Vaccine 33 (2015) 1459–1464
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Four-segmented Rift Valley fever virus induces sterile immunity in sheep after a single vaccination Paul J. Wichgers Schreur a,∗ , Jet Kant a , Lucien van Keulen a , Rob J.M. Moormann a,b , Jeroen Kortekaas a a b
Department of Virology, Central Veterinary Institute, Part of Wageningen University and Research Centre, Lelystad, The Netherlands Department of Infectious Diseases and Immunology, Virology Division, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands
a r t i c l e
i n f o
Article history: Received 9 December 2014 Received in revised form 23 January 2015 Accepted 25 January 2015 Available online 7 February 2015 Keywords: Rift Valley fever virus Four-segmented Sheep Vaccination Challenge
a b s t r a c t Rift Valley fever virus (RVFV), a mosquito-borne virus in the Bunyaviridae family, causes recurrent outbreaks with severe disease in ruminants and occasionally humans. The virus comprises a segmented genome consisting of a small (S), medium (M) and large (L) RNA segment of negative polarity. The Msegment encodes a glycoprotein precursor (GPC) protein that is co-translationally cleaved into Gn and Gc, which are required for virus entry and fusion. Recently we developed a four-segmented RVFV (RVFV4s) by splitting the M-genome segment, and used this virus to study RVFV genome packaging. Here we evaluated the potential of a RVFV-4s variant lacking the NSs gene (4s-NSs) to induce protective immunity in sheep. Groups of seven lambs were either mock-vaccinated or vaccinated with 105 or 106 tissue culture infective dose (TCID50 ) of 4s-NSs via the intramuscular (IM) or subcutaneous (SC) route. Three weeks post-vaccination all lambs were challenged with wild-type RVFV. Mock-vaccinated lambs developed high fever and high viremia within 2 days post-challenge and three animals eventually succumbed to the infection. In contrast, none of the 4s-NSs vaccinated animals developed clinical signs during the course of the experiment. Vaccination with 105 TCID50 via the IM route provided sterile immunity, whereas a 106 dose was required to induce sterile immunity via SC vaccination. Protection was strongly correlated with the presence of RVFV neutralizing antibodies. This study shows that 4s-NSs is able to induce sterile immunity in the natural target species after a single vaccination, preferably administrated via the IM route. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Rift Valley fever virus (RVFV) is responsible for devastating disease in ruminants and occasionally humans. The virus causes recurrent outbreaks on the African continent, the Arabian Peninsula and several islands off the coast of Southern Africa. During outbreaks, abortion storms as well as high mortality among newborn ruminants are observed frequently. The majority of infected humans display flu-like symptoms whereas a small percentage of individuals develop severe disease, which may result in death.
Abbreviations: RVFV, Rift Valley fever virus; GPC, glycoprotein precursor; SC, subcutaneously; IM, intramuscular; IV, intravenously; RVFV-4s, four-segmented RVFV; 4s-NSs, four-segmented RVFV lacking the NSs gene; S, small; M, medium; L, large; TCID50 , tissue culture infective dose 50%; FFU, focus forming unit; N, nucleocapsid protein; NSs, non-structural protein S-segment; NSm, non-structural protein M-segment. ∗ Corresponding author. Tel.: +31 320 238423. E-mail address: [email protected] (P.J. Wichgers Schreur). http://dx.doi.org/10.1016/j.vaccine.2015.01.077 0264-410X/© 2015 Elsevier Ltd. All rights reserved.
Transmission among ruminants is predominantly mediated by Aedine and Culicine mosquito vectors. Since these vectors are not confined to RVFV endemic areas there is a significant risk that RVFV will expand its territory in the near future [1–3]. Although humans can also be infected via mosquito bite, the majority of infections are attributed to exposure to infected tissues or body fluids [4]. RVFV is a member of the Bunyaviridae family, genus Phlebovirus, and comprises a segmented genome consisting of a small (S), medium (M) and large (L) RNA segment of negative polarity [5,6]. The L segment encodes the RNA-dependent RNA polymerase, responsible for transcription and replication of the viral genome. The M-segment encodes a glycoprotein precursor (GPC) protein that is co-translationally cleaved into two major structural glycoproteins; Gn and Gc, which are required for virus–cell attachment and membrane fusion. Additionally, the M-segment encodes a non-structural protein (NSm), with anti-apoptotic function [7,8] and a 78-kDa protein of unknown function that was shown to be incorporated into virions produced by mosquito cells [9]. The S segment encodes a nucleocapsid protein (N) in genomic-sense
1460
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
orientation and a nonstructural protein (NSs) in antigenomic-sense orientation. The N protein protects the viral genomic RNA from degradation and NSs, the major virulence factor of the virus, is involved in antagonizing host innate immune responses [10–14]. In African countries where the virus is endemic, RVFV outbreaks are controlled by vaccination with either inactivated vaccines or with the live-attenuated Smithburn and Clone 13 vaccines [4,15,16]. The formalin-inactivated vaccines offer optimal safety, but require multiple administrations to induce protective immunity, whereas the Smithburn and Clone 13 vaccines are effective with one administration [17,18]. The Smithburn vaccine was created by multiple intracerebral passaging of the virus in mice [19] and the Clone 13 vaccine is a natural RVFV isolate that contains a large (70%) internal deletion in the NSs gene [20]. Importantly, the Smithburn vaccine is not safe in pregnant and young target animals [21,22]. So far no safety concerns have arisen from a limited number of reported studies with Clone 13 in target animals [18,23]. A Clone 13-based vaccine was commercialized in South Africa by Onderstepoort Biological Products in 2010. Since
that time, several million doses were applied in the field. However, as far as the authors are aware of, the experiences with application of this vaccine in the field await to be reported. Currently, alternative live-attenuated viruses are being developed aiming to generate live vaccines with an even higher safety profile. The establishment of RVFV reverse genetics in 2006 has boosted this development [24]. In 2011, we developed intrinsically safe RVFV replicon particles which are highly effective in inducing a protective immune response [25,26]. However, large scale application of this nonspreading RVFV (NSR) in livestock requires optimization of the production process. More recently, as part of a study on RVFV genome packaging, we developed a novel liveattenuated RVFV. This virus, referred to as four-segmented RVFV (RVFV-4s), was found to be highly immunogenic in mice [27]. RVFV4s was constructed by splitting the wild-type M-genome segment into two M-type genome segments encoding either the Gn or Gc protein. Here, we evaluated the potential of a RVFV-4s variant lacking the NSs gene (4s-NSs) to induce protective immunity in sheep,
Fig. 1. Rectal temperatures of vaccinated and mock-vaccinated lambs. Rectal body temperatures (◦ C) were determined daily during the experimental period. Fever was defined as a body temperature above 40.5 ◦ C (dashed line). Data are depicted as averages (n = 7) with standard deviation. Rectal body temperatures of mock-vaccinated lambs determined 23, 24 and 27 days post-vaccination are depicted as averages of 6, 5 and 4 measurements, respectively, since a lamb from this group died on each of these days.
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
the major natural target species of RVFV. We evaluated two different doses and administered the vaccine via either subcutaneous (SC) or intramuscular (IM) route. The results show that 4s-NSs is able to induce sterile immunity in lambs after a single vaccination, preferably administrated via the IM route.
2. Results 2.1. Vaccination with 4s-NSs protects lambs from viremia, fever and mortality To evaluate the vaccine potential of 4s-NSs, a vaccinationchallenge experiment was performed in lambs. A total of 35 lambs were randomly divided into 5 groups of 7 animals. Lambs of groups 1 and 2 were vaccinated SC with 105.1 or 106.1 TCID50 (50% tissue culture infective dose) of 4s-NSs whereas lambs of groups 3 and 4 were vaccinated IM with 105.1 or 106.1 TCID50 , respectively. Group 5 was mock-vaccinated and served as a challenge control. Neither infectious vaccine virus nor vaccine RNA was detected in any of the plasma samples of vaccinated animals in the first week post-vaccination (data not shown). Three weeks post-vaccination all lambs were challenged with a high dose (105 TCID50 ) of RVFV rec35/74, administered via the intravenous (IV) route. As expected, mock-vaccinated lambs developed fever (>40.5 ◦ C) starting two days post-challenge (Fig. 1). The induction of fever was strongly correlated with the presence of virus in plasma samples, as evidenced by qRT-PCR (up to ≈1010 RNA copies/ml plasma) (Fig. 2) and virus
1461
isolation (data not shown). Three lambs eventually succumbed to the RVFV infection; two lambs 3 days after challenge and one lamb 7 days after challenge. In contrast, no fever was observed in vaccinated lambs (Fig. 1) and no challenge virus could be recovered from plasma samples collected from these animals. In addition, no viral RNA was detected in plasma samples of lambs vaccinated with either dose via the IM route or with a dose of 106 TCID50 administered via the SC route (Fig. 2). Only minor levels of viral RNA were detected in plasma samples of two animals in the 105 SC group. Collectively, these results show that a single administration of 4sNSs induces sterile immunity in the major natural target species of RVFV.
2.2. IM vaccinated lambs display higher neutralizing antibody responses compared to SC vaccinated lambs Pre- and post-challenge sera were evaluated for the presence of RVFV-specific neutralizing antibodies using a recently developed RVFV-4s based highly sensitive virus neutralization test (VNT) (Wichgers Schreur et al., manuscript in preparation). As expected, neutralizing antibodies could not be detected in any of the sera collected on the day of vaccination or in sera collected from mockvaccinated animals before challenge (Fig. 3). In contrast, high levels of neutralizing antibodies were detected in sera obtained 2 and 3 weeks post 4s-NSs vaccination. Interestingly, VNT titers in sera derived from animals vaccinated with the 105 dose of 4s-NSs via the IM route were significantly higher than those in sera derived
Fig. 2. Monitoring of viremia in vaccinated and mock-vaccinated lambs. Plasma samples were analyzed by qRT-PCR for the presence of RVFV-specific RNA at the indicated days post-challenge. The dashed line indicates the limit of detection. Infectious virus could be isolated from plasma samples containing >105 RNA copies/ml.
1462
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
2.4. No viral RNA or infectious virus detected in organs of vaccinated animals
Fig. 3. RVFV VNT titers of vaccinated and mock-vaccinated lambs. Weekly obtained serum samples were evaluated for the presence of neutralizing antibodies using a 4s-NSs based VNT. Error bars represent standard deviations. The arrow marks the day of challenge and the asterisk indicates a statistical difference. The y-axis crosses at the detection limit of the assay (y = 10).
from animals vaccinated with the same dose via the SC route. The results also indicate that sterile immunity was obtained in lambs vaccinated with 105 or 106 TCID50 , administered via IM route and in lambs vaccinated with 106 TCID50 administered via SC route. In these groups, no anamnestic immune responses following challenge infections were observed. The two viremic animals in the 105 (SC) group showed relatively low VNT titers after vaccination and high titers after challenge, suggesting that protection was correlated with neutralizing antibody responses. Altogether, these results show that 4s-NSs vaccination, preferably administered via the IM route, results in the induction of a protective humoral immune response. 2.3. N-antibody levels progressively increased after vaccination and challenge In addition to the evaluation of neutralizing antibody responses, we monitored the kinetics of the induction of anti-N antibodies using the commercially available ID Screen® Rift Valley Fever Competition ELISA (ID-VET, Montpellier, France). As expected, Nantibody levels in mock-vaccinated animals could not be detected before challenge but rapidly increased after challenge infection (Fig. 4). In contrast, N-antibody levels in vaccinated animals increased progressively during the complete course of the experiment. Similar as observed in the VNT experiments, sera from the 105 SC group showed lower N-antibody levels compared to the 105 IM, 106 IM and the 106 SC groups.
Fig. 4. Detection of anti-N antibodies by ELISA. Weekly obtained sera were evaluated for the presence of N-specific antibodies by ELISA. Titers are expressed as percentage competition (% S/N). All values lower than 40% are considered positive, between 40% and 50% are considered doubtful and above 50% are considered negative. The 50% boundary is represented by a dashed line. Error bars represent standard deviations. No statistical differences were observed between vaccinated groups. The arrow marks the day of challenge.
To evaluate whether challenge virus was present in organs at the end of the experiment, we examined liver and spleen homogenates for viral RNA and infectious virus. As a positive control, liver and spleen homogenates from mock-vaccinated animals that did not survive the challenge infection were used. In animals that succumbed to the infection, high levels of viral RNA were present in both liver and spleen homogenates (Fig. 5). Despite that no challenge virus could be detected in liver samples of surviving animals in the mock-vaccinated group, spleen samples still contained viral RNA and infectious virus 2 weeks after challenge. In both liver and spleen homogenates of vaccinated animals neither infectious virus nor viral RNA could be detected. Altogether these results indicate that challenge virus is effectively cleared in 4s-NSs vaccinated animals.
3. Discussion Commercially available RVFV vaccines suffer from efficacy or safety concerns which could complicate their acceptance in countries at risk of future incursions. This explains the need for well-characterized vaccines that combine high efficacy with high safety and low production costs [4,28]. Here, we evaluated the efficacy of a highly safe RVFV vaccine (4s-NSs) in young lambs using two different vaccine doses (105 and 106 TCID50 ) and two different administration routes (SC and IM). A hypothetical risk of using live-attenuated RVF vaccines is the occurrence of reassortment of vaccine virus with virulent wild-type virus. Fear for such an event is, however, poorly founded. First, RVFV reassortment events in nature are extremely rare [29]. Second, there is no evidence to support differences in virulence for target animals between RVFV isolates, suggesting that any reassortment event in nature, even among two virulent field strains, is unlikely to result in more pathogenic variants. Nevertheless, to minimize the risk of reassortment, live-attenuated RVF vaccines preferably should not cause viremia in vaccinated animals. With the observation that 4s-NSs does not induce post-vaccination viremia in lambs we demonstrate that 4s-NSs will not be prone to reassort with virulent viruses when used in the field. In addition to reassortment, live-attenuated vaccines could be sensitive for recombination. Although recombination has not been described for RVF viruses, we designed the RVFV-4s strains in such a way that they lack any complementary sequences that could, so far only theoretically, facilitate a recombination event [27]. Altogether, 4s-NSs, can be considered as a novel live-attenuated vaccine with an optimal safety signature. The lamb model which was employed in this study was validated in earlier studies [17,26]. In this model, intravenous inoculation of challenge virus in young lambs reproducibly results in high fever and viremia. Vaccine-induced immunity can be determined relatively easily by measuring the level of neutralizing antibodies before challenge and by monitoring fever and viremia post-challenge. We here report that a single vaccination of lambs with the 4s-NSs vaccine provides sterile immunity and that the IM route is the most efficacious administration route. Traditionally, sheep vaccines are applied via the SC route. Administering live-attenuated vaccines via the SC route may, from a vaccine efficacy perspective, be rather inefficient. With the IM administration route, vaccine virus is expected to reach susceptible cells much easier, such as those present in the blood, resulting in a more robust immune response. With the direct comparison of the SC and IM administration routes in this study we show that IM vaccination of sheep can be superior over SC vaccination, at
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
1463
Fig. 5. (A) Survival curve of mock-vaccinated group and (B) viral RNA and infectious virus in livers and spleens of mock-vaccinated animals. Livers and spleens were collected at necropsy from lambs that succumbed (†) to the infection or from surviving lambs 2 weeks post-challenge. Tissue homogenates were prepared as described in Section 4 and the presence of RVFV was evaluated by qRT-PCR and virus isolation. From samples marked with an asterisk (*) infectious virus was recovered. Only results of mock-vaccinated animals are shown, as organs of vaccinated animals were all negative.
least for this particular RVFV vaccine tested. We expect that the IM administration route of live-attenuated RVFV vaccines might also be favorable for cattle. The latter is supported by a recent vaccination study in calves with an MP-12 variant (arMP-12NSm21/384). In this study, the authors showed that IM vaccination of arMP12NSm21/384 resulted in a higher RVFV VNT response at 91 days post-vaccination compared to SC vaccination [30]. The overall pattern of the RVFV neutralizing immune responses correlated very well with the overall pattern of the anti-N antibody responses. Nevertheless, the anti-N responses in the 4s-NSs vaccinated groups appeared to be slightly lower when compared to Clone 13 induced anti-N responses [17]. Possibly, lower anti-N responses are correlated with reduced N-expression in 4s-NSs infected cells. We previously showed that cells infected with 4sNSs express reduced levels of the N protein when compared with wild-type virus infected cells [27]. Remarkably, and in contrast to the N-expression, we also showed that intracellular Gn levels are higher in RVFV-4s infected cells [27]. Since most of the natural RVFV neutralizing antibodies target Gn we postulate that the high efficacy of the 4s-NSs vaccine (sterile immunity with 105 TCID50 ) is at least partly the result of increased Gn expression. Collectively, the results reported here provide further evidence that the 4s-NSs vaccine optimally combines efficacy with safety. We already showed that RVFV-4s is completely avirulent in mice, even in the presence of the main virulence factor NSs [27]. With this study we show that 4s-NSs does not induce post-vaccination viremia or other untoward effects in target animals and induces sterile immunity after a single vaccination. Additional studies are needed to investigate the longevity of the protective immune response. 4. Materials and Methods 4.1. Ethics statement The animal experiment was conducted in accordance with the Dutch Law on Animal Experiments (Wod, ID number BWBR0003081) and approved by the Animal Ethics Committee of the Central Veterinary Institute (Permit Number: 2014019). 4.2. Preparation vaccine and challenge virus A previously described RVFV-4s strain not containing the NSs gene named RVFV-LMMSdelNSs , here referred to as 4s-NSs, was used as a vaccine [27]. The virus was grown in BSR-T7 cells in the presence of CO2 -independent medium (CIM, Invitrogen) supplemented with 5% FBS and 1% penicillin–streptomycin (Invitrogen), hereafter referred to as complete CIM medium. The recombinant RVFV strain 35/74 (RVFV rec35/74) was used as challenge virus [25].
Titers were determined as TCID50 and focus forming units (FFU) using the Spearman–Kärber algorithm.
4.3. Vaccination and challenge of lambs Conventional 10–12 week-old lambs were randomly divided into five groups of seven animals. These lambs were offspring from Texel–Swifter ewes and a Suffolk ram. After 1 week of acclimatization, lambs of groups 1–4 were vaccinated via either the IM (right thigh) or SC route (right axilla) with either 105.1 or 106.1 TCID50 (corresponding to 104.6 and 105.6 FFU) of the 4s-NSs virus. Lambs of group 5 were mock-vaccinated with complete CIM. Three weeks post-vaccination, all lambs were challenged via the IV route (jugular vein) with 105 TCID50 of RVFV rec35/74. Vaccine and challenge viruses were administered in 1 ml complete CIM medium. Prior to IV challenge, animals were sedated by IM administration of medetomidine (40 g/kg medetomidine hydrochloride, Sedator® , Eurovet, The Netherlands). Rectal temperatures were determined daily and serum blood samples were obtained weekly. EDTA blood samples were taken weekly but during the first 6 days post-vaccination and 11 days post-challenge additional daily EDTA blood samples were taken. At the end of the experiment (2 weeks post-challenge), animals were euthanized by exsanguination, after being anesthetized with 50 mg/kg sodium pentobarbital (Euthasol® , ASTfarma BV, The Netherlands) applied via the IV route. Plasma samples were analyzed for the presence of RVFV RNA with quantitative real-time PCR (qRT-PCR) as described previously [31].
4.4. VNT Virus neutralization titers were determined using a RVFV-4sbased VNT (Wichgers Schreur et al., manuscript in preparation). Briefly, three-fold serial dilutions of sera were made in micro titer plates and mixed with a fixed amount of RVFV-LMMSeGFP (∼200 TCID50 ) [27]. After a 2-h incubation period, BHK-21 cells were added to the serum–virus mixtures. After an incubation of 2 days at 37 ◦ C and 5% CO2 eGFP expression was evaluated with the EVOS-FL microscope. VNT titers were calculated using the Spearman–Kärber algorithm.
4.5. Preparation of tissue homogenates Tissue homogenates were prepared using the ULTRA TURRAX system in combination with DT-20 tubes (IKA, Staufen, Germany). Briefly, 5 ml culture medium was added to 0.5 g of tissue. Samples were homogenized for 40 s and cell debris was subsequently removed by centrifugation (10 min 4500 rpm). Resulting
1464
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
homogenates were used to infect BHK-21 cells (for virus isolation) or were used for RNA isolation as previously described [25].
[13]
4.6. Statistical analysis Data were statistically analyzed using the Mann–Whitney test in GraphPad Prism. P values
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Four-segmented Rift Valley fever virus induces sterile immunity in sheep after a single vaccination Paul J. Wichgers Schreur a,∗ , Jet Kant a , Lucien van Keulen a , Rob J.M. Moormann a,b , Jeroen Kortekaas a a b
Department of Virology, Central Veterinary Institute, Part of Wageningen University and Research Centre, Lelystad, The Netherlands Department of Infectious Diseases and Immunology, Virology Division, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands
a r t i c l e
i n f o
Article history: Received 9 December 2014 Received in revised form 23 January 2015 Accepted 25 January 2015 Available online 7 February 2015 Keywords: Rift Valley fever virus Four-segmented Sheep Vaccination Challenge
a b s t r a c t Rift Valley fever virus (RVFV), a mosquito-borne virus in the Bunyaviridae family, causes recurrent outbreaks with severe disease in ruminants and occasionally humans. The virus comprises a segmented genome consisting of a small (S), medium (M) and large (L) RNA segment of negative polarity. The Msegment encodes a glycoprotein precursor (GPC) protein that is co-translationally cleaved into Gn and Gc, which are required for virus entry and fusion. Recently we developed a four-segmented RVFV (RVFV4s) by splitting the M-genome segment, and used this virus to study RVFV genome packaging. Here we evaluated the potential of a RVFV-4s variant lacking the NSs gene (4s-NSs) to induce protective immunity in sheep. Groups of seven lambs were either mock-vaccinated or vaccinated with 105 or 106 tissue culture infective dose (TCID50 ) of 4s-NSs via the intramuscular (IM) or subcutaneous (SC) route. Three weeks post-vaccination all lambs were challenged with wild-type RVFV. Mock-vaccinated lambs developed high fever and high viremia within 2 days post-challenge and three animals eventually succumbed to the infection. In contrast, none of the 4s-NSs vaccinated animals developed clinical signs during the course of the experiment. Vaccination with 105 TCID50 via the IM route provided sterile immunity, whereas a 106 dose was required to induce sterile immunity via SC vaccination. Protection was strongly correlated with the presence of RVFV neutralizing antibodies. This study shows that 4s-NSs is able to induce sterile immunity in the natural target species after a single vaccination, preferably administrated via the IM route. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Rift Valley fever virus (RVFV) is responsible for devastating disease in ruminants and occasionally humans. The virus causes recurrent outbreaks on the African continent, the Arabian Peninsula and several islands off the coast of Southern Africa. During outbreaks, abortion storms as well as high mortality among newborn ruminants are observed frequently. The majority of infected humans display flu-like symptoms whereas a small percentage of individuals develop severe disease, which may result in death.
Abbreviations: RVFV, Rift Valley fever virus; GPC, glycoprotein precursor; SC, subcutaneously; IM, intramuscular; IV, intravenously; RVFV-4s, four-segmented RVFV; 4s-NSs, four-segmented RVFV lacking the NSs gene; S, small; M, medium; L, large; TCID50 , tissue culture infective dose 50%; FFU, focus forming unit; N, nucleocapsid protein; NSs, non-structural protein S-segment; NSm, non-structural protein M-segment. ∗ Corresponding author. Tel.: +31 320 238423. E-mail address: [email protected] (P.J. Wichgers Schreur). http://dx.doi.org/10.1016/j.vaccine.2015.01.077 0264-410X/© 2015 Elsevier Ltd. All rights reserved.
Transmission among ruminants is predominantly mediated by Aedine and Culicine mosquito vectors. Since these vectors are not confined to RVFV endemic areas there is a significant risk that RVFV will expand its territory in the near future [1–3]. Although humans can also be infected via mosquito bite, the majority of infections are attributed to exposure to infected tissues or body fluids [4]. RVFV is a member of the Bunyaviridae family, genus Phlebovirus, and comprises a segmented genome consisting of a small (S), medium (M) and large (L) RNA segment of negative polarity [5,6]. The L segment encodes the RNA-dependent RNA polymerase, responsible for transcription and replication of the viral genome. The M-segment encodes a glycoprotein precursor (GPC) protein that is co-translationally cleaved into two major structural glycoproteins; Gn and Gc, which are required for virus–cell attachment and membrane fusion. Additionally, the M-segment encodes a non-structural protein (NSm), with anti-apoptotic function [7,8] and a 78-kDa protein of unknown function that was shown to be incorporated into virions produced by mosquito cells [9]. The S segment encodes a nucleocapsid protein (N) in genomic-sense
1460
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
orientation and a nonstructural protein (NSs) in antigenomic-sense orientation. The N protein protects the viral genomic RNA from degradation and NSs, the major virulence factor of the virus, is involved in antagonizing host innate immune responses [10–14]. In African countries where the virus is endemic, RVFV outbreaks are controlled by vaccination with either inactivated vaccines or with the live-attenuated Smithburn and Clone 13 vaccines [4,15,16]. The formalin-inactivated vaccines offer optimal safety, but require multiple administrations to induce protective immunity, whereas the Smithburn and Clone 13 vaccines are effective with one administration [17,18]. The Smithburn vaccine was created by multiple intracerebral passaging of the virus in mice [19] and the Clone 13 vaccine is a natural RVFV isolate that contains a large (70%) internal deletion in the NSs gene [20]. Importantly, the Smithburn vaccine is not safe in pregnant and young target animals [21,22]. So far no safety concerns have arisen from a limited number of reported studies with Clone 13 in target animals [18,23]. A Clone 13-based vaccine was commercialized in South Africa by Onderstepoort Biological Products in 2010. Since
that time, several million doses were applied in the field. However, as far as the authors are aware of, the experiences with application of this vaccine in the field await to be reported. Currently, alternative live-attenuated viruses are being developed aiming to generate live vaccines with an even higher safety profile. The establishment of RVFV reverse genetics in 2006 has boosted this development [24]. In 2011, we developed intrinsically safe RVFV replicon particles which are highly effective in inducing a protective immune response [25,26]. However, large scale application of this nonspreading RVFV (NSR) in livestock requires optimization of the production process. More recently, as part of a study on RVFV genome packaging, we developed a novel liveattenuated RVFV. This virus, referred to as four-segmented RVFV (RVFV-4s), was found to be highly immunogenic in mice [27]. RVFV4s was constructed by splitting the wild-type M-genome segment into two M-type genome segments encoding either the Gn or Gc protein. Here, we evaluated the potential of a RVFV-4s variant lacking the NSs gene (4s-NSs) to induce protective immunity in sheep,
Fig. 1. Rectal temperatures of vaccinated and mock-vaccinated lambs. Rectal body temperatures (◦ C) were determined daily during the experimental period. Fever was defined as a body temperature above 40.5 ◦ C (dashed line). Data are depicted as averages (n = 7) with standard deviation. Rectal body temperatures of mock-vaccinated lambs determined 23, 24 and 27 days post-vaccination are depicted as averages of 6, 5 and 4 measurements, respectively, since a lamb from this group died on each of these days.
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
the major natural target species of RVFV. We evaluated two different doses and administered the vaccine via either subcutaneous (SC) or intramuscular (IM) route. The results show that 4s-NSs is able to induce sterile immunity in lambs after a single vaccination, preferably administrated via the IM route.
2. Results 2.1. Vaccination with 4s-NSs protects lambs from viremia, fever and mortality To evaluate the vaccine potential of 4s-NSs, a vaccinationchallenge experiment was performed in lambs. A total of 35 lambs were randomly divided into 5 groups of 7 animals. Lambs of groups 1 and 2 were vaccinated SC with 105.1 or 106.1 TCID50 (50% tissue culture infective dose) of 4s-NSs whereas lambs of groups 3 and 4 were vaccinated IM with 105.1 or 106.1 TCID50 , respectively. Group 5 was mock-vaccinated and served as a challenge control. Neither infectious vaccine virus nor vaccine RNA was detected in any of the plasma samples of vaccinated animals in the first week post-vaccination (data not shown). Three weeks post-vaccination all lambs were challenged with a high dose (105 TCID50 ) of RVFV rec35/74, administered via the intravenous (IV) route. As expected, mock-vaccinated lambs developed fever (>40.5 ◦ C) starting two days post-challenge (Fig. 1). The induction of fever was strongly correlated with the presence of virus in plasma samples, as evidenced by qRT-PCR (up to ≈1010 RNA copies/ml plasma) (Fig. 2) and virus
1461
isolation (data not shown). Three lambs eventually succumbed to the RVFV infection; two lambs 3 days after challenge and one lamb 7 days after challenge. In contrast, no fever was observed in vaccinated lambs (Fig. 1) and no challenge virus could be recovered from plasma samples collected from these animals. In addition, no viral RNA was detected in plasma samples of lambs vaccinated with either dose via the IM route or with a dose of 106 TCID50 administered via the SC route (Fig. 2). Only minor levels of viral RNA were detected in plasma samples of two animals in the 105 SC group. Collectively, these results show that a single administration of 4sNSs induces sterile immunity in the major natural target species of RVFV.
2.2. IM vaccinated lambs display higher neutralizing antibody responses compared to SC vaccinated lambs Pre- and post-challenge sera were evaluated for the presence of RVFV-specific neutralizing antibodies using a recently developed RVFV-4s based highly sensitive virus neutralization test (VNT) (Wichgers Schreur et al., manuscript in preparation). As expected, neutralizing antibodies could not be detected in any of the sera collected on the day of vaccination or in sera collected from mockvaccinated animals before challenge (Fig. 3). In contrast, high levels of neutralizing antibodies were detected in sera obtained 2 and 3 weeks post 4s-NSs vaccination. Interestingly, VNT titers in sera derived from animals vaccinated with the 105 dose of 4s-NSs via the IM route were significantly higher than those in sera derived
Fig. 2. Monitoring of viremia in vaccinated and mock-vaccinated lambs. Plasma samples were analyzed by qRT-PCR for the presence of RVFV-specific RNA at the indicated days post-challenge. The dashed line indicates the limit of detection. Infectious virus could be isolated from plasma samples containing >105 RNA copies/ml.
1462
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
2.4. No viral RNA or infectious virus detected in organs of vaccinated animals
Fig. 3. RVFV VNT titers of vaccinated and mock-vaccinated lambs. Weekly obtained serum samples were evaluated for the presence of neutralizing antibodies using a 4s-NSs based VNT. Error bars represent standard deviations. The arrow marks the day of challenge and the asterisk indicates a statistical difference. The y-axis crosses at the detection limit of the assay (y = 10).
from animals vaccinated with the same dose via the SC route. The results also indicate that sterile immunity was obtained in lambs vaccinated with 105 or 106 TCID50 , administered via IM route and in lambs vaccinated with 106 TCID50 administered via SC route. In these groups, no anamnestic immune responses following challenge infections were observed. The two viremic animals in the 105 (SC) group showed relatively low VNT titers after vaccination and high titers after challenge, suggesting that protection was correlated with neutralizing antibody responses. Altogether, these results show that 4s-NSs vaccination, preferably administered via the IM route, results in the induction of a protective humoral immune response. 2.3. N-antibody levels progressively increased after vaccination and challenge In addition to the evaluation of neutralizing antibody responses, we monitored the kinetics of the induction of anti-N antibodies using the commercially available ID Screen® Rift Valley Fever Competition ELISA (ID-VET, Montpellier, France). As expected, Nantibody levels in mock-vaccinated animals could not be detected before challenge but rapidly increased after challenge infection (Fig. 4). In contrast, N-antibody levels in vaccinated animals increased progressively during the complete course of the experiment. Similar as observed in the VNT experiments, sera from the 105 SC group showed lower N-antibody levels compared to the 105 IM, 106 IM and the 106 SC groups.
Fig. 4. Detection of anti-N antibodies by ELISA. Weekly obtained sera were evaluated for the presence of N-specific antibodies by ELISA. Titers are expressed as percentage competition (% S/N). All values lower than 40% are considered positive, between 40% and 50% are considered doubtful and above 50% are considered negative. The 50% boundary is represented by a dashed line. Error bars represent standard deviations. No statistical differences were observed between vaccinated groups. The arrow marks the day of challenge.
To evaluate whether challenge virus was present in organs at the end of the experiment, we examined liver and spleen homogenates for viral RNA and infectious virus. As a positive control, liver and spleen homogenates from mock-vaccinated animals that did not survive the challenge infection were used. In animals that succumbed to the infection, high levels of viral RNA were present in both liver and spleen homogenates (Fig. 5). Despite that no challenge virus could be detected in liver samples of surviving animals in the mock-vaccinated group, spleen samples still contained viral RNA and infectious virus 2 weeks after challenge. In both liver and spleen homogenates of vaccinated animals neither infectious virus nor viral RNA could be detected. Altogether these results indicate that challenge virus is effectively cleared in 4s-NSs vaccinated animals.
3. Discussion Commercially available RVFV vaccines suffer from efficacy or safety concerns which could complicate their acceptance in countries at risk of future incursions. This explains the need for well-characterized vaccines that combine high efficacy with high safety and low production costs [4,28]. Here, we evaluated the efficacy of a highly safe RVFV vaccine (4s-NSs) in young lambs using two different vaccine doses (105 and 106 TCID50 ) and two different administration routes (SC and IM). A hypothetical risk of using live-attenuated RVF vaccines is the occurrence of reassortment of vaccine virus with virulent wild-type virus. Fear for such an event is, however, poorly founded. First, RVFV reassortment events in nature are extremely rare [29]. Second, there is no evidence to support differences in virulence for target animals between RVFV isolates, suggesting that any reassortment event in nature, even among two virulent field strains, is unlikely to result in more pathogenic variants. Nevertheless, to minimize the risk of reassortment, live-attenuated RVF vaccines preferably should not cause viremia in vaccinated animals. With the observation that 4s-NSs does not induce post-vaccination viremia in lambs we demonstrate that 4s-NSs will not be prone to reassort with virulent viruses when used in the field. In addition to reassortment, live-attenuated vaccines could be sensitive for recombination. Although recombination has not been described for RVF viruses, we designed the RVFV-4s strains in such a way that they lack any complementary sequences that could, so far only theoretically, facilitate a recombination event [27]. Altogether, 4s-NSs, can be considered as a novel live-attenuated vaccine with an optimal safety signature. The lamb model which was employed in this study was validated in earlier studies [17,26]. In this model, intravenous inoculation of challenge virus in young lambs reproducibly results in high fever and viremia. Vaccine-induced immunity can be determined relatively easily by measuring the level of neutralizing antibodies before challenge and by monitoring fever and viremia post-challenge. We here report that a single vaccination of lambs with the 4s-NSs vaccine provides sterile immunity and that the IM route is the most efficacious administration route. Traditionally, sheep vaccines are applied via the SC route. Administering live-attenuated vaccines via the SC route may, from a vaccine efficacy perspective, be rather inefficient. With the IM administration route, vaccine virus is expected to reach susceptible cells much easier, such as those present in the blood, resulting in a more robust immune response. With the direct comparison of the SC and IM administration routes in this study we show that IM vaccination of sheep can be superior over SC vaccination, at
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
1463
Fig. 5. (A) Survival curve of mock-vaccinated group and (B) viral RNA and infectious virus in livers and spleens of mock-vaccinated animals. Livers and spleens were collected at necropsy from lambs that succumbed (†) to the infection or from surviving lambs 2 weeks post-challenge. Tissue homogenates were prepared as described in Section 4 and the presence of RVFV was evaluated by qRT-PCR and virus isolation. From samples marked with an asterisk (*) infectious virus was recovered. Only results of mock-vaccinated animals are shown, as organs of vaccinated animals were all negative.
least for this particular RVFV vaccine tested. We expect that the IM administration route of live-attenuated RVFV vaccines might also be favorable for cattle. The latter is supported by a recent vaccination study in calves with an MP-12 variant (arMP-12NSm21/384). In this study, the authors showed that IM vaccination of arMP12NSm21/384 resulted in a higher RVFV VNT response at 91 days post-vaccination compared to SC vaccination [30]. The overall pattern of the RVFV neutralizing immune responses correlated very well with the overall pattern of the anti-N antibody responses. Nevertheless, the anti-N responses in the 4s-NSs vaccinated groups appeared to be slightly lower when compared to Clone 13 induced anti-N responses [17]. Possibly, lower anti-N responses are correlated with reduced N-expression in 4s-NSs infected cells. We previously showed that cells infected with 4sNSs express reduced levels of the N protein when compared with wild-type virus infected cells [27]. Remarkably, and in contrast to the N-expression, we also showed that intracellular Gn levels are higher in RVFV-4s infected cells [27]. Since most of the natural RVFV neutralizing antibodies target Gn we postulate that the high efficacy of the 4s-NSs vaccine (sterile immunity with 105 TCID50 ) is at least partly the result of increased Gn expression. Collectively, the results reported here provide further evidence that the 4s-NSs vaccine optimally combines efficacy with safety. We already showed that RVFV-4s is completely avirulent in mice, even in the presence of the main virulence factor NSs [27]. With this study we show that 4s-NSs does not induce post-vaccination viremia or other untoward effects in target animals and induces sterile immunity after a single vaccination. Additional studies are needed to investigate the longevity of the protective immune response. 4. Materials and Methods 4.1. Ethics statement The animal experiment was conducted in accordance with the Dutch Law on Animal Experiments (Wod, ID number BWBR0003081) and approved by the Animal Ethics Committee of the Central Veterinary Institute (Permit Number: 2014019). 4.2. Preparation vaccine and challenge virus A previously described RVFV-4s strain not containing the NSs gene named RVFV-LMMSdelNSs , here referred to as 4s-NSs, was used as a vaccine [27]. The virus was grown in BSR-T7 cells in the presence of CO2 -independent medium (CIM, Invitrogen) supplemented with 5% FBS and 1% penicillin–streptomycin (Invitrogen), hereafter referred to as complete CIM medium. The recombinant RVFV strain 35/74 (RVFV rec35/74) was used as challenge virus [25].
Titers were determined as TCID50 and focus forming units (FFU) using the Spearman–Kärber algorithm.
4.3. Vaccination and challenge of lambs Conventional 10–12 week-old lambs were randomly divided into five groups of seven animals. These lambs were offspring from Texel–Swifter ewes and a Suffolk ram. After 1 week of acclimatization, lambs of groups 1–4 were vaccinated via either the IM (right thigh) or SC route (right axilla) with either 105.1 or 106.1 TCID50 (corresponding to 104.6 and 105.6 FFU) of the 4s-NSs virus. Lambs of group 5 were mock-vaccinated with complete CIM. Three weeks post-vaccination, all lambs were challenged via the IV route (jugular vein) with 105 TCID50 of RVFV rec35/74. Vaccine and challenge viruses were administered in 1 ml complete CIM medium. Prior to IV challenge, animals were sedated by IM administration of medetomidine (40 g/kg medetomidine hydrochloride, Sedator® , Eurovet, The Netherlands). Rectal temperatures were determined daily and serum blood samples were obtained weekly. EDTA blood samples were taken weekly but during the first 6 days post-vaccination and 11 days post-challenge additional daily EDTA blood samples were taken. At the end of the experiment (2 weeks post-challenge), animals were euthanized by exsanguination, after being anesthetized with 50 mg/kg sodium pentobarbital (Euthasol® , ASTfarma BV, The Netherlands) applied via the IV route. Plasma samples were analyzed for the presence of RVFV RNA with quantitative real-time PCR (qRT-PCR) as described previously [31].
4.4. VNT Virus neutralization titers were determined using a RVFV-4sbased VNT (Wichgers Schreur et al., manuscript in preparation). Briefly, three-fold serial dilutions of sera were made in micro titer plates and mixed with a fixed amount of RVFV-LMMSeGFP (∼200 TCID50 ) [27]. After a 2-h incubation period, BHK-21 cells were added to the serum–virus mixtures. After an incubation of 2 days at 37 ◦ C and 5% CO2 eGFP expression was evaluated with the EVOS-FL microscope. VNT titers were calculated using the Spearman–Kärber algorithm.
4.5. Preparation of tissue homogenates Tissue homogenates were prepared using the ULTRA TURRAX system in combination with DT-20 tubes (IKA, Staufen, Germany). Briefly, 5 ml culture medium was added to 0.5 g of tissue. Samples were homogenized for 40 s and cell debris was subsequently removed by centrifugation (10 min 4500 rpm). Resulting
1464
P.J. Wichgers Schreur et al. / Vaccine 33 (2015) 1459–1464
homogenates were used to infect BHK-21 cells (for virus isolation) or were used for RNA isolation as previously described [25].
[13]
4.6. Statistical analysis Data were statistically analyzed using the Mann–Whitney test in GraphPad Prism. P values
