Portal de Periódicos da CAPES

Prospective field study of transovarial dengue-virus transmission by two different forms of Aedes aegypti in an urban area of Bangkok, Thailand

2011; Wiley; Volume: 36; Issue: 1 Linguagem: Inglês

10.1111/j.1948-7134.2011.00151.x

1948-7134

Supatra Thongrungkiat, Pannamas Maneekan, Ladawan Wasinpiyamongkol, Samrerng Prummongkol,

Malaria Research and Control

A prospective field study was conducted to determine transovarial dengue-virus transmission in two forms of Aedes aegypti mosquitoes in an urban district of Bangkok, Thailand. Immature Aedes mosquitoes were collected monthly for one year and reared continuously until adulthood in the laboratory. Mosquitoes assayed for dengue virus were processed in pools and their dengue virus infection status was determined by one-step RT-PCR and nested-PCR methods. Of a total 15,457 newly emerged adult Ae. aegypti, 98.2% were dark and 1.8% of the pale form. The results showed that the minimum infection rate (MIR) by transovarial transmission (TOT) of dengue virus during the one-year study ranged between 0 to 24.4/1,000 mosquitoes. Dengue virus TOT increased gradually during the hot summer months, reaching a peak in April-June, while dengue cases peaked in September, a rainy month near the end of the rainy season. Therefore, mosquito infections due to TOT were prevalent four months before a high incidence of human infections. TOT dengue virus infections occurred in both forms of Ae. aegypti. All four dengue serotypes were detected, with DEN-4 predominant, followed by DEN-3, DEN-1, and DEN-2, respectively. Dengue is the most important mosquito-borne viral disease of humans in the world as a result of the geographic spread of dengue epidemics with increasing frequency and severity. Dengue is now endemic in all WHO regions, except the WHO European region (WHO 2009). The incidence of dengue is still on the rise in many tropical and subtropical countries, especially in Southeast Asia. It is caused by any of four distinct dengue virus serotypes (DEN-1 to –4) and is transmitted to humans through the bite of infected female Aedes mosquitoes. Aedes aegypti and Aedes albopictus are the known primary vectors of dengue viruses. However, Ae. aegypti is considered the main vector of the dengue virus in Asia, due to its wide distribution in both urban and rural areas. In endemic areas, dengue viruses are partially maintained by the human-Aedes mosquito-human cycle (Gubler and Trent 1994). Current virological surveillance, by detection and serotyping of dengue viruses in both humans and wild-caught Aedes mosquitoes, has been widely undertaken (Lanciotti et al. 1992, Chow et al. 1998, Urdaneta et al. 2005, Guedes et al. 2010). Ae. aegypti mosquitoes are able to transmit the dengue virus vertically, or transovarially, to their offspring after becoming infected, either by oral feeding or intrathoracic inoculation under laboratory conditions (Lee et al. 1997, Mourya et al. 2001, Joshi and Sharma 2001, Joshi et al. 2002, Wasinpiyamongkol et al. 2003). In addition, transovarial transmission has been reported in nature by the detection of the dengue virus in the field-collected larvae and field-caught adult male mosquitoes (Khin and Than 1983, Chung et al. 2001, Lee and Rohani 2005, Arunachalam et al. 2008, Angel and Joshi 2008, Guedes et al. 2010). Therefore, transovarial transmission (TOT) has been suggested as a potential influence on the epidemiology of dengue infection. It has been proposed as an important mechanism for the maintenance of the dengue virus in vector populations during inter-epidemic periods (Rhodain and Rosen 1997). Integrated vector management uses well-established methods for suppressing the vector populations to prevent and control dengue outbreaks. However, improved knowledge of dengue transmission and effective proactive surveillance systems are needed to improve dengue prevention and control strategies. Dengue is endemic in Thailand, where the trend of morbidity from 2004 to 2008 increased from an annual incidence of 62.6/100,000 to 141.8/100,000 (Bureau of Epidemiology). Bangkok, the largest metropolitan area in the central region, has also been reported as the one with the highest prevalence of dengue. Ae. aegypti has been incriminated as the key dengue vector in dengue outbreaks. Three forms of Ae. aegypti, which differ by variations in the adult abdominal tergal white scale patterns, have long been recognized (Mattingly 1957, 1958). Only two forms of Ae. aegypti (the dark form, or Ae. aegypti type form, and the pale form, or form queenslandensis) have been reported in Bangkok. (Sheppard et al. 1969, Sucharit and Surathin 1994). Both are susceptible to dengue virus type 2 (DEN-2) and have been shown to be capable of transovarial transmission in laboratory experiments (Wasinpiyamongkol et al. 2003). In this study, we demonstrated the natural transovarial transmission (TOT) of dengue viruses by conducting a prospective field study to determine dengue virus infections in the two forms (dark and pale) of Ae. aegypti mosquitoes. This is the first report to show the evidence of TOT together with the dengue cases in an endemic, urbanized area of Bangkok, Thailand. The study was conducted in Bang Khun Thian District, an urbanized residential area of Bangkok that has reported dengue outbreaks almost every year. This district was among the top five for dengue cases in Bangkok during the study year. Emergency measures were undertaken to interrupt transmission, space-spraying insecticide for adult mosquitoes in houses with reported dengue and four to five neighboring houses. Mosquito larvae were collected monthly in the field for one year (October, 2007 to September, 2008), from patients' houses and 8–12 nearby houses with permission. Immature mosquitoes were captured from domestic water storage containers and discarded trash, such as bottles and used tires, in and around houses. The larvae were reared continuously until adult emergence at 28° C in the laboratory of the Department of Medical Entomology, Faculty of Tropical Medicine, Mahidol University in Bangkok, Thailand. Prior to adulthood, male and female pupae were separated according to their sizes, males being markedly smaller, and allowed to emerge in separate containers. Newly emerged adults were anaesthetized with cold, identified to species, and the Ae. aegypti were classified morphologically into two forms: the dark form (Ae. aegypti type form) and the pale form (form queenslandensis) based on the dorsal scaling on the abdominal tergite (Mattingly 1957, 1958). The type form has only a white scale on the first abdominal tergite, while the form queenslandensis has extended white scaling on the abdominal tergites beyond the first. After identification, they were kept at –80° C until assayed for dengue virus infection by RT-PCR. Mosquitoes assayed for dengue virus were processed in pools and sorted by collection month and forms, resulting in variable pool sizes of between 9–40 mosquitoes/pool. Dengue infection status was determined by one-step RT-PCR and dengue viral serotype by nested PCR techniques, using the methods described by Lanciotti et al. (1992), with some modification. Briefly, the mosquito pools were ground in 200 μl of cold PBS at pH 7.4 in a 1.5 ml microfuge using a sterilized micropestle, then centrifuged at 13,000 rpm, at a temperature of 4° C, for 30 min. The supernatants were then collected. Viral RNA was extracted with a viral RNA mini kit (QIAmp, QIAGEN, GmbH, Hilden, Germany) according to the manufacturer's protocol to extract the RNA from each mosquito specimen. Then, a one-step RT-PCR kit (QIAGEN, GmbH, Hilden, Germany) was used with the extracted RNA as in the RT amplification step at 42°C for 1 h, following by 35 cycles of denaturation (94° C, 30 s), annealing (55° C, 1 min), and extension (72° C, 2 min). The one step RT-PCR products, 511 bp amplified of C and PrM of DENV with primer D1 (forward, 5′- TCAATATGCTGAAACGCGCGAGAAACCG-3′) and D2 (reverse, 5′-TTGCACCAACAGTCAATGTCTTCAGGTTC-3′), were then processed in the second round of amplification to identify each dengue virus serotype using Taq polymerase (QIAGEN, GmbH, Hilden, Germany). These one-step RT-PCR products were diluted to 1:50 with water and accompanied with primer for DENV 1–4 (D1 and TS1 (5′-CGTCTCAGTGATCCGGGGG-3′), TS2 (5′-CGCCACAAGGGCCATGAACAG-3′), TS3 (5′-TAACATCATCATGAGACAGAGC-3′), and TS4 (5′-CTCTGTTGTCTTAAACAAGAGA-3′)), followed by 25 cycles of denaturation (94° C, 30 s), annealing (55° C, 1 min), and extension (72° C, 2 min). The DEN 1– 4 products were 482, 119, 290, and 392 bp, respectively. A 5 μl volume of each reaction was run in 1.5% agarose gel containing ethidium bromide, then electrophoresed. The gel was exposed to a UV transilluminator and photographed using SynGene gel documentation equipment. TOT dengue-virus infections are estimated using variable-sized pools of mosquitoes (Walter et al. 1980). It is expressed as minimum infection rate (MIR) per 1,000 mosquitoes, which is calculated as the ratio of the total number of positive pools to the total number of mosquitoes determined, multiplied by 1,000. A total of 15,457 mosquito larvae was collected during the study period from October, 2007 to September, 2008. All of the emergent adults were Ae. aegypti as expected, since the study environment was an exclusive habitat for this species. Of the 15,457 mosquitoes collected, 15,179 (98.2%) were the Ae. aegypti type form, and 278 (1.8%) were the pale form (Figure 1). Of these, 283 pools sorted by collection month and form were identified with dengue virus infection by one-step RT-PCR and nested-PCR methods (Figure 2). The majority of the pools contained 35 mosquitoes/pool. The results showed that natural transovarial transmission (TOT) of dengue virus infection occurred in the study area. TOT dengue virus appeared in mosquitoes almost every month, except for December, for which no positive pool was detected. The minimum infection rate (MIR) during the one-year study period ranged between 0–24.4 per 1,000 mosquitoes (Figure 3). In Thailand, dengue has been reported year round, but outbreaks tend to occur during the rainy season. Maximum transmission starts in July-September each year. In the study area, it was found that the dengue-virus TOT increased gradually during the hot summer season, reaching a peak in April-June, and then declined. No correlation was found between TOT dengue-virus-infection rate and either dengue cases or rainfall, r = 0.221 p = 0.490, and r = 0.311 p = 0.325, respectively. However, it appeared that a high rate of TOT dengue-virus-infection occurred four months before a high incidence of human cases, with a case peak in September, 2008 (Figure 3). TOT dengue virus infection was present in both forms of Ae. aegypti. The MIR for dark and pale forms were 15.6/1,000 and 12.9/1,000 mosquitoes, respectively. During the study period, all four serotypes were simultaneously identified in the positive pools. Of 283 pools, 119 were positive for dengue-virus infections, with 57 (47.9%) positive for DEN-4, 16 (13.4%) DEN-3, 6 (5.04%) DEN-1, 4 (3.4%) DEN-2, and 36 (30.3%) positive for mixed dengue serotypes. Aedes aegypti, dark form (left) and pale form (right). Gel electrophoresis, illustrating Aedes aegypti mosquito samples by nested PCR method (lanes 3–14). Lanes 1 and 17 are 100 bp ladder. The positive controls in lanes 2 and 15 are a combination of 4 DEN 1–4 serotypes 482, 119, 290, and 392 bp, respectively. The negative control is H2O (lane 16). Monthly TOT dengue-virus infections, expressed as MIR/1,000 mosquitoes, number of dengue cases, and rainfall. Transovarial transmission (TOT) of dengue virus infection has been reported in Aedes mosquito vectors and has become a topic of research interest in many endemic countries around the world. It has been demonstrated that the dengue viruses can undergo TOT over several generations of Aedes aegypti mosquitoes under laboratory conditions (Joshi et al. 2002, Wasinpiyamongkol et al. 2003, Rohani et al. 2008). In addition, Mourya et al. (2001) showed that transovarially infected female mosquitoes could transmit dengue virus orally. The study also found that vertical transmission increased after several weeks of TOT-egg incubation at room temperature. Therefore, it was possible that these events might occur in nature, and could have important epidemiological consequences for dengue transmission. Many surveillance studies of natural TOT dengue virus have been performed. The findings from the present study of TOT were consistent with those of other studies finding that TOT of dengue virus has occurred. This is the first report from a longitudinal field study on the occurrence of TOT of dengue virus by Ae. aegypti in an urban area of Bangkok, Thailand. Based on the morphological variants of Ae. aegypti, the present results revealed that the dark form comprised a higher proportion (98.2%) than the pale form (1.8%) in the study area. This supported the previous study by Sheppard et al. (1969) that Ae. aegypti populations in Bangkok predominantly consisted of the dark type form. Transovarial transmission (TOT) of dengue virus infection was detected in both forms of Ae. aegypti. Thus, although the dark-form population was significantly higher than the pale form, the latter remains important, since it has been shown that a similar TOT dengue virus infection rate, with an especially high rate of TOT dengue virus occurred in the male pale-form population (unpublished data). Moreover, from the previous studies, the pale form has been found to be more susceptible to DEN-2 than the dark form under laboratory conditions (Wasinpiyamongkol et al. 2003). Several surveillance studies have suggested that TOT occurs at a very low rate (Khin and Than, 1983, Gunther et al. 2007, Akbar et al. 2008). Failure to detect natural TOT dengue-virus infection has also been reported (Hutamai et al. 2007, Zeidler et al. 2008). Numerous factors affect the detection of dengue viruses in mosquito samples, including the sensitivity of the testing methodologies and the appropriate field-work, such as the locality of mosquito collection, duration of collection, mosquito sampling, etc. The RT-PCR technique we used is a powerful, highly sensitive, virus surveillance tool that can be used rapidly and accurately to detect dengue virus serotypes in pools of infected mosquitoes. The mosquito collection locality was important since recently, mosquito surveys of repeatedly epidemic areas for dengue in northern Thailand have found no TOT dengue virus despite a high incidence of dengue cases during the study period (Hutamai et al. 2007). This may be associated with imported dengue viruses from humans visiting an area that then facilitates transmission to other communities where Ae. aegypti are abundant. However, our selected study area had a history of dengue outbreaks every year. Therefore, TOT may facilitate the persistence of dengue virus in a re-epidemic area, and is notable as a focus for the dissemination to areas at risk of dengue. Dengue infection still lacks an available and effective vaccine/drug treatment and effective warning parameters are needed to monitor the dengue situation, thus enabling an early outbreak response. Traditional dengue surveillance in most countries normally relies on case-based surveillance, laboratory-based surveillance, and surveillance of Aedes population densities. It is likely that the incidence of dengue is frequently underestimated, and an epidemic may have occurred before it was recognized. The currently available mosquito-density assessment methods employ Ae. aegypti larval sampling surveys and are expressed as three larval indices, but these do not provide sensitive indicators for forecasting a dengue outbreak. Previous virologic surveillance, which determined dengue-virus infection in field-caught adults or TOT dengue-virus infection either in larvae or adult male mosquitoes, were analyzed in correlation with actual dengue cases. The results showed that dengue virus-infected field-caught Ae. aegypti were detected six to eight weeks before the start of a dengue outbreak among humans (Chow et al. 1998, Urdaneta et al. 2005). Interestingly, our findings also showed a positive relationship between high TOT dengue-virus-infection rates in emerged adults from field-collected larvae, four months before reports of high numbers of human dengue cases. The present study showed that the TOT dengue-virus infection rate was high in April (dry summer season) when dengue cases began to rise. Our results were consistent with the study of Angel and Joshi (2008) that found high transovarial transmission of dengue virus occurred in mosquitoes during the summer season, before the active dengue-transmission season during the following rainy season. Therefore, the results of the present study and earlier studies support the occurrence of high TOT dengue-virus-infection in mosquito vectors as being a significant epidemiological signal for potential application as a surveillance tool in an early-warning system for dengue outbreaks. Whether the emergence of adults infected by TOT triggers or enhances the cycle of dengue outbreaks remains uncertain. However, the increase in the TOT dengue-virus-infection rate may reflect the number of infected parent mosquitoes, either by horizontal or vertical transmission, in a particular area. For practical reasons, the methods of collecting adult mosquitoes is more laborious and time-consuming, due to low infestation levels, besides being less safe than the approach using Aedes larvae including sampling pupae. TOT dengue-virus surveillance using collected larvae thus enables easier and earlier detection than virological surveillance of field-caught infected adult mosquitoes. Moreover, Guedes et al. (2010) reported that dengue virus surveillance of immature forms provided useful information about the adults regarding dengue-virus serotypes circulating in the Ae. aegypti population. Studies of endemic situations have shown that several factors influence the dynamics of dengue transmission, including seasonal climate (e.g., temperature, rainfall, humidity) and non-climatic factors (e.g., herd immunity, social factors). Climatic factors directly influence the biology of vectors and dengue viruses, and are thus important determinants of dengue epidemics (WHO 2009). However, the relationships between these factors are complex and depend on which ones are more prevalent. A strong correlation between dengue cases and rainfall (r=0.899, p < 0.001) was also observed in the study area, as reported by another study conducted in Bangkok (Halstead 2008). Several studies revealed that dengue hemorrhagic fever in Thailand has a substantial rainy-season pattern due to increased adult survival and longevity related to temperature and humidity rather than mosquito density. Temperature also affects the gonotrophic cycle and extrinsic incubation period (Thammapalo et al. 2005, Halstead 2008, Barbazan et al. 2010). In conclusion, the surveillance of dengue virus TOT in immature mosquito forms from the natural environment suggests some potential for identifying disseminated dengue outbreak areas quickly. More importantly, rising TOT dengue-virus-infection rates provide an early warning signal of an impending dengue epidemic so that pre-epidemic control interventions can be implemented to suppress disease transmission and to prevent the disease spreading to new areas. However, the TOT indicator should be examined further with larger areas and over longer times, also taking into account various factors that could influence the TOT mechanism, such as rainfall, humidity, and temperature. The authors thank Mrs. Phassapong Nimsumlee, Bang Khun Thian district officer, for her help in identifying the study area and providing dengue case and rainfall data. Thanks also to the staff of the Department of Medical Entomology, Faculty of Tropical Medicine, Mahidol University, for their assistance in both the field and laboratory work. Thanks to Mr. Paul Adams for corrections to the English language in the manuscript. This research was supported by the National Research Council of Thailand.