The Rift Valley fever (RVF) vaccine candidate 40Fp8 shows an extreme attenuation in IFNARKO mice following intranasal inoculation

Abstract

Rift Valley fever (RVF) is an important zoonotic viral disease affecting several species of domestic and wild ruminants, causing major economic losses and dozens of human deaths in various geographical areas of Africa, where it is endemic. Although it is not present in Europe, there is a risk of its introduction and spread linked to globalisation and climate change. At present, the only measure that could help to prevent the disease is vaccination of flocks in areas at risk of RVF. Available live attenuated vaccines are an effective means of controlling the disease, but their use is often questioned due to residual virulence, particularly in susceptible hosts such as pregnant sheep. On the other hand, no vaccine is currently licensed for use in humans. The development of safe and effective vaccines is therefore a major area of research. In previous studies, we selected under selective mutagenic pressure a highly attenuated RVFV 56/74 virus variant called 40Fp8. This virus showed an extremely attenuated phenotype in both wild-type and immunodeficient A129 (IFNARKO) mice, yet was still able to induce protective immunity after a single inoculation, thus supporting its use as a safe, live attenuated vaccine. To further investigate its safety, in this work we have analysed the attenuation level of 40Fp8 in immunosuppressed mice (A129) when administered by the intranasal route, and compared it with other attenuated RVF viruses that are the basis of vaccines in use or in development. Our results show that 40Fp8 has a much higher attenuated level than these other viruses and confirm its potential as a candidate for safe RVF vaccine development.

Fig 1. Intranasal inoculation of A129 mice with attenuated viruses.

Groups of n = 8 A129 mice 8–12 weeks old were inoculated by the IN route with a low (L) or high (H) dose (50 or 1000 pfu, respectively, full and empty symbols) of the indicated viruses: 40Fp8 (blue), rG1 (green) and rMP-12 (black). Each group included 4 Female (F) and 4 Male (M). Doses were confirmed by titration of the inocula. Animals were monitored daily to check for signs of disease. (a) Survival curves. Percentages of survival are represented using Kaplan-Meier plots. Survival distributions were compared by Log-rank (Mantel-Cox) test. The survival curves are significantly different with p <0.0001. (b) Weight variation. For each individual mouse, weight at the beginning of the experiment was considered as 100% (thick line) and used to calculate the percentage of variation along the experiment. Results of the group are represented in the “box and whiskers” format, with individual values as black points, and the median plotted as a line in the middle of the box. Results are only shown for the groups inoculated with 40Fp8. Left panel (full boxes), animals receiving the high dose; right panel (empty boxes), animals receiving the low dose. (c) Clinical signs. Signs observed were classified as mild (ocular watery discharge and/or ruffled hair), or severe (strongly reduced activity and/or paralysis and/or tremors) that led to euthanasia of the animal. FD = found dead. The area corresponding to days 2 to 5 pi has been red-shaded to highlight the period when deaths occured in the rG1 and rMP-12 groups.

Fig 2. Histopathological studies.

Animals succumbing or euthanized along the experiment at the days pi indicated were selected for histopathological (HP) and immunohistochemical (IHC) studies. Results shown correspond to selected/representative individuals within each virus group, identified by sex (M/F) and viral dose received (H/L) as in the legend of Fig 1. Data for each animal are shown in S1 Table. (a) Individual representation of the cumulative scores of cells immunolabelled against RVF virus antigen in tissue samples obtained on different days pi from selected mice within each virus group. Immunolabeled cell score (y-axis); Mice evaluated in each virus group (x-axis). The borders of the bars have been coloured according to the virus inoculated: green: rG1; black: rMP-12; blue: 40Fp8. For graphical purposes, animals inoculated with 40Fp8 have been assigned a value of 0.5 although all animals had a value of 0. (b-i) Representative images of tissue sections immunolabelled against RVF virus antigen. Massive presence of cells immunolabelled against RVF virus antigen in the liver (b, c) and spleen (d, e) of rG1 (b, d) and rMP-12 (c, e) inoculated mice. Note a slightly lower presence of viral antigen in the rMP-12-inoculated mouse (c, e). Observe also the occasional presence of immunolabelled cells in the kidney of rG1 (f; arrows) inoculated mice, as well as the presence of labelled circulating mononuclear cells within the meningeal blood vessels (g; arrows). Neither in the liver (h) nor in the brain (i) of the 40Fp8-inoculated mouse euthanised on day 9 pi were cells immunolabelled against the viral antigen detected. Note the severe menigitis consisting mainly of lymphocyte infiltrates (i; arrow). Liver (b) and spleen (d) of rG1(FD) F_H_3 mouse (3 dpi); Liver (c) and spleen (e) of rMP-12(EU) F_H_1 mouse (3 dpi); Kidney (f) of rG1(EU) M_L_3 mouse (3 dpi); Brain (g) of rG1(FD) F_L_2 mouse (4 dpi); Liver (h) and brain (i) of 40Fp8 (EU) F_H_4 mouse (9 dpi). IHC, scale bars: 50 micrometers (f, g, i); 200 micrometers (b, c, d, e, h).

Fig 3. Analysis of samples from IN inoculated mice.

(a) Viral load as determined by RT-qPCR in blood samples collected at day 3 pi from mice inoculated with a high (H, full symbols) or low (L, empty symbols) dose (1000 or 50 pfu, respectively) of virus 40Fp8 (blue), rG1 (green) or rMP-12 (black). Each point corresponds to an individual. Bars indicate the mean value. The area defined by negative control samples (negative controls in each round of PCR giving Cq values between 33 and 40) has been shaded. A line has been depicted at Cq = 37, considered arbitrarily as the detection level of the assay. Statistical significance determined by multiple t-tests (Holm-Sidak method). P values: 40Fp8 vs rG1: 0.000005 (high dose), 0.0005 (low dose); 40Fp8 vs rMP-12: 0.0025 (high dose), 0.4676 (low dose). (b) Neutralizing antibodies in survivors inoculated with 40Fp8. Serum dilution (log2) at which 50% of the wells were protected from cpe as observed in control wells. H (filled squares): animals inoculated with the high dose, i.e. 1000 pfu; L (empty squares): low dose, 50 pfu. Each point corresponds to an individual. Bars indicate the mean value. Samples were tested from 1/10 dilution (sensitivity limit of the assay, indicated by a dotted line) and then serially diluted 2-fold to 1/320. For the purpose of the graph, samples in which cpe was observed at the first dilution tested were considered negative and assigned an arbitrary value of 1/5 dilution; similarly, samples that still showed total neutralisation at the last dilution tested were not further tested and assigned a value of 1/640 dilution. Statistical significance determined by Mann-Whitney test. P value: 0.0023. (c) AntiN ELISA was performed only on samples from the low dose group (empty circles). Samples were tested in serial 3-fold dilutions from 1/50; graph shows the OD reading values given by samples diluted 1/150. The line corresponds to the cut-off value of 0.150, established as 2x the reading of the blank.

Fig 4. Histopathological studies on schedule euthanized mice.

11–15 week old IFNARKO mice, including both males and females, were IN inoculated with 10 microliters of viral suspension containing 1000 pfu of rG1 (n = 2), rMP-12 (n = 2) and 40Fp8 (n = 7). Animals were assessed for signs of disease and survival only, and euthanized on days 2, 5 and 9 pi. Based on previous survival results, groups receiving rG1 and rMP-12 were scheduled only for day 2. Brain and liver samples were collected for virus isolation and RT-qPCR while liver, brain, spleen and kidney samples were fixed in 4% buffered formalin solution for histopathological (HP) and immunohistochemical (IHC) studies. (a) Individual representation of the cumulative scores of cells immunolabelled against RVF virus antigen in tissue samples obtained on day 2 pi from mice infected with rG1, rMP-12 and 40Fp8. Immunolabeled cell score (y-axis); Mice evaluated in each virus group (x-axis). The borders of the bars have been coloured according to the virus inoculated: green: rG1; black: rMP-12; blue: 40Fp8; (b-g) Representative images of liver sections immunolabelled against RVF virus antigen. Note the occasionally presence of immunolabeled cells, mainly hepatocytes, in the liver of rG1 (b, inset; arrows) and rMP-12 (c; arrows) inoculated mice on day 2 pi, as well as the absence of viral antigen in the liver of 40Fp8 inoculated mice on the same date (d). In mice inoculated with 40Fp8 and euthanized on day 5 pi (e, f), viral antigen was only detected ocassionally in the liver of one mouse (f, inset; arrows). Viral antigen was not present in any of the tissue sections taken from the mice euthanized on day 9 pi (g). (b) Liver, rG1#2 mouse, 2 dpi; (c) Liver, rMP-12#1 mouse, 2 dpi; (d) Liver, 40Fp8#1 mouse, 2 dpi; (e) Liver, 40Fp8#1 mouse, 5 dpi; (f) Liver, 40Fp8#2 mouse, 5 dpi; (g) Liver, 40Fp8#1 mouse, 9 dpi. IHC, scale bars: 50 micrometers (c); 100 micrometers (insets: b,f); 200 micrometers (b,d,e,f,g).