|
Potential Vectors of Dirofilaria immitis (Leidy, 1856) in Itacoatiara, Oceanic Region of Niteroi Municipality, State of Rio de Janeiro, Brazil Norma Labarthe/^+, Maria Lucia Serrao*, Yuri Fontenele Melo, Sebastiao Jose de Oliveira**, Ricardo Lourenco-de-Oliveira*
Faculdade de Veterinaria, Centro de Ciencias Medicas, Universidade Federal
Fluminense, Rua Vital Brazil Filho 64, 24230-340 Niteroi, RJ, Brasil Received 6 January 1998; Accepted 17 April 1998
Code Number:OC98083 Dirofilaria immitis is a widespread mosquito-borne parasite that causes dirofilariasis, a commonly diagnosed disease of dogs that is rarely reported in cats and humans. A mosquito survey was conducted in Itacoatiara in the State of Rio de Janeiro, from March 1995 to February 1996, using canine, feline and human baits. A total of 3,667 mosquitoes were dissected for D. immitis larvae, representing 19 species and 10 genera. From those, Ae. scapularis, Ae taeniorhynchus, Cx. quinquefasciatus, Cx. declarator, Cx. saltanensis and Wy. bourrouli were found infected with D. immitis parasites, and among those, only the first three harbored infective larvae. The majority of larvae were found in the Malpighian tubules (889/936), and larval melanization was observed in the two Aedes species. In descending order, the best vectors were Ae. scapularis, Ae. taeniorhynchus, and Cx. quinquefasciatus which alternate seasonally in importance. Cx. quinquefasciatus is suggested to be a vector to cats. The potential transmission of D. immitis parasites by these three vectors to man is discussed. Key words: Dirofilaria immitis - mosquitoes - heartworm - potential vector
Dirofilaria immitis (Leidy 1856) is a widespread mosquito-borne nematode parasite of dogs. While canine heartworm is enzootic in many areas (Guerrero et al. 1992a), feline dirofilariasis is much less common. Feline heartworm infections parallels that of dogs in a defined area. While feline dirofilariasis is at a lower infection rate than canine dirofilariasis (Dillon 1988, Elkins & Kadel 1988), it is thought to be increasing in prevalence and distribution (Guerrero et al. 1992b). Heartworm is also a zoonosis (OMS 1979). Heartworm is common in dogs in Rio de Janeiro, Brazil (21.3%). In the coastal region of the county of Niteroi an even higher prevalence is observed (37.5%), and microfilaremic dogs, the source of mosquito infections, account for 25% of the local population (Labarthe et al. 1997a). Dogs may have high levels of microfilaremia (10^3 to 10^5/ml) (Lok 1988). Mosquitoes blood-feeding on dogs with even moderate numbers of microfilariae frequently die (Sauerman 1980). Mosquitoes need to survive D. immitis infection in order to support the extrinsic cycle of the nematode through development of infective third stage larvae (L3). Vector survival seems to rely on biochemical reactions which limit the nematode burden in the mosquito (Christensen 1977, 1978, 1981). Intrinsic barriers to D. immitis development in mosquitoes include: larval damage due to the cibarial armature (Coluzzi & Trabucchi 1968) or to the presence of oxyhaemoglobin crystals formed by bloodmeal coagulation in the midgut (Nayar & Sauerman 1975, Lowrie 1991); trapping of larvae in the coagulated bloodmeal in the mosquito's midgut (Kartman 1953); lysis of larval cuticle by host cells (Talluri & Cancrini 1994) and other immune responses, i.e., encapsulation and/or melanization of the parasite in the mosquito Malpighian tubules (Lindemann 1977, Christensen 1981, Christensen et al. 1989). Among extrinsic factors, temperature is the most important and has been shown to regulate the duration of parasite development in the mosquito (Kutz & Dobson 1974, Christensen & Hollander 1978). Laboratory and field data show that many mosquito species, from different geographic areas worldwide, are susceptible and yield infective larvae under field or laboratory conditions. Although more than 60 mosquito species have been identified as potential vectors of D. immitis worldwide (Ludlam et al. 1970), their vectorial capacity varies (Sauerman 1980). In the Americas, mosquito species belonging to the subgenus Ochlerotatus of Aedes are considered the best vectors of heartworm (Yen 1938, Kershaw et al. 1953, Ludlam et al. 1970, Christensen 1977, Arnott & Edman 1978, Buxton & Mullen 1980, Walters & Lavoipierre 1982, Sauerman & Nayar 1983, Ernst & Slocombe 1984, Hribar & Gerhardt 1985, Roberts et al. 1985, Johnson & Harrell 1986, Parker 1986, 1993, Scoles et al. 1993, Loftin et al. 1995). In Brazil, there are two reports: one under experimental conditions with Ae. fluviatilis (Lutz) that suggested that although infective larvae have been found in some individuals, this species is not likely to be an efficient vector in nature (Kasai & Williams 1986) and another in which Ae. taeniorhynchus (Wiedemann) and Ae. scapularis (Rondani) were shown to be suitable natural potential vectors of heartworms in Rio de Janeiro (Lourenco-de-Oliveira & Deane 1995). In nature, the complete life cycle of the mosquito must be considered before concluding either that a given species or strain is a probable vector of D. immitis. The present study examines infection rates of mosquito species with D. immitis at Niteroi in relation to their biology (feeding behavior and seasonal biting frequency). MATERIALS AND METHODS Mosquitoes were collected in Itacoatiara, State of Rio de Janeiro, where the average prevalence of canine microfilaremic heartworm is 31.7% (unpublished data). Data on the collection sites and methods are available in Labarthe et al. (1998). Briefly, mosquitoes were captured four days each month using a dog, a cat and two human volunteers as baits, from March 1995 until February 1996. Mosquitoes were kept in cylindrical cages of 8.5 cm diameter at 28 C, 80% RH and provided with a 10% glucose solution. Dissection of live females was initiated immediately after collections, and all mosquitoes were dissected within five days of collection. After chloroform anesthesia, each mosquito was identified using taxonomic keys of Lane (1953) and Consoli and Lourenco-de-Oliveira (1994). Subsequently, the head was placed in a small saline droplet on a microscope slide, and the alimentary tract and Malpighian tubules were drawn from the body into another saline droplet by gently pulling the terminal segments backwards with hypodermic needles. The thorax was teased apart in a third droplet of saline. All preparations were examined for the presence of worms under microscopic magnification after placement of a coverslip. Larvae found in the mosquitoes were identified as D. immitis based on: morphological characteristics previously described (Taylor 1960) and those observed in experimental infections (Macedo et al. 1998); the Malpighian tubule developmental site of worms, known only among species of the genus Dirofilaria of Onchocercidae (Symes 1960, Walters & Lavoipierre 1982, Sauerman & Nayar 1983); the only Dirofilaria species known from the lowlands of Rio de Janeiro is D. immitis (Lourenco-de-Oliveira & Deane 1995); and the mosquito collecting site is an active D. immitis transmission focus (Labarthe et al. 1997a). The infection rate was determined as the percentage of numbers of a species infected with any stage larvae (Chandra et al. 1996). The transmission capacity of each species was determined by the annual transmission potential (ATP) that corresponds to the sum of the monthly transmission potentials (MTP) (WHO 1987), where:
no. of L3 no. in the head and proboscis x no. of days in the month MPT = mosquitoes x -------------------------------------------------------- caught no. of mosquitoes dissected x no. of catching days The frequencies of infected and infective mosquitoes (with L3 in head and proboscis) were analyzed by chi-square or when values were less than 5, Fisher's exact test was used (Mood & Graybill 1963, Rodrigues 1993). RESULTS A total of 3,667 mosquitoes belonging to 19 species and 10 genera was captured and dissected. From those, six species were found naturally infected with various larval stages. Only Ae. taeniorhynchus, Cx. quinquefasciatus and Ae. scapularis harbored infective larvae in the head and proboscis, and so were rated as infective ( Table I). Traditional statistical analysis crossing species versus infective and infected mosquitoes showed little or no relevance to the objective of the present study as chi-squares and Fisher tests became less statistically significant as relevant species were segregated and crossed among each other. When the number of uninfected Ae. taeniorhynchus, Cx. quinquefasciatus, Ae. scapularis and Cx. declarator captured were analyzed versus their infected members, Ae. taeniorhynchus was the most infected, followed by Ae. scapularis, Cx. quin-quefasciatus and Cx. declarator (chi-square 19.1, p<0.01). When infectivity was analyzed, Fisher test showed p>0.07 for all combinations among the three species found infective. A total of 936 larvae was found among the dissected mosquitoes. The majority of larvae (889/936) was found in the Malpighian tubules (Fig. a) and half of the infective mosquitoes (4/8) still harbored either L2 or L3 larvae in the tubules. Twenty eight infective larvae were found in the head and proboscis of mosquitoes: eighteen were found in Ae. scapularis (Fig. b) in January 1996; four in April and three in July 1995 in Ae. taeniorhynchus (Fig. c); and two in August and one in November 1995 in Cx. quinquefasciatus. Melanization of larvae was observed only in the two Aedes species. In Ae. taeniorhynchus and Ae. scapularis, 9.5% and 16.7% of the harbored larvae were melanized, respectively (Table I). All Ae. taeniorhynchus with some melanized larvae had at least 45 larvae, except for one with 55 larvae in the cells of the Malpighian tubules, none of which were melanized. Among Ae. scapularis, melanization was observed even when the mosquito had as few as eight larvae. However, melanization of larvae was not observed in six Ae. scapularis harboring 9-20 larvae.
Infected mosquitoes were found in every month of the year, but February, while infective mosquitoes were found sparsely throughout the year (Table II). DISCUSSION To vector D. immitis, mosquitoes must live long enough to allow complete filarial development. Multivoltine species would probably make better vectors than univoltine species, and a specie's flight range and host seeking preference can influence the importance of a species as a vector (Ludlam et al. 1970, Otto & Jachowski 1980). In the present survey, Ae. taeniorhynchus, Ae. scapularis and Cx. quinquefasciatus were found to fulfill these prerequisites and to be natural vectors of D. immitis in Itacoatiara. Ae. taeniorhynchus and Cx. quinquefasciatus have been known as natural vectors of D. immitis in other areas (Villavaso & Steelman 1970, Sauerman & Nayar 1983, Russell 1985, Parker 1986, 1993). In Rio de Janeiro, both Ae. taeniorhynchus and Ae. scapularis have already been found naturally infected with presumed D. immitis larvae and considered to be potential vectors of the parasite (Lourenco-de-Oliveira & Deane 1995). In the surveyed area the annual transmission potential (ATP) for Ae. scapularis was approximately three times that of Ae. taeniorhynchus and six times more than that of Cx. quin-quefasciatus. The differences in ATP values is related to variations in mosquito population density, biting frequency and distribution throughout the year (Labarthe et al. 1998), showing that Ae. scapularis is the most important vector in Itacoatiara, followed closely by Ae. taeniorhynchus. Cx. quinquefasciatus is a secondary vector. In localities in the State of Rio de Janeiro (FEEMA 1983) where the hemisynanthropic primary vectors are abundant, canine heartworm frequency is high while where the endophilic secondary vector predominates, canine heartworm frequency is low (Labarthe et al. 1992, Souza 1992). Cx. declarator, Cx. saltanensis and Wy. bourrouli did not harbor infective larvae and have never been described as potential vectors of D. immitis. Therefore, their infections are thought to be a dead end and if so, have no epidemiological importance in heartworm transmission in the surveyed area. Melanization of larvae seems to be an important survival reaction of Aedes mosquitoes to D. immitis infection (Lindemann 1977, Christensen 1977, 1978, 1981, Christensen et al. 1989). Melanized larvae were seen in all heavily infected Ae. taeniorhynchus mosquitoes but one, which had all 55 larvae in the cells of the Malpighian tubules, suggesting that the infection was too recent for melanization to have taken place. Ae. scapularis mosquitoes melanized a larger proportion of larvae than Ae. taeniorhynchus, and melanization did not seem to be related to the number of larvae. Since Cx. quinquefasciatus mosquitoes were never observed with melanized larvae but always had smaller numbers of larvae than Ae. scapularis and Ae. taeniorhynchus, it seems that they control larval burden by other mechanisms, such as by reducing the number of larvae by the cibarial armature (Coluzzi & Trabucchi 1968). Transmission at Itacoatiara can potentially occur throughout the year, since infected and infective mosquitoes were found, respectively, in 11 and 5 months of the year. Environmental temperature directly influences the rate of larval development in the mosquito (Kutz & Dobson 1974, Christensen & Hollander 1978) so, transmission potential can change seasonally with temperature. Different species of mosquitoes were infective during different months: Ae. taeniorhynchus in autumn and early winter, Cx. quinquefasciatus in winter and spring and Ae. scapularis in the summer. These data strongly suggest that in the surveyed area, monthly heartworm chemoprophylaxis should be given to dogs year-round, in contrast with north temperate latitudes, where chemoprophylaxis is recommended only during some months, depending mainly on the recorded temperatures of each region (Knight & Lok 1995, Slocombe et al. 1995). Cats, in spite of being susceptible to D. immitis (McCall et al. 1992), are rarely found naturally infected. For instance, in urbanizing sections of Rio de Janeiro, where the canine heartworm prevalence is high (30.9%) (Labarthe et al. 1997a), feline heartworm prevalence is 1.6% (Labarthe et al. 1997b). This expressive difference in prevalence may be explained by the fact that the primary vectors of the parasite, Ae. scapularis and Ae. taeniorhynchus, are hemisynanthropic mosquito species which seek cats almost accidentally (Labarthe et al. 1998). Cx. quinquefasciatus, herein considered a secondary vector, is the only vector species commonly associated with cats (Genchi et al. 1992, Labarthe et al. 1998). That is, the potential mosquito vector for cats is a modest D. immitis vector in Rio de Janeiro. Furthermore, both cats and Cx. quinquefasciatus show marked nocturnal behavior: free cats are active at the time when the vectors are seeking blood meals, making it more difficult for mosquitoes to feed on them. Although somehow protected from mosquitoes, 33 to 36% of heartworm positive cats are indoors (Atkins 1997). Cx. quinquefasciatus is an endophilic species (Deane 1951, Rachou 1956), therefore, in places where their density is high, once an infective mosquito comes in the house, it can feed on cat and infect it. Since D. immitis is infective to man (OMS 1979), is frequently diagnosed among Brazilians (Campos et al. 1997), the three natural vectors (Ae. scapularis, Ae. taeniorhynchus and Cx. quin-quefasciatus) have been shown to seek humans in the studied heartworm focus (Labarthe et al. 1998), human cases of dirofilariasis may be expected in the study area. Also, since both the canine and culicidae fauna are similar along the lowland areas in the State of Rio de Janeiro (Labarthe et al. 1997a, Lourenco-de-Oliveira 1985a,b), health professionals should more seriously consider dirofilariasis among the many possible causes of solitary lesions of the human lung (Levinson et al. 1979, Campos et al. 1997). Acknowledgments To Carlos Alberto Coimbra and Antonio Bernardo da Costa for discussing data. To LP Lounibos for the critical reading of the manuscript. To Genilton Vieira and Heloisa MN Dinis for assistence with images. REFERENCES
Copyright 1998 Fundacao Oswaldo Cruz - Fiocruz The following images related to this document are available:Photo images[oc98083c.jpg] [oc98083a.jpg] [oc98083b.jpg] |
|