Seasonal Activity of Adult Mosquitoes (Diptera: Culicidae) in a Focus of Dirofilariasis and West Nile Infection in Northern Iran.

Background
Mosquito-borne arboviruses such as West Nile, dengue, Rift Valley fever, and Sindbis viruses and the nematode Dirofilaria are reported in Iran, but there is little information on the seasonal activity of their vectors in the country. We aimed to determine the seasonal activity of adult mosquitoes (Diptera: Culicidae) in a focus of dirofilariasis and West Nile infection in Guilan Province, northern Iran.


Methods
Collections were carried out using light traps in seven counties at least two times from random sites and every two weeks from a fixed site (Pareh Village, Rudbar County) during Aug-Dec 2015 and Apr-Oct 2016.


Results
Overall, 16327 adult mosquitoes comprising 18 species representing seven genera were identified. The most prevalent species were Cx. theileri (23.59%), Cx. tritaeniorhynchus (20.75%), Cx. pipiens (19.37%), Ae. vexans (18.18%), An. pseudopictus (10.92%) and An. maculipennis s.l. (5.48%). Aedes pulcritarsis and Cx. perexiguus were found for the first time in Guilan Province. The active season of adult mosquitoes extended from early May to early Oct in the fixed site. There was no significant regression between the abundance of adult mosquitoes and the meteorological data during active season in the fixed site (P> 0.05, R2= 0.31).


Conclusion
Though no significant regression between the abundance of mosquitoes and the meteorological data was observed during active season, temperature and rice fields had a great influence in starting and ending active season in the region.


Introduction
West Nile virus (WNV) (Flaviviridae: Flavivirus) and its subtype Kunjin is distributed in Eurasia, Africa, North and Central America and Australia. Mosquitoes (Diptera: Culicidae), especially ornithophilic species, are the principal vectors of the virus and some virus isolations have been reported from soft and hard ticks. Wild birds, especially wetland species, are the principal vertebrate hosts, the virus has also been isolated from mammals and frogs (1,2).
A large amount of available data on mosquitoes in Iran is based on collections and ecology of larvae (27,(36)(37)(38)(39)(40) and many other references cited by aforementioned articles). Different methods of collecting adult mosquitoes, such as using light traps, aspirators, pit shelters and total catch (Pyrethrum space spray), have been used mostly in relation to anopheline vectors of malarial protozoa (41)(42)(43)(44)(45)(46)(47). There are a few published documents in the country that deal with adult sampling, especially using light traps, which include culicines (27,(48)(49)(50), but there are no studies of seasonal activity. That is why there is very little information about the seasonal activity of culicine adults in Iran.
This study was carried out by means of light traps to study the seasonal activity of mosquitoes, especially probable and proven vectors of WNV and Dirofilaria, in Guilan Province, northern Iran.

Study area
Guilan Province locates in the Caspian Sea littoral of northern Iran, between the Caspian Sea and the Alborz Mountain range. It has coastal, plain, foothill, and mountainous areas with an area of approximately 14,700 square kilometers. The province is bordered by Mazandaran Province in the east, Ardabil Province in the west and Zanjan and Qazvin provinces in the south. It is also bordered by the Republic of Azerbaijan in the north as well as Russia across the Caspian Sea (Fig. 1). The province has a temperate climate and relatively warm-humid summer. It is located between

Specimen and data collection
In seven counties (including different topographical areas of the province) adult collections were carried out at least two times from  Table  2. Two CDC light traps were used in each variable and fixed site. The light traps were suspended from the ceiling in animal shelters from sunset to sunrise, i.e. from 1800 to 0600 hrs. The electricity of traps was provided by 6-volt rechargeable batteries. Moreover, ad hoc collections were carried out using manual aspirators (hand catch) in the fixed and random sites. The specimens were identified using the morphological-based keys (29). The abbreviations of mosquito genera and subgenera follow Reinert (51). The specimens are deposited at the Museum of Medical and Veterinary Entomology, School of Health, Guilan University of Medical Sciences, Rasht, Iran.

Mapping collected mosquitoes and statistical analysis
ArcGIS 10.3 was used to create a geo-database of mosquitoes and to map the collection sites and the distributions of the most medically important species. The statistical analysis of mosquito abundance and meteorological data was carried out using the linear regression test of SPSS software (ver. 16 for Windows, SPSS Inc., Chicago, IL).

Seasonal activity and the fluctuations of rainfall and temperature
In general, the active season of adult mosquitoes extended from early May to early Oct in the fixed site (Pareh Village of Rudbar County). The peak of activity was late June for Cx. theileri, mid-July for An. maculipennis s.l., An. pseudopictus and Cx. pipiens, and late July for Cx. tritaeniorhynchus. While the peak of activity of most adult mosquitoes was late June to mid-July, and the abundance dramatically decreased after that, the monthly mean temperature increased by Aug. Also after Apr rainfall decreased in the fixed site during Jun and Jul and the rainy season started in Sep (Figs. [3][4][5][6]. There is no significant regression between the abundance of adult mosquitoes and the meteorological data ( Table 2) during active season in the fixed site (P> 0.05, R 2 = 0.31).     *

Mosquito fauna
During the present investigation, 18 species representing seven genera of mosquitoes were collected in Guilan Province in which Ae. pulcritarsis and Cx. perexiguus were new records for the province. Thus, the number of species recorded in the province increased from 30 (36)  (Culicella) morsitans (Theobald) whose larvae were collected during recent years in Guilan Province (36), were not captured in the present study. The reason is the rarity of some aforementioned species and the tendency of some, such as Cx. hortensis, Cx. territans and Cs. morsitans, to feed on birds, amphibians or reptiles (54), thus they were not attracted to light traps used in animal (cattle and sheep) shelters during the study.
Also, there are seven species of the Maculipennis Group in Guilan Province differentiated by egg patterns or the polymerase chain reaction (PCR) technique (36), which could not be differentiated with the morphologybased keys of females and larvae (29) used in the present study.

Species dominance structure
In the present study, the most abundant species were Cx. theileri, Cx. tritaeniorhynchus, Cx. pipiens, Ae. vexans, An. pseudopictus and An. maculipennis s.l., respectively. With the exception of An. maculipennis s.l., which is subdominant, they are all dominant according to the classification (52,53) (Table 3). This is concordant with the previous findings in the province based on collections of larvae (36)(37)(38), as well as adults (37,60). The exception is Cx. theileri. The species had been found less often in the larval stage than any other Culex in the province (38). One reason is probably due to sampling, the heavy rainfall in the province results in a great number of different natural larval habitats that are favorable for Cx. theileri (38), but those habitats are not easily located and sampled. On the other hand, the favorable larval habitats of some species, such as Cx. pipiens and Cx. tritaeniorhynchus, i.e. artificial containers and rice fields, respectively, are easier to find and sample. Another reason is probably the biology of the species, some species such as Cx. hortensis and Cx. territans, which do not bite humans and mammals but mostly feed on amphibians, reptiles or birds (54), have been collected very often as larvae (38). They were not collected during the present investigation by means of aspirators and light traps from animal shelters which attract Cx. theileri ( Table 3). The most prevalent species of the province, An. maculipennis s.l., An. pseudopictus, Ae. vexans, Cx. pipiens, Cx. theileri and Cx. tritaeniorhynchus are known vectors of both WNV and Dirofilaria (1,57,59).

Seasonal activity and the fluctuations of rainfall and temperature
During the present investigation, Ae. vexans was one of most abundant species and most prevalent aedine species (Table 3), as noted previously (37,60). However, most specimens were collected from Anzali (Table 4) and the species was not collected from the fixed site, so its seasonal activity is not discussed here. Anopheles maculipennis s.l. showed the peak of activity in the mid-Jul (Figs. 3, 4). The peak of monthly activity of anophelines (including An. maculipennis s.l. and An. superpictus) was reported during Jul-Aug in Kalaleh County of Golestan Province, northern Iran (47). The most An. maculipennis s.l. was captured in Aug in Aras Valley, Turkey, adjoining Iran (61). There are no records for the seasonal activity of culicine adults in Iran. The peak of activity of Cx. pipiens was found in Jul in northern Italy (62). That is in concordance with the result of the present study (Figs. 5, 6). However, the peak of activity of Cx. pipiens was recorded in Aug in the Belek Region and Aras Valley of Turkey (61,63). Most Cx. theileri was found in Jun in Aras Valley of Turkey (61), which is similar to the present study (Figs. 5, 6), however, the peak activity of this species was reported in Aug in Ankara, Turkey (64). Moreover, most Cx. tritaeniorhynchus was captured in Aug in Belek Region of Turkey (63), while the peak of activity was observed in Jul in the present study (Figs. 5,6). Differences between the results of the present investigation and the findings in other regions may be explained by differences in the topography and climates (especially temperature) which influence the bionomics of mosquitoes. On the other hand, some differences are due to sampling regimes. For example, mosquito abundance was reported based on weekly catches (62), and on monthly catches ( (61) and in this study. Though the mean monthly temperature of about 16 °C is a limiting factor in the activity of adult mosquitoes in the study area (Figs. 3,5), no significant regression was observed between different meteorological data ( Table  2) and the abundance of adult mosquitoes during active season (P> 0.05, R 2 = 0.31). During the present study, the peak of activity of most adult mosquitoes was late Jun to mid-Jul, only the peak of activity of Cx. tritaeniorhynchus was in late Jul (Figs. [3][4][5][6], after which the abundance dramatically decreased as temperature increased by Aug (Figs. 3, 5). High temperature (> 35 °C) is generally a limiting factor for the abundance of adult mosquitoes, especially in localities with warm climate such as southern Iran (41) and Saudi Arabia (65). However, it does not seem that temperature was a key factor in decreasing the abundance of adult mosquitoes in the area of the present study, because the temperature does not exceed 34 °C and the monthly mean temperature is lower than 30 °C in the fixed site in Aug (Figs. 3, 5). In addition, rainfall decreased in the fixed site during Jun and Jul while the abundance increased. The rainy season started in Sep while the abundance of mosquitoes dramatically decreased (Figs. 4, 6). A key factor mentioned here is rice fields, the main larval habitats of the most prevalent species, are dry during Aug. Temperature decreases significantly during Sep and Oct, consequently, the prevalence of mosquitoes decreases.
In view of integrated vector management, ecological data, especially seasonal activity, is very important for intervention measurements. On the other hand, one of main intervention measurements is using pesticides yet. There is little-published data about the susceptibility sta-tus of mosquitoes, especially culicines (66,67), in northern Iran. This subject can be a goal for forthcoming studies in Guilan Province.

Conclusion
Though there was no significant regression between the abundance of adult mosquitoes and the meteorological data in the fixed site during active season, temperature and rice fields had a great influence in starting and ending active season in the region. The seasonal activity of the important species Ae. vexans, other species found less abundant in this study, host preference analysis and filarial and arbovirus screening should be the subjects of future investigations in the region.