One way to help stem the spread of antimicrobial resistance is to reduce the number of cases of disease that must be treated with therapeutic agents, thereby reducing the selective pressure that often leads to such resistance. Toward this goal, insecticides play a major role in controlling major disease vectors—the arthropods that pass pathogens to humans.
The use of insecticides, however, presents a problem parallel to the use of antimicrobials—the chemical agents themselves frequently promote resistance among the vectors they are intended to control. Indeed, resistance to insecticides has appeared in every major species of arthropod vectors— including mosquitoes, ticks, fleas, lice, and sand flies—and various vectors have developed resistance to every class of pesticide. The focus of this session of the workshop was an examination of the role that vectors play in a variety of diseases and how management efforts are being used to better control vector populations.
Malaria represents a classic example of the difficulties associated with vector-control efforts. In 1955, the World Health Organization (WHO) called for the global eradication of malaria through the use of DDT. However, the mosquitoes that carry the disease soon developed resistance to DDT, and in 1976 WHO shifted its goal from eradication to control. The resistance problems continued with the switch to newer insecticides, such as the organophosphates and pyrethroids. Many experts maintain that new insecticides, coupled with the development and implementation of various
other control measures, will be needed to regain long-term control over the mosquitoes that spread malaria. Some experts, however, argue that DDT has unfairly been declared an unacceptable hazard, and that it can play an important—and significantly larger—role in controlling not only malaria but also such other widespread diseases as dengue.
Research on resistance among vectors now takes two basic approaches. One approach is to study the molecular mechanisms involved. Scientists already have a relatively good understanding of the basic mechanisms underlying resistance to commonly used insecticides, and they are now using recently developed molecular techniques to begin to dissect these mechanisms at the DNA level. The next challenge will be to use this molecular understanding to develop novel vector-control methods that avoid or minimize resistance problems.
The second approach to research involves resistance management— that is, developing and implementing control methods that minimize the likelihood that vectors will evolve strong resistance to important insecticides. Two such promising control strategies are alternating the use of different insecticides and applying them in mosaic patterns over a given geographic area. Scientists are now conducting, in Mexico, a large-scale trial of these two strategies, compared to the use of DDT or a pyrethroid insecticide used in the conventional manner. Information from such trials may enable scientists and policy makers to establish rational strategies for long-term insecticide use.
Expanding the use of integrated pest management, or IPM, also holds promise for improving vector-control programs. Tackling a pest problem in numerous ways at once—including the use of pesticides in a timely manner at select locations—will likely yield more thorough and longer-lasting control than would result from any single method applied individually.
Developing and applying mathematical models can provide a solid basis for identifying new ways to effectively control the emergence of resistance among vector populations—and also to identify methods that are not effective. Among their benefits, models enable researchers to examine trends in data, explore questions of population dynamics, compare various management options, and generate new hypotheses for study. A major roadblock to progress in this area is the lack of interdisciplinary work among modelers studying insect resistance and those studying antimicrobial resistance in pathogens. Indeed, increased scientific collaboration across disciplines could foster scientific progress in numerous areas related to antimicrobial resistance. Likewise, policy makers developing programs to control the spread of antimicrobial resistance among pathogens could benefit by examining the successes and failures of current vector-control programs.
INSECTICIDE RESISTANCE IN INSECT VECTORS OF HUMAN DISEASE
Janet Hemingway, Ph.D.
Liverpool School of Tropical Medicine Liverpool, United Kingdom
Insecticides play a central role in controlling major vectors of diseases. In 1955 the WHO Assembly proposed the global eradication of malaria with DDT. However, the shift from malaria eradication to control in 1976 was prompted by the appearance of DDT resistance in many mosquito vectors. The resistance problems continued with the switch to newer insecticides such as the organophosphates (OPs), carbamates, and pyrethroids. Operationally, many control programs have now switched from blanket spraying of house interiors to focal use of insecticides on bednets. Focal spraying reduces the amount of insecticide used, but also limits the choice of insecticides to pyrethroids due to the need for rapid insect kill and high mammalian safety margin.
Today the major emphasis in research into resistance is on the molecular mechanisms of resistance, and rational resistance management, with a view to controlling the spread and development of resistant vector populations. The level of resistance in insect populations is dependent on the amount and frequency of insecticides used, and the inherent characteristics of the insect species selected. For example, for decades tsetse flies have been controlled by DDT treatment, but resistance has never developed in this species; the same is true of triatomid bugs. In both species the major factor influencing insecticide resistance development is the life cycle of the insect pest, in particular the long life cycles for the bugs, and the production of very small numbers of young by the tsetses. Mosquitoes, in contrast, have all the characteristics suited to rapid resistance development including short life cycles and abundant progeny.
The major mosquito vectors span the Culex, Aedes, and Anopheles genera. Culex are the vectors of filariasis and Japanese encephalitis, Aedes of dengue, dengue hemorrhagic fever, and yellow fever, and Anopheles of malaria. The range of many of these species is still expanding.
DDT was first introduced for mosquito control in 1946. In 1947 the first cases of DDT resistance occurred in Aedes (Brown, 1986). Since then over a hundred species of mosquito have become resistant to one or more insecticide (WHO, 1992). Insecticides used for malaria control have in
cluded ?-benzine hexachloride (BHC), organophosphorus, carbamate, and pyrethroid insecticides. Other insecticide groups, such as the benzylphenyl ureas and Bacillus thuringiensis (Bti), have had limited use against mosquitoes.
?-BHC/dieldrin resistance is still widespread, despite the lack of use of these insecticides for many years. OP resistance, in the form of either broad-spectrum resistance or malathion-specific resistance, occurs in many vectors (Hemingway, 1982, 1983; Hemingway and Georghiou, 1983; Herath et al., 1987). OP resistance is widespread in all the major Culex vectors (Hemingway and Karunaratne, 1998) and pyrethroid resistance occurs in C. quinquefasciatus (Chandre et al., 1998). Pyrethroid resistance is widespread in Ae. aegypti (Hemingway et al., 1989) and cases of OP and carbamate resistance also occur in this species (Mourya et al., 1993). The development of pyrethroid resistance in An. gambiae is particularly important given the recent emphasis on the use of pyrethroid impregnated bednets for malaria control (Vontas et al., 2001).
The peridomestic vectors of Leishmania are primarily controlled by insecticides throughout their range. The control of these sandflies is often a by-product of anti-malarial housespraying. The only insecticide resistance reported to date in sandflies is to DDT (El-Sayed et al., 1989).
The body louse Pediculus humanus has developed widespread resistance to organochlorines (Brown and Pal, 1971), is malathion-resistant in parts of Africa (WHO, 1992), and has “low-level” resistance to pyrethroids in several regions (Fine, 1963). Resistance to DDT and lindane occurs in the human head lice in Israel, Canada, Denmark, and Malaysia (WHO, 1992). Permethrin has been extensively used for head lice control since the early 1980s. The first reports of control failure with this insecticide were in the early 1990s in Israel (Mumcuoglu et al., 1995), the Czech Republic (Rupes et al., 1994), and France (Chosidow et al., 1994).
The Simulium damnosum complex, vectors of onchocerciasis, have been subjected to long-term insecticide-based control in West Africa since 1974. Temephos resistance prompted a switch to chlorphoxim, but resistance to this insecticide occurred within a year (Hemingway et al., 1991). Resistance in Simulium is currently being managed by a rotation of temephos, Bti, and permethrin, the insecticide usage being determined by the rate of water discharge in the major breeding sites of these vectors.
The Biochemistry of Resistance
Three major enzyme groups are responsible for metabolically based resistance to organochlorines, OPs, carbamates, and pyrethroids. DDT-
dehydrochlorinase is a glutathione S-transferase (Clark and Shamaan, 1984). Esterases are involved in OP, carbamate, and pyrethroid resistance. Monooxygenases are involved in the metabolism of pyrethroids, carbamate, and the activation and/or detoxication of OPs.
Esterase-Based Resistance: The esterase-based resistance mechanisms have been most extensively studied in Culex mosquitoes and the aphid Myzus persicae. Broad-spectrum OP resistance is conferred by the elevated esterases, which act by rapidly binding and slowly turning over the insecticide (Kadous et al., 1983). Two common esterase loci, esta and estß, are involved in Culex (Vaughan et al., 1997). In C. quinquefasciatus the most common elevated esterase phenotype involves esta21 and estß21 (Vaughan and Hemingway, 1995). Smaller numbers of Culex have elevated estß1 alone, elevated esta1 alone, or co-elevated estß1 and esta3 (DeSilva et al., 1997; Hemingway and Karunaratne, 1998). Differences of up to 1,000-fold in the inhibition kinetic constants of these esterases in resistant and susceptible insects occur for the oxon analogues of various OPs (Karunaratne et al., 1995). The superiority of insecticide binding in enzymes from resistant insects suggests that there has been a positive insecticide selection pressure to maintain elevation of favorable esterase.
In contrast to the situation in Culex a number of Anopheles species have a non-elevated esterase mechanism that confers resistance specifically to malathion through increased rates of metabolism (Hemingway, 1982, 1983, 1985; Malcolm and Boddington, 1989).
Glutathione S-Transferase-Based Resistance: GSTs are multifunctional enzymes that detoxify a large range of xenobiotics. They catalyze the nucleophilic attack of reduced glutathione (GSH) on the electrophilic centers of lipophilic compounds. Multiple forms of these enzymes have been reported for the mosquito, housefly, Drosophila, sheep blowfly, and grass grub (Clark et al., 1984; Clark et al., 1985; Toung et al., 1990).
There are at least three families of insect GST and all have a role in insecticide resistance. In Ae. aegypti at least two GSTs are elevated in DDT-resistant insects (Grant and Matsumura, 1989), while in An. gambiae a large number of different class III GSTs are elevated (Prapanthadara et al., 1993). The Ae. aegypti and An. gambiae GSTs in resistant insects are constitutively over-expressed. The GST-2 of Ae. aegypti is over-expressed in all tissues except the ovaries of resistant insects (Grant and Hammock, 1992).
Monooxygenase-Based Resistance: The monooxygenases are a complex family of enzymes involved in the metabolism of xenobiotics. The P450 monooxygenases are the rate-limiting enzyme step in the chain. P450s are involved in the metabolism of virtually all insecticides, leading to activation of OPs, or, more generally, detoxication. Elevated monooxygenase activity
is associated with pyrethroid resistance in An. stephensi, An. gambiae (Vulule et al., 1994), and C. quinquefasciatus (Kasai et al., 1998).
Target Site Resistance
The OPs, carbamates, organochlorines, and pyrethroids all target the nervous system. Newer classes of insecticides are now coming onto the market place for vector control, but the high cost of developing and registering them inevitably means that insecticides are developed initially for the agricultural market and then utilized for public health vector control, where their activities and safety profile are appropriate. Compounds targeting the nicotinic acetylcholine receptor have recently made this transition from agriculture into public health.
Acetylcholinesterase: The OPs and carbamates target acetylcholinesterase (AChE). Alterations in AChE in resistant insects result in a decreased sensitivity to insecticide inhibition of the enzyme (Hemingway and Georghiou, 1983). The OPs are converted to their oxon analogues, via the action of monooxygenases before acting as AChE inhibitors.
GABA Receptors: Resistance to dieldrin occurred in the 1950s, but the involvement of the GABA receptors in this resistance was not elucidated until 1990. The GABA receptor in insects is a widespread inhibitory neurotransmission channel in the central nervous system and in neuromuscular junctions (Bermudez et al., 1991). It is a site of action for pyrethroids and avermectins as well as cyclodienes (Kadous et al., 1983; Bloomquist, 1994).
Sodium Channels: The pharmacological effect of DDT and pyrethroids is to cause persistent activation of the sodium channels by delaying the normal voltage-dependent mechanism of inactivation (Soderlund and Bloomquist, 1989). In mosquitoes there have been many reports of suspected ”kdr”-like resistance inferred from cross-resistance between DDT and pyrethroids, which act on the same site within the sodium channel. These reports have been validated by electrophysiological measurements in Ae. aegypti and An. stephensi (Hemingway et al., 1989; Vatandoost et al., 1996).
The Molecular Biology of Resistance
Mutations in Structural Genes
Resistance to malathion is caused by a single Trp251-Leu substitution within the E3 esterase of Lucilia cuprina (Campbell et al., 1998). A similar phenotype has been observed in malathion-resistant Anopheles (Hemingway and Georghiou, 1983). A second Gly137-Asp substitution in E3 confers
broad cross-resistance to many OPs, but not to malathion (Newcomb et al., 1997).
Non-silent point mutations within structural genes are the cause of target site resistance. The mutations reduce insecticide binding without causing a loss of primary function of the target site. The number of possible amino acid substitutions is limited and identical resistance-associated mutations are found across divergent taxa. The degree to which function is impaired by the mutations is reflected in the fitness of resistant individuals in the absence of insecticide selection. This fitness cost is important in the persistence of resistance in the field.
An alanine to serine substitution in the channel-lining domain of the GABA receptor confers resistance to cyclodienes (ffrench-Constant et al., 1998). The mutation occurs in a broad range of dieldrin-resistant insects (Thompson et al., 1993), occasionally in an alanine to glycine form. Despite the widespread switch away from cyclodienes for agricultural and public health use, the resistance allele is still found at relatively high frequencies in insect field populations (Aronstein et al., 1995).
A reduction in the sensitivity of the voltage gated sodium channel to insecticide binding causes the kdr resistance phenotype. Kdr mutants are more variable than those in the GABA receptors, but are still limited to a small number of regions on this large channel protein. The first mutation to be characterized in kdr insects was a leucine to phenylalanine mutation in the S6 transmembrane segment of domain II in the sodium channel sequence (Martinez-Torres et al., 1998) that produces 10- to 20-fold resistance to DDT and pyrethroids. In “super-kdr” houseflies, this mutation is combined with a second methionine to threonine substitution further up-stream in the same domain resulting in >500-fold resistance (Williamson et al., 1996). A PCR-based diagnostic discriminates between homozygous-susceptible, homozygous-resistant, and heterozygotes with the Leu to Phe mutation (Martinez-Torres et al., 1998). Since kdr is partially recessive the ability to detect heterozygotes is of paramount importance in the early detection and management of resistance in the field.
The number of changes may be constrained by the possible modifications that can influence pyrethroid/DDT binding to the sodium channels. However, a note of caution is needed, as there is a tendency to investigate pyrethroid-resistant insects with a PCR approach confined to regions where a kdr mutation has already been seen. Hence changes in other parts of the sodium channel gene could be missed. A different approach to isolating kdr-type mutants has been used in Drosophila, utilizing the relative ease with which large numbers of mutants in the para sodium channel gene can be isolated based on their temperature sensitivity. Two classes of these mutations are in positions equivalent to the kdr and super-kdr mutations in
different domains. The third class is in a novel position (ffrench-Constant et al., 1998).
A range of different amino acid substitutions in the acetylcholinesterase Ace genes of Drosophila and the housefly M. domestica cause resistance (Feyereisen, 1995). Many of these mutations lie close to or within the active site gorge. Five point mutations associated with OP and carbamate resistance have been identified in D. melanogaster. AChE and site directed mutagenesis of the sex-linked AChE from Ae. aegypti have demonstrated that these same mutations also confer resistance in the mosquito enzyme, but none of these mutations have been identified in field-collected insects.
In three Culex species an estß gene is amplified in resistant insects (Vaughan et al., 1995; Whyard et al., 1995; Karunaratne et al., 1998). The commonest amplified esterase-based mechanism involves the co-amplification of two esterases, esta21 and estß21 in C. quinquefasciatus and other members of the C. pipiens complex (Vaughan et al., 1997). Other strains of Culex have amplified esta3 and estß1 co-amplified (DeSilva et al., 1997), while the TEM-R strain has amplified estß1 alone (Mouches et al., 1986).
The esta and estß genes have arisen as the result of an ancient gene duplication. The genes are in a head-to-head arrangement approximately 1.7 kb apart in susceptible insects (Vaughan et al., 1997). In resistant insects the amplified esta21/estß21 genes are 2.7 kb apart; the difference is accounted for by expansion with three indels in the intergenic spacer (Vaughan et al., 1997), which have introduced further gene regulators (Hemingway et al., 1998). Amplification of these alleles has occurred once and spread worldwide by migration (Raymond et al., 1991).
Amplified esterases are expressed at different levels. For example, there is four-fold more estß than esta in resistant C. quinquefasciatus, although the genes are present in a 1:1 ratio. This difference in expression is reflected at the protein and mRNA level (Karunaratne et al., 1995).
GSTs metabolize DDT and OPs (Hemingway et al., 1985) and moderate the toxic effects of pyrethroid-generated radicals (Vontas et al., 2001). GST-based DDT resistance is common in Anopheles, reflecting decades of heavy use of DDT for malaria control. Molecular characterization of GSTs is most developed in An. gambiae and An. dirus (Hemingway et al., 1998; Prapanthadara et al., 1998). Three classes of insect GSTs are important in insecticide metabolism in insects. Aedes aegypti GST-2, is overexpressed in
a DDT-resistant GG strain, where the resistance mutation leads to disruption of a trans-acting repressor (Grant and Hammock, 1992).
In An. gambiae multiple class I and class III GST genes are clustered in separate single locations. One gene, aggst1a, is alternatively spliced to produce four distinct mRNA transcipts each of which shares a common 5’ exon but differing 3’ exons (see Figure 3-1) (Ranson et al., 1998). The organizations of the class I GST gene family in insecticide-resistant and -susceptible An. gambiae are similar.
Insect P450s belong to six families. Increased transcription of genes belonging to the CYP4, CYP6, and CYP9 families occurs in insecticide-resistant strains. It is not known which enzymes are responsible for insecticide metabolism in mosquitoes. Helicoverpa armigera has different resistance-associated P450s between strains; CYP6B2 is overexpressed in a pyrethroid-resistant strain (Xiao-Ping and Hobbs, 1995), whereas CYP4G8 is overexpressed in another (Pittendrigh et al., 1997).
P450-based resistance in M. domestica (housefly) and D. melanogaster is mediated by mutations in trans-acting regulatory genes. CYP6A8 is highly expressed in the DDT-resistant 91-R strain of D. melanogaster but not detectable in the uninduced 91-C susceptible strain (Maitra et al., 1996; Dombrowski et al., 1998). Hybrids between the two strains show low levels of expression suggesting that the 91-C strain carries a repressor that suppresses transcription of CYP6A8 (Liu and Scott, 1997).
Management of Insecticide-Resistant Vector Populations
The use of an insecticide until resistance becomes a limiting factor is rapidly eroding the number of suitable insecticides for insect control. Rotations, mosaics, and mixtures have all been proposed as resistance management tools. Numerous mathematical models have been produced to estimate how these tools should be optimally used (Tabashnik, 1989). However, these models have rarely been tested under field conditions due to the practical difficulties in estimating changes in resistance gene frequencies in large samples of insects (Hemingway et al., 1997). With the advent of different biochemical and molecular techniques for resistance gene frequency estimation, field trials of resistance management strategies have become more feasible. A large-scale trial of the use of rotations or mosaics of insecticides compared to single use of DDT or a pyrethroid is currently under way in Mexico (Penilla et al., 1998). Changes in resistance gene frequencies in An. albimanus are being monitored over a four-year period. Information resulting from such large-scale trials may allow us to establish rational strategies for long-term insecticide use. As our ability to manipulate the insect genome improves and our understanding of the regulation of insecticide resistance mechanisms increases, new strategies should be devised for incorporation into these control programs.
MANAGING THE EMERGENCE OF PESTICIDE RESISTANCE IN VECTORS
William G. Brogdon, Ph.D.
Centers for Disease Control and Prevention, Atlanta, GA
A linked four-stage process, involving investment, surveillance, interpretation, and remediation, is described for management of insecticide resistance in vectors. Insecticide resistance has been a problem in all insect groups that serve as vectors of emerging diseases. Although mechanisms by which insecticides become less effective are similar across all vector taxa, each resistance problem is potentially unique and may involve a complex pattern of resistance foci. The main defense against resistance is close surveillance of the susceptibility of vector populations. This requires the investment of programs in surveillance. Once surveillance data have been gathered, it is crucial that the results be interpreted practically, in terms of control efficacy. Only then can strategies for remediation be undertaken.
There are fundamental differences between the management of pesticide and that of antimicrobial resistance. Rapid generation times in mi
crobes can lead to rapid selection of highly resistant populations, even in the course of a single infection. Generation times in vectors are significantly longer, resulting in more gradual selection of resistance. Among the more rapid emergences of insecticide resistance we have observed was the appearance and intensification of fenitrothion resistance over a three-month period at one location in Haiti following an insecticide spray cycle (Brogdon et al., 1988).
The focus of antibiotic resistance management is on infections in individual patients or in specific institutions. The focus of insecticide resistance management is on vector populations in a geographic control area. The use of antibiotics follows similar principles throughout the world. The application strategies for insecticides are radically different around the world. For control of malaria, wall spraying and the use of insecticide-impregnated bednets are the most used methods of control. In the United States, ground and aerial application of ultra-low volumes of insecticide are widely used.
Changes in antibiotic use have little environmental impact. Changes in insecticide use can have significant environmental impact. Collection, culture, and testing of resistant microbe strains is not technically difficult or expensive. Collection, culture, and testing of resistant insects can be both difficult and expensive. Generally, teams of individuals must go into the field and collect through the night, sometimes for days, to obtain an adequate number of insects for resistance testing. Collections may include adults, larvae, or eggs of vectors.
Control failure in antibiotic use can have immediate fatal consequences for the patient. Insecticide control failure may have less obvious impact on disease control. The continued survival of a vector species may be less noticeable against a background of pest species that have been controlled, giving the illusion of adequate vector control. For example, good control of pest species of mosquitoes in Florida gave the impression that control with a particular insecticide was adequate. Further testing revealed that Culex nigripalpus, the principal vector species, did not respond to that insecticide (Brogdon, unpublished data).
The most significant feature shared by antimicrobial and insecticide resistance management is that the choice of new chemical options is dangerously limited. It is crucial that legislators and health officials make the investment necessary to manage resistance. Funding allows resistance problems to be detected and assessed through the use of resistance surveillance. Going through the motions of resistance surveillance is a waste of resources unless the data obtained are subjected to professional and timely interpreta tion. Only after the problem has been adequately detected and interpreted can the program develop a strategy for remediation of the problem.
Investment in Resistance Management
The investment that is being applied to antimicrobial resistance becomes obvious if one simply does an Internet search with the keywords “antimicrobial resistance.” Key sites that are immediately located include those established by WHO, the European Antimicrobial Resistance Surveillance System, and the Centers for Disease Control and Prevention (CDC). These sites provide timely data on the detection, assessment, and distribution of antimicrobial resistance throughout the world.
Corresponding sites dedicated to insecticide resistance are not found. The extent of the problem was revealed by a fortunate investment under the Emerging Infections Program at CDC that funded an Insecticide Resistance Surveillance Laboratory. Through collaborative work with over 20 states, it became clear that insecticide resistance is widespread, though highly focal in the United States (Brogdon, unpublished data). More troubling is the observation that knowledge of insecticide resistance status is virtually non-existent in vector control programs. Similar difficulties apply overseas. In general, insecticide susceptibility testing is the last activity funded in control programs. If funded, it is the first activity cut. A serious commitment to resistance management is not made. Though there are exceptions, these observations are generally applicable.
Reversing the trend regarding investment in insecticide resistance testing will require training of the appropriate health officials in the techniques that may be used to detect and assess resistance problems. This may be done through training courses or through training-oriented websites. One such website for insecticide resistance testing has been produced at the CDC (http://www.cdc.gov/ncidod/wbt/resistance/).
Insecticide Resistance Surveillance
The most fundamental feature of insecticide resistance problems may be that each such problem is potentially unique, given the broad variety of resistance mechanisms available to the insect and the great disparity in insecticide use history and thus selection pressure on natural populations of vectors. Given the diversity of problem types and the focal nature of resistance, it becomes the primary goal of resistance surveillance to measure resistance as it exists, at a particular place, at a particular time. Fortunately, an array of bioassay, biochemical assay, and molecular methods already exist (and many others are under active development) to facilitate detection and assessment of resistance problems.
The first line of defense (and the most cost-effective) is a bioassay method, such as the bottle bioassay developed at the CDC (Brogdon and
McAllister, 1998a). Using this technique, the specific formulation of insecticide(s) used in a program may be used in a simple test of susceptibility of the vector species. A discriminating insecticide dosage is empirically determined, a resistance threshold (upper range limit of time of survival of a susceptible population) is established, and testing can proceed to rapidly characterize populations in the control area. Use of insecticide resistance enzyme inhibitors (synergists) can provide information on the mechanism of insecticide resistance in bottle assays, but, typically, the most detailed information on resistance mechanisms comes through application of biochemical resistance assays, generally conducted in microtiter plates. Knowledge of the mechanism allows a program to make an informed decision on how to combat the problem. For example, some mechanisms may be rendered ineffective through the judicious use of synergists. In other cases, it becomes necessary to switch insecticides. Knowing the resistance mechanism protects the program from falling prey to cross-resistance, where resistance to the insecticide used also crosses to insecticides that have not been used, but that are susceptible to the same resistance mechanism.
Biochemical microplate assays are available for the oxidase, esterase, glutathione S-transferase, and insensitive acetylcholinesterase resistance mechanisms. These assays give clear evidence of resistance mechanisms, provided their results are carefully correlated to bioassay data from the same vector populations. One drawback of these techniques is that the insects to be tested must be fresh or frozen at dry ice (–80°C or better) temperatures. This becomes a real problem in field situations, particularly in the tropics. Moreover, no biochemical assay yet exists (though one is under development) for the target site resistance mechanism (kdr) that afflicts pyrethroid use.
Biochemical data are interpreted similarly to bioassays, in that an upper range limit for resistance enzyme activity is established as a threshold, allowing individuals with higher than normal activities to be classified as less susceptible (given corroborative bioassay data). The biochemical data should account for a similar proportion of a vector population above the resistance threshold as bioassay tests on the same population. An advantage of biochemical assays is that they may be conducted as a complete panel or battery of tests, generally conducted in triplicate, on each insect. Data from multiple tests may be plotted on the same graph, clearly revealing instances of cross-resistance.
Increasing understanding of insect molecular biology is transforming the field of resistance detection, through direct sequencing of resistance genes and the applied use of both qualitative and quantitative PCR and RTPCR. Advanced techniques such as fluorescence PCR are allowing effi
cient detection of resistance gene copy number (multiple esterase copies are a frequent cause of resistance), level of gene expression (important in the increased expression of oxidase genes), and the efficient detection of point mutations in insecticide target genes (Brogdon and McAllister, 1998b).
There are two huge advantages to the use of molecular resistance detection techniques. First, preserved (in alcohol or desiccated) specimens may now be used in the analysis, allowing much easier logistics in the collection and transport of collected vector insects. Also, molecular analysis allows the use of insects collected as pools. Collection of vector samples is expensive in both time and personnel. It would be invaluable if the same pools of insects collected for routine viral surveillance, for example, could also be used for identification of species (and their relative number), measurement of resistance mechanism indices, and correlation of resistance-positive and disease-positive pools. This exciting scenario should be within reach of current molecular technology.
Given investment in surveillance, we now have the technology to provide the necessary raw data for us to manage insecticide resistance. These techniques are making it possible to deal with the greatest emerging problem in resistance surveillance, that of multiple resistance. Multiple resistance is the resistance to an insecticide through multiple resistance mechanisms. Additional resistance mechanisms may enhance, reduce, or modify the insecticide specificity of the resistance based upon a single mechanism. Such complex problems make proper interpretation of data a critical factor in resistance management.
Interpretation of Resistance Data
It is an unfortunate part of the history of vector control that so much routine resistance surveillance data was detected in the absence of subsequent informed analysis. Frequently such testing was done as a routine, with the methods carefully followed, but with the data filed away uninterpreted. It is the most frequent failing in the process of conducting resistance management.
One particular problem that arises is that field personnel do not know how to cope with interpretation when they lack access to an established susceptible strain of the insect being tested. All that is required is for the program to establish a susceptibility baseline using the current target population. These data are interpreted though evaluation of the level of current control success on that population. The mission then becomes to monitor for changes in resistance baseline and in control success. The presence of even high levels of insecticide resistance may or may not be relevant to
vector control. If the efficacy of control is not being concurrently assessed, it serves no purpose to conduct resistance susceptibility testing. Even if resistance is “detected,” its significance will remain unknown. All successful control programs carefully assess the efficacy of their program, whether by plotting incidence of disease cases or by counting adult or larval collections at informative sites. These data must be integrated with the resistance surveillance data for effective remediation of problems to be contemplated.
Remediation of Resistance
Just as every resistance problem may be unique, every solution is potentially unique. Management schemes take one of two forms. The simplest is surveillance-response. The most inexpensive strategy, it involves establishment of susceptibility baselines, periodic susceptibility testing, correlation of changes with control efficacy, and change of control strategy when the data indicate the necessity of doing so. Response choices are to switch chemicals, to apply chemicals focally, or where chemicals are ineffective, to use source reduction or concentrate on personal protection.
The more complicated and expensive strategy, integrated pest management (IPM), presupposes a deeper understanding of the disease and vector. Strategies for IPM include rotations of chemicals, timed application of chemicals, use of mixtures, and the provision of refugia (mosaic application of insecticides) for resistance genes. Thus far, the best example of a resistance management program in a disease vector is the Onchocerciasis Control Program in Africa. There, 11 countries invested (through the assistance of 80 development partners) in resistance surveillance, interpretation, and remediation in an area where 35 million people are at risk for the disease (WHO, 1997). Where there has been success in one program lies the potential for success in others.
WHAT IS THE ROLE OF INSECTICIDE RESISTANCE IN THE RE-EMERGENCE OF MAJOR ARTHROPOD-BORNE DISEASES?
Donald R. Roberts, Ph.D.*
Department of Preventive Medicine and Biometrics Uniformed Services University of the Health Sciences, Bethesda, MD
Paul B. Hshieh, Ph.D.
Office of Biostatistics and Epidemiology, Center for Biologics Evaluation and Research Food and Drug Administration, Rockville, MD
Twenty years of spectacular and continual growth in cases (see Figure 3-2) and geographical spread of human malaria (Roberts et al., 1997) and dengue fever (Gubler, 1998) in the Americas should send a resounding message that something is horribly wrong in modern approaches to disease control. Within the last few months dengue has spread to Hawaii and Easter Island, and an outbreak is occurring in Argentina. In fact, outbreaks are occurring commonly throughout the Americas (ProMED-mail, 2002). Aedes aegypti is the primary vector of dengue. This vector has re-invaded and proliferated in almost all areas of the New World where it had previously been eliminated. Modern vector control methods have not measurably slowed re-conquest of the Americas by this important vector of dengue and yellow fever viruses.
The insecticide and methods previously used to exert decades of control over malaria and to eliminate Ae. aegypti are no longer (or rarely) used. Malaria and dengue were controlled in different ways; but the insecticide (DDT) was the same for both. The environmental movement has forced the move away from proven methods of disease control. Environmental activists have campaigned to stop use of DDT and other insecticides in public health (Amato, 1993; World Wildlife Fund [WWF], 1999). Claims that resistance neutralized the effectiveness of insecticides have been part of the environmental campaign, as illustrated in the claim that “DDT resistance has been a factor in the failure of many national malaria eradication programs.” (WWF, 1999).
Decades of research have been employed to reduce the complex phenomena of insecticide resistance in disease vectors into simpler subsystems. As comprehensively reviewed by Hemingway and Ranson (2000), resistance has been studied in many vector species, especially in vectors of malaria and dengue viruses, and at many levels, to include its molecular basis. This reductive approach has not led to accurate models and predictive theories that are broadly applicable to disease control operations. As recently shown with insecticide-treated nets, resistance does not necessarily counter the benefits of insecticide use (Sina and Aultman, 2001). The same has been shown for the continuing effectiveness of sprayed houses in controlling malaria, even when the vectors are resistant to the insecticide that is being used (Roberts and Andre, 1994).
Failure of predictive accuracy has occurred because research has fo
cused almost completely on toxicological resistance, which, in turn, is based on the erroneous premise that insecticides control disease transmission by killing insects. No example of this fundamental error is more revealing than in the resistance of malaria vector mosquitoes to DDT.
A recent probability model (Roberts et al., 2000) quantifies DDT’s mode of action in controlling malaria transmission inside houses. The model shows that greatest impact of DDT residues stems from a repellent action. A contact irritant action is next in order of importance. The model was tested against field data for important malaria vectors in the Americas, Africa, and Asia. The results showed that mortality accounted for less than 10 percent of the total impact of DDT residues against two vectors, from 10 percent to about 36.2 percent against a third vector, and 18.3 percent against a fourth vector. In a separate study, the model was field validated in experimental hut studies against a fifth vector species in southern Belize (Grieco et al., 2000). This model represents the first use of stochastic methods to elucidate actual functions of DDT residues, but it was certainly not the first demonstration of DDT’s powerful repellent and irritant actions.
From the very beginning of DDT’s use, it was known as a slow-acting poison (Metcalf et al., 1951) that strongly influenced insect behavior. In 1947 Kennedy published a paper (Kennedy, 1947) entitled “The Excitant and Repellent Effects on Mosquitoes of Sub-Lethal Contacts with DDT.” This paper warrants careful consideration because its 1947 publication date means that the research was actually conducted before wide-scale use of DDT in malaria control (which started in 1946). Kennedy’s experiments demonstrated repellent and irritant actions of DDT against a vector of malaria (Anopheles maculipennis atroparvus) and Ae. aegypti. Kennedy’s work was published five years before the first report of DDT resistance in a malaria vector mosquito.
Kennedy introduced his paper as follows:
During the war it was necessary to concentrate on the practical development of DDT as an insecticide without waiting for studies of its mode of action. Working assumptions had to be made, especially with regard to the possible repellency of DDT deposits. Buxton (1945) sums up what has been the accepted opinion on this subject, as follows: “DDT does not act as a repellent to any insect, so far as is known” (Kennedy, 1947).
He went on to say that “… the existence of contact repellency has been firmly established.” Buxton (1945) mentioned “restlessness” as an early symptom of DDT poisoning. This has been described under semi-field conditions by Gahan and others (1945) and by Metcalf and others (1945), who observed also that mosquitoes excited by contact with DDT no longer stayed in dark corners but made for the windows.
However, such repellency has been discounted as of no practical signifi
cance because, in Buxton’s (1945) words: “So far as is known, once visible symptoms develop, death follows: recovery from an early stage of poisoning does not occur.”
“This idea is perhaps even more widespread than the idea that DDT is not repellent” (Kennedy, 1947).
Kennedy ended his introductory paragraph with the statement that “Our laboratory observations did not bear out these statements” (Kennedy, 1947).
Kennedy then reported on a series of laboratory experiments showing that mosquitoes were strongly repelled by DDT residues (for example, see Figure 3-3), and that the test specimens often recovered completely after showing preliminary symptoms of poisoning. With this documentation he concluded that “DDT must be regarded as acting in two contradictory ways simultaneously. It acts as a lethal agent on the one hand, and as an excitant and thereby sometimes a repellent on the other” (Kennedy, 1947).
Kennedy’s work was conducted at a time when Professor George Macdonald was establishing the mathematical foundations of malaria control. Kennedy commented on Macdonald’s views:
Macdonald (in discussion on Buxton, 1945, p. 394) was convinced that the reduction in anopheline infestation of sprayed rooms was due to actual destruction of mosquitoes and “not due to repellent effect, a point of extreme importance.” His evidence was the presence of dead mosquitoes and the lack of any increase in the infestation of adjoining, untreated rooms. Nevertheless he remarked that there was no evidence of any reduction in such untreated rooms for which, as he said, “one might legitimately hope” in view of the passage of mosquitoes from room to room. The solution of this puzzle is surely that there was both destruction and repulsion of the mosquitoes (Kennedy, 1947).
Six years later, in 1953, Macdonald and Davidson published what became the founding principles for using insecticides in malaria control. Their analyses were based on field studies from 1947 to 1952. They did not cite Kennedy’s paper and chemical actions favoring mosquito survival were characterized as undesirable. Macdonald and Davidson struggled valiantly to reconcile DDT’s slow killing power with its spectacular ability to reduce malaria.
The authors knew of DDT’s complex influence on vector behavior, as revealed in assessments that:
“DDT was the most irritant of the three insecticides (DDT, BHC, and Dieldrin), and a very large proportion of the mosquitoes escaped its action.”
“The marked irritant properties of DDT were evident in all these treatments, the greater proportion, and sometimes as much as 90 per cent, of the mosquitoes being caught in the window traps.”
“The marked irritant effect of DDT on mosquitoes makes adequate dosage … imperative” (Macdonald and Davidson, 1953).
In the end, Macdonald and Davidson concluded that the answer to DDT’s “irritant” action was to increase dosage so that mosquitoes would absorb a lethal dose before chemical repulsion. A surprising conclusion considering that Kennedy (1947) reported that “… excitation appears quickly, often in a matter of seconds after the insects are brought into contact with DDT …” (Kennedy, 1947).
One might wonder how Macdonald and Davidson, with their comprehensive knowledge of malaria, could so seriously err in understanding DDT’s mode of action? In their defense, key elements of field data were missing, as revealed in their assessment that “In most, if not all, … entomological assessment of the efficiency of the insecticides has been based
on records of the reduction in numbers of the daytime-resting population of treated shelters. No account has been taken of mosquitoes entering and leaving the shelters during the night” (Macdonald and Davidson, 1953).
Macdonald and Davidson stated that, of their tentative conclusions, “the one most clearly brought to light is that a considerable expansion of our knowledge is urgently needed—an expansion in the fields of basic theory, anopheline habits, the physical chemistry of insecticides, and their mode of action” (Macdonald and Davidson, 1953).
Unfortunately, rapid acceptance of their model virtually guaranteed that research and expansion of knowledge, particularly regarding chemical mode of action, would not occur. Macdonald and Davidson allowed that DDT killed mosquitoes that entered houses and reduced longevity of malaria vector populations below what is needed to maintain malaria transmission. Acceptance of this concept ended debate and support for research on DDT’s mode of action. With preeminence of the idea that DDT reduced vector population longevity, Kennedy’s research, and that of many others, was relegated to obscurity. The facts that DDT was a powerful repellent, a functional irritant, and only a slow-acting poison were ignored. Those facts have now been resurrected in the probability model, described above.
To fully appreciate the probability model, it is important to understand that insects have extraordinary abilities to detect and respond to chemicals; for example, certain species are capable of detecting a single molecule in 1017 molecules of air. Insects have developed these abilities as an adaptation to secondary chemicals that occur in nature. Plants and animals produce the so-called “secondary” chemicals to repel, deter, irritate, or poison herbivores, predators, and parasites (Schoonhoven, 1985). DDT is an organohalogen compound, and organohalogens are produced as secondary chemicals in nature (Gribble, 1999). The similarity of DDT to natural organohalogens is suggested by the finding that males of a species of euglossine bee are attracted to DDT and will travel to houses and collect DDT from sprayed walls (Roberts et al., 1982). The bees store (relative to their size) large amounts of DDT in hind tibial pouches and, remarkably, are not harmed. It is also worth noting that we now know that some natural organohalogens are persistent in the environment, are lipophilic, bioaccumulate, and may be present in greater abundance than synthetic compounds like DDT (Vetter et al., 2001). As a final point, winged adult insects can easily and rapidly move away from a poison (given that the chemical is not useful to the insect) and avoid the biological cost of premature death or metabolic detoxification.
When a malaria vector is inside a house it is exposed to human hosts, outdoors it is exposed to a variety of hosts. Thus, it is generally understood that malaria transmission becomes most efficient when vectors and humans come together inside the confined spaces of homes. When the vector is
outside, it only has access to indirect human host stimuli emanating from inside the house. Once the vector is indoors, host proximity and unobstructed “view” of the human host (no intervening wall) alters the balance between factors that inhibit versus factors that stimulate a vector to bite. These factors and host availability influence probabilities of disease transmission.
By definition, a repellent action occurs when the mosquito is repelled without making physical contact with the chemical. In contrast, a mosquito is irritated only after making physical contact with a sprayed surface. Physical contact that results in an irritant response (escape response) can occur after very brief physical contact (as shown by Kennedy, 1947). A toxic action requires more prolonged physical contact with a sprayed surface (Roberts and Andre, 1994). The vector’s actions, for example, moving to a house, entering the house, resting on walls, and biting, can be aligned sequentially with repellent actions that prevent house entering; irritant actions that promote premature exiting and prevent biting indoors; and toxic actions that produce mortality (Roberts et al., 2000).
We used the following notations to simulate interactions of repellent, irritant, and toxic actions of insecticides:
T: treatment house.
C: control house.
pe: probability a mosquito will enter the house.
pb|e: probability a mosquito will bite, given that it enters the house.
pbe: probability a mosquito will enter the house and bite indoors.
ps|be: probability a mosquito survives, given that it entered the house and took a blood meal indoors.
psbe: probability a mosquito will enter the house, bite indoors, and survive.
Applying the multiplication law of probability (Rosner, 1995), we have the following equations
pbe = pb|e × pe (1)
psbe= ps|be × pb|e × pe (2)
As stated above, we assume that probabilities for different vector activities in the control house each equate to unity. This assumption leads to the relationship of and implies that ; see equations (1) and (2). The influence of insecticide on mosquito activities can be described as the probability difference between control and treatment for entering, biting, and surviving. For the act of entering the house, the probability difference is ; and for the biting and surviving,
and, respectively. Actually, based on conditional probabilities, is the cumulative effect due to insecticide repellency, is the cumulative effect due to repellency plus irritancy, and is the cumulative effect due to repellency, irritancy, and toxicity. The probability of entering, , can be estimated by
Using comparative data from the control house, pb|e can be estimated by the observed proportion of mosquitoes that bite or become blood engorged within the treated house, and ps|be can be estimated by the observed proportion of mosquitoes that survived after they entered and fed in the treated house. To model field data, we would further adjust probabilities to account for natural differences in numbers collected in houses as a result of location differences.
Our simulations (formulas not presented) partitioned the total impact of a sprayed house by the percent of effect that can be attributed to each chemical action. Overall, the combined effects of repellent and irritant actions () did not drop below a cumulative level of 50 percent effectiveness until probabilities of entering and biting indoors were very high, and (i.e., almost no repellent or irritant action).
In experimental hut studies the repellent actions of DDT residues can function at levels above 90 percent compared to unsprayed huts (Grieco et al., 2000; Roberts et al., 2000). No repellent action was detected in a deltamethrin-sprayed hut, but a pronounced irritant action drove mosquitoes out of the deltamethrin-sprayed hut several hours before they exited the control hut. The combined message from model simulations and experimental hut studies is that DDT functions as a repellent, secondarily as an irritant, and lastly as a poison. In contrast, deltamethrin seems to function as an irritant and secondarily as a poison. These findings highlight a simple truth, an insecticide is just one more chemical in the mosquito’s chemical world. The mosquito can detect and respond to an insecticide in many ways, perhaps the least likely being physical contact and death.
We have shown that repellent and irritant actions of DDT did not suddenly surface after decades of DDT use; but were actually recognized as major actions of DDT residues from the very start of DDT’s use as a public health insecticide. In conclusion, funding agencies should formally recognize behavioral actions of vectors to insecticides as a priority area for future support.
Amato I. 1993. The crusade against chlorine. Science261:152–154.
Aronstein K, Ode P, ffrench-Constant RH. 1995. PCR based monitoring of specific Droso phila (Diptera: Drosophilidae) cyclodiene resistance alleles in the presence and absence of selection. Bulletin of Entomological Research85:5–9.
Bermudez I, Hawkins CA, Taylor AM, Beadle DJ. 1991. Actions of insecticides on the insect GABA receptor complex. Journal of Receptor Research11:221–232.
Bloomquist JR. 1994. Cyclodiene resistance at the insect GABA receptor chloride channel complex confers broad cross-resistance to convulsants and experimental phenylpyrazole insecticides. Archives of Insect Biochemistry and Physiology26:69–79.
Brogdon WG and McAllister JC. 1998a. Simplification of adult mosquito bioassays through use of time-mortality determinations in glass bottles. Journal of the American Mosquito Control Association14:159–164.
Brogdon WG and McAllister JC. 1998b. Insecticide resistance and vector control. Emerging Infectious Diseases4:605–613.
Brogdon WG, Hobbs JH, St. Jean Y, Jacques JR, Charles LB. 1988. Microplate assay analysis of reduced fenitrothion susceptibility in Haitian Anopheles albimanus. Journal of the American Mosquito Control Association4:152–158.
Brown AWA. 1986. Insecticide resistance in mosquitoes: a pragmatic review. Journal of the American Mosquito Control Association2:123–140.
Brown AWA and Pal R.1971. Insecticide resistance in arthropods. WHO Monograph Series 38.
Campbell PM, Newcomb RD, Russell RJ, Oakeshott JG. 1998. Two different amino acid substitutions in the ali-esterase, E3, confer alternative types of organophosphorus insecticide resistance in the sheep blowfly, Lucilia cuprina. Insect Biochemistry and Molecu lar Biology28:139–150.
Chandre F, Darriet F, Darder M, Cuany A, Doannio JMC, Pasteur N, Guillet P. 1998. Pyrethroid resistance in Culex quinquefasciatus from West Africa. Medical and Veteri nary Entomology12:359–366.
Chosidow O, Chastang C, Brue C, Bouvet E, Izri M, Monteny N, Bastuji-Garin S, Rousset JJ, Revuz J. 1994. Controlled study of malathion and d-phenothrin lotions for Pediculus humanus var capitis-infested schoolchildren. Lancet344:1724–1727.
Clark AG and Shamaan NA. 1984. Evidence that DDT-dehydrochlorinase from the house fly is a glutathione S-transferase. Pesticide Biochemistry and Physiology22:249–261.
Clark AG, Dick GL, Martindale SM, Smith JN. 1985. Glutathione S-transferases from the New Zealand grass grub, Costelytra zealandica.Insect Biochemistry15:35–44.
Clark AG, Shamaan NA, Dauterman WC, Hayaoka T. 1984. Characterization of multiple glutathione transferases from the housefly, Musca domestica (L).Pesticide Biochemistry and Physiology22:51–59.
DeSilva D, Hemingway J, Ranson H, Vaughan A. 1997. Resistance to insecticides in insect vectors of disease: Esta3, a novel amplified esterase associated with estß1s from insecticide resistant strains of the mosquito Culex quinquefasciatus.Experimental Parasitology 87:253–259.
Dombrowski SM, Krishnan R, Witte M, Maitra S, Diesing C, Waters LC, Ganguly R. 1998. Constitutive and barbital-induced expression of the CYP6A2 allele of a high producer strain of CYP6A2 in the genetic background of a low producer strain. Gene221:69–77.
El-Sayed S, Hemingway J, Lane RP. 1989. Susceptibility baselines for DDT metabolism and related enzyme systems in the sandfly Phlebotomus papatasi (Scopoli) (Diptera: Psy chodidae).Bulletin of Entomological Research79:679–684.
Feyereisen R. 1995. Molecular biology of insecticide resistance. Toxicology Letters82:83–90.
ffrench-Constant RH, Pittendrigh B, Vaughan A, Anthony N. 1998. Why are there so few resistance-associated mutations in insecticide target genes?Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences353:1685–1693.
Fine BC. 1963. The present status of resistance to pyrethroid insecticides. Pyrethrum Post 7:18–21.
Grant DF and Hammock BD. 1992. Genetic and molecular evidence for a trans-acting regulatory locus controlling glutathione S-transferase-2 expression in Aedes aegypti.Molecular and General Genetics234:169–176.
Grant DF and Matsumura F. 1989. Glutathione S-transferase 1 and 2 in susceptible and insecticide resistant Aedes aegypti.Pesticide Biochemistry and Physiology33:132–143.
Gribble GW. 1999. Chlorine—element from hell or gift from God? The scientific side of the chlorine story. Technology6:193–201.
Grieco JP, Achee NL, Andre RG, Roberts DR. 2000. A comparison study of house entering and exiting behavior of Anopheles vestitipennis (Diptera: Culicidae) using experimental huts sprayed with DDT or deltamethrin in the southern district of Toledo, Belize, C.A. Journal of Vector Ecology25:62–73.
Gubler DJ. 1998. Resurgent vector-borne diseases as a global health problem. Emerging Infectious Diseases4:442–450.
Hemingway J. 1982. The biochemical nature of malathion resistance in Anopheles stephensi from Pakistan. Pesticide Biochemistry and Physiology17:149–155.
Hemingway J. 1983. Biochemical studies on malathion resistance in Anopheles arabiensis from Sudan. Transactions of the Royal Society of Tropical Medicine and Hygiene 77:477–480.
Hemingway J. 1985. Malathion carboxylesterase enzymes in Anopheles arabiensis from Sudan. Pesticide Biochemistry and Physiology23:309–313.
Hemingway J and Georghiou GP. 1983. Studies on the acetylcholinesterase of Anopheles albimanus resistant and susceptible to organophosphate and carbamate insecticides. Pes ticide Biochemistry and Physiology19:167–171.
Hemingway J and Karunaratne SH. 1998. Mosquito carboxylesterases: a review of the molecular biology and biochemistry of a major insecticide resistance mechanism. Medical and Veterinary Entomology12:1–12.
Hemingway J and Ranson H. 2000. Insecticide resistance in insect vectors of human disease. Annual Review of Entomology45:371–391.
Hemingway J, Boddington RG, Harris J, Dunbar SJ. 1989. Mechanisms of insecticide resistance in Aedes aegypti (L.) (Diptera: Culicidae) from Puerto Rico. Bulletin of Entomo logical Research79:123–130.
Hemingway J, Callaghan A, Kurtak DC. 1991. Biochemical characterization of chlorphoxim resistance in adults and larvae of the Simulium damnosum complex (Diptera: Simulidae). Bulletin of Entomological Research81:401–406.
Hemingway J, Hawkes N, Prapanthadara L, Jayawardena KG, Ranson H. 1998. The role of gene splicing, gene amplification and regulation in mosquito insecticide resistance. Philo sophical Transactions of the Royal Society of London. Series B: Biological Sciences 353:1695–1699.
Hemingway J, Malcolm CA, Kissoon KE, Boddington RG, Curtis CF, Hill N. 1985. The biochemistry of insecticide resistance in Anopheles sacharovi: comparative studies with a range of insecticide susceptible and resistant Anopheles and Culex species. Pesticide Biochemistry and Physiology24:68–76.
Hemingway J, Penilla RP, Rodriguez AD, James BM, Edge W, Rogers H, Rodriguez MH. 1997. Resistance management strategies in malaria vector mosquito control. A large scale field trial in Southern Mexico. Pesticide Science51:375–382.
Herath PR, Hemingway J, Weerasinghe IS, Jayawardena KG. 1987. The detection and characterization of malathion resistance in field populations of Anopheles culicifacies B in Sri Lanka. Pesticide Biochemistry and Physiology29:157–162.
Kadous AA, Ghiasuddin SM, Matsumura F, Scott JG, Tanaka K. 1983. Difference in the picrotoxinin receptor between the cyclodiene-resistant and susceptible strains of the German cockroach. Pesticide Biochemistry and Physiology19:157–166.
Karunaratne SH, Hemingway J, Jayawardena KG, Dassanayaka V, Vaughan A. 1995. Kinetic and molecular differences in the amplified and non-amplified esterases from insecticide-resistant and susceptible Culex quinquefasciatus mosquitoes. Journal of Biological Chem istry270:31124–31128.
Karunaratne SH, Vaughan A, Paton MG, Hemingway J. 1998. Amplification of a serine esterase gene is involved in insecticide resistance in Sri Lankan Culex tritaeniorhynchus. Insect Molecular Biology7:307–315.
Kasai S, Weerasinghe IS, Shono T. 1998. P450 Monooxygenases are an important mechanism of permethrin resistance in Culex quinquefasciatus say larvae. Archives of Insect Bio chemistry and Physiology37:47–56.
Kennedy JS. 1947. The excitant and repellent effects on mosquitoes of sub-lethal contacts with DDT. Bulletin of Entomological Research37:593–607.
Liu N and Scott JG. 1997. Inheritance of CYP6D1-mediated pyrethroid resistance in house fly (Diptera: Muscidae). Journal of Economic Entomology90:1478–1481.
Macdonald G and Davidson G. 1953. Dose and cycle of insecticide applications in the control of malaria. Bulletin of the World Health Organization9:785–812.
Maitra S, Dombrowski SM, Waters LC, Gunguly R. 1996. Three second chromosome-linked clustered CYP6 genes show differential constitutive and barbital-induced expression in DDT-resistant and susceptible strains of Drosophila melanogaster. Gene180:165–171.
Malcolm CA and Boddington RG. 1989. Malathion resistance conferred by a carboxylesterase in Anopheles culicifacies Giles (Species B) (Diptera: Culicidae). Bulletin of Entomologi cal Research79:193–199.
Martinez-Torres D, Chandre F, Williamson MS, Darriet F, Berge JB, Devonshire AL, Guillet P, Pasteur N, Pauron D. 1998. Molecular characterization of pyrethroid knockdown resistance (kdr) in the major malaria vector Anopheles gambiae s.s.Insect Molecular Biology7:179–184.
Metcalf CL, Flint WP, Metcalf RL. 1951. Destructive and Useful Insects: Their Habits and Control. New York: McGraw-Hill Book Company. P. 289.
Mouches C, Pasteur N, Berge JB, Hyrien O, Raymond M, de Saint Vincent BR, de Silvestri M, Georghiou GP. 1986. Amplification of an esterase gene is responsible for insecticide resistance in a Californian Culex mosquito. Science233:778–780.
Mourya DT, Hemingway J, Leake CJ. 1993. Changes in enzyme titres with age in four geographical strains of Aedes aegypti and their association with insecticide resistance. Medical and Veterinary Entomology7:11–16.
Mumcuoglu KY, Miller J, Uspensky I, Hemingway J, Klaus S, Ben-Ishai F, Galun R. 1995. Pyrethroid resistance in the head louse Pediculus humanus capitis from Israel. Medical and Veterinary Entomology9:427–432.
Newcomb RD, Campbell PM, Russell RJ, Oakeshott JG. 1997. cDNA cloning, baculovirus-expression and kinetic properties of the esterase, E3, involved in organophosphorus resistance in Lucilia cuprina. Insect Biochemistry and Molecular Biology27:15–25.
Penilla RP, Rodriguez AD, Hemingway J, Torres JL, Arredondo-Jimenez JI, Rodriguez MH. 1998. Resistance management strategies in malaria vector mosquito control. Baseline data for large scale field trial against Anopheles albimanus in Mexico. Medical and Veterinary Entomology12:217–233.
Pittendrigh B, Aronstein K, Zinkovsky E, Andreev O, Campbell BC, Daly J, Trowell S, ffrench-Constant RH. 1997. Cytochrome P450 genes from Helicoverpa armigera: expression in a pyrethroid-susceptible and -resistant strain. Insect Biochemistry and Mo lecular Biology27:507–512.
Prapanthadara L, Hemingway J, Ketterman AJ. 1993. Partial purification and characterization of glutathione S-transferase involved in DDT resistance from the mosquito Anoph eles gambiae. Pesticide Biochemistry and Physiology47:119–133.
Prapanthadara LA, Ranson H, Somboon P, Hemingway J. 1998. Cloning, expression and characterization of an insect class I glutathione S-transferase from Anopheles dirus species B. Insect Biochemistry and Molecular Biology28:321–329.
ProMED-mail. 2002. Dengue/DHF Update (13): 5 Apr 2002. [Online]. Available: http://www.promedmail.org/pls/askus/wwv_flow.accept [accessed July 26, 2002].
Ranson H, Collins FH, Hemingway J. 1998. The role of alternative mRNA splicing in generating heterogeneity within the Anopheles gambiae class I glutathione S-transferase family. Proceedings of the National Academy of Sciences95:14284–14289.
Raymond M, Callaghan A, Fort P, Pasteur N. 1991. Worldwide migration of amplified insecticide resistance genes in mosquitoes. Nature350:151–153.
Roberts DR and Andre RG. 1994. Insecticide resistance issues in vector-borne disease control. American Journal of Tropical Medicine and Hygiene50 (6 Suppl):21–34.
Roberts DR, Alecrim WD, Heller JM, Ehrhardt SR, Lima JB. 1982. Male Eufriesia purpurata, a DDT-collecting uglossine bee in Brazil. Nature297:62–63.
Roberts DR, Alecrim WD, Hshieh P, Grieco JP, Bangs M, Andre RG, Chareonviriphap T. 2000. A probability model of vector behavior: effects of DDT repellency, irritancy, and toxicity in malaria control. Journal of Vector Ecology25:48–61.
Roberts DR, Laughlin LL, Hshieh P, Legters LJ. 1997. DDT, global strategies, and a malaria control crisis in South America. Emerging Infectious Diseases3:295–302.
Rosner B. 1995. Fundamentals of Biostatistics. Belmont, CA: Duxbury Press. P. 56.
Rupes V, Moravec J, Chmela J, Ledvidka J, Zelenkova J. 1994. A resistance of head lice (Pediculus capitis) to permethrin in Czech Republic. Central European Journal of Public Health3:30–32.
Schoonhoven LM. 1985. Insects in a chemical world. In: Morgan ED, Mandava NB, eds. Handbook of Natural Pesticides: Volume VI: Insect Attractants and Repellents. Boca Raton, FL: CRC Press. Pp. 1–21.
Sina BJ and Aultman K. 2001. Resisting resistance. Trends in Parasitology17:305–306.
Soderlund DM and Bloomquist JR. 1989. Neurotoxic action of pyrethroid insecticides. An nual Review of Entomology34:77–96.
Tabashnik BE. 1989. Managing resistance with multiple pesticide tactics: theory, evidence and recommendations. Journal of Economic Entomology82:1263–1269.
Thompson M, Shotkoski F, ffrench-Constant RH. 1993. Cloning and sequencing of the cyclodiene insecticide resistance gene from the yellow fever mosquito Aedes aegypti. FEBS 325:187–190.
Toung YS, Hsieh T, Tu CD. 1990. Drosophila glutathione S-transferase 1-1 shares a region of sequence homology with maize glutathione S-transferase III. Proceedings of the Na tional Academy of Sciences87:31–35.
Vatandoost H, McCaffery AR, Townson H. 1996. An electrophysiological investigation of target site insensitivity mechanisms in permethrin-resistant and susceptible strains of Anopheles stephensi. Transactions of the Royal Society of Tropical Medicine and Hy giene90:216.
Vaughan A and Hemingway J. 1995. Mosquito carboxylesterase Esta21 (A2). Cloning and sequence of the full length cDNA for a major insecticide resistance gene worldwide in the mosquito Culex quinquefasciatus. Journal of Biological Chemistry270:17044– 17049.
Vaughan A, Hawkes N, Hemingway J. 1997. Co-amplification explains linkage disequilibrium of two mosquito esterase genes in insecticide-resistant Culex quinquefasciatus. Biochemical Journal325:359–365.
Vaughan A, Rodriguez M, Hemingway J. 1995. The independent gene amplification of indistinguishable esterase B electromorphs from the insecticide resistant mosquito Culex quinquefasciatus. Biochemical Journal305:651–658.
Vetter W, Scholz E, Gaus C, Muller JF, Haynes D. 2001. Anthropogenic and natural organohalogen compounds in blubber of dolphins and dugongs (Dugong dugon) from northeastern Australia. Archives of Environmental Contamination and Toxicology 41:221–231.
Vontas JG, Small GJ, Hemingway J. 2001. Glutathione S-transferases as antioxidant defence agents confer pyrethroid resistance in Nilaparvata lugens. Biochemical Journal357:65– 72.
Vulule JM, Beach RF, Atieli FK, Roberts JM, Mount DL, Mwangi RW. 1994. Reduced susceptibility of Anopheles gambiae to permethrin associated with the use of permethrin-impregnated bednets and curtains in Kenya. Medical and Veterinary Entomology8:71– 75.
WHO (World Health Organization). 1992. Vector resistance to pesticides. Fifteenth report of the Expert Committee on Vector Biology and Control. In: WHO Technical Report Series818.
WHO. 1997. Onchocerciasis Control Programme in West Africa (OCP). [Online]. Available: http://www.who.int/ocp/ocp002.htm.
Whyard S, Downe AE, Walker VK. 1995. Characterization of a novel esterase conferring insecticide resistance in the mosquito Culex tarsalis. Archives of Insect Biochemistry and Physiology29:329–342.
Williamson MS, Martinez-Torres D, Hick CA, Devonshire AL. 1996. Identification of mutations in the housefly para-type sodium channel gene associated with knockdown resistance (kdr) to pyrethroid insecticides. Molecular and General Genetics252:51–60.
WWF (World Wildlife Fund). 1999. Disease Vector Management for Public Health and Conservation. Washington, DC: WWF. Pp. 1–10.
Xiao-Ping W and Hobbs AA. 1995. Isolation and sequence analysis of a cDNA clone for a pyrethroid inducible P450 from Helicoverpa amrigera. Insect Biochemistry and Molecu lar Biology25:1001–1009.