In this study, we designed a new tent trap; the BioDiVector (BDV) tent trap, consisting of two rectangular tents that use human bait without endangering the technical personnel. The daily activity pattern of Aedes aegypti and Aedes albopictus in intra, peri, and extradomiciliary sites was studied in an endemic area of dengue in southern Mexico by using the BDV tent trap. Totals of 3,128 individuals of Ae. aegypti and 833 Ae. albopictus were captured. More Ae. aegypti males than females were caught, while the opposite was true with Ae. albopictus. The activity of both mosquito species was affected by the interaction between the collection site and time of day. In general, more individuals of both mosquito species were captured at the extradomicillary sites than at the peri and intradomicillary sites. Mosquitoes showed two peaks of activity, one in the morning and the other in the afternoon, but in general this only occurred at the extradomicillary sites, whereas no peak of activity was observed at the intra and peridomicillary sites. Overall, Ae. aegypti had a higher indirect biting rate than Ae. albopictus. Finally, due to its efficiency, simplicity, and low cost, we suggest the use of this innovative tool for entomological surveillance, bionomics and vector incrimination studies in geographical areas where dengue and other arboviruses are present.
Studies on the bionomics of Aedes mosquitoes include systematic laboratory and field evaluations to identify biological, ecological, and ethological aspects of suspected or proven vectors of dengue with the aim of defining the vectorial capacity of a mosquito population (FernándezSalas and FloresLeal 1995. From the epidemiological perspective, the biorhythm and intensity of biting as well as the feeding periodicity of gonoactive females are entomological components that need to be determined for the incrimination of a mosquito species as a vector, since these parameters provide evidence related to the type of relationship established for hostvector contact, and the feeding preferences for animals or humans (Dye 1986).
The recent introduction of Aedes albopictus (Skuse) to southern Mexico (CasasMartínez and TorresEstrada 2003) and its subsequent establishment in the urban habitats occupied by Aedes aegypti (L.) has led to speculations among experts on the role of the Asian tiger mosquito in dengue epidemiology, on account of its evident invasive capacity and the recognized vectorial capacity of this mosquito in the transmission of many viral pathogens, including dengue fever, West Nile virus, Yellow fever virus, St. Louis encephalitis, and Chikungunya fever (IbáñezBernal et al. 1985, Holick et al. 2002, Paupy et al. 2011). Thus, Ae. albopictus provides an interesting model for determining its role as a vector of arboviruses in an area where this species is already established; and so efficient entomological tools are required to investigate the basic components of its bionomics.
The methods used for the surveillance and control of dengue vectors are based on techniques that were developed for the eradication of Ae. aegypti and prevention of yellow fever (Scott and Morrison 2003). Although most of the traps used for catching adult mosquitoes are efficient for entomological sampling (MacieldeFreitas et al. 2008, Hoel et al. 2009), these are not specific for studying the daily pattern of biting or humanvector contact. Due to ethical considerations and biosecurity measures recommended by national and international organizations for the protection of personnel involved in the sampling of vectors of arboviruses in endemic zones, alternative methods are being developed to replace the technique that uses humans as bait during field entomological evaluations (WHO 2011). In this context, there is an urgent need for the development of an efficient trap for monitoring adult mosquitoes (Sivagnaname and Gunasekaran 2012).
In this study, we designed a BDV tent trap which uses human bait without endangering the technical personnel. By using this trap, we investigated the diurnal activity pattern and determined the indirect biting rate of Ae. aegypti and Ae. albopictus in an endemic urban area of dengue fever in southern Mexico.
MATERIALS AND METHODS
The observations were performed in the city of Tapachula, which has a population of approximately 320,000 inhabitants. It is located in the southern part of the state of Chiapas on the Soconusco coastal plain, near the border with Guatemala, at 14°54’29” N, 92°15’38” W, 177 masl (INEGI 2010). This region experiences a subhumid warm climate with rains during the summer (Aw). The rainy season normally occurs from May to October (García 1986). Dengue is endemic in this urban area, and during the last few years dengue outbreaks have taken place in the rainy season, favoring an increase in the population of Ae. aegypti and Ae. albopictus despite the efforts of health institutions to control the mosquitoes thorough the implementation of antilarval measures and the spatial application of insecticides to combat adults. Ae. Albopictus, introduced to Tapachula in 2002, has shown great ecological capacity to naturalize and coexist successfully with Ae. aegypti in southern Mexico (CasasMartínez and TorresEstrada 2003). During the last decade, this species has expanded its geographic range distribution to various locations in urban, suburban, and rural areas on the Pacific coastal plain and the Soconusco region of Chiapas (CasasMartínez et al. 2003), colonizing extradomicillary and peridomicillary habitats (CasasMartínez et al. 2003).
The model of the trap consisted of two tents made of fine mesh (tricot) with a 2.2 × 1.9 × 1.8 m white exterior and a 1.0 × 0.7 × 1.8 m black interior (Figure 1). The interior net is attached to the roof of the exterior net. For trap installation, all that was required was the suspension of the exterior tent from its four top corners using ropes attached to nails inserted in the wall. The trap was operated by a technician or volunteer seated within the interior tent. Before and after mosquito collection, a manual lift mechanism raised and lowered the exterior tent. In the center of the ceiling and the upper and lower corners of the exterior tent (white), a system of rings and ropes was installed in order to lift (0.3 m) and extend the mesh to the floor during the collection period. This controlled the opening and closing of the trap by a person inside the interior tent (black) during mosquito collection. After each collection period, mosquitoes were captured with the help of an entomological net. During the study, we used six traps of the same size and manufacturer. This sampling device did not require a power source for its operation, and its transportation and installation was quick and easy at any sampling site. The average trap installation time was 15 min.
Daily pattern of activity of mosquitoes
Aedes aegypti and Ae. albopictus activity was studied during two days per month over a period of nine months, from March, 2010 to January, 2011. The study sites were located in houses and cemeteries. The samplings were performed between 06:00 and 18:00 at intervals of 2 h, with 1 h of collection and 1 h of resting. On the first day of sampling, three traps were located in rooms (intradomicillary) and three traps were placed in the yards (peridomicillary) of the houses. The next day, the traps were installed at two different points in three cemeteries (extradomicillary). Two of the cemeteries, Jardín (14º53’36.80” N, 092º14’48.20” W, 158 masl) and Municipal (14º54’10.80” N, 092º16’17.20” W, 168 masl), are located within the urban area of Tapachula city, whereas the cemetery Raymundo Enríquez (14º51’40.00” N, 092º19’27.60” W, 66 masl) is located on the outskirts of Tapachula, approximately 1.3 km from the nearest group of houses. The dominant vegetation of the Jardín cemetery consists of herbaceous plants, native and introduced shrubs, and trees that cast shade throughout the day. Vegetation density and cover at the Municipal cemetery is lower compared to the previous site, thus there are sections with increased sun exposure due to the lack of trees. Vegetation density and cover at the Raymundo Enríquez cemetery is low, however, crops of cocoa (Theobroma cacao L.), rambutan (Nephilium lappaceum L.), coconut (Cocus nucifera L.), and red ginger (Alpinia purpurata K. Schum.) surround the cemetery.
Identification of mosquitoes
All collected mosquitoes were separated according to the time of collection, stored in disposable plastic containers covered with fine mesh fastened with rubber bands (Figure 1E), and transported to the laboratory, where mosquitoes were identified and quantified per sex and sampling site. Specimen identification was performed using the morphological characteristics described by Savage and Smith 1995.
Data per species were subjected to ln (x + 1) and analyzed by a twoway analysis of variance (ANOVA) with collection site (intra, peri, and extradomicillary) and time of day as the two factors. Extradomicillary data was separated into three collection sites, which corresponded to the three cemeteries since each present different environmental characteristics. The Jardín cemetery was considered extradomicillary site 1, the Municipal cemetery extradomicillary site 2, and the Raymundo Enríquez cemetery extradomicillary site 3. Due to the rare presence of Ae. aegypti males at extradomicillary site 3, these were not included in the analysis. Data from indirect biting rates were analyzed by a oneway ANOVA. When significant effects were observed, means were separated by a Tukey test. The statistical analyses were performed using the R statistical software (version 1.1–3, 2012) and the Agricolae package (Mendiburu 2012).
Daily activity pattern of Ae. aegypti
A total of 3,128 individuals (1,015 females and 2,108 males) of Ae. aegypti was collected during this study. The activity of Ae. aegypti females was affected by collection site (F = 95.18; df = 3.402; P < 0.001), time of day (F = 3.58; df = 5,402; P < 0.01), and the interaction term collection site*time of day (F = 2.01; df = 15,402; P < 0.05). The interaction between collection site*time of day is shown in Figure 2. For collection site, the multiple comparisons showed that there were no significant differences in the activity of Ae. aegypti at the intradomicillary site, peridomicillary site, and the extradomicillary site 2. At the extradomicillary site 1, female activity was higher in the morning (06:00–10:00) and late afternoon (16:00–18:00) than at 10:00–16:00 (Figure 2A). For time of the day, the multiple comparisons revealed that at 06:00–8:00, 8:00–10:00, 14:00–16:00, and 16:00–18:00, female activity was higher at the extradomicillary sites than at the peri and intradomicillary sites. At 10:00–12:00 and 12:00–14:00, mosquito activity was higher at extradomicillary site 2 than in peri and intradomicillary sites. Female activity at extradomicillary site 1 was intermediate and not significant to that observed at extradomicillary site 2 and intradomicillary site but it was different to that recorded at the peridomicillary site (Figure 2B).
The activity of Ae. aegypti males was affected by collection site (F = 159.66; df = 3,402; P < 0.01), time of day (F = 7.66; df = 5,402; P < 0.001), and the interaction term collection site*time of the day (F = 2.23; df = 15,402; P < 0.05). The interaction between collection site*time of day is shown in Figure 3. For collection site, multiple comparisons revealed that there were no significant differences in mosquito activity between intra and peridomicillary sites. Male activity at extradomicillary site 1 was higher at 08:00–10:00 than at 16:00–18:00, 12:00–14:00, and 14:00–16:00. Male activity at 10:00–12:00 and 06:00–08:00 was intermediate and not significant to that at 08:00–10:00, 14:00–16:00, and 12:00–14:00. At extradomicillary site 2, mosquito activity was higher in the morning (06:00–12:00) and late afternoon (16:00–18:00) than at midday/midafternoon (12:00–16:00) (Figure 3A) For time of day, multiple comparisons showed that at whatever time of day, male activity was higher at the extradomicillary site than at the peri and intradomicillary sites (Figure 3B).
Daily activity pattern of Ae. albopictus
A total of 833 Ae. albopictus (538 females and 295 males) were caught during this study. Activity of female Ae. albopictus was affected by collection site (F = 47.61; df = 4,456; P < 0.001), and time of day (F = 3.73; df = 5,456; P < 0.01), but the interaction term collection site*time of day was not significant (F = 0.89; df = 20,456; P > 0.05). Female activity was higher at extradomicillary sites 3 and 1 when compared to activity observed at extradomicillary site 2, peri, and intradomicillary sites (Figure 4A). Females were more active in the morning and afternoon than at midday. The activity showed by females at 10:00–12:00 and 14:00–16:00 was intermediate and not significant to that at 06:00–08:00, 08:00–10:00, 16:00–18:00, and 12:00–14:00 (Figure 4B).
Aedes albopictus male activity was affected by collection site (F = 26.04; df = 4,456; P < 0.01), time of day (F = 6.47; df = 5,456; P < 0.001), and the interaction term collection site*time of day (F = 2.38; df = 20,456; P < 0.01). The interaction between collection site*time of day is shown in Figure 5. For collection site, multiple comparisons showed that there were no significant differences in the activity of Ae. albopictus among the intradomicillary site, peridomicillary site, and extradomicillary site 2. At the extradomicillary site 1, male activity was higher during early morning (06:00–08:00) and late afternoon (16:00–18:00) than between 10:00 and 16:00. At the extradomicillary site 3, males were more active at 16:00–18:00 than during the rest of the day (Figure 5A). For time of day, multiple comparisons revealed that there were no significant differences in male activity between the time intervals of 08:00–10:00, 10:00–12:00, and 12:00–14:00. At 06:00–08:00, males were more active in the extradomicillary sites 1 and 3 than in the peri and intradomicillary sites. Male activity at extradomicillary site 2 was intermediate and not significantly different to that displayed by males at extradomicillary sites 2 and 3, as well as the peri and intradomicillary sites. During 14:00–16:00, male activity was higher at extradomicillary site 1 than at the peridomicillary site. Males displayed intermediate activity at extradomicillary sites 1 and 2 and the intradomicillary site. This was not significant when compared to extradomicillary site 3 and the peridomicillary site. At 16:00–18:00, males were more active at extradomicillary site 3 than at extradomicillary site 2 and the intra and peridomicillary sites. Mosquito activity was intermediate at extradomicillary site 1 and not significantly different to activity displayed at extradomicillary sites 2 and 3, peri, and intradomicillary sites (Figure 5B).
Indirect biting rate
As an indirect measure of vectorhuman contact, the biting rate, expressed as the number of mosquito females/human/day, was calculated for both species for the five collection sites (intra, peri, extradomicillary 1, 2, and 3). In general, Ae. aegypti females presented a higher indirect biting rate than Ae. albopictus females (Table 1).
Table 1. Indirect biting rate (no. mosquito females/human/day) for the five different collection sites using a humanbaited tent trap. Means within a column followed by the same letter are not significantly different (P > 0.05, Tukey test).
Extradomicillary (1, 2, and 3)
Preventive and control measures against dengue have been directed towards critical places at specific times. They require efficient methods for entomological surveillance as well as sensitive techniques to forecast and/or detect sudden increases of mosquito populations in real time (Regis et al. 2008). Therefore, sampling methods and devices to detect mosquitoes are important tools for obtaining ecological and behavioral data, such as population density, daily and seasonal abundance, and spatial distribution of vectors or the efficiency of mosquito control measures. All this information is fundamental to understanding the epidemic potential and to establish strategies for effective and timely control programs (Krockel et al. 2006). Due to the ecological characteristics of Ae. aegypti and Ae. albopictus populations, the use of ovitraps and/or the collection of adults have been the most appropriate sampling methods to determine the times and places where action should be focused in order to prevent or decrease the severity of disease outbreaks (Focks 2003, Regis et al. 2008). However, these types of traps do not allow the study of different aspects of hostvector interactions, such as daily pattern of activity periodicity and host preferences.
In the present study, we found that the BDV tent trap could be used as a sensitive entomological technique to monitor Ae. aegypti and Ae. albopictus at low population densities, regardless of collection site. This characteristic of the trap may be attributed to its design that considered two important factors. First, the visual capacity of Aedes mosquitoes to distinguish the contrast of black and white targets (Muir et al. 1992, Hoel et al. 2009) and the fact that these mosquitoes prefer to rest on black, nonreflective surfaces such as clothing (Fay and Prince 1970, Edman et al. 1997). The second aspect considered was the availability of humans, who acted as a source of chemical stimuli attractive to mosquitoes during the different phases of hostseeking behavior (Sutcliffe 1987). Generally, human landing/biting collection is the most efficient technique for capturing Aedes mosquitoes (Jones et al. 2003). However, this technique has come under scrutiny because of the ethical considerations of exposing collectors to infected mosquitoes (Gimnig et al. 2013). These restrictions on biosafety and ethical considerations for human use for sampling mosquito vectors of arboviruses were overcome in this study, because the risk of being bitten by an infective female was significantly reduced when using the double tent trap. Similar tent traps that avoid the exposition of the human baits to mosquitoes have been designed and evaluated in endemic areas of malaria in Africa (Mathenge et al. 2002, Govella et al. 2009). Another advantage of the trap designed in this study was the mechanism for raising and lowering the exterior tent by one person at the end of the collection interval; a characteristic that allows optimization and the comparative standardization of the sampling effort by the technical personnel. Additionally, the trap designed in this study is inexpensive, costing no more than US$30. This price is very competitive compared to other devices used to collect diurnal mosquitoes. For example, some traps such as CDC gravid trap, FayPrice trap, and BG sentinel traps cost US$100 or more (John W. Hock Company, http://www.JohnWHock.com; Biogents AG, http://www.biogents.com). Although the trap designed for this study had very specific purposes, i.e, daily activity and indirect humanvector contact of Aedes, it can also be used for studies of biodiversity of Culicidae, host preference, and interspecific competition. Future studies will compare the efficacy of the BDV tent trap against traps commonly used for monitoring Aedes mosquitoes.
In this study, we found that the activity of both mosquito species was influenced by the interaction between collection site and time of day. For example, mosquitoes showed two peaks of activity, one in the morning and another in the afternoon, but in general this only occurred at the extradomicillary sites, whereas no peak of activity was observed in the intra and peridomicillary sites. These results suggest that some environmental conditions affected mosquito behavior. Temperature and light seem to be the most important factors affecting mosquito activity (Rowley and Graham 1968, Taylor and Jones 1969, Chadee and Martinez 2000, Kawada et al. 2005). Several studies have reported that the activity pattern of sexes of Ae. aegypti and Ae. albopictus is diurnal and bimodal, with a peak after sunrise (06:00–07:00), and another before sunset (17:00–18:00) (Trpis et al. 1973, Chadee 1988, Thavara et al. 2001, LimaCamara 2010). In contrast, other studies have reported that Ae. aegypti has a trimodal (morning, midday, and afternoon) pattern of activity (Corbet and Smith 1974, Chadee and Martinez 2000). In addition, Chadee and Martinez 2000 found that the activity of the Trinidad strain of Ae. aegypti is both diurnal and nocturnal at indoor and outdoor urban sites, but this mosquito species was only active during the day at rural sites. The last authors suggest that the extension of the activity of Ae. aegypti during nocturnal hours may have occurred as an adaptation of mosquitoes to light, but another study suggested that the nocturnal activity of this species is due to its intrinsic reaction to light more than an evolutionary adaptation to light (Kawada et al. 2005). A laboratory study showed that the nocturnal hostseeking behavior of Ae. aegypti and Ae. albopictus was positively correlated with increasing light intensity, with the former species more sensitive to light than the latter (Kawada et al. 2005).
We found that more females and males of both species of mosquitoes were captured at the extradomicillary sites than at the peri and intradomicillary sites. A similar situation was reported by Chadee and Martinez 2000, who found that at both urban and rural sites, larger numbers of Ae. aegypti were captured outside than inside houses in Trinidad, West Indies. Furthermore, we captured more individuals of Ae. aegypti than Ae. albopictus at the intradomicillary site, peridomicillary site, and extradomicillary sites 1 and 2. In contrast, only Ae. albopictus was caught at extradomicillary site 3. A possible explanation for this result is that extradomicillary site 3 is surrounded by vegetation, providing many potential breeding sites (e.g., tree holes, leaf axils, coconut shells) for Ae. albopictus females. This site is also far from homes. Generally, Ae. albopictus is primarily an exophagic mosquito and typically breeds outdoors, whereas Ae. aegypti is primarily a domestic and endophagic mosquito that shows a greater preference for indoor breeding (Hawley 1988, Edman et al. 1997, Thavara et al. 2001, Paupy et al. 2009). Thus, differences in the abundance of both mosquito species in Tapachula may be influenced by the ecological heterogeneity of the urban landscape and the degree of urbanization of the sites where sampling took place. Previous studies have shown that although both species can be found in urbanized areas, Ae. albopictus is commonly found in suburban and rural areas where open spaces with vegetation is prevalent (Tsuda et al. 2006, Honorio et al. 2009).
We found that more Ae. aegypti males than females were caught, while the opposite was true for Ae. albopictus. Nelson et al. 1976, using human bait, found that consistently more Ae. aegypti males than females were collected in Jakarta, Indonesia. In Madagascar, it was found that more Ae. aegypti and Ae. albopictus females than males were collected by the human bait trap method (Raharimala et al. 2012). In Florida, traps baited with CO2, Llactic acid, or 1octen3ol caught more Ae. albopictus females than males (Hoel et al. 2009). According to the literature, no trap attracts males as efficiently as human volunteers (Hartberg 1971). Like females, males actively flew around the black tent but often had brief contacts over the mesh that protected humans. In several species of Aedes mosquitoes, males assembled in the vicinity of the hosts presumably to intercept females coming to feed (Hartberg 1971, Jaenson 1985). For example, Ae. aegypti males formed swarms in the laboratory, triggered by the onset of the photophase or by the presence of host odors (Cabrera and Jaffe 2007). The swarm attracted both males and females and increased mating activity. Observations using an olfactometer showed that swarming males produced a volatile pheromone that stimulates the flying activity of females at a distance (Cabrera and Jaffe 2007).
In summary, based on the evidence described above, we suggest the use of the BDV tent trap designed in this study for laboratory and field studies, including entomological surveillance, bionomics, and vector incrimination of diurnal mosquitoes in geographic areas with dengue and other arboviruses. The activity of Ae. aegypti and Ae. albopictus was influenced by the interaction between collection site and time of day. Finally, the BDV trap can be considered as an entomological tool that is ideal for its features and advantages to complement epidemiological risk assessments as part of prevention programs, surveillance, and control of dengue vectors.
We thank José Luis Aguilar Rodríguez, Martín Vázquez Castillo, José Luis Espinosa Aguilar, and Eufronio Díaz Espinosa for their technical support during the mosquito field collections and Rafael Ángel Avendaño Rabiella, Jorge Aurelio Torres Monzón, and José Asunción Nettel Cruz for their help and assistance in the entomological laboratory activities (all are members of the research group BioDiVector at INSP/Centro Regional de Investigación en Salud Pública, Chiapas, Mexico). The English text was corrected by Julian Flavell. This investigation was supported by a grant (FOSISS87617) from CONACyT, Mexico.