WO2010101462A2

WO2010101462A2 - Behaviour-modifying odorant mixture for malaria mosquitoes

Info

Publication number: WO2010101462A2
Application number: PCT/NL2010/050110
Authority: WO
Inventors: Joop J.A. Van Loon
Original assignee: Wageningen Universiteit; Ifakara Health Institute
Priority date: 2009-03-05
Filing date: 2010-03-05
Publication date: 2010-09-10
Also published as: WO2010101462A3

Abstract

The invention comprises a new agent for attracting mosquitoes and the use thereof. The agent consists of a mixture comprising L-lactic acid, ammonia and octanoic acid, and at least 2 and not more than 6 further carboxylic acids as a behaviour-modifying agent for mosquitoes, wherein the further carboxylic acids are selected from the group consisting of propionic acid, butanoic acid, 3- methyl butanoic acid, pentanoic acid, heptanoic acid and tetradecanoic acid. Especially preferred is a blend comprising 5 and more preferably 2 of said carboxylic acid compounds except for 3-methyl butanoic acid. Alternatively, a mixture of comprising L-lactic acid, ammonia and tetradecanoic acid and a fourth component selected form the group consisting of 2-methylbutanoic acid, 3-methylbutanoic acid, 2-methylbutanal, 3-methylbutanal, octanal, nonanal, decanal, 1-butanol, 2-methyl-1-butanol, 3-methyl-1-butanol, 3-hydroxy-2- butanone, 2,3-butanedione, methyl-2-methylbutanoate, butylisobutyrate, butylbutyrate, butylacetate, 1-butylamine, 1-pentylamine, dimethyldisulphide and dimethylpentasulphide can be used.

Description

Title: Behaviour-modifying odorant mixture for malaria mosquitoes

The invention relates to means for insect control. More particularly, the present invention relates to compounds that have a behaviour- disrupting effect on mosquitoes for control and/or monitoring purposes.

BACKGROUND Insect control has been in great demand throughout human history.

It is necessary to control harmful insects like mosquitoes, to prevent the spread of disease, such as malaria, dengue and yellow fever. Public health authorities everywhere have expended intense effort on eliminating mosquito- related disease; however, this effort has not been wholly successful, largely because of the difficulty of eliminating mosquitoes.

Female mosquitoes seek a human host from which they obtain a blood meal for egg development. Mosquitoes locate hosts through a combination of chemicals that are characteristic of the hosts. In this host seeking, volatile organic compounds (VOCs) of human origin provide essential cues for anthropophilic mosquitoes such as Anopheles gambiae sensu stricto and Aedes aegypti and other mosquito species that bite man. Humans are differentially attractive to mosquitoes because of the odours they emit (Lindsay et al. 1993, Knols et al. 1995). Skin microflora plays an important role in the production of human body odours (Noble, W.C., 2004, "The skin microflora and microbial skin disease", Cambridge Univ. Press). If host selection by mosquitoes is based on the species composition, metabolic activity and/or density of the skin microflora, then this will bear a direct impact on the number of bites received per person and the resulting risk of infection (Takken and Knols 1999). The non-random nature of host selection remains poorly understood yet bears an important impact on exposure to disease (Smith et al. 2006). Selective control measures aimed at high-risk groups thus require a detailed understanding of differential attractiveness and the factors governing it.

A review of the research that thus far has been performed on attractants for mosquitoes is given by E.A. Rebollar-Tellez (2005, Folia Entomol. Mex. 44(2): 247). This research was further expanded by Bernier et al (2002, JAMCA 18: 186-195), Qiu et al (2006, Med.Vet.Entomol. 20(3):280287) and Logan et al. (2008, J.Chem. Ecol 34(3):308-322) . It has appeared that many VOCs show an attractant action, among which acetone, butyric acid and lactic acid seem to give the strongest responses (see e.g., US 6,267,953, DE 197 03 133, EP-O 582 915). Several compounds present in human sweat, breath and skin washings are recognized as attractive to different mosquito species (Takken and Knols 1999). The most widely researched examples include carbon dioxide, a final metabolite of vertebrate respiratory processes and ammonia, a component of human sweat (Costantini et al. 1996, Braks et al. 2001). Further, skin microflora produce carboxylic acids that were held to be responsible for an attractant effect {Knols, 1997 #19045}. It has also been shown that a mixture, in this field often indicated as 'blend', of L-lactic acid, NH3 and 12 aliphatic carboxylic acids has a synergistic effect in attracting mosquitoes compared to the isolated individual compounds (Smallegange et al. 2005).

Although many compounds and combinations of compounds are already known as mosquito attractants, a functional attractant composition which outperforms the human skin is urgently needed. SUMMARY OF THE INVENTION

The present invention relates to the use of a mixture comprising L- lactic acid, ammonia and tetradecanoic acid and a fourth component selected form the group consisting of 2-methylbutanoic acid, 3-methylbutanoic acid, 2- methylbutanal, 3-methylbutanal, octanal, nonanal, decanal, 1-butanol, 2- methyl-1-butanol, 3-methyl-l-butanol, 3-hydroxy-2-butanone, 2,3- butanedione, methyl-2-methylbutanoate, butylisobutyrate, butylbutyrate, butylacetate, 1-butylamine, 1-pentylamine, dimethyldisulphide and dimethylpentasulphide for attracting mosquitoes of the genus Anopheles gambiae.

Alternatively the invention relates to the use of a mixture comprising L-lactic acid, ammonia, octanoic acid and at least 2 and not more than 6 further aliphatic carboxylic acids as a behaviour- modifying agent for mosquitoes. More specifically, in this mixture the further carboxylic acids are selected from the group consisting of propionic acid, butanoic acid, 3-methyl butanoic acid, pentanoic acid, heptanoic acid and tetradecanoic acid. Preferably, all 6 further carboxylic acids are used. In a most preferred embodiment of the invention, the mixture contains 5, more preferably 2 of these carboxylic acids, and not 3-methyl butanoic acid.

In a preferred embodiment the mixture is evaporated in the presence of CO2.

In another embodiment, the invention relates to an agent for attracting mosquitoes comprising a mixture comprising L-lactic acid, ammonia, octanoic acid, and at least 2 and not more than 6 further carboxylic acids. Specifically, the further carboxylic acids are selected from the group consisting of propionic acid, butanoic acid, 3-methyl butanoic acid, pentanoic acid, heptanoic acid and tetradecanoic acid. Preferably, all 6 further carboxylic acids are used. In the most preferred embodiment, the agent contains 5, more preferably 2 of said further carboxylic acids, and not 3- methyl butanoic acid. Alternatively, the invention relates to an agent for attracting mosquitoes comprising a mixture comprising L-lactic acid, ammonia and tetradecanoic acid and a fourth component selected form the group consisting of 2-methylbutanoic acid, 3-methylbutanoic acid, 2- methylbutanal, 3-methylbutanal, octanal, nonanal, decanal, 1-butanol, 2- methyl- 1-butanol, 3-methyl-l-butanol, 3-hydroxy-2-butanone, 2,3- butanedione, methyl-2-methylbutanoate, butylisobutyrate, butylbutyrate, butylacetate, 1-butylamine, 1-pentylamine, dimethyldisulphide and dimethylpentasulphide.

Further, the invention relates to a method to combat mosquitoes comprising attracting them by using an agent according to the invention and subsequently inactivating them by trapping or intoxication.

LEGENDS TO THE FIGURES

Figure 1: The MM-X® trap set up. The picture on the left shows the experimental installation on one side of the screen house while the picture on the right shows the process of baiting the traps using the glass vials containing the candidate odorants.

Figure 2: The nylon strips used to deliver the blend constituents. Nylon strips have replaced the glass vials shown in Figure 1, as they cause a greater odour release area. Each strip was soaked in a solution containing the optimal concentration of a single test compound. Bl had a batch of 10 strips, while B2 had 6 strips and B3 had 9 strips. To test each blend, the respective batch was hung inside the attractant plume tube of the MM-X^® trap.

Figure 3: Set up for testing blends in the experimental huts. The left picture shows an experimental hut while the picture on the right shows how the blends were delivered from within the untreated bed nets. The automated CDC-LT is set beside the bed net to collect mosquitoes every one and half hour. Figure 4: Modeled change in EIR when traps baited with odours as attractive as blend 1 and blend 3 are used in combination with 60% insecticide net coverage.

Figure 5: Hourly collections of Anopheles gambiae in experimental hut trials using blends 1 and 2 versus human volunteers (Hl and H2). The data show the close approximation in numbers of mosquitoes collected using the blends and human beings.

DETAILED DESCRIPTION

An 'attractant' as used herein is a compound (agent) or combination of compounds (or blend or mixture) that has an attracting capacity for mosquitoes, which is able to attract mosquitoes vis-a-vis a control attractant wherein that compound is absent.

The inventors now have found that addition of a limited set of carboxylic acids to a standard blend of ammonia and lactic acid increases the attractivity of the blend for mosquitoes to levels of attractivity that are equal to human beings. The carboxylic acids that are added to achieve this effect are at least three carboxylic acids selected from the group consisting of propionic acid (C3), butanoic acid (C4), 3-methyl butanoic acid (3mC4), pentanoic acid (C5), heptanoic acid (C7), octanoic acid (C8) and tetradecanoic acid (C 14). Especially the effect of C 14 is noteworthy, since up till now this compound has not been recognized as a major attractant.

To identify the optimal concentration of each individual compound, it is preferred — as has been done in the experiments reported in the experimental section - to add an individual carboxylic acid to the basic blend in different concentrations and to identify which concentration produces the highest attractivity. This is repeated for each compound. Then, for each carboxylic acid the optimal concentration is known and this concentration can be applied in making a blend comprising several carboxylic acids.

In testing several combinations, it has been found that the blends of the present invention were as attractive to mosquitoes as humans. Such a competitive attractiveness is unique; although the behaviour-modifying action of many compounds on insects has in the meantime been established, the human skin up till now formed the unparalleled attractant for malaria mosquitoes. It will be clear that this invention will have a wide range of applications, for example baiting mosquito killing devices (including insecticides) to improve their effectiveness, diverting host-seeking malaria vectors from susceptible humans to alternative odour sources, and trapping of insects (for whatever reason).

In another embodiment, the present inventors found that addition of several compounds to a blend consisting of lactic acid, ammonia and tetradecanoic acid (C14) unexpectedly increased the attractiveness of the blend. Furthermore, it has been found that this activity is concentration specific, meaning that in many cases a higher concentration of the same addition showed not or less to be active. Anyhow, it has been found than any of the below compounds, when individually added to the blend consisting of lactic acid, ammonia and tetradecanoic acid (C 14) had a significant positive effect on the level of attraction: 2-methylbutanoic acid, 2-methylbutanol, 3- methylbutanol, octanal, 1-butanol, 2- methyl- 1-butanol, 3- methyl- 1-butanol, 3- hydroxy-2-butanone, methyl-2-methylbutanoate, butylisobutyrate, butylbutyrate, butylacetate, 1-butylamine, 1-pentylamine, dimethyl disulphide and dimethylpentasulphide. In Table 3 the specific concentrations that were used in the experimental design are indicated.

The agent of the invention can be formulated in the form of a liquid, gel or solid, provided that the attractant compounds can vaporize from said formulation. Preferably, the compounds forming the composition are dissolved in distilled water or ethanol. Additionally surface active compounds may be added.

Alternatively, the agent of the invention may be mixed with an insecticide and used as a spray and applied jointly with the release of CO2. The use of CO2 in this respect is advantageous, since this gas is a kairomone which is active over a relatively large distance. CO2 is known to activate the insects to start flying against the wind and thus increasing the chance to be attracted by the agent of the invention.

As can be seen from a field trial reported in the experimental part, two blends, the one consisting of lactic acid, ammonia, CO2 and all the mentioned carboxylic acids (C3, C4, 3mC4, C5, C7, C8 and C14), the other having only 3 carboxylic acids (C5, 3mC4 and C8), showed the same level of attractiveness as human controls. Surprisingly, when the compound 3mC4 was omitted from the first blend, the resulting blend (blend 3) showed an increase in attractiveness, thereby outperforming the human controls. This is surprising, since the compound 3mC4 seemed to contribute significantly to the performance of the first two blends and the only reason to try a blend without 3mC4 was to reduce the foul odour from the blend (and thus making it more acceptable for use). Accordingly, the invention comprises the use of any of these blends for changing the behaviour of mosquitoes, while the use of the blend consisting of lactic acid, ammonia, CO2, propionic acid, butanoic acid, pentanoic acid, heptanoic acid, octanoic acid and tetradecanoic acid (blend 3) is preferred.

The agent may be used to attract or otherwise modify the behavioural responses of mosquitoes to humans, preferably mosquitoes of the genus Anopheles, more preferably Anopheles gambiae sensu stricto, Anopheles arabiensis, Anopheles funestus, mosquitoes of the genus Aedes, more preferably Aedes aegypti or Aedes albopictus mosquitoes of the genus Culex, more preferably Culex quinquefasciatus or mosquitoes of the genus Mansonia, more preferably Mansonia africana and Mansonia uniformis. They can be used to combat the mosquitoes by applying them as attractant. The agent may be used in a trap, wherein mosquitoes are elicited to move into a trap. Such a trap can be equipped with means for catching and optionally killing the mosquitoes, such as, for example, a vacuum inlet. Such an insect trap works according to the counter-flow principle (Kline 1999a), where the odorant air stream is blown downwards and surrounded by an upwind airstream sucking the insects into the trap, whereby the inlet opening is such that the insects can not escape back into the free air. Also other known apparatuses for catching and/or killing insects like electrocution or a trap provided with sticky glue can be used, like those listed in Table 10 of US 6,267,953 or a trap whose surface is covered with sticky insect glue (for example Tanglefoot®) http://www.tanglefoot.com/products/ttcoating.htm, accessed 2 February 2009). It is also possible that insecticides or other toxic compounds for mosquitoes are located in the trap. Suitable traps are commercially available from BioQuip products, Gardena, CA, USA; John W. Hock Company, Gainesville, FIa, USA; BioSensory Inc., Putnam, CT, USA; Mills technology center, Wilimatic, CT, USA; or BioGents AG, Regensburg, Germany.

Alternatively, an insecticide may be admixed with the attracting agent, thus forming a lethal combination of attracting and killing compounds.

Insecticides that may be used in this respect are, for instance, but not limited to abamectin, acephat, acrinathrin, alanyacarb, aldicarb, allethrin, alphametrin, amitraz, avermectin, AZ 60541, azadirachtin, azamethiphos, azinphos A, azinphos B, azocyclotin, barthrin, bendiocarb, benfluracarb, bensultap, bifenthrin, bioallethrin, S-bioallethrin, bioethanomethrin, biophenothrin, bioresmethrin, BPMC, brofenprox, bromethrin, bromophos A, bufencarb, buprofezin, butocarboxin, butethrin, butylpyridaben, cadusphos, carbaryl, carbofuran, carbophenothion, carbosulfan, cartap, chloethocarb, chlorethoxyphos, chlorphenapyr, chlorphenvinphos, chlorpyriphos, chlorpyriphos M, cismethrin, clocythrin, clophentazin, cyanophos, cyclethrin, cycloprothrin, cyhalothrin, cyhexatin, cypermethrin, cyphenothrin, cyromazin, deltamethrin, demeton M, demeton S, demeton S-methyl, diafenthiuron, diazinone, dichlorfenthion, dichlorphos, dicliphos, dicrotophos, diethione, diphlubenziron, dimethoate, dimethrin, dimethylvinphos, dioxathione, disulphotone, ediphenphos, emamectin, esphenvalerate, ethiophencarbv, ethione, ethophenprox, ethoprophos, etrimphos, fipronil, fluazinam, fluazuron, flucycloxurone, flucythrinate, fluphenoxurone, fluphenprox, flumethrin, fluorethrin, fluvalinate, formothione, fubphenprox, furathiocarb, furethrin, GH-601, HCH, heptenophos, hexaflumurone, hexathiazox, imidacloprid, iprobenphos, isazophos, isophenphos, isoprocarb, isoxathione, ivermectin, lambda-cyalothrin, luphenurone, malathion, mecarbam, mervinphos, mesulphenphos, metaldehyd, methacriphos, metamidophos, methidathione, methiocarb, methomyl, metolcarb, milbemycin, monocrotophos, moxidectin, naled, NC 184, NCI-85913, nitenpyram, NRDC-105, NRDC-108, NRDC-132, NRDC-134, NRDC-139, NRDC-140, NRDC-141, NRDC-142, NRDC-146, NRDC-147, NRDC-148, NRDC-156, NRDC-157, NRDC-158, NRDC-159, NRDC-160, NRDC-163, NRDC-165, NRDC-167, NRDC-168, NRDC-169, NRDC-170, NRDC-171, NRDC-172, NRDC-173, NRDC-174, NRDC-181, NRDC-182, omethoate, oxamyl, oxydimethon m, oxydeprophos, parathion A, parathion M, permethrin, phenamiphos, phenazaquine. phenbutatinoxide, phenfluthrin, phenitrothione, phenobucarb, phenothiocarb, phenothrin, phenoxycarb, phenpyrad, phenproximate, phenthion, phenthoate, phenvalerate, phonophos, phorate, phosalone, phosmet, phosphamidone, phosthiazate, phoxim, phtalthrin, pirimicarb, pirimiphos A, pirimiphos M, PP- 321, prophenophos, promecarb, propargyl-rethrin, propaphos, proparthrin, propoxur, prothiophos, prothoate, prothrin, pymetozine, pyrachlorphos, pyresmethrin, pyrethrum, pyridaphenthione, pyridaben, pyrimidiphen, pyriproxyphen, quinalphos, resmethrin, RU-12457, RU-15525, RU-24501, S- 2852, S-5439, salathione, sebuphos, silafluophen, sulphotep, sulprophos, super- pynamine, tebuphenozide, tebuphenpyrad, tebupirimiphos, teflubenzurone, tefluthrin, temephos, terrallethrin, teralomethrin, terbam, terbuphos, tetrachlorvinphos, thiaphenox, thiodicarb, thiophanox, thiomethone, thioanazine, thuringiensin, tralomethrin, transfluthrin, transresmethrin, triarithen, triazophos, triazurone, trichlorphon, triflumurone, trimethacarb, trimethrin, vamidothione, WL- 85871, XMC, xylylcarb, Y-4042, zetamethrin, ZR-3903. Preferred insecticides are deltamethrin and permethrin.

In a specific embodiment, the behaviour-modifying agent is used in combination with a biocontrol agent, like a biocontrol agent selected from the group of entomopathogenic fungi, in particular from the group of Metarhizium anisopliae and Beauυeria bassiana (Scholte et al. 2004).

The invention will be illustrated by the following examples, which are not meant to limit the invention in any way.

EXAMPLES

Example 1

This example describes the experimental procedure and results of an experiment in which three different blends of odorants were tested on their attractiveness to anopheline mosquitoes in the presence of human volunteers under natural environmental conditions in Tanzania (Table 1 and Table 2).

Materials and methods

Mosquitoes: The semi-field assays were conducted using laboratory reared An. gambiae s.s. The larvae were fed on Tetramin^® fish food and at maintained at temperature of 27±1°C. The adult mosquitoes were kept inside mosquito cages measuring 30cm x 30cm x 30cm in a separate room, where temperatures were maintained at 27⁰C and relative humidity at 70-90%. The adults were fed on 10% glucose solution delivered through Whatman^® filter paper. The insectary was set to a photoperiod of 12 hours darkness and 12 hours light.

The semi-field system: Experiments prior to the field trials were conducted within a semi-field enclosure, also referred to as the screen house, at the Ifakara Health Institute (IHI). The screen house measures 28.8m x 21m x 4.1m and has three equal-area compartments one of which was available for this research. This system has been described in detail by (Ferguson et al. 2008).

Trapping Device: We used a counter flow geometry trap (the MM-X^® model) made by the American Biophysics Corporation to comparatively evaluate the mosquito responses elicited by the odor compounds (Kline 1999b, Njiru et al. 2006). This trap consists of an oval shaped plastic casing (the collection container) enclosing an extended inner tubing where the bait is inserted (the attractant plume tube). It has two fans blowing air in opposite directions. The smaller fan (the attractant plume fan) located directly on top of the attractant plume tube blows air out while the larger fan (the exhaust fan) which is located near the top of the trap sucks air upwards through the trap, thereby creating a counter current suction mechanism. Attracted mosquitoes trace the path of the blown-out air current, which carries the volatiles from the bait. When the insects reach near the lower end of the trap, they are sucked into the collection container by the more powerful current of the exhaust fan. At the end of the experiment, the collection tube is closed using a plastic seal after which the trap is disconnected from the 12 volt battery that powers it.

Candidate odor compounds: The test compounds were all synthetically manufactured and included carbon dioxide (CO2) gas, aqueous ammonia (NH3OH), L-Lactic acid, propionic acid (C3), butanoic acid (C4), pentanoic acid (C5), 3-methyl-butanoic acid (3mC4), heptanoic acid (C7), octanoic acid (C8) and tetradecanoic acid (C14). These compounds were prioritized at the Laboratory of Entomology, Wageningen University, The Netherlands through a series of electrophysiological and behavioural experiments. Ammonia was available as 25% aqueous solution while L-Lactic acid was available as 85% solution also prepared in water. The other carboxylic acids were available as absolute formulations with purity levels ranging between 99.7% and 99.9%. The CO2 gas was only available as industrial grade and therefore its exact purity was difficult to ascertain. All the compounds were purchased from Sigma Aldrich^® (United Kingdom), except CO2 which was supplied by Tanzania Oxygen Company Ltd^®. In these studies CO2 was added to uniformly activate the experimental mosquitoes but also as a synergist to improve the attractiveness of the other compounds (Gillies 1980, Gibson et al. 1997, Dekker et al. 2005). Distilled water was incorporated both as a solvent for ammonia, L- Lactic acid and the aliphatic carboxylic acids C3-C7, but also to raise humidity levels in the odor plumes. The other carboxylic acids C8 and C14 were dissolved in 50% ethanol.

Field study area: The field study was conducted in Lupiro Village (8.01⁰S and 36.63⁰E), Ulanga District, in the south-eastern part of Tanzania. This area lies 300 metres above sea level on the flood plains south of Kilombero Valley and is approximately 26 km south of Ifakara town, where IHI is located. The village experiences very high malaria transmission with recent entomological inoculation rate estimated at over 400 infectious bites per year (Killeen et al. 2007). The house types are mainly mud and brick walled with thatched roofs but interspersed with a few iron-sheet roofs. The village borders an active perennial swamp used by the community for rice cultivation. Annual rainfall is approximately 1200- 1800mm and annual temperature ranges between 20⁰C and 32.6⁰C. Experimental huts: The experimental huts were similar in average-dimension and shape to the local huts in the study area (S.J. Moore et al., unpublished). They had a galvanized iron frame-work and corrugated iron sheet roofs overlaid with thatch (Figure 3). The walls were made of canvas on the outside and wood panels coated with mud on the inside. Each hut had one door, two windows and open eaves all round as was the case with the local huts. Indoor temperature in the experimental huts was comparable to that in the local huts, averaging at 28° C.

Experimental procedures

Semi-field experiments

Tests involving individual odor compounds: The determination of optimal concentrations of the carboxylic acids involved iterative evaluations of their attractiveness with reference to the attractiveness of a standardized control blend. A weakly attractive primary blend (consisting 2.5% NH3OH and distilled water) was first formulated for use as control blend, and also as the basic mixture upon which other odorants were added. This blend was first tested in a competitive binary assay against natural human foot odors collected in worn nylon socks. By adding any individual compound to the primary blend we obtained a secondary blend for which the effect of changing the concentrations of the added compound was evaluated relative to the attractiveness of the primary mixture. This bubble-sort procedure required choosing on a logarithmic scale a concentration (for each candidate compound) and reducing or increasing it until the optimally attractive concentration was reached (figure 1, main text).

The solutions were provided as ImI aliquots in separate open glass vials. The vials were fitted at the lower end of the attractant plume tube of the MM-X^® traps and supported by specially fabricated vial holders. Two MM-X^® traps placed 20 metres apart (at the opposite ends inside the screen-house) were baited with either the primary blend (serving as the control) or a secondary blend (serving as the treatment). CO2 gas was concurrently delivered to both control and treatment traps at 500ml/min measured through calibrated gas flow meters (Glass Precision Engineering Ltd, UK). The MM-X® trap set up is shown in figure 2. Two hundred hungry female An. gambiae s.s. aged between 3 and 8 days old were released at the centre of the screen house and equidistant from the two traps. Initially, six-dual choice experiments where conducted in which the test trap was baited with the primary blend plus L- Lactic acid at different concentrations, iteratively selected from a logarithmic scale. The control trap on the other hand was baited with the primary blend alone. For each experiment (i.e. for each concentration) four replicates were conducted each lasting six hours. The most attractive secondary blend was determined to be the one containing the 85% L-Lactic acid. Therefore in the rest of the assays, the different concentrations of the other aliphatic carboxylic acids were all tested in combination with 85% L-Lactic acid.

Tests comparing the odor combinations with natural host odors collected on worn nylon socks: Once the optimal concentration for each carboxylic acid was determined, the compounds were all added together with the primary blend at their respective optimal concentrations to form a new blend designated Blend 1 (Bl). Bl was tested against natural foot odours collected in worn nylon socks, a highly attractive host odor source initially used by Pates et al. (2001) and first tested in the semi field system by Njiru et al (2006) against An. gambiae s.s. To collect the foot odors, the nylon socks were worn for ten hours prior to the experiment. The blend was then subjected to further refinement through a process where individual carboxylic acids were sequentially replaced with blanks and the resulting blends compared to the original. Carboxylic acids whose replacement resulted in significantly less attractive blends were then incorporated together with the primary blend to formulate another blend (B2). B2 was similarly tested against natural foot odours collected on worn nylon socks. Tests with worn nylon socks were conducted using the same binary assay procedure as in the first experiments with 500ml/min CO2 added concurrently to both traps.

Tests comparing different delivery methods for the blends: In the initial experiments, the test compounds were delivered from open glass vials as shown in figure 1. To compare the attractiveness of the blends to the worn nylon socks, we decided to use a delivery method that would provide the same surface area as the socks. Nylon fabric made of 15 denier microfibres (90% Polyamide and 10% Spandex) and having a total surface area equal to that of sock, was cut into equal sized strips (the number of strips depending on the number of blend constituents). The strips were soaked in the individual test compounds, then removed and kept at 24°C for 4 hours, so that they were semi-dry by the start of the experiment. Bl delivered from the nylon strips was tested six times against Bl delivered from open glass vials in a binary assay similar to the ones described above. We also tested Bl in nylon strips against Bl in 0.002 millimeter-thickness Low Density Polyethylene (LDPE) bags, a slow release method previously used for delivering odorants to attract tsetse flies (Torr et al. 1997). The nylon strips are shown in figure 2. We observed that more mosquitoes were trapped when the blends were delivered from soaked nylon strips than either the open glass vials or the LDPE bags. Therefore, all experiments with Bl, B2 and B3 including the field trials were conducted using the nylon strips as the delivery method for the blend constituents except CO2 gas. The strips were lcm wide and 25cm long.

Conceptual model of blend development and evaluation: The entire semi-field methodology undertaken in developing the odor blends, including the evaluation of individual carboxylic acids, is summarized in the conceptual model in figure 1. The flow processes are described in the legend.

Field experiments

Field experiments to compare the attractiveness of the synthetic blends to humans. The three blends, Bl, B2 and B3 were field-tested at village level using experimental huts. Sampling was conducted nightly in a 4 x 4 Latin square design replicated 4 times as follows: The synthetic blends (Bl and B2) and the human volunteers (Person 1 and Person 2) were allocated to the four experimental huts respectively. The human volunteers slept under untreated bed-nets in the respective huts. Similarly the blends were delivered under untreated bed-nets in the other two huts (Figure 3). To deliver the odor blends from under the bed nets, we modified MM-X® trap by closing the collection tube to eliminate suction. The blends were provided in soaked nylon strips inserted in the attractant tube of the MM-X® trap. 500ml/min CO2 was added to the trap through a rubber tubing. Unlike in the semi-field experiments, the MM-X® trapwas used here only to release the odors but not to trap the mosquitoes. Instead, automated Centers for Disease Control Light Traps (CDC-LT) were set in all the huts and programmed so that mosquitoes collected every one and half hours were stored in separate bags. The use of the same kind of trap in all huts also ensured equal trapping efficiencies. The set up was rotated every night so that at the end of the experiment each blend and each volunteer had occupied each hut four times. All the collections were done between 1900Hrs and 0700Hrs.

Field experiments to determine whether Bl can reduce human exposure to mosquito bites at household level: Finally, we tested whether our most potent synthetic lure, Bl could be used as alternative odor source to reduce people's exposure to malaria mosquitoes at household level. We introduced either Bl or an 'unbaited blank' into experimental huts in which there was a human volunteer sleeping under an untreated bed-net. The blend was released from under untreated bed net located between 3 to 5 meters away from the sleeping volunteer while the 'blanks' consisted of empty untreated bed nets at the same distance from the volunteer. Two CDC light traps were used to collect mosquitoes inside each hut; one near the human volunteer and the other near the blank or Bl. Two pairs of experimental huts were used such that in two huts the human volunteers were paired with Bl while in another two huts, the volunteers were paired with no bait. The volunteers remained in the same huts but Bl was shifted between the two paired huts every night so as to account for any spatial variations of mosquito catches. These cross-over experiments were repeated 8 times for a total of sixteen nights.

Data analysis Analysis of data from the semi-field experiments: We used version 15 of the Statistical Package for Social Sciences (SPSS Inc, Chicago) for data analyses. Optimally attractive concentrations of the candidate compounds were determined and quantified as a relative improvement on the attractiveness of the control. The preference by mosquitoes to fly to either the treatment trap or the control trap was coded as 1 and 0 respectively, and then weighted by the number of mosquitoes caught per trap. The proportion of mosquitoes caught in the treatment trap (Pt) was computed for each candidate odorant at each test concentration, taking the total number of mosquitoes collected in both traps as the denominator. Data was fitted to a binary logistic regression and Pt was estimated as a function of the categorical variables, trap location (X₁) and phase of the night (x2). The intercept obtained was exponentiated to determine the odds for the treatment compared to the control. The odds were then used to calculate the estimated probability that the mosquitoes would be trapped preferentially in the trap baited with the test blend {i.e. estimated probability = odds / (1+odds)}. The observed proportions were calculated as a ratio of mosquito catches in treatment trap (Ct) and the total of the mosquito catches in both treatment (Ct) and control traps (C_c), i.e. observed proportion = Ct / (Ct + Cc). To verify the optimum concentrations at which individual candidate compounds were most attractive, total counts of the mosquito catches were subjected to a Chi- square analysis, with the test concentration contributing the highest X² value being considered the optimum concentration. Similar logistic regression techniques were used for tests involving worn nylon socks as well as the subtraction assays.

Analysis of data from the field experiments: The field data was fitted to a

Generalized Linear Model (GLM) as follows: Mosquito catches were modeled as a function of two fixed factors, bait and hut. We treated date as a random factor to account for fluctuations in daily mosquito numbers and any resultant density dependent effect on trap efficiencies. Due to skewness, the mosquito counts were log-linked in the generalized estimating equations to make the data amenable to assumptions of the standard normal distribution. The reported trap efficiencies were therefore estimated by exponentiating the best fit model parameters for each bait. We also performed Tukey's Honestly Significant Difference test to measure the significance between the geometric means of mosquitoes collected in the huts when different baits are used.

Protection of participants

After a comprehensive explanation of the risks involved and the objectives of the study, written informed consents were obtained from the two volunteers. All participants were provided with immediate access to weekly screening for malaria parasites by light microscopy and treatment with artemether- lumefantrine. As it turned out, no participant was affected during the study. Ethical review and approval of the study was provided by the Medical Research Coordination Committee of the National Institute for Medical Research of the United Republic of Tanzania (NIMR/HQ/R.8a/Vol.IX/710). Results

From the semi-field experiments, two blends, Bl and B2 respectively, were selected as these were shown to provide the highest degree of attraction of mosquitoes to odour-baited MM-X® traps in the semi-field system. A further blend, B3, was prepared in which pentanoic acid and octanoic acid were the only two carboxylic acids, next to ammonia, lactic acid, water. As with Bl and B2, B3 was complemented by CO2 from a gas tank.

The number of mosquitoes collected depended on both the characteristics of the hut and the nature of the bait. Blend 3 caught three times more mosquitoes than humans (Figure 4b) (p<0.0001). Blend 1 and Blend 2 caught equal numbers of mosquitoes to humans (figure 6). The mosquitoes collected in blend-baited as well as human-baited huts were nearly 100% unfed. We also observed a similar pattern of mosquito-entry into the huts for both the blends and the humans with two peaks one at the beginning of the night and another at the end of the night. This means the blends attracted mosquitoes of a similar physiological state as those attracted by humans and did not modify the natural house entry behavior of the vectors, indicating their suitability as representative mosquito sampling tools. We tested whether the synthetic lures could be used as alternative odor source and to reduce people's exposure to malaria mosquitoes at household level. We introduced either Bl or a 'no-bait blank' into experimental huts in which one human volunteer was sleeping under an untreated bed-net (SOM). We measured the total mosquitoes entering the huts as well as the relative proportions that were attracted to the synthetic baits or humans (using CDC LT). In this way we measured the increase in the numbers of mosquitoes entering the huts whenever the blend was inside as well as the probability that once the mosquitoes were inside the hut, they would be attracted away from the human and towards the odor trap. We observed that introducing the blend into a hut occupied by a human volunteer increased the number of mosquitoes entering that hut by a factor of 2.6 (figure 4). Once inside the hut, An. gambiae preferentially selected the human host rather than the blend with an estimated mean of 134.9 (95% C.I. 121.7-148.0) versus 47.5 (95% C.I. 35.6- 59.4) An. gambiae per night (p <0.0001). This blend can therefore be better described as a long range attractant as opposed to a short range attractant. Using these first data on the effectiveness of blends we modeled the data to estimate the impact of such interventions in rural Tanzania in an area of stable year-round malaria transmission where An. gambiae is the primary malaria vector. Using a baseline of 60% insecticide net coverage and traps that attract 1 times (Bl and B2) or 3 times (B3) as many mosquitoes as humans as we have shown that even at a coverage of 10% odor-baited traps could have a dramatic effect on lowering malaria transmission (figure 5) .

Example 2

In a further experiments, the effects of adding individual low molecular chemical compounds, selected from different classes (carboxylic acids, aldehydes, alcohols, ketones, esters, amines and sulphur-containing compounds) to a basic attractant blend of ammonia, lactic acid and tetradecanoic acid have been established using a dual-port olfactometer, described in detail by Smallegange et al., 2005. The control port released the basic attractant blend from three separate Low Density Polyethylene (LDPE) sachets; the treatment port releases this basic blend plus the compound listed, i.e. in total 4 individual dispensers are used, one for each individual compound of which the release is mixed before outlet into the olfactometer. The use of LPDE sachets is advantageous, since this ensures a controlled and stable release of the compound(s) contained in the sachets (see a.o. Torr, 1997). The concentration of the compounds of the basic blend was as follows: Ammonia: 25% in water, released from LDPE sachets of 0.03 mm wall thickness. Lactic acid: pure compound (undiluted), released from LDPE sachets of 0.05 mm wall thickness. Tetradecanoic acid: pure compound (undiluted), released from LDPE sachets of 0.05 mm wall thickness. 'The compounds added to the basic blend are diluted either in paraffin oil or water, depending on their polarity, and released from LPDE sachets of 0.02 mm wall thickness. Three dilutions were used: a dilution of 1% (dilution factor 100), a dilution of 0.1 % (dilution factor 1000) and a dilution of 0.01% (dilution factor 10000). Of course, the applicable concentrations that are to be used may vary slightly (± 20%) with respect to the indicated concentrations.

Of course it is possible, and from a commercial point of view advantageous, to combine the components of the basic blend and the additional compound in fewer than 4 sachets, if this is allowable on basis of reactivity and solvent. It should also be understood that if a compound from either the basic blend or an additional compound in such a case will be released from a sachet with a different wall thickness, the concentration of that component in the sachet should be adjusted to be able to obtain a similar release pattern as obtained through the sachet that was used for the present experiment.

The results of these tests are depicted in Table 3. One of the interesting finds is that the activity is concentration dependent: for some of the compounds used a higher concentration would not automatically yield and increased effect. In some cases even a higher concentration meant reversal of the effect (e.g. octanal, diemethyldisulphide, butylisobutyrate) or even a repellent effect of the mixture (2,3-butanedione).

References

Braks, M. A. H., J. Meiierink, and W. Takken. 2001. The response of the malaria mosquito. Anopheles gambiae, to two components of human sweat, ammonia and L-lactic acid, in an olfactometer. Physiological

Entomology 26: 142-148. Costantini, C, G. Gibson, N. Sagnon, A. d. Torre, J. Brady, and M.

Coluzzi. 1996. Mosquito responses to carbon dioxide in a West African

Sudan savanna village. Medical and Veterinary Entomology 10: 220- 227.

Dekker, T., M. Geier, and R. T. Carde. 2005. Carbon dioxide instantly sensitizes female yellow fever mosquitoes to human skin odours.

Journal of Experimental Biology 208: 2963-2972.

Ferguson. H. M.. K. R. Ng'habi. T. Walder. D. Kadungula. S. J. Moore. I. Lyimo, T. L. Russell, H. Urassa, H. Mshinda, G. F. Killeen, and B.

G. Knols. 2008. Establishment of a large semi-field system for experimental study of African malaria vector ecology and control in

Tanzania. Malar J 7: 158.

Gibson, G., C. Constantini, F. Sagnon, A. Torre, and M. Coluzzi. 1997. The responses of Anopheles gambiae, and other mosquitoes in Burkina

Faso, to COy-the start of a search for synthetic human odour. Annals of

Tropical Medicine and Parasitology 91: 123-124. Gillies, M. T. 1980. The role of Carbondioxide in host finding by mosquitoes

(Diptera: culicidae): A review. Bulletin of Entomology 70: 525-532. Killeen. G. F.. A. Tami. J. Kihonda. F. O. Okumu. M. E. Kotas. H.

Grundmann, N. Kasigudi, H. Ngonyani, V. Mayagaya, R. Nathan,

S. Abdulla. J. D. Charlwood. T. A. Smith, and C. Lengeler. 2007.

Hybrid promotion strategies combining targeted public subsidies with private sector delivery achieve high bednet coverage and reduced malaria transmission in Kilombero Valley, Southern Tanzania. BMC

Infectious Diseases 7. Kline, D. L. 1999a. Comparison of two American mosquito traps: the professional and a new counterflow geometry trap. Journal of the

American Mosquito Control Association 15: 276-282. Kline, D. L. 1999b. Comparison of two American biophysics mosquito traps: the professional and a new counterflow geometry trap. J. Am. Mosq.

Control Assoc. Knols, B. G. J. 1996. On human odour, malaria mosquitoes, and Limburger cheese. The Lancet 348. Knols. B. G. J.. R. d. Jong, and W. Takken. 1995. Differential attractiveness of isolated humans to mosquitoes in Tanzania.

Transactions of the Royal Society of Tropical Medicine and Hygiene 89:

604-606. Lindsay, S. W., J. H. Adiamah, J. E. Miller, R. J. Pleass, and J. R.

Armstrong. 1993. Variation in attractiveness of human subjects to malaria mosquitoes (Diptera: Culicidae) in The Gambia. Journal of medical entomology 30: 368-73. Niiru. B. N.. W. R. Mukabana. W. Takken. and B. G. Knols. 2006.

Trapping of the malaria vector Anopheles gambiae with odour-baited

MM-X traps in semi-field conditions in western Kenya. Malar J 5: 39. Pates. H. V.. W. Takken. K. Stuke. and C. F. Curtis. 2001. Differential behaviour of Anopheles gambiae sensu stricto (Diptera: Culicidae) to human and cow odours in the laboratory. Bulletin of Entomological

Research 91: 289-296. Scholte. E. J.. B. G. J. Knols. R. A. Samson, and W. Takken. 2004.

Entomopathogenic fungi for mosquito control: A review - art, no. 19.

Journal of insect science 4: Nil 168- Nil 191. Smallegange, R. C, Y. T. Qiu, J. J. van-Loon, and W. Takken. 2005.

Synergism between ammonia, lactic acid and carboxylic acids as kairomones in the host-seeking behaviour of the malaria mosquito

Anopheles gambiae sensu stricto (Diptera: Culicidae). Chemical senses

30: 145-52. Smith, T., N. Maire, K. Dietz, G. F. Killeen, P. Vounatsou, L. Molineaux, and M. Tanner. 2006. Relationship between the entomologic inoculation rate and the force of infection for Plasmodium falciparum malaria. American journal of tropical medicine and hygiene 75 (2)

Suppl. S: 11-18. Takken, W., and B. G. J. Knols. 1999. Odor-mediated behavior of afrotropical malaria mosquitoes. Annual Review of Entomology 44: 131-

157. Torr. S. J.. D. R. Hall. R. J. Phelps. and G. A. Vale. 1997. Methods for dispensing odour attractants for tsetse flies (Diptera: Glossimdae). Bulletin of Entomological Research 87: 311.

Claims

1. Use of a mixture comprising L-lactic acid, ammonia and tetradecanoic acid and a fourth component selected form the group consisting of 2-methylbutanoic acid, 3-methylbutanoic acid, 2- methylbutanal, 3-methylbutanal, octanal, nonanal, decanal, 1- butanol, 2-methyl-l-butanol, 3-methyl-l-butanol, 3-hydroxy-2- butanone, 2,3-butanedione, methyl-2-methylbutanoate, butylisobutyrate, butylbutyrate, butylacetate, 1-butylamine, 1- pentylamine, dimethyldisulphide and dimethylpentasulphide as an attractant agent for mosquitoes, preferably mosquitoes of the genus Anopheles gambiae.

2. Use according to claim 1, wherein the concentration of ammonia is about 25% in water, L-lactic acid and tetradecanoic acid are pure or substantially pure compounds and the fourth component is present in a dilution of about 1%, about 0.1% or about 0.01%, in particular according to the following scheme:

2-methylbutanoic acid 1%

3-methylbutanoic acid 0.01%

2-methylbutanal 0.01%

3-methylbutanal 1% octanal 0.1 or 1% nonanal 1% decanal 1%

1-butanol 0.01%

2-methyl-l-butanol 0.01%

3- methyl- 1 -butanol 0.01%

3-hydroxy-2-butanone 0.01% 2,3-butanecαone 0.01% methyl-2 - methylbutanoate 1% butylisobutyrate 0.1% butylbutyrate 0.1 or 1% butylacetate 1%

1-butylamine 0.01%

1-pentylamine 0.01% dimethyl disulphide 0.01 or 0.1% dimethylpentasulphide 1%.

3. Use according to claim 1 or 2, wherein the compounds are released from LDPE sachets, more particularly, wherein the ammonia is released from a sachet with 0.03 mm wall thickness, L-lactic acid and tetradecanoic acid are released from a sachet with 0.05 mm thickness and the fourth component is released from a sachet with

0.02 mm thickness.

4. Use of a mixture comprising L-lactic acid, ammonia and octanoic acid, and at least 2 and not more than 6 further carboxylic acids as a behaviour-modifying agent for mosquitoes.

5. Use according to claim 4, wherein the further carboxylic acids are selected from the group consisting of propionic acid, butanoic acid, 3- methyl butanoic acid, pentanoic acid, heptanoic acid and tetradecanoic acid.

6. Use according to claim 5, wherein all 6 further carboxylic acids are used.

7. Use according to claim 5, wherein 5 further carboxylic acids are used, and not 3-methyl butanoic acid.

8. Use according to any of claims 4-7, wherein the mixture is released in the presence of CO2.

9. Use according to any of claims 4-8, wherein the mixture is an attractant for mosquitoes of the genus Anopheles gambiae.

10. Agent for attracting mosquitoes comprising a mixture comprising L- lactic acid, ammonia, octanoic acid, and at least 2 and not more than 6 further carboxylic acids or a mixture comprising L-lactic acid, ammonia and tetradecanoic acid and a fourth component selected form the group consisting of 2-methylbutanoic acid, 3- methylbutanoic acid, 2-methylbutanal, 3-methylbutanal, octanal, nonanal, decanal, 1-butanol, 2-methyl-l-butanol, 3-methyl- 1- butanol, 3-hydroxy-2-butanone, 2,3-butanedione, methyl-2- methylbutanoate, butylisobutyrate, butylbutyrate, butylacetate, 1- butylamine, 1-pentylamine, dimethyl disulphide and dimethylpentasulphide.

11. Agent according to claim 10, wherein the further carboxylic acids are selected from the group consisting of propionic acid, butanoic acid, 3-methyl butanoic acid, pentanoic acid, heptanoic acid and tetradecanoic acid.

12. Agent according to claim 10, wherein all 6 further carboxylic acids are used.

13. Agent according to claim 10, wherein 5 further carboxylic acids are used, and not 3-methyl butanoic acid.

14. Method to combat mosquitoes comprising attracting them by using an agent according to any of claims 10-13 and subsequently inactivating them by trapping or intoxication.