The 40Fp8 vaccine strain is safe and protects pregnant ewes from a virulent RVFV challenge

Abstract

In the present study, we evaluated the immunogenicity, safety and protective efficacy of the attenuated RVFV-40Fp8 strain in natural hosts (non-pregnant ewes) and in a highly susceptible host infection model such as pregnant ewes in the first third of pregnancy. Our results confirm the immunogenicity of 40Fp8 administration in non-pregnant and pregnant ewes, as well as the absence of foetal damage even after a high-dose vaccination regime in pregnant ewes. In addition, the ewes and their foetuses were protected against a virulent RVFV-56/74 strain challenge, as shown by comparative evaluation of blood, serum and different tissue samples from vaccinated and non-vaccinated pregnant ewes. These results confirm the potential use of 40Fp8 as a RVF live-attenuated vaccine candidate and pave the way for further clinical developments.

Introduction

Rift Valley fever (RVF) is a zoonotic arboviral disease that causes abortions and deaths in livestock (cattle, sheep, goats and camels), and in wild ruminants^1–3. The disease is caused by the Rift Valley fever virus (RVFV), a well-known phlebovirus (class Bunyaviricetes, Order Hareavirales, family Phenuiviridae), that is transmitted and perpetuated in nature by competent mosquito vectors of the Aedes and Culex genera^4–7. RVFV virions display an icosahedral symmetry^8,9 and contain at least three ssRNA genome segments of different size and negative (large and medium segments) or ambisense (small segment) polarities. The large (L) segment encodes for the virion associated RNA polymerase (RdRp). The medium (M) segment contains five in-frame AUG codons that are alternatively used for the synthesis of two major structural glycoproteins, Gn and Gc, and at least two accessory proteins, NSm, a 13-14-kDa cytosolic protein, and a 78-kDa glycoprotein equivalent to a fused NSm-Gn ORF¹⁰. The small (S) segment encodes the nucleocapsid (N) protein and the accessory protein NSs in opposite orientation separated by an intergenic region (IGR). The disease is endemic in Africa and southern regions of the Arabian Peninsula. To ameliorate the impact of the disease in rural and peri urban communities, vaccination of susceptible livestock has been proposed as a “One-Health” approach to prevent disease outbreaks^11–13.

Live attenuated vaccines (LAVs) are excellent inducers of protective and memory immune responses. Their success is largely based on limited replication in the infected host and the induction of a functional immune response. In some cases, the use of LAVs may not be advisable due to potential adverse effects, particularly in immunosuppressed individuals or during pregnancy, either because of impaired immune responses that may not control virus replication, or because of the greater susceptibility of developing foetal tissues to viral invasion. Several live-attenuated vaccine candidates have been reported in the past for RVFV, some of which are commonly used in African countries as a means of disease control or are in early clinical development^14,15. Live-attenuated vaccines are excellent inducers of protection, and immunity generally lasts a long time after a single shot. Therefore, they may be preferred over inactivated vaccines that require multiple boosters, if safety concerns, mainly regarding stability, residual virulence, and reassortment events, are properly addressed. The guidelines for a LAV development for RVF require, among other several considerations, to ensure safety (see (https://www.who.int/publications/m/item/who-target-product-profiles-for-rift-valley-fever-virus-vaccines), therefore a careful determination of residual virulence should be carried out first.

In previous studies, we obtained and characterized a mutagenized variant of the virulent 56/74 isolate by serial passaging in the presence of the antiviral agent favipiravir. This virus variant was termed 40Fp8 and displayed an excellent profile as a live attenuated vaccine due to its extreme attenuation and minimal replication, yet robust immunogenicity in murine models. 40Fp8 showed a total of 24 amino acid substitutions in different viral proteins compared to the parental strain^16,17. We identified that the combination of only 3 of these changes, the P82L mutation in the NSs protein, and mutations G924S and A1303T in the L-protein, was enough to render the virus attenuated in both BALB/c mice and in sheep, while its capacity to induce an immune protective response was unaffected^18,19.

It was noteworthy that the inoculation of immunodeficient A129 IFNAR^-/- transgenic mice, considered as a highly susceptible infection model, with 40Fp8 was not pathogenic, even after intranasal inoculation, while other attenuated vaccine candidates such as rMP12 or NSs-deleted virus induced clear pathological damage²⁰. This higher level of attenuation displayed by 40Fp8 in the immunodeficient murine model was therefore an important asset to exploit its suitability as a safer LAV.

In this study, we aimed to test the immunogenicity, safety and the protective efficacy of the RVFV-40Fp8 strain against a virulent challenge using non-pregnant ewes together with pregnant ewes in the first third of gestation, the latter considered as an animal model extremely susceptible to RVFV infection, by monitoring infectivity in ewes, reproductive tissues and foetuses. The results shown here confirm the immunogenicity of the vaccine candidate in ewes as well as its safety and protective capability in this target species.

Results

Preliminary evaluation of the safety and immunogenicity of a high dose of 40Fp8 strain inoculation in ewes (experiment 1).

We previously demonstrated that the 40Fp8 virus was highly attenuated in mice. To confirm attenuation in a natural host, a high dose of 40Fp8 (10⁷ pfu) was inoculated subcutaneously in ewes (Fig. 1a). Then, clinical, virological, immunological and biochemical parameters were compared with control ewes that received the same dose of the parental virulent RVFV strain 56/74. There was a clear increase in rectal temperature at 1 dpi that peaked at 2 dpi in ewes that received 56/74 virus (#9, #12, #69). In contrast, neither of the two ewes inoculated with 40Fp8 (#21, #30) showed hyperthermia at any time after inoculation (Fig. 2a). Isolation of infectious virus was achieved in two of the three ewes on the first day after inoculation with 56/74, and in all three ewes at 2 and 3 dpi, while no infectious virus was detected in ewes inoculated with 40Fp8 throughout the experiment (Fig. 2b). AST, GGT and LDH serum levels, indicative of liver/tissue damage, increased from 2 dpi in ewes inoculated with 56/74 in contrast to animals receiving 40Fp8 (Fig. 2c). In the group inoculated with 40Fp8 it was possible to detect induction of specific antibodies although slightly delayed and of lower magnitude with respect to the group inoculated with 56/74 (Fig. 2d). Taken together, the absence of clinical disease, viremia and hepatic damage markers in ewes inoculated with a high dose of 40Fp8 virus, together with the specific and neutralizing antibody responses, confirmed its attenuated phenotype and suggested its immunogenic character.

Fig. 1. Experimental procedures in sheep.

Analysis of immunogenicity (a, b), safety (c) or protective efficacy (d) of the 40Fp8 vaccine strain in pregnant animals. Sheep identification numbers (#) are indicated. GD gestational day, IV intravenous inoculation, SC subcutaneous inoculation, dpi days post-immunization, dpc days post challenge.

Fig. 2. Comparative evaluation of rectal temperatures, viremia levels, liver enzymes and RVFV-specific antibodies.

a Sheep rectal temperature records before and after virus inoculation. The horizontal black dotted line marks hyperthermia. b Viremia levels obtained upon titration in cell culture. The values displayed correspond to 50% tissue culture infectious dose per mL (TCID_50/mL). Sensitivity limit of the assay is indicated by a horizontal dotted line. For the purposes of this graph, negative samples were assigned an arbitrary value of 10⁰. c Kinetics of hepatic enzymes plasmatic levels after virus inoculation. AST Aspartate transaminase; GGT Gamma-glutamyltransferase and LDH Lactate dehydrogenase. The shaded areas depict normal range values. Values given in International Units (IU) per mL. d Kinetics of serum antibody responses. Each symbol represents values for individual sheep at the days of sacrifice (4, 7, 8), identified as in (a). Left panel: neutralizing antibodies. The values correspond to the reciprocal of serum dilution causing a 50% reduction of infectivity in cell culture. Right panel: anti-nucleoprotein antibodies. Values correspond to competition percentages as determined by the IDVet ELISA test. Shaded area between 40 and 50% indicates the doubtful status as defined by the provider. Red symbols: sheep inoculated with the wild-type 56/74 strain (n = 3). Blue symbols: sheep inoculated with 40Fp8 virus (n = 2). Individual identification of the 5 animals are indicated in (a).

Comparative evaluation of the immunogenicity induced by a high dose of 40Fp8 strain administered by different routes of inoculation (experiment 2).

An additional 40Fp8 inoculation experiment was performed in three ewes (#76, #809, #868) using the intravenous (IV) route instead, with a high dose of 1.0 × 10⁷ pfu (3.2 × 10⁷ after back titration). In parallel, three ewes (#4, #11, #70) were also SC inoculated with the same dose to compare the effect of the route of administration (Fig. 1b). On the same day of inoculation (d0), the sheep #11, included in the group to be inoculated subcutaneously, showed a slight hyperthermia. However, no further temperature increases were observed after inoculation with 40Fp8 in any of the animals (Fig. 3a). In general, the values of biochemistry parameters (hepatic transaminases, total protein and blood urea nitrogen) showed no evident alterations with respect to normal levels, as observed in the previous experiment and in contrast to the values usually found in ewes inoculated with the virulent RVFV 56/74 isolate²¹. Assessment of RNAemia by RT-qPCR rendered negative Cq values (≥40). Only one blood sample (day 5, ewe #809, IV route) rendered a Cq value below 40 (35.88). However, viral isolation attempts using blood from this sheep were negative upon three consecutive blinded passages in cell culture, indicating a very low or null systemic dissemination of 40Fp8 irrespective of the delivery route. By days 5–7 post inoculation, most of the ewes’ sera had detectable neutralizing antibody (nAb) titers at or above the threshold values of the assay. These data also suggested active virus replication, as confirmed by the kinetics of anti-N antibody induction by competition ELISA (Fig. 3b), also confirming the proper induction of virus-specific antibody responses. No differences were found in the kinetics of induction of humoral immunity in relation to the route of administration. To assess virus-induced cellular immunity, an IFNγ ELISA assay was performed after re-stimulation of whole blood collected at 14 or 21 dpi. IFNγ secretion was clearly detected after re-stimulation with Gn and Gc specific peptides in 2 of the ewes inoculated IV but only at 21 dpi, and in both groups when recombinant proteins (Gn or N) were used as stimuli (Fig. 3c). Taken together, these data point to the ability of 40Fp8 strain to elicit a complete immune response in ewes, despite its high level of attenuation.

Fig. 3. Comparison of intravenous (IV) or subcutaneous (SC) 40Fp8 delivery.

a Evolution of rectal temperature in ewes after SC (light blue) or IV (dark blue) inoculation with 40Fp8 virus. Identification numbers of the 6 animals, 3 per group, are indicated. The horizontal black dotted line marks hyperthermia. b Serological determination of neutralizing antibody titers by TCID₅₀ microneutralization assay (left panel) and competition ELISA anti-N (right panel). Both graphs show mean plus SD values. The dotted line depicts the sensitivity threshold (1/10 dilution). The gray shaded area depicts the doubtful status as defined by the IDVet ELISA test. c Detection of IFNγ in plasma by capture ELISA. Blood samples taken at 14 or 21 days after SC or IV 40Fp8 inoculation were re-stimulated with recombinant RVFV Gn or N proteins or peptides derived from the Gn (#19 and #21) or Gc (#226 and #253) protein sequences. Each symbol denotes individual values (after subtraction of values derived from unstimulated plasma) and horizontal bars denote mean values. Sheep identification as in (a).

Vaccine safety study in pregnant ewes (experiment 3)

Since inoculation of 40Fp8 was immunogenic both at the humoral and cellular level in ewes, we were then aimed to test its vaccine potential in this host species using a more relevant and susceptible animal model such as pregnant sheep. RVF causes stillbirths and abortions in pregnant ruminants, therefore a live attenuated vaccine should be free of any residual virulence that could induce pregnancy failure in such susceptible hosts. To rule out residual virulence, we administered a subcutaneous high dose of 40Fp8 in 12 pregnant ewes at gestational day 56 when sheep are more susceptible to reproductive failure (Fig. 1c). Apart from a mild temperature increase on day 1 after immunization (Fig. 4a), the ewes showed no clinical signs or reproductive failure leading to foetal death, congenital malformations or miscarriages during the experiment. During necropsies carried out between 4 and 30 dpi, the ewes showed no relevant macroscopic or histopathological lesions in target organs of the virus (placentomes and liver) (Fig. 4c and Supplementary Figs. 1 & 2) and the developmental parameters of the foetuses were as expected for their gestation period. No remarkable differences were observed in weight or length of foetuses (Fig. 4b) obtained from immunized ewes and control ewes at the end of the experiment (30 dpi; 86 days of gestation). In addition, all foetuses obtained throughout the experiment from immunized ewes (a total of 27) as well as those from control ewes at the end of the experiment (a total of 3), presented a normal appearance during macroscopic evaluations (Fig. 4c; supplementary Table 2).

Fig. 4. Effect of 40Fp8 inoculation in pregnant ewes.

a Rectal temperature kinetics (mean ± SD) in ewes inoculated subcutaneously with 10⁷ pfu of the 40Fp8 virus (n = 12) and in mock-inoculated ewes (n = 2). The horizontal dotted line marks hyperthermia. b Evaluation of foetal development. Mean and individual values for size and weight of foetuses found after necropsy at different days of gestation (GD). Number of foetuses for each gestation day is shown at Supplementary Table 2/GD 60: n = 7; GD 63: n = 9; GD 70: n = 8; GD 86: n = 3; mock samples: n = 3. Blue symbols, vaccinated; gray symbols, mock control (*). c Macroscopic images showing representative examples of the livers, foetuses and placentas of immunized ewes and non-immunised control ewes at 30 days post-immunisation (dpi)/86 gestational days (GD). Note the absence of the scattered necrotic lesions, characteristic of infected RVFV animals, the normal appearance of the placentas, and the presence of live foetuses with equivalent developmental parameters.

Only low levels of viral RNA were detected by qPCR in some animals in the early stages after immunization in blood samples from umbilical and jugular veins (ewes) (Fig. 5a). The viral genome was practically undetectable from 14 dpi onwards. The presence of the viral genome was also assessed in tissue samples obtained from foetuses (liver, spleen and brain) and ewes (liver, spleen and placentomes) where levels were also very low or undetectable throughout the experiment (Fig. 5b). Accordingly, no infectious virus was isolated from any of these samples with Cqs over the threshold after three blind serial passages in Vero cell cultures. Additional histopathological studies revealed no remarkable histopathological lesions in any of the tissues evaluated, including target organs of the virus (liver and placentomes). Immunohistochemical studies were also negative for the presence of virus antigens in the tissues under investigation (Supplementary Fig. 2).

Fig. 5. Molecular detection of 40Fp8 after inoculation in pregnant ewes (n = 12).

a Detection of viral RNA by RT-qPCR in blood samples taken from umbilical (foetus) or jugular veins of pregnant ewes at various days after inoculation with the 40Fp8 virus. b Detection of viral RNA by RT-qPCR in foetal and/or maternal tissues at various times after inoculation with the 40Fp8 virus. Data from liver, spleen and placentome (cotyledon) collected from pregnant ewes as well as data from liver, spleen and brain taken from foetuses are shown. Foetal spleen samples from 7 to 14 dpi were not retrieved. Blue symbols, vaccinated; gray symbols, mock control (*). The shaded areas indicate the sensitivity limit of the assay. The dotted lines indicate the relation between the Cq obtained and the corresponding viral infectious load or pfu equivalents.

Immunized ewes seroconverted between 4 and 7 dpi, as shown by the kinetics on anti-nucleoprotein antibodies in serum, developing a clear neutralising antibody response from 4 dpi, which reached levels (serum dilution over 1/40) correlating with protection from 7 dpi onwards (Fig. 6). These results demonstrated that a high dose of RVFV-40Fp8 did not cause clinical signs or reproductive failure in pregnant ewes, but induced levels of immunity correlated with protection, thus appearing as a safe vaccination approach against RVFV.

Fig. 6. Humoral responses upon 40Fp8 inoculation in pregnant ewes (n = 12).

Kinetics of neutralizing antibody levels assessed by microneutralization assay in sheep sera obtained on the indicated dates after vaccination with 40Fp8 virus. The dotted line indicates the sensitivity limit of the assay (left panel). Kinetics of induction of anti-nucleoprotein N antibodies assessed by competitive ELISA (IDVet RVFV screen test). The shaded area indicates the doubtful status as defined by the provider (right panel). Median and interquartile range as well as individual values are represented. Blue symbols, vaccinated; gray symbols, mock control (*).

40Fp8 vaccine protective efficacy study in pregnant ewes (experiment 4)

The previous data advocated for the use of 40Fp8 strain as a promising vaccine candidate since pregnant ewes immunized with 40Fp8 were free of adverse effects and developed nAbs from days 4-7. To test whether the 40Fp8 strain induced immunity was indeed sufficient to protect both ewes and foetuses from infection, a new group of pregnant ewes at gestational day 45 was inoculated with 10⁷ pfu of 40Fp8, and three weeks after vaccination were challenged subcutaneously with 5 × 10⁶ pfu of the virulent RVFV 56/74 strain (see experimental outline Fig. 1d). To evaluate the occurrence of lesions that could affect both pregnant ewes and foetuses after challenge, groups of vaccinated and non-vaccinated control animals were necropsied sequentially on days 4, 7 and 21 post-challenge (dpc). When present, lesions were scored according to their extent in the tissues and the severity of the lesion itself between 1 (focal/minimal lesions) and 4 (diffuse/severe lesions). One vaccinated ewe (#9) scheduled to be euthanized on day 4 post challenge developed hyperthermia on day 21 post immunization (Fig. 7a) and aborted prior to challenge (24 days post-immunization). To ascertain if this abortion was induced by 40Fp8 vaccination, the aborted foetus and placental residues recovered from this ewe were subjected to RT-qPCR as well as isolation of RVFV in cell culture, rendering negative results in both assays (not shown). To further investigate the possible cause of this outcome, the aborted tissue samples were also analyzed by qPCR for the presence of several abortifacient pathogens (i.e., Border disease virus, Chlamydia abortus, Coxiella burnetii, Toxoplasma gondii, Neospora caninum, Listeria monocytogenes, Campylobacter fetus, and Brucella spp). No positive hit was found to identify a causative pathogen for this outcome. On day 2 after challenge with the virulent RVFV strain 56/74, all mock-immunized ewes (#4, #5 and #6) displayed higher temperatures than immunized ewes, with values above 40 °C for ewe # 6. Ewe #5 also exhibited temperatures above 40 °C on 4 dpc (Fig. 7a). At necropsy, following a macroscopic pathological analysis, non-immunized control ewes euthanized at 4 (ewe #5) and 7 dpc (ewe #6) exhibited severe scattered foci of necrosis in the liver and increased volume of amniotic fluid (Fig. 7b, c). Although the growth parameters of their foetuses were normal, the foetuses of control ewe #6 euthanized at 7 dpc were found dead and were collected before imminent abortion (Fig. 7c). Non-immunized control ewe #4 aborted on 17 dpc, being euthanized that same day. At necropsy, there was evidence of vaginal mucopurulent discharge associated with acute necrotic haemorrhagic endometritis, without the presence of macroscopic hepatic lesions. Remains of placenta, but not of the foetuses, were found. In contrast, no liver or reproductive organ lesions were observed in any of the immunized ewes euthanized throughout the experiment, and their foetuses (a total of 11) showed normal growth parameters and appearance (Fig. 7c; supplementary Table 2).

Fig. 7. Analysis of protection after RVFV virulent challenge in 40Fp8 vaccinated pregnant ewes.

a Rectal temperatures of individual ewes before and after virus challenge (red vertically dotted line). The horizontal black dotted line marks hyperthermia. b Macroscopic lesions and scores observed in sheep upon 56/74 challenge. When present, lesions were scored according to their extent in the tissues and the severity of the lesion itself between 1 (focal/minimal lesions) and 4 (diffuse/severe lesions). c Representative images of liver, foetuses and placentas found in immunized (upper images) and non-immunized sheep (bottom) after challenge. Scattered foci of necrotic lesions were found in the livers of non-immunized control ewes (white arrows) along with the presence of foetuses found dead and collected before imminent abortion. In contrast, livers without lesions and live foetuses were found in the 40Fp8 immunized ewes. Acute purulent metritis (yellow arrow) was observed in a non-immunized control ewe that aborted at 17 dpc, in contrast to the normal aspect placenta of a protected immunized ewe that was euthanized at 21 dpc.

Histopathological evaluation of the livers of non-immunized ewes euthanized at 4 (ewe #5) and 7 dpc (ewe #6) revealed findings characteristic of RVFV infection, with multifocal necrotic foci filled with cellular debris and both degenerated and viable neutrophils. Viral infection was also confirmed via immunohistochemistry, with hepatocytes within the necrotic foci appearing as the main cells immunolabelled against viral antigen (Fig. 8a, e). Viral antigen was also occasionally observed in the placentomes at 4 dpc (non-immunized ewe #5), specifically in maternal epithelial cells (Fig. 8b–d). However, at 7 dpc, antigen-positive cells were visible throughout the placentomes of non-immunized ewe #6, both in the maternal epithelial cells/syncytial cells and in the foetal trophoblasts (Fig. 8f, g). These cells, mainly maternal epithelial cells, were enlarged and showed necrosis accompanied by viable, degenerated neutrophil infiltrates. Neutrophil infiltrates were also often observed at the base of the maternal villi. As for the tissues collected from non-immunized ewe foetuses at 4 and 7 dpc, viral antigen was only detected in brain, kidney and liver of foetuses before imminent abortion sampled at 7 dpc from ewe #6 ewe (Fig. 8h-m).

Fig. 8. Pathologic findings in non-vaccinated sheep and foetal tissues.

Representative images of histopathological lesions (haematoxylin-eosin staining) and immunohistochemical detection of RVFV antigen in different tissues obtained from pregnant ewes not immunized with the 40Fp8 strain and their foetuses at different days after challenge. Note the presence of hepatocytes immunolabelled against viral antigen within necrotic foci in the liver of non-immunized ewes at 4 and 7 dpc (a, e, 40x). Viral antigen was also occasionally observed in the placentomes at 4 dpc (b, 10x; c, d, 40x), specifically in maternal epithelial cells (red arrows). However, at 7 dpc, antigen-positive cells were visible throughout the placentomes of non-immunized ewe #6, both in the maternal epithelial cells/syncytial cells (red arrows) and in the foetal trophoblasts (black arrows) (f, 10x; g, 40x). Viral antigen was detected in brain, kidney and liver of foetuses before imminent abortion sampled at 7 dpc from ewe #6, but not in foetuses sampled at 4 dpc (h-m, 40x). Caruncular endometrium of ewe #4 aborted on 17 dpc revealed the presence of areas of coagulative necrosis and haemorrhages (asterisk), foci of mineralization (black arrowhead) and lymphoplasmacytic and neutrophilic infiltrates (black arrow) (n, 10x; o, magnified area, 40x). Remains of placenta from the aborted foetus of ewe #4 (p, 10x; q, magnified area, 40x). Note the presence of lymphoplasmacytic and neutrophilic infiltrates (black arrow) as well as the presence of viral antigen in the placental epithelial cells (black arrowheads). (f) foetal villus; (m); maternal villus; (mct): maternal connective tissue.

Histopathological evaluation of the caruncular endometrium of ewe#4 aborted on 17dpc revealed the presence of areas of coagulative necrosis with loss of tissue architecture and haemorrhages, foci of mineralization, focally extensive to multifocal lymphoplasmacytic and neutrophilic infiltrates together with fibrosis at the periphery of the lesion (Fig. 8n, o). The placental remains found corresponding to the aborted foetus of ewe #4 also showed areas of necrosis together with diffuse lymphoplasmacytic and neutrophilic infiltrates. By immunohistochemistry, viral antigen was detected in the placental epithelial cells of this aborted foetus, but not in the uterine caruncles of ewe #4 (Fig. 8p, q). Finally, the liver of ewe #4 showed neither lesions nor presence of viral antigen.

In contrast, tissue samples taken from 40Fp8 vaccinated ewes and their foetuses showed no remarkable histopathological lesions and no presence of viral antigen, which evidences that vaccination with 40Fp8 controls the spread of RVFV to reproductive tissues and foetuses.

Analysis of blood samples for RVFV RNA detection was performed by RT-qPCR. As expected, viral genome was not detected in any of the blood samples taken after immunization with 40Fp8. After challenge, all immunized ewes remained negative, and only samples from non-vaccinated ewes yielded positive results in the 25-35 Cq range between 2-4 dpc (Fig. 9a). These RNA-positive blood samples yielded positive virus isolation in Vero cell culture after the first passage, while samples from immunized ewes were negative for virus isolation after three passages. Blood collected from umbilical vein during necropsies was also subjected to viral RNA detection. Cq values above the limit of detection were found in all foetuses, but upon virus isolation no infectious virus was recovered except for mock immunized ewes’ foetal samples at 7 dpc (Fig. 9a). Liver tissue was also evaluated by RT-qPCR for the presence of RVFV RNA in both ewes and foetuses. The lowest Cq values corresponded to the livers of control ewes tested in the 4 and 7 dpc groups. Livers from foetuses carried by RVFV-RNA positive ewes were also RNA-positive on day 7. Liver samples from vaccinated ewes and their foetuses displaying Cq values above threshold levels were negative to viral isolation attempts. Other tissues obtained from vaccinated animals (placentome, foetal brain and spleen) were evaluated and in all cases yielded negative to virus isolation, despite some background RNA detection by RT-qPCR. In contrast, viral isolation was successful in those samples tested on foetuses carried by non-immunized control ewes (Fig. 9b). These results agree with viral antigen detection described by immunohistochemistry.

Fig. 9. Molecular analysis after virus challenge.

a RT-qPCR analysis of viral RNA in vaccinated ewes and foetal (umbilical vein) blood before and after challenge. b Detection of viral RNA in sheep (liver and placentome) or foetal (liver, spleen, brain) tissue samples by RT-qPCR. In a and b, the shaded area indicates the sensitivity limit of the assay. Red daggers indicate that no foetuses were found in control sheep #4 (aborted on 17 dpc, and euthanized). Blue symbols, 40Fp8-vaccinated; red symbols, mock-vaccinated.

As expected, by day 7-14 all vaccinated ewes had seroconverted with protective nAb titers that increased upon the virulent challenge. The kinetics of anti-N antibodies followed a similar pattern, indicating effective priming of the immune system by the vaccine candidate (Fig. 10).

Fig. 10. Humoral responses in vaccinated and challenged ewes.

Neutralizing anti-RVFV antibodies (nAbs) detected in sera by virus microneutralization assay in cell culture. Blue symbols, 40Fp8-vaccinated; red symbols, mock-vaccinated. The dotted line indicates sensitivity threshold (left panel). Detection of serum anti-nucleoprotein antibodies by competitive ELISA (IDVet RVF screen Kit). The shaded area indicates the doubtful status as defined by the provider (right panel) Median and interquartile range (for n > 3) as well as individual values are represented.

Discussion

One of the hallmarks of RVF outbreaks in Africa is the massive occurrence of abortions in sheep flocks²². During the breeding seasons nearly 100% of the pregnant sheep abort upon a RVFV infection, resulting in heavy economic losses and social consequences^23–25. Vaccination can help to contain the impact of the disease. However, vaccines must be safe for foetuses and capable of providing sufficient maternal protection after immunization to prevent the virus spreading to the reproductive organs and foetuses. If vaccination is carried out using a live-attenuated vaccine (LAV), it should not retain residual virulence to ensure the completion of the gestational cycle. Several LAVs such as Smithburn, clone 13 or MP12 can be safely administered in adult sheep but their use during pregnancy has been questioned due to potential adverse effects on the foetuses^26–28.

LAVs can be optimal candidates for livestock vaccination programmes; however, the potential for reassortment events, reversion to (or the presence of residual) virulence has frequently been identified as concerns that could potentially contribute to disease spread, thereby limiting their use in immunosuppressed hosts or in pregnant animals^29–31. To overcome these problems, novel, rationally designed, reverse genetics-based vaccine candidates as well as other strategies have been recently developed and some of them are in early phases of clinical developments¹⁴.

We showed in previous works that the mutagenized RVFV variant 40Fp8 virus is highly attenuated even in an immunodeficient mouse model^17,20. We showed here that the inoculation with the 40Fp8 virus does not alter the basal biochemical parameters, in sharp contrast to what occurs when a virulent strain is inoculated through the same route in adult sheep, indicating that 40Fp8 can be administrated safely in sheep. However, a high level of attenuation could hamper the elicitation of an effective immune response. Therefore, here we tested firstly the immunity induced in sheep to assess its potential as a novel candidate vaccine. For these experiments we used a fixed viral stock derived from a single clone that was Sanger sequenced to confirm homogeneity of sequence. This was performed in two pilot experiments using a reduced number of sheep. Due to restrictions on animal movement, we ended up using sheep from three different Spanish breeds, sourced from various suppliers and geographical locations. All the breeds derived from a common ancestral origin and, as expected, no differences in response to the vaccine were observed. Taken together, it was possible to confirm that at the doses used here, the parenteral inoculation of 40Fp8 was able to trigger an immune response of a sufficient magnitude to induce a neutralizing antibody response to titres usually over the described protective threshold. This is important since the induction of neutralizing antibodies is a clear correlate of protection against RVF^32–34. However, the minimal immunization dosage of 40Fp8 able to induce protective antibody titers in the sheep model should be further investigated. In previous works, RVF LAV doses of 10⁴ -10⁵pfu were shown to be protective in gestational sheep models^35,36. In our hands, doses of 40Fp8 ranging 10³-10⁴ pfu, fully protected mice from a virulent challenge (unpublished observations). Due to the highly attenuated character of 40Fp8 it may be necessary to inoculate with high virus titers to ensure full protection using a single dose only. Of note, our results also show that the inoculation of high doses of the attenuated 40Fp8 virus induces cell-mediated immune responses since IFN-γ was detected after in vitro re-stimulation of 40Fp8-immune blood cells. Nonetheless, a more in-depth characterization will be needed, to ascertain if memory T-cell responses can be also detected long after immunization.

The subcutaneous immunization of pregnant ewes with RVFV-40Fp8 induced immunity levels which correlated with protection in the absence of clinical disease and low to undetectable levels of viral RNA. After virulent RVFV-56/74 challenge, no adverse effects were recorded in 40Fp8-vaccinated ewes, in contrast to mock-vaccinated animals that showed clear liver and reproductive organ pathology, foetal affectation and/or abortion. In some cases, Cq values above threshold were obtained by RT-qPCR in umbilical vein blood and in some foetal tissues. However, virus isolation trials in these samples was always negative. A possible explanation for the detection of viral RNA should be related with presence of residual RNA genomes derived from the challenge inoculum, since the dose administered was quite high, or to the presence of viral replication intermediates since, in all cases, antibodies against the viral nucleoprotein N were detected. Failure to isolate infectious virus may also indicate that it is rapidly neutralized by vaccine-induced maternal antibodies, preventing widespread dissemination to reproductive organs or more susceptible foetal tissues.

In our experimental setting, the pregnant ewes vaccinated with 40Fp8 were sacrificed at various times post-inoculation in order to screen for potential pathological alterations in reproductive and foetal tissues. Therefore, the pregnancy was not allowed to progress to full term ending with lamb delivery. Based on the results described here it should be expected that pregnant ewes vaccinated with 40Fp8 should deliver healthy lambs after a subsequent virulent RVFV challenge. More analysis will be needed to assess the duration of protection conferred by the administration of a single dose, the effect of a booster dose or the use of immunostimulatory molecules or adjuvants to improve the protective capabilities and, furthermore, reversion to virulence studies according to regulatory issues may be also addressed. These experiments should be performed ideally in a natural, RVFV-endemic setting, warranting further efforts in additional 40Fp8 safety studies, such as reversion to virulence upon serial passage in animals or vector competence studies that would impact the clinical development of 40Fp8 as a novel RVF vaccine candidate. Nonetheless, we have observed sequence homogeneity upon 10 serial passages in Vero cells, indicating that vaccine stocks can be generated in this cell line approved for vaccine manufacturing. Other studies are being conducted on the effect of reassortment events on virulence. Our investigations concluded that, although the chance of reassortment events is theoretically possible, their occurrence would be highly improbable given that the viral load in the blood is practically non-existent. Furthermore, if reassortment did occur, it would confer attenuation to the receptor virus. Finally, the dissemination of 40Fp8 in mosquitoes is also being addressed to analyse the chances of reversion in these hosts. Altogether these data will help to complete the safety studies of this vaccine candidate.

Methodology

Ethics statement

Procedures involving the use of animals were conducted in accordance with Spanish regulations on animal experimentation (RD 53/2013) and European regulations (EU directive 2010/63/EU) on the protection of animals used for scientific purposes. The animal research ethics committee (Consejo Superior de Investigaciones Científicas-CSIC) and the regional authorities of the Community of Madrid (permit number PROEX/22/06.79) approved the experimental procedures.

Biosafety

All experiments were conducted in biosafety level 3 (BSL-3) laboratories as well as in the animal biosafety level 3 (ABSL-3) facilities at the CISA-INIA. The Spanish biosafety commission granted approval for work with Rift Valley fever virus.

Cells and viruses

Vero and/or Vero E6 cell lines (ATCC #CCL-81 and #CRL-1586) were grown and subcultured in DMEM supplemented with 2-5% foetal bovine serum under conventional conditions of temperature (37 °C) and humidity (95%) with a 5% CO₂ atmosphere. The RVFV-40Fp8 virus was previously described^17,20. The original isolate was propagated upon low moi (0.02) Vero cell inoculation²⁰. The supernatant was clarified, titrated, and used as inoculum for experiments 1 and 2 (see below). For pregnant sheep inoculation (experiments 3 and 4), a larger stock of 40Fp8 was produced by amplification of a plaque-purified clone from the original isolate obtained after 8 passages in Vero cells in the presence of 40 µM of favipiravir^16,17. Briefly, several biological clones picked up from infected Vero cells were first grown in the presence of the drug, supernatants titrated and further amplified at a moi of 0.1 in the absence of the drug. 72 hpi supernatants obtained were collected, clarified, titrated, and conserved in aliquots at -80 °C. A stock designated as c1/2022 was finally selected as inoculum for the animal experiments. The genetic stability of c1/2022 was confirmed by Sanger sequencing as described³⁷. Challenge virus was isolated from an experimentally infected sheep that succumbed to infection with the virulent RVFV 56/74 South African origin isolate^38,39. This virus was further grown and propagated in C6/36 mosquito cells (CISA repository) as previously described^16,21.

Animal study design

Four experiments were conducted to confirm the immunogenicity, safety and protective efficacy of the live attenuated 40Fp8 prototype vaccine against RVFV in non-pregnant ewes and pregnant ewes of different Spanish breeds, as the ovine species is a natural host of the virus and highly susceptible to the disease (Fig. 1). To assess immunogenicity after 40Fp8 inoculation, two pilot experiments were performed using Churra ewes (experiments 1 & 2). Safety was further tested in pregnant Manchegan ewes (experiment 3) while, to test protective efficacy upon virus exposure, Castilian pregnant ewes vaccinated with 40Fp8 were challenged with the virulent 56/74 isolate (experiment 4). Animals were supplied by commercial farms with high sanitary standards. All animals underwent a 7-day acclimatization period in the BSL-3 animal facility of the Animal Health Research Center (CISA-INIA/CSIC) before starting the experiments. Clinical signs were recorded following a clinical score chart (supplementary Table 1). Scores were assigned to parameters such as hyperthermia, loss of appetite, behavior, ocular discharge, respiratory distress, digestive disturbances, coordination, and reproductive disorders on a severity scale of 0 to 3: no signs (0), mild (1), moderate (2), most severe (3). All parameter scores were added together. Total clinical scores (CS), out of a theoretical total of 70 points, were also used to define humane endpoints. Pre-determined humane endpoints included animals displaying severe clinical signs such as hyperthermia above 41 °C, lack of intake, recumbence, respiratory distress, digestive disturbances, gait abnormalities or nervous signs for more than two consecutive days, or a total CS > 16 in a single day for an animal.

Euthanasia protocols were performed in accordance with current legislation. All animals were sedated with a combination of ketamine (Anesketin, 2 mg/kg BW) and xylazine (Xilagesic, 0.1 mg/kg BW) administered intravenously. Animals were then euthanized with an overdose of intravenous pentobarbiturate (Dolethal, 20 ml).

Experiment 1 (immunogenicity). Churra ewes (n = 5), aged 2 years, were used. In this experiment, two Churra ewes (#21 and #30) were inoculated subcutaneously (SC) in the neck area near the pre-scapular lymph node with 1 ml containing 10⁷ plaque forming units (pfu) of the RVFV-40Fp8 virus stock. These ewes were euthanized on days 4 and 8 post inoculation (dpi), respectively. A second group of three ewes (#9, #12 and #69) received identical dose of the parental RVFV strain 56/74. These animals were euthanized at 4, 7, and 8 dpi for pathological evaluation. Clinical evaluation was performed daily after inoculation. EDTA blood and serum samples were taken from the jugular vein (Fig. 1a).

Experiment 2 (immunogenicity). Churra ewe lambs (n = 6), aged 4 months, were used. In this experiment, three animals (#4, #11 and #70) were SC inoculated in the neck area near the pre-scapular lymph node with 1 ml containing 1 × 10⁷ pfu of the 40Fp8 virus. Another three ewe lambs (#76, #809 and #868) were inoculated intravenously (IV) with an identical dose of 40Fp8. Animals were clinically monitored for three weeks after inoculation. EDTA blood and serum samples were taken daily during the first week or every other day (9, 11, and 13 dpi) during the second week. Two final blood and serum samples were taken at 14 and 21 dpi (Fig. 1b).

In experiments 1 and 2, the presence of virus in blood samples was examined by either virus isolation or RT-qPCR. Serum samples were used for biochemical analysis of enzymes indicative of hepatic injury and for detection of anti-nucleoprotein and/or anti-RVFV neutralizing antibodies. In experiment 2, antigen-specific cellular responses were also monitored by IFN-γ ELISA analysis after in vitro antigenic re-stimulation of EDTA blood samples taken at 14 and 21 dpi.

Experiment 3 (vaccine safety). Pregnant Manchegan ewes (n = 14), between 3 and 5 years of age, oestrus-synchronized and naturally mated, were used. Ewes were reproductively synchronized using intravaginal sponges following previously described protocols⁴⁰ and then mated naturally for 3 consecutive days, 24 h after the sponges were removed. Pregnancy was diagnosed by transabdominal ultrasonography five weeks after mating and confirmed shortly before the animals were transferred to the BSL-3 facility and by a pregnancy test on the day of inoculation (Idexx Rapid Visual Pregnancy Test kit). Although the gestation test performed on the day of inoculation revealed that ewe #18 was not pregnant, this animal was not excluded from the study. After the acclimatization period, ewes at gestational day 56 were inoculated SC with 1 ml containing 1 x 10⁷pfu of the RVFV-40Fp8 strain and euthanized in groups of n = 3 on days 4 (#12, #23, and 91 N), 7(#65, #73 and 91B), 14 (#10, #15 and #17) and 30 (#18, #26 and #40) post-immunization (dpi). Two pregnant ewes (#92 and #94) used as non-immunized controls were inoculated with mock-infected cell culture medium and euthanized at the end of the experiment (30 dpi). Throughout the experiment, clinical signs and rectal temperatures were examined daily. At the days of necropsy, blood and serum samples were collected for determination of viral loads (either by RT-qPCR or cell culture isolation) and for detection of anti-nucleoprotein and/or anti-RVFV neutralizing antibodies (Fig. 1c).

During the necropsies of the ewes and their foetuses, macroscopic evaluations of the lesions were performed. When present, lesions were scored according to their extent in the tissues and the severity of the lesion itself between 1 (focal/minimal lesions) and 4 (diffuse/severe lesions). To compare the level of foetal development, all foetuses found were measured and weighed. Samples were aseptically taken from the liver, spleen umbilical cord and placentomes of ewes, and from the liver, spleen and brain of foetuses. Amniotic fluid and blood from umbilical cords were also sampled. Additional samples were collected from any other organ in which macroscopic changes were observed. All samples were kept on ice during the necropsies and subsequently stored at −80 °C. Tissue samples were also fixed in 4% buffered formalin solution for 72 h, routinely processed and embedded in paraffin wax for subsequent routine histopathological studies (haematoxylin and eosin staining), as well as immunohistochemistry to visualize viral antigen using a polyclonal rabbit anti-RVFV serum, as described¹⁹.

Experiment 4 (vaccine protective efficacy). Pregnant Castilian ewes (n = 12), between 4 and 6 years of age, reproductively synchronized and naturally mated, were used. Pregnancy was confirmed by ultrasound prior to transferring the animals to the BSL-3 animal facility and by a pregnancy test on the day of inoculation (Idexx Rapid Visual Pregnancy Test kit). After the acclimatization period, nine ewes at gestational day 45 were inoculated with the RVFV-40Fp8 virus following the same immunization protocol as for experiment 3. Three pregnant ewes (#4, #5 and #6) were used as mock-immunized controls. Twenty-four days after immunization all animals were SC challenged with 1 mL containing 10⁷ pfu of the RVFV virulent strain 56/74. Throughout the experiment, animals were monitored clinically daily. Blood and serum samples were taken on different days after immunization (dpi) and after challenge (dpc) and used for the same assessments as described in experiment 3. Programmed necropsies were performed on 3 immunized ewes and 1 non-immunized control ewe at day 4 (ewe #5, #7, #8, #9), 7 (ewe #6, #10, #11, #12), and 21 (ewe #1, #2, #3, #4) post-challenge (dpc). The level of foetal development, macroscopic evaluations, tissue sampling and subsequent histopathological and immunohistochemical assessments were similar to those described for experiment 3.

Blood chemistry

Blood samples were collected in BD Vacutainer tubes without anticoagulant. For separation of serum, the tubes were centrifuged at 1.267 × g for 10 min and stored at −80 °C until use. The sheep serum levels of aspartate aminotransferase (AST), gamma-glutamyltransferase (GGT) and lactate dehydrogenase (LDH), collected between 0 and 7 days post inoculation (experiment 1), were determined with a Saturno 100 analyzer, (Crony instruments) using specific reagents (BioAnalítica SL) according to the manufacturer’s instructions.

Molecular, virological and serological assays

Different organ and blood samples were prepared and tested for the presence of RVFV RNA by RT-qPCR. Organ samples were processed to obtain 10% weight/volume homogenates in DMEM, by running one homogenization cycle (50 Hz oscillation frequency (50 cycles/s) for 4 minutes) in a Tissue lyser LT (Qiagen). Tubes were then centrifuged at 5.000 rpm for 5 min, supernatants collected and stored at -80 °C until use. Viral RNA was isolated with the Speedtools system according to the manufacturer’s instructions (Biotools, Spain) from either 0.5 mL of blood or 0.5 mL of a 10% organ homogenate. One to five microliters of purified RNA were subsequently used for RT-qPCR using the Scriptools RT-qPCR assay (Biotools, Madrid, Spain) in combination with an Eco-Real time PCR thermocycler (Illumina). One-tube RT-qPCR was performed using the forward L segment primer (5’-TTCTTTGCTTCTGATACCCTCTG-3’), L segment reverse primer (5’-GTTCCACTTCCTTGCATCATCTG-3’) and a FAM-BHQ1-labeled probe (5’-FAM-TTGCACAAGTCCACACAGGCCCCT-BHQ1-3’) as described⁴¹.

In vitro virus isolation was attempted in those RT-qPCR positive samples to confirm or to rule out the presence of infectious virus. Briefly, 0.1-0.5 mL of whole blood or clarified organ homogenates was adsorbed for 1 hour to semi confluent Vero cell monolayers in 25 cm² tissue culture flasks. After adsorption, cells were extensively washed with PBS and incubated for 5 days. In case of absence of cytopathic effect (CPE), this procedure was repeated up to three blind passages.

Detection of RVFV serum nucleoprotein N-specific antibodies was determined by an ID Screen® Rift Valley Fever Competition Multi-species (IDVet, France). The kit discriminates positive from negatives base on an inhibition percentage. The percentage of competition is calculated as the ratio between OD (sample)/OD (negative control) x 100. Values under 40% determine a positive status, while those above 50% determine a negative status. Values between 40% and 50% are recorded as “doubtful”. The titers of RVFV serum neutralizing antibodies (nAbs) were determined using a virus microneutralization test (VNT). Briefly, in a 96-well plate format, serial dilutions (50 μl) of heat-inactivated sera (2 h, 56 °C), in 2–4 replicas, were incubated with 50 μl of the attenuated RVFV-56/74 or 40Fp8 strain (100 TCID₅₀/ml) for 2 h at room temperature. Subsequently, 2.5 × 10⁴ Vero cells (in 50 μl) were added to each well. Plates were incubated for 3–4 days at 37 °C and 5% CO₂ and CPE scored under microscope observation. VNT₅₀ titres were calculated using the Reed and Muench 50% end-point method.

IFN-γ detection assay

An in vitro re-stimulation assay was performed in MW24 plate wells using 0.5 mL of whole EDTA blood samples per well, which were incubated with 1 µg/mL of either recombinant purified proteins Gn, or N, or 5 µg of Gn-derived peptides #19 (203-FQSYAHHRTLLEAVH-217) and #21 (211-TLLEAVHDTIIAKAD-225), or Gc-derived peptides #226 (1031-AAFLNLTGCYSCNAG-1045) and #253 (1139-SWNFFDWFSGLMSWF-1153)²¹. The plates were incubated at 37 °C in a 5% CO₂ incubator for three days. The plasma was recovered by centrifugation of plates (300 x g) and tested in a IFNγ capture sandwich ELISA assay as described^19,21. Briefly the plasma from both SC and IV inoculated ewes (experiment 2) were collected at 14 and 21 dpi and added to the ELISA plate wells previously coated with anti-IFNγ mAb purified (MT17.1, Mabtech). Upon incubation and washing steps a detector anti-bovine IFNγ mAb conjugated with biotin (MT307, Mabtech) was added. Streptavidin-HRPO (Becton-Dickinson) was used for detection of immunocomplexes upon addition of TMB peroxidase substrate (Sigma-Aldrich). Absorbance values were determined at 450 nm by an automated plate reader (BMG, Labtech).

Data visualization and analysis

This study was exploratory in nature, using a low number of individuals for complying with ethics, welfare and strict biosafety regulations. Statistical methods were not applied due to the insufficient sample size considering high variability between subjects. We focused instead on describing trends and patterns rather than seeking statistical significance. The valuable information obtained in this work is therefore based mainly on descriptive and qualitative analyses. The graphical representation of the results, both individual and mean values, was carried out with GraphPad Prism Version 7.0 (GraphPad Software, La Jolla, CA, USA).

Author contributions

B.B. produced the vaccine stock. A.B., P.S.-C. and B.B. searched sheep sources, designed and supervised the animal experiments. C.A., S.M., G.L., E.C.-P., B.B., A.B. and P.S.-C., performed analytical assays (molecular, biochemical, immunological and virological), conducted necropsies, and provided pathology, histopathology and immunohistochemical data. A.B., B.B. and P.S.-C. supervised data analysis. A.B., B.B. and P.S.-C. wrote drafts and final manuscripts. A.B. and B.B. obtained the necessary financial resources for the project, while A.B. was responsible for overseeing its management. All authors read and agreed to the published version of the manuscript.All authors reviewed the manuscript.

Data availability

Data supporting the findings of this study are available within the paper and its Supplementary Information. Any other secondary data source is available upon request to authors.

Competing interests

A.B., B.B. and S.M. are the inventors of patent WO/2021/245313 (application PCT/ES2021/070403), which was filed with the Spanish National Research Council (Agencia Estatal Consejo Superior de Investigaciones Científicas, CSIC) and is currently being extended to several third countries. This manuscript supports the safety and efficacy of a vaccine candidate that is the subject of patent protection. All other authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41541-025-01250-6.