Dengue virus in humans produces a wide symptomatic spectrum ranging from a mild clinical profile to dengue with warning signs, and can evolve to severe dengue. Factors related to outcome are usually associated with age, secondary infection, immunologic status, dengue serotype, orgenotype [1]. There is significant global diversity among dengue virus (DENV) strains. The four serotypes of DENV (DENV-1, DENV-2, DENV-3, and DENV-4) belong to the genus Flavivirus of the family Flaviviridae. They are genetically distinct, but cause similar diseases and share epidemiological features [2].

In the past 50 years, the incidence of dengue fever (DF) has increased 30-fold with increasing geographic expansion into new countries [3], including Brazil which had no reports of dengue cases for almost 40 years because of the successful elimination of Aedes aegypti attained in the 50’s. However, the virus was reintroduced in early 80’s in Brazil and rapidly spread all over. The majority of annual dengue notifications in the Americas are now from Brazil [4,5]. Dengue has been a major public health problem in the State of Rio Grande do Norte (RN) and other states in Brazilfor over twenty years. The first laboratory-confirmed cases in RN occurred in 1994. All four DENV serotypes have been found, and the frequencies of these serotypes within the population vary by geographic region of the state and year [6]. A total of 32,328 dengue cases and 40 deaths were reported in Natal Cityfrom 2011 to 2014, and within this period the highest number of cases occurred in 2012 (S1 Fig).

Transmission of DENV largely occurs from infected mosquitos to humans, defined as horizontal transmission, and it is largely considered the major route by which the virus remainsin an area. However, vertical transmission within the mosquito (infected female mosquito to infected offspring) has been suggested as a mechanism that ensures conservation of the virus during conditions that would be adverse for horizontal transmission (i.e. harsh winters, inter-epidemic stages) [7], and potentially influence on the epidemiology of dengue infection [8], thus leading to persistence of the virus without apparent need of human infection. Vertical transmission has been reported in A. aegypti under experimental [9,10] and natural conditions in different locationsin the world such as Myanmar [11], Mexico [12], Brazil [13] and Cuba [14].

Like A. aegypti, Aedes albopictus has now been implicated as an important vector for DENV [15–17], Chikungunya virus (CHIKV), and potentially Zika virus (ZIKV) transmission in African and European countries [15,18–20], and in mainland India [21]. More recently, CHIKV and ZIKV have become highly endemic in Latin America after their introduction in Brazil, and these viruses have rapidly spread worldwide.

The monitoring of dengue viruses in immature and adult mosquitoes represents an important strategy for identifying the beginning of the epidemic period and direct control actions, especially in difficult-to-access areas [22,23]. Therefore, the purpose of this study was to determine the incidence of adult and larval/pupae of A. aegypti and A.albopictus with DENV in urban areas in the state of Rio Grande do Norte, Brazil, in order to estimate the potential magnitude of dengue infection persistence in these insects.

The state of Rio Grande do Norte (RN), Brazil, has reported cases of dengue fever since October 1994, when the first autochthonous cases were reported in the municipality of Assú after an out-of-season carnival in the region. Since then, this state has demonstrated an endemic profile for dengue, with increasingly frequent and intense epidemic years due to several aspects, such as an emergence or reemergence of DENVserotypes. In 2015, Zika and Chikungunya emerged with great impact, including an increase in Microcephaly and Guillain-Barre Syndrome cases. Vector control remains the sole effective method to prevent DENV transmission, although a vaccine for dengue has recently become available and testing of its efficacy and coverage isbeing performedin multiple places. Entomological surveillance is a key factor in alerting authorities to possible outbreaks, but until now natural DENV infection of mosquito populations has been scarcely used as an early warning system to monitor fluctuating prevalence of infected mosquitoes [14]. We have focused ondeterming the magnitude of horizontal infection in DENV vector transmitters A. aegypti and A. albopictus. This study provides the first description of the natural transovarian transmission of dengue virus in A. aegypti collected in urban areas of the state of Rio Grande do Norte, Brazil. The transovarial transmission of dengue virus by Aedes mosquitoes is a well documented phenomenon in endemic areas worldwide (S1 Table). In this study, DENV-4 was identified in 4 pools of A. aegypti larvae/pupae with an infection rate of 3.37, a similar value found by Rohani et al. [27] in Malaysia (MIR = 3.9–100), and by Gunther et al. [12] in Mexico (MIR = 4.6), both in A. aegypti.

The presence of A. albopictus was identified in four urban areas in Natal: Felipe Camarão, Nova Descoberta, Potengi and Quintas. This may be explained by the environmental characteristics of the study site. Natal contains a large preservation area of wooded dunes that is a remnant of the Atlantic Forest ecosystem in Brazil called the "Dunes State Park" (05 ° 46'S, 35 ° 12'W). This park is crucial for water supply, outdoor activity and cultural activity. It encompasses 1,172.8 ha (15 x 2 km) and borders the Atlantic Ocean along its east bank andthe urban area of Natal along the west side of the park. Tropical fauna includes birds, marsupials, and marmosets. There are more than 350 species of native flora including trees, shrubs, bromeliads, orchids and grasses [28]. The proximity of the Dunas State Park to the urban area can promote adequate environmental conditions for the circulation of A. albopictus in both areas. No DENV was found in A. albopictus larvae. In contrast, a study conducted in Malaysia showed that 18/363 (4.9%) of larvae pools from A. albopictus were positive for dengue virus [27].

This study allowed the detection of dengue virus type 4 during the study period. A previous study conducted in the same geographic region in 2011showed DENV-1, DENV-2 and DENV-4 co-circulating, thus highlighting the important introduction of DENV-4 in the State of Rio Grande do Norte in May 2011 after a religious event held in Santa Cruz, a municipality located 100 km from Natal. DENV-4 was detected in Natal right after this event, demonstrating the speed with which these viruses can be transmitted with migratory or population movement, allowing rapid dissemination of the virus once the vectors are available. In 2012, this same study reported that the circulation of only DENV-4 was confirmed [6]. Co-circulation of serotypes 1, 2 and 4 was also shown in the years 2013 and 2014, although predominantly DENV-4 [29]. The present study reinforces the importance of applying molecular techniques such as RT-PCR in the detection of dengue virus in larvae and pupae collected in the field. Similar findings were also reported by Chao et al. [30] and Rohani et al. [27]. These techniques are practical diagnostic tools for surveillance of DENV in Aedes mosquitoes, serving as a warning sign during the onset of dengue transmission and could be of use to predict potential complications of multiple DENV serotypes circulating.

This work represents the most complete study to date on the interaction between dengue viruses and A. aegypti and A. albopictus mosquitoes in the State of Rio Grande do Norte, and raises important questions about a possible role of A. albopictus in the transmission of dengue virus in Brazil. Currently, the vector-fighting strategy in Brazilhas been mainly focused on A. aegypti; however, few studies have been conducted to know the role of other species in dengue transmission.

It was possible to identify three dengue serotypes in A.aegypti and A. albopictusadults: DENV-1, 2 and 4, with DENV-4 being the most prevalent, being identified in 18.4% of the studied pools. DENV-2 was only identified in A. aegypti adults. DENV-1 and DENV-4 were identified in A. aegypti and A. albopictus adults. Despite this important finding, it was not possible to evaluate the vector capacity of the investigated mosquito species. Thus, additional studies should be performed for viral identification in the salivary gland of the mosquito.

The co-circulation of three of the four dengue serotypes (DENV-1, DENV-2 and DENV-4) was clearly demonstrated in our study. This finding is in fact concerning, and should serve as an alert to health authorities for the possibility of co-infection events by more than one serotype, and the current problem is further aggravated by the co-circulation of dengue, zika and chikungunya viruses in Brazil.

In addition, the minimum infection rate in adults of the Aedes genus was 19.8, considering both species. Considering only the A.aegypti species, the MIR was 16.2. In analyzing the infection individually by DENV-1, DENV-2 and DENV-4 in A.aegypti, the MIR was 0.7, 0.7 and 14.6, respectively. Similar minimal infection rates were found in A. aegypti by Urdaneta et al. [31] in Venezuela (MIR = 15.9), and Garcia-Rejon et al. [32] in Mexico (MIR = 18) (S2 Table). It was not possible to calculate the MIR of the A. albopictus species, since less than 1000 individuals of this species were collected.

In conclusion, this study reveals the importance of arbovirus research in A. aegypti, A. albopictus and other species. The early detection of these viruses in vectors during entomological and virological surveillance could allow for more intense targeted control actions.