Fig. 3: Effects of bloodmeal with and without nearby ZIKV related-odour (invasion or transmission blends) on mosquito feeding behaviour and fitness.

a–g Females were fed on bloodmeal by membrane feeder without (control) or with nearby placed ZIKV related-odour, invasion blend or transmission blend. a Schematic model of experiments. b Average meal size was determined by the amount of hematin excreted and normalized individually to wing length [χ²₁ = 50.60, p < 0.001; Control vs Invasion z = 3.15, df = 1, p < 0.001; Invasion vs Transmission z = 4.62, df = 1, p < 0.001; Control vs Transmission z = 7.01, df = 1, p < 0.001]. c Blood feeding proportion (%) after exposure to feeders impregnated with the ZIKV related-odours at invasion [Control vs Invasion z = 4.25, df = 1, p < 0.001] and transmission [Invasion vs Transmission z = 4.59, df = 1, p < 0.001; Control vs Transmission z = 6.59, df = 1, p < 0.001] stages in comparison to control [χ²₁ = 106.62, p < 0.001]. d Accumulated blood feeding proportion (%) of mosquitoes was monitored per 10 seconds (Sec) at the first minute of the experiment and then continued per minute on feeders with/without nearby ZIKV related-odour at the invasion stage (invasion blend) until 800 Sec, [Cox hazard proportional model, χ²₁ = 217·50, p < 0.0001]. e Accumulated blood feeding proportion (%) was monitored per 10 seconds (Sec) at the first minute of the experiment and then continued per minute on feeders with/without nearby ZIKV related-odour at transmission stage until 800 Sec, [Cox hazard proportional model, χ²₁ = 22.68, p < 0.0001]. f The relationship between fecundity (number of eggs per mosquito) and blood meal size (hematin) [treatment*hematin: χ²₂ = 6.27, p < 0.001; Invasion vs Transmission: χ²₁ = 0.19, p = 0.66; Control vs Transmission/Invasion: χ²₁ = 6.91, p = 0.009/χ²₁ = 7.83, p = 0.005] of individual mosquitoes. g Survival of mosquitoes was monitored daily post feeding until natural death, [Cox hazard proportional model, χ²₁ = 100.71, p < 0.001; Invasion vs Transmission: χ²₁ = 12.19, p < 0.001; Control vs Transmission: χ²₁ = 114.22, p < 0.001; Control vs Invasion: χ²₁ = 15.09, p < 0.001]. These experiments were done in triplicate (random effect). Significant effects were determined using GLMM models (lmer) or Cox hazard proportional model [in all experiments: n = 30/experimental replication]. Bars represented by β-estimation generated by the mixed model ± SE (β-lmer ± SE); asterisks denote significant differences (*p < 0.05; **p < 0.01; ***p < 0.001).