Experimental infection of rhesus macaques and common marmosets with a European strain of West Nile virus

Abstract

West Nile virus (WNV) is a mosquito-borne flavivirus that infects humans and other mammals. In some cases WNV causes severe neurological disease. During recent years, outbreaks of WNV are increasing in worldwide distribution and novel genetic variants of the virus have been detected. Although a substantial amount of data exists on WNV infections in rodent models, little is known about early events during WNV infection in primates, including humans. To gain a deeper understanding of this process, we performed experimental infections of rhesus macaques and common marmosets with a virulent European WNV strain (WNV-Ita09) and monitored virological, hematological, and biochemical parameters. WNV-Ita09 productively infected both monkey species, with higher replication and wider tissue distribution in common marmosets compared to rhesus macaques. The animals in this study however, did not develop clinical signs of WNV disease, nor showed substantial deviations in clinical laboratory parameters. In both species, the virus induced a rapid CD56dimCD16bright natural killer response, followed by IgM and IgG antibody responses. The results of this study show that healthy rhesus macaques and common marmosets are promising animal models to study WNV-Ita09 infection. Both models may be particularly of use to evaluate potential vaccine candidates or to investigate WNV pathogenesis.

Figure 1. Time schedule.

On each time line, blue arrows indicate the time-points of the implantation of the data loggers for temperature registration, the intradermal WNV infection, and euthanasia. Red arrows indicate the bleeding time-points in the follow-up period. The numbers on the time line represent the days post-infection. Names of the animals are depicted with the “R”-animals being rhesus macaques and the “M”-animals being common marmosets.

Figure 2. Viral kinetics and body temperature.

Viral kinetics and body temperature of animals during (A) 14-days follow-up, and (B) 3-days follow-up. Dark blue areas represent the quantification of the WNV RNA load (RNA copies per ml, plotted on the left y-axis). The yellow lines represent the body temperature (in °C on the right Y-axis). Both parameters are plotted against time in days (x-axis). Due to technical failure of the temperature registration software only the body temperature registration of R01034 could be retrieved from the data logger. (C) Peak virus load was calculated for rhesus macaques and compared with the peak virus load of marmosets. (D) Total virus production in rhesus macaques and marmosets from the 14-days infection study calculated from the area under the curve (AUC). (E) Total virus production in rhesus macaques and marmosets from the 3-days infection study calculated from the AUC.

Figure 3. Humoral response after WNV infection.

WNV-E-protein-specific IgM and IgG levels of the individual animals during WNV infection. The antibody binding was calculated as the absorbance at 450 nm minus the absorbance at 520 nm. The mean value of two independent measurements of 1∶50 diluted samples is depicted in the figure.

Figure 4. Characterization of NK subsets during WNV infection.

(A) Representative example of the gating strategy. CD3−, CD45+, CD14− cells were selected from the lymphogate. Depending on the expression of CD56 and CD16 NK-cells were divided into three subsets. (B) Full circles represent the total NK population and from each individual animal the fraction of CD56^bright, CD16^bright and D16^negCD56^neg of total NK population is shown at 4 time points after WNV infection. (C) Percentage of CD16^bright NK-cells of total lymphocyte population. (D) CD161 expression on CD16^bright NK-cells. (E) NKG2A expression on CD16^bright NK-cells. (F) NKp44 expression on CD16^bright NK-cells.