Immune interactions between mosquitoes and their hosts

Parasite Immunology, 2006, 28, 143–153

Review Article Mosquitoes ORIGINAL and ARTICLE host immunity Blackwell Publishing, Ltd.

Immune interactions between mosquitoes and their hosts P. F. BILLINGSLEY,1 J. BAIRD,1 J. A. MITCHELL1,3 & C. DRAKELEY2,3 1

School of Biological Sciences, University of Aberdeen, Aberdeen, UK, 2Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, UK and 3Joint Malaria Programme, Moshi, Tanzania

SUMMARY

INTRODUCTION

The intimate contact between mosquitoes and the immune system of their hosts is generally not considered important because of the transient nature of mosquito feeding. However, when hosts are exposed to many feeding mosquitoes, they develop immune responses against a range of salivary antigens. Understanding the importance of these responses will provide new tools for monitoring vector populations and identifying individuals at risk of mosquito-borne diseases, and allow the development of novel methods for monitoring control and mosquito-release programmes. Antibodies targeting the mosquito midgut are also important in the development of mosquito vaccines. The feasibility of this approach has been demonstrated and future research opportunities are considered in this review. The potential impact of mosquito vaccines is also discussed. Our understanding of the interplay between mosquitoes and the immune system of their hosts is still in its infancy, but it is clear that there is great potential for exploiting this interplay in the control of mosquito-borne diseases.

When mosquitoes ingest blood they initiate a transient but intimate two-way interaction with many different consequences. Host blood may contain pathogenic organisms, protozoans, nematodes and viruses (1,2), which can make their way from the bloodmeal to infect the fly. The host blood, apart from its obvious nutritional value, brings with it a range of factors that can remain active in the midgut after ingestion. Some of these can modulate the normal mosquito physiological responses to blood feeding by virtue of the fact that they are exogenous homologues of endogenous molecules. For example, if mammalian TGF-β is presented in the bloodmeal, it will activate mosquito-endogenous pathways modulated by a TGF-β equivalent (3). Conversely, the presence of serum proteinase inhibitors will temporarily inhibit the activity of trypsin in the midgut lumen (4). Additionally, host factors such as leucocytes and antibodies can remain active in the mosquito midgut for considerable periods (5,6). In Boophilus microplus, antibodies directed against a specific midgut molecule (Bm86) will kill the tick (7). In mosquitoes, however, the transient nature of feeding and the comparatively short digestion period have meant that the potential to harness host antibodies to work within the mosquito midgut have been largely disregarded except in the context of transmission blocking immunity against malaria parasites (8). In this review, we consider the current status of efforts to target antibodies against the mosquito, particularly against the midgut, in a drive to develop an effective mosquito vaccine. We also assess the epidemiological potential of such a vaccine and its future prospects. The immune interaction between host and mosquito is, of course, bi-directional: the feeding mosquito injects into its host a cocktail of bioactive molecules that comprise the saliva (9). Again, comparisons with ticks that are in longterm, close association with their hosts (10) have meant that the immune responses to mosquito feeding are rarely considered. However, in many natural settings the scale of mosquito feeding is difficult to comprehend. For example, in

Keywords antivector vaccine, concealed antigen, epidemiology, immune response, mosquito, saliva

Correspondence: Dr Peter F. Billingsley, School of Biological Sciences, University of Aberdeen, Tillydrone Avenue, Aberdeen AB24 2TZ, UK (e-mail: [email protected]). Received: 13 July 2005 Accepted for publication: 1 September 2005 © 2006 The Authors Journal compilation © 2006 Blackwell Publishing Ltd

143

P. F. Billingsley et al.

Parasite Immunology

Tanzania, nightly averages of approximately 3000 Anopheles gambiae plus > 500 Anopheles funestus plus large numbers of other species were recorded from single households at the height of the rainy season (11). This intensity of exposure to mosquito salivary gland secretions inevitably results in a complementary antibody response, the characterization and exploitation of which is still in its infancy. Here we also consider the antibody responses of human hosts to mosquito feeding and examine their epidemiological implications and applications.

IMMUNE RESPONSES TO MOSQUITO SALIVA Mosquito salivary glands Female mosquito paired salivary glands are localized in the thorax and their secretions facilitate the acquisition of blood from a host. Each gland is formed by two similarly constructed lateral lobes, and one shorter and wider medial lobe connected to a common salivary duct (12,13). The cannulated feeding of blood by a mosquito takes several minutes (14) and requires the repeated probing of host skin with its mouthparts until it locates and pierces a blood vessel, allowing the blood to be removed as though being drawn through a needle. The secretion of saliva by a mosquito during feeding is important for the successful location of host blood vessels and manipulation of host haemostatic and immune responses (14 –16). The feeding mosquito introduces into the host a cocktail of salivary proteins that include enzymes and inhibitory proteins (15), many of which are antigenic. Within saliva, the 5′ nucleotidase family is composed of apyrase and 5′ nucleotidase, both of which prevent the aggregation of platelets following blood vessel wall disruption (17 – 20). Apyrase can hydrolyse adenosine diphosphate (ADP) and adenosine triphosphate (ATP), both of which are important for the platelet-mediated clotting of ruptured blood vessels (14). Apyrase is also of particular importance in the probing and location of blood vessels in the host (16,17), and this is reflected in the differing amounts of activity between species (16). In Aedes aegypti, the main antiplatelet activity is because of apyrase (16,17), whereas Anopheles stephensi, Anopheles albimanus and Culex quinquefasciatus have little apyrase activity (16). Additional active molecules aid with taking blood from a host. Anopheles mosquitoes compensate for the low apyrase activity by using a potent vasodilatory salivary peroxidase and anticlotting molecule, antithrombin (9,16,21,22). Other identified antihaemostatic salivary proteins include an antifactor Xa-directed protein from the salivary glands of Ae. aegypti (23), platelet-activating factor that hydrolyses phospholipase C required for platelet aggregation from the

144

salivary glands of C. quinquefasciatus (24) and adenosine deaminase and nucleosidase that have vasodilatory and antiplatelet abilities from the salivary glands of Ae. aegypti and C. quinquefasciatus (25,26). The D7 family of proteins is the most abundant in the salivary glands of mosquitoes (27), and long and short forms have been identified in several species including An. gambiae, Anopheles arabiensis, An. stephensi, Anopheles darlingi, Ae. aegypti and C. quinquefasciatus (20,28–30). The function of these proteins is not known. However, they may bind and or carry ligands, allowing the capture of host haemostatic or inflammatory response molecules (27,28). Another suggestion is that they function as carriers or transporters, similar to the nitrophorins from the salivary glands of Rhodnius prolixus (28). These transport vesicles contain nitric oxide that is discharged into host tissues. Nitrophorins also bind histamine, speeding up the release of NO, thus achieving vasodilation, inhibition of platelet aggregation and suppressing the inflammatory effects of histamine (28). D7 molecules are highly divergent polypeptides, so it is anticipated that they will have evolved diverse biochemical properties and binding affinities (27,28).

Immune responses to mosquito saliva Host haemostatic responses consist of multiple pathways and responses (16) requiring arthropods to secrete a diverse array of active compounds and inhibitors in order to successfully ingest a bloodmeal (9). The host has several mechanisms that serve to protect against arthropod attachment and feeding. These include the host skin, haemostatic pathways; inflammatory (cellular) responses, antigen processing pathways; hypersensitivity reactions and long-lived systemic (humoral) responses (31). Mosquito salivary secretions are directly responsible for skin reactions to mosquito bites causing host hypersensitivity responses (32–34) and local irritating skin lesions, which vary from small papules to large swellings (33). The degree of host response is dependent on the duration and intensity of exposure to biting mosquitoes and the immunological profile of the host, such as its previous allergic or exposure history. In normal individuals, five stages of clinical reactivity occur against mosquito bites (35–38): Stage 1. On first contact with the mosquito, no clinical reactivity is seen. Stage 2. After repeated interactions between mosquito and host, there is a delayed type III hypersensitivity response that manifests itself as a skin wheal-and-flare response after 24 h. Stage 3. As exposure to mosquito bites continues, immediate and delayed responses are observed. Stage 3 presents as types I and III hypersensitivity responses, respectively,

© 2006 The Authors Journal compilation © 2006 Blackwell Publishing Ltd, Parasite Immunology, 28, 143–153

Volume 28, Number 4, April 2006

where skin papules are visible 10 –15 min postbite, followed by additional bite marks being evident after 12 h. Stage 4. An immediate response is generated that subsides within a couple of hours postbite. Stage 5. The host becomes tolerant to the mosquito and no clinical responses are seen. The development of each stage is exposure rather than age dependent. Thus, a child highly exposed to mosquito bites could be at a more advanced reaction stage than an adult who receives a small number of bites each year (39). Additional to the local skin wheal-and-flare reactions, latephase allergic reactions, delayed Arthus-like IgG reactions, cell-mediated delayed type III hypersensitivity and systemic host responses also occur (40). Systemic reactions include generalized skin hives, swelling and in severe cases anaphylactic shock (33,41). Reactions to mosquito bites have been compared to similar skin and hypersensitivity responses seen in asthma, pollen and bee venom allergies (39,42,43). Mosquito feeding also generates circulating antimosquito antibodies (35,44). The presence of specific antibodies can be equated to the type of skin / hypersensitivity response. People with low IgG and IgE responses have mild skin responses. Strong IgE responsive individuals have local immediate skin reactions consistent with type I hypersensitivity; such people typically have a history of allergic reactions. When there is a strong IgG response, skin reactions appear to be type III delayed hypersensitive Arthus responses, and when both IgE and IgG antimosquito responses are strong, immediate and delayed skin reactions are evident (45). Immunoblotting techniques have been used to characterize the recognition by host antimosquito antibodies of salivary antigens, demonstrating important variations associated with species and mosquito population density (39,45–49). The intensity of IgG and IgE antimosquito antibodies recognizing a number of salivary antigens increases over the mosquito season (39). The major antigens (22, 30 and 36 kDa) of Aedes communis were recognized by IgG1,4 and IgE antibodies of exposed hosts (46), and antigens of similar masses in Ae. aegypti (31, 36, 46, 64 – 66 kDa) and An. stephensi (46 kDa) were also prominent and recognized by host sera (47). Exposure of a host to a single mosquito species may also result in salivary antigens of other species being recognized by host sera, probably because of cross-reactive epitopes or antigens (49). Sera from Canadian residents exposed naturally to Aedes vexans recognized salivary antigens of Ae. vexans, Ae. aegypti and C. quinquefasciatus; both shared and species-specific antigens were present (49). However, antigens of the same molecular weights from salivary glands of different species may not necessarily share molecular identity (50). An important step forward has been the cataloguing of expressed proteins from the salivary glands of several species of mosquitoes (19,20,22,51). For An. gam-

Mosquitoes and host immunity

biae, some of these proteins have been matched against salivary antigens, and sera of naturally and experimentally exposed hosts recognized what are thought to be 5′ nucleotidase, apyrase, calreticulin, a D7-related protein, lysozyme, and several unnamed hypothetical and salivary gland proteins (52). Immunoblotting of An. gambiae salivary antigens indicated that there was no difference in the molecular weights of antigens recognized most frequently by sera collected in northern Tanzania in pre- and postmosquito seasons. However, two antigens (51 and 162 kDa) were recognized mainly by sera collected prior to the main premosquito season sera (52), suggesting that salivary markers of exposure may ultimately be identifiable. In a further refinement, sera against recombinant Ae. aegypti apyrase and a 37-kDa protein of unknown function recognized 68-kDa protein in Ae. aegypti, Ae. vexans and Anopheles albopictus, and a 37-kDa protein in Ae. aegypti, Ae. vexans and C. quinquefasciatus (48,49). Thus, the major steps have been taken in identifying salivary antigens, producing recombinants and measuring reactivity against them in exposed host sera. The next steps will be to correlate these with exposure and exploit the reactivities for epidemiological studies (see next section). Titres of antimosquito IgG and but not IgE differed between pre- and postmosquito season against Ae. vexans antigens (33). Recognition of mosquito salivary antigens by IgG and IgE was greater in areas of greater mosquito density with male host antimosquito responses positively associated with greater IgG responses (33). Length of mosquito exposure in terms of the age of a host resident in highdensity mosquito environment is thought to determine antimosquito antibody responses, e.g. high exposure to Aedes albopictus saw IgG antimosquito recognition positively correlate with age up to 2 years before levelling off (36).

EPIDEMIOLOGICAL APPLICATIONS OF ANTIMOSQUITO RESPONSES Measuring malaria transmission Measures of malaria transmission are needed not only to define malaria endemicity and risk, but also to assess interventions and monitor changes over time. Additionally, knowledge of vector species may influence the choice of control strategy. Population level measures of exposure are routinely used, yet all are subject to caveats. The gold standard measure for (mosquito–man) transmission intensity is the entomological inoculation rate (EIR) (53). This is classically derived from the density of man-biting anopheline mosquitoes, sporozoite rate and the human blood index, and represents the number of infectious bites an individual is likely to be exposed to over a defined period of time,


145


Parasite Immunology

usually 1 year (ib/p/year). However, trapping methods differ in efficiency and use may be limited because of ethical and logistical constraints. Most trapping methods use adult volunteers, but for many reasons, extrapolation to incidence in children has limitations (54). Additionally, obtaining reliable and reproducible estimates of EIR is subject to seasonal or meteorological fluctuations and, most importantly when assessing individual risk, microheterogeneity. Studies have recorded more than three log orders difference in mosquito densities between neighbouring households, demonstrating significant clustering of infected mosquitoes (55 – 57), which further affects the reliability of EIR estimates. Not unexpectedly, these caveats are particular pronounced at low transmission intensities, at altitude and areas of unstable malaria where the mosquito densities are very low and infected mosquito densities much lower still (58). Transmission estimates based on the prevalence of human infection (the parasite rate, PR) are informative as they represent actual (rather than potential) infections and are a direct measure of the disease within the community. PR has been used to define malaria endemicity and been correlated with EIR. However, PR is also susceptible to microheterogeneity caused by climatic factors and the socioeconomic determinants of health-seeking behaviour (59). Clinical correlates of transmission such as spleen rates or haemoglobin levels (Hb) that reflect longer-term trends in exposure to malaria may also be as a result of other diseases or nutrient deficiency. More recently, serological correlates of transmission intensity have been described yet these represent longterm rather than short-term exposure (60). Although incidence of disease may be the closest logical correlate of the burden of disease on health systems, it can be subject to variability between sites and may not be appropriate for evaluation of early phase studies of vector control nor will it be suitable for epidemic prediction. Obviously, exposure to an infected mosquito is a prerequisite for development, but the success of an infective bite is influenced by a variety of factors including age, body size and immunity of the host (54).

Epidemiology of salivary gland antibody responses The potential epidemiological importance of exposure of hosts to the saliva of vectors has been previously noted. Antibody responses to salivary antigens of the tick Ixodes dammini were higher in Lyme disease seropositives than seronegatives and correlated with self-reported tick exposure (10). Antibody levels peaked 3 –5 weeks after exposure and declined significantly after 3 months without exposure (61). Antibody responses to salivary gland homogenates and recombinant salivary proteins from the sand fly Lutzomyia longipalpis correlated with delayed-type hypersensitivity

146

responses to Leishmania chagasi (62). Similarly, responses to Triatoma salivary gland preparations were higher in Chagas patients than in controls, and higher in people from Triatomine-infested areas (63).

Anti-anopheline saliva responses There are fewer data on human exposure to anopheline saliva and its interaction with malaria transmission. Remoué et al. (64) have recently shown that elevated antibody responses to An. gambiae saliva correlated with increased rainfall and mosquito density. Importantly, they suggest that raised antibody responses can predict clinical cases in children. Thus, there is clear potential for the use of antisaliva antibodies in defining those most at risk, and this could be refined by combining and comparing with the standard entomological measures (65,66) and with geographical information (67). Individual antibody responses can be further correlated with demographic features and socioeconomic data to update guidelines on personal protection for example with house construction or improved mosquite net use (68). More fundamental research questions could be addressed such as the suggestion that in areas of seasonal malaria, the dramatic increase in the density of uninfected mosquito bites results in an increased infectiousness of the human reservoir to the mosquito. At the population level, the technique can be used to define risk, particularly at low-transmission intensities or in epidemic prone areas. Clear seasonal differences in antimosquito saliva antibody prevalence and titres were seen in low and moderate altitude villages in Tanzania (52), suggesting exposure-related immune responses similar to those documented for ticks. It is conceivable that a community threshold for positivity or magnitude of response could be defined for epidemic prediction, which would be far less intensive to undertake than entomological surveillance in areas where vector densities are very low (69). Like the responses to ticks, preliminary mosquito data (52) suggest that antibody titres wane after periods of nonexposure. Antibody responses to malaria antigens are markers of long-term trends in transmission (60); it is unknown as yet whether antimosquito saliva antibodies can be applied in a similar way nor how tightly titres will be correlated with exposure to single and mixed vector species. A refined antisaliva assay has the potential to evaluate vector control programmes. For example, antimosquito antibody levels in the population may have value in helping to define what coverage of insecticide treated nets (ITNs) is necessary to achieve a mass effect (70). A comparison of antibody levels in individuals without ITNs at different levels of coverage will show where maximum vector biting reduction is achieved and could provide important information for governments instigating large ITN campaigns (71).




Key further developments Major progress is being made in identifying the constituent salivary proteins of vectors of disease (19,51). Knowledge of the salivary gland proteome (or sialome) will allow production of recombinant forms of the most immunogenic proteins for epidemiological studies; large-scale production of antigen has been an obstacle to their evaluation. Existing data with salivary extracts from An. gambiae and An. arabiensis indicate that there are distinct proteins even between these sibling species, and an individual’s antibody response is highest to the most prevalent vector (52). Assays could therefore be developed to incriminate the most common vectors and monitor their seasonal contributions, but there are a vast number of variables in human antibody responses (both intensity and type), mosquito distribution and seasonality that need to be characterized to make such assays useful. Clearly, any fine-tuning of estimating vector composition from host challenge can be correlated with existing entomological measures (speciation, blood index, sporozoite rate) and allow more judicious and targeted use of interventions as high-risk individuals are identified. Similarly, post-intervention monitoring may identify species that have altered behaviour, have developed resistance or have been reintroduced to the area. Indeed, expression of a transgene in the salivary gland may itself be a useful marker for the extent of feeding by genetically modified mosquitoes.

INDUCED IMMUNITY TO MOSQUITOES When a mosquito feeds, it ingests host immune factors, both soluble and cellular, that remain active in the midgut (6). In contrast to salivary antigens, the host will not normally have been exposed to antigens from the internal organs of the mosquito, which lead to the idea of using these ‘concealed’ antigens as vaccine targets (7). The best source of concealed antigens is the midgut, because after feeding it contains the blood meal with its component immune effector molecules

and cells. The key to an effective mosquito vaccine is finding an antigen that elicits a high-titre antibody response in the host that will reduce lifespan or fecundity of the mosquito, or prevent parasite transmission from host to vector. The length of the sporogonic cycle dictates that older mosquitoes are much more likely to be carrying parasites (11), and evidence suggests that the most effective vaccines are those that would significantly reduce mosquito lifespan (72,73).

Mosquito vaccines – shotgun approaches Early efforts to find a mosquito vaccine used crude tissue homogenates to elicit polyclonal immune responses. Antibodies (in a variety of hosts) to homogenates of whole Ae. aegypti, C. tarsalis, C. quinquefasciatus, Anopheles tessellatus or An. stephensi, or to midgut and Malpighian tubules in some cases increased mosquito mortality, or reduced fecundity (74 –76). Midguts and whole-body preparations of Ae. aegypti induced high antibody titres in mice, and the increased mortality of mosquitoes that fed on them was correlated with antibody that bound to midgut microvilli (77). Similarly, antibodies to Ae. aegypti midguts reduced fecundity of the mosquitoes feeding upon immunized rabbits (78) and inhibited the transmission of two arboviruses – Ross River and Murray Valley encephalitis viruses – through the mosquito (79). Immunization with tissue homogenates of An. stephensi induced a complex antibody response (80) and again demonstrated that the midgut is an effective target for immune control of this species (81). More consistent effects have been observed when antimosquito antibodies have targeted malaria parasite passage through the vector (Table 1). Antibodies against the midgut reduced Plasmodium berghei transmission through Anopheles farauti (82), and interestingly, blocked oocyst to sporozoite transition in An. stephensi (83). Surprisingly, few attempts have been made to simplify tissue fractions prior to immunizations. Antibodies to An. tessellatus midgut glycoproteins that contain GlcNAc side

Table 1 Potential impact of mosquito vaccines on the basic reproductive number (R 0) Percent reduction Mosquito species

Plasmodium species

Antibody/antigen

m

p

b

R0

Anopheles stephensi

– Plasmodium berghei Plasmodium vivax Plasmodium vivax Plasmodium vivax

Serum against mosquito head Monoclonal MG24C against midgut Monoclonal MG4B against midgut Monoclonal MG25E against midgut Monoclonal MG25E against midgut

3 90 80 50 –

– 19 2 12 –

– 93 45 20 66

3·00 99·97 92·01 93·96 66

Anopheles gambiae

Percent reductions in m, p and b are based on data from 81, 83 and 90 and the potential effect on R 0 estimated using the Ross–MacDonald model equation 73; R0 = m · a2 · pn · b/−r · loge(p); m = bites per person per night; n = length of sporogonic cycle (days); a = human biting index; p = daily survival rate; b = susceptibility to infection; r = rate of recovery. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing Ltd, Parasite Immunology, 28, 143–153

147


Parasite Immunology

chains inhibited infectivity of the two major human malaria parasites, Plasmodium falciparum and Plasmodium vivax (84) and adversely affected the formation of the peritrophic matrix in the midgut after feeding (85), but no effects on mosquito survival were reported. Alternatively, when immunized with subcellular fractions of An. stephensi midguts, mice with high antibody titres to a microvillar preparation inhibited survival, fecundity and (modestly) P. berghei transmission (86). The immunizing fraction though, remained too complex for any meaningful analysis and this is also the case for fractions prepared by lectin affinity (87). A more direct way to test for responses to a defined antigen is by in vivo injection of DNA, which can result in the host organism expressing heterologous proteins with consequent induction of an antibody response. Direct use of DNA in this way also circumvents the need for in vitro expression and purification of recombinant proteins for vaccination. IgG responses were induced in mice injected with a midgut cDNA library of An. gambiae, and high antibody titres achieved after boosting with midgut proteins (88). Mosquitoes feeding on these mice showed reproducible reductions in survival and fecundity, but interestingly cellular rather than humoral responses seemed to be responsible for the mosquitocidal effects. Together, these studies demonstrate clearly the feasibility and potential of the mosquito vaccine approach, but they have all encountered similar problems. Even within experiments, variability can be high and reproducibility of effect difficult to achieve (81). Immunization with a complex mixture of either midgut protein extracts or a midgut cDNA library has meant that protective target antigens have not been identified. If such antigens are in low abundance, antibodies against them may not be induced consistently (89). The problems are further exacerbated by the strong immunoreactivity of carbohydrate epitopes associated with the midgut microvilli, antibodies against which seem to be consistently ineffective in terms of mosquitocidal effect. It is therefore important that single, effective antibodies and their complementary antigens are identified.

Specific antimosquito midgut antibodies Monoclonal antibodies (MAbs) against An. stephensi midgut components were tested for their bioactivity and transmission-blocking effects (90) (Table 1). Two MAbs reduced infection of P. vivax to several anopheline species, suggesting that a common ligand for parasite invasion may have been blocked. The same MAbs also reduced fecundity and survival of An. stephensi, and it is surprising therefore that the antigens targeted by these MAbs remain unidentified. Characterization of target molecules has been achieved to some degree with a monoclonal antibody (MG96) raised

148

against Ae. aegypti midgut extracts, which recognizes a glycotype composed of Manα1–6 proximal to Galβ1– 4GlcNAc-α-O-R glycans (91). This glycotype is a potential transmission-blocking vaccine candidate for vector-borne diseases and further reinforces the importance of understanding the effects of carbohydrate epitopes in development of a mosquito vaccine. A specific, mosquitocidal immune response against An. gambiae midgut mucin has been induced by immunization with the respective cDNA and recombinant protein (88). This represents an important advance in that for the first time a candidate molecule of known nucleotide sequence has been positively tested; a consideration for future studies may be to test whether combinations of cDNAs elevate the mosquitocidal response.

Phage antibodies – a rapid screening approach Phage display technology has been used to produce singlechain antibody (scAb) clones against vector antigens (92). This approach allows antigen-specific scAbs to be selected very rapidly and then tested for antimosquito activity. Clones have been selected that bind to mosquito midgut from a high-diversity phage displayed scAb library constructed from naïve human donors (93). However, the clones showed similar reactivity towards midguts of several mosquito species and on close examination were all identical, binding to a carbohydrate moiety on the luminal side of the midgut. Such anticarbohydrate scAbs may be useful in blocking Plasmodium ookinetes, which use carbohydrate residues for recognition and binding to the midgut (94), but to reduce lifespan or fecundity, it is probably more effective to target protein epitopes. Currently, a novel panning technique is being used to select for clones that recognize only protein epitopes within the mosquito midgut (Baird et al. unpublished). Traditionally, antimosquito antibodies have been tested for bioactivity by feeding them individually to cages of mosquitoes in a time-consuming process that is a potential bottleneck in the screening and testing process. This bottleneck will be exacerbated when strategies such as phage display lead to the production of many antibodies. A novel method to avoid this is single-insect feeding. Here, each well in a 96-well plate is modified to act as a membrane feeder containing blood, an antibody clone and cell line (e.g. Escherichia coli) containing the antibody gene in an expression plasmid (95). As well as being used to identify biological activity in live insects, the nucleotide sequence coding for the clone can be recovered from midguts of dead mosquitoes dissected postfeeding. This offers a relatively fast alternative to traditional testing, but does not overcome the major problems – carbohydrate reactivity and identification of low-abundance antigens.




Future prospects The development of an effective mosquito vaccine will require the identification and testing of defined antigens. DNA vaccination offers a rapid, reproducible means of testing a range of prospective target molecules, but some sort of rational screening approach is needed. A combination of genomics and gene silencing could be used to identify molecules with a critical role in parasite transmission or midgut function. If silencing a gene blocks transmission or compromises mosquito fitness, the protein coding for that gene becomes a likely candidate for vaccine development. Identifying such target molecules may not in itself lead to an effective antimosquito vaccine; any antigen will have to elicit an immune response sufficient to disrupt function, and satisfy criteria such as accessibility to host antibody on feeding, and low variability.

Figure 1 Potential impact of mosquito vaccines on the basic reproductive number (R0) for malaria. R0 estimation is repeated over several cycles, representing mosquito movements from hosts to oviposition sites. Vaccine-induced reductions in p (mosquito survival) of 5% (squares), 10% (circles) and 20% (triangles) are modelled for full coverage (filled symbols) and for 20% coverage affecting 80% of the mosquitoes in each oviposition cycle (open symbols) (97). The model does not include any new recruitment of mosquitoes to the population.

POTENTIAL APPLICATION OF MOSQUITO VACCINES Mosquito vaccines can be considered most simply as immune insecticides offering the ideal characteristics of high specificity, long-term persistence and minimal environmental impact. Thus, any potential impact predicted by models will largely match that of traditional insecticides and be analogous to treatments with bednets. The potential impact of mosquito vaccines has been examined using the Ross– MacDonald model (72,76) and a more sophisticated, individual-based model for malaria transmission (73,96,97). These models demonstrate some interesting features. Mosquito vaccines have far greater potential impact than any other vaccines being developed for malaria control because of their mosquitocidal effects, and this is particularly so when the mosquito species targeted has a strong propensity for feeding on humans (e.g. An. gambiae, An. funestus) rather than other hosts (e.g. An. arabiensis, An. stephensi). Blood stage, liver stage and even transmission-blocking vaccines have effects that are linearly proportional to their efficacies and coverage (73). In stark contrast, a mosquito vaccine of even relatively poor efficacy (72) and low coverage (73) is predicted to significantly reduce R0 (the basic reproductive number) because the mosquito survival rate during each feeding cycle is compromised (Table 1). In the model shown, percentage reductions in survival (p) would theoretically reduce R0 to below one even in intense transmission areas. In epidemiological terms, this is good news, suggesting that control efforts using mosquito vaccines should focus attention on high-risk groups. Indeed, if the spatial heterogeneity of transmission is taken into account (98), there is the interesting possibility that about 80% of the mosquitoes can be targeted by vaccinating around 20% of the human popula-

tion. Movement of mosquitoes between the feeding and oviposition sites (99) also means that, at any given time, a large proportion of the hungry adult population will be found aggregated into a few households. Thus, on each round of feeding, this large proportion of mosquitoes would be feeding on a small proportion of hosts, and if these few hosts were vaccinated, they would have a major impact on malaria transmission (Figure 1). Control approaches targeting households with high mosquito densities have been advocated previously (e.g. 11), but are an aspect of vector control that has remained largely untested. As mentioned previously, integrating antimosquito saliva assays with data from geographical information systems (GIS) may help define why some individuals are prone to high levels of exposure and how best to target this group. The true value of mosquito vaccines may lie in being used as part of a more integrated control approach. Reduced vector survival contributes significantly to local extinction of malaria parasites, and enhances the success of parasite control strategies (100). The additional pressure on the parasite to complete its life cycle in the presence of antimosquito vaccines would be indirect, as these act against the vector, but would complement the direct effects imposed by vaccines against liver, blood and sexual stages of Plasmodium. The environmental issues related to mosquito vaccines are negligible and concern only the downstream effects of potentially altering the ecological balance as mosquito populations are reduced. However, given the acknowledged plastic behavioural and physiological adaptations that are regularly demonstrated by mosquitoes, it is likely that there would be selective pressure towards feeding on alternative


149


Parasite Immunology

hosts. This in turn might have mixed outcomes; the presence of cattle has been considered a risk factor for malaria by increasing the population densities of some vector species (101). However, modelling indicates that numbers of cattle may have less influence on malaria transmission than previously thought and that it is the extent of human–mosquito contact and mosquito survival that drive endemic and epidemic transmission (102). Consequently, although mosquito vaccines may conceivably select for feeding behaviour on nonhuman hosts, the switch in behaviour ought not to increase malaria transmission. Most of the laboratory-based studies reviewed here have performed immunization and feeding efficacy studies on a single species of mosquito at a time; the degree of specificity of any antimosquito antibodies remains largely untested. However, there is demonstrable potential for multispecies effects of mosquito vaccines. Although cross-reactivity and bioactivity of antisera did not bridge the Anopheles and Culex species gap (103), there are antimosquito MAbs that reduce significantly the transmission of P. vivax by An. stephensi, An. freeborni, An. albimanus and An. farauti, and reduce fecundity and survival of An. stephensi (104). In practically every malaria epidemiological setting, more than one vector species is responsible for transmission, sometimes of more than one parasite. There is the intriguing possibility that a mosquito vaccine targeting multiple species of mosquito and reducing transmission of more than one species of malaria might be developed. Antigenic cross-reactivity does occur between mosquito species (105), but the nature of such cross-reactivities remains unclear. The limited carbohydrate repertoire of the midgut cell surfaces noted earlier (87) is a good source of common epitopes, antibodies against which would most likely affect transmission rather than mosquito longevity (84,91). Another similar important epidemiological consideration is that we know little about the degree of polymorphism in putative antigens or functional molecules of mosquitoes in natural populations. Genes associated with the mosquito immune system show high frequencies of single nucleotide polymorphisms (SNPs) (106), but these do not necessarily all result in changes to the protein structure. Studies on protein/ antigenic variation in natural populations will be important once protective mosquito antigens have been identified. One problem that mosquito vaccines share with the malaria zygote/ookinete stage vaccines is the absence of natural boosting. In both cases, the vaccines would be effective only for a limited period and boosting would need to be by re-immunization. This is a major problem, but if there are any epitopes shared between saliva and the midgut, sufficient natural boosting could occur (see Nuttall et al., this issue), and this will happen in the most exposed section of the population. It is worthwhile, therefore, to examine if any

150

of the salivary gland antigens are protective if presented in the appropriate way. Although field data suggest not, anecdotal observations from rearing mosquito colonies suggest that rodents subjected to high mosquito exposure will eventually cause poor survival and fecundity in mosquitoes fed upon them. The immune interactions between mosquitoes and hosts, especially human hosts, in natural disease transmission settings, remains largely unexplored. The completion of the An. gambiae genome (107) and the near completion of the genome sequences of other mosquito species offer opportunities for more rapid advancement. Midgut and salivary antigens can now be readily identified, and their efficacy in terms of antimosquito effects or as epidemiological markers screened reasonably rapidly. Clearly, monitoring antibody responses to saliva will be of limited epidemiological use in places where mosquito populations are extremely high, whereas the converse is true for mosquito vaccines. The study of mosquito–host interactions links vaccinology, malariology and traditional entomology; the results to date strongly support a sustained investment in more research.

ACKNOWLEDGEMENTS This work was supported in part by the European Commission, NERC and the Bill and Melinda Gates Foundation. The Joint Malaria Programme is a collaboration among the National Institute for Medical Research, Kilimanjaro Christian Medical College, the London School of Hygiene & Tropical Medicine and the Centre for Medical Parasitology; University of Copenhagen. C. J. D. is supported by a Wellcome Trust Research Training Fellowship.

REFERENCES 1 Billingsley PF & Sinden RE. Determinants of malariamosquito specificity. Parasitol Today 1997; 13: 297 – 301. 2 Vaughan JA, Trpis M & Turell MJ. Brugia malayi microfilariae (Nematoda: Filaridae) enhance the infectivity of Venezuelan equine encephalitis virus to Aedes mosquitoes (Diptera: Culicidae). J Med Entomol 1999; 36: 758 –763. 3 Lieber MJ & Luckhart S. Transforming growth factor-betas and related gene products in mosquito vectors of human malaria parasites: signaling architecture for immunological cross-talk. Mol Immunol 2004; 41: 965 – 977. 4 Borovsky D. Proteolytic enzymes and blood digestion in the mosquito, Culex nigripalpus. Arch Insect Biochem Physiol 1986; 3: 147 –160. 5 Sinden RE & Smalley ME. Gametocytes of Plasmodium falciparum: phagocytosis by leucocytes in vivo and in vitro. Trans R Soc Trop Med Hyg 1976; 70: 344 – 345. 6 Vaughan JA, Wirtz RA, do Rosario VE & Azad AF. Quantitation of antisporozoite immunoglobulins in the haemolymph of Anopheles stephensi after bloodfeeding. Am J Trop Med Hyg 1990; 42: 10 –16.



7 Willadsen P, Eisemann CH & Tellam RL. ‘Concealed’ antigens: expanding the range of immunological targets. Parasitol Today 1993; 9: 132 –135. 8 Kaslow DC. Transmission-blocking vaccines: uses and current status of development. Int J Parasitol 1997; 27: 183 –189. 9 Champagne DE & Valenzuela JG. Pharmacology of haematophagous arthropod saliva. In The Immunology of HostEctoparasitic Arthropod Relationships, ed. Wikel SK. Wallingford: CAB International; 1996: 85 –106. 10 Schwartz BS, Ribeiro JM & Goldstein MD. Anti-tick antibodies: an epidemiologic tool in Lyme disease research. Am J Epidemiol 1990; 132: 58–66. 11 Charlwood JD, Kihonda J, Sama S et al. The rise and fall of Anopheles arabiensis (Diptera: Culicidae) in a Tanzanian village. Bull Entomol Res 1995; 85: 37 – 44. 12 Clements AN. The Biology of Mosquitoes. London: Chapman and Hall; 1992. 13 Sauer JR, Bowman AS, McSwain JL & Essenberg RC. Salivary gland physiology of blood-feeding arthropods. In The Immunology of Host-Ectoparasitic Arthropod Relationships, ed. Wikel SK. Wallingford: CAB International; 1996: 62 – 84. 14 Ribeiro JMC, Rossignol PA & Spielman A. Role of mosquito saliva in blood vessel location. J Exp Biol 1984; 108: 1– 7. 15 Ribeiro JMC. Role of saliva in blood-feeding by arthropods. Ann Rev Entomol 1987; 32: 463 – 478. 16 Ribeiro JMC. Blood-feeding in mosquitoes: probing time and salivary gland anti-haemostatic activities in representatives of three genera (Aedes, Anopheles, Culex). Med Vet Entomol 2000; 14: 142–148. 17 Champagne DE, Smartt CT, Ribeiro JMC & James AA. The salivary gland-specific apyrase of the mosquito Aedes aegypti is a member of the 5′-nucleotidase family. PNAS 1995; 92: 694–698. 18 Ribeiro JMC & Valenzuela JG. The salivary purine nucleosidase of the mosquito, Aedes aegypti. Insect Biochem Mol Biol 2003; 33: 13–22. 19 Valenzuela JG, Pham VM, Garfield MK, Francischetti IMB & Ribeiro JMC. Toward a description of the sialome of the adult female mosquito Aedes aegypti. Insect Biochem Mol Biol 2002; 32: 1101–1122. 20 Valenzuela JG, Francischetti IMB, Pham VM, Garfield MK & Ribeiro JMC. Exploring the salivary gland transcriptome and proteome of the Anopheles stephensi mosquito. Insect Biochem Mol Biol 2003; 33: 717 –732. 21 Waidhet-Kouadio P, Yuda M, Ando K & Chinzei Y. Purification and characterization of a thrombin inhibitor from the salivary glands of a malarial vector mosquito, Anopheles stephensi. Biochim Biophys Acta 1998; 1381: 227 –233. 22 Valenzuela JG, Francischetti IMB & Ribeiro JMC. Purification, cloning, and synthesis of a novel salivary anti-thrombin from the mosquito Anopheles albimanus. Biochemistry 1999; 38: 11209–11215. 23 Stark KR & James AA. Isolation and characterization of the gene encoding a novel factor Xa-directed anticoagulant from the yellow fever mosquito, Aedes aegypti. J Biol Chem 1998; 273: 20802–20809. 24 Ribeiro JMC & Francischetti IMB. Platelet-activating-factorhydrolyzing phospholipase C in the salivary glands and saliva of the mosquito Culex quinquefasciatus. J Exp Biol 2001; 204: 3887–3894. 25 Ribeiro JMC, Charlab R & Valenzuela JG. The salivary adenosine deaminase activity of the mosquito Culex quinquefasciatus and Aedes aegypti. J Exp Biol 2001; 204: 2001 –2010.


26 Ribeiro JMC & Valenzuela JG. The salivary purine nucleosidase of the mosquito, Aedes aegypti. Insect Biochem Mol Biol 2003; 33: 13 – 22. 27 Valenzuela JG, Charlab R, Gonzalez EC et al. The D7 family of salivary gland proteins in blood-sucking Diptera. Insect Mol Biol 2002; 11: 149 –155. 28 Arca B, Lombardo F, Lanfrancotti A et al. A cluster of four D7-related genes is expressed in the salivary glands of the African malaria vector Anopheles gambiae. Insect Mol Biol 2002; 11: 47 –55. 29 Calvo E, deBianchi AG, James AA & Marinotti O. The major acid soluble proteins of adult female Anopheles darlingi salivary glands include a member of the D7-related family of proteins. Insect Biochem Mol Biol 2002; 32: 1419 –1427. 30 Malafronte RDS, Calvo E, James AA & Marinotti O. The major salivary gland antigens of Culex quinquefasciatus are D7-related proteins. Insect Biochem Mol Biol 2003; 33: 63–71. 31 Wikel SK. Immunology of the skin. In The Immunology of Host-Ectoparasitic Arthropod Relationships, ed. Wikel SK. Wallingford: CAB International; 1996: 1 – 29. 32 Hudson AL, Bowman AL & Orr CWM. Effect of absence of saliva on blood feeding by mosquitoes. Science 1960; 131: 1730 –1731. 33 Peng ZN, Rasic N, Liu Y & Simons FER. Mosquito salivaspecific IgE and IgG antibodies in 1059 blood donors. J Allergy Clin Immunol 2002; 110: 1 –5. 34 Sandeman RM. Immune responses to mosquitoes and flies. In The Immunology of Host-Ectoparasitic Arthropods Relationships, ed. Wikel SK. Wallingford: CAB International; 1996. 35 Cabrera R, Guarda R & Gonzalez S. Parasitic Infections. In Skin Immune System (SIS). London: CRC Press; 1997: 605– 616. 36 Konishi E. Distribution of immunoglobulin G and E antibody levels to salivary gland extracts of Aedes albopictus (Diptera: Culicidae) in several age groups of Japanese population. J Med Entomol 1990; 27: 519 – 522. 37 McKiel JA & West AS. Effects of repeated exposures of hypersensitive humans and laboratory rabbits to mosquito antigens. Can J Zoo 1961; 39: 597 – 603. 38 Mellanby K. Man’s reaction to mosquito bites. Acta Allergol 1946; 158: 554. 39 Palosuo K, Brummer-Korvenkontio H, Mikkola J, Sahi T & Reunala T. Seasonal increase in human IgE and IgG4 antisaliva antibodies to Aedes mosquito bites. Allerg Immunol 1997; 114: 367 –372. 40 Wikel SK. Immune responses to arthropods and their products. Ann Rev Entomol 1982; 27: 21– 48. 41 McCormack DR, Salatu KF & Hershey JN. Mosquito bite anaphylaxis: immunotherapy with whole body extracts. Ann Allergy Asthma Immunol 1995; 74: 39 – 44. 42 Gluck JC & Palosuo T. Asthma from mosquito bites: a case report. Ann Allergy Asthma Immunol 1986; 56: 492 – 493. 43 Reunala T, Brummer-Korvenkontio H, PalosuoK et al. Frequent occurrence of IgE and IgG4 antibodies against saliva of Aedes communis and Aedes aegypti mosquitoes in children. Int Arch Allergy Immunol 1994; 104: 366 – 371. 44 ReunalaT, Brummer-Korvenkontio H & Palosuo T. Are we really allergic to mosquito bites? Ann Allergy 1990; 56: 492– 493. 45 Shen H-D, Chen C-C, Chang H-N & Chang L-Y. Human IgE and IgG antibodies to mosquito proteins detected by the immunoblot technique. Ann Allergy 1989; 63: 143 – 146. 46 Brummer-Korvenkontio H, Lappalainen P, Reunala T & Palosuo T. Detection of mosquito saliva-specific IgE and IgG4


151


47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

152

Parasite Immunology

antibodies by immunoblotting. J Allergy Clin Immunol 1994; 93: 551–555. Brummer-Korvenkontio H, Palosuo T, Francois G & Reunala T. Characterization of Aedes communis, Aedes aegypti and Anopheles stephensi mosquito saliva antigens by immunoblotting. Allergy Immunol 1997; 112: 169 –174. Peng Z & Simons FER. Cross-reactivity of skin and serum specific IgE responses and allergen analysis for three mosquito species with worldwide distribution. J Allergy Clin Immunol 1997; 100: 192–198. Peng Z, Li H & Simons FER. Immunoblot analysis of salivary allergens in 10 mosquito species with worldwide distribution and the human IgE responses to these allergens. J Allergy Clin Immunol 1998; 101: 498 – 505. Peng Z, Li H & Simons FER. Extracts of Aedes vexans, Culiseta inornata and Culex tarsalis. Arch Allergy Immunol 1996; 110: 46–51. Francischetti IMB, Valenzuela JG, Pham VM, Garfield MK & Ribeiro JMC. Toward a catalogue for the transcripts and proteins (sialome) from the salivary gland of the malaria vector Anopheles gambiae. J Exp Biol 2002; 205: 2429 – 2451. Mitchell JA. Antibody responses to ectoparasitic arthropods as an indication of exposure history and disease risk. PhD Thesis. Aberdeen: University of Aberdeen; 2005. Hay SI, Rogers DJ, Toomer JF & Snow RW. Annual Plasmodium falciparum entomological inoculation rates (EIR) across Africa: literature survey, Internet access and review. Trans R Soc Trop Med Hyg 2000; 94: 113 –127. Smith T, Killeen G, Lengeler C & Tanner M. Relationships between the outcome of Plasmodium falciparum infection and the intensity of transmission in Africa. Am J Trop Med Hyg 2004; 71: 80–86. Drakeley CJ, Schellenberg D, Kihonda J et al. An estimation of the entomological inoculation rate for Ifakara: a semi-urban area in a region of intense malaria transmission in Tanzania. Trop Med Int Health 2003; 8: 767 –774. Charlwood JD, Smith T, Lyimo E et al. Incidence of Plasmodium falciparum infection in infants in relation to exposure to sporozoite-infected anophelines. Am J Trop Med 1998; 59: 243–251. Mbogo CN, Snow RW, Khamala CP et al. Relationships between Plasmodium falciparum transmission by vector populations and the incidence of severe disease at nine sites on the Kenyan coast. Am J Trop Med Hyg 1995; 52: 201 –206. Beier JC, Killeen GF & Githure JI. Short report: entomologic inoculation rates and Plasmodium falciparum malaria prevalence in Africa. Am J Trop Med Hyg 1999; 61: 109 –113. Drakeley C, Carneiro I, Reyburn H et al. Altitude-dependent and altitude-independent variations in Plasmodium falciparum prevalence in north-eastern Tanzania. J Infect Dis 2005; 191: 1589–1598. Drakeley C, Corran PH, Coleman PG et al. Estimating medium and long-term trends in malaria transmission using serological markers of malaria exposure. PNAS 2005; 102: 5108–5113. Schwartz BS, Ford DP, Childs JE, Rothman N & Thomas RJ. Anti-tick saliva antibody: a biologic marker of tick exposure that is a risk factor for Lyme disease seropositivity. Am J Epidemiol 1991; 134: 86 – 95. Barral A, Honda E, Caldas A et al. Human immune response to sand fly salivary gland antigens: a useful epidemiological marker? Am J Trop Med Hyg 2000; 62: 740 – 745.

63 Nascimento RJ, Santana JM, Lozzi SP, Araujo CN & Teixeira AR. Human IgG1 and IgG4: the main antibodies against Triatoma infestans (Hemiptera: Reduviidae) salivary gland proteins. Am J Trop Med Hyg 2001; 65: 219 – 226. 64 Remoué F, Cisse B, Ba F et al. Evaluation of the antibody response to Anopheles salivary antigens as a potential marker of risk of malaria? Trans Roy Soc Trop Med Hyg 2005; in press. 65 Lindblade KA, Walker ED & Wilson ML. Early warning of malaria epidemics in African highlands using Anopheles (Diptera: Culicidae) indoor resting density. J Med Entomol 2000; 37: 664 – 674. 66 Trape JF, Lefebvre-Zante E, Legros F et al. Vector density gradients and the epidemiology of urban malaria in Dakar, Senegal. Am J Trop Med Hyg 1992; 47: 181 –189. 67 Staedke SG, Nottingham EW, Cox J, Kamya MR, Rosenthal PJ & Dorsey G. Short report: proximity to mosquito breeding sites as a risk factor for clinical malaria episodes in an urban cohort of Ugandan children. Am J Trop Med Hyg 2003; 69: 244 – 216. 68 Lindsay SW, Jawara M, Paine K, Pinder M, Walraven GE & Emerson PM. Changes in house design reduce exposure to malaria mosquitoes. Trop Med Int Health 2003; 8: 512–517. 69 Bodker R, Akida J, Shayo D et al. Relationship between altitude and intensity of malaria transmission in the Usambara Mountains, Tanzania. J Med Entomol 2003; 40: 706 –717. 70 Hawley WA, Phillips-Howard PA, ter Kuile FO et al. Communitywide effects of permethrin-treated bednets on child mortality and malaria morbidity in western Kenya. Am J Trop Med Hyg 2003; 68: 121 –127. 71 Curtis C, Maxwell C, Lemnge M et al. Scaling-up coverage with insecticide-treated nets against malaria in Africa: who should pay? Lancet Infect Dis 2003; 3: 304 – 307. 72 Billingsley PF. Approaches to vector control: new and trusted. Molecular targets in the insect midgut. Trans Roy Soc Trop Med Hyg 1994; 88: 136 – 140. 73 Foy BD, Killeen GF, Magalhaes T & Beier JC. Immunological targeting of critical insect antigens. Am Entomol 2002; 48: 150 –163. 74 Alger NE & Cabrera EJ. An increase in death rate of Anopheles stephensi fed on rabbits immunised with mosquito antigens. J Econ Entomol 1972; 65: 165 –168. 75 Sutherland GB & Ewan AB. Fecundity decrease in mosquitoes ingesting blood from specifically sensitised animals. J Insect Phys 1974; 20: 655 – 660. 76 Ramasamy MS, Srikrishnaraj KA, Wijekoone S, Jesuthasan LSB & Ramasamy R. Host immunity in mosquitoes: effect of antimosquito antibodies on Anopheles tessellatus and Culex quinquefasciatus (Diptera: Culicidae). J Med Entomol 1992; 29: 934 –938. 77 Hatfield P. Anti-mosquito antibodies and their effects on feeding, fecundity and mortality of Aedes aegypti. Med Vet Entomol 1988; 2: 331 –338. 78 Ramasamy MS, Ramasamy R, Kay BH & Kidson C. Antimosquito antibodies decrease the reproductive capacity of Aedes aegypti. Med Vet Entomol 1988; 2: 87 – 93. 79 Ramasamy MS, Sands M, Kay BH, Fanning ID, Lawrence GW & Ramasamy R. Anti-mosquito antibodies reduce the susceptibility of Aedes aegypti to arbovirus infection. Med Vet Entomol 1990; 4: 40 – 55. 80 Almeida APG & Billingsley PF. Induced immunity against the mosquito Anopheles stephensi Liston (Diptera: Culicidae):



81

82

83

84

85

86

87

88

89

90

91

92

93

reactivity and characteristics of immune sera. Med Vet Entomol 1999; 13: 53–64. Almeida APG & Billingsley PF. Induced immunity against the mosquito Anopheles stephensi Liston (Diptera: Culicidae): effects on mosquito survival and fecundity. Int J Parasitol 1998; 28: 1721–1731. Ramasamy MS & Ramasamy R. Effect of anti-mosquito antibodies on the infectivity of the rodent malaria parasite Plasmodium berghei to Anopheles farauti. Med Vet Entomol 1990; 4: 161–166. Lal AA, Schriefer ME, Sacci JB et al. Inhibition of malaria parasite development in mosquito-midgut antibodies. Infect Immun 1994; 62: 316 – 318. Ramasamy R, Wanniarachchi IC, Srikrishnaraj KA & Ramasamy MS. Mosquito midgut glycoproteins and recognition sites for malaria parasites. Biochim Biophys Acta 1997; 1361: 114–122. Ramasamy MS, Rschid L, Srikrishnaraj KA & Ramasamy R. Antimidgut antibodies inhibit peritrophic membrane formation in the posterior midgut of Anopheles tessellates (Diptera: Culicidae). J Med Entomol 1996; 33: 162 – 164. Almeida APG & Billingsley PF. Induced immunity against the mosquito Anopheles stephensi: effects of cell fraction antigens on longevity, fecundity and Plasmodium berghei transmission. J Med Entomol 2002; 39: 207 – 214. Wilkins S & Billingsley PF. Oligosaccharides on midgut microvillar glycoproteins of the mosquito, Anopheles stephensi Liston. Insect Biochem Mol Biol 2001; 31: 937 – 948. Foy BD, Magalhaes T, Injera WE et al. Induction of mosquitocidal activity in mice immunized with Anopheles gambiae midgut cDNA. Infect Immun 2003; 71: 2032 – 2040. Jacobs Lorena M & Lemos FJA. Immunological strategies for the control of insect disease vectors: a critical assessment. Parasitol Today 1995; 11: 144 – 147. Lal AA, Patterson PS, Sacci JB et al. Anti-mosquito midgut antibodies block development of Plasmodium falciparum and Plasmodium vivax in multiple species of Anopheles mosquitoes and reduce vector fecundity and survivorship. PNAS 2001; 98: 5228–5223. Dinglasan RR, Fields I, Shahabuddin M, Azad AF & Sacci JB Jr. Monoclonal antibody MG96 completely blocks Plasmodium yoelii development in Anopheles stephensi. Infect Immun 2001; 71: 6995–7001. Willadsen P, Billingsley PF. Immune intervention against blood-feeding insects. In The Biology of the Insect Midgut, eds Lehane MJ, Billingsley PF. London: Chapman and Hall; 1999: 323–344. Foy BD, Killeen GF, Frohn RH, Impoinvil D, Williams A & Beier JC. Characterization of a unique human single-chain antibody isolated by phage-display selection on membranebound mosquito midgut antigens. J Immunol Methods 2002; 261: 73–83.


94 Zieler H, Nawrocki JP & Shahabuddin M. Plasmodium gallinaceum ookinetes adhere specifically to the midgut epithelium of Aedes aegypti by interaction with a carbohydrate ligand. J Exp Biol 1999; 202: 485 – 495. 95 Killeen GF, Foy BD, Shahabuddin M et al. Tagging bloodmeals with phagemids allows feeding of multiple-sample arrays to single cages of mosquitoes (Diptera: Culicidae) and the recovery of single recombinant antibody fragment genes from individual insects. J Med Entomol 2000; 37: 528 – 533. 96 Killeen GF, McKenzie FE, Foy BD, Billingsley PF & Beier JC. A simplified model for predicting entomological inoculation rates from vector population characteristics and infectious reservoir size. Am J Trop Med Hyg 2000; 62: 535 – 544. 97 Killeen GF, McKenzie FE, Foy BD, Billingsley PF & Beier JC. The potential impacts of integrated malaria transmission control on entomological inoculation rate in highly endemic areas. Am J Trop Med Hyg 2000; 62: 545 – 551. 98 Woolhouse MEJ, Dye C, Etard JF et al. Heterogeneities in the transmission of infectious agents: implications for the design of control programs. PNAS 1997; 94: 338 – 342. 99 Le Menach A, McKenzie FE, Flahault A & Smith DL. The unexpected importance of mosquito oviposition behaviour for malaria: non-productive larval habitats can be sources for malaria transmission. Malar J 2005; 4: 23. 100 McKenzie FE, Baird JK, Beier JC, Lal AA & Bossert WH. A biologic basis for integrated malaria control. Am J Trop Med Hyg 2002; 67: 571 –577. 101 Bouma M & Rowland M. Failure of passive zooprophylaxis: cattle ownership in Pakistan is associated with a higher prevalence of malaria. Trans R Soc Trop Med Hyg 1995; 89: 351 – 353. 102 Saul A. Zooprophylaxis or zoopotentiation: the outcome of introducing animals on vector transmission is highly dependent on the mosquito mortality while searching. Malar J 2003; 2: 32. 103 Ramasamy R, Nadesalingam P & Ramasamy MS. Antigenic similarity between the mosquito vectors of malaria and filariasis. J Med Entomol 1991; 28: 760 – 762. 104 Lal AA, Patterson PS, Sacci JB et al. Anti-mosquito midgut antibodies block development of Plasmodium falciparum and Plasmodium vivax in multiple species of Anopheles mosquitoes and reduce vector fecundity and survivorship. PNAS 2001; 98: 5228 – 5233. 105 Moorthy SAV, Ramasamy R & Ramasamy MS. Antigenic relationships between adult and larval Anopheles tessellatus midgut glycoproteins and the midguts of other vector mosquitoes. Med Vet Entomol 2003; 17: 26 – 32. 106 Morlais I, Poncon N, Simard F, Cohuet & Fontenille D. Intraspecific nucleotide variation in Anopheles gambiae: new insights into the biology of malaria vectors. Am J Trop Med Hyg 2004; 71: 795 – 802. 107 Holt RA, Subramanian GM, Halpern A et al. The genome sequence of the malaria mosquito Anopheles gambiae. Science 2002; 298: 129 –149.


153