Skip to main content

Advanced Studies towards Knowledge on Lyssavirus Encephalitis
Pathogenesis Improving Options for Survival

Final Report Summary - ASKLEPIOS (Advanced Studies towards Knowledge on Lyssavirus EncephalitisPathogenesis Improving Options for Survival)

Executive Summary:
ASKLEPIOS has addressed the topic Health.2013.2.3.4-1: Neglected infectious diseases of Central and Eastern Europe. The ultimate goal of projects under this call was to deliver knowledge that can contribute to the future prevention, treatment or diagnosis of neglected diseases that are disproportionally affecting Central and Eastern Europe. This knowledge should focus on biological mechanisms and pathology of the diseases in question.

Rabies has a strong impact at the scientific, clinical, and cultural levels. It is one of the most ancient known diseases that can be effectively prevented by means of vaccination, yet is still one of the most devastating diseases . Morbidity and mortality caused by RABV outcompetes that of Ebola and Marburg virus infections, being the only virus known to kill approximately 100% of the affected persons.

Within ASKLEPIOS we have identified a number of compounds that can (significantly) increase survivorship of experimentally infected mice and have therefore taken a significant step forward in the identification of rationally designed treatment strategies for rabies.

Project Context and Objectives:
Rabies encephalitis is considered an invariably fatal disease. A small number of patients with rabies have recovered from the disease, albeit with (severe) neurological sequelae. All approaches to treat rabies have been based on the “trial-and-error” approach and therefore the reasons for these few successful treatments are poorly understood and have not been reproduced. One of the main bottle-necks in progressing with the identification of successful treatment strategies is the fact that the pathogenesis of the disease is poorly understood, which limits the possibilities of rationally designing intervention strategies.

ASKLEPIOS aims to address this point by bringing together existing knowledge in order to understand the mechanisms that govern pathogenesis of rabies and subsequently design novel intervention strategies. These novel intervention strategies are based on rational choices of compounds with a high likelihood for success, rather than bed-side, trial-and-error approaches.
Three issues remain of paramount importance and will be addressed within ASKLEPIOS:

1. Rabies pathogenesis is a multifactorial process. As a result, successfully treating rabies most likely will require a combination of compounds that will influence different determinants of pathogenesis. ASKLEPIOS proposes a combination therapy regime against rabies.
2. Previous studies (including studies from the partners involved in ASKLEPIOS) have identified that both the virus and the host response play an equally important yet detrimental role during rabies encephalitis. As a result, a successful combination therapy against rabies most likely will require compounds that inhibit virus replication and prevent detrimental host responses. ASKLEPIOS proposes the combination of compounds against both these factors as a rationally designed intervention strategy against rabies.
3. Rabies virus is exclusively neurotropic and the disease caused by this virus is restricted to the central nervous system (CNS). As a result, a successful treatment should be able to reach the brain even if given peripherally. ASKLEPIOS is using an innovative approach to address this issue by using blood-brain-barrier (BBB) openers to allow for combination therapy to get into the CNS and be effective on both eliminating the virus but also the host responses that have a detrimental effect on the brain.

By using this two-hit-approach model and the innovative BBB-openers, ASKLEPIOS aims for a rational, innovative combination therapy against rabies that when successful can be used in all patients with high chances of reproducibility.
Project Results:
Hereunder the principal Science & Technology results are reported per partner. Where relevant, some information has not been included because at the time of writing this report, these data are considered as confidential information.

Erasmus Medical Centre Rotterdam (EMC, partner 1)

EMC has been involved mainly in WP2 and WP4 as well as WP5 (management) and WP6. During the kick off meeting of the consortium it was agreed that EMC will also take over one of the tasks of partner 5 in WP1.

WP1: Task 1.4 Testing type-I IFNs, Viperin and MAP kinase inhibitors
The inhibitory effect of Viperin was tested by EMC
At EMC, plasmids expressing murine Viperin were obtained through external collaboration and propagated. Murine Viperin was transfected in N2A cells or BHK-21-C13 cells, and transfection efficiency was determined. Virus stocks (CVS-11) were titrated on N2A cells transfected or non-tranfected with Viperin: No reduction in viral titre was observed in tranfected cells. When N2A and BHK-21-C13 cells were transfected with Viperin and subsequently infected with CVS-11, a slight reduction of viral titre was observed in Viperin-transfected N2A cells on day 3 post-infection. Therefore we choose to continue experiments with N2A cells only. N2A cells were transfected with Viperin and subsequently infected with CVS-11 or influenza virus (positive control of viral inhibition): 1-2 logs reduction of viral titre was observed in Viperin-transfected N2A cells on days 2, 3 and 4 post-infection. Finally, N2A cells were infected with CVS-11 and 24 hours later transfected with viperin: 0.5 log reduction of viral titer was observed in N2a viperin-transfected cells on day 2 post infection (day 1 post transfection).
EMC was not able to convincingly demonstrate a substantial inhibition of viral replication in Viperin transfected cells and was therefore decided not to include this molecule in subsequent in vivo experiments in WP4.

WP1: Task 1.5 Exploring combinations of siRNA etc.
EMC tested a combination of different molecules (T705, ribavirin, FDA-1) in an in vitro model of inhibition of viral replication. Protocols were wet up at the lab of partner 2 and were shared across the consortium. EMC identified that FDA-1 molecule could reduce viral replication in vitro but only on cytotoxic concentrations. Ribavirin had a synergistic effect on viral inhibition when combined with T705. It was therefore decided that both these molecules should be taken forward to WP4.

WP2: Task 2.1 Inhibition of pyroptosis
Approval to conduct animal experiments obtained from Animal Ethics Committee of EMC on 27 March 2014. As agreed during the kick off meeting in Rotterdam (January 2014), partner 1 grew and distributed sufficient amount of virus stocks for the in vivo experiments of WP2. For this reason, Eastern European fox virus was obtained from Dr. Thomas Muller whereas Silver-haired bat rabies virus (SHBRV-18) was available in the lab of partner 1. Eastern European fox virus and SHBRV-18 were grown, titrated and distributed to partner 4.
At the consortium meeting held in Prague (October 2014), under the strong recommendations of the SAB, it was decided that all host response inhibitors will be tested against one virus only: SHBRV-18. Animal experiments were conducted in Pulawy, Poland, at the laboratory of partner 4 in the presence of investigators from both EMC and ARTEMIS. Animals were infected intra muscularly with SHBRV-18 virus and on day 5, 7 and 9 post infection were given CASPASE-1 (CASP-1) inhibitor or IL-1β inhibitor intra-cranially.
CASP1 inhibitor significantly extended the mean survival time of treated animals by 1.5 days compared to non-treated virus infected control animals (p=0.0371 Log-rank Mantel-Cox test). IL-1β inhibitor did not have an effect on the survival time of treated mice compared to the non-treated controls. Viral titers were recovered from all parts of the CNS, similarly with the titers recovered from mice treated with CASP-1 or IL-1β or the non-treated controls.
Next we measured mRNA of several markers involved in the pyroptotic pathway, namely CASP-1, IL-1β, IL-18 and TNF-α. The effect on mRNA expression was more pronounced in mice treated with IL-1β inhibitor, compared to mice treated with CASP-1 inhibitor. This differential mRNA expression (mostly seen as down-regulation of the genes tested) was seen both in comparison with non-infected treated mice.
Histopathological screening of the brains that were collected from the mice revealed that infected animals had no apparent visible abnormalities. Occasionally, multifocal mild to moderate lymphocytic meningoencephalitis was seen. Intracytoplasmic eosinophilic inclusion bodies (Negri bodies) were not seen neither necrosis was seen. Occasionally rabies-antigen positive neurons showed signs of degeneration as demonstrated by loss of Nissl substance with shrinkage. Pyknosis and karyorhexis without influx of inflammatory cells was present, although not abundant. and with infected, non-treated mice.

WP4: Task 4.2: Evaluation of combination therapy for rabies infections
EMC prepared two different batches of SHBRV-18 virus which was distributed to all partners. The first batch was for the first set of experiments used at LD100 and the second batch was used for the second set of experiments and was used at LD50. EMC worked towards the deliverables of this task in collaboration with ARTEMIS. Two sets of experiments were done under this task.
These experiments fall under the information that is confidential.

Animal and Plant Health Agency (APHA, partner 2)

APHA were tasked with numerous work package deliverables, in principle WP1 and WP4. These are outlined below with the actions involving collaboration with other consortium members are indicated. For much of the in vivo studies, decisions regarding almost all experiments were decided at consortia level with extensive ethical discussion. Where significant outputs were generated the details are underlined and bolded

WP1: Virus replication inhibitors
1.2 Development of siRNA
APHA investigated the antiviral activity of siRNAs targeting three different genes: nucleoprotein (N), glycoprotein (G) and polymerase (L) in collaboration with the IP. siRNA constructs were tested in collaboration with Changchun Veterinary Research Institute, China, and evaluated by transfection assay. These constructs had previously been shown to inhibit RABV (CVS) replication (Yang et al., 2012). However, APHA was unable to reproduce these results, and did not observe any knockdown of CVS. Additionally, adaptations of previously-published siRNAs were manufactured by Sigma (UK): RNA124 (Brandao et al., 2007), sh-RNA-G7 (Meshram et al., 2013) and RV-L (Sonwane et al., 2012). APHA transferred the manufactured siRNA to IP for testing. Despite a limited antiviral response, these molecules were highly cytotoxic - therefore due to this reason, along with the high cost, these siRNA were not considered to be a viable option for further assessment.

1.3 Development of RABV-neutralizing aptamers (APHA) and antiviral peptides dermaseptins
RABV-neutralising aptamers:
APHA tested a number of RNA aptamers for antiviral activity, which target the viral glycoprotein. These were developed within another APHA collaborative project, in collaboration with The University of Leeds, UK. When administered to N2A cells 1 hour after infection with CVS-11, all of the RNA aptamers tested did exhibit limited antiviral activity after 48-hour incubation (post-infection method), although higher concentrations than those tested would be required to achieve 100% knock-down of virus. However, due to the limited quantity/concentration obtained from transcription, and the subsequent relatively high cost, RNA aptamers are unlikely to be considered a viable option for further in vivo assessment or a potential clinical application.

Dermaseptin peptides:
APHA also tested dermaceptin peptides S4, S4(M4K) and S4(1-16), using sequences obtained from IP. Both the original and TAT-conjugated versions of the peptides were manufactured by Alta Biosciences, Birmingham, UK. Prior to the assessment of dermaseptin peptides, IP hosted Karen Mansfield (APHA) for one week during June/July 2014 (months 7/8), for training in working with dermaseptin peptides. Following assessment at APHA, when administered to N2A cells 1 hour after infection with CVS-11, peptides S4 and S4(M4K) demonstrated significant antiviral activity after a 48 hour incubation (post-infection method), confirming previous observations at IP. TAT-conjugation did not increase the antiviral activity. Peptides S4 and S4 (M4K) were considered candidates for further in vivo assessment.

Additional compounds introduced by partner 2 (APHA)
Additionally, APHA has introduced to the consortium and tested favipiravir (T-705), the MAPK inhibitor SB203580 and Ribavirin (not specified in the project deliverables) – a highly significant antiviral effect was observed for all of these compounds against CVS-11 in N2A cells, when tested using the APHA post-infection method. Both of these compounds were therefore considered strong candidates for further combination and in vivo testing. T-705 was initially tested in preliminary in vivo experiments at APHA, which suggested a potential improvement to survival of RABV-infected mice (IC challenge) following IV administration of T-705 at 3 days post infection. Following numerous discussions with both the academic board and the rest of the ASKLEPIOS consortia, several molecules were tested extensively in vivo as part of WP4.

Additional work undertaken by APHA
APHA propagated and disseminated CVS-11 stocks to partner laboratories within the consortium. Additionally, a protocol for the in vitro assessment of antiviral compounds was developed at APHA. This protocol was subsequently disseminated throughout the consortium to ensure that all partners were using the same method , in order to generate comparable results. APHA also hosted two visitors from partner laboratories, Lucie Dufkova (Partner 3, VRI) and Andras Marosi (Partner 5, SzIE). Miss Dufkova and Mr. Marosi both visited for two weeks, for training in rabies diagnostic techniques.

WP4: In vivo testing of therapeutic molecules
Some of the work performed by APHA under this WP is considered as confidential information.
Task 4.1 Evaluation of viral replication inhibitors
Following successful evaluation of numerous candidate antiviral molecules in vitro (WP1-3) candidate molecules were tested in vivo at APHA using an established mouse model. As well as assessing individual molecules APHA has tested numerous delivery methods for different antiviral preparations under this WP. Each stage of in vivo experimentation required extensive ethical discussion and review.

Assessment of survivorship following intracranial (IC) inoculation with the challenge virus strain (CVS) of rabies and IC inoculation of different antiviral molecules.
The toxicity test did not result in any adverse effect to the mice from the IC treatment and as such the experiment was allowed to progress. In total, 185 female CD1 mice of 3-4 weeks in age were utilised for the study with, from the 160 that were inoculated with virus, only 3 surviving infection. The survivors were in groups treated with T-705 (1 mouse) and single chain antibody fragment (ScFv; 2 mice). All other mice infected IC with 100TCID50 of CVS succumbed with clinical disease by day 5 with the vast majority succumbing either on days 5 or 6. Nevertheless, although not statistically significant (p>0.05) this study gave us the impetus to reassess the effect, if any, of antiviral treatment, but using a peripheral route of virus infection to reduce the severity of the challenge model.

Tolerability assessment and investigation into the effect of T-705 delivery on mouse survival following rabies infection.
A treatment regimen using different routes of antiviral inoculation was used to assess whether the delivery of the antiviral molecules by different routes influenced the outcome of infection. To enable assessment of the route of inoculation of antiviral molecules both intravenous (IV) and intraperitoneal (IP) inoculations were administered daily. Animals were inoculated with antiviral molecules both before during and after the virus infection to maximise the concentration of antivirals present during the virus inoculation phase. The study demonstrated both tolerability to the antivirals and that the administration of antiviral molecules had an effect on the outcome of the onset of clinical disease and survivorship.

Assessment of AR-12 as an antiviral molecules in vivo (ASUF70/7394-2-007)
During the life of the WP an opportunity arose to test a further novel antiviral molecule AR-12 as a potential antiviral molecule against rabies virus and, after extensive in vitro assessment and ethical clearance, a small study was performed in vivo. Unfortunately the administration of AR-12 post inoculation had no effect on survivorship. Following successful evaluation of numerous individual candidate antiviral molecules in vivo ethical clearance obtained to assess the effect of combination therapies, using combinations of molecules identified in deliverable 4.1. Assessment of survivorship following peripheral (IM) inoculation with SHBV and IP inoculation of different combinations of antiviral molecules with mannitol. The results of these experiments fall under confidential information.

Veterinary Research Institute Vyzkumny Ustav Veter (VUV, partner 3)

WP3 Task 1: Adaptation and evaluation of the in vitro BBB model
In vitro BBB models are valuable and easy to use supporting tools that can precede and complement animal and human studies. Two in vitro models of BBB were developed in this study. Both models are based on co-culture of human microvascular brain endothelial cells with human astrocytes in transwell system. The first model combines primary human microvascular endothelial cells (HMEBC) and primary human cortical astrocytes (available commercially from the Applied Cell Biology Research Institute/Cell Systems, Kirkland, WA, USA). In this model, the HMEBC are cultured inside the culture inserts, the astrocytes are grown on the bottom of the well in 6-well plates. The second one combines immortalized human microvascular endothelial brain cells hCMEC/D3 (kindly provided by Dr Pierre-Olivier COURAUD from Institut COCHIN, Paris, France) and primary human astrocytes. In this model, we culture the hCMEC/D3 inside the transwell inserts and the astrocytes are grown at the abluminal side of the transwell membrane in the 6-well plates. The culture conditions of both models were optimized to reach properties typical for BBB in vivo. Our aim was to reach highly differentiated tight junctions, high electrical transendothelial resistance, and low permeability for hydrophilic molecules as sucrose or inulin. Specific transporters for molecules such as aminoacids and glucose, specific enzymatic activities of the BBB (γ-glutamyl transpeptidase, monoamine oxidase), and P-glycoprotein were developed in the cells forming the in vitro BBB models. These data indicate that our BBB models closely mimic the in vivo situation by reproducing some of the complexities of the cellular environment that exist in vivo, while retaining the experimental advantages associated with tissue culture.

WP3 Task 2: Testing of selected compounds for toxicity and penetration through BBB using the in vitro model
The toxicity of labradimil and bradykinin on the both primary human microvascular endothelial brain cells and hCMEC/D3 was investigated by the MTT assay. Mannitol was not tested since this compound was used in in vivo experiments only and instead of mannitol, labradimil as a bradykinin agonist was used as mentioned above. Serial dilutions of the stock solution of the analyzed compounds ranging from 1-50 µM (labradimil) and 1-10 µM (bradykinin) were prepared in culture medium for the treatment of the tested cells. After incubation, 20 μL of MTT solution (5 mg/mL in PBS) were added to each well in cultivation plate and the plates were incubated for 4 h at 37 °C. The plates were subsequently centrifuged and purple formazan product was dissolved by the addition of 100 μL of DMSO to each well. The absorbance was monitored at 570 (measurement) and 630 nm (reference). In both cell systems, no cytotoxity was observed when compared the values from drug-treated and mock-treated cells in the investigated ranges of concentrations.
These in vitro BBB models described above were used to test the BBB permeability after treatment with BBB openers bradykinin, and labradimil. The cells grown in our models were treated with a range of concentrations of the BBB openers, 1-50 µM (labradimil) and 1-10 µM (bradykinin), for time intervals ranging from 15 to 60 mins. After 15, 30, 45, and 60 mins, transendothelial electrical resistance was measured by the EVOM2 TEER instrument. The EVOM2 not only qualitatively measures cell monlayer health, but also quantitatively measures cellular confluence of the cellular monolayer. The experiments were performed in triplicates and the whole experiments repeated four times. Despite relatively high variability of the data, there were significantly lower values of transendothelial electrical resistance in both culture systems treated with 10 µM bradykinin at time intervals 15 and 30 mins. Treatment with 50 µM solution of labradimil increased significantly BBB permeability as measured by transendothelial electrical resistance at most investigated time intervals.
Data from measurements of the transendothelial electrical resistance were subsequently evaluated by transport experiments. Cells in the both BBB models were treated with the BBB openers as described above. After 15, 30, 45, and 60 mins, we determined the permeability of the cell monolayer using transport experiments by Lucifer Yellow. Lucifer Yellow was used as a paracellular marker. This small hydrophilic molecule presents a low cerebral penetration and its endothelial permeability coefficient reveals the endothelial cell monolayer integrity. All experiments were performed in triplicates. The amount of Lucifer Yellow was determined in representative samples from each lower compartment, and in triplicate from the initial solution-containing compound. To obtain a concentration dependent transport parameter, the clearance principle was used. For each time, the increment in cleared volume between the successive sampling events was calculated by divided the amount of transported solute by the donor chamber concentration. The total volume cleared was calculated by summing the incremental cleared volumes up to the given time points. We observed relatively high inter-experiment variability in particular in case of the in vitro BBB model based on the primary human microvascular encothelial brain cells. The second BBB model provided more stable data and confirmed our findings from the measurement of transendothelial electrical resistance; i.e. 10 µM bradykinin at time intervals 15 and 30 mins, and 50 µM solution of labradimil increases significantly BBB permeability in the in vitro system.

WP3 Task 3: Testing the “openers” of the BBB using an in vivo model
Blood-brain barrier (BBB) openers were tested in vivo for their ability to increase BBB permeability in adult laboratory C57Bl/6 mice. Three compounds were selected on the basis of results from the in vitro experiments; i.e. bradykinin, labradimil, and mannitol. The BBB permeability was assayed using sodium fluorescein. Sodium fluorescein (Sigma-Aldrich), a low molecular mass molecule (376 Da), was used to detect fluid shifts between the circulation and CNS, which occur when BBB permeability becomes enhanced. Mice were injected with 10 mg sodium fluorescein in 0.1 ml sterile saline i.p. Thirty minutes later, animals were anesthesized with ketamine-HCl (100–200 mg/kg), cardiac blood was collected, and mice were perfused with PBS to remove blood from the intravascular compartment. Brains were removed, individually weighed, and stored at −70°C until processing. Homogenization of brain tissues was performed in 1 ml sterile PBS using TissueLyser II (Qiagen). The homogenate was clarified by centrifugation at 14000× g for 10 min, at 4°C. The amount of fluorescein in each sample was determined using standards ranging from 125 to 3000 µg on an Infinite M200 fluorometer (Tecan) using an excitation wavelength of 480 nm, and fluorescence was read at 538 nm. The uptake ratio was expressed as the ratio of the amount of sodium fluorescein measured in the brain to the amount measured in serum.
Bradykinin and labradimil were tested at a range of concentrations from 5 to 50 µg/kg. The compounds were administered intraperioneally and in other experiments intravenously 15 mins after intraperitoneal infection of sodium-fluorescein solution. Control mice were injected with sterile vehiculum. 30 mins after the administration of the drugs, mice were processed as described above. Independently on the dose and way of administration, there was no increase of sodium-fluorescein concentration in drug-treated mice in comparison to controls.
Mannitol was injected intraperitoneally at a range of doses (25% solution of mannitol in saline, volume from 200 to 600 µl) to mice 15 mins after intraperitoneal injection of sodium-fluorescein. Controls were injected intraperitoneally with the same volume of saline. 30 min after the treatment, mice were processed as described above. Mice injected with 300, 500, and 600 µl of mannitol solution exhibited significantly increased concentrations of sodium-fluorescein in brain tissue. The treatment had no side effects on the mice.
Once we identified mannitol as promising BBB opener, we tested if repeated treatment has any effect on the intensity of BBB permeation and any side effects on mice. Groups of mice were treated with (i) a single dose of 500 µl of 25% mannitol, (ii) with 500 µl of 25% mannitol every second day for one week, and (iii) with 500 µl of 25% mannitol on a daily basis for one week. Control mice were treated with 500 µl of saline on a daily basis for one week. At the last day, all mice were processed as described above and the BBB permeability was assayed. Based on our results, we can conclude that repeated treatment on mice with mannitol has no effect on the intensity of BBB permeability versus mice treated with only one dose. In all groups, the fluorescence signal in the brain was significantly higher in the treated mice in comparison to controls. Moreover, the mice exhibited no side effects after the repeated treatment.
Next, we compared three dosage regimens; i.e. one dose, two doses and three doses per day. Previously, we identified that 500 ul i.p. of 25% mannitol in saline is the most effective way how to increase (temporarily) the permeability of the BBB. We found that administration of mannitol once a day opens effectively BBB; this can be repeated every day without any loss of the activity of mannitol and without any side effects. However, administration of mannitol two- or three-times a day causes severe side effects or death of the treated mice.

WP4 Task 2: Evaluation of combination therapy for rabies infections.
Part of the work that VUV performed under this task falls under confidential information.
We tested if mannitol treatment has any effect on the development of RABV-infection in mice. The Silver-haired bat rabies virus (SHBRV-18) provided by the Erasmus MC grown on N2a cells. The titer of the virus was determined on BHK-21-C13 cells and calculated with the Karber method to be 106.8 TCID50/ml. We found, that administration of mannitol once a day (500 ul i.p. 25%) had no effect on the development of rabies in mice.
Państwowy Instytut Weterynaryjny National Veterinary Research Institute (PIWET, partner 4)
Task 2.2 Inhibition of detrimental host response
Approval for experiments in animals from Animal Ethics Commission obtained on 28 March 2014. Individually ventilated cages purchased and installed in the BACL3 area. Protocols for experiment prepared, discussed and accepted, strains of rabies virus were obtained from partner 1, multiplied, titrated – ready for use in the experiments. Drugs that we identified were: TNF-α, IL-6, MAPks inhibitors identified and purchased. Disposables - tubes, syringes, needles, gloves etc. and two pieces of laboratory equipment – PCR and deep freezer were purchased. The experiment was performed in January 2015.
Simultaneously partner 1 and 8 performed experiments with inhibitors of pyroptosis. TNF-α inhibitor (Remicade) significantly extended the median survival time of treated mice by 1.5 days compared to non-treated virus controls. MAPk inhibitor (Sorafenib/Nexavar) significantly extended the median survival time of treated mice by 1.5 days compared to non-treated mice. IL-6 inhibitor (Roactemra) did not influence the survival time of treated mice versus non-treated controls.
The results of the WP2 pointed out to Remicade and Sorafenib as the potential drugs influencing the survival of mice infected with rabies virus. These drugs along with antiviral drugs suggested by the results of WP1 and WP2 generated by other partners of the consortium and after the detailed discussion on the GA meetings and the approval by SAB were used in WP4 experiments.

WP4: Task 4.2: Evaluation of combination therapy for rabies infections
Part of this work from PIWET falls under confidential information.
The results of the experiments point out to the significant extension of survival time of mice treated with all drugs as listed above. Also the titer of rabies virus in the brain was significantly lower in medicated mice than in control mice. The therapy slow down the infection and this may be the indication for the experimental treatment of patients showing signs of rabies.
In a second set of experiments under this task, Partner 4 tested the combination of four compounds given either at 4 hours prior to infection or 48 hours or 96 hours post infection with a sublethal dose of SHBRV-18 virus. Significantly increased survivorship was observed with the combination of these four compounds when treatment was initiated at any time point tested. These data make a significant contribution in future decisions for treating rabies patients.

Partner 5 Szent István University (SZIE, partner 5) and Magyar Tudományos Akadémia- Magyar Tudományos Akadémia Agrártudományi Kutatóközpont (MTA-ATK, partner 6)

The two partners worked together for the scientific part of this project. Partner 5 provided the expertise whereas partner 6 provided the facilities and some man-power to conduct the experiments described below.

WP1 – Virus replication inhibitors
Task 1.1: Development of in vitro system for testing inhibition of viral replication
In vitro studies of ASKLEPIOS include a cell culture-based antiviral assay of certain candidate therapeutic compounds, followed by an immunofluorescent method of virus titration (FFA – fluorescent focus assay) and real-time reverse transcription PCR (qRT-PCR) assay to evaluate the viral loads in the supernatant samples of cell cultures. Therefore, the cell culturing system for two different cell lines (mouse neuroblastoma – N2A, and baby hamster kidney – BHK21 cells) was established. The cell lines were obtained from collaborating partners (partner 1 and 2); optimal conditions and equipment for large-scale propagation and storage were determined. For the antiviral assay, N2A cells were used, which were infected with rabies virus (RABV) using an adsorption method. For back-titration of the supernatant samples, both cell lines were found appropriate; in case of BHK21 a suspension method of virus titration was used. For in vitro experiments the CVS-11 (Challenge Virus Standard) RABV strain was chosen, which was received from partner 2 and was later successfully propagated in N2A cells.
The experimental setting for FFA tests were based on the publication of Chopy and al. (J. Neurovirol. 2011, 17:353-367) and the system was tested with type I interferons. There was no antiviral effect observed, which was due to the too high MOI. After excessive discussion with collaborating partners (on the first annual general meeting) the protocol was modified and a new SOP with lower MOI was established by partner 2 and distributed to all partners.
For qRT-PCR tests a differential TaqMan assay for RABV RNA detection was used. The system was tested with CVS-11 and with cloned and expressed RABV RNA (provided by partner 1) using three different thermocyclers and RT-PCR kits. The assay was optimized for later experiments.

Task 1.4: Testing type-I IFNs, Viperin and MAP kinase inhibitors for viral replication
The aim of task 1.4 was to test molecules with possible inhibitory effect against RABV replication in vitro. As candidate compounds type-I mouse interferons (IFN-α and -β), a MAP kinase inhibitor (sorafenib) as well as different antivirals provided by partner 7 (ribavirin, favipiravir – T-705, FDA-1) were chosen.
Cytotoxicity Detection Kit Plus® (Roche) was used to determine the non-cytotoxic concentrations of each compound on N2A cells (6.25×104 cells/ml). The highest concentration of the drugs that did not show high cytotoxicity were the following: Ribavirin and FDA-1: 10 μg/ml; T-705: 100 μg/ml; mouse interferon-α and -β: 10 IU/ml; Sorafenib: 50 μmol/l (in form of sorafenib-tosylate).
In the antiviral assay N2A cells were treated with non-cytotoxic concentrations of the different compounds (IFN-α and -β, sorafenib, ribavirin, T-705, FDA1) 1h after inoculation with CVS-11 (500 TCID50/ml). After 48h incubation time the supernatants were collected and titrated using fluorescent focus assay. qRT-PCR was also performed to determine viral RNA loads.
The studies showed that FDA1 has minimal inhibitory effect on RABV multiplication: viral titres were reduced only by 1 log at the highest applied concentration (10 μg/ml). Therefore, this compound was excluded from later experiments.
All of the other compounds showed concentration-dependent antiviral effect. T-705 has a moderate effect with ~2.75 logs decrease in viral titres in the highest concentration (10 μg/ml). Ribavirin reduced RABV titre by approximately 3.75 logs in the highest dose, 10 μg/ml. In case of type-I interferons, IFN-β was more effective: ~5.75 logs decrease was observed when applied in 10 IU/ml (highest concentration), while IFN-α caused only ~3.5 logs decrease. The inhibitory potential of sorafenib was comparable to that of IFN-β, with a similar, ~5.75 logs decrease in RABV titres. In lower concentrations the inhibition was weaker in all compounds. In case of T-705, the lowest concentration used in the experiment, 0.01 μg/ml caused no significant change in virus replication.
According to these results, the IC50 values for the different compounds are the following:
IFN-α: 7.05 IU/ml
IFN-β: 1.22 IU/ml
Ribavirin: 14.32 μg/ml
T-705: 792.16 μg/ml
FDA1: 26,580.17 μg/ml
Sorafenib: 1.6 μmol/l
All of the data mentioned above are results of FFA tests on the supernatants collected form the plates of the antiviral assay. qRT-PCR tests were also performed on the same samples, with similar results, showing the same tendency; however, the RNA titres determined with PCR method were slightly higher than the titres measured with FFA. This is caused by the fact that only complete, functional virions can be detected by titration, while in case of PCR viral RNA present in the sample (including mRNA and RNA of incomplete virions) can be amplified.
In conclusion, sorafenib and IFN-β have the most pronounced antiviral effect against rabies infection in vitro. Ribavirin, IFN-α and T-705 are still effective, but FDA1 cannot significantly reduce RABV replication.
Viperin was studied by partner 1 and 8 because strict GMO regulation in Hungary did not make possible to conduct research on that subject for SzIE and MTA-ATK (partner 5 and 6).

Task 1.5: Exploring combinations of siRNA, aptamers, Type-I IFNs, Viperin, MAP kinase inhibitors and small molecules identified in the SILVER consortium
Combinations of IFN-β, sorafenib, ribavirin and T705 were analysed for their effect of interfering with RABV replication in N2A cells. Every combination consisted of two drugs, and two concentrations were used of each compound. Therefore, 24 different combinations were included in task 1.5 experiments. The experimental protocol was the same as described for task 1.4.
According to FFA tests none of the combinative treatments resulted in a synergistic antiviral effect. However, additive effect was found when IFN-β was combined with sorafenib, ribavirin or T-705. The most pronounced inhibition was caused by the combination of the higher concentrations of IFN-β and sorafenib (1 IU/ml and 5 μmol/l) with more than 6 logs decrease in RABV titres. In other combinations there was no improvement in the reduction of virus titres compared to the effect of the components alone. The qRT-PCR results show the same trend as described for FFA, but the calculated titre values are notably higher (see task 1.4).

WP4 – In vivo testing of therapeutic molecules
Task 4.2: Evaluation of combination therapy for rabies infections
Two sets of mouse experiments were conducted at the BSL-3 laboratory and rodent facility of partner 6 (MTA-ATK). The first set of experiment was based upon the “Christmas tree” concept, which means that every compound that showed promising effect in WP1 and WP2 was included in the therapeutic combination, and BBB opener (mannitol) was also used. The results of these experiments fall under confidential information.

University of Leuven (KU Leuven, partner 7)

KU Leuven was in the years preceding ASKLEPIOS involved in another EU FP7 consortium named SILVER. In this consortium novel small molecule antivirals were discovered for a variety of viruses including molecules that inhibited rabies virus replication. At the start of ASKLEPIOS partner 7 was able to synthesise and purify these molecules at a ~10 mg scale after which they were shipped to several of the ASKLEPIOS consortium partners. These partners investigated if the anti-rabies activities of these molecules could be confirmed on the rabies strains set-forward within ASKLEPIOS (Duvenahage virus (DUVV) and European bat lyssaviurses (EBLV) 1 and 2). Despite the many antiviral assays performed at the different sites of the partners, no significant antiviral activities of any of the molecules from the SILVER program could be detected. This indicates that the rabies virus inhibitors discovered in the SILVER program are very specific to the rabies strain CVS which was use in this program. Later in this discovery program partner 7 set-up in-house antiviral assays with the rabies strain ERA to further look into this issue. Also these investigations showed that the anti-rabies molecules discovered during SILVER must be very specific for strain CVS as no significant activity on strain ERA could be found. These results are very important for future anti-rabies virus work at partner 7 as they have now a clear view on the antiviral spectrum of their molecules and on how they should further develop them to broad spectrum anti-rabies virus inhibitors.
During the continuation of ASKLEPIOS partner 2 (APHA) discovered the anti-rabies activity of Favipiravir (T-705) in vitro. Partner 7 had already a lot of experience with the antiviral activity of this molecule on other viruses including in animal models. Partner 7 was therefore able to collect in a timely manner sufficient amounts of Favipiravir for animal studies and to exchange all information needed on formulation and pharmacokinetics of this molecule. During the ASKLEPIOS program many batches of Favipiravir were purchased by partner 7 and shared with several of the partners to be used in combinations with other molecules. Also the formulation and use of Favipiravir was exchanges with the different partners on several occasions. In this way, the in vitro discovery of partner 2 on the anti-rabies activity of Favipiravir could very rapidly be followed up by the full consortium in animal experiments as a single drug or in combinations with additional therapeutic approaches.
During the ASKLEPIOS program several of the partners were in need of infliximab, a TNF-alpha inhibitor, to be used in antiviral testing. Also here, partner 7 contributed significantly by coordinating the purchase and logistics of shipping of this molecule to the different partners.

Artemis One Health (Artemis, partner 8)

WP2: Task 2.1 Inhibition of pyroptosis
Approval to conduct animal experiments obtained from Animal Ethics Committee of EMC on 27 March 2014. As agreed during the kick off meeting in Rotterdam (January 2014), partner 1 grew and distributed sufficient amount of virus stocks for the in vivo experiments of WP2. For this reason Eastern European fox virus obtained from Dr. Thomas Muller whereas Silver-haired bat rabies virus (SHBRV-18) was available in the lab of partner 1. Eastern European fox virus and SHBRV-18 were grown, titrated and distributed to partner 4.
At the consortium meeting held in Prague (October 2014), under the strong recommendations of the SAB, it was decided that all host response inhibitors will be tested against one virus only: SHBRV-18. Animal experiments were conducted in Pulawy, Poland, at the laboratory of partner 4 in the presence of investigators from both EMC and ARTEMIS. Animals were infected intra muscularly with SHBRV-18 virus and on day 5, 7 and 9 post infection were given CASPASE-1 (CASP-1) inhibitor or IL-1β inhibitor intra-cranially.
CASP1 inhibitor significantly extended the mean survival time of treated animals by 1.5 days compared to non-treated virus infected control animals. IL-1β inhibitor did not have an effect on the survival time of treated mice compared to the non-treated controls. Viral titers were recovered from all parts of the CNS, similarly with the titers recovered from mice treated with CASP-1 or IL-1β or the non-treated controls.
Next we measured mRNA of several markers involved in the pyroptotic pathway, namely CASP-1, IL-1β, IL-18 and TNF-α. The effect on mRNA expression was more pronounced in mice treated with IL-1β inhibitor, compared to mice treated with CASP-1 inhibitor. This differential mRNA expression (mostly seen as down-regulation of the genes tested) was seen both in comparison with non-infected treated mice and with infected, non-treated mice.

WP3 Task 4: Development of siRNA delivery system
A well-described challenge in treatment of neuro-degenerative diseases is to get pharmaceutical compounds across the BBB. Artemis investigated the possibility to use exosomes to get effective compounds across the BBB. The working hypothesis was that exosomes produced by cells residing in the brain are more likely to fuse with neurons. Therefore, we have used different cell lines to identify the ones that produce more exosomes; microglia, astrocyte en neurons. To this end, the respective cell lines were cultured using medium without FBS, medium with normal FBS and medium containing exosome-free FBS. Subsequently, we have compared different methods:
1. Ultracenfugation to pellet exosomes
2. Optiprep gradient for isolation of exosomes
3. Size exclusion chromatography
4. Optiprep gradient for isolation of exosomes followed by size exclusion chromatography
Presence and purity of exosome preparations were analysed by electron microscopy, detection of exosome-specific miRNA, detection of exosome-specific proteins.
Results: None of the methods employed resulted in pure exosome preparations. In addition, all the exosome preparations contained a lot of aggregates, with methods 3 and 4 resulting in relatively purer and less aggregates. However, the yield was extremely low. Because none of the methods resulted in exosome preparations that were pure enough and with enough yield for further experiments, it was decided by the consortium to abandon this task at the end of year 2. In addition, exosomes were envisaged as delivery vehicles for aptamers and siRNA, two strategies that were also abandoned because of unsuccessful results in the fist two years. Despite the negative results with the exosomes, valuable knowledge was obtained that would allow further optimization of effective procedures for exosome isolation. Considering the potential use of exosomes for treatment of for instance neuro-degenerative diseases, it is of paramount importance to further build on the preliminary results of ASKLEPIOS by investing more efforts into the field of exosomes.

WP4: Task 4.2: Evaluation of combination therapy for rabies infections
ARTEMIS worked towards the deliverables of this task in collaboration with EMC. Two sets of experiments were done under this task. The results of these experiments fall under confidential information.
ARTEMIS measured the effect of treatment on the mRNA expression of relevant (neuronal) markers. We observed some differences between virus infected and treated groups compared to treated only or virus only groups both in the mRNA levels of pyroptotic and apoptotic markers. In general, these differences were more pronounced (or even statistically significant) in the mice that did not receive mannitol with treatment. As expected, all pyroptotic markers that we tested (CASP-1, IL-1β, IL-18, PYCARD) were significantly different in treated groups (without mannitol) compared to their treatment control counterparts or virus control mice. Some apoptotic markers were also differentially expressed in these mice (CASP-3 and CASP-8).

Institute Pasteur (IP, partner 9)

During the course of the Asklepios Program, Institut Pasteur moved from the position of SAB member to the position of Partner n°9 (IP) in the consortium (decision of the Executive Board / General Assembly in Rotterdam on 10th January 2014). This “entry” occurred with 6 months delay (summer 2014) but the official signature was only completed in April 2015.
Partner n°9 (IP) is developing since about 10 years Antiviral Strategies against Rabies with the following logic in mind:
1. Searching for inhibitory opportunities at both viral and cellular side.
2. Targeting different steps of the viral cycle to ultimately synergise antiviral effects.
3. Following in parallel (i) a random approach using high-throughput screening methods on infected cells, pseudotypes or minireplicons in order to screen separately the different steps; (ii) a cognitive approach to deduce precise targets for inhibition from the analysis of the functional interactions required during viral cycle. These two approaches are symmetrical and complementary: the random approach can lead to identify the inhibited function and by the way can contribute fundamental knowledge on the RABV infection itself; the cognitive approach offers opportunities based on fundamental/structural data.
4. Looking for large spectrum antivirals since they would be more attractive for further development by industrial partners. Indeed, rabies remains a neglected disease, preventable if post-exposure treatment is administrated shortly post-infection. Both features do not play in favour or an antiviral strategy. However (i) the long period between the biting (an obvious risk event) and the symptoms (2 months in average) and (ii) the lack of any efficient therapy once the symptomatic period is reached, leave opportunity to develop an efficient antiviral response, even if reaching the virus while travelling in neurons is not an easy task.

Because of the late entry date in the Asklepios program, Partner 9 (IP) was associated and collaborated with the tasks from Partner 2 (APHA) in the context of WP1 (Virus Replication Inhibitors) even if more targets than only the replication step were considered. Active interactions were established with APHA through exchange of material and personal. In particular, K. Mansfield AHVLA visited IP in July 2014.

Task 1.2 Development of siRNA (APHA) and Institut Pasteur (IP)
Inhibition of cellular partners of rabies proteins by shRNA:
Partner n°9 (IP) is exploring since a long time the interactions that Lyssavirus proteins establish with cellular partners. This approach may also serve for antiviral purpose targeting host functions. Four partners of the L polymerase (confidential at this stage) are studied both to understand their role during the viral cycle, and to evaluate their interest in a antiviral strategy. Specific shRNAs targeting each cellular partners have been stably transduced in 293T cells using an HIV-derived vector to inhibit the relevant protein/mRNA expression. At least one partner may be involved in the microtubule-dependent routing of the viral RNP from the Negri bodies to the cytoplasmic membrane where RABV buds. This candidate cellular protein was previously involved in Influenza virus replication what may promote its interest in a large spectrum antiviral approach, currently under investigation.

Task 1.3 Development of RABV-neutralizing aptamers (APHA) and antiviral dermaceptin peptides (APHA and IP)
Blocking entry and post-entry: dermaseptins
Dermaseptins are small amphiphiic peptides, about 30 aa in size, secreted by the skin of amphibians. They are easy to synthesize and their antimicrobial activity has been broadly demonstrated against bacteria, yeast, fungi and intracellular parasites. Concerning viruses, essentially enveloped particles such as HIV and HSV have been explored due to the intrinsic capacity of dermaseptins to shape into alpha helixes, then to self-associate for creating “holes” into membranes what is probably the main expression of their inhibitory effect.
We have investigated the anti-rabies activity in vitro (BSR cells) of two dermaseptins from Phyllomedusa sauvagei : S3 (30aa) and S4 (28aa). S4 being more active against RABV infection than S3, we have further analysed the molecular basis of its activity by punctual mutations or deletion. This demonstrated that the 5 NH2-aa of S4 are crucial for its inhibitory potential. Assuming that the structure in alpha helix accounts for the inhibitory activity, we have introduced a point mutation in position 4 (S4M4K) to strengthen the alpha-helix shape. This resulted in a significantly reduction in cytotoxicity without significant loss in inhibition. However, the inhibition was totally lost when a penetratin (Tat peptide) was synthesized in tandem to facilitate the dermaseptin penetration into the cell. This demonstrates that dermaseptin S4 and derivatives essentially inhibit the early steps of viral entry (binding) at the membrane level. Thus, they can represent a possible alternative to rabies Immunoglobulins (RIGs) for the treatment of rabies exposures. It must be outlined that RIGs are recommended by WHO for all category III rabies exposures (>90% of the cases). However, the main vaccine producers have stopped their production, waiting to develop more up to date and less allergic alternatives such as humanized monoclonal antibodies. However, at the present time, this attitude has resulted in a dramatic shortage and increased cost of RIGs worldwide, particularly in the developing countries where rabies is almost exclusively present. In this perspective, alternatives must be proposed and we explored the in vivo potential (in Swiss female mice) of our two best candidates S4 and S4M4K.
To mimic a post-exposure treatment, mice were first inoculated at the same gastrocnemian muscle area with a lethal dose of RABV, then, 1h later, with 100μg or 200μg of S4 or S4M4K. Interestingly, S4M4K appeared more efficient than S4 at inhibiting RABV in vivo, probably because of a better uptake than S4.

Blocking the replication complex: Arbidol and Ribavirin:
Along the Asklepios program, Partner n°9 (IP) has tested in vitro several promising molecules active in other viral models or blocking relevant cellular pathways, as potential anti-rabies candidates. We used (1) several cell lines or minireplicon: hamster fibroblasts (BSR), rodent (Neuro-2A) or human (SK-N-SH) neuronal cells, (2) different dynamics of delivery, pre-, co- or post-infection.
One molecule, Arbidol, drug licensed in Russia and China against Influenza virus was compared to Ribavirin, well known as a large spectrum inhibitor of RNA viruses. In combination at different doses these two molecules showed a clear synergistic effect by the Bliss independence model and the Loewe additivity model (95-100 % inhibitory effect against RABV). In collaboration with partners out of the Asklepios consortium, combination between other molecules (targeting RABV or cell pathways) were tested in the purpose of optimizing the antiviral potential of the cocktail.

Blocking the transcription/replication complex: peptides mimicking the P-N°-L interaction domain.
All Negative Strand RNA Viruses (NSRV) share a similar transcription/replication complex: the template (RNA genome intimately linked to the nucleoprotein N) is transcribed/replicated by the polymerase (L polymerase + cofactor P). This complex is then an ideal target to develop a large wide-spectrum antiviral strategy, efficient on both neglected (RABV) and more bankable (i.e. RSV) viruses.
The P protein plays a particular key role for a least two functions: (1) it mediates the physical link between the L polymerase and the N-RNA template; (2) its acts as a chaperone, forming N°-P complexes that prevent N° from binding to cellular RNA and remains it available for specific encapsidation of the nascent viral RNA during replication. P protein is thus a target of choice for an inhibitory strategy of the replication complex. Using a combination of 2-hybrid and structural approaches, we have shown that the N-extremity of P is particularly important: (1) residues 4-40 are binding to N° (Mavrakis et al, 2004); (2) residues 40-70 (extended up to 100) encampass the main L binding domain (Castel et al, 2009).
We used these fundamental data to develop inhibitory peptides mimicking the N-extremity of P. When transfected or delivered (in tandem with a Tat peptide) into neuronal (SK-N-SH) or fibroblastic (BSR) cells infected with RABV or hosting a RABV minireplicon, these peptides inhibit infection and this inhibition is valid for different Lyssaviruses.
The recent crystal structure of the P-N° complex, allowed us to design peptide segments better fitting the functional pockets promoting interactions. These peptides (about 20 aa in size) have been synthesized in tandem with the penetratin (TAT-sequence) and their anti-rabies effect in vitro in being evaluated. The best candidates for inhibition will be used as models for a peptido-mimetic approach to design simpler molecules with the same inhibitory effect.
University of Veterinary Medicine Budapest (UVMB, partner 10)
In the course of ASKLEPIOS, the legal entity represented by partner 5 (SziE) changed into UVMB. All data collected after 1 June 2016 from partner 5 are officcially the contribution of partner 10 to ASKLEPIOS.

Task 4.2: Evaluation of combination therapy for rabies infections (All partners)
Part of this work of UVMB falls under confidential information.
There was no detectable toxic effect of the treatment, every clinically healthy mouse maintained its weight or gained weight. There was no mortality in the uninfected (drug control) groups. Due to the lower dose of virus inoculated, a number of mice survived rabies infection. These animals were terminated at the end of experiment, on day 28. Out of a group number of 13, there were 6 survivors in the untreated (virus control) group, 8 in the 4h pre-infection treatment group, 9 and 10 in the 48h- and 96h post-infection treatment groups, respectively. The difference in survival between the virus control and the treated groups is statistically significant (p=0.002) if the treated groups are analysed together. This shows that the treatment with the combination of the three compounds listed above can increase the survival in rabies-infected mice. The timing of the start of treatment is not clearly correlated with survival rates; interestingly in groups where the drugs were administered from a later time point, slightly more animals survived (at the group numbers of 13 this difference was not significant). According to qRT-PCR and IHC results from CNS samples of experimental mice the viral loads are not notably different between experimental groups in animals that succumbed to rabies (e.g. similar RNA copy numbers can be measured in mice belonging to treated and virus control groups). In case of surviving mice that were infected, 4 mice were found positive for RABV RNA in the brain, but at a much lower copy number.
Blood samples were taken from all surviving animals before their termination, which makes possible the detection of anti-RABV neutralizing antibodies in the serum of mice. Hence, virus neutralization assay (FAVN) is planned to continue the study.
We can conclude that combination treatment with inhibitors of the detrimental host response can increase the survival chance in case of RABV infection. Based on results of other collaborating partners, with the addition of antivirals or HRIG the combination can be even more effective, but possible toxicity issues need to be considered carefully.
Potential Impact:
Rabies continues to cause more than 60,000 deaths every year in rabies-endemic countries. Where populations are aware of the disease, rabies is one of the most feared infections as its association with horrific disease symptoms and death have cemented it’s place in human history. Importantly, rabies is one of the few pathogens for which there is no treatment following the onset of clinical disease, with the outcome of infection being death in almost 100% of cases. This factor drives societal fear of infection and has given rabies virus a global reputation that overshadows other diseases. Frustratingly, rabies virus can be readily combatted using simple tools and procedures following a suspected exposure. Indeed, awareness of the disease and suitable actions to take following potential exposure can vastly increase the likelihood of preventing the virus from establishing infection. Sadly, this knowledge is often lacking in areas where the virus is endemic with the result that needless deaths occur. Alongside, simple wound washing procedures that can greatly reduce the potential for development of disease, vaccines and post exposure tools are also available to combat the virus. However, whilst vaccines for pre-immunisation and the combinatory approach of vaccination and immunoglobulin administration for post-exposure treatment are available, no options have been identified that can reduce or prevent rabies virus replication once clinical disease has initiated. The search for effective antiviral molecules to treat those that have already developed clinical disease associated with rabies virus infection is considered one of the most important goals in rabies virus research.
The work performed here has identified two molecules that may serve as options for treatment of humans experiencing clinical disease following interaction with a rabid animal. T-705 may well have a future role in treatment options following the development of clinical disease although experimentation in non-human primates is warranted to assess this further. Furthermore, the plant derived ScFv molecule assessed may have a future role in both treatment of clinical disease as well as potentially in post exposure activities where it may serve as an alternative to rabies immunoglobulin. Alongside the identification of other molecules through the work of the remaining members of the consortia, the outputs from this project, following further evaluation may well have profound societal implication for the treatment of this devastating disease.

WP1 – Virus replication inhibitors
Different molecules were tested for their antiviral activity against rabies infection in cell cultures. Our results demonstrate that Type-I interferons, the MAP kinase inhibitor sorafenib, ribavirin and favipiravir (T-705) can interfere with RABV multiplication in cell cultures in non-cytotoxic concentrations. Interferons and ribavirin were already used in human patients with rabies, and were found unsuccessful. However, when IFN-β is combined with ribavirin, favipiravir or Sorafenib, an additive inhibitory effect can be detected. In case of other diseases (e.g. chronic hepatitis C) interferons and ribavirin are usually administered in combination therapy. Our data indicate that the same approach should be considered for rabies treatment a well. The antiviral effect of T-705 against RABV is already published but has never been used in patients. Until now, there was no information about the in vitro anti-rabies activity of sorafenib, and its pronounced effect in combination with IFN-β is even more promising. These compounds can be candidates for a future rabies therapeutic protocol, possibly together with inhibitors of proinflammatory host responses. Other combinations and other compounds could be tested based upon the results and utilizing the methods of ASKLEPIOS, e.g. different MAP kinase inhibitors, new antiviral compounds (mainly those which are active against other negative sense single-stranded RNA viruses). If new compounds or combinations are found to be effective against RABV replication, they should be tested in animal models to find out whether they are active under in vivo conditions, and if there are any toxicity or pharmacokinetic/pharmacodinamic concerns.

WP2 – Host response inhibitors
Partner 1, EMC and partner 2 (APHA) previous to the ASKLEPIOS project were also involved in another FP-7 funded project (ANTIGONE) in which pathogenic mechanisms of lyssavirus infections were identified by studying global mRNA profile of lyssavirus infections. The results of the studies of EMC led to the formulation of the main objective of ASKLEPIOS, namely, the identification of molecules that inhibit not only virus replication but also host responses that have a detrimental effect and are partly responsible for the severity of rabies encephalitis. Within WP2, several host response inhibitors in an in vivo model of rabies infections were tested. Compounds (some of them are already registered drugs for human use) that could inhibit detrimental host responses during rabies and prolong survivorship of experimentally infected animals treated with the different host modulators were identified. Despite the fact that survivorship did not increase, the results of these studies paved the way for the subsequent design of combination therapy and the experiments performed under WP4. In addition, together with the results obtained within ANTIGONE project, within this WP, Partner 1 demonstrated that pyroptosis is a crucial pathway in the pathogenesis of lyssavirus infections and intervention strategies aiming on this pathway may have a positive effect on the course of clinical rabies.

WP3- Blood brain barrier openers
Here, we identified mannitol as a compound, which can effectively increase permeability of the blood-brain barrier and facilitate the delivery of therapeutical agents into the CNS. However, mannitol treatment has no effect on the development of rabies neuroinfection in mice. Therapy of RABV-infected mice with a combination of RABV-specific antibodies and immunomodulatory compounds, when initiated 4 h before infection, has a significant positive effect on mouse survival and the therapy significantly reduces virus titers both in brain and spinal cord. These results provide a platform for a rational development of novel therapeutical approaches against rabies in humans.

WP4 – In vivo testing of therapeutic molecules
Our in vivo experiments clearly demonstrated that the combinatory approach of rabies therapy has a good potential. Using inhibitors of detrimental host responses the survival chances of rabies-infected mice were improved. Their anti-RABV effect may broaden the timeframe when post-exposure prophylaxis can applied successfully. Together with the findings of recent studies about rabies pathogenesis and the molecular background of rabies encephalitis, the knowledge acquired during ASKLEPIOS may provide essential information to scientists and clinicians facing future cases of human rabies. The compounds and combinations that found to be effective during the project should be considered among the possible therapeutic and post-exposure options in human rabies. However, further extensive research should follow the ASKLEPIOS project to find other molecules, biologicals that interfere with rabies pathogenesis and reduce the damage to neural tissues. The use of combinations studied in the current project should be optimized considering the timing of administration and best pharmacokinetics (e.g. sufficient therapeutic concentration in the brain).

Overall impact of ASKLEPIOS
The innovative aspect of ASKLEPIOS lies in the “two-hit” approach in which both the virus and the host are tackled as a novel intervention strategy to improve clinical rabies and increase survivorship of this otherwise lethal disease. All the partners of the consortium significantly contributed to the successful completion of all tasks described. In addition, the results obtained within ASKLEPIOS show a lot of potential for the identification of novel treatment strategies. In conclusion, despite the fact that no “cure” against rabies was found during the three years of ASKLEPIOS, we have made a significant step forward towards treatment of rabies. Most of the compounds identified under ASKLEPIOS to be potentially beneficial are already registered drugs for human use; implementation of such novel intervention strategies is therefore a step further.

List of Websites: