Evaluation of Different Strategies for Post-Exposure Treatment of Ebola Virus Infection in Rodents

Zaire Ebola virus (ZEBOV) is a pathogen that causes severe hemorrhagic fever in humans and non-human primates. There are currently no licensed vaccines or approved treatments available against ZEBOV infections. The goal of this work was to evaluate different treatment strategies in conjunction with a replication deficient, recombinant human adenovirus serotype 5 (Ad-CAGopt)-based vaccine expressing the Zaire Ebola virus glycoprotein (ZGP) in Ebola infected mice and guinea pigs. Guinea pigs were treated with Ad-CAGoptZGP in combination with different treatment strategies after challenge with guinea pig adapted-ZEBOV (GA-ZEBOV). B10.BR mice were used to further characterize efficacy and immune responses following co-administration of Ad-CAGoptZGP with the most effective treatment: AdHu5 expressing recombinant IFN-α (hereafter termed DEF201) after challenge with a lethal dose of mouse adapted-ZEBOV (MAZEBOV). In mice, DEF201 treatment was able to elicit full protection against a lethal dose of MA-ZEBOV when administered 30 minutes after infection. In guinea pigs the Ad-CAGoptZGP and DEF201 combination therapy elicited full protection when treated 30 minutes post-exposure and were a superior treatment to Ad-CAGoptZGP supplemented with recombinant IFN-α protein. Further analysis of the immune response revealed that addition of DEF201 to AdCAGoptZGP enhances the resulting adaptive immune response against ZGP. The results highlight the importance of the innate immune response in the prevention of ZEBOV pathogenesis and support further development of the AdCAGoptZGP with DEF201 treatment combination for post-exposure therapy against ZEBOV infection. Copyright: © 2011 Richardson JS, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. J o u r na l o f B iote rrorism&iodefe n s e ISSN: 2157-2526 Journal of Bioterrorism & Biodefense Citation: Richardson JS, Wong G, Pillet S, Schindle S, Ennis J, et al. (2011) Evaluation of Different Strategies for Post-Exposure Treatment of Ebola Virus Infection in Rodents. J Bioterr Biodef S1:007. doi:10.4172/2157-2526.S1-007


Introduction
Ebola virus (EBOV) is a member of the family Filoviridae. They are enveloped, single-stranded, negative-sense RNA viruses, which cause severe hemorrhagic fever and are associated with highly lethal infections in humans. Several strains of EBOV have been identified to date including the Zaire, Sudan, Bundibugyo, Ivory Coast and Reston Ebola virus [1]. The most aggressive strain identified is the Zaire Ebola virus (ZEBOV) with mortality rates reported to be as high as 90% [2]. Outbreaks of EBOV infection have occurred sporadically in Africa in the past and caused substantial fatalities within the affected communities. Several factors including high lethality rates in humans, the ease of in vitro propagation and the potential for aerosol dissemination make EBOV a causative agent for biological warfare [3]. While significant progress has been made in understanding the pathogenesis of EBOV infection there is still no clinically approved EBOV vaccine or treatment available. Thus, the development of an effective post-exposure therapy is considered a high priority despite the limited impact of EBOV on the human population worldwide.
Macrophages and dendritic cells (DCs) are important components of the innate immune system and known to be the primary early targets for EBOV infection [22]. Upon infection of macrophages with ZEBOV, sustained cytokine and chemokine production was observed but with little or no interferon-alpha (IFN-α) response in vitro [23]. ZEBOV-infected DCs also did not produce IFN-α [24], but contrary to macrophages, infected DCs do not become fully activated and hence do not secrete pro-inflammatory cytokines, upregulate co-stimulatory molecules or properly stimulate T-cells [25]. This lack of stimulation results in a poor adaptive immune response. EBOV infection results in the suppression of a normal stimulation of the host interferon response through VP35 and VP24 viral proteins. VP35 has been shown to block IFN-α/β production by small ubiquitin-like modifier (SUMO)ylation of interferon regulatory factor 7 (IRF-7) [26] and inhibition of the IRF-3 kinases, namely IKK-ε and TBK-1 [27]. VP24 can interact with host cell importin-α proteins which prevent the nuclear import and accumulation of phosphorylated STAT-1 [28]. Since the STAT-1 transcription factor is utilized by both IFN-α/β and IFN-γ signaling pathways, the presence of VP24 inhibits cellular responses to both Type I and II IFN [28]. As a result the dysregulated innate immune response becomes ineffective at limiting and clearing viral infection, promotes non-productive inflammation as well as negatively impacting the potency of subsequent specific adaptive immune responses. The robust non-specific inflammatory response is suspected to contribute to the progression to shock-like symptoms, coagulation abnormalities and multiple organ failure ultimately causing a fatal outcome to the infected host. Modulating the innate immune system during the early stages of EBOV infection may therefore improve host immune defenses and lead to a more positive outcome. Bolus administration of recombinant interferon was shown to delay EBOV replication in vitro [29] and induce survival in infected mice [30] but this strategy was not successful in NHPs [29]. However, constitutive in situ production of IFN-α via a replication deficient adenovirus has demonstrated strong antiviral efficacy against SARS [31] and Yellow Fever Virus [32].
The present study compares several treatment strategies, some which modulate the innate immune system and investigates whether these treatments are able to enhance survival against a lethal ZEBOV challenge in mice and guinea pigs. The strategies evaluated include: isopropanol, dextrose, azithromycin, CD40 ligand (CD40L), AdHu5-iMYD88.CD40 with the AP1903 dimerization drug, recombinant IFN-α protein (rIFN-α), and an adenovirus expressing recombinant IFN-α (DEF201).

Construction and production of adenoviral vectors
Molecular clones of E1-deleted human adenovirus serotype 5 vectors (AdHu5) expressing ZEBOV glycoprotein (ZGP) were generated and called Ad-CAGoptZGP as described previously [42]. The authenticity of each vector was confirmed by sequencing and the recombinant virus was rescued by transfecting the linearized DNA into HEK 293 cells maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 1% penicillin, 1% streptomycin, 1% L-glutamine, 1% sodium pyruvate, and 10% fetal bovine serum. Largescale infections (5 x 10 8 cells) were initiated from positive transfectants and purified by cesium chloride as previously described [43]. Genome structures of vectors were analyzed by restriction digestions of viral DNA and compared with those of the molecular clones as previously described [43]. Particle number and infectivity (IFU) of vectors were determined by standard optical density and immunodetection of the AdHu5 hexon protein respectively, following infection of HEK 293 cells with limiting dilutions of each vector preparation according to the recommendations by the manufacturer (Adeno-X rapid titer kit, Clontech, Mountain View, CA). Several Ad-CAGoptZGP preparations were generated and quantified for both infectious particle and total particle number. Preparations with a ratio of at least 1:200 infectious to total particle were used in this study. AdHu5 expressing recombinant IFN-α (DEF201) were provided by Defyrus Incorporated, ON and manufactured as described [44].

Animal models, vaccination and challenge
Groups of 5 or 10 B10.BR mice (Jackson Laboratory, ME) were challenged by intraperitoneal (I.P.) injection with 1000 x LD 50 of mouse-adapted ZEBOV strain Mayinga [45]. They were then treated by intramuscular (I.M.) injection with 1 x 10 7 infectious particles of either empty recombinant adenoviral vector diluted to a total of 500μl volume with PBS or DEF201 via the I.M. or I.N. routes 30 minutes post challenge. The mice were then weighed every day for 14 days and monitored for clinical signs of disease for 30 days following challenge.

Neutralizing antibody assay
Sera harvested from rodents 28 days post vaccination were inactivated at 56 o C for 45 minutes. Twofold serial dilutions of each sample (1:10, 1:20, 1:40, etc, in 50 ml of DMEM) was mixed with equal volume of ZEBOV encoding the eGFP reporter gene (100 transducing units/well) and incubated at 37 o C for 60 minutes. The mixture was then transferred onto sub-confluent VeroE6 cells in 96-well flat-bottomed plates and incubated for 90 minutes at 37 o C in 5% CO 2 . Control wells were infected with equal amounts of ZEBOV-eGFP with or without the addition of serum. 100 ml of DMEM supplemented with 20% FBS was then added to each well and plates were incubated at 37 o C in 5% CO 2 for 48 hr. Dilutions were scored as positive for the presence of NAbs when there was a greater than a 50% reduction in the number of cells expressing eGFP in a field of view using a fluorescent microscope. The highest serum dilution scoring positive for NAb was recorded and neutralization titers were reported as the reciprocal of this dilution. Positive and negative control sera used in this experiment were harvested from mice or guinea pigs 28 days post immunization with a VSV vaccine expressing ZGP, or naïve mice or guinea pig sera respectively. All infectious in vitro work was performed in the biocontainment level 4 laboratory of the NML, PHAC.

ZGP-specific IgG ELISA
Immulon 2 HB 96 well flat bottom MicroTiter ELISA Plates (Thermo Scientific) were coated overnight with 50µl/well of 1 µg/ ml His-ZEBOV-GP capture antigen diluted in PBS [47]. Plates were washed three times with wash buffer (PBS 0.1% TWEEN 20) and then blocked for 90 min with blocking buffer (5% skim milk powder/ PBS/0.2% Tween 20) at 37ºC [47]. Plates were washed three times with wash buffer. Sera collected from either mice or guinea pigs

Flow cytometry analysis
The frequency of CD4+ or CD8+ cells producing IFN-γ TNF-α and IL-2 was assessed by flow cytometry. Mouse splenocytes were isolated day 10 post Ad-CAGoptZGP vaccination by harvesting the spleens and grinding spleen tissues. Splenocytes were seeded at 2 x 10 6 cells per well in DMEM media supplemented with 10% FBS 1% penicillin, 1% streptomycin, 1% L-glutamine, 1% non-essential amino acids, 1% sodium pyruvate, 1% HEPES buffer, 5 × 10 −3 M 2-mecaptoethanol. Cells were stimulated for 5 hours with 5μg/ml of the peptide carrying the immunodominant MHC class I epitope of ZGP (TELRTFSI) for mice with the H-2 k haplotype [48] in the presence of 1μl/ml of the protein transport inhibitor GolgiPlug (BD Biosciences). Splenocytes were then stained with PacBlue-conjugated rat anti-mouse CD4 and PerCp-Cy5.5-conjugated rat anti-mouse CD8 antibodies (BD Biosciences), followed by a 20 minute incubation in Cytofix/Cytoperm (BD Biosciences). Intracellular cytokines were detected after staining with PE-conjugated anti-mouse IFN-γ, PECy7-conjugated anti-mouse  Vaccines Development x 10 10 infectious forming units (IFU) of Ad-CAGoptZGP per animal in conjunction with either dextrose, AdHu5-iMYD88.CD40 with dimer drug, CD40L, rIFN-α, DEF201 or 6 hours later with isopropanol, or azithromycin from days 1-5 post-challenge. Control guinea pigs were injected I.M. with PBS. Infection of control guinea pigs resulted in rapid weight loss combined with 100% mortality by day 7 post infection. Delays to time of death were observed for the isopropanol treatment (day 12), dextrose (day 12), azithromycin (day 10), AdHu5-iMYD88.CD40 with the dimer drug (day 15) and CD40L (day 11) but survival was not observed for these treatment groups (Figures 2A and  2B). Treatment of guinea pigs with Ad-CAGoptZGP in combination with rIFN-α resulted in 2 of 3 survivors where surviving animals demonstrated moderate clinical symptoms with on average a maximum weight loss of 11% (Figures 2A and 2B). When DEF201 was mixed with the Ad-CAGoptZGP vaccine and injected I.M. into guinea pigs, complete protection (6 survivors out of 6) was observed with little to no weight loss or clinical symptoms ( Figures 1A and 1B). Interestingly, treatment with DEF201 alone did not provide protection in guinea pigs against a lethal challenge with GA-ZEBOV (data not shown).

Survival and weight loss in mice following Ebola virus challenge and treatment
There are currently only limited reagents available for studying the immune response in guinea pigs. Therefore mice were used to better define the additive effect of DEF201, either directly or indirectly, to Ad-CAGoptZGP. Thirty minutes after challenge with 1000 x LD 50 of MA-ZEBOV, B10.BR mice were administered with either a single I.M. or intranasal (I.N.) dose of 1 x 10 7 IFU of DEF201 per animal. Control B10.BR mice were administered PBS. An additional control consisting of mice administered replication-deficient AdHu5 vector not expressing any transgene (AdHu5-empty) I.M. was added to evaluate the contribution the vector alone since AdHu5 particles were shown to modulate the innate immune response [34]. Infection of PBS control mice resulted in rapid weight loss and 100% mortality between days 6 and 8 post-challenge. In contrast, complete survival was observed in MA-ZEBOV challenged mice following treatment with either I.M. or I.N. administration of 1 x 10 7 IFU of DEF201 with no weight loss or clinical symptoms ( Figure 3A). Interestingly, treatment of B10.BR mice with AdHu5-empty at the same dose resulted in 40% survival. Surviving mice displayed signs of illness with an average maximum weight loss of 17% ( Figure 3B).

Immune responses in mice following treatment
The increased protection observed in guinea pigs with the addition of DEF201 to Ad-CAGoptZGP could be due to an independent antiviral effect from the expression of IFN-α directly on ZEBOV, and/or that IFN-α further improves the adaptive immune response generated by Ad-CAGoptZGP. To address this last possibility, cellular and humoral immune responses were evaluated in the mouse model. The flow cytometry analysis revealed that vaccination with Ad-CAGoptZGP primed a T-cell response characterized by functional CD8+ effector cells producing IFN-γ and TNF-α after restimulation with the immunodominant ZGP peptide, and the absence of these cytokines in CD4+ cells (Figure 4, middle and left panels respectively). A significant amount of those CD8+/IFN-γ/TNF-α+ cells also produced IL-2 ( Figure  4, right panel). Co-treatment with DEF201 increased the proportion of the CD8+/IFN-γ+ cells to 5.42% from 2.82% and CD8+/IFN-γ/IL-2+ cells to 0.42% from 0.25% with Ad-CAGoptZGP alone (Figure 4, lower panel). As expected, no ZGP-specific response was observed in cells from mice treated with DEF201 alone. TNF-α (BD Biosciences), and APC-conjugated anti-mouse IL-2 diluted in PermWash buffer (BD Biosciences). At least 300 000 events were analysed using the 17-color flow cytometer (LSR II Flow Cytometer, BD Biosciences).

Survival and weight loss in guinea pigs following Ebola virus infection and treatment
Complete survival was previously observed in mice administered Ad-CAGoptZGP 30 minutes after infection with 1000 x LD 50 of MA-ZEBOV [33]. However, the same strategy resulted in delayed time to death with only 1 survivor out of 3 guinea pigs after challenge with 100 x LD 50 of GA-ZEBOV in a pilot experiment (Figures 1A and 1B). In an attempt to improve the survival rate, several treatments were evaluated in guinea pigs when administered 30 minutes after lethal challenge with GA-ZEBOV. Selected strategies hypothesized to improve survival included modulators of the innate immune response; fluids or an antibiotic each evaluated in conjunction with the Ad-CAGoptZGP vaccine. Guinea pigs were first infected with 100 x LD 50 of GA-ZEBOV, followed 30 minutes later by a single intramuscular (I.M.) injection of 1 The humoral immune response was analyzed on mice serum harvested at 28 days post treatment via total anti-ZGP IgG ELISAs and neutralizing antibody (NAb) assays. Significant levels of total anti-ZGP IgG antibodies were detected for both the Ad-CAGoptZGP with and without DEF201 treatment groups, with corresponding A405 values at 1.31±0.03 and 1.24±0.02 respectively. Background levels of IgG were observed for the DEF201 treatment group with an A405 value of 0.19+0.01 ( Figure 5A). Neutralizing assays revealed that sera from DEF201 treated B10.BR mice did not have detectable NAb levels on day 28, where the limit of detection was set at 10 reciprocal dilutions. Significant levels of NAb were detected from sera of mice treated with the Ad-CAGoptZGP with or without DEF201, with average NAb titers of 42±13 or 23±6 reciprocal dilutions respectively on day 28 ( Figure 5B).

Humoral immune responses in guinea pigs following treatment
The humoral immune response was also analyzed on guinea pig serum harvested 28 days post vaccination via total anti-ZGP IgG ELISAs and neutralizing antibody (NAb) assays. Significant levels of total anti-ZGP IgG antibodies were detected from sera of animals treated with Ad-CAGoptZGP with or without DEF201, with corresponding A405 values of 2.40±0.09 or 2.17±0.06 respectively ( Figure 6A). Similarly, significant levels of NAb were detected in the serum of guinea pigs treated with Ad-CAGoptZGP alone or in combination with DEF201 with corresponding values at 20±12 and 45±21 respectively ( Figure 6B).

Discussion
EBOV viruses are among the most deadly infectious agents in humans and it is thought that their virulence is in part due to their capacity to suppress and evade host IFN responses [1]. Consequently, it was hypothesized that productive modulation of the innate immune response with interferon could suppress EBOV replication in vivo sufficiently enough to allow the immune system to mount a protective adaptive response.
Several strategies which modulate the innate immune response were investigated in this study. Isopropanol administration has been shown to have immunosuppressive properties that prevent toxic shock [35], a common suspected occurrence in EBOV human cases. Dextrose administration induces a temporary state of acute hyperglycemia which causes endothelial progenitor cells to adapt a pro-inflammatory phenotype and enhance naïve T-cell activation [36]. Azithromycin is a macrolide antibiotic with the ability to suppress the pro-inflammatory activity of macrophages and is used to treat chronic inflammatory diseases [37]. Since loss of intestinal integrity is common in advanced EBOV infection, it also has the added benefit of preventing the normal intestinal flora from causing further infection after traversing the gastrointestinal barrier. CD40 ligand (CD40L) can partially mimic CD4+ T-cell activation of the CD40 receptor on DCs which aids in the maturation of immature DCs [36]. AdHu5-iMYD88.CD40 treatment in combination with the AP1903 dimerization drug targets the sustained activation of DCs and results in higher antigen-specific CD8+ T-cell responses [36]. IFN-α has been shown to be a clinically effective general immunotherapeutic antiviral in the past, where it is responsible for priming the host innate immune response and then transitioning into an effective adaptive immune response by regulating various cytokines and their receptors [38]. Furthermore, administration of recombinant IFN-α was shown to prevent death in EBOV-infected mice but not in nonhuman primates thus far [29].
Protection against ZEBOV infection was demonstrated in mice following I.M. and I.N. administration with DEF201. Surprisingly partial protection was observed within the AdHu5-empty control group, where 2 of 5 mice survived despite being challenged with a uniformly lethal dose of MA-ZEBOV. This suggests that the AdHu5 vector itself also contributes to the stimulation of a protective immune response against ZEBOV. Indeed, the AdHu5 vector has been shown to promote both transduced and bystander DC maturation as well as stimulate Type I IFN production [39].
While protection was not observed with the isopropanol, dextrose, azithromycin, CD40L, and AdHu5-iMYD88.CD40 treatments on guinea pigs, delays in the occurrence of clinical symptoms and median time to death were observed for all these treatments, suggesting that they participated in delaying ZEBOV-induced pathogenesis. It is possible that further optimization of some of these interventions could lead to better outcome and in turn be useful notably for improving Figure 5: Humoral immune responses in vaccinated mice with or without DEF201 treatment. A) Serum ZEBOV GP-specific IgG ELISAs. The harvested serum samples were also tested for the presence of IgG antibodies against ZEBOV GP at a 1:50 dilution ratio. The data is reported as the optical density measured by absorbance at 405nm. B) Serum ZEBOV GP-specific neutralizing antibody response. Groups of 3 to 6 mice were vaccinated and treated with or without DEF201, where serum was harvested 28 days post-vaccination. Serum neutralizing antibody levels were reported as the reciprocal of the dilution where greater than 50% viral neutralization was observed. Figure 6: Humoral immune responses in vaccinated guinea pigs with or without DEF201 treatment. A) Serum ZEBOV GP-specific IgG ELISAs. The harvested serum samples were also tested for the presence of IgG antibodies against ZEBOV GP at a 1:50 dilution ratio. The data is reported as the optical density measured by absorbance at 405nm. B) Serum ZEBOV GP-specific neutralizing antibody response. Groups of 4 guinea pigs were vaccinated and treated with or without DEF201, where serum was harvested 28 days postvaccination. Serum neutralizing antibody levels were reported as the reciprocal of the dilution where greater than 50% viral neutralization was observed. patient management. In all aspects, the present study further support the concept that IFN-α is important early in EBOV infection with regards to survival. This was also a conclusion of previous work, which showed that I.M. administration of IFN-α2β protein into cynomolgus macaques was able to delay viremia and death from lethal ZEBOV infection by several days [29]. Unfortunately, recombinant IFN-α has a short half-life of 8 hours in vivo [40] and can be extremely costly. Effective treatment of EBOV infection may require sustained levels of IFN-α for days in order to increase efficacy and give more time to the adaptive immunity to mount a protective response. DEF201 can produce constitutively high levels of IFN-α for days in a cost efficient manner. In addition to higher and more sustained expression of IFN-α, the AdHu5 particles themselves resulted in improved survival in mice when administered 30 minutes after lethal MA-ZEBOV infection. The addition of DEF201 to the Ad-CAGoptZGP vector resulted in improved survival in comparison to recombinant IFN-α together with Ad-CAGoptZGP. The lack of survival in guinea pigs administered only DEF201 further highlights the enhanced protection of the combination therapy.
Analysis of the resulting T-and B-cell immune response following Ad-CAGoptZGP administration with DEF201 showed overall elevated cellular and humoral responses relative to Ad-CAGoptZGP alone. CD8+ cells primed by the Ad-CAGoptZGP vaccination were functional according to their ability to produce IFN-γ and TNF-α after ex vivo restimulation. The presence of a significant proportion of these cells producing IL-2 is indicative of resting memory T cells [41]. Therefore, DEF201 treatment could contribute to a better clearance of the virus and also promote the establishment of memory cells. However, the positive effect observed with DEF201 alone also indicates that the antiviral state induced by added expression of IFN-α plays a role in controlling or delaying EBOV replication. Interestingly, even only delaying replication is likely to be helpful in providing more time for the generation of a protective immune response. Taken together the results further emphasize the role of IFN-α and the innate immune response in early ZEBOV infection. Future work will need to evaluate the combination of DEF201 with Ad-CAGoptZGP as a treatment to nonhuman primates already infected with a lethal dose of ZEBOV.