SiRNA Inhibits Replication of Langat Virus, a Member of the Tick-Borne Encephalitis Virus Complex in Organotypic Rat Brain Slices

Tick-borne encephalitis virus is the causative agent of tick-borne encephalitis, a potentially fatal neurological infection. Tick-borne encephalitis virus belongs to the family of flaviviruses and is transmitted by infected ticks. Despite the availability of vaccines, approximately 2000–3000 cases of tick-borne encephalitis occur annually in Europe for which no curative therapy is available. The antiviral effects of RNA mediated interference by small interfering RNA (siRNA) was evaluated in cell culture and organotypic hippocampal cultures. Langat virus, a flavivirus highly related to Tick-borne encephalitis virus exhibits low pathogenicity for humans but retains neurovirulence for rodents. Langat virus was used for the establishment of an in vitro model of tick-borne encephalitis. We analyzed the efficacy of 19 siRNA sequences targeting different regions of the Langat genome to inhibit virus replication in the two in vitro systems. The most efficient suppression of virus replication was achieved by siRNA sequences targeting structural genes and the 3′ untranslated region. When siRNA was administered to HeLa cells before the infection with Langat virus, a 96.5% reduction of viral RNA and more than 98% reduction of infectious virus particles was observed on day 6 post infection, while treatment after infection decreased the viral replication by more than 98%. In organotypic hippocampal cultures the replication of Langat virus was reduced by 99.7% by siRNA sequence D3. Organotypic hippocampal cultures represent a suitable in vitro model to investigate neuronal infection mechanisms and treatment strategies in a preserved three-dimensional tissue architecture. Our results demonstrate that siRNA is an efficient approach to limit Langat virus replication in vitro.


Introduction
Tick-borne encephalitis virus complex represents a group of closely related viruses endemic in Europe and Asia causing serious neuroinfections and hemorrhagic fevers [1]. Tick-borne encephalitis virus (TBEV) belong to the Flaiviridae family and can be divided in three main subtypes, the European, the Siberian and Far Eastern subtypes [2]. TBEV are transmitted to humans by the bite of infected ticks and, in rare cases, through the consumption of infected unpasteurized milk [3,4]. Following an incubation period of 3-8 days after a tick bite, the virus replicates locally in epidermal dendritic cells and spreads via lymph vessels to the blood stream where a short but significant viremia occurs and extraneural tissues are infected. It is during this viremic phase that the virus crosses the blood-brain barrier and invades the central nervous system (CNS) by a still unknown mechanism. Once the virus has reached the CNS, active infection causes inflammation, lysis of cells and cellular dysfunction [5,6]. Clinical manifestations of TBEV infections typically follow a biphasic course. The first viremic phase is characterized by flu-like symptoms such as fever, headache and muscle pain. After 3-7 days of an asymptomatic phase, 20-30% of the patients develop the second meningoen-cephalitic phase with neurological disorders of varying severity [5,7,8,9]. The most common clinical feature of TBE patients is ataxia followed by paresis or paralysis of one or more extremities [10].
Two efficient formalin-inactivated whole virus vaccines are available in Europe and provide a high degree of protection against the disease. Nevertheless more than 2500 TBE cases have been registered yearly in Europe in the past ten years and a continuous increase in TBE morbidity was observed in the last years [11,12]. Up to today no specific therapeutic options are available for flaviviral infections and an effective therapy for TBEV infection would be highly desirable [13]. Results from in vitro and in vivo studies indicate that therapeutics based on RNA interference (RNAi) could be effective against viral infections, and small interfering RNA (siRNA) molecules are promising candidates for future clinical applications [14,15,16,17]. Indeed, several RNAi-based antiviral drugs are currently being tested in clinical trials [18,19,20,21].
RNA interference is a conserved post-transcriptional gene silencing process which leads to the specific degradation of RNA within the cytoplasm of eukaryotic cells [22,23,24]. Long doublestranded RNA molecules are initially processed by the enzyme Dicer into 21-25 nucleotides long small interfering RNA. In the cytoplasm siRNAs are incorporated into the multiprotein RNAinduced silencing complex (RISC) which results in sequence specific association and degradation of the complementary mRNA [24,25]. SiRNA mediated intervention strategies take advantage of this conserved mechanism by artificially introducing siRNA molecules into the cytoplasm which ultimately leads to the degradation of target RNA molecules. It was shown that the application of siRNA molecules with sequences complementary to viral genomic RNA or RNA replication intermediates allows to significantly reduce the number of virus progeny in infected cells. Using this strategy the replication of human pathogenic flaviviruses such as West Nile virus (WNV), Dengue virus (DENV) and Japanese encephalitis virus (JEV) was successfully inhibited both in vitro and in vivo [14,26,27,28,29,30] and recently inhibition of TBEV by RNAi was demonstrated in human embryonic kidney cells (HEK293T) [31]. Sensitive target sequences were identified within the structural (Capsid (C), membrane protein (M) and envelope (E)), non-structural (mostly NS5 and NS3) and the 39 untranslated region of the flavivirus 11 kb open reading frame (ORF).
We used organotypic hippocampal brain slice cultures (OHCs) infected with Langat virus (LGTV) as a model to analyze the effectiveness of RNAi in the inhibition of the virus replication. OHCs is a well established model based on in vitro cultured brain slices which was adapted for this virus encephalitic model from previous work [32]. Organotypic cultures offer unique advantages over other in vitro models, in that they reproduce important aspects of the in vivo situation. Organotypic cultures retain the three-dimensional tissue architecture of the brain with a preserved cellular composition including neurons, microglial cells and astrocytes [32,33]. The use of OHCs allowed the culture of Langat virus in an in vitro system reproducing some aspects of TBEV infection of the brain.
LGTV is a member of the tickborne encephalitis virus complex sharing high nucleotide homology with TBEV. It has an attenuated virulence for the human host and was used as a vaccine candidate against TBE for several years, but was abandoned due to the sporadic occurrence of encephalitis cases (1:18000) [34,35]. Because of its relative avirulence for humans it can be handled under BSL-2 conditions. LGTV retains neurovirulence for rodents and is therefore considered a suitable virus-model for the pathogenic TBEV in mice and rats [36,37,38,39]. Using Langat virus in the organotypic brain culture model we could demonstrate the effectiveness of siRNA in inhibiting TBE virus replication in a three-dimensional neuronal tissue architecture.

Evaluation of siRNA Sequences Targeting Langat Virus Genome
A total of 19 siRNA sequences (Q1-Q6 and D1-D13) targeting genes within the structural (S), the non-structural (NS) and the 39 and 59 untranslated regions (UTRs) of Langat genome (Table 1 and Figure 1 A) were tested. All siRNA sequences were evaluated for their capacity to inhibit replication of Langat virus in HeLa cells when transfected 4 h before the infection occurred. Fourteen out of the 19 specific siRNA sequences were able to induce a significant reduction of viral genome copy numbers in comparison to the nonsense siRNA used as a negative control ( Figure 1B).
Inhibition of virus replication was most efficient with siRNA molecules targeting sequences located within the structural region (Q5, Q6, D5, D12) and the 59 UTR (D3 and D8), reducing viral genome copy numbers by up to 85%. Only Q1, D1 and D10, all targeting the 39 UTR, had no effect on virus replication. A comparative analysis of all siRNA target sequences with the genomes of five members of the Tick-borne encephalitis virus complex revealed that siRNA sequences D1, D3, D8, D10 and D13 located within the 39UTR, 59UTR and the core region were strongly conserved with 1 to 2 nucleotide divergence between different viral subtypes. SiRNA sequence D3 was the only one with 100% sequence identity in all three TBEV subtypes and in the Omsk hemorrhagic fever virus (Table S1). SiRNA sequence D3 targeting 59UTR was selected for further experiments on HeLa cells and organotypic cultures.

Inhibition of Virus Replication by siRNA D3 in HeLa Cells
To determine the inhibitory effect of siRNA sequence D3 over time the replication profile of Langat virus was assessed daily during the 6 days incubation period in cells treated with specific siRNA D3 and in cells treated with nonsense siRNA as controls. While virus replication was continuous in HeLa cells treated with nonsense siRNA, the number of viral genome copies was reduced by 96.5% (1.4 logs) ( Figure 2A) and the number of infectious particles by 98.8% (2 logs) ( Figure 2B), from day 2 to day 6 post infection in cells transfected with siRNA D3. To exclude unspecific inhibition of virus replication trough an effect of the transfection reagent on cell viability we performed a LDH-based cytotoxicity assay on cells under the different treatment regimes. No significant difference in cytotoxicity was observed between transfected and infected cells and untransfected and uninfected cells demonstrating that cell viability was comparable between the four groups ( Figure S1).
We analyzed the effectiveness of siRNA on viral replication when administered after the initiation of the infection. To this end HeLa cells were first infected with Langat virus at MOI of 10, 1, 0.1 or 0.01 and 1 h later transfected with siRNA sequence D3. Compared to cells transfected with the nonsense siRNA sequence, treatment of infected HeLa cells with siRNA sequence D3 resulted in a reduction of viral genome copies by 1.6 log (MOI of 10), 2.4 log (MOI of 1), 2.4 log (MOI = 0.1) and 5 log (MOI = 0.01), respectively which corresponds to an inhibition of more than 98% for all the infectious doses. At the lowest infectious dose of 0.01 MOI the virus titer was reduced below the detection limit ( Figure 3).

Langat Virus Replication in the Organotypic Hippocampal Cell Culture Model
OHCs are an excellent model for the investigation of neuronal infection mechanisms and new treatment strategies. The ability of Langat virus to infect and replicate in OHCs was demonstrated by real-time RT-PCR and immunofluoresence ( Figure 4). Using a red fluorescent antibody, cells expressing Langat virus proteins were specifically stained.

Antiviral Activity of siRNA in Infected Organotypic Hippocampal Cell Cultures
To assess the capacity of siRNA to inhibit virus replication in a complex three-dimensional neuronal network, inhibition experiments were performed on OHCs. Transfection of OHCs with the specific siRNA sequence D3 before and after the infection led to an efficient inhibition of virus replication reducing the number of viral genome copies in OHC homogenates from 5.4610E5 to 1.7610E3 copies/ml (2.7 logs) and the number of infectious particles from 2.7610E2 to an average of 0.5 ffu/ml (4.3 logs) ( Figure 5 A and B) corresponding to a reduction of more than 99.6% as compared to nonsense siRNA treated OHCs. No

Off-target Effects
Several studies suggest that transfection of cells with siRNAs may result in the activation of the interferon pathway and affect virus replication via interferon mediated antiviral activity [40,41,42]. To investigate whether the treatment of OHCs with siRNA leads to the activation of the IFN system, expression levels of IFN-b mRNA were quantified by real time RT-PCR in noninfected OHCs treated with different siRNA sequences and compared to untreated or poly I:C treated cultures. While treatment of HeLa cells with poly I:C, used as a positive control, leds to a strong upregulation of IFN-b mRNA, none of the siRNA sequences lead to an increase in the IFN-b mRNA expression 24 h after the second transfection ( Figure S3).

Discussion
Therapeutic options for the treatment of TBE are lacking and specific therapies are urgently needed since the incidence of TBE is rising. Several studies have shown that siRNA molecules have the potential to be used as a specific therapeutic strategy against viral infections. Due to difficulties in delivery, toxicity and the stimulation of unspecific immune response few approaches were continued in vivo. For TBE virus infections detailed investigations determining optimal target sequence and inhibitory capacity for synthetic siRNA are lacking. The majority (11/19) of the 19 siRNA sequences screened in this study were capable of reducing Langat virus titer by more than 80% in comparison to nonsense siRNA when transfected into the cells before the infection (Figure 1  B). The most effective reduction in viral replication was achieved with siRNA sequences targeting the 39 UTR and the structural genes. SiRNA sequence D3 (59UTR) and Q6 (Envelope) showed the highest capacity to inhibit viral replication, reducing the number of viral RNA genome copies from 1610E6/ml (nonsense siRNA) to 5.5610E4/ml (siRNA D3) and 6.6610E4/ml (siRNA Q6) respectively, corresponding to a reduction of more than 93% (Figure 1 B). In addition to its strong antiviral activity sequence D3 is highly conserved between different members of the tick-borne encephalitis virus complex and was therefore selected for further experiments on HeLa cells and organotypic brain cultures. Using the siRNA sequence D3 we could demonstrate that the application of siRNA after the infection was still effective. In fact the inhibitory effect was comparable to the results obtained when siRNA was applied before infection. These results indicate that it is feasible, at least in the paradigm tested herein, to inhibit replication of TBEV in vitro once the infection has been initiated.
For further in vitro analysis we developed an encephalitis model based on infection of rat organotypic hippocampal brain cultures with Langat virus. The use of an attenuated virus strain has several advantages compared to the work with human pathogenic TBE strains. Despite its relative avirulence for the human host, LGTV shares a high nucleotide homology with pathogenic TBEV. Nevertheless, LGTV can be handled under BSL-2 conditions. In rodents Langat virus is able to cause viremia, infect the central nervous system and thereby causing encephalitis and neuronal lesions replicating clinical and histopathological features observed in human cases of TBE [36,43]. Our immunohistochemical results about infection of neuronal cells by LGTV is in agreement with previous findings of fatal TBE human cases where neurons represent the main target for TBEV infection [44]. By measuring virus replication in organotypic hippocampal cultures and staining  viral proteins within the neuronal cells we could demonstrate that Langat virus productively infects neurons of rat brain slices ( Figure 4C, D, E, F, G, H). In many aspects OHCs fill the gap between dissociated cell cultures and in vivo animal experiments. OHCs are ex vivo brain slices and constitute an intact neuronal network with a well-preserved representation of the most important brain cells, including neurons, astrocytes and microglia [45,46]. This in vitro model of neuronal network allows the investigation of the mechanisms underlying neuropathogenicity of TBE and, as we have shown in this study, the efficiency of new antiviral drugs under conditions resembling those encountered in vivo [47,48]. Our results obtained in OHCs transfected with the siRNA sequence D3 support findings from several in vivo studies showing that neuronal cells could be transfected with siRNA and are competent to develop a strong siRNA mediated antiviral activity [14,30,49,50]. Applying siRNA sequence D3 on organotypic cell cultures an inhibition of Langat virus replication by more than 99.6% was achieved, both on the level of viral RNA and on the number of infectious particles. After entering the human body, tick-borne flaviviruses develop a high viremia which is considered to be a prerogative for the subsequent invasion of the CNS [51]. The antiviral potential of our siRNA sequences may help to attenuate the viremia and thus prevent the development of encephalitis or, if encephalitis is already established, to limit the number of infected neurons to a minimum thereby minimizing the damage in the brain caused by either the virus itself or the immune response to the infection.
The siRNA sequence D3 targets a well conserved stretch of the 59UTR of Langat virus sharing 100% homology with all 3 subtypes of TBEV, the European, the Siberian and the Far Eastern subtype as well as with the Omsk hemorrhagic fever virus and 95% (18 nucleotides out of 19) with Louping ill virus (see Table S1). We therefore consider D3 a promising candidate for future in vivo studies on TBEV infections. Other siRNA sequences such as D5, D8, D12, D4, Q5 and Q6 were also highly effective in our in vitro analysis and may be used in the combination with D3. As it has been suggested in other studies the combination of 2 or more specific siRNAs simultaneously targeting multiple viral genome regions may enhance the inhibitory effect and reduce an eventual viral escape [27,52].
An unspecific stimulation of the interferon response by the siRNA sequence D3 is unlikely since no upregulation of the IFNb gene was detected. Cationic lipid reagents, such as Lipofectamine RNAiMAX and Dharmafect 2 used in this study, are promising carriers for the siRNA delivery in vivo [53,54]. However they are known to cause side effects such as cytotoxicity and nonspecific activation of intracellular signalling pathways [55,56]. To determine toxicity of the transfection regimen the LDH release was assessed in several experiments but no increase in LDH level was observed in HeLa cells or in organotypic cell cultures treated with the transfection reagents and siRNA.
Despite these encouraging results, development of reliable transfection methods remains the biggest hurdle in the progress of in vivo antiviral RNAi-based drug therapies. In case of encephalitogenic viruses such as TBEV, carriers for therapeutic siRNA molecules must be able to cross the blood-brain barrier [57]. For these reasons the development of new strategies which allow highly efficient transfection of siRNA molecules is a crucial step on the way towards the clinical use of RNAi-based therapies. Organotypic cultures represent a valuable tool to evaluate the efficiency of different transfection strategies with siRNA and to determine their potential to induce the interferon system in a conserved neuronal network.
In summary, our study provides further support for the use of RNAi technology in the development of antiviral drugs against encephalitogenic tick-borne flaviviruses. Organotypic brain cultures were used for the first time as a successful in vitro approach in an RNAi-based antiviral therapy.

Cell Culture and Virus Amplification
Vero and HeLa cells were cultured in Eagle's minimum essential medium (MEM) supplemented with 10% foetal bovine serum, 1.25% L-Glutamin, 1% Non-essential amino acids, 1% Penicillin-Streptomycin and 0.5% Neomycin-Bacitracin (Biochrom AG, Berlin, Germany). Langat virus strain TP21 was kindly provided by Daniel Růžek (University of South Bohemia, Č eské Budějovice, Czech Republic). Virus was cultured on 80% confluent Vero cells in 75-cm 2 culture flasks (Sigma-Aldrich, Buchs, Switzerland) and grown for 7 days at 37uC in cell culture medium containing 2% FBS. Virus containing cell culture supernatant was titrated and used for infection and transfection assays.

Organotypic Cultures
Organotypic hippocampal cultures were prepared from 5 daysold Wistar rats as previously described by us [33]. Briefly, rat pups were sacrificed by a lethal dose of Pentobarbital i.p (G. Streuli & Cie. SA, Uznach, Switzerland). The brain was removed and submerged in ice-cold dissection medium consisting of Hank's balanced salt solution (HBSS; Gibco Life Technologies, Basel, Switzerland) with 6 mg/ml glucose and 10 mg/ml Penicillin-Streptomycin. Hippocampus was isolated and cut perpendicular to the axis into 400 mm thick-sections by a McIlwain tissue chopper (Mickle Laboratory, Giuldford, UK). Slices with intact hippocampal morphology were selected and individually transferred on a semiporous (0.4 mm) membrane of the Transwell inserts (Corning Inc., Corning, NY). Inserts were placed in a 24-well plate in contact with 200 ml serum-free Neurobasal TM medium (Gibco) supplemented with B27 Supplement (20 ml/ml, Gibco). Before transfection, slices were incubated at 37uC with 5% CO 2 for 4 days. Medium containing B27 Supplement was changed every day for the first two days.
Comparative analysis of all siRNA sequences with genomes of tick-borne flaviviruses was performed using GeniusPro Version 5.5.7 (Table S1). When infection was performed before transfection, HeLa cells were infected with Langat virus at a MOI of 10, 1, 0.1 or 0.01. One hour after the infection the inoculum was removed and cells were transfected with 200 nM siRNA/Lipofectamine RNAiMAX complexes as described above. The siRNA/Lipofectamine complexes were removed after 4 hours and cells incubated for 6 days at 37uC. Virus titer was assessed by determining the number of genome copies in 100 ml cell supernatant by real-time RT-PCR 6 days after infection.

Transfection and Infection Assays
Organotypic hippocampal cultures were prepared as described above. After slicing OHCs were cultured for 4 days to allow recovery. OHCs were transfected with 800 nM siRNA complexed with the lipid-based transfection reagent Dharmafect 2 (Thermo Fisher Scientific, Lafayette, USA) for 24 h according to the manufacturer's instruction. Lipid-siRNA complexes were prepared in 200 ml antibiotic-free Neurobasal Medium (NBM) whereof 100 ml were given to the medium underneath the membrane and 100 ml were added drop wise onto the slice. One day after incubation the OHC were infected with 2610E6 FFU Langat virus in 200 ml NBM. After 1 h inoculum was removed and OHCs transfected for the second time. Viral genome copy number and number of infectious particles were assessed 2 days after the second transfection step.

LDH Cytotoxicity Test
HeLa cells transfected with 200 nM siRNA (D3 or nonsense) and untransfected cells were infected with Langat virus 4 h after transfection and incubated for 6 days at 37uC. During 6 days samples of 100 ml supernatant were taken daily and LDH activity was quantified according to the manufacturer's instructions applying the Cytotoxicity Detection Kit PLUS (LDH) (Roche Diagnostics, Rotkreuz, Switzerland) by measuring optical density (OD) at 490 nm.

Viral RNA Extraction
Nucleic acid extraction from cell supernatant or OHC homogenate was performed using the BioRobot EZ1 Workstation (Qiagen). 100 ml cell culture supernatant were inactivated in 400 ml AVL viral lysis buffer (Qiagen) and viral RNA isolated applying the EZ1 Virus mini kit v2.0 (Qiagen). OHCs were homogenized in 800 ml Qiazol (Qiagen) by the TissueLyser Sytem (Qiagen) for 2 min at 25 Hz. Extraction was performed with the Universal RNA Tissue Kit (Qiagen) following the manufacturer's instructions. RNA was eluted in a final volume of 50 ml.

Primers
The specific primers and probes for the detection of Langat virus strain TP21 were designed using Primer Express software v3.0 (Applied Biosystems, Foster City, CA) and purchased from Microsynth (Balgach, Switzerland).

One
Step Real-time Reverse Transcriptase PCR Viral RNA was quantified by one-step real-time RT-PCR using the QuantiFast probe RT-PCR kit (Qiagen) according to the manufacturer's protocol with the following cycling conditions: reverse transcription at 50uC for 10 min, an initial PCR activation step at 95uC for 5 min, 40 cycles of two-step cycling for 10 s at 95uC and 30 s at 60uC.

Fluorescence Microscopy
Organotypic hippocampal slices were fixed in 4% paraformaldehyde (PFA) in phosphate-buffered saline for 90 min at 4uC and cryopreserved in 18% sucrose solution in PBS for 3-4 days. Slices were cut into 12 mm thick sections with a Jung CM 1800 cryostat (Leica Microsystems, Glattbrug, Switzerland) according to a procedure described elsewhere [33]. The sections were collected in PBS, transferred to a chrome-alum-gelatine-coated glass slide, air dried and kept in PBS. Organotypic sections were incubated with primary antibodies directed against Langat virus (LGT-I-13A10; 1:1000 Fort Detrick, USAMRIID, USA) and FOX3 (rabbit polyclonal; 1:1000, Millipore). Primary and secondary antibodies were diluted in TBS (pH 7.6) containing 0.5% bovine serum albumin. Sections were washed three times with PBS and then exposed to the secondary antibody for 45 min at RT in the dark.