Influence of RVFV Infection on Olfactory Perception and Behavior in Drosophila melanogaster

Abstract

In blood-feeding dipterans, olfaction plays a role in finding hosts and, hence, in spreading pathogens. Several pathogens are known to alter olfactory responses and behavior in vectors. As a mosquito-borne pathogen, Rift Valley Fever Virus (RVFV) can affect humans and cause great losses in livestock. We test the influence of RVFV infection on sensory perception, olfactory choice behavior and activity on a non-biting insect, Drosophila melanogaster, using electroantennograms (EAG), Y-maze, and locomotor activity monitor. Flies were injected with RVFV MP12 strain. Replication of RVFV and its persistence for at least seven days was confirmed by quantitative reverse transcription-PCR (RT-qPCR). One day post injection, infected flies showed weaker EAG responses towards 1-hexanol, vinegar, and ethyl acetate. In the Y-maze, infected flies showed a significantly lower response for 1-hexanol compared to uninfected flies. At days six or seven post infection, no significant difference between infected and control flies could be found in EAG or Y-maze anymore. Activity of infected flies was reduced at both time points. We found an upregulation of the immune-response gene, nitric oxide synthase, in infected flies. An infection with RVFV is able to transiently reduce olfactory perception and attraction towards food-related odors in Drosophila, while effects on activity and immune effector gene expression persist. A similar effect in blood-feeding insects could affect vector competence in RVFV transmitting dipterans.

Keywords: Rift Valley Fever Virus; arbovirus; electroantennography; neuro-immune-interaction; olfactory choice test.

Figure 1

Experimental schedule. Female cnbw flies were injected with 1000 focus forming units (FFU) of Rift Valley Fever Virus (RVFV) and either incubated for 1 or 6 to 7 days. Changes in odor perception was measured via electroantennography (EAG). For EAG measurement, flies were fixed in a cut pipette tip. Odor stimuli were delivered in an air stream and antennal response was recorded via electrodes placed on the antenna. Changes in behavior towards odors were tested in the Y-maze. 20 flies were put into the small start tube connected via a Y-piece to the odor tubes. After 24 h, flies were counted and responsive indices (RI) determined. Changes in locomotor activity were traced with a locomotor activity monitor (LAM). Single flies were placed into a glass tube with sucrose agar. Movements were detected by three infrared light beams for 24 h. Infection status was confirmed via quantitative reverse transcription-PCR (RT-qPCR).

Figure 2

Electroantennography. Stimulus-response curves of flies infected with RVFV on 1 day post injection (dpi) or 7 dpi. (A) Recording example of traces from a dilution series of 1-hexanol with marked stimulus delivery of 0.5 s (grey line) with 10 s interstimulus intervals in between. (B–E) EAG measurements of responses to 1-hexanol (B,C) or ethyl acetate (D,E) and apple cider vinegar (ACV) as positive control. Mean antennal response ±SEM. Dotted lines indicate threefold SD of the response to paraffin oil (blank). (B) n(control) = 11; n(RVFV) = 9; (C) n(control) = 13; n(RVFV) = 12; (D) n(control) = 10; n(RVFV) = 8. (E) n(control) = 13; n(RVFV) = 10. Statistical analysis: multiple t-tests, ns ≥ 0.05, p < 0.05 (*).

Figure 3

Y-maze odor attraction assay. RIs of flies confronted with different concentrations of tested compounds. RIs of naive flies tested with different concentrations of (A) 1-hexanol, (B) ACV and (C) ethyl acetate against diluent as control. (A) n(125 µg/mL) = 21; n(60 µg/mL) = 30; n(12.5 µg/mL) = 31 (B) n(100%; 50%; 10%) = 10 (C) n(1000 µg/mL) = 12; n(60; 6; 1 µg/mL) = 5; n(0.1 µg/mL fed or starved) = 15. Error bars display SEM. Statistical test: two-sample t-test (between concentrations), one-sample t-test (concentrations compared to zero), ns ≥ 0.05, p < 0.05 (*), p < 0.01 (**), p < 0.001 (***).

Figure 4

Y-maze odor attraction assay. RIs to 1-hexanol (A) and ACV (B). RIs of controls and RVFV injected flies tested with 1-hexanol (60 µg/mL) and ACV (100%) against diluent as control at 1 and 7 dpi. Shown are five biological replicates each representing the average of five Y-mazes ±SEM. Statistical test: t-test, ns ≥ 0.05, p < 0.05 (*).

Figure 5

Locomotor activity at 1 dpi (A) and 7 dpi (B). Averaged (William’s mean) activity counts per hour of control (blue) and RVFV (red) infected flies ± SEM (shaded areas). At each time point, each group consists of 3 biological replicates; each replicate includes 16 flies. Grey boxes indicate significant differences between groups at individual hours post start at marked time points. Statistical analysis: t-tests, ns ≥ 0.05, p < 0.001 (***).

Figure 6

Fold change of dNOS expression in RVFV-injected flies normalized to controls. 2 dpi: 1.49 ± 0.41 (±SEM, n = 5); 8 dpi: 1.96 ± 0.56 (±SEM, n = 4).