Total mosquito activity at untreated and treated nets

A total of 3725 mosquitoes were assessed over 745 WHO cone tests using strains of KS, NG, BF and VK7 exposed to P2, OS, P3, IG2 and UT nets. Mosquitoes exposed to UT nets had a propensity to crawl along the net surface (Fig.2a). The mean total movement events observed during cone tests on UT nets was 4175 (SD 2154) in KS, 4671 (SD 1655) in NG, 3636 (SD 1619) in BF and 1975 (SD 1769) in VK7 (Fig.3). Susceptible strains typically had higher total activity than resistant strains. During exposure to treated nets, mosquito activity became dispersed throughout the cone and crawling on the net surface was reduced (Fig.2b). The mean total movement events (Fig.3) during ITN tests by strain was 4905 (SD 1120) in KS, 5326 (SD 904) in NG, 4022 (SD 1470) in BF, and 2737 (SD 1582) in VK7. Significant differences between the total movement comparing UT nets and ITNs by strain are indicated in Additional file 2: Table S1.

Composite outputs from individual ViCTA analyses demonstrate behavioural differences between untreated and insecticide-treated bednets. a composite summary image using the Kisumu strain of An. gambiae (KS) on an untreated net constructed by sampling video every 0.1s and merging darkest components of each frame, b Composite summary output of KS on net treated with Interceptor G2 (IG2), c composite image of Ngousso strain of An. coluzzii (NG) on untreated net, d Composite summary output of NG exposed to PermaNet 3

Total mosquito movement activity for all net treatments grouped by strain. Baseline behaviour on UT net is at the left side of each panel; 95% CI are indicated by error bars

The regional occupancy of mosquitoes in cones revealed variation in the responses of Anopheles spp. to different ITNs in terms of absolute (Fig.4a) and relative (Fig.4b) occupancy. During exposure to P2, mosquitoes were 1.482.11 times more active in the upper half of the cone (UHC) than in the UT net (Table 1, P2 vs UT upper region: KS OR 2.11; 95% CI 1.64, 2.7; P0.0010; NG OR 1.78; 95% CI 1.39, 2.27; P0.0010; BF OR 1.48; 95% CI 1.14, 1.92; P0.0010; VK7 OR 1.67; 95% CI 1.29, 2.15, P0.0010). Although activity was increased, there were no significant differences in the proportion of time spent in the upper and lower parts of the cone compared to the UT net (borderline result for KS Table 2 P2 vs UT: KS OR 0.51; 95% CI 0.25, 1.02; P=0.0582).

Regional activity in ViCTA analysis. a Mean absolute regional activity in LHC (bottom) vs UHC (top) regions for all strains and treatments. b Mean proportional regional activity. Baseline behaviour on UT net is shown in the top rows. 95% CI are indicated by error bars

In OS tests, all strains spent 1.552.67 times more time in the UHC than in the UT net, and the proportional LHC occupancy was 6675% lower (Table 1, OS vs UT upper region: KS OR 2.67; 95% CI 2.1, 3.41; P0.0010; NG OR 2.54; 95% CI 2.01, 3.2; P0.0010; BF OR 1.75; 95% CI 1.31, 2.34; P0.0010; VK7 OR 1.55; 95% CI 1.21, 1.98, P0.0010; Table 2 OS vs UT KS OR 0.34; 95% CI 0.25, 1.02; P=0.0582; NG OR 0.59; 95% CI 0.29, 1.18; P=0.1922; BF OR 0.56; 95% CI 0.26, 1.22; P=0.2117; VK7 OR 0.72; 95% CI 0.36, 1.42, P=0.5579); in P3 tests, mosquitoes spent 1.483.6 times more time in the UHC and were 65% less likely to move in the LHC (Table 1, P3 vs UT upper region: KS OR 2.79; 95% CI 2.16, 3.60; P0.0010; NG OR 3.38; 95% CI 2.67, 4.27; P0.0010; BF OR 1.48; 95% CI 1.13, 1.94; P=0.0015; VK7 OR 2.43; 95% CI 1.91, 3.08, P0.0010; Table 2, P3 vs UT: KS OR 0.35; 95% CI 0.18, 0.71; P0.0010; NG OR 0.23; 95% CI 0.11, 0.46; P0.0010; BF OR 0.55; 95% CI 0.25, 1.21; P=0.1930; VK7 OR 0.23; 95% CI 0.11, 0.45, P0.0010). Insecticide-susceptible and insecticide-resistant strains exhibited different behaviours during exposure to IG2: in the susceptible strains, between 50 and 74% less activity was observed in the LHC during IG2 tests compared to the UT net (Table 2 IG2 vs UT: KS OR 0.50; 95% CI 0.25, 0.99; P=0.0470; NG OR 0.26; 95% CI 0.13, 0.52; P0.0010), which was not observed in the two resistant strains (Table 2 IG2 vs UT: BF OR 1.74; 95% CI 0.8, 3.75; P=0.2372; VK7 OR 0.55; 95% CI 0.28, 1.08; P=0.1025).

Although mosquito resting could be indirectly inferred as the reciprocal of mosquito movement activity counts per frame (5n, where n is the number of moving mosquitoes detected per frame), a stricter measure was used. A resting frame was defined as a video frame where none of the five mosquitoes moved, compared to the previous frame. The total number of resting frames was calculated for each assay. Resting behaviour showed stark differences between susceptible and resistant strains (Fig.5) with VK7 mosquitoes found to be resting for a large proportion of the cone assays.

Mean mosquito resting period by strain and treatment. Seconds count based on number of measured frames where none of the five mosquitoes moved compared to the previous frame, out of a total of 1800 frames over the 3min of the cone test. Numbers adjacent to each bar indicate mean number of seconds resting

Over 92% of mosquitoes successfully blood-fed after tests with UT nets, with most mosquitoes feeding at 1h [KS 91.7% (176/192); NG 96% (192/200); BF 95.88% (93/97) and VK7 93.43% (199/213)]. After ITN tests, none of the mosquitoes exposed to P3, or the susceptible strains exposed to P2 and OS, survived long enough to feed. Of the few KS and NG mosquitoes that lived long enough to take a blood meal after IG2 tests, approximately half fed at 1h PE (PE) [KS 53.85% (7/13), NG 53.33% (16/30)]. In the resistant strains, compared to UT nets, BF mosquitoes were at least 90% less likely to feed at 1h PE when exposed to ITNs (Additional file 2: Table S2 BF: IG2 OR 0.10; 95% CI 0.00, 2.56; P=0.2325; OS OR 0.01; 95% CI 0.00, 0.25; P=0.0028; P2 OR 0.00; 95% CI 0.00, 0.02; P0.0010). P2 had the largest immediate inhibitory effect [blood-feeding success=9.8% (9/92)], followed by OS [48.57% (17/35)] and IG2 [61.86% (73/118)]. At 24h PE, approximately one third of BF mosquitoes (31.11%, 14/45) that were unfed at 1h after IG2 exposure, and all mosquitoes that were unfed at 1h after exposure to P2 and OS tests, blood-fed (P2 n=83; OS n=18).

ITN exposure did not have a significant effect on the blood-feeding behaviour of VK7 mosquitoes at either 1h or 24h PE. Most mosquitoes successfully fed at 1h [P2: 93.90% (77/82); OS: 87.50% (42/48); IG2 62.07% (18/29)].

The mean total weight of blood meals taken after tests with UT netting was 12.47g (SD=7.64). BMW per strain on UT netting was as follows: KS 12.26g (SD=7.89), NG 11.74g (SD=6.13), BF 10.48g (SD=6.42) and VK7 13.10g (SD=8.93). After ITN exposure, blood meal weights decreased (range 6.86g to 12.20g Fig.6 and Additional file 2: Table S3), significantly so in NG and VK7 strains (NG 2=5.47; df=1; P=0.0193; VK7 2=13.94; df=3; P=0.0030), where, compared to UT, at least 4.0g less blood was ingested (Additional file 2: Table S3 NG IG2=4.44g; 95% CI 8.23, 0.65; P=0.0225, VK7: OS=4.06g; 95% CI 6.84, 1.27; P=0.0045 and P2=4.19g; 95% CI 7.12, 1.36; P=0.0042). Activity in the LHC during tests was significantly associated with smaller blood meal weights in VK7 mosquitoes, and blood meals became 9.3 times smaller as activity increased (estimate=9.35g; 95% CI 17.76, 0.93; P=0.0298) regardless of treatment.

Mean haematin concentration of blood-fed mosquitoes which took a blood meal at 1h and 24h. Error bars at 95% CI are included only for cases where n>2. Numbers adjacent to each bar indicate sample size

In VK7 mosquitoes that could not feed at 1h PE but which recovered to feed at 24h PE, blood meal weights were 1.45 times smaller blood meals compared to those that fed at 1h PE (estimate=1.45g; 95% CI 2.93, 0.03; P=0.0551). The sizes of individual mosquitoes had a significant effect on blood meal weights in the NG, BF and VK7 strains (NG 2=5.94; df=1; P=0.0148; BF 2=17.55; df=1; P<0.0001; and VK7 2=11.18; df=1; P=0.0008) and as size increased, so did the size of the blood meal (NG estimate=5.49; 95% CI 0.91, 10.03; P=0.0201, BF estimate=8.28; 95% CI 4.09, 12.47; P0.0010 and VK7 estimate=4.19; 95% CI 1.73, 6.65; P0.0010).

After UT net tests, the 24-h mortality was 0.485.45% and the median longevity was 1214days (Additional file 2: Table S4, Additional file 2: S5). For all ITNs, 24-h mortality for KS and NG was 85.37100% and 3.4620.83% for resistant BF and VK7 (Additional file 2: Table S4). No mosquitoes survived P3 exposure and no susceptible mosquitoes survived P2. The median longevity for resistant strains after ITN tests was 914days (Additional file 2: Table S5). KS and NG mosquitoes were at least 47.55 times more likely to die when exposed to IG2 or OS compared to a UT net (Additional file 2: Table S4).

The longevity of KS mosquitoes was influenced by activity during tests and mosquitoes were 11.93 times more likely to die as the LHC activity increased regardless of treatment (including UT). In VK7 mosquitoes, observed longevity also was influenced by the activity in the cone during tests (treatment*activity interaction: 2=9.11; df=3; P=0.0278) and as the proportion of activity in the LHC increased, the likelihood of dying increased for UT and P2 (Additional file 2: Fig. S3a). In IG2 tests, mosquitoes were more likely to die when greater activity was recorded in the UHC, and, compared with UT, mosquitoes were significantly more likely to die when the proportion of activity close to the net was less than 50% (Additional file 2: Fig. S3b).

Mosquitoes that did not blood-feed at 1h PE were at least 17 times more likely to die at 24h PE (Additional file 2: Table S4 KS 2=0.00; df=1; P=0.9987, NG 2=25.80; df=1; P<0.0001, BF 2=19.40; df=1; P<0.0001 and VK7 2=24.54; df=1; P<0.0001). The amount of blood ingested had a slight impact on longevity: BF and VK7 mosquitoes were 7% and 2% less likely to die as the BMW increased, respectively (Additional file 2: Table S5 BF HR 0.93; 95% CI 0.90, 0.93; P<0.0001 and VK7 HR 0.98; 95% CI 0.97, 0.99; P=0.0012). The effect of treatment on longevity varied with wing length for all strains, for example, smaller mosquitoes lived longer PE to UT netting (all strains) and were more likely to die if they were exposed to IG2 for all the strains except BF (Additional file 2: Fig. S2 a, b).

Here is the original post:
Video augmentation of the WHO cone assay to quantify mosquito ... - Parasites & Vectors