in Ad5-seropositive macaques as in Ad5- ICS assay using a flow cytometer. Macaques immunized seronegative macaques (Fig. 4a, b). Comparable IFN-γ with rAd2-ZGP produced ZEBOV GP-specific CD8 SFCs were detected in macaques with or without pre- T cells producing IFN-γ, IL-2, and TNF-α (Fig. 3f, Sup- existing anti-Ad5 neutralizing antibodies (Fig. 4c), sug- plementary Figure S5). In contrast, macaques immunized gesting that pre-existing anti-Ad5 immunity did not affect with ZGP and ZVLPs generated IFN-γ-, IL-2-, and TNF- the efficacy of rAd2-ZGP in inducing ZEBOV GP-specific + + α-secreting CD8 T cells at much lower levels (Fig. 3f, CMI responses. rAd2-ZGP provoked similar CD8 T cell + + Supplementary Figure S5). We also measured IFN-γ responses with a Th1 cytokine profile of IFN-γ , IL-2, and ELISpot responses to ZEBOV GP peptides, ZEBOV GP, TNF-α. Pre-existing anti-Ad5 neutralizing antibodies did Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 7 of 12 Fig. 4 Antibody and cell-mediated immune responses to ZEBOV induced by Ad2-ZGP in Ad5-seropositive rhesus macaques. Chinese rhesus macaques seropositive for Ad5 neutralizing antibodies were immunized intramuscularly with 1 × 10 vp rAd2-ZGP. After 4 weeks, these macaques received a booster immunization of the same vaccine. a Serum samples were collected and subjected to ELISA analysis of IgG antibodies that bind to ZEBOV GP. The titers were calculated as reciprocal endpoints. Control serum samples were run every time the assay was performed. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 SDs. b Serum samples were measured for the neutralizing activities to ZEBOV GP pseudo-typed lentivirus. Pseudo-typed lentiviruses at 100 TCID were incubated with 8 serial twofold dilutions of serum samples from each group and infected into Huh-7 cells. The neutralizing activity was measured as the decrease of luciferase expression relative to negative sera. The IC50 was calculated by the dose–response inhibition function in GraphPad Prism 7.00. Data are presented as the mean ± SD (n = 4). c PBMCs were collected 2 weeks after each immunization. PBMCs were stimulated with a peptide pool derived from ZEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million PBMCs. Data are presented as the mean ± SD (n = 4). d Qualitative profiles of ZEBOV GP-specific CD8 T cell responses in PBMCs 2 weeks after a booster immunization. GP-specific CD8 T cells secreting IFN-γ, IL-2, and TNF-α were determined with ICS analysis. Data are presented as the mean ± SD (n = 4) not dampen the immunogenicity of rAd2-ZGP. There- We investigated in two animal species the antibody and fore, rAd2-ZGP is similarly effective in inducing GP- CMI responses induced by rAd2-ZGP and, in parallel, by specific antibody and CMI responses in both Ad5- two protein-based EBOV vaccines, ZGP and ZVLPs. We seronegative and Ad5-seropositive subjects. found the following: (i) rAd2-ZGP generated robust GP- binding IgG antibody and neutralizing antibody responses Discussion to autologous ZEBOV (Figs. 1–3); (ii) rAd2-ZGP elicited Preventing EBOV epidemic outbreaks requires pro- robust GP-specific CMI responses with CD8 T cells phylactic vaccines that rapidly induce protective host secreting Th1 cytokines in rhesus macaques, whereas immune responses, including neutralizing antibody and ZGP and ZVLPs induced weak CMI responses to EBOV 17–19 cytotoxic lymphocyte responses . The trimeric surface (Fig. 3); (iii) rAd2-ZGP, but not ZGP or ZVLPs, induced GP of EBOV, which mediates viral entry and contains the some cross-reactive antibody and CMI responses to het- major epitopes for neutralizing antibodies, has been erologous SEBOV (Figs. 1–3); and (iv) pre-existing anti- adopted as the principal antigen for EBOV vaccines, Ad5 immunity showed no attenuation of the immuno- 27 28 including those based on purified GP , VLPs , DNA genicity of rAd2-ZGP (Fig. 4). vector and recombinant viral vectors such as recombi- Replication-incompetent adenovirus vectors showed nant adenoviral vectors and the vesicular stomatitis viral good performance in inducing antibody and T cell 9–11, 14, 30, 31 vector . These vaccine candidates, tested responses, which both play important roles in the pro- 17– individually in different laboratories, exhibited varying tection of rAd-based vaccines against EBOV infection degrees of immunogenicity and protective efficacy in . Recently, an rAd5-vectored EBOV vaccine was proved mice, guinea pigs, and NHPs . to be safe and immunogenic in clinical trials and was 14, 15 In this study, we constructed rAd2-ZGP, a recombinant approved by the Chinese FDA (NCT02326194) . replication-incompetent rAd2-vectored EBOV vaccine. However, the declining tendency of the immune response Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 8 of 12 at 6 months after rAd5-GP immunization in vaccinees response that was limited to autologous ZEBOV without suggested that repeated immunization might be needed to significant CMI responses (Figs. 1–3). Further modifica- boost and ensure protection in humans . In this study, tions of the protein antigen or its combination with we demonstrated that an EBOV vaccine based on rAd2, potent adjuvants might be considered to enhance the which has also been used extensively in human gene immunogenicity and protective efficacy of GP protein- therapy trials and has good safety records in humans, based vaccines. ZVLPs, consisting of GP and matrix protein VP40, have been shown to induce EBOV-specific could induce robust antibody and CMI responses against 28, 39 autologous EBOV in mice and rhesus macaques (Figs. 1– antibody and CMI responses in mice and NHPs . 3). These antibody and CMI responses also showed sig- Immunization with a mammalian-derived VLP conferred nificant cross-reactivity to a heterologous SEBOV protection against autologous EBOV in both small ani- (Figs. 1–3). These results suggested that rAd2-ZGP, if mals and NHPs, but much larger dosages, multiple proved to be safe and protective in further preclinical and immunizations and the presence of adjuvants were 39, 40 clinical studies, could be a potential candidate vaccine to required . In our study, ZVLPs could elicit CMI prevent EBOV infection. responses, but the responses were much weaker than Previous studies suggested that rAd-vectored EBOV those for rAd2-ZGP (Figs. 1–3), possibly due to the low vaccines based on different serotypes might exhibit dif- dosage, the limited immunization, and the absence of ferent levels of immunogenicity and therefore might differ adjuvants. Therefore, ZVLPs alone cannot elicit efficient 26, 31 in protective efficacy . An rAd5-based EBOV vaccine protective immune responses after a single immunization conferred effective protection against EBOV infection in and might not be effective in inducing cross-reactive rhesus macaques , whereas rAd26- or rAd35-based neutralizing antibodies and CMI responses. vaccines did not exhibit the same level of protection . Although the exact mechanisms of protection of an One possible explanation is that these adenovirus ser- effective EBOV vaccine remain to be clarified, it is likely otypes have different cell tropisms and might use different that neutralizing antibody and CMI responses play cell surface receptors for viral entry, which could affect important roles in controlling viral infection. In this study, the target cell preference and therefore the antigen we found that a significant proportion of ZEBOV GP- 33 + expression and presentation . We selected Ad2 because specific CD8 T cells secreting IFN-γ, IL-2, and TNF-α, it shares the same cellular receptor with Ad5 and might which have been shown to be associated with protection, thus have similar cell tropism as well as immunogenicity were generated in macaques immunized with rAd2-ZGP to rAd5-based vaccines (Fig. 3). This result highlighted the property of rAd2-ZGP . Ad2 and Ad5 belong to ade- novirus subgroup C, use the coxsackievirus and adeno- in eliciting Th1-biased T cell responses, which might virus receptors for virus attachment and use the α β or contribute to the clearance of EBOV-infected host cells. v 3 α β integrins for virus internalization . rAd2-ZGP Pre-existing anti-Ad5 immunity is a concern for the v 5 induced high levels of GP-binding and neutralizing anti- application of rAd5-vectored vaccines because it could bodies in mice and rhesus macaques, which is comparable attenuate the rAd5-mediated antigen expression and 6, 13 41 to the results in other studies using rAd5-GP . Because therefore decrease the immunogenicity . In some African this level of immune response showed complete protec- countries, ~90% of residents are seropositive for anti-Ad5 tion against EBOV challenge in the context of an rAd5- neutralizing antibodies . This high seroprevalence might based vaccine, rAd2-ZGP might have similar protective weaken the capacity of rAd5-vectored vaccines in indu- 6, 9, 13, 18 effects against EBOV infection . The antibody and cing EBOV-specific immunity, as observed in a clinical CMI responses induced by rAd2-ZGP also showed some trial conducted in this population in comparison with cross-reactivity to heterologous SEBOV, which has not other populations with relatively lower anti-Ad5 neu- 14, 15 been reported for rAd5-vectored EBOV vaccines. These tralizing antibody seroprevalence . The rAd2-ZGP features can be valuable for vaccines to be used in an described in this study, however, could potentially cir- EBOV epidemic region with multiple circulating strains . cumvent the pre-existing anti-Ad5 neutralizing anti- Importantly, the rapid generation of protective immune bodies. rAd2-ZGP induced robust GP-specific responses rendered rAd2-ZGP suitable for pre- and post- neutralizing antibody and CMI responses in macaques exposure immunization in the case of an EBOV outbreak. with pre-existing anti-Ad5 immunity (Fig. 4). The anti- There is a report that purified GP provided only partial body and CMI responses were comparable to those in protection against EBOV infection when given alone or as Ad5-seronegative macaques. Although Ad2 and Ad5 a booster vaccine with a DNA vaccine , but GP could might co-circulate in populations of developing countries, protect rodents from lethal EBOV challenge when fused their neutralizing antibodies do not cross-neutralize each with an Fc fragment or formulated as Ebola immune other. People who are seropositive for anti-Ad5 neu- 27, 38 complexes . Our study also showed that the GP tralizing antibodies are not always seropositive for anti- protein-based vaccine primarily induced an antibody Ad2 neutralizing antibodies . Therefore, rAd2-ZGP Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 9 of 12 might be considered an alternative EBOV vaccine candi- placed into the adenovirus shuttle vector pGA1 to obtain date for people with pre-existing anti-Ad5 neutralizing pGA1-ZGP. The pGA1-ZGP was linearized and subjected antibodies or could be used as a booster vaccine for to homologous recombination with the linearized Ad2 people who have been previously immunized with an backbone with deletion of the E1 and E3 genes in com- rAd5-based vaccine. petent E. coli BJ5183 cells (Invitrogen, CA, USA). The We also noted that a booster immunization with rAd2- resultant adenoviral plasmid pAd2-ZGP was linearized and transfected into HEK293 cells to rescue rAd2-ZGP. ZGP did not significantly increase ZEBOV GP-specific binding antibodies, which was similar to the findings of a To generate purified ZGP, the coding sequence corre- previous study by another laboratory . However, a booster sponding to residues 33–311 and 464–632 of GP from the immunization with rAd2-ZGP enhanced the titer of ZEBOV Makona strain (GenBank accession number neutralizing antibodies against ZEBOV (Fig. 3c), sug- KJ660346) was integrated into a baculovirus vector, with a gesting that a booster immunization can increase the gp67 signal sequence fused to the N-terminus. ZGP quality of GP-specific antibodies. The booster immuni- protein was expressed in Sf9 cells and purified as pre- zation also significantly increased the cross-reactive viously described . To generate ZVLPs, a recombinant binding antibodies to SEBOV GP (Fig. 3b). Moreover, baculovirus expressing full-length GP and VP40 from the the CMI responses to ZEBOV and SEBOV were increased ZEBOV Makona strain (GenBank accession number after a booster immunization (Fig. 3d, e). These results KJ660346) was generated and produced in Sf9 cells . suggested that although the pre-existing anti-Ad2 anti- ZVLPs were purified through a discontinuous sucrose bodies might exert some dampening effects on repeated gradient (10, 30, and 50%) at 28,000 × g for 90 min at 4 °C. immunization with rAd2-ZGP, the quality of antibody For morphology analysis, the ZVLP sample was set on a responses and the quantity of CMI responses can still be carbon-coated copper grid, stained with 1% phospho- enhanced. tungstic acid and examined on a Tecnai G2 Spirit (FEI, We generated an rAd2-vectored EBOV vaccine OR, USA) at an accelerating voltage of 120 kV. expressing GP and evaluated its immunogenicity in mice and rhesus macaques. We showed that rAd2-ZGP Western blot analysis of GP induced robust GP-specific antibody and CMI responses To analyze the rAd2-ZGP-mediated expression of GP, 6 9 to autologous EBOV, even in the presence of anti-Ad5 1×10 Vero cells were infected by rAd2-ZGP at 5 × 10 immunity. Importantly, immune responses induced by viral particles (vp). At 12, 24, 36, 48, 60, and 72 h post rAd2-ZGP showed cross-reactivity to heterologous infection, the cells were harvested. One-tenth of the rAd2- SEBOV. The rapid generation of EBOV-specific immune ZGP-infected cell lysates and GP standard samples (0.08, responses renders rAd2-ZGP suitable for pre- and post- 0.16, 0.32, 0.64, 1.28, and 2.56 µg) were subjected to SDS- exposure immunization in the case of an EBOV outbreak. PAGE followed by western blot analysis. To verify the GP rAd2-ZGP is worth further investigating in EBOV- component in the candidate vaccines, the rAd2-ZGP- challenged NHP models. infected cell lysates at 72 h post infection, purified ZGP and ZVLPs were subjected to SDS-PAGE followed by Materials and methods western blot analysis with anti-GP polyclonal antibodies Ethics statement (Sino Biological, China) and anti-VP40 antibodies All animal experiments were performed at the Animal (Abcam, MA, USA). Experimental Center of the Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences. Evaluation of immunogenicity of vaccine candidates in The experimental protocols were approved by the Insti- mice tutional Animal Care and Use Committee (IACUC# Seven-week-old female Balb/c mice were injected 2015006). The Chinese rhesus macaques that participated intramuscularly with two immunization strategies as fol- in this study were free of simian immunodeficiency virus, lows: (1) single immunization with either a lower dosage simian T lymphotropic virus type 1, and simian retrovirus. or a higher dosage of each vaccine candidate, including 8 9 rAd2-ZGP (5 × 10 vp or 5 × 10 vp per mouse in Vaccine preparation phosphate-buffered saline, PBS), ZGP (2 or 20 µg per rAd2-ZGP, a replication-incompetent adenovirus type 2 mouse with alum), ZVLPs (2 or 20 µg per mouse in PBS), 8 9 carrying the EBOV glycoprotein-encoding gene, was and rAd2 empty vector (rAd2-Empty, 5 × 10 vp or 5 × 10 constructed as previously described . ZGP, the coding vp per mouse in PBS), and (2) prime-boost immunization sequence for GP from the contemporary ZEBOV Makona with a lower dosage of each vaccine candidate, including strain (GenBank accession number KJ660346), was rAd2-ZGP (5 × 10 vp per mouse in PBS), ZGP (2 µg per codon-optimized for mammalian expression and synthe- mouse with alum), ZVLPs (2 µg per mouse in PBS), and sized (Genscript Inc., Nanjing, China). The ZGP gene was rAd2-Empty (5 × 10 vp per mouse in PBS). At 3 weeks Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 10 of 12 after the first immunization, blood samples were collected. previously described . In brief, EBOV GP pseudo-typed Mice receiving a single immunization with the higher lentiviruses were prepared by co-transfection of the HIV- - − dosages were euthanatized for isolation of splenocytes for 1 proviral vector pNL4–3.Luc.R E and expression vec- further analysis. At 6 weeks after the first immunization, tors encoding GP from the ZEBOV Makona strain mice receiving a lower dosage of vaccine candidates were (GenBank accession number KJ660346) or GP from boosted similarly as with the first immunization. At SEBOV (GenBank accession number KR063670) into 3 weeks after the booster immunization, the mice were 293T cells. Subsequently, the pseudo-typed viruses were euthanatized. Serum samples were collected, and spleno- harvested, titered, and then mixed with serially diluted cytes were isolated for further analysis. serum samples at 100 TCID per well. After incubation at 37 °C for 30 min, 90–100% confluent Huh-7 cells were Evaluation of immunogenicity of vaccine candidates in infected with the mixture for 4 h. Then, the infection rhesus macaques mixture was replaced with complete DMEM medium. Twenty 4–6-year-old Chinese rhesus macaques were After 48 h of incubation, the luciferase activity was randomly divided into five groups with both male and detected with a Luciferase Assay System (Promega, WI, female macaques in each group. Each group received two USA). The neutralizing activity was measured as the intramuscular vaccinations at 4-week intervals as follows: decrease of luciferase expression relative to negative (1) four macaques received 1 ml PBS as a mock treatment control sera. The half maximal inhibitory concentration control; (2) four macaques who had not been exposed to (IC50) values were then calculated as the reciprocal of the adenovirus received two doses of rAd2-ZGP (1 × 10 vp serum dilution at which 50% neutralization was achieved in PBS); (3) four Ad5-seropositive macaques, who were using GraphPad Prism 7.00 . immunized with rAd5 over 1 year ago, received two doses The titers of neutralizing antibodies against Ad5 and of rAd2-ZGP (1 × 10 vp in PBS); (4) four macaques Ad2 were detected with an MN assay as previously received two doses of purified ZGP (400 µg in PBS with described . Briefly, HEK293 cells were seeded into 96- alum); and (5) four macaques received two doses of well plates at 3 × 10 cells per well. After 24 h, serial ZVLPs (400 µg in PBS). Blood samples were collected, and dilutions of heat-inactivated serum samples were incu- the serum samples and peripheral blood mononuclear bated with Ad2-SEAP or Ad5-SEAP at 4 × 10 vp per well cells were isolated for further analysis as previously for 1 h at 37 °C. The mixtures were added to the 96-well described . Macaques receiving rAd2-ZGP were tested plates and incubated for 24 h at 37°C. Subsequently, supernatants were harvested, and the SEAP activity was for serum neutralizing antibodies to Ad2 and Ad5 before and after immunization of rAd2-ZGP. detected using a Phospha-Light System according to the manufacturer’s instructions (Thermo Fisher Scientific, IL, Enzyme-linked immunosorbent assay USA). The relative light units (RLUs) were recorded, and EBOV GP-specific binding antibodies were analyzed by the neutralizing titers were calculated as the reciprocal enzyme-linked immunosorbent assay (ELISA) as descri- dilutions that inhibited 50% of the RLU values. bed previously . In brief, 96-well flat-bottom plates were coated with 100 µl purified EBOV GP at 1 µg/ml in PBS at ELISpot assay 4 °C overnight. After washing and blocking, serum sam- IFN-γ ELISpot assays were performed using freshly ples were serially diluted in twofold increments and added isolated splenocytes or PBMCs. In brief, sterile 96-well in duplicate. After incubation at room temperature for 2 microtiter plates (Merck Millipore, MA, USA) were h, the plates were washed, and HRP-labeled secondary coated with IFN-γ coating antibody (R&D Systems, MN, antibody was added (Proteintech, IL, USA). After incu- USA) at 4 °C overnight. The plates were then washed and bation for another 1 h at room temperature, the plates blocked at 37 °C for 2 h. Mouse splenocytes or monkey were washed and developed with TMB/E substrate PBMCs were isolated using a density gradient medium TM 5 (Merck Millipore, MA, USA). Finally, the reaction was (Lymphoprep , Vancouver, Canada), seeded at 5 × 10 stopped, and the OD values were read. The ELISA cells per well and stimulated with a peptide pool of GP titers were calculated as a reciprocal endpoint. A negative from the ZEBOV Makona strain (GenBank accession serum control was run each time the assay was per- number KJ660346) (Chinese Peptide Company, China) at formed. A cutoff value for a positive result was calculated 2 µg/ml per peptide or with ZEBOV GP or SEBOV GP as the mean optical density (at a 1:100 dilution) for the (GenBank accession number AKB09538) (Sino Biological, negative sera plus 3 standard deviations (SDs). China) at 20 µg/ml. After incubation for 24 h, the plates were incubated with biotinylated detection antibodies Micro-neutralization assay and developed with alkaline phosphatase-conjugated The micro-neutralization (MN) assays were performed streptavidin (BD PharMingen, CA, USA) and NBT/ using lentivirus pseudo-typed with EBOV GP as BCIP reagent (Pierce, IL, USA). Finally, the spots were Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 11 of 12 counted with an ELISpot reader (Bioreader 4000, BIO- G.G. and X.X. read and revised the manuscript. All authors edited and approved the final manuscript. SYS, Germany). Conflict of interest ICS assay The authors declare that they have no conflict of interest. ICS assays were performed as previously described .In Supplementary Information accompanies this paper at (https://doi.org/ brief, freshly isolated mouse splenocytes or monkey 10.1038/s41426-018-0102-5). PBMCs were seeded into 96-well plates (2 × 10 cells per well) and incubated with an EBOV GP peptide pool (2 µg/ Received: 16 February 2018 Revised: 3 May 2018 Accepted: 6 May 2018 ml). One hour later, brefeldin A (BD Phamingen, CA, USA) was added and incubated for another 10 h. The cells were then harvested, stained with surface antibodies (CD3-Pacific blue, CD8-APC-Cy7, CD4-FITC; BD Bios- References 1. Feldmann,H.&Geisbert,T.W.Ebola haemorrhagic fever. Lancet 377,849–862 ciences, CA, USA) for 1 h, then permeabilized and stained (2011). with intracellular antibodies (IFN-γ-PE, IL-2-APC, TNF- 2. Feldmann, H., Jones, S.,Klenk,H. D.& Schnittler,H. J.Ebola virus: from dis- α-PE-Cy7; BD Biosciences, CA, USA). Finally, the cells covery to vaccine. Nat. Rev. Immunol. 3,677–685 (2003). 3. Baize, S. et al. Emergence of Zaire Ebola virus disease in Guinea. N. Engl.J.Med. were detected with an LSR Fortessa SORP instrument 371,1418–1425 (2014). (BD Biosciences, CA, USA). 4. Hoelscher, M. A. et al. Development of adenoviral-vector-based pandemic influenza vaccine against antigenically distinct human H5N1 strains in mice. Lancet 367,475–481 (2006). Statistical analysis 5. Zhang, Y. et al. Effects of the fusion design and immunization route on the Flow cytometric data were analyzed using FlowJo ver- immunogenicity of Ag85A-Mtb32 in adenoviral vectored tuberculosis vaccine. Hum. Vaccin. Immunother. 11,1803–1813 (2015). sion 7.6 (Tree Star, Inc., Ashland, USA). Statistical ana- 6. Sullivan, N. J. et al. Accelerated vaccination for Ebola virus haemorrhagic fever lyses and graphical presentations were conducted with in non-human primates. Nature 424,681–684 (2003). GraphPad Prism version 6.0 (GraphPad Software, Inc., 7. Creech, C.B.etal. Randomized,placebo-controlled trial to assess the safety and immunogenicity of an adenovirus type 35-based circumsporozoite CA, USA). Differences among groups were tested with malaria vaccine in healthy adults. Hum. Vaccin. Immunother. 9,2548–2557 one-way ANOVA. Differences in the same vaccine groups (2013). were determined by Student’s t test. Throughout the text, 8. Wang, H. et al. Ebola viral glycoprotein bound to its endosomal receptor Niemann-pick C1. Cell 164,258–268 (2016). figures, and legends, the following terminology is used to 9. Wu, S. et al. An adenovirus vaccine expressing Ebola virus variant makona show statistical significance: *, P < 0.05; **, P < 0.01; and glycoprotein is efficacious in guinea pigs and nonhuman primates. J. Infect. ***, P < 0.001. Dis. 214,S326–S332 (2016). 10. Milligan, I. D. et al. Safety and immunogenicity of novel adenovirus type 26– and modified vaccinia ankara–vectored ebola vaccines: a randomized clinical Data availability trial. JAMA 315,1610–1623 (2016). The authors declare that all relevant data are available 11. Tapia, M. D. et al. Use of ChAd3-EBO-Z Ebola virus vaccine in Malian and US from the corresponding author upon request. adults, and boosting of Malian adults with MVA-BN-Filo: a phase 1, single- blind, randomised trial, a phase 1b, open-label and double-blind, dose- escalation trial, and a nested, randomised, double-blind, placebo-controlled Acknowledgements trial. Lancet Infect. Dis. 16,31–42 (2016). We thank the staff at the Animal Center of GIBH for their excellent technical 12. Ledgerwood, J. E. et al. A replication defective recombinant Ad5 vaccine assistance. This study was supported by the National Natural Science expressing Ebola virus GP is safe and immunogenic in healthy adults. Vaccine Foundation of China (Nos. 31470892 and 91442102), the National Key Research 29,304–313 (2010). and Development Project (2016YFC1200900), the National Science and 13. Sullivan,N.J.etal. CD8+ cellular immunity mediates rAd5 vaccine protection Technology Major Project of the Ministry of Science and Technology of China against Ebola virus infection of nonhuman primates. Nat. Med. 17,1128–1131 (2015ZX09102025002), the Guangzhou Health Care and Cooperative (2011). Innovation Major Project (Nos. 201508020252 and 201803040004), and a grant 14. Zhu, F. C. et al. Safety and immunogenicity of a recombinant adenovirus type- of the CAS Youth Innovation Promotion Association to Liqiang Feng (No. 5 vector-based Ebola vaccine in healthy adults in Sierra Leone: a single-centre, 2014328). randomised, double-blind, placebo-controlled, phase 2 trial. Lancet 389, 621–628 (2016). Author details 15. Zhu, F. C. et al. Safety and immunogenicity of a novel recombinant adenovirus State Key Laboratory of Respiratory Disease, Guangzhou Institutes of type-5 vector-based Ebola vaccine in healthy adults in China: preliminary Biomedicine and Health, Chinese Academy of Sciences, Guangzhou 510530, report of a randomised, double-blind, placebo-controlled, phase 1 trial. Lancet 2 3 China. University of Chinese Academy of Sciences, Beijing 100049, China. The 385, 2272–2279 (2015). Guangzhou 8th People’s Hospital, The First Affiliated Hospital of Guangzhou 16. Xiang, Z. et al. Chimpanzee adenovirus antibodies in humans, sub-Saharan Medical University, Guangzhou 510060, China. Key Laboratory of Jilin Province Africa. Emerg. Infect. Dis. 12, 1596–1599 (2006). for Zoonosis Prevention and Control, Institute of Military Veterinary, Academy 17. Wong, G. et al. Immune parameters correlate with protection against ebola of Military Medical Sciences, Changchun 130122, China. CAS Key Laboratory of virus infection in rodents and nonhuman primates. Sci. Transl. Med. 4, 146–158 Pathogenic Microbiology and Immunology, Institute of Microbiology, Chinese (2012). Academy of Sciences, Beijing 100101, China 18. Sullivan,N.J., Martin,J.E., Graham, B.S.&Nabel, G. J. Correlatesofprotective immunity for Ebola vaccines: implications for regulatory approval by the animal rule. Nat. Rev. Microbiol. 7,393–400 (2009). Author contributions L.C., L.F., and Y. F. conceived and designed this study. Y.F., C.L., P.H., Q.W., X.Z., Y. 19. Choi, J. H. et al. A single dose respiratory recombinant adenovirus-based Z., Y.S., S.Y., C.Y., Y.F., C.W., and L.Q. performed the experiments and collected vaccine provides long-term protection for non-human primates from lethal and analyzed the data. L.C., L.F., and Y. F. wrote the manuscript. W.X., Y.L., C.S., F. Ebola infection. Mol. Pharm. 12,2712–2731 (2015). Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 12 of 12 20. Richardson, J. S. et al. Enhanced protection against Ebola virus medi- 33. Barouch,D.H.etal. Immunogenicity of recombinant adenovirus serotype 35 ated by an improved adenovirus-based vaccine. PLoS ONE 4, e5308 vaccine in the presence of pre-existing anti-Ad5 immunity. J. Immunol. 172, (2009). 6290–6297 (2004). 21. Sun, C. et al. Epidemiology of adenovirus type 5 neutralizing antibodies in 34. Parks, R., Evelegh, C. & Graham, F. Use of helper-dependent adenoviral vectors healthy people and AIDS patients in Guangzhou, southern China. Vaccine 29, of alternative serotypes permits repeat vector administration. Gene Ther. 6, 3837–3841 (2011). 1565–1573 (1999). 22. Sumida, S. M. et al. Neutralizing antibodies to adenovirus serotype 5 vaccine 35. Zhang, Y. & Bergelson, J. M. Adenovirus receptors. J. Virol. 79,12125–12131 vectors are directed primarily against the adenovirus hexon protein. J. (2005). Immunol. 174, 7179–7185 (2005). 36. Park, D. J. et al. Ebola virus epidemiology, transmission, and evolution during 23. Chroboczek, J., Bieber, F. & Jacrot, B. The sequence of the genome of ade- seven months in Sierra Leone. Cell 161,1516–1526 (2015). novirus type 5 and its comparison with the genome of adenovirus type 2. 37. Mellquist-Riemenschneider, J. L. et al. Comparison of the protective efficacy of Virology 186,280–285 (1992). DNA and baculovirus-derived protein vaccines for EBOLA virus in guinea pigs. 24. Fernandes,P., Silva, A. C.,Coroadinha, A. S. &Alves,P.M.inDavid T. Curiel Virus Res. 92,187–193 (2003). Adenoviral Vectors for Gene Therapy (Second Edition) Ch. 6 (Academic Press, 38. Phoolcharoen, W. et al. A nonreplicating subunit vaccine protects mice against San Diego, 2016). lethal Ebola virus challenge. Proc. Natl Acad. Sci. USA 108,20695–20700 (2011). 25. Li, Q. et al. Neutralizing antibodies against adenovirus type 2 in normal and 39. Warfield, K. L. et al. Ebola virus-like particle-based vaccine protects nonhuman HIV-1-infected subjects: implications for use of Ad2 vectors in vaccines. Hum. primates against lethal Ebola virus challenge. J. Infect. Dis. 196,S430–437 Vaccin. Immunother. 13,1–8 (2017). (2007). 26. Stanley, D. A. et al. Chimpanzee adenovirus vaccine generates acute and 40. Warfield, K. L. et al. Induction of humoral and CD8+ T cell responses are durable protective immunity against ebolavirus challenge. Nat. Med. 20, required for protection against lethal Ebola virus infection. J. Immunol. 175, 1126–1129 (2014). 1184–1191 (2005). 27. Konduru, K. et al. Ebola virus glycoprotein Fc fusion protein confers 41. Gilbert, S. C. Adenovirus-vectored Ebola vaccines. Expert. Rev. Vaccin. 14, protection against lethal challenge in vaccinated mice. Vaccine 29, 1347–1357 (2015). 2968–2977 (2011). 42. Xiao, L. et al. Enhancement of SIV-specific cell mediated immune responses by 28. Warfield, K. L. et al. Ebola virus-like particles protect from lethal Ebola virus co-administration of soluble PD-1 and Tim-3 as molecular adjuvants in mice. infection. Proc. Natl Acad. Sci. USA 100, 15889–15894 (2003). Hum. Vaccin. Immunother. 10,724–733 (2013). 29. Li, W. et al. Characterization of immune responses induced by Ebola virus 43. Zheng, X. et al. Treatment with hyperimmune equine immunoglobulin or glycoprotein (GP) and truncated GP isoform DNA vaccines and protec- immunoglobulin fragments completely protects rodents from Ebola virus tion against lethal Ebola virus challenge in mice. J. Infect. Dis. 212, infection. Sci. Rep. 6, 24179 (2016). S398–403 (2015). 44. Sun,C.J. etal. Mucosalpriming with areplicating-vaccinia virus-based vaccine 30. Marzi,A.etal. VSV-EBOV rapidlyprotects macaques against infection with the elicits protective immunity to simian immunodeficiency virus challenge in 2014/15 Ebola virus outbreak strain. Science 349,739–742 (2015). rhesus monkeys. J. Virol. 87, 5669–5677 (2013). 31. Geisbert, T. W. et al. Recombinant adenovirus serotype 26 (Ad26) and Ad35 45. Zhang, Q. et al. Potent neutralizing monoclonal antibodies against Ebola virus vaccine vectors bypass immunity to Ad5 and protect nonhuman primates infection. Sci. Rep. 6, 25856 (2016). against Ebola virus challenge. J. Virol. 85, 4222–4233 (2011). 46. Zheng, X. et al. Seroprevalence of neutralizing antibodies against adenovirus 32. Keshwara, R.,Johnson,R.F.& Schnell, M. J. Toward an effectiveEbola virus type 14 and 55 in healthy adults in Southern China. Emerg. Microbes Infect. 6, vaccine. Annu.Rev.Med. 68, 371–386 (2017). e43 (2017). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Emerging Microbes & Infections Springer Journals

An adenovirus serotype 2-vectored ebolavirus vaccine generates robust antibody and cell-mediated immune responses in mice and rhesus macaques

Free
12 pages

Loading next page...
 
/lp/springer_journal/an-adenovirus-serotype-2-vectored-ebolavirus-vaccine-generates-robust-icJUtJ34Xv
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Biomedicine; Biomedicine, general; Immunology; Medical Microbiology; Microbiology; Antibodies; Vaccine
eISSN
2222-1751
D.O.I.
10.1038/s41426-018-0102-5
Publisher site
See Article on Publisher Site

Abstract

Ebolavirus vaccines based on several adenoviral vectors have been investigated in preclinical studies and clinical trials. The use of adenovirus serotype 2 as a vector for ebolavirus vaccine has not been reported. Herein, we generated rAd2- ZGP, a recombinant replication-incompetent adenovirus serotype 2 expressing codon-optimized Zaire ebolavirus glycoprotein, and evaluated its immunogenicity in mice and rhesus macaques. rAd2-ZGP induced significant antibody and cell-mediated immune responses at 2 weeks after a single immunization. The glycoprotein (GP)-specific immune responses could be further enhanced with a booster immunization. Compared to protein antigens, Zaire ebolavirus GP and Zaire ebolavirus-like particles, rAd2-ZGP could induce stronger cross-reactive antibody and cell-mediated immune responses to heterologous Sudan ebolavirus in mice and rhesus macaques. In rAd2-ZGP-immunized macaques, GP- specific CD8 T cells could secret IFN-γ and IL-2, indicating a Th1-biased response. In adenovirus serotype 5 seropositive macaques, rAd2-ZGP could induce robust antibody and cell-mediated immune responses, suggesting that the efficacy of rAd2-ZGP is not affected by pre-existing immunity to adenovirus serotype 5. These results demonstrated that rAd2-ZGP can be considered an alternative ebolavirus vaccine for use in adenovirus serotype 5 seropositive subjects or as a sequential booster vaccine after the subjects have been immunized with a recombinant adenovirus serotype 5-based vaccine. Introduction diseases with high fatality . The 2013–2016 epidemic Ebolavirus (EBOV) belongs to a genus of the Filoviridae caused by ZEBOV in West Africa has become the largest family and consists of five species, including Zaire, Sudan, outbreak recorded, with more than 28,000 people affected 1 3 Bundibugyo, Reston, and Taï Forest EBOVs . Among and a death toll of at least 12,000 . The economic and these species, Zaire EBOV (ZEBOV) and Sudan EBOV health burdens posed by EBOV highlighted the need to (SEBOV) have the highest pathogenicity in humans . develop safe and effective prophylactic vaccines. EBOV infection usually leads to severe hemorrhagic fever Recombinant adenoviruses (rAd) have been extensively explored as vaccine vectors for many pathogens, including HIV, mycobacterium tuberculosis, malaria, influenza Correspondence: Liqiang Feng (feng_liqiang@gibh.ac.cn)or 4–7 virus, and EBOV . rAd vectors expressing EBOV gly- Ling Chen (chen_ling@gibh.ac.cn) State Key Laboratory of Respiratory Disease, Guangzhou Institutes of coprotein (GP), the major surface protein mediating the Biomedicine and Health, Chinese Academy of Sciences, Guangzhou 510530, attachment and entry of EBOV , effectively protected China mice and non-human primates (NHPs) against lethal University of Chinese Academy of Sciences, Beijing 100049, China Full list of author information is available at the end of the article. © The Author(s) 2018 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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 theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license 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 license, visit http://creativecommons.org/licenses/by/4.0/. 1234567890():,; 1234567890():,; Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 2 of 12 6, 9 EBOV infection . At least three rAd-vectored vaccines, evaluated the immunogenicity of purified GP of ZEBOV including recombinant human adenovirus serotype 5 (ZGP) and virus-like particles of the ZEBOV Makona (rAd5), human adenovirus serotype 26 (rAd26), human strain (ZVLPs). adenovirus serotype 35 (rAd35), and chimpanzee adeno- 10–13 virus type 3 (ChAd3), have been investigated . Results Recently, the rAd5-vectored vaccine expressing GP of the Characterization of rAd2-ZGP, ZGP, and ZVLPs ZEBOV Makona strain (GenBank accession number We generated rAd2-ZGP, a recombinant replication- KJ660346) underwent a phase II clinical trial in Sierra incompetent adenovirus serotype 2 carrying the GP of the Leone and showed good safety and immunogenicity . contemporary ZEBOV Makona strain (GenBank acces- The rAd5-GP-induced antibody response to EBOV sion number KJ660346) (Supplementary Figure S1A). The peaked at day 28 after immunization but declined by 85% antigen components in three EBOV vaccine candidates 14, 15 during the following 6 months , suggesting that a were confirmed by SDS-PAGE followed by western blot booster immunization later might be necessary. Of par- analysis as described in the Methods (Supplementary ticular note, rAd5-GP vaccine-induced GP-specific anti- Figure S1B). We estimated that the accumulated level of body and T cell responses were weakened by the presence GP expressed by 5 × 10 vp rAd2-ZGP on cultured Vero of pre-existing anti-Ad5 antibody in a phase I trial con- cells was ~1 μg (between 0.64 and 1.28 μg) by 72 h post ducted in China and a phase II trial conducted in Sierra infection (Supplementary Figure S1C). The GP of ZEBOV Leone, which suggested that pre-existing anti-Ad5 2014 (ZGP) with a truncated mucin-like domain and a immunity might dampen the immunogenicity of rAd5- transmembrane domain (~75 kDa) was produced in Sf9 14–16 vectored EBOV vaccines . Therefore, there is a need cells using the baculovirus expression system. The ZVLPs to evaluate more EBOV vaccines based on other rAd were generated by expressing full-length GP and VP40 of serotypes for immunization in Ad5-seropositive people the ZEBOV Makona strain. Under electron microscopy, and for immunization in people who previously received these filamentous ZVLPs were ~70 nm in diameter and rAd5-vectored vaccines. 500–1000 nm in length, resembling the filovirus particles Both neutralizing antibody and cell-mediated immune in size and morphology (Supplementary Figure S1D). (CMI) responses play important roles in the protection against EBOV infection by rAd-vectored vaccines in rAd2-ZGP elicited specific antibody responses to 17–19 rodent and NHP models . A strong association has homologous ZEBOV and heterologous SEBOV in mice been observed between the titers of GP-specific neu- To evaluate whether rAd2-ZGP can elicit GP-specific tralizing antibodies and the protective efficacy of rAd5-GP antibody responses, female Balb/c mice were intra- 18–20 in rodent and NHP challenge models . It was also muscularly immunized with either a lower dosage of 5 × + 8 9 demonstrated that rAd5-GP-induced CD8 T cell 10 vp rAd2-ZGP or a higher dosage of 5 × 10 vp rAd2- responses are critical for preventing the establishment of ZGP. Mice injected with either 2 µg ZGP, 20 µg ZGP, 2 µg infection through clearing of EBOV-infected host cells in ZVLPs, or 20 µg ZVLPs were evaluated in parallel. Mice 13 8 9 NHPs . Although rAd5 appeared to be a safe and effec- injected with either 5 × 10 vp rAd2-Empty, 5 × 10 vp tive vector for an EBOV vaccine, there is a high pre- rAd2-Empty, or PBS were used as mock controls (Fig. 1a). valence of pre-existing anti-Ad5 neutralizing antibodies in The antibody response was evaluated at 3 weeks after the 16, 21, 22 8 human populations, especially in Asia and Africa . single immunization. Although 5 × 10 vp rAd2-ZGP is We therefore investigated whether recombinant human estimated to produce ~1 µg GP in cultured Vero cells, it adenovirus serotype 2 (rAd2) could be used as another generated a much higher level of GP-binding IgG anti- adenovirus vector for EBOV vaccines. Ad2 and Ad5 both bodies to autologous ZEBOV GP than both the 2 and 20 belong to human adenovirus subgroup C but are distinct µg ZGP and ZVLPs. We also assessed if the GP-specific in sero-reactivity . rAd2 has been shown to be a safe and antibody response has any cross-reactivity to heterologous efficient vector in humans in many gene therapy clinical SEBOV GP. Although ZEBOV GP shares 67% amino-acid trials . Given that pre-existing anti-Ad5 neutralizing identity with SEBOV GP (Supplementary Figure S1E), antibodies have minimal cross-neutralizing activity to rAd2-ZGP could elicit significant cross-reactive anti- Ad2 and people who are Ad5 seropositive might not be bodies to heterologous SEBOV GP (Fig. 1b). To further Ad2 seropositive , we proposed that Ad2 could be assess the quality of GP-specific antibodies, we measured exploited as another vector for an EBOV vaccine. the neutralizing antibodies using a neutralization assay In this study, we generated an rAd2 vector carrying a based on reporter lentiviruses pseudo-typed by ZEBOV codon-optimized GP gene encoding the contemporary GP or SEBOV GP. A single immunization with rAd2-ZGP ZEBOV Makona strain (rAd2-ZGP). We investigated the elicited a significantly higher level of neutralizing anti- immunogenicity of rAd2-ZGP by assessing GP-specific bodies to ZEBOV than did the ZGP and ZVLP vaccines antibody and CMI responses in mice and NHPs. We also (Fig. 1c, Supplementary Figure S2). A single immunization Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 3 of 12 Fig. 1 Antibody and cell-mediated immune responses to homologous ZEBOV and heterologous SEBOV after a single immunization in 8 9 mice. a Seven-week-old Balb/c female mice were intramuscularly immunized with 5 × 10 vp rAd2-ZGP, 5 × 10 vp rAd2-ZGP, 2 µg ZGP, 20 µg ZGP, 2 8 9 µg ZVLPs, or 20 µg ZVLPs. Mice injected with PBS, 5 × 10 vp or 5 × 10 vp rAd2-Empty were used as control groups. b Three weeks after immunization, serum samples were collected and subjected to ELISA analysis of IgG antibodies that bind to ZEBOV GP and SEBOV GP. The titers were calculated as reciprocal endpoints. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 SDs. c Three weeks after immunization, the serum samples were measured for the neutralizing activities to ZEBOV GP pseudo- typed lentivirus or SEBOV GP pseudo-typed lentivirus. Pseudo-typed lentiviruses at 100 TCID were incubated with 8 serial twofold dilutions of serum samples from each group and infected into Huh-7 cells. The neutralizing activity was measured as the decrease of luciferase expression relative to negative control sera. The IC50 was calculated by the dose–response inhibition function in GraphPad Prism 7.00. Data are presented as the mean ± SD (n = 5). d Mice immunized with higher dosage groups were sacrificed 3 weeks after immunization. Splenocytes were isolated and stimulated with a peptide pool derived from ZEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay and imaged with an ELISpot reader. Data are shown as the number of SFCs in one million splenocytes. e CD8 T cells secreting IFN-γ were determined with an ICS assay after stimulation with a peptide pool derived from ZEBOV GP. Data are shown as the mean ± SD (n = 5). f Splenocytes were stimulated with either ZEBOV GP or SEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million splenocytes. Comparisons between groups were performed by one-way ANOVA, comparisons of the neutralizing antibody titers between low- and high-dose immunizations in the rAd2-ZGP group were performed by Student’s t test, and a P-value < 0.05 was considered statistically significant. *, P < 0.05; **, P < 0.01; ***, P < 0.001 with rAd2-ZGP, but not ZGP or ZVLPs, induced apparent antibodies to heterologous SEBOV (Fig. 2b, c, Supple- cross-reactive neutralizing antibodies to heterologous mentary Figure S2F). SEBOV (Fig. 1c, Supplementary Figure S2). Both GP- specific binding antibodies and neutralizing antibodies rAd2-ZGP elicited cell-mediated immune responses to increased with a higher dosage of immunization (Fig. 1b, homologous ZEBOV and heterologous SEBOV in mice c, Supplementary Figure S2). The rAd5-vectored EBOV vaccine has been shown to We also tested a prime-boost immunization regime generate CMI responses, which also contribute to the 8 13, 19 usingeither5×10 vp rAd2-ZGP, 2 µg ZGP, or 2 µg protection against EBOV infection . We therefore ZVLPs with a booster immunization of the same vaccine examined whether rAd2-ZGP induces CMI responses. 6 weeks after the first immunization. Mice injected with Mice immunized with higher dosages (5 × 10 vp rAd2- PBS and 5 × 10 vp rAd2-Empty were used as mock ZGP, 20 µg ZGP, or 20 µg ZVLPs) were sacrificed at controls (Fig. 2a). Although all vaccines generated bind- 3 weeks after a single immunization. Mice immunized ing antibodies to ZEBOV GP after the two immuniza- with lower dosages (5 × 10 vp rAd2-ZGP, 2 µg ZGP, or 2 tions, rAd2-ZGP appeared to elicit the strongest IgG µg ZVLPs) were sacrificed at 3 weeks after the second antibodies that bind to homologous ZEBOV GP (Fig. 2b). immunization. Splenocytes were stimulated with a rAd2-ZGP stimulated the strongest neutralizing anti- ZEBOV GP peptide pool. rAd2-ZGP induced significant bodies to homologous ZEBOV, especially after the ZEBOV GP-specific IFN-γ spot forming cells (SFCs). In booster immunization (Fig. 2c, Supplementary Fig- contrast, ZVLPs only induced weak ZEBOV GP-specific ure S2C). A booster immunization of rAd2-ZGP induced IFN-γ SFCs, and ZGP induced little ZEBOV GP-specific + + cross-reactive IgG binding antibodies and neutralizing IFN-γ SFCs (Figs. 1d and 2d). We also measured IFN-γ Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 4 of 12 Fig. 2 Antibody and cell-mediated immune responses to homologous ZEBOV and heterologous SEBOV after prime-boost immunization in mice. a Seven-week-old Balb/c female mice were intramuscularly injected with 5 × 10 vp rAd2-ZGP, 2 µg ZGP, or 2 µg ZVLPs. Six weeks later, mice received a booster immunization of the same vaccine. Mice injected with PBS or 5 × 10 vp rAd2-Empty were used as control groups. b Three weeks after each immunization, serum samples were collected and subjected to ELISA analysis of IgG antibodies that bind to ZEBOV GP and SEBOV GP. The titers were calculated as reciprocal endpoints. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 SDs. c Three weeks after each immunization, serum samples were measured for neutralizing activities to the ZEBOV GP pseudo-typed lentivirus or SEBOV GP pseudo-typed lentivirus. Pseudo-typed lentiviruses at 100 TCID were incubated with 8 serial twofold dilutions of serum samples from each group and infected into Huh-7 cells. The neutralizing activity was measured as the decrease of luciferase expression relative to negative sera. The IC50 was calculated by the dose–response inhibition function in GraphPad Prism 7.00. Data are presented as the mean ± SD (n = 5). d Mice were sacrificed 3 weeks after the booster immunization. Splenocytes were isolated and stimulated with a peptide pool derived from ZEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay and imaged with an ELISpot reader. Data are shown as the number of SFCs in one million splenocytes. e CD8 T cells secreting IFN-γ were determined using an ICS assay after stimulation with a peptide pool derived from ZEBOV GP. Data are shown as the mean ± SD (n = 5). f Splenocytes were stimulated with either ZEBOV GP or SEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million splenocytes. Comparisons between groups were performed by one-way ANOVA; comparisons of the neutralizing antibody titers between prime and booster immunizations in the rAd2-ZGP group were performed by Student’s t test. A P-value < 0.05 was considered statistically significant. *, P < 0.05; **, P < 0.01; ***, P < 0.001 CD8 T cells in mice that received either one immuni- rAd2-ZGP induced robust antibody responses to zation of the higher dosage or two immunizations of the homologous ZEBOV and heterologous SEBOV in rhesus lower dosages. rAd2-ZGP induced the most robust GP- macaques + + specific IFN-γ CD8 T cell responses compared with To evaluate the antibody response elicited by rAd2-ZGP ZGP and ZVLPs (Figs. 1e and 2e, and Supplementary in NHPs, Chinese rhesus macaques were injected intra- Figure S3). muscularly with 1 × 10 vp rAd2-ZGP, 400 µg ZGP, 400 To assess whether the CMI response is cross-reactive to µg ZVLPs, or PBS. The same macaque received a booster heterologous SEBOV, mouse splenocytes were stimulated immunization of the same vaccine 4 weeks after the first with purified ZEBOV GP or SEBOV GP. Although the use immunization (Fig. 3a). rAd2-ZGP rapidly induced a high of GP protein is not as efficient as that of GP peptides for level of IgG antibodies that bind to ZEBOV GP 2 weeks measuring GP-specific IFN-γ SFCs, rAd2-ZGP appeared after one immunization, whereas ZGP and ZVLPs required to induce IFN-γ SFCs responsive to ZEBOV GP, and two immunizations to elicit comparable antibody respon- rAd2-ZGP also elicited, to a less extent, IFN-γ SFCs ses (Fig. 3b, Table 1). rAd2-ZGP generated a high titer of responsive to SEBOV GP (Figs. 1f and 2f). In contrast, neutralizing antibodies against autologous ZEBOV 2 weeks ZGP and ZVLPs induced no significant IFN-γ SFCs after one immunization (Fig. 3c, Supplementary Fig- responsive to SEBOV GP. These results suggested that ure S4A, Table 1). Although no significant enhancement in rAd2-ZGP could elicit strong CMI responses, including ZEBOV GP-binding antibodies was observed, a booster + + IFN-γ CD8 T cell responses to autologous ZEBOV. immunization with rAd2-ZGP further increased the titer of rAd2-ZGP can also elicit cross-reactive CMI responses to neutralizing antibodies against ZEBOV (Fig. 3c, Supple- heterologous SEBOV in immunized mice. mentary Figure S4B, Table 1). In contrast, ZGP and ZVLPs Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 5 of 12 Fig. 3 Antibody and cell-mediated immune responses to homologous ZEBOV and heterologous SEBOV after immunization in rhesus macaques. a Chinese rhesus macaques were divided into four groups and immunized intramuscularly with either 1 × 10 vp rAd2-ZGP, 400 µg ZGP, 400 µg ZVLPs, or PBS. After 4 weeks, macaques received a booster immunization of the same vaccine. b At 0, 2, 4, and 6 weeks after the first immunization, serum samples were collected and subjected to ELISA analysis of IgG antibodies to ZEBOV GP and SEBOV GP. The titers were calculated as reciprocal endpoints. Control serum samples were run each time the assay was performed. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 SDs. c Serum samples were measured for the neutralizing activities to ZEBOV GP pseudo-typed lentivirus or SEBOV GP pseudo-typed lentivirus. Pseudo-typed lentiviruses at 100 TCID were incubated with 8 serial twofold dilutions of serum samples from each group and infected into Huh-7 cells. The neutralizing activity was measured as the decrease of luciferase expression relative to negative sera. The IC50 was calculated by the dose–response inhibition function in GraphPad Prism 7.00. Data are presented as the mean ± SD (n = 4). d PBMCs were collected 2 weeks after the first and second immunizations and were stimulated with a peptide pool derived from ZEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million PBMCs. e PBMCs were collected at 2 weeks after each immunization. PBMCs were stimulated with either ZEBOV GP or SEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million PBMCs. f Qualitative profiles of ZEBOV GP-specific CD8 T cell + + responses in PBMCs at 2 weeks after a booster immunization. GP-specific CD8 T cells secreting IFN-γ , IL-2, and TNF-α were determined with ICS analysis. Data are presented as the mean ± SD (n = 4). Comparisons between groups were performed by one-way ANOVA; comparisons of the neutralizing antibody titers between prime and booster immunizations in the rAd2-ZGP group were performed by Student’s t test. A P-value < 0.05 was considered statistically significant. *, P < 0.05; **, P < 0.01; ***, P < 0.001 generated weaker GP-binding antibodies and neutralizing increased further after a booster immunization, but at a antibodies than did rAd2-ZGP after one immunization. much lower magnitude than to ZEBOV (Fig. 3b, c, Sup- Even after a booster immunization, the neutralizing anti- plementary Figure S4C-D, Table 1). In contrast, ZGP and bodies elicited by ZGP and ZVLPs were still lower than ZVLPs generated little neutralizing antibodies to SEBOV those by rAd2-ZGP (Fig. 3b, c, Supplementary Figure S4A- even after two immunizations (Fig. 3c, Supplementary B, Table 1). Consistent with the results in mice, rAd2-ZGP Figure S4C-D, Table 1). These results demonstrated that also induced a cross-reactive antibody response to SEBOV rAd2-ZGP can effectively and rapidly generate antibody in rhesus macaques. Binding antibodies to SEBOV GP and responses to autologous ZEBOV. rAd2-ZGP can elicit neutralizing antibodies to SEBOV were detected in maca- some cross-reactive GP-binding antibodies and neutraliz- ques 2 weeks after one immunization with rAd2-ZGP and ing antibodies to heterologous SEBOV. Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 6 of 12 Table 1 Reciprocal ELISA titers and neutralizing antibody titers 2 weeks after each immunization in rhesus macaques Group Day 14 Day 42 a b ELISA IgG titer nAb titer (IC50) (range) ELISA IgG titer nAb titer (IC50) ZEBOV SEBOV ZEBOV SEBOV ZEBOV SEBOV ZEBOV SEBOV PBS <100 <100 <10 <10 <100 <100 <10 <10 rAd2-ZGP 45,050 ± 5577 2575 ± 790 276 (236–293) 147 (72–253) 68,452 ± 7597 6100 ± 3189 485 (365–560) 278 (187–390) ZGP 3550 ± 1700 152 ± 36 67 (42–88) 56 (38–66) 22,000 ± 5530 1642 ± 300 110 (89–127) 90 (28–139) ZVLP 8300 ± 1400 1050 ± 320 64 (48–83) 47 (26–71) 60,800 ± 19,800 3000 ± 2144 193 (84–288) 83 (45–118) rAd2-ZGP 57,650 ± 26,000 3700 ± 717 241 (196–281) 114 (98–124) 80,290 ± 18,000 5675 ± 3220 390 (298–457) 180 (155–215) (Ad5 seropositive) The ELISA titers were calculated as reciprocal endpoints. Control serum samples were run each time the assay was performed. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 standard deviations The neutralizing activity was measured as the decrease of luciferase expression relative to negative sera. IC50 was calculated by dose–response inhibition function in GraphPad Prism 7.00 rAd2-ZGP elicited cell-mediated immune responses to and SEBOV GP in macaques immunized with rAd2 vector autologous ZEBOV and heterologous SEBOV in rhesus carrying an unrelated Zika virus antigen, rAd2-zika, and macaques we found no significant GP-specific CMI response in To evaluate CMI responses elicited by rAd2-ZGP in macaques immunized with rAd2-zika (Supplementary NHPs, GP-specific enzyme-linked immunospot (ELISpot) Figure S6). assay and intracellular cytokine staining (ICS) assays were Taken together, these results demonstrate that rAd2- performed using freshly isolated peripheral blood mono- ZGP could elicit significant CMI responses against auto- nuclear cells (PBMCs) from these rhesus macaques logous ZEBOV and generate cross-reactive CMI respon- (Fig. 3a). After the first immunization, macaques immu- ses to heterologous SEBOV. nized with rAd2-ZGP developed the strongest IFN-γ ELISpot response to ZEBOV GP peptides, which were rAd2-ZGP could induce antibody and cell-mediated further enhanced after a booster immunization, whereas immune responses in Ad5-seropositive rhesus macaques macaques immunized with ZGP and ZVLPs only gener- To investigate whether rAd2-ZGP is effective in indu- ated a weak IFN-γ ELISpot response after a booster cing GP-specific immune responses in subjects who have immunization (Fig. 3d). To assess whether the CMI pre-existing anti-Ad5 neutralizing antibodies, rhesus response was cross-reactive to heterologous SEBOV, macaques that had been previously immunized with rAd5 macaque PBMCs were stimulated with either autologous were immunized with 1 × 10 vp rAd2-ZGP followed by a ZEBOV GP or heterologous SEBOV GP protein. rAd2- booster immunization 4 weeks later. These Ad5- ZGP generated IFN-γ SFCs responsive to ZEBOV GP, seropositive macaques were injected with Ad5 over 1 especially after a booster immunization (Fig. 3e). rAd2- year ago and had Ad5 neutralizing antibody (mean titer ZGP also induced, but to a less extent, IFN-γ SFCs 1:2630) but little cross-neutralization to Ad2 (mean titer responsive to SEBOV GP. In contrast, ZGP and ZVLPs 1:173) before immunization with rAd2-ZGP (Supple- induced no significant cross-reactive CMI response to mentary Table S1). rAd2-ZGP-induced GP-specific anti- SEBOV GP. Given that antigen-specific CD8 T cells body and CMI responses were compared with Ad5- secreting cytokines play important roles in the clearance seronegative macaques immunized with the same 13, 18, 26 of EBOV-infected cells , we further analyzed immunization regimen. rAd2-ZGP induced similar levels ZEBOV GP-specific CD8 T cells for the production of of GP-specific IgG antibodies and neutralizing antibodies Th1 cytokines, including IFN-γ, IL-2, and TNF-α,by an to ZEBOV in Ad5-seropositive macaques as in Ad5- ICS assay using a flow cytometer. Macaques immunized seronegative macaques (Fig. 4a, b). Comparable IFN-γ with rAd2-ZGP produced ZEBOV GP-specific CD8 SFCs were detected in macaques with or without pre- T cells producing IFN-γ, IL-2, and TNF-α (Fig. 3f, Sup- existing anti-Ad5 neutralizing antibodies (Fig. 4c), sug- plementary Figure S5). In contrast, macaques immunized gesting that pre-existing anti-Ad5 immunity did not affect with ZGP and ZVLPs generated IFN-γ-, IL-2-, and TNF- the efficacy of rAd2-ZGP in inducing ZEBOV GP-specific + + α-secreting CD8 T cells at much lower levels (Fig. 3f, CMI responses. rAd2-ZGP provoked similar CD8 T cell + + Supplementary Figure S5). We also measured IFN-γ responses with a Th1 cytokine profile of IFN-γ , IL-2, and ELISpot responses to ZEBOV GP peptides, ZEBOV GP, TNF-α. Pre-existing anti-Ad5 neutralizing antibodies did Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 7 of 12 Fig. 4 Antibody and cell-mediated immune responses to ZEBOV induced by Ad2-ZGP in Ad5-seropositive rhesus macaques. Chinese rhesus macaques seropositive for Ad5 neutralizing antibodies were immunized intramuscularly with 1 × 10 vp rAd2-ZGP. After 4 weeks, these macaques received a booster immunization of the same vaccine. a Serum samples were collected and subjected to ELISA analysis of IgG antibodies that bind to ZEBOV GP. The titers were calculated as reciprocal endpoints. Control serum samples were run every time the assay was performed. A cutoff value for a positive result was calculated as the mean optical density (at a 1:100 dilution) for the control serum sample plus 3 SDs. b Serum samples were measured for the neutralizing activities to ZEBOV GP pseudo-typed lentivirus. Pseudo-typed lentiviruses at 100 TCID were incubated with 8 serial twofold dilutions of serum samples from each group and infected into Huh-7 cells. The neutralizing activity was measured as the decrease of luciferase expression relative to negative sera. The IC50 was calculated by the dose–response inhibition function in GraphPad Prism 7.00. Data are presented as the mean ± SD (n = 4). c PBMCs were collected 2 weeks after each immunization. PBMCs were stimulated with a peptide pool derived from ZEBOV GP. IFN-γ SFCs were assessed with an ELISpot assay. Data are shown as the number of SFCs in one million PBMCs. Data are presented as the mean ± SD (n = 4). d Qualitative profiles of ZEBOV GP-specific CD8 T cell responses in PBMCs 2 weeks after a booster immunization. GP-specific CD8 T cells secreting IFN-γ, IL-2, and TNF-α were determined with ICS analysis. Data are presented as the mean ± SD (n = 4) not dampen the immunogenicity of rAd2-ZGP. There- We investigated in two animal species the antibody and fore, rAd2-ZGP is similarly effective in inducing GP- CMI responses induced by rAd2-ZGP and, in parallel, by specific antibody and CMI responses in both Ad5- two protein-based EBOV vaccines, ZGP and ZVLPs. We seronegative and Ad5-seropositive subjects. found the following: (i) rAd2-ZGP generated robust GP- binding IgG antibody and neutralizing antibody responses Discussion to autologous ZEBOV (Figs. 1–3); (ii) rAd2-ZGP elicited Preventing EBOV epidemic outbreaks requires pro- robust GP-specific CMI responses with CD8 T cells phylactic vaccines that rapidly induce protective host secreting Th1 cytokines in rhesus macaques, whereas immune responses, including neutralizing antibody and ZGP and ZVLPs induced weak CMI responses to EBOV 17–19 cytotoxic lymphocyte responses . The trimeric surface (Fig. 3); (iii) rAd2-ZGP, but not ZGP or ZVLPs, induced GP of EBOV, which mediates viral entry and contains the some cross-reactive antibody and CMI responses to het- major epitopes for neutralizing antibodies, has been erologous SEBOV (Figs. 1–3); and (iv) pre-existing anti- adopted as the principal antigen for EBOV vaccines, Ad5 immunity showed no attenuation of the immuno- 27 28 including those based on purified GP , VLPs , DNA genicity of rAd2-ZGP (Fig. 4). vector and recombinant viral vectors such as recombi- Replication-incompetent adenovirus vectors showed nant adenoviral vectors and the vesicular stomatitis viral good performance in inducing antibody and T cell 9–11, 14, 30, 31 vector . These vaccine candidates, tested responses, which both play important roles in the pro- 17– individually in different laboratories, exhibited varying tection of rAd-based vaccines against EBOV infection degrees of immunogenicity and protective efficacy in . Recently, an rAd5-vectored EBOV vaccine was proved mice, guinea pigs, and NHPs . to be safe and immunogenic in clinical trials and was 14, 15 In this study, we constructed rAd2-ZGP, a recombinant approved by the Chinese FDA (NCT02326194) . replication-incompetent rAd2-vectored EBOV vaccine. However, the declining tendency of the immune response Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 8 of 12 at 6 months after rAd5-GP immunization in vaccinees response that was limited to autologous ZEBOV without suggested that repeated immunization might be needed to significant CMI responses (Figs. 1–3). Further modifica- boost and ensure protection in humans . In this study, tions of the protein antigen or its combination with we demonstrated that an EBOV vaccine based on rAd2, potent adjuvants might be considered to enhance the which has also been used extensively in human gene immunogenicity and protective efficacy of GP protein- therapy trials and has good safety records in humans, based vaccines. ZVLPs, consisting of GP and matrix protein VP40, have been shown to induce EBOV-specific could induce robust antibody and CMI responses against 28, 39 autologous EBOV in mice and rhesus macaques (Figs. 1– antibody and CMI responses in mice and NHPs . 3). These antibody and CMI responses also showed sig- Immunization with a mammalian-derived VLP conferred nificant cross-reactivity to a heterologous SEBOV protection against autologous EBOV in both small ani- (Figs. 1–3). These results suggested that rAd2-ZGP, if mals and NHPs, but much larger dosages, multiple proved to be safe and protective in further preclinical and immunizations and the presence of adjuvants were 39, 40 clinical studies, could be a potential candidate vaccine to required . In our study, ZVLPs could elicit CMI prevent EBOV infection. responses, but the responses were much weaker than Previous studies suggested that rAd-vectored EBOV those for rAd2-ZGP (Figs. 1–3), possibly due to the low vaccines based on different serotypes might exhibit dif- dosage, the limited immunization, and the absence of ferent levels of immunogenicity and therefore might differ adjuvants. Therefore, ZVLPs alone cannot elicit efficient 26, 31 in protective efficacy . An rAd5-based EBOV vaccine protective immune responses after a single immunization conferred effective protection against EBOV infection in and might not be effective in inducing cross-reactive rhesus macaques , whereas rAd26- or rAd35-based neutralizing antibodies and CMI responses. vaccines did not exhibit the same level of protection . Although the exact mechanisms of protection of an One possible explanation is that these adenovirus ser- effective EBOV vaccine remain to be clarified, it is likely otypes have different cell tropisms and might use different that neutralizing antibody and CMI responses play cell surface receptors for viral entry, which could affect important roles in controlling viral infection. In this study, the target cell preference and therefore the antigen we found that a significant proportion of ZEBOV GP- 33 + expression and presentation . We selected Ad2 because specific CD8 T cells secreting IFN-γ, IL-2, and TNF-α, it shares the same cellular receptor with Ad5 and might which have been shown to be associated with protection, thus have similar cell tropism as well as immunogenicity were generated in macaques immunized with rAd2-ZGP to rAd5-based vaccines (Fig. 3). This result highlighted the property of rAd2-ZGP . Ad2 and Ad5 belong to ade- novirus subgroup C, use the coxsackievirus and adeno- in eliciting Th1-biased T cell responses, which might virus receptors for virus attachment and use the α β or contribute to the clearance of EBOV-infected host cells. v 3 α β integrins for virus internalization . rAd2-ZGP Pre-existing anti-Ad5 immunity is a concern for the v 5 induced high levels of GP-binding and neutralizing anti- application of rAd5-vectored vaccines because it could bodies in mice and rhesus macaques, which is comparable attenuate the rAd5-mediated antigen expression and 6, 13 41 to the results in other studies using rAd5-GP . Because therefore decrease the immunogenicity . In some African this level of immune response showed complete protec- countries, ~90% of residents are seropositive for anti-Ad5 tion against EBOV challenge in the context of an rAd5- neutralizing antibodies . This high seroprevalence might based vaccine, rAd2-ZGP might have similar protective weaken the capacity of rAd5-vectored vaccines in indu- 6, 9, 13, 18 effects against EBOV infection . The antibody and cing EBOV-specific immunity, as observed in a clinical CMI responses induced by rAd2-ZGP also showed some trial conducted in this population in comparison with cross-reactivity to heterologous SEBOV, which has not other populations with relatively lower anti-Ad5 neu- 14, 15 been reported for rAd5-vectored EBOV vaccines. These tralizing antibody seroprevalence . The rAd2-ZGP features can be valuable for vaccines to be used in an described in this study, however, could potentially cir- EBOV epidemic region with multiple circulating strains . cumvent the pre-existing anti-Ad5 neutralizing anti- Importantly, the rapid generation of protective immune bodies. rAd2-ZGP induced robust GP-specific responses rendered rAd2-ZGP suitable for pre- and post- neutralizing antibody and CMI responses in macaques exposure immunization in the case of an EBOV outbreak. with pre-existing anti-Ad5 immunity (Fig. 4). The anti- There is a report that purified GP provided only partial body and CMI responses were comparable to those in protection against EBOV infection when given alone or as Ad5-seronegative macaques. Although Ad2 and Ad5 a booster vaccine with a DNA vaccine , but GP could might co-circulate in populations of developing countries, protect rodents from lethal EBOV challenge when fused their neutralizing antibodies do not cross-neutralize each with an Fc fragment or formulated as Ebola immune other. People who are seropositive for anti-Ad5 neu- 27, 38 complexes . Our study also showed that the GP tralizing antibodies are not always seropositive for anti- protein-based vaccine primarily induced an antibody Ad2 neutralizing antibodies . Therefore, rAd2-ZGP Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 9 of 12 might be considered an alternative EBOV vaccine candi- placed into the adenovirus shuttle vector pGA1 to obtain date for people with pre-existing anti-Ad5 neutralizing pGA1-ZGP. The pGA1-ZGP was linearized and subjected antibodies or could be used as a booster vaccine for to homologous recombination with the linearized Ad2 people who have been previously immunized with an backbone with deletion of the E1 and E3 genes in com- rAd5-based vaccine. petent E. coli BJ5183 cells (Invitrogen, CA, USA). The We also noted that a booster immunization with rAd2- resultant adenoviral plasmid pAd2-ZGP was linearized and transfected into HEK293 cells to rescue rAd2-ZGP. ZGP did not significantly increase ZEBOV GP-specific binding antibodies, which was similar to the findings of a To generate purified ZGP, the coding sequence corre- previous study by another laboratory . However, a booster sponding to residues 33–311 and 464–632 of GP from the immunization with rAd2-ZGP enhanced the titer of ZEBOV Makona strain (GenBank accession number neutralizing antibodies against ZEBOV (Fig. 3c), sug- KJ660346) was integrated into a baculovirus vector, with a gesting that a booster immunization can increase the gp67 signal sequence fused to the N-terminus. ZGP quality of GP-specific antibodies. The booster immuni- protein was expressed in Sf9 cells and purified as pre- zation also significantly increased the cross-reactive viously described . To generate ZVLPs, a recombinant binding antibodies to SEBOV GP (Fig. 3b). Moreover, baculovirus expressing full-length GP and VP40 from the the CMI responses to ZEBOV and SEBOV were increased ZEBOV Makona strain (GenBank accession number after a booster immunization (Fig. 3d, e). These results KJ660346) was generated and produced in Sf9 cells . suggested that although the pre-existing anti-Ad2 anti- ZVLPs were purified through a discontinuous sucrose bodies might exert some dampening effects on repeated gradient (10, 30, and 50%) at 28,000 × g for 90 min at 4 °C. immunization with rAd2-ZGP, the quality of antibody For morphology analysis, the ZVLP sample was set on a responses and the quantity of CMI responses can still be carbon-coated copper grid, stained with 1% phospho- enhanced. tungstic acid and examined on a Tecnai G2 Spirit (FEI, We generated an rAd2-vectored EBOV vaccine OR, USA) at an accelerating voltage of 120 kV. expressing GP and evaluated its immunogenicity in mice and rhesus macaques. We showed that rAd2-ZGP Western blot analysis of GP induced robust GP-specific antibody and CMI responses To analyze the rAd2-ZGP-mediated expression of GP, 6 9 to autologous EBOV, even in the presence of anti-Ad5 1×10 Vero cells were infected by rAd2-ZGP at 5 × 10 immunity. Importantly, immune responses induced by viral particles (vp). At 12, 24, 36, 48, 60, and 72 h post rAd2-ZGP showed cross-reactivity to heterologous infection, the cells were harvested. One-tenth of the rAd2- SEBOV. The rapid generation of EBOV-specific immune ZGP-infected cell lysates and GP standard samples (0.08, responses renders rAd2-ZGP suitable for pre- and post- 0.16, 0.32, 0.64, 1.28, and 2.56 µg) were subjected to SDS- exposure immunization in the case of an EBOV outbreak. PAGE followed by western blot analysis. To verify the GP rAd2-ZGP is worth further investigating in EBOV- component in the candidate vaccines, the rAd2-ZGP- challenged NHP models. infected cell lysates at 72 h post infection, purified ZGP and ZVLPs were subjected to SDS-PAGE followed by Materials and methods western blot analysis with anti-GP polyclonal antibodies Ethics statement (Sino Biological, China) and anti-VP40 antibodies All animal experiments were performed at the Animal (Abcam, MA, USA). Experimental Center of the Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences. Evaluation of immunogenicity of vaccine candidates in The experimental protocols were approved by the Insti- mice tutional Animal Care and Use Committee (IACUC# Seven-week-old female Balb/c mice were injected 2015006). The Chinese rhesus macaques that participated intramuscularly with two immunization strategies as fol- in this study were free of simian immunodeficiency virus, lows: (1) single immunization with either a lower dosage simian T lymphotropic virus type 1, and simian retrovirus. or a higher dosage of each vaccine candidate, including 8 9 rAd2-ZGP (5 × 10 vp or 5 × 10 vp per mouse in Vaccine preparation phosphate-buffered saline, PBS), ZGP (2 or 20 µg per rAd2-ZGP, a replication-incompetent adenovirus type 2 mouse with alum), ZVLPs (2 or 20 µg per mouse in PBS), 8 9 carrying the EBOV glycoprotein-encoding gene, was and rAd2 empty vector (rAd2-Empty, 5 × 10 vp or 5 × 10 constructed as previously described . ZGP, the coding vp per mouse in PBS), and (2) prime-boost immunization sequence for GP from the contemporary ZEBOV Makona with a lower dosage of each vaccine candidate, including strain (GenBank accession number KJ660346), was rAd2-ZGP (5 × 10 vp per mouse in PBS), ZGP (2 µg per codon-optimized for mammalian expression and synthe- mouse with alum), ZVLPs (2 µg per mouse in PBS), and sized (Genscript Inc., Nanjing, China). The ZGP gene was rAd2-Empty (5 × 10 vp per mouse in PBS). At 3 weeks Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 10 of 12 after the first immunization, blood samples were collected. previously described . In brief, EBOV GP pseudo-typed Mice receiving a single immunization with the higher lentiviruses were prepared by co-transfection of the HIV- - − dosages were euthanatized for isolation of splenocytes for 1 proviral vector pNL4–3.Luc.R E and expression vec- further analysis. At 6 weeks after the first immunization, tors encoding GP from the ZEBOV Makona strain mice receiving a lower dosage of vaccine candidates were (GenBank accession number KJ660346) or GP from boosted similarly as with the first immunization. At SEBOV (GenBank accession number KR063670) into 3 weeks after the booster immunization, the mice were 293T cells. Subsequently, the pseudo-typed viruses were euthanatized. Serum samples were collected, and spleno- harvested, titered, and then mixed with serially diluted cytes were isolated for further analysis. serum samples at 100 TCID per well. After incubation at 37 °C for 30 min, 90–100% confluent Huh-7 cells were Evaluation of immunogenicity of vaccine candidates in infected with the mixture for 4 h. Then, the infection rhesus macaques mixture was replaced with complete DMEM medium. Twenty 4–6-year-old Chinese rhesus macaques were After 48 h of incubation, the luciferase activity was randomly divided into five groups with both male and detected with a Luciferase Assay System (Promega, WI, female macaques in each group. Each group received two USA). The neutralizing activity was measured as the intramuscular vaccinations at 4-week intervals as follows: decrease of luciferase expression relative to negative (1) four macaques received 1 ml PBS as a mock treatment control sera. The half maximal inhibitory concentration control; (2) four macaques who had not been exposed to (IC50) values were then calculated as the reciprocal of the adenovirus received two doses of rAd2-ZGP (1 × 10 vp serum dilution at which 50% neutralization was achieved in PBS); (3) four Ad5-seropositive macaques, who were using GraphPad Prism 7.00 . immunized with rAd5 over 1 year ago, received two doses The titers of neutralizing antibodies against Ad5 and of rAd2-ZGP (1 × 10 vp in PBS); (4) four macaques Ad2 were detected with an MN assay as previously received two doses of purified ZGP (400 µg in PBS with described . Briefly, HEK293 cells were seeded into 96- alum); and (5) four macaques received two doses of well plates at 3 × 10 cells per well. After 24 h, serial ZVLPs (400 µg in PBS). Blood samples were collected, and dilutions of heat-inactivated serum samples were incu- the serum samples and peripheral blood mononuclear bated with Ad2-SEAP or Ad5-SEAP at 4 × 10 vp per well cells were isolated for further analysis as previously for 1 h at 37 °C. The mixtures were added to the 96-well described . Macaques receiving rAd2-ZGP were tested plates and incubated for 24 h at 37°C. Subsequently, supernatants were harvested, and the SEAP activity was for serum neutralizing antibodies to Ad2 and Ad5 before and after immunization of rAd2-ZGP. detected using a Phospha-Light System according to the manufacturer’s instructions (Thermo Fisher Scientific, IL, Enzyme-linked immunosorbent assay USA). The relative light units (RLUs) were recorded, and EBOV GP-specific binding antibodies were analyzed by the neutralizing titers were calculated as the reciprocal enzyme-linked immunosorbent assay (ELISA) as descri- dilutions that inhibited 50% of the RLU values. bed previously . In brief, 96-well flat-bottom plates were coated with 100 µl purified EBOV GP at 1 µg/ml in PBS at ELISpot assay 4 °C overnight. After washing and blocking, serum sam- IFN-γ ELISpot assays were performed using freshly ples were serially diluted in twofold increments and added isolated splenocytes or PBMCs. In brief, sterile 96-well in duplicate. After incubation at room temperature for 2 microtiter plates (Merck Millipore, MA, USA) were h, the plates were washed, and HRP-labeled secondary coated with IFN-γ coating antibody (R&D Systems, MN, antibody was added (Proteintech, IL, USA). After incu- USA) at 4 °C overnight. The plates were then washed and bation for another 1 h at room temperature, the plates blocked at 37 °C for 2 h. Mouse splenocytes or monkey were washed and developed with TMB/E substrate PBMCs were isolated using a density gradient medium TM 5 (Merck Millipore, MA, USA). Finally, the reaction was (Lymphoprep , Vancouver, Canada), seeded at 5 × 10 stopped, and the OD values were read. The ELISA cells per well and stimulated with a peptide pool of GP titers were calculated as a reciprocal endpoint. A negative from the ZEBOV Makona strain (GenBank accession serum control was run each time the assay was per- number KJ660346) (Chinese Peptide Company, China) at formed. A cutoff value for a positive result was calculated 2 µg/ml per peptide or with ZEBOV GP or SEBOV GP as the mean optical density (at a 1:100 dilution) for the (GenBank accession number AKB09538) (Sino Biological, negative sera plus 3 standard deviations (SDs). China) at 20 µg/ml. After incubation for 24 h, the plates were incubated with biotinylated detection antibodies Micro-neutralization assay and developed with alkaline phosphatase-conjugated The micro-neutralization (MN) assays were performed streptavidin (BD PharMingen, CA, USA) and NBT/ using lentivirus pseudo-typed with EBOV GP as BCIP reagent (Pierce, IL, USA). Finally, the spots were Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 11 of 12 counted with an ELISpot reader (Bioreader 4000, BIO- G.G. and X.X. read and revised the manuscript. All authors edited and approved the final manuscript. SYS, Germany). Conflict of interest ICS assay The authors declare that they have no conflict of interest. ICS assays were performed as previously described .In Supplementary Information accompanies this paper at (https://doi.org/ brief, freshly isolated mouse splenocytes or monkey 10.1038/s41426-018-0102-5). PBMCs were seeded into 96-well plates (2 × 10 cells per well) and incubated with an EBOV GP peptide pool (2 µg/ Received: 16 February 2018 Revised: 3 May 2018 Accepted: 6 May 2018 ml). One hour later, brefeldin A (BD Phamingen, CA, USA) was added and incubated for another 10 h. The cells were then harvested, stained with surface antibodies (CD3-Pacific blue, CD8-APC-Cy7, CD4-FITC; BD Bios- References 1. Feldmann,H.&Geisbert,T.W.Ebola haemorrhagic fever. Lancet 377,849–862 ciences, CA, USA) for 1 h, then permeabilized and stained (2011). with intracellular antibodies (IFN-γ-PE, IL-2-APC, TNF- 2. Feldmann, H., Jones, S.,Klenk,H. D.& Schnittler,H. J.Ebola virus: from dis- α-PE-Cy7; BD Biosciences, CA, USA). Finally, the cells covery to vaccine. Nat. Rev. Immunol. 3,677–685 (2003). 3. Baize, S. et al. Emergence of Zaire Ebola virus disease in Guinea. N. Engl.J.Med. were detected with an LSR Fortessa SORP instrument 371,1418–1425 (2014). (BD Biosciences, CA, USA). 4. Hoelscher, M. A. et al. Development of adenoviral-vector-based pandemic influenza vaccine against antigenically distinct human H5N1 strains in mice. Lancet 367,475–481 (2006). Statistical analysis 5. Zhang, Y. et al. Effects of the fusion design and immunization route on the Flow cytometric data were analyzed using FlowJo ver- immunogenicity of Ag85A-Mtb32 in adenoviral vectored tuberculosis vaccine. Hum. Vaccin. Immunother. 11,1803–1813 (2015). sion 7.6 (Tree Star, Inc., Ashland, USA). Statistical ana- 6. Sullivan, N. J. et al. Accelerated vaccination for Ebola virus haemorrhagic fever lyses and graphical presentations were conducted with in non-human primates. Nature 424,681–684 (2003). GraphPad Prism version 6.0 (GraphPad Software, Inc., 7. Creech, C.B.etal. Randomized,placebo-controlled trial to assess the safety and immunogenicity of an adenovirus type 35-based circumsporozoite CA, USA). Differences among groups were tested with malaria vaccine in healthy adults. Hum. Vaccin. Immunother. 9,2548–2557 one-way ANOVA. Differences in the same vaccine groups (2013). were determined by Student’s t test. Throughout the text, 8. Wang, H. et al. Ebola viral glycoprotein bound to its endosomal receptor Niemann-pick C1. Cell 164,258–268 (2016). figures, and legends, the following terminology is used to 9. Wu, S. et al. An adenovirus vaccine expressing Ebola virus variant makona show statistical significance: *, P < 0.05; **, P < 0.01; and glycoprotein is efficacious in guinea pigs and nonhuman primates. J. Infect. ***, P < 0.001. Dis. 214,S326–S332 (2016). 10. Milligan, I. D. et al. Safety and immunogenicity of novel adenovirus type 26– and modified vaccinia ankara–vectored ebola vaccines: a randomized clinical Data availability trial. JAMA 315,1610–1623 (2016). The authors declare that all relevant data are available 11. Tapia, M. D. et al. Use of ChAd3-EBO-Z Ebola virus vaccine in Malian and US from the corresponding author upon request. adults, and boosting of Malian adults with MVA-BN-Filo: a phase 1, single- blind, randomised trial, a phase 1b, open-label and double-blind, dose- escalation trial, and a nested, randomised, double-blind, placebo-controlled Acknowledgements trial. Lancet Infect. Dis. 16,31–42 (2016). We thank the staff at the Animal Center of GIBH for their excellent technical 12. Ledgerwood, J. E. et al. A replication defective recombinant Ad5 vaccine assistance. This study was supported by the National Natural Science expressing Ebola virus GP is safe and immunogenic in healthy adults. Vaccine Foundation of China (Nos. 31470892 and 91442102), the National Key Research 29,304–313 (2010). and Development Project (2016YFC1200900), the National Science and 13. Sullivan,N.J.etal. CD8+ cellular immunity mediates rAd5 vaccine protection Technology Major Project of the Ministry of Science and Technology of China against Ebola virus infection of nonhuman primates. Nat. Med. 17,1128–1131 (2015ZX09102025002), the Guangzhou Health Care and Cooperative (2011). Innovation Major Project (Nos. 201508020252 and 201803040004), and a grant 14. Zhu, F. C. et al. Safety and immunogenicity of a recombinant adenovirus type- of the CAS Youth Innovation Promotion Association to Liqiang Feng (No. 5 vector-based Ebola vaccine in healthy adults in Sierra Leone: a single-centre, 2014328). randomised, double-blind, placebo-controlled, phase 2 trial. Lancet 389, 621–628 (2016). Author details 15. Zhu, F. C. et al. Safety and immunogenicity of a novel recombinant adenovirus State Key Laboratory of Respiratory Disease, Guangzhou Institutes of type-5 vector-based Ebola vaccine in healthy adults in China: preliminary Biomedicine and Health, Chinese Academy of Sciences, Guangzhou 510530, report of a randomised, double-blind, placebo-controlled, phase 1 trial. Lancet 2 3 China. University of Chinese Academy of Sciences, Beijing 100049, China. The 385, 2272–2279 (2015). Guangzhou 8th People’s Hospital, The First Affiliated Hospital of Guangzhou 16. Xiang, Z. et al. Chimpanzee adenovirus antibodies in humans, sub-Saharan Medical University, Guangzhou 510060, China. Key Laboratory of Jilin Province Africa. Emerg. Infect. Dis. 12, 1596–1599 (2006). for Zoonosis Prevention and Control, Institute of Military Veterinary, Academy 17. Wong, G. et al. Immune parameters correlate with protection against ebola of Military Medical Sciences, Changchun 130122, China. CAS Key Laboratory of virus infection in rodents and nonhuman primates. Sci. Transl. Med. 4, 146–158 Pathogenic Microbiology and Immunology, Institute of Microbiology, Chinese (2012). Academy of Sciences, Beijing 100101, China 18. Sullivan,N.J., Martin,J.E., Graham, B.S.&Nabel, G. J. Correlatesofprotective immunity for Ebola vaccines: implications for regulatory approval by the animal rule. Nat. Rev. Microbiol. 7,393–400 (2009). Author contributions L.C., L.F., and Y. F. conceived and designed this study. Y.F., C.L., P.H., Q.W., X.Z., Y. 19. Choi, J. H. et al. A single dose respiratory recombinant adenovirus-based Z., Y.S., S.Y., C.Y., Y.F., C.W., and L.Q. performed the experiments and collected vaccine provides long-term protection for non-human primates from lethal and analyzed the data. L.C., L.F., and Y. F. wrote the manuscript. W.X., Y.L., C.S., F. Ebola infection. Mol. Pharm. 12,2712–2731 (2015). Feng et al. Emerging Microbes & Infections (2018) 7:101 Page 12 of 12 20. Richardson, J. S. et al. Enhanced protection against Ebola virus medi- 33. Barouch,D.H.etal. Immunogenicity of recombinant adenovirus serotype 35 ated by an improved adenovirus-based vaccine. PLoS ONE 4, e5308 vaccine in the presence of pre-existing anti-Ad5 immunity. J. Immunol. 172, (2009). 6290–6297 (2004). 21. Sun, C. et al. Epidemiology of adenovirus type 5 neutralizing antibodies in 34. Parks, R., Evelegh, C. & Graham, F. Use of helper-dependent adenoviral vectors healthy people and AIDS patients in Guangzhou, southern China. Vaccine 29, of alternative serotypes permits repeat vector administration. Gene Ther. 6, 3837–3841 (2011). 1565–1573 (1999). 22. Sumida, S. M. et al. Neutralizing antibodies to adenovirus serotype 5 vaccine 35. Zhang, Y. & Bergelson, J. M. Adenovirus receptors. J. Virol. 79,12125–12131 vectors are directed primarily against the adenovirus hexon protein. J. (2005). Immunol. 174, 7179–7185 (2005). 36. Park, D. J. et al. Ebola virus epidemiology, transmission, and evolution during 23. Chroboczek, J., Bieber, F. & Jacrot, B. The sequence of the genome of ade- seven months in Sierra Leone. Cell 161,1516–1526 (2015). novirus type 5 and its comparison with the genome of adenovirus type 2. 37. Mellquist-Riemenschneider, J. L. et al. Comparison of the protective efficacy of Virology 186,280–285 (1992). DNA and baculovirus-derived protein vaccines for EBOLA virus in guinea pigs. 24. Fernandes,P., Silva, A. C.,Coroadinha, A. S. &Alves,P.M.inDavid T. Curiel Virus Res. 92,187–193 (2003). Adenoviral Vectors for Gene Therapy (Second Edition) Ch. 6 (Academic Press, 38. Phoolcharoen, W. et al. A nonreplicating subunit vaccine protects mice against San Diego, 2016). lethal Ebola virus challenge. Proc. Natl Acad. Sci. USA 108,20695–20700 (2011). 25. Li, Q. et al. Neutralizing antibodies against adenovirus type 2 in normal and 39. Warfield, K. L. et al. Ebola virus-like particle-based vaccine protects nonhuman HIV-1-infected subjects: implications for use of Ad2 vectors in vaccines. Hum. primates against lethal Ebola virus challenge. J. Infect. Dis. 196,S430–437 Vaccin. Immunother. 13,1–8 (2017). (2007). 26. Stanley, D. A. et al. Chimpanzee adenovirus vaccine generates acute and 40. Warfield, K. L. et al. Induction of humoral and CD8+ T cell responses are durable protective immunity against ebolavirus challenge. Nat. Med. 20, required for protection against lethal Ebola virus infection. J. Immunol. 175, 1126–1129 (2014). 1184–1191 (2005). 27. Konduru, K. et al. Ebola virus glycoprotein Fc fusion protein confers 41. Gilbert, S. C. Adenovirus-vectored Ebola vaccines. Expert. Rev. Vaccin. 14, protection against lethal challenge in vaccinated mice. Vaccine 29, 1347–1357 (2015). 2968–2977 (2011). 42. Xiao, L. et al. Enhancement of SIV-specific cell mediated immune responses by 28. Warfield, K. L. et al. Ebola virus-like particles protect from lethal Ebola virus co-administration of soluble PD-1 and Tim-3 as molecular adjuvants in mice. infection. Proc. Natl Acad. Sci. USA 100, 15889–15894 (2003). Hum. Vaccin. Immunother. 10,724–733 (2013). 29. Li, W. et al. Characterization of immune responses induced by Ebola virus 43. Zheng, X. et al. Treatment with hyperimmune equine immunoglobulin or glycoprotein (GP) and truncated GP isoform DNA vaccines and protec- immunoglobulin fragments completely protects rodents from Ebola virus tion against lethal Ebola virus challenge in mice. J. Infect. Dis. 212, infection. Sci. Rep. 6, 24179 (2016). S398–403 (2015). 44. Sun,C.J. etal. Mucosalpriming with areplicating-vaccinia virus-based vaccine 30. Marzi,A.etal. VSV-EBOV rapidlyprotects macaques against infection with the elicits protective immunity to simian immunodeficiency virus challenge in 2014/15 Ebola virus outbreak strain. Science 349,739–742 (2015). rhesus monkeys. J. Virol. 87, 5669–5677 (2013). 31. Geisbert, T. W. et al. Recombinant adenovirus serotype 26 (Ad26) and Ad35 45. Zhang, Q. et al. Potent neutralizing monoclonal antibodies against Ebola virus vaccine vectors bypass immunity to Ad5 and protect nonhuman primates infection. Sci. Rep. 6, 25856 (2016). against Ebola virus challenge. J. Virol. 85, 4222–4233 (2011). 46. Zheng, X. et al. Seroprevalence of neutralizing antibodies against adenovirus 32. Keshwara, R.,Johnson,R.F.& Schnell, M. J. Toward an effectiveEbola virus type 14 and 55 in healthy adults in Southern China. Emerg. Microbes Infect. 6, vaccine. Annu.Rev.Med. 68, 371–386 (2017). e43 (2017).

Journal

Emerging Microbes & InfectionsSpringer Journals

Published: Jun 6, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off