Prospects for suppressing dengue transmission by means of biological agents.


A biological agent that would most effectively reduce risk of dengue transmission would act against the adult stage of the vector mosquito. Beyond anecdotal references to swallows, dragonflies and ants, however, no biological agent has yet been developed specifically for use against a domestic vector of an arbovirus.


many locations, Pantropical

Problem Overview:

Human populations subject seek sustainable and environmentally friendly relief from dengue outbreaks.



Adulticidal interventions designed to reduce the force of transmission of dengue virus frequently fail because the adult stage of Aedes aegypti, the main vector mosquito, rests sequestered in scattered sites sheltered Samong residential structures. The disused automobile tires and assorted trash that constitute the main peridomestic breeding sites of these mosquitoes are difficult to identify, and the containers of potable water that are their main domestic sites must not be contaminated. For these reasons, biological agents have attracted attention in the ongoing efforts to protect human populations against dengue transmission. Needed, are efficacious agents that can readily be introduced into inaccessible sites as well as others that are neither harmful nor repugnant. The cost of the application must permit its long-term use. It is significant that Service (1983) suggested that biological agents hold little promise as the basis for interventions against dengue transmission and Gubler (1990) mentioned no biological agent in the anti-dengue prospectus that he formulated for the American Society of Tropical Medicine and Hygiene for the 1990's.

Agents that can be included under the rubric of "biological control" include a broad array of organisms, extracts and other biological products, many of which are self-disseminating or environmentally nonintrusive. The objective of this discussion is to categorize those that have been developed for use against A. aegypti and to evaluate their ability to penetrate into inaccessible sites as well as their acceptability in potable water. Other characteristics, such as persistence and cost, will also be discussed.


A biological agent that would most effectively reduce risk of dengue transmission would act against the adult stage of the vector mosquito. Beyond anecdotal references to swallows, dragonflies and ants, however, no biological agent has yet been developed specifically for use against a domestic vector of an arbovirus. On the contrary, the adulticidal chemicals that have been used in houses destroy such "nontarget" predatory arthropods as spiders and ants; sprayed houses tend to lose their invertebrate fauna. Thereby, residual spraying might ultimately increase the force of transmission. Needed are agents that act directly on longevity, narrowness of host-range, feeding frequency, abundance or vector competence of adult A. aegypti, the major variables that affect vectorial capacity.


Aedes aegypti (female), primary vector of dengue-hemorrhagic fever

Zooprophylactic agents constitute an exception to this judgment. Indeed, Hess and Hays (1970), in their classical formulation of the concept of "forage ratio" observed that A. albopictus mosquitoes in Hawaii frequently feed on cats and remarked that this effect would inhibit dengue transmission. This early suggestion that biological diversity might protect people against dengue infection remains to be tested.


Aside from destroying the larval stages of A. aegypti, the ultimate objective of a public health program based on antilarval biological agents is to reduce the abundance of the adult stage of this vector mosquito. An increment of reduction in the abundance of larval mosquitoes may not be reflected in a similar reduction in the abundance of adults (Agudelo-Silva and Spielman 1984; Washburn et al. 1990). A food-limited infestation of larvae produces more adults and more vital adults than would otherwise develop. In this manner, destruction of a portion of the larva in a site may increase risk of human disease. This inverse relationship between the density of larvae in a small container and the resulting production of adults imposes a requirement for careful evaluation on any "biological control" effort. Apparent virtue may induce ultimate harm.


Chinese health authorities have used various kinds of fish to exclude A. aegypti mosquitoes from breeding in large cisterns or other containers of potable water (Lu 1989). "..Small fishes, such as Claris fuscus, Tilapia nilotica, and Macropodus spp., have...been used in many regions to eliminate the larvae in the domestic water containers with considerable success." The use of catfish appears to be particularly effective (Neng et al. 1987). Such fish may have potential elsewhere in the world where the introduction can be maintained and where the human population does not object to the presence of such obvious companion organisms.


Omnivorous tadpoles constitute another potentially beneficial vertebrate predator of peridomestic mosquitoes (Spielman and Sullivan 1974). These larvae of the giant Cuban tree frog (Hyla septentrionalis) voraciously destroy aquatic insects and browse on algae growing on the margins of water containers. They rarely are found in containers that hold less than several liters of water and seem to have more impact on Culex breeding than on that of small-container Aedes mosquitoes. Due to their presence, however, few larval A. aegypti are found in cisterns or water-barrels on Grand Bahama Island. These large and noisy frogs became evident in the Miami area during the 1980's where they exist to this day as an example of a particularly notorious introduced organism (W. F. Loftus in Ewell 1984). These accidentally introduced frogs may somewhat impede the force of transmission of any dengue introduction that may occur.


Although various larval mosquitoes are adapted as predators of other mosquitoes, including certain Psorophora and Culex (Fuscus) species, the toxorhynchitines have attracted the most attention as biological control agents. Toxorhynchites mosquitoes enjoy 2 advantages for this purpose: they breed in the same kinds of containers that are exploited by A. aegypti, and they are nonhematophagous.

A mathematical model representing the dynamics of the A. aegypti - Toxorhynchites interaction suggested that 1 predatory larva in 80% of the containers in a site would reduce the density of adult A. aegypti by 80% in 20 days (Focks et al. 1980). In a subsequent small-scale field trial, adult emergence was reduced by 74% in containers receiving 1 or 2 first-instar predatory larvae every 10 days for 50 days. (Focks et al. 1982). Larval Toxorhynchites were placed, by hand. This confirmed an earlier inundative release on St. Maarten in which placement of 1 laboratory-reared larva in each water container temporarily reduced the abundance of larval A. aegypti (Gerberg and Visser 1978). The effect on adult mosquitoes was not recorded. The technique began to lose luster when it was recognized that the predatory efficiency of Toxorhynchites rutilus rutilus inversely correlates with prey density. An interference effect became evident when the prey was present in excess (Hubbard et al. 1988). Although numerous larval T. rutilus were present in tires observed near New Orleans, larval A. albopictus remained about as numerous there as in other tires (Marten 1990a). An evaluation based on larval breeding density may fail to reflect public health realities.


Discarded tires, a common urban habitat of Aedes aegypti and Aedes albopictus

A series of Toxorhynchites releases, performed in Indonesia, were evaluated by recording the density of adult mosquitoes. Here, it was learned that sustained release of first instar predatory larvae into virtually all household water storage basins failed to reduce the abundance of adult A. aegypti. This lack of effect may have been due to an inability of first instar larvae to survive in the absence of prey (Annis et al. 1989). Another realistic trial produced similar results (Annis et al. 1990). Although all 380 water-storage containers in a Javan village were seeded every other week with 5-10 second instar predatory larvae, no effect was noted on the indoor biting abundance of adult Aedes mosquitoes. A. albopictus was present in the area, and may have confounded these results because it breeds mainly in bamboo nodes, where no predators were distributed. This mosquito feeds out-of-doors, however, rather than in the indoor sites in which A. aegypti feeds.

Other predatory insects have been used to reduce the abundance of A. aegypti mosquitoes. Dragonfly larvae, in particular, may be useful for this purpose (Sebastian et al. 1990). In this Myanmar (=Burma) experiment, numerous larval Crocothemis servila, a pond-dwelling libellulid, were reared from field derived adults. Larval mosquitoes were provided as food to maintain these subadult dragonflies prior to their release into containers of potable water in the treated community. Target larvae virtually disappeared immediately after 2 dragonflies were placed in each container, and the density of target adults declined about 6 weeks later. Although no explanation for this delay in the suppression of adults is evident, the experiment is considered as a success. These observations suggest that a predator that is satisfactory for use in tanks of potable water must be large enough to consume many larvae in a short time, cling to the sides of the container so that it will not be lost when water is removed, develop slowly enough that it need not be replaced frequently and be able to survive for months when no larval mosquitoes are available as food. This model dragonfly release effort may be too expensive to serve as a basis for operational anti-dengue programs, particularly where our "throw-away society" provides alternative breeding sites that may be too small, numerous and difficult to identify.


Certain mosquitoes, most notable A. albopictus, appear to establish parapatric relationships with A. aegypti. (Gilotra et al. 1967). Indeed, the former species has displaced the latter from large portions of the southern United States, and only within a few years (Rai 1991). A. albopictus is a highly competent vector of dengue as well as a variety of other arboviruses and is a notorious pest species. Funeral vases serve as a peculiarly effective vehicle for transporting A. albopictus into new sites (O'Meara et al. 1992a).


Frontal poses of male and female each of Aedes aegypti (left) and Aedes albopictus (right)

In Asia, A. albopictus and A. aegypti rarely are found in the same breeding sites. On Taiwan, for example, A. albopictus is found in the dengue-free north and A. aegypti in the south, where dengue is endemic (JC Lien, personal communication). The zone of overlap there is exceptionally narrow. In India, A. aegypti seems to displace A. albopictus in urban sites, while the reverse occurs in rural sites (Gilotra et al, 1967). These populations seem to coexist "in a state of equilibrium" in gardens.

A. aegypti is similarly non-sympatric with another mosquito that breeds in small containers, A. bahamensis (Spielman and Feinsod 1978). These mosquitoes infest different towns on Grand Bahama Island. Recently, A. bahamensis invaded Florida, where a similar parapatric distribution came into being (O'Meara et al. 1989). Although some element of interspecies displacement appears to be in effect, no change in the distribution of these populations has occurred within the past few years.

Has the invasion of these exotic aedine mosquitoes partially freed Florida and certain other portions of the southern United States from the threat of epidemic dengue? A. bahamensis is noncompetent as a vector of a variety of arboviruses (Llewellyn et al. 1970), and it is autogenous (Spielman and Weyer, 1965). Indeed, although these mosquitoes are capable of feeding on vertebrate hosts, they rarely do so and appear never to do so before laying their first clutch of eggs (O'Meara et al. 1992b). A. albopictus, on the other hand is a highly competent host for dengue as well as LaCrosse virus (Rai 1991) and is notorious as a pest. Its indiscriminate pattern of host-seeking, however, probably negates its capacity as a vector of any infection, whether anthroponotic or zoonotic, in all but the most host-restricted site.


Aedes albopictus (female), the Asian Tiger Mosquito, a recent invader mosquito into North and South America, Africa and Europe; a rural, secondary vector of dengue.

In order for one mosquito to displace another, they must interact in some substantive manner. If that interaction were to take the form of competition for trophic resources, interpopulational competition must be keener than intrapopulational competition, an unlikely event. Such a mode of interaction seems unlikely because genetically related organisms are more likely to share essential resources than are more distantly related organisms. Indeed, crowding among small-container Aedes does appear to be more intense within species than between species (Walker et al. 1987; Ho et al. 1998). A. albopictus seems to compete tropically with A. triseriatus, perhaps more effectively in nutrient-poor disused automobile tires than in the more permissive tree hole environment (Livdahl and Willey 1991). Convincing evidence of interpopulational trophic competition, however, does not exist.

Sexual interaction may provide another mode of interspecies displacement. In the event that the males of one kind of mosquito were to mate with, but not effectively inseminate, the females of another, the fertility of those females might be impaired (Spielman and Kitzmiller 1967). The name "Satyr effect" has been applied to this hypothetical relationship (Ribeiro and Spielman 1986. Perhaps sexual profligacy helps populations to perpetuate.

Parasitism represents yet another forum for interpopulational displacement. In spite of recent arguments to the contrary (Read and Schrag 1991), pathogens generally are less virulent when well adapted to their hosts than when the host-parasite relationship is transient. An amazingly clear demonstration of this principal lies in the interrelationship between gregarines and their mosquito hosts, to be discussed below. By differentially destroying alien mosquitoes these parasites create a potentially asymmetrical relationship between different kinds of mosquitoes (Beier and Craig 1985). Such a pest as A. albopictus, for example, or more innocuous an insect as A. bahamensis, may thereby displace A. aegypti.


Fashion dictates that any discussion of "vector control" should refer to the potential of molecular genetics for use against mosquito-borne infection (Curtis 1990). The development of "designer gene" release strains by modifying mosquitoes is frequently cited optimistically as providing the best hope for intervening against malaria and dengue. The MacArthur Foundation, for example, has established a major interinstitutional consortium that was originally designed to create such release strains. Of course, a long history of "genetic control" projects precedes the current program; these efforts of the 1960's and 70's failed largely because of the difficulty of reintroducing a laboratory-adapted strain of organisms into the field. These methods included sterile male release, translocation heterozygote release, search for meiotic drive mechanisms, use of hybrid male sterility and cytoplasmic incompatibility. Most were imaginative and logical, and many were evaluated in the field.

The present efforts to develop release strains of vector-incompetent mosquitoes are based on the techniques of molecular genetics, a field component is rarely included. The prevailing objective is to modify a laboratory strain of mosquitoes in order to reduce its vector competence and eventually to release these mosquitoes in nature such that the wild-type vector-competent strain will eventually be replaced. Three major factors generally fail to be addressed: 1) No mechanism of displacement is identified; 2) Vector competence is a weak component in the force of transmission; and 3) A man-biting pest insect cannot be nurtured (at least "not in my back-yard"). Proponents generally omit mention of the history of failure and fail to demonstrate the logic of their proposed release. Prospects for this technology seem dim.


A "third generation of insecticides" was heralded in a classical program of work originally housed in the late Carroll Williams laboratory at Harvard University. Williams anticipated that enormously active hormonomimetic compounds could be designed that would block development of particular kinds of insects (Williams 1967). The insecticide would match the target. The first such hormonomimetic insecticide was developed in the mid 1960's (Spielman and WIlliams 1966). This derivative of farnesoic acid prevents metamorphosis; treated larvae produce imagos that become locked as pharate adults. A simulated field study demonstrated efficacy (Spielman 1970). The derivative of farnesoic acid used in these experiments was later modified to produce a more active and stable compound that is commercially available under the name methoprene or altocid (Staal 1975). Methoprene now promises to be the most environmentally acceptable larvicidal chemical usable against mosquitoes. Its apparent absence of mammalian toxicity provides promise for use in potable water.


The ability of certain cyclopoid copepods to destroy larval mosquitoes was noted in 1938 (Hurlbut). These ubiquitous "water fleas" were seen preying on newly hatched larvae. If such mosquitoes had developed to the third instar, however, they would have been too large to be attacked by these minute predators. Field experiments in Rongaroa (French Polynesia) later demonstrated that Mesocyclops can be used in larvicidal interventions against A. aegypti and polynesiensis but not Culex (Riviere et al. 1987). These crab-hole applications reduced the abundance of adult A. polynesiensis by about 76%. Indeed, in Colombia the abundance of copepods in natural settings inversely correlates with the abundance of larval anopheline mosquitoes (Marten et al. 1989). Cyclopoid copepods appear to limit the abundance of certain kinds of mosquitoes.

Not all copepods destroy all mosquitoes. Predation efficacy of 7 different cyclopoids was tested in various kinds of containers (Marten 1990). Macrocyclops albidus was most voracious in a New Orleans trial, killing virtually all freshly hatched larval A. albopictus under all experimental conditions. No Culex, Orthopodomyia or Toxorhynchites, however, were killed. In order to select a suitable match of predator and prey, a routine was developed for comparing the efficacy of cyclopoids as predators of mosquitoes, based on predation efficacy and rate of increase at 30?C, the temperature of natural breeding sites (Brown et al, 1991). Conditions have been standardized for the production of large masses of predatory copepods based on the use of protozoal infusions as food.

Field trials have been conducted to determine whether copepods can usefully destroy larval Stegomyia mosquitoes. Copepods (Macrocyclops albidus) were released into each of about 200 tires arranged in 2 linear stacks of about 100 discarded tires each located near New Orleans (Marten 1990a). A third stack remained untreated. Larval A. albopictus that were numerous in the treated tires at the beginning of the experiment virtually disappeared within 2 months. Adults disappeared about 1 month later and remained scarce for at least another year. These predators, however, did not reduce the abundance of Culex salinarius.

A series of other field trials are currently being conducted in (1) Fortaleza, Brazil using Mesocyclops longisetus against A. aegypti (Vasconselos, Sleigh and Kay, personal communication); (2) Naning, China using Mesocyclops quangxiensis against A. albopictus (Wu and Kay, personal communication), (3) Darnley Island in Quensland, Australia using Mesocyclops MB3 sp. against A. aegypti and A. scutellaris (Kay, personal communication); (4) Brisbane, Australia using Mesocyclops asper against A. aegypti (Kay, personal communication); (5) Stradbroke Island in Quensland, Australia using Mesocyclops asper against A. aegypti (Kay, personal communication); (6) Progresso, Honduras using one of 4 Mesocyclops sp against A. aegypti; and (7) New Orleans, Louisianna using Macrocyclops albidus against A. albopictus (Marten, personal communication). A recent evaluation in Tahiti of the efficacy of Mesocyclops aspercornis against A. aegypti has been completed (Lardeux et al. 1992). Preliminary results from each of these evaluations have been encouraging. Crucial elements in these evaluations will be observations on the biting-density of adult mosquitoes of the target species.

A recent small-scale field trial in Anguilla produced results that were somewhat less encouraging (M.F. Suarez, personal communication). "Mesocyclops aspericornis and Mesocyclops albidus did not completely eliminate larval A. aegypti seven weeks after application. The major difficulty was that the community cleaned and refilled their water storage containers..."

Cyclopoids now appear to offer high promise as biological control agents for A. aegypti mosquitoes. In practice, a local cyclopoid isolate is chosen, evaluated against the target species and established in a fermentation-like production system. No international quarantine procedure is required. Suspensions of these organisms are then applied as appropriate by means of a coarse-nozzled sprayer. The technology is "appropriate" and costs appear to be modest. Little environmental or toxicological risk is evident.

Certain drawbacks of a copepod release program have been identified: 1) Those cyclopoids that have until now been used for this purpose do not resist desiccation. They must frequently be reapplied when the climate is arid. Various Macrocyclops and Acanthocyclops species, however, can resist extended periods of drying. 2) Because they kill only newly hatched larvae, BTI (to be discussed below) must be included in the formulation. The main effect of a cyclopoid application effort is delayed, occurring only after some subsequent rainfall when a new lot of eggs hatch. In the event that the breeding site becomes too dry for the copepods to survive, the treatment will fail. The recent Rongoroa Island evaluation encountered just such a difficulty (Lardeux et al. 1992). Although highly promising, the practicality of these biological agents in anti-dengue interventions remains to be established.


The gregarine parasites of mosquitoes provide fascinating prospects for the destruction of small-container breeding mosquitoes. The nicely euphonious name Ascogregarina now replaces the problematic former designation Lankesteria (Ward et al. 1982). Once ingested by larval mosquitoes, the oocysts of these parasites release sporozoites that invade the gut wall (Beier and Craig 1985). There the trophozoites develop into gamonts by the time of pupation and migrate to the lumen of the malphigian tubules. After sexual union, they form oocysts that mature at the time of ecdysis and are ultimately released in the excreta. The burden of parasitism in the normal host seems light (Walker et al. 1987). In other kinds of mosquitoes, however, hosts tend to die. As. culicis normally parasitizes A. aegypti, and As. taiwanensis parasitizes A. albopictus (Munsterman and Wesson 1990). The usefulness of these entomopathogens as biological control agents of mosquitoes lies in their possible role in the displacement of one kind of mosquito by another. Gregarines may represent the long-sought forum of interaction.


Until recently, the microsporidian pathogens of mosquitoes appeared to hold little promise for use against A. aegypti mosquitoes because they appeared to be poorly adapted to the small-container habitat. Edhazardia aedis, however, was discovered in Puerto Rican Aedes aegypti and reisolated from similar mosquitoes collected in Thailand (Sweeney and Becnel, 1991). Although other microsporidians require copepod intermediated hosts for the meiospore stage, this organism develops without the need for such an alternation of hosts. Larval mosquitoes die if they become heavily infected by uninucleate spores and release another generation of uninucleate spores. Those that sustain a light infection complete development and produce adults that are infected by binucleate spores that are transmitted to virtually all progeny. For a biocontrol effort against A. aegypti microsporidian-infected eggs might be seeded into natural breeding sites. Although, many larvae of the target species would escape the lethal effects of this agent, the infection should burden those mosquitoes that escape.

Amblyospora connecticus naturally infects Aedes cantator but can infect certain closely related mosquitoes in the haploid segment of its cycle. Copepods serve as alternate hosts in the haploid cycle. The diploid cycle can develop solely in the natural mosquito host (Andreadis 1989). Although few males survive infection, infected imaginal female mosquitoes appear to suffer little burden and their progeny serve as the vehicle for introducing the parasite into new sites. Perhaps such microsporidian meiospores might be inoculated into potential breeding sites in order to produce a high level of infection in native copepods (Sweeney et al. 1988). These pathogens, however, appear to have little potential in a vector reduction effort because they cannot yet be grown in "fermentation tanks" and female adults develop normally.

The possibility remains that the synergistic effects of parasitism might offer some advantage in a biological intervention against vector mosquitoes. A ciliate pathogen of mosquitoes, for example, potentiates the lethal effects of an opportunistic fungus (Washburn et al. 1988). Such complex interrelationships illustrate the difficulties as well as opportunities that await the alert research worker. BACTERIA

BTI is a proven, environmentally safe mosquito larvicide that is nontoxic for people. Bacillus thuringiensis israelensis (BTI) was discovered only in the 1970's (Goldberg and Margalit 1977). It has become commercially available under such names as Teknar, Vectobac and Bactimos. The beauty of this material is that an application destroys larval mosquitoes but spares any predators that may be present. The toxin, however, is destroyed by sunlight. This imposes a necessity for frequent application. Weekly visits to each breeding site may be required, thereby making it more expensive.


Of the numerous fungi that have been described, only Lagenidium giganteum seems to have the ability to possibly stop the spread of disease transmission through mosquitoes. Indeed, the federal Environmental Protection Agency registered several such formulations for use against larval mosquitoes on 16 August 1991. Oospores that can resist droughts can readily be produced in bulk (Kerwin et al. 1986). Oospores survive for many years in soil, but reactivate only about a month after flooding. In practice, the spores are activated by staying in water for 1-2 weeks before being sprayed onto the surface of the site to be treated. About half of the target mosquito population becomes infected by the oospores and larvae continue to die for the next several weeks. The infection may continue for a period of time. This material holds promise for use against dengue vectors breeding in potable (drinking) water; it should soon become commercially available.

Although these preparations seem to persist for a long time in water, there are certain additional problems. It can be difficult and costly to apply. The fungus infects and kills only a part of target larval mosquitoes, as well. The level may not be large enough to decrease risk of human dengue infection.

Various other species have been considered for use against mosquitoes, but there have been problems with these also (Whistler et al. 1974, Chapman 1985, Lacey and Undeen 1986).

Other fungi, such as Leptolegnia chapmani (Zattau and McInnis 1987) and Toplocladium cylindrosporum (Ravellec and Riba 1989), present some potential for anti-dengue programs.


The uses of various plant extracts to decrease the transmission of dengue are very traditional (Williams 1967). Chrysanthemum flowers, the neem tree, marigolds, Swartzia madagascariensis -- a variety of an African plant, and the Ethiopian soap-berry vine all create elements that can be used as insecticides (Spielman and Lemma 1973, Minjas and Sarda 1986, Green et al. 1991). Local traditions provide guides to the identification of such plants. Thus, many plants contain things that destroy mosquitoes; they can be useful in village-level self-help anti-dengue programs where there are a large number of people willing to help and where the plants are cheap.


Certain algae seem to fill the guts of larval mosquitoes with a mass of material that they are not able to digest, and this is said to hurt growth (Marten 1986: 1987). The promise of such material against the mosquitoes carrying the dengue virus, however, seems limited because these mosquitoes usually breed in sites that are too dark for algae.


Louis Roth's classical 1948 study on the role of sound in A. aegypti looking for mates created a highly productive series of research efforts that focused on the possibility of gathering mosquitoes by attracting them to a source of odor or sound. These efforts continue to this day


A broad range of predators and parasites provide potential for developing anti-dengue agents. Different types of worms and spiroplasms could decrease the ability of mosquitoes to breed and carry dengue, as well (Case and Washino 1979, Kerwin and Washino 1985).


This case study is another in a long line of reviews of the potential of biological agents for decreasing transmission of one or another vector-borne agent, and, like these others, concludes with a set of recommendations.

The major requirement of a program that will help to stop transmission of dengue is the ability to get into obscure bodies of water scattered within and around groups of humans where A. aegypti mainly breeds. An untended guard-dog or a hidden automobile tire presents an obstacle to any effort in a society that respects privacy. Only biological agents carry the potential for overcoming this difficulty, and the most likely agents are those represented by closely related organisms. Toward this end, we require a program of biological research aiming toward an understanding of the factors that limit the number of mosquitoes.

A second requirement for an anti-dengue program is the potential for use in potable (drinking) water. Household sites for collecting and storing potable water generally can be visited in a way that will permit an effort to introduce biological agents. Biological agents can avoid the requirements of toxicity and absence of odor that such an application requires. Toward this end, we require a program of biological research aiming to develop methods for formulating BTI such that the material will not rapidly sediment.

The most vulnerable point in the dengue transmission cycle lies in the length of its incubation period. Toward this end, we require an understanding of the factors that influence the length of life of these vectors.

Nearly as sensitive in the continuation of dengue transmission is the requirement for a vector-host relationship (a relationship between a mosquito and the host where it feeds) that focuses on mammals that have the ability to maintain the basic reproductive rate of the thing that causes the disease. Toward this end, we require an understanding of the host-seeking behavior of these mosquitoes.

The requirement for a long-term, sustainable effort makes cost an important factor in any public health effort. For this reason, intervention agents cannot generally be produced in living biological systems. Production should be possible in certain tanks, and a minimum number of steps should be required for preparation of the final formulation. Extended shelf-life is crucial, as is the ability to kill all mosquitoes that have been treated. Pesticide development efforts should aim toward these objectives.

Most critical in an approach toward suppressing or containing any vector-borne infection is the ability to select research directions on the basis of practical matters. Any anti-dengue campaign that is designed should pursue attainable objectives that are worthwhile and sustainable. Failure to satisfy any of these criteria may result in the worst public health harm, an ultimate increase in the disease.




Agudelo-Silva, F and A Spielman. 1984. Paradoxical effects of simulated larviciding on production of adult mosquitoes. Am. J. Trop. Med. Hyg. 33:1267-1269.

Hess AD and RO Hays. 1970. Relative potentials of domestic animals for zooprophylaxis against mosquito vectors of encephalitis. Am. J. Trop. Med. Hyg. 19:327-334.

Gubler, DJ. 1990. Aedes aegypti and Aedes aegypti-borne disease control in the 1990'S. Am. J. Trop. Med. Hyg. 40:571-578.

Lacey, LA and CM Lacey. 1990. The medical importance of riceland mosquitoes and their control using alternatives to chemical insecticides. J. AM. Mosq. Cont. Assoc. 6(Suppl. 2):1-93.

Rawlins, SC. 1989. Biological control of insect pests affecting man and animals in the tropics. CRC Critical Revs Microbiol. 16:235-252.

Nasci, RS. 1988. Biology of Aedes triseriatus (Diptera: Culicidae) developing in tires in Louisiana. J. Med. Ent. 25:402-5.

Service, MW. 1983. Biological control of mosquitos - has it a future? Mosq. News 43:113-

Washburn, JO, DR Mercer and JR Anderson. 1991. Regulatory role of parasites: impact on host population shifts with resource availability. Science 253:185-188.

Williams, CM. 1967. Third-generation pesticides. Sci. Amer. 217:13-17.


Lu BL. ca 1989. Recent studies on the dengue vectors of China. Working paper, pp 89-92.

Neng, W, W Shusen, H. Guangxin, X Rongman, T Guangkun, and Q Chen. 1987. Control of Aedes aegypti larvae in household water containers by Chinese cat fish. Bull. Wld Hlth Org. 65:503-506.


Spielman, A and JJ Sullivan. 1974. Predation on peridomestic mosquitoes by hylid tadpoles on Grand Bahama Island. Am. J. Trop. Med. Hyg. 23:704-709.


Gilotra, SK, LE Rozeboom and NC Bhattacharya. 1967. Observations on possible competitive displacement between populations of Aedes aegypti Linnaeus and Aedes albopictus Skuse in Calcutta. Bull. Wld. Hlth. Org. 37:437-446.

Ho, BC, A Ewert and LM Chew. 1989. Interspecific competition among Aedes aegypti, Ae. albopictus, and Ae. triseriatus (Diptera: Culicidae): larval development in mixed cultures. J. Med. Entomol. 26:615-23.

Lardeux, F, F Riviere, Y Sechan and B Kay. 1992. Field Release of Mesocyclops aspercornis (Copepoda: Cyclopidae) for control of Aedes polynesiensis on an atol in French Polynesia. J. Med. Entmol. (in press).

Livdahl, TP and MS Willey. 1992. Prospects for an invasion: Competition between Aedes albopictus and native Aedes triseriatus. Science 253:189-191.

Llewwellyn, CH, A Spielman and TE Frothingham. 1970. Survival of arboviruses in Aedes albonotatus, a peridomestic Bahamian mosquito. Proc. Soc. Exp. Biol. Med. 133:551-559.

O'Meara, GF, VL Larson, DH Mook, MD Latham. 1989. Aedes bahamensis: its invasion of south Florida and association with Aedes aegypti. J. Am. Mosq. Control Assoc. 5:1-5.

O'Meara, GF, AD Gettmen, LF Evans Jr and FD Scheel. 1992a. Invasion of cemeteries in Florida by Aedes albopictus. J. Amer. Mosq. Cont. Assoc. 7:in press.

O'Meara, GF, VL Larsen and DH Mook. 1992b. Blood feeding and autogeny in the peridomestic mosquito, Aedes bahamensis (Diptera: Culicidae). J. Med. Entomol. 29:in press.

Rai, KS. 1991. Aedes albopictus in the Americas. Ann. Rev. Entomol. 36:459-84.

Ribeiro, JMC and A. Spielman. 1986. The Satyr effect: a model predicting parapatry and species extinction. Amer. Naturalist 128:513-528.

Spielman, A and JG Kitzmiller. 1967. The genetics of populations of medically important arthropods. in JR Wright and R. Pal (eds), Genetics of Insect Vectors of Disease. Elsevier, Amsterdam pp 459-485.

Spielman, A and FM Feinsod. 197?. Differential distribution of peridomestic Aedes mosquitoes on Grand Bahama Island. Trans. Roy. Soc. Trop. Med. Hyg. 73:381-384.

Walker, ED, RS Copeland, SL Paulson, LE Munsterman. 1987. Adult survivorship, population density, and body size in sympatric populations of Aedes triseriatus and Aedes hendersoni (Diptera: Culicidae). J. Med Ent. 24:485-93.


Annis, B, S Krisnowardojo, S Atmosoedjono and P Supardi. 1989. Suppression of larval Aedes aegypti populations in household water storage containers in Jakarta, Indonesia, through release of first instar Toxorhynchites splendens larvae. J. Am. Mosq. Control Assoc. 5:235-238.

Annis, B, S. Nalim, Hadisuwono, Widiarti and DT Boewono. 1990. Toxorhynchites amboinensis larvae released in domestic containers fail to control dengue vectors in domestic containers in a rural village in Central Java. J. Amer. Mosq. Control Assoc. 6:75-78.

Focks, DA, DA Dame, AL Cameron and MD Boston. 1980. Predator-prey interaction between insular populations of Toxorhynchites rutilus rutilus and Aedes aegypti. Environ. Entomol. 9:37-42.

Focks, DA, SR Sackett and DL Bailey. 1982. Field experiments on the control of Aedes aegypti and Culex quinquefasciatus by Toxorhynchites rutilus rutilus (Diptera: Culicidae). J. Med. Entomol. 19:336-339.

Gerberg, EJ and WM Visser. 1978. Preliminary field trial for the biological control of Aedes aegypti by means of Toxorhynchites brevipalpis, a predatory mosquito larva. Mosq. News 38:197-200.

Hubbard, SF, SLC O'Malley and R Russo. 1988. The functional response of Toxorhynchites rutilus rutilus to changes in the population density of its prey, Aedes aegypti. Med. Vet. Entomol. 2: 279-283.

Sebastian, A, MM Sein, MM Thu and PS Corbet. 1990. Bull. Ent. Res. 80:223-232.

Hormonomimetic compounds

Spielman, A, and CM Williams. 1966. Lethal effects of synthetic juvenile hormone on larvae of the yellow fever mosquito, Aedes aegypti. Science. 154:1043-1044.

Spielman, A. 1970. Synthetic juvenile hormones as larvicides for mosquitoes. Industry Trop. Hlth. 7:67-70.

Staal, GB. 1975. Insect growth regulators with juvenile hormone activity. Ann. Rev. Entomol. 16:417-460.


Brown, MD, BH Kay and JK Hendrix. 1991. Evaluation of Australian Mesocyclops (Cyclopoida: Cyclopidae) for mosquito control. J. Med. Entomol 28:618-623.

Urlbut, HS. 1938. Copepod observed preying on first instar larva of Anopheles quadrimaculatus Say. J. Parasitol. 24:281.

Marten, GG, R. Astaiza, MF Suarez, C Monje and JW Reid. 1989. Natural control of larval Anopheles albimanus (Diptera: Culicidae) by the predator Mesocyclpos (Copepoda: Cyclopoida). J. Med. Entomol. 26:624-627.

Marten, GG. 1990. Evaluation of cyclopoid copepods for Aedes albopictus control in tires. J. Amer. Mosq. Control Assoc. 6:681-688.

Marten, GG. 1990. Elimination of Aedes albopictus from tire piles by introducing Macrocyclops albidus (Copepoda, Cyclopidae). J. Amer. Control Assoc. 6:689-693.

Marten, GG. 1990. Elimination of Aedes albopictus from tire piles by introducing Macrocyclops albidus (Copepoda Cyclopidae). J. Amer. Mosq. Control Assoc. 6:689-93.

Marten, GG. 1990. Evaluation of cyclopoid copepods for Aedes albopictus control in tires. J. Amer. Mosq. Control Assoc. 6:681-87.

Marten, GG. 1984. Impact of the copepod Mesocyclops leuckarti pilosa and the green alga Kirchneriella irregularis upon larval Aedes albopictus (Diptera: Culicidae). 1984. Bull. Soc. Vector Ecol. 9:1-5.

Marten GG. 1990 Larvivorous Copepods. New Orleans Mosquito Control Board Monthly Report.

Marten, GG. 1991. Biological Control (Copepods). New Orleans Mosquito Control Board Monthly Report.

Marten GG. 1991. Pilot project on the use of Cyclops for Aedes aegypti control in Progress, Honduras: Summary report for the first year and work plan for the second year.

Riviere, F, BH Kay, JM Klein and Y Sechan. 1987. Mesocyclops aspericornis (Copepoda) and Bacillus thuringiensis var. israelensis for the biological control of Aedes and Culex vectors (Diptera: Culicidae) breeding in crab holes, tree holes and artificial containers. J. Med. Entomol. 24:425-430.


Perich, MJ, PM CLair and LR Boobar. 1990. Integrated use of planaria (Dugesia dorotocephala and Bacillus thuringiensis var. israelensis against Aedes taeniorhynchus: a laboratory bioassay. J. Am. Mosq. Control Assoc. 6:667-671.


Beier, JC and GB Craig Jr. 1985. Gregarine parasites on mosquitoes. Integr. Mosq. Cont. Meth. 2:167-184.

Munstermann, L. 1990. First record of Ascogregarina taiwanensis (Apicomplexa: Lecunidae) in North American Aedes albopictus. J. Am. Mosq. Cont. Assoc. 6:235-243.

Read, AF and SJ Schrag. 1991. The evolution of virulence: experimental evidence. Parasit. Today 7:296-297.

Washburn, JO, DE Egerter, JR Anderson and GA Saunders. 1991. Density reduction in larval mosquito (Diptera: Culicidae) populations by interactions between a parasitic ciliate (Ciliophora: Tetrahymenidae) and an opportunistic fungal (Ooemycetes: Pythiaceae) parasite. J. Med. Entomol. 25:307-314.

Walker, ED, SJ Poirier and WT Veldman. 1987. Effects of Ascogregarina barretti (Eugregarinida: Lecudinidae) infection on emergence success, development time, and size of Aedes triseriatus (Diptera: Culicidae) in microcosms and tires. J. Med. Entomol. 24:303-9.


Andreadis, TG. 1989. Host specificity of Amblyospora connecticus (Microsporida: Amblyosporidae), a polymorphic microsporidian parasite of Aedes cantator (Diptera: Culicidae). J. Med. Entomol. 26:140-145.

Chapman, HC. 1985. Ecology and use of Coelomomyces species in biological control, a review. in JN Couch and CE Bland (eds) The genus Coelomomyces. Academic Press, NY. pp 361-369.

Kerwin, JL, CA Simmons and RK Washino. 1986. Oosporogenesis by Lagenidium giganteum in liquid culture. J. Invert. Pathol. 47:258-270.

Kerwin, JL and RK Washino. 1988. Lagenidium giganteum (Oomycetes: Lagenidium) and description of a natural epizootic involving a new isolate of the fungus. J. Med. Ent. 25:453-460.

Ravellec, M, A Vey and G Riba. 1989. Infection of Aedes albopictus by Toplocladium cylindrosporum. J. Invert. Pathol. 53:7-11.

Sweeney, AW and JJ Becnel. 1991. Potential of microsporidia for the biological control of mosquitoes. Parasit. Today 7:217-220.

Sweeney, AW, SL Doggett and G. Gullick. 1988. Bioassay experiments on the dose response of Mesocyclops sp. copepods to meiospores of Amblyospora dyxenuides produced in Culex annulirostris mosquito larvae. J. Invert. Pathol. 53:118-120.

Whistler, HC, SL Zebold and JA Semanchuk. 1974. Alternate host for the mosquito parasite Coelomomyces 251:715-716.

Wilson, ML, F Agudelo-Silva and A Spielman. Increased abundance , size and longevity of food-deprived mosquito populations exposed to a fungal larvicide. Am. J. Trop. Med. Hyg. 43:551-536.

Zattau, WC and T McInnis, Jr. 1987. Life cycle and mode of infection of Leptolegnia chapmanii (Oomycetes) parasitizing Aedes aegypti. J. Invert. Pathol. 50:134-145.


Goldberg, LJ, and J. Margalit. 1977. A bacterial spore demonstrating rapid larvicidal activity against Anopheles sergentii, Uranotaenia unguiculata, Culex univittatus, Aedes aegypti and Culex pipiens. Mosq. News. 37:355-358.

Lacey, LA and AH Undeen. 1986. Microbial control of black flies and mosquitoes. Ann. Rev. Entomol. 31:265-96.

Ohana, B, J Margalit and Z Barak. 1987. Fate of Bacillus thuringiensis under simulated field conditions. Appl. Environ. Microbiol. 53:828-831.

Plant extracts

Green, MM, JM Singer, DJ Sutherland and CR Hibben. 1991. Larvicidal activity of Tagetes minuta (Marigold) toward Aedes aegypti. J. Amer. Mosq. Control Assoc. 7:182-286.

Spielman, A and A Lemma. 1973. Endod extract, a plant derived moluscicide: toxicity for mosquitoes. Am. J. Trop. Med. Hyg. 22:802-804.

Minjas, JN and RK Sarda. 1986. Laboratory observations on the toxicity of Swartzia madagascariensis (Leguminosae) to mosquito larvae. Trans. Roy. Soc. Trop. Med. Hyge. 80:460-461.


Marten, GG. 1987. The potential of mosquito-indigestible phytoplankton for mosquito control. Oper. and Sci Notes. Op. and Sci. Notes. 3:105-6

Marten, GG. 1986. Mosquito control by plankton management: the potential of indigestible green algae. J. Trop. Med. Hyg. 89:213-222.

Minjas, JN and RK Sarda. 1986. Laboratory observations on the toxicity of Swartzia madagascariensis (Leguminosae) extract to mosquito larvae. Trans. Roy. Soc. Trop. Med. Hyge.. 80:460-461.


Dawson, GE, BR Laurence, JA Pickett, MM PiLE and LJ Wadhams. 1989. A Note on the mosquito oviposition pheromone. Pestic. Sci. 37:000-000.

Otieno, WA, TO Onyango, MM Pile, BR Laurence, GW Dawson, LJ Wadhams and JA Pickett. 1988. A field trial of the synthetic oviposition pheromone with Culex quinquefasciatus Say (Diptera: Culicidae) in Kenya. Bull. Ent. Res. 78:463-470.


Case, TJ and RK Washino. 1979. Flatworm control of mosquito larvae in rice fields. Science 206:1412-1414.

Humphrey-Smith, I, O Grulet and C Chastel. 1991. Pathogenicity of Spiroplasma taiwanense for larval Aedes aegypti mosquitoes. Med Vet Entomol 5:229-232.

Kerwin, JL and RK Washino. 1985. Recycling of Romanomermis culicivorax (Mermithidae: nematoda) in rice fields in California, USA. J. Med. Entomol. 22:637-643.

a8.genetic control

Submitted by:

Professor Andrew Spielman*
Harvard School of Public Health
665 Huntington Ave.
Boston, MA 02115
(617) 432-2058

*Diseased December 2006

Edited by:

Elizabeth Wiley
E.S. Harkness Dorm
367 Cedar St.
New Haven, CT 06510

All photos courtesy of and copyright:

Leonard E. Munstermann, Ph.D.
Scientist (Epidemiology), Yale School of Medicine

Latest articles


Air Pollution



Endangered Species




Global Climate Change

Global Health


Natural Disaster Relief

News and Special Reports

Oceans, Coral Reefs



Public Health



Toxic Chemicals


Waste Management


Water and Sanitation

Yale Himalaya Initiative