The study was carried out in the municipality of Caxias, state of Maranhão, located at 4º51'32" South latitude and 43º21'22" West longitude. The municipal headquarters is 66 meters above sea level, with 5,196.769 km² area. Caxias is the fifth largest city in the state of Maranhão, Brazil, with 166.1591 inhabitants (IBGE, 2022). The climate is tropical, hot and humid and has two defined seasons, rainy and dry, with average temperatures ranging from 17 to 27ºC, varied rainfall of 1,100 and 1,500 mm year^-1 (GERPLAN, 2002; Sousa, Meneses, & Vianna, 2015).

The experiments were conducted at the Laboratory of Medical Entomology – LABEM, Universidade Estadual do Maranhão – UEMA. A. aegypti and A. albopictus females were acquired from the insectary kept at LABEM, in which the breeding of mosquitoes followed the Consoli and Lourenço-de-Oliveira (1994) protocol, adapted by Pinheiro and Tadei (2002).

Sixty fully engorged females with a mean age of three days were used and transferred to an aluminum/acrylic cage with dimensions of 40 x 40 cm in a room under controlled conditions, with an average temperature of 28 ± 2ºC, relative humidity around 85% and photophase of 12 hours (Navarro et al., 2003;Gomes, Sciavico, & Eiras, 2006). For each species, separately, three traps were placed into the breeding cages, which consisted of plastic 180 mL glass lined with filter paper, combined with biolarvicides prepared as follows: one based on S. spinosa, tablet formulation, NatularTM 20EC - 20.65% Spinosad a.i., Clarke Mosquito Control Products, Inc. - spinosad, and the other consisting of B. thuringiensis sub. israelensis, granule formulation, VectoBac® WG - Bti.

The combination with Bti consisted of the addition of 0.003 grams of larvicide in each glass, with the addition of 100 mL tap water. To prepare the combination with spinosad (Natular^TM DT, batch 1309190010), ¼ (0.21g L^-1) of tablet corresponding to 1.87% spinosyn was added, with the addition of 100 mL tap water, following the manufacturer recommendations.

As a positive control, 100 mL 10% grass infusion was used, this substance is the standard used as an attraction for engorged females of A. aegypti. This solution was prepared from dehydrated grass leaves, with the addition of approximately 20 L tap water, packed in a plastic container with a capacity of 25 L. The solution was sealed and stored for seven days for fermentation, right after the solution was filtered and diluted to a final concentration of 10% (Reiter, Amador, & Colon, 1991).

Glasses were daily clockwise rotated and, after four days, eggs were removed for counting under a stereomicroscope (Zeiss stemiDV4). The experiment was repeated four times; each repetition had a new group of females and lasted five days, in accordance with Barbosa and Silva (Barbosa & Silva, 2002) with modifications.

Field experiments were conducted in two distinct areas, area I, located in the central area, and area II, located in the south of the city. Field experiments were repeated in two seasons, dry (from August to October, 2017) and rainy (from January to March, 2018). These areas were selected for presenting cases of dengue in 2017, and for presenting distinct characteristics regarding vegetation, which is important considering the biological characteristics of the two species under study.

Ten houses were chosen in each area, with an average distance of about 400 m apart (Monteiro, Carvalho, & Souto, 2014; Soares, Silva, Oliveira, & Abreu, 2015). All residences were geo-referenced using the Global Positioning System (GPS) (Figure 1A and B). Three ovitraps were installed in each house, two combined with biological insecticides one based on S. spinosa, tablet formulation, NatularTM DT - spinosad, and the other consisting of B. thuringiensis sub. israelensis, granule formulation, VectoBac® WG – Btiand, containing 10% grass infusion.

Figure 1. Maps showing the sites where the ovitraps were installed in the municipality of Caxias, state of Maranhão, Brazil: A) Area I; B) Area II. Source: author himself

Ovitraps combined with biolarvicides were prepared as follows: one S. spinose-based, tablet formulation, NatularTM DT – spinosad, and the other consisting of B. thuringiensis sub. israelensis, granule formulation, VectoBac® WG - Bti. The Bti combination consisted of adding 0.003 grams of the larvicide in each trap, with an addition of 500 mL tap water. While for preparing the combination with spinosad, ¼ tablet that corresponds to 1.87% spinosyns was added to the same amount of water as the previous biolarvicide. The quantities used of the biolarvicides were defined according to the manufacturer recommendation.

As a positive control, 500 mL 10% grass infusion was used, this substance is used as a standard for attraction of A. aegypti engorged females in ovitraps. This solution was prepared from dehydrated grass leaves, with an addition of approximately 20 L tap water, packaged in a plastic container with a capacity of 25 L. The solution was sealed and stored for seven days for fermentation. Right after that, the solution was filtered and diluted to a final concentration of 10% (Reiter et al., 1991).

Ovitraps used in the field experiment, consisted of a black plastic container, with a hole about 12 cm wide, and a capacity of 800 mL (Reiter, Amado, Anderson, & Clark, 1995). Inside of them, a plywood reed was placed, Duratree, 12 cm x 2.5 cm wide, fixed vertically by a steel clip #08 for oviposition and eggs adherence.

Ovitraps were installed in the peridomicile, limited to the residence immediate surroundings, on the ground, 20 cm apart, preferably positioned, where possible, in places protected from the sun and rain. After a four-day exposure period, ovitraps were collected and forwarded to the Laboratory of Medical Entomology – LABEM, properly identified. After drying the reeds, eggs were counted under a stereomicroscope (Zeiss model stemiDV4). Between each collection, there was an interval of 25 days.

In each study area, by season, three repetitions were performed, with 30 ovitraps installed in each, totaling 90 traps. Considering both seasons, the sampling effort was 180 ovitraps for each area, totaling 360 traps used in the two areas.

The positive ovitrap index (POI) is the ration of the positive traps number and the number of examined traps multiplied by one hundred and the Egg Density Index (EDI) is the ratio of the number of eggs to the number of positive traps (Gomes, 1998). To assess the attractiveness or repellency, the Oviposition Activity Index (OAI) was used, represented by the formula OAI = (Nt - Nc)/(Nt + Nc), where Nt is the number of eggs laid in the treatment and Nc is the number of eggs placed in the control trap (Karmer & Mulla, 1979). OAI ≥ +0.3 indicates attractiveness, while OAI ≤ -0.3 indicates that the treatment is repellent to oviposition.

Data were assessed for normal distribution by Lilliefors’ Test, and for homoscedasticity, by Levene’s Test (Zar, 2014). To check for differences in averages and medians for the egg density index between the treatment groups and the control, ANOVA (F) or Kruskal-Wallis (H) was applied, in cases of normal and non-normal distribution, respectively. In case of difference between averages and medians, Tukey and Dunn’s tests were continued a posteriori, respectively. For all analyses, a value of p < 0.05 was considered statistically significant (Siqueira & Tibúrcio, 2011).

The analyses w ere run using the BioEstat version 5.0 statistical packages (Ayres, Ayres, Ayres, & Santos, 2007), GraphPad Prism 5.0 and Microsoft Office Excel 2013.

In laboratory conditions, A. aegypti females laid an average of 536.25 eggs in all four replications, with an average of 582.5 (0.3321) in ovitraps containing spinosad and Bti, with an average of 430 (0.2458) eggs and grass infusion with an average of 596.25 eggs. However, no significant differences were detected for the averages of eggs obtained between treatments (F = 0.8521, p = 0.5388) (Figure 2A).

Although there was no difference between the averages of eggs, the oviposition activity index showed that spinosad was attractive for A. aegypti (OAI = 0.32 and OAI = 0.38) and for Bti only in one presentation presented OAI ≥ + 0.3 (OAI = 0.46).

Regarding A. albopictus oviposition, 147.75 average eggs were obtained in the four replications. The highest average of eggs was obtained for the grass infusion (294), with a value about seven times higher than spinosad (43.5) and about three times higher compared to Bti (105.75).

When comparing egg averages between treatments, there was a statistically significant difference (F = 13.1888, p = 0.0025). In grass-infused ovitraps, higher number of eggs were obtained with both biolarvicides, spinosad (**) and for Bti (*) (Figure 2B). However, there was no statistically significant difference for number of eggs between biolarvicides.

Figure 2. Comparison of averages of eggs by ovitraps combined with biolarvicides under laboratory conditions: (A) Aedes aegypti; (B) Aedes albopictus.

The calculation of OAI indicated that both biolarvicides, spinosad and Bti, are attractive to females of A. albopictus, for spinosad with OAI = 0.96 and for Bti of OAI = 0.57.

In Area I, a total of 22,479 eggs were obtained and the number of eggs obtained during the rainy season (19,555) was about seven times greater than in the dry season (2,924). Regarding the number of eggs in the treatments, the combination with spinosad reached about twice as many eggs as in ovitraps with Bti, in the two seasons. On the other hand, the infusion of grasses (positive control) obtained a greater number of eggs than the two treatments, in both seasons.

Regarding the positive ovitrap index - POI, between the combinations, similar values were observed in both seasons. However, regarding the EDI, there were variations between treatments and control, in both seasons, in which the grass infusion showed higher egg density than the treatments.

In the dry season, there was no significant difference in averages of eggs in the combination (F = 5.0986, p = 0.0131); ovitraps with grass infusion had higher means than treatments with Bti and spinosad (p < 0.05) (Figure 3).

Figure 3. Average of eggs of Aedes aegypti and Aedes albopictus collected using ovitraps combined with biolarvicides in area I: A) Dry season (August to October, 2017); B) Rainy season (January to March, 2018).

In Area II, 23,005 eggs were obtained, and 21,132 were obtained for the rainy season, 11 times higher than the value for the dry season, 1,873. Ovitraps with spinosad presented a higher number of eggs than the traps with Bti, in both seasons. Ovitraps with grass infusion showed a higher number of eggs than Bti, in both seasons, with values about three times higher in the rainy season and two times higher in the dry season.

The Positive Ovitrap Index - POI were similar in both periods, for all treatments. However, the Egg Density Index – EDI in both periods, was higher for grass infusion, corresponding about twice more than the other treatments.

The average of eggs obtained in the dry season did not show a significant difference between treatments (F = 1.1206, p = 0.3415) (Figure 4). On the other hand, in the rainy season, there was a significant difference between treatments (H = 10.5494, p = 0.0051), in which the grass infusion obtained a median significantly higher than spinosad and Bti (Dunn’s Test p < 0.05) (Figure 4).

Figure 4. Average of eggs of Aedes aegypti and Aedes albopictus collected with ovitraps combined with biolarvicides in area II: A) Dry season (August to October, 2017); B) Rainy season (January to March, 2018).

In relation to the studied areas, the results obtained in the field showed through the Oviposition Attractiveness Index that both biolarvicides were attractive to Aedes females, during the two analyzed seasons, dry and rainy. The lowest value for spinosad was OAI = 0.33 and the highest, OAI = 0.76, and for Bti, the highest was OAI = 0.32 and the lowest, OAI = 0.81.

Regarding the egg density correlated with climatic factors, in the dry season, it was observed that there was no record of rainfall during the collection days; the average temperature was 28.2ºC and relative humidity was 58%. On the other hand, in the rainy season, temperatures of 26.3ºC, relative humidity of 78.6% and average rainfall of 11.8 mm were recorded. During this season, EDI was obtained with higher values. When comparing EDI between the analyzed seasons, for treatments and grass infusion, EDI was higher in the rainy season.