key: cord-0274439-6u67mte7 authors: Bhatia, Bharti; Furuyama, Wakako; Hoenen, Thomas; Feldmann, Heinz; Marzi, Andrea title: Ebola virus glycoprotein domains associated with protective efficacy date: 2021-05-22 journal: bioRxiv DOI: 10.1101/2021.05.22.445257 sha: c26d63400558e58486900eeeb4dacb7550603c4f doc_id: 274439 cord_uid: 6u67mte7 Ebola virus (EBOV) is the cause of sporadic outbreaks of human hemorrhagic disease in Africa, and the best-characterized virus in the filovirus family. The West Africa epidemic accelerated the clinical development of vaccines and therapeutics leading to licensure of vaccines and antibody-based therapeutics for human use in recent years. The most widely used vaccine is based on vesicular stomatitis virus (VSV) expressing the EBOV glycoprotein (GP)(VSV-EBOV). Due to its favorable immune cell targeting, this vaccine has also been used as base-vector for the development of second generation VSV-based vaccines against Influenza, Nipah, and Zika viruses. However, in these situations it may be beneficial if the immunogenicity against EBOV GP is minimized to induce a better protective immune response against the other foreign immunogen. Here, we analyzed if EBOV GP can be truncated to be less immunogenic yet still able to drive replication of the vaccine vector. We found that the EBOV GP glycan cap and the mucin-like domain are both dispensable for VSV-EBOV replication. The glycan cap domain, however, appears critical for mediating a protective immune response against lethal EBOV challenge in mice. Medicines Authority (EMA) in 2019 [5, 6] . This licensed, live-attenuated vaccine is based on 52 vesicular stomatitis virus (VSV); its glycoprotein was replaced with the EBOV glycoprotein GP, 53 which is the main immunogen of the virus [7] . The vaccine VSV-EBOV, also known as rVSV-54 ZEBOV and marketed under the brand name Ervebo®, has been shown to protect nonhuman 55 primates (NHPs) from lethal disease after administration of a single dose [8] . Mechanistic 56 studies revealed that antibodies specific to the EBOV GP are the main mediators of protection 57 [9] , however, the fast-acting nature of the vaccine is likely due to a combination of strong innate 58 followed by adaptive immune responses [10] . In recent years, we have developed second generation vectors based on VSV-EBOV. The concept 61 is founded on the favorable immune cell targeting of the EBOV GP which has been hypothesized 62 to be important for the fast-acting nature of VSV-EBOV [10, 11] . VSV-EBOV-based vectors have 63 been successfully developed as vaccine candidates for a number of different viruses including 64 influenza, Nipah (NiV) and Zika viruses (ZIKV) [12] [13] [14] . Most recently, a vaccine against SARS-CoV- Here, we investigate if the immunogenicity of the EBOV GP can be reduced without 74 compromising vector replication. For this we generated VSV-EBOV vectors expressing GPs 75 harboring deletions of the two most immunogenic domains, the mucin-like domain (MLD) and 76 the glycan cap (GC) (Fig. 1A) . We found that all vectors replicated well in vitro, however, 77 protective efficacy against lethal challenge in the EBOV mouse model was reduced when the GC 78 was lacking. Our data suggests that the GC of the EBOV GP is associated with protective 79 immunity of GP-based EBOV vaccines. VSV vector expressing EBOV GP lacking the GC and MLD VSV-EBOVDGCDMLD (deletion of aa 106 228-489) was constructed and recovered from plasmid as previously described [13] . VSV 107 wildtype (VSVwt) and VSV-EBOV were used as control vaccines [18] . Mouse-adapted EBOV (MA-108 EBOV) was used for the challenge study in mice [19] . All viruses were propagated and titered on 109 Vero E6 cells, sequenced confirmed, and stored at -80 °C. Transcription and replication competent virus-like particles (trVLPs) were produced in HEK293 155 cells as described previously [25] . In place of the T7-driven tetracistronic minigenome plasmid an 156 RNA polymerase II-driven tetracistronic EBOV minigenome plasmid (pCAGGS-4cis-EBOV-eGFP) 157 expressing GFP as reporter and hammerhead and hepatitis delta virus ribozymes flanking the 158 minigenome was used (details about the cloning strategy will be provided upon request). The 159 mouse sera were heat-inactivated for 30 min at 50°C before use. For neutralization assays, showed initial signs of disease including ruffled fur and weight loss (Fig. 2B ). Over the next 5 202 days all mice in the VSVwt and control groups succumbed to the infection, as did 5/6 mice in the 203 VSV-EBOV∆GC∆MLD group (Fig. 2C ). Mice vaccinated with VSV-EBOV or VSV-EBOV∆MLD never 204 showed signs of disease and survived indicating complete protection (Fig. 2B, C) . On day 4 post-205 challenge, 4 mice per group were euthanized for sample collection to determine virus loads and 206 antibody levels ( Fig. 2A) . We found that the virus replicated to high titers in the blood, liver and (Fig. 3A) . The EBOV GP-specific titers in sera of mice vaccinated with 220 VSV-EBOV∆GC∆MLD were lower on day 0 compared to the VSV-EBOV or VSV-EBOV∆MLD 221 groups, however, the difference was not statistically significant (Fig. 3A) . On day 4, the level of 222 EBOV GP-specific IgG in the serum of VSV-EBOV∆GC∆MLD-vaccinated mice was comparable to 223 control groups and significantly lower to the other EBOV GP-based vaccines (Fig. 3A) . Next, we determined differences in neutralizing antibody titers using a well-characterized 226 surrogate system [26] . Similar to the EBOV GP-specific IgG ELISA data, the neutralization is 227 strongest after vaccination with VSV-EBOV, followed by VSV-EBOV∆MLD (Fig. 3B) . Mice 228 vaccinated with VSV-EBOV∆GC∆MLD, VSVwt and unvaccinated controls had limited to no 229 neutralizing activity in their serum (Fig. 3B ). Neutralization at 50% occurred at the following 230 serum concentrations VSV-EBOV >1:1280, VSV-EBOV∆MLD ~1:160, VSV-EBOV∆GC∆MLD ~1:20. examined the neutralizing activity in serum samples collected from the surviving mice and found 233 that the neutralizing titers for the VSV-EBOV group were comparable to day 0 (Fig. 3C ). However, the neutralizing activity in surviving mice vaccinated with VSV-EBOV∆MLD and the 235 single surviving animal vaccinated with VSV-EBOV∆GC∆MLD was clearly boosted compared to 236 day 0 indicating MA-EBOV replication following challenge (Fig. 3C) . The boosting effect through 237 challenge was less clear with animals vaccinated with VSV-EBOV indicating more potent 238 protection (Fig. 3C) . NiV and ZIKV [12] [13] [14] . In fact, the clinical development of a VSV-NiV vaccine based on VSV-EBOV 249 expressing NiV G has been funded by the Coalition for Epidemic Preparedness Innovations(CEPI) 250 and is well on its way [29] . It was previously shown that pre-existing immunity to the VSV∆G vector does not hinder re-253 vaccination with another VSV∆G-based vaccine [30] . Little is known, however, about the 254 influence of pre-existing immunity to EBOV or EBOV GP on vaccine efficacy using VSV-EBOV or 255 VSV-EBOV-based vaccine candidates. Vaccination utilizing EBOV GP has previously been shown 256 to elicit antibodies predominantly targeting the GC and MLD [31, 32] . These antibodies are 257 generally non-neutralizing as both domains get cleaved off during EBOV entry into cells before 258 receptor binding and fusion occurs [33] . However, with the increase in antibody-based 259 therapeutics development a few GC domain-binding antibodies have been described to potently 260 neutralize EBOV infection and to be polyfunctional [34, 35] , which is in general associated with 261 increased vaccine efficacy [36] . In addition, the monoclonal antibody cocktail ZMapp contains 262 the GC-binding antibody 13c6 [37] . Recently, inhibition of cathepsin cleavage by these GC-binding antibodies was proposed as a mechanism [38] . Therefore, while a deletion of the GC and 264 MLD might not interfere with the GP's ability to drive VSV replication, it could result in reduced 265 protective immunity against EBOV infection. We compared the VSV vectors expressing EBOV GP [8], the EBOV GP∆MLD [17] , and the EBOV GP∆GC∆MLD, finding no significant differences in 267 replication in cell culture (Fig. 1C) , which highlights that both domains, GC and MLD, are indeed 268 dispensable for in vitro replication of the VSV vectors. Similarly, the neutralizing titer in the single surviving mouse in the VSV-EBOV∆GC∆MLD group 293 increased. In contrast, the neutralizing activity in the serum from VSV-EBOV-vaccinated and 294 surviving mice remained the same or slightly decreased by day 42 (Fig. 3B, C) because the vaccine-induced immune responses controlled challenge virus replication as shown by no 296 viremia (Fig. 2D ) and, therefore, no boost effect was observed. In summary, our data support and expand existing data sets about the importance of antibodies 299 directed to the EBOV GP GC for protection from EVD [31, 32, 38] . Our findings also demonstrate 300 that both the GC and MLD are dispensable for VSV-EBOV-based vaccine vector replication in 301 vitro. If the EBOV GP is only used to target important immune cells ("vehicle"), a VSV vector 302 expressing the EBOV GP∆GC∆MLD might be of advantage as the immune response will likely be 303 biased, as intended, towards the second foreign immunogen. However, when developing a truly 304 bivalent vaccine with the aim to similarly protect against EBOV and a second pathogen, a VSV ICTV Virus Taxonomy 361 Profile: Filoviridae. The Journal of general virology Ebola virus disease Dis Primers 2020 World Health Organization. Ebola situation report Lessons on Vaccine Development Development of Pandemic Vaccines: ERVEBO Case Study Properties of replication-375 competent vesicular stomatitis virus vectors expressing glycoproteins of 376 filoviruses and arenaviruses Live attenuated 379 recombinant vaccine protects nonhuman primates against Ebola and Marburg 380 viruses Antibodies are 383 necessary for rVSV/ZEBOV-GP-mediated protection against lethal Ebola virus 384 challenge in nonhuman primates VSV-EBOV 389 rapidly protects macaques against infection with the 2014/15 Ebola virus 390 outbreak strain Peri-exposure 392 protection against Nipah virus disease using a single-dose recombinant vesicular 393 stomatitis virus-based vaccine Single-dose live-396 attenuated Nipah virus vaccines confer complete protection by eliciting 397 antibodies directed against surface glycoproteins A VSV-400 based Zika virus vaccine protects mice from lethal challenge A single dose of a vesicular stomatitis virus-404 based influenza vaccine confers rapid protection against H5 viruses from 405 different clades Rapid protection from 408 COVID-19 in nonhuman primates vaccinated intramuscularly but not intranasally 409 with a single dose of a recombinant vaccine The signal peptide of the ebolavirus glycoprotein 413 influences interaction with the cellular lectins DC-SIGN and DC-SIGNR Mucin-Like Domain of Ebola Virus Glycoprotein Enhances Selective 417 Oncolytic Actions against Brain Tumors Vesicular stomatitis virus-based Ebola vaccines with improved 421 cross-protective efficacy. The Journal of infectious diseases A mouse model for 424 evaluation of prophylaxis and therapy of Ebola hemorrhagic fever. The Journal of 425 infectious diseases 1999 Stat1-Deficient Mice Are Not an Appropriate Model for 428 Efficacy Testing of Recombinant Vesicular Stomatitis Virus-Based Filovirus 429 Vaccines. The Journal of infectious diseases Cathepsin B & L are not required for 435 ebola virus replication. PLoS neglected tropical diseases 2012, 6, e1923 Gamma 438 Irradiation as an Effective Method for Inactivation of Emerging Viral Pathogens The American journal of tropical medicine and hygiene Enzyme-linked 443 immunosorbent assay for detection of filovirus species-specific antibodies. 444 Clinical and vaccine immunology : CVI Generation and Optimization 447 of a Green Fluorescent Protein-Expressing Transcription and Replication-448 Competent Virus-Like Particle System for Ebola Virus Serological 452 Evidence for the Circulation of Ebolaviruses in Pigs From Sierra Leone. The 453 459 28. World Health Organization. WEEKLY BULLETIN ON OUTBREAKS AND OTHER 460 EMERGENCIES. 2021, 4-5. 461 29. Coalition for Epidemic Preparedness Innovations (CEPI) Vesicular 466 stomatitis virus-based vaccines against Lassa and Ebola viruses Antibody Repertoires to the Same Ebola Vaccine 470 Antigen Are Differentially Affected by Vaccine Vectors Protective Mechanisms Induced by a Trivalent Parainfluenza Virus-Vectored 475 Ebolavirus Vaccine Structures of protective antibodies 478 reveal sites of vulnerability on Ebola virus Systematic 483 Analysis of Monoclonal Antibodies against Ebola Virus GP Defines Features that Isolation of potent neutralizing 488 antibodies from a survivor of the 2014 Ebola virus outbreak Polyfunctional Fc-492 effector profiles mediated by IgG subclass selection distinguish RV144 and 493 VAX003 vaccines Mechanism of Binding to Ebola Virus Glycoprotein by the 497 Convergence of a common 501 solution for broad ebolavirus neutralization by glycan cap-directed human 502 antibodies Assessment of a vesicular stomatitis virus-based 505 vaccine by use of the mouse model of Ebola virus hemorrhagic fever. The Journal 506 of infectious diseases