Skip to main content

European Network for Advanced Research on Olfaction for Malaria Transmitting Insect Control

Final Report Summary - ENAROMATIC (European network for advanced research on olfaction for malaria transmitting insect control)

Executive summary:

Malaria, the deadliest of the known insect-born diseases, is transmitted to humans by Anopheline mosquitoes. The goal of ENAROMATIC has been to discover compounds of natural or synthetic origin capable of interfering with the olfactory function of mosquitoes and their capacity to detect the presence of humans in their environment and transmit malaria parasites through blood meals.

Toward this objective, the consortium has undertaken a series of studies and has achieved the main objectives listed in the research plan as follows.

1. Several odorant binding proteins (OBPs) and odorant receptors (ORs) of Anopheles gambiae, which participate in the process of human host odor recognition, have been cloned, expressed in bacteria and/or insect cells and used in screening and crystallization studies.

2. Protein (for OBPs) and cell-based (for ORs) screening assays allowing fast analyses of synthetic and natural (aromatic plant-derived) compound collections for the presence of olfaction-based behavior modifiers have been established and relevant screens were carried out.

3. To enhance the effectiveness of the screening process, specific reagents and methods were developed, which allowed the establishment of co-expression patterns for a number of the selected OBPs and ORs in female antennal sensilla. Because co-expression of OBPs and ORs recognizing common ligands is prerequisite for efficient odor recognition, co-expressed proteins received priority in the screening efforts.

4. The capacity of essential oils (EOs), selected by OBP in vitro screens, to trigger changes in female A. gambiae behaviour was assessed by in vivo behavioural and electrophysiological assays and a total of 35 EOs were found to have significant repellent activities. Nine EO constituent compounds and derivatives having strong repellent properties were also identified. The precise repellent effects of the five most active compounds and binary and ternary mixtures thereof have been established.

5. The crystallization of several OBPs and structure determination of five ligand-bound OBPs served as the basis for modelling studies, which resulted in predictions for compounds, including derivatives of the identified repellents, with better OBP binding potential. Of twenty such synthetic compounds that were assessed for binding to specific OBPs and in vivo functionality, five were found to have strong repellent activities.

6. The four most active natural repellent compounds identified by in vivo assessments and mixtures thereof have been evaluated for acute neurotoxicity and chronic cytotoxicity effects on rodent cortex and midbrain cultures grown on the surface of microelectrode array chips. The acute toxicity assays suggested that the newly discovered mosquito repellents should not to be toxic unless ingested in considerable quantities. The chronic neurotoxicity assays suggested that direct administration of the two most active repellent compounds to the cultures produces changes in midbrain network activity similar to those of DEET, whereas administration of the other two causes no impairment of midbrain network activity.

7. Efficacy assessments of the same four most active natural repellent compounds, carried out initially in model huts in Nigeria, have shown one of these compounds to display repellency equivalent to DEET, while specific binary mixtures of the new compounds produced significantly higher repellent effects. The testing of the same compounds in rooms of family-occupied dwellings in a selected village in Nigeria produced results essentially identical to those obtained from the model huts.

8. Procedures for the protection of intellectual property associated with the discovery of the new natural repellent compounds are reaching completion and discussions with relevant companies that expressed an interest in their commercial exploitation are in progress.

Overall, these results have demonstrated both the feasibility of isolating new, safe-to-use olfaction-based mosquito behavior modifiers from natural sources and the validity of the principle that rational design of novel compounds with enhanced bioactivities based on modeling studies that capitalize on known OBP three dimensional structures is feasible. Thus, it is felt that the ENAROMATIC consortium has achieved fully its mission for a contribution toward the reduction in malaria transmission to humans by targeting the olfactory capacity of the mosquito disease vector.

Project Context and Objectives:

Blood-feeding insects carry many potentially lethal diseases, including encephalitis, malaria, and yellow fever. The most deadly of the insect-born diseases is malaria. It is caused by the parasite Plasmodium falciparum, which is transmitted to humans by female anopheline mosquitoes, particularly Anopheles gambiae, during blood feeding. The most recent estimates for human losses to malaria currently stand at over 400 million of new clinical cases and approximately 1 million deaths each year, the latter involving particularly children under the age of five.

To date malaria control has relied mainly on repressing the insect vector–human target interaction. Insecticides have helped to contain malaria in the past and even now, commercially available products to control mosquitoes are mostly insecticides and insecticide-treated bed nets. However, because insecticides proved to have severe adverse effects on public health and the general ecosystem, as well as being ineffective after prolonged use due to the rise of insecticide resistance in the targeted vector populations, new methods for controlling the disease vector are needed.

The means that could potentially be employed to control the sizes of mosquito disease vector populations and frequency of contact between vectors and human hosts rely on three distinct areas of research:
(i) development of new effective but environmentally safe insecticides;
(ii) development of genetically modified mosquitoes that maintain unreduced survival vigor in natural environments but are refractive to parasite propagation and survival; and
(iii) discovery of new compounds, mostly volatiles, for use as agents disrupting the normal olfactory capacity of female mosquitoes thus preventing them from identifying their human hosts, obtaining blood meals from them, and transmitting the malaria parasite in the process.

ENAROMATIC's effort has focused on the third of these areas of discovery.

ENAROMATIC has investigated the mechanisms providing the specificity of odor recognition with emphasis on the initial stages of this process in anopheline mosquitoes. Besides the acquisition of important new knowledge on this important physiological process, the ultimate goal of the project has been the identification of novel compounds capable of interfering with the mosquito's capacity to recognize odors of human origin and transmit the malaria parasite to humans through blood feeding.

Until the onset of the ENAROMATIC project, most efforts to understand anopheline mosquito olfaction had focused on either of the two major groups of proteins involved in odor recognition, odorant binding proteins (OBPs) and odorant receptors (ORs), which are usually studied separately. In contrast, because it was clear that the olfactory process requires the coordinated action of both of these components for efficient perception of odor stimuli, ENAROMATIC's strategy has been to tackle the target of interfering with the mosquito olfactory function by manipulating concurrently both components of this system. Thus, at the heart of ENAROMATIC research laid the fundamental principle that optimal odor recognition tuning and sensitivity are dependent on the availability of odor-matched OBP-OR pair combinations within single olfactory sensilla.

It has been anticipated that successful development of the proposed studies would advance significantly the understanding of the mechanisms that control odor recognition in mosquitoes. Most importantly, it was felt that the identification of multiple disruptors of host seeking behavior of female mosquitoes would provide new and effective tools to be employed in the effort to reduce the incidence of contact between the human host and the mosquito disease vector and produce a commensurate reduction in the spread of malaria transmission. Because a natural pathway essential for reproduction was targeted, the probability of occurrence of resistance at the level of OBPs or ORs is minimal because of the impact of the relevant mutations on the organism's capacity to perceive and respond to bona fide host odors. Furthermore, the availability of multiple high affinity OR ligands, would permit their sequential use for short periods of time, thus further minimizing the (already minimal) risk of resistance development in the mosquito populations due to prolonged exposure to single control agents.

To achieve ENAROMATIC's objectives, a significant number of prerequisites had to be fulfilled including the availability of:

- Rationally selected OBP and OR targets that likely participate in the host seeking process;
- Collections of synthetic or natural compounds to be examined with appropriate screening methods for the presence of desired functional characteristics;
- Fast in vitro screening methods for reliable initial analyses of large collections of compounds for the purpose of defining subsets displaying physiologically relevant binding capacity for specific OBP and OR targets and specific effects on the function of cognate ORs;
- Efficient methods for precise evaluation of the consequences of application of the selected subset of compounds on mosquito behavior in vivo;
- Methods for analyzing important structural features of OBP ligand binding pockets and determining ligand/OBP structure-function relationships;
- Capacity for reliable modeling of ligand derivatives with enhanced binding affinity and function;
- Capacity to synthesize the ligands and modeled ligand derivatives in native and radio-labeled forms needed for the purpose of evaluating their binding and functional properties.

Additionally, prior to establishing unequivocally their usefulness as agents that may interfere with the female mosquito's capacity to identify its human host, the compounds selected as the strongest candidate modifiers of olfaction-based female mosquito behavior should be evaluated, at least at an initial level, for:

- Lack of cytotoxicity;
- Efficacy under field conditions.

Accordingly, the following broad scientific objectives were defined:

1. Cloning and expression of OBPs and ORs judged to play important roles in the process of human host odor recognition by the female mosquito, the vector of the disease;
2. Discovery of disruptors of the host seeking behavior using a number of novel as well as previously developed in vitro high-throughput screening (HTS) systems in conjunction with collections of synthetic compounds and natural compound mixtures derived from aromatic plants, which are rich in volatiles;
3. Identification of OBP-OR combinations recognizing common ligands and potential disruptors of host seeking behavior and determination of their distribution in the female mosquito antennal sensilla;
4. Determination of in vivo functional properties of newly identified lead compounds by electro-physiological and behavioral assays involving female (and male) mosquitoes;
5. Crystallization and crystal-based structure determination of a number of model OBPs selected on structure and functionality-related criteria, modeling of ligand fitting into ligand-binding pockets and design of derivatives having better binding properties;
6. Synthesis of ligand derivatives and validation of improved binding to cognate OBPs and enhanced functionality in the in vivo electrophysiological and behavioral asssays;
7. Evaluation of lead compounds deemed to be effective with regard to disruption of normal female mosquito behavior for potential cytotoxicity on mammalian cells using model mammalian neuronal network biosensors;
8. Testing of the most promising disruptors of olfaction-based mosquito behavior at sites in an African country where malaria is endemic to determine efficacy under conditions that simulate the sites of eventual application of newly developed products.

These objectives would be fulfilled through the realization of 5 interconnected working Sections each encompassing specific work packages and tasks therein, as summarized below.

Section 1 objectives:

The objectives of this section were to develop and employ high throughput in vitro screening assays for identification of OBP- and OR-specific ligands; define ligand-matched OBP-OR pairs; and employ additional in vitro tests to assess the functional relevance of ligand-matched OBP-OR pair combinations in the olfactory process. Different groups of natural or synthetic chemical molecules or mixtures thereof were to be screened against the defined OBP and OR targets for ligand lead activities. These objectives were pursued in the context of the 3 workpackages, which are briefly described below.

WP1: identification of specific ligands for 10 selected OBPs shown to have a strong expression bias toward female mosquito antennae. Tasks under WP1 included the:
- Expression and purification of recombinant OBPs (Task 1.1);
- Screening for ligands for specific OBPs amongst a number of chemicals known to elicit electrophysiological responses in female mosquito antennae and derivatives thereof (Task 1.2);
- Screening of extracts (essential oils) derived from available collection of plants and known to contain OBP-specific binding activities (Task 1.3).

WP2: identification of ligands with specific agonist and antagonist activities 11 selected ORs. The targets included 9 ORs displaying a pronounced bias for expression in female antennae, OR1, which has a known ligand and OR7 (Orco), the heterodimerization partner of all ORs. Tasks under WP2 included the:
- Generation of transformed cell lines expressing stably the selected ORs and relevant G? subunits of trimeric G protein complexes (Task 2.1);
- Cell-based high throughput screening for the presence of OR ligands amongst (i) chemical compounds known to trigger odorant-related reponses in mosquitoes and (ii) other chemicals shown to possess specific OBP binding activities in WP1 (Task 2.2);
- Screening for the presence of OR ligands amongst the members of a combinatorial chemical library shown to possess OBP ligand activity (Task 2.3);
- Preparation of extracts from available endemic plant collection (Task 2.4);
- Screening of plant extracts for the presence of OR ligands (Task 2.5).

WP3: confirmation of functional relevance of defined ligand matched OBP-OR pairs using in vitro methodologies. This confirmation was important because it is the prelude to the more detailed studies required for the development of novel ligands capable of interfering effectively with the mosquito's olfactory competence.

Tasks under WP3 will included the:
- Assessment of enhancement of in vitro ligand-dependent OR activity by ligand-matched OBPs (Task 3.1);
- Generation of anti-OBP and anti-OR antibodies (Task 3.2);
- Examination of co-expression patterns of ligand-matched OBP-OR pairs in female olfactory sensilla (Task 3.3).

Section 2 objectives:

The objectives of this section have been to:
(i) crystallize and obtain crystal structures for selected OBPs;
(ii) model interactions of ligands in ligand binding pockets;
(iii) design and synthesize ligand derivatives with enhanced potential; and
(iv) assess functional properties of ligand derivatives in vitro and in vivo.

These objectives were pursued in the context of the 3 workpackages, which are briefly described below.

WP4: Crystallization and structure determination of OBPs of known ligand binding specificities. Tasks under WP4 included the:
- Production and purification of recombinant OBPs (Task 4.1);
- Crystallization of recombinant proteins (Task 4.2);
- Structure determination of crystallized proteins (Task 4.3).

WP5: Modelling of known and newly discovered ligands into cognate OBP ligand binding pockets and designing of ligand mimetics predicted to have improved properties. Tasks under WP5 included the:
- Modeling of OBP-ligand interactions and predictive design of ligands with enhanced binding characteristics (Task 5.1);
- Site-directed mutagenesis to confirm QSAR predictions arising from modeling studies (Task 5.2).

WP6: Synthesis of native and radiolabeled ligand derivatives and assessment of their in vitro binding characteristics and functional properties. Tasks under WP6 included the
- Synthesis of radioactive ligands and ligand derivatives predicted to have increased binding affinities (Task 6.1);
- Binding assays using wild type and mutant OBPs and radioactive and non-radioactice ligand derivatives (Task 6.2);
- In vitro functional assays using ligand derivatives with desirable binding properties (Task 6.3)

Section 3 objectives:

The objectives of this section were to (i) validate the functionality of identified ligands and cognate OBPs and ORs determined through the work carried out under WPs 1-6 using in vivo electrophysiological and behavioral assays; and (ii) classify the identified ligands in terms of in vivo effects on female mosquitoes. A final objective of this Section was to examine the in vivo effects of selective OBP expression in the antennae of male and female mosquitoes and correlate the reduction in specific OBP expression to physiological responses against administration of the relevant ligands. These experiments would not only establish unequivocally the role of specific OBPs in the process of host recognition but also deduce the extent of functional redundancy existing in the mosquito olfactory system. These objectives were pursued in the context of a single workpackage, which is briefly described below.

WP7: Testing the biological activity of all identified ligands and cognate OBPs and ORs using in vivo electrophysiological and behavioral assays. Tasks under WP7 included:
- Electrophysiological recordings (Task 7.1);
- Behavioral assays (Task 7.2);
- Specific OBP and OR RNAi injections and functional responses of mosquitoes expressing reduced levels of cognate OBPs to ligand administration (Task 7.3).

Section 4 Objectives:

The objective of this section was to undertake a preliminary safety evaluation of newly discovered olfactory disruptors for lack of cytotoxicity effects through the use of neuronal network biosensors. These objectives were pursued in the context of a single workpackage, which is briefly described below.

WP8: Employment of mammalian neuronal network microelectrode array (MEA) neurochips for establishing the lack of cytotoxicity of selected ligands for mammalian cells. Tasks under WP8 included the
- Culture of rodent cortex and spinal cord on the surface of microelectrode array neurochips (Task 8.1);
- Analysis of the electrical activity patterns obtained from neuronal networks grown on MEA neurochips after application of ligand leads. (Task 8.2).

Section 5 Objectives:

The objectives of this section were to assess the efficacy of a small number of olfactory disruptors, selected on the basis of the in vivo experimentation results of WP7 and the preliminary safety assessments of WP8, at model field sites in Africa. These objectives were pursued in the context of a single workpackage, which is briefly described below.

WP9: Application of selected ligand mimetics under field conditions in an African country (Nigeria) where malaria is endemic and assessment of field efficacies. Tasks under WP7 included the:
- Protocol development for experimental hut studies and A. gambiae population structure survey in experimental test sites (Task 9.1);
- Analysis of effects of olfactory disruptors on host seeking behavior of A. gambiae under experimental hut conditions (Task 9.2);
- Assessment of efficacy of olfactory disruptors on host seeking behavior of A. gambiae under field conditions (Task 9.3).

Project Results:

The progress that has been achieved toward the realization of the 8 project objectives described in the previous section may be summarized as follows:

1. Cloning and expression of OBPs and ORs

For this objective, a total of 12 OBPs and 11 ORs selected as the basis of the ENAROMATIC studies based on previously determined expression profiles in male and female antennae and demonstration that they are expressed predominantly in female olfactory sensilla, have been cloned in appropriate expression vectors and expressed in bacteria (OBPs) and lepidopteran insect cells (OBPs and ORs).

Recombinant OBPs expressed in bacteria were purified after being subjected to a denaturation-renaturation treatment and their structure and purity determined by circular dichroism (CD) and gas chromatography-mass spectroscopy (GC-MS) analyses. OBP expression in insect cells resulted in secretion of native polypeptides in the media of the expressing cells and purification from the latter using affinity purification protocols for appropriate tags added to their termini. Subsequent studies (summarized below) have demonstrated that the isolated OBPs were also functional in terms of ligand binding properties.

For the case of the ORs, successful functional expression in insect cells has been achieved as documented by a number of criteria, which included (i) biochemical assays (for quantitative assessments) and immunolocalization tests (for confirmation of cellular plasma membrane localization) involving custom made antibodies against native proteins (for two ORs), antibodies against specific epitope tags attached to the recombinant OR termini (for all ORs); and (ii) functional assays (for establishment of functionality) using either known specific ligands [for 4 de-orphanized ORs co-expressed with OR7 (Orco)] or a generic, non-specific ligand activating all receptors (for 6 orphan ORs co-expressed with Orco).

2. Discovery of ligands that may disrupt mosquito host seeking behavior

For this target objective, protein-based HTS assays based on fluorescent reporter molecules were established for the cloned OBPs. These assays allowed undertaking of screening tests in microtiter plates with readouts being read in automated plate readers.

Subsets of the purified OBPs have been employed for the screening of:
(i) a large combinatorial library of chemical compounds of known structures available to one of our partners;
(ii) an additional collection of 120 pure chemicals known to be active on mosquitoes, which was available to two of our other partners;
(iii) a collection of approximately 200 plant essential oils and fractions thereof containing mixtures of naturally occurring compounds, which have been generated by yet another member of the ENAROMATIC consortium from various aromatic plants;
(iv) pure constituent compounds obtained from OBP-binding essential oils and identified by relevant biochemical fractionation;
(v) a small set of chemical compounds with representatives known to have specific bioactivities against specific arthropods including mosquitoes, which were provided to us by one of our collaborating chemical companies; and
(vi) another small set of purchased chemical compounds that were predicted to have high binding capacity for specific OBPs (and maybe biological active as well against mosquitoes) based on modeling studies.

The screening tests resulted in the identification of a number of synthetic chemicals as well as multiple plant extracts and fractions thereof containing compounds, which exhibited good binding properties against specific OBPs. They also resulted in the identification of subsets of the collaborating company compounds that exhibited enhanced binding properties against specific OBPs. Moreover, based on the OBP binding assessments, the binding screens confirmed that modeling studies can be used successfully for predicting compound derivatives having enhanced binding properties against targeted OBPs as compared to the parental compounds. Finally, the OBP binding studies revealed that co-expression of OBPs may lead to OBP-OBP interactions that may extend the odor code of the individual OBPs.

The combined results revealed distinct spectra of binding for each of the examined OBPs and suggested that synthetic compounds and natural ones derived from essential oils may be active on mosquitoes. As described further below, some of these compounds were tested for their biological activity using relevant physiological and behavioral assays and were found to be biologically active against mosquitoes, primarily as powerful repellents.

For the identification of OR ligands, lepidopteran cell lines producing stably the selected ORs and, in some instances, accessory Ga proteins predicted to facilitate and/or maximize functional responses of the ORs upon ligand addition have been generated. These were used as insect cell-based HTS platforms for the screening and identification of (i) aromatic plant extracts containing compounds with ligand-like activities against specific ORs and (ii) new chemical compounds behaving as specific OR ligands. As a possible additional heterologous expression platform for low throughput ligand screening, mammalian cell lines expressing an accessory Ga15 protein were also generated which carried integrated copies of the coding regions of two deorphanized ORs under mammalian promoter control in their genomes.

Prior to actual compound screening, a number of different reporter systems were tested in conjunction with two of the generated cell lines expressing stably ORs in order to identify the reporter system providing the most robust and reproducible functional responses in the transformed cell lines. The different reporter systems tested included (i) Ca2+-fluorescent reporter probes (Fluo3 and Fluo4, evaluated by epifluorescence microscopy); (ii) Ca2+-activated luminescent proteins (aequorin and Photina cytoplasmic and mitochondrial versions, evaluated through the use of a microplate reader); and (iii) membrane potential fluorescent dyes (Red and Blue FLIPR dyes, evaluated both by scanning confocal microscopy and through the use of a monochromator microplate reader).

Based on sensitivity of responses and ease of use, the system selected for further exploitation was the insect cell-based assay that relied on the expression of a mitochondrially localized photoprotein, which is activated upon Ca2+ entry into the cells resulting from the activation of the selected ORs (ligand-gated ion channels), and was amenable to adaptation to HTS formats. For the mammalian cell lines expressing the accessory Ga15 protein, which were considered as an alternative screening system, initial calcium imaging experiments of OR-expressing cells showed that although ORs appeared mediate cell responses to their cognate ligands, their reactivity in calcium imaging experiments was erratic. Moreover, these cell lines proved to be problematic in terms of viability and ease of use. As a result, obtaining reliable results with the mammalian cells has been judged to be of limited value for ligand screening.

Compared to other major systems used for the characterization of insect ORs, which have rather complex and time-consuming assay preparation and handling requirements and require specialized monitoring instrumentation, the newly developed insect cell-based assay, which employs the Ca2+-sensitive photoprotein for receptor response readout, has considerable advantages. Thus, it yields robust and quantitative readouts of physiological responses of A. gambiae ORs upon administration of odors and other relevant ligands without requirements for elaborate instrumentation, and is adaptable to use as a medium to high throughput screening platform for fast discovery of OR-specific ligands.

Besides being used for the screening of the synthetic and natural compound collections described above for the identification of specific ligands, agonists and antagonists, capable of acting as modifiers of olfaction-based mosquito behaviors, the insect cell-based screening system afforded the opportunity for discoveries that shed light into previously unknown details of ligand-dependent olfactory receptor function. Specifically, we have documented that a specific class of reagents can be used as a set of generic tools for establishing the functionality of orphan ORs whose ligands have yet to be determined and obtaining enhanced responses from characterised ones without sacrificing specificity. With the use of such generic activation tools, which act on all mosquito OR-Orco heteromers in a non-specific manner, and by modifying appropriately the basic ligand screening protocol we have also been able to identify in our compound collections those ligands acting both as receptor agonists or antagonists for both de-orphanized and orphan receptors. Importantly, for compounds proven to act as major mosquito repellents in vivo, we were able to establish that these act on many mosquito receptors and deduced the molecular basis for our findings. The latter represents an important discovery because it affords the modeling of new compounds that could have enhanced and longer lasting repellent properties for Anopheles as well as other mosquito disease vectors. Therefore, the newly established insect cell-based assay system for mosquito ORs, which employed a genetically-encoded reporter photoprotein, proved to be a convenient, reliable and information-rich tool suitable for screening efforts and pharmacological characterization of mosquito receptors. Among others, this provided specific insights into the molecular basis underlying the function of some of the olfaction-based mosquito behavior modifiers identified through the combination of OBP screens described above, and the physiological and behavioral assays described further below.

3. Identification and distribution of OBPs and ORs recognizing common ligands

The aim of this objective has been to evaluate ENAROMATIC-selected OBPs and ORs with respect to a possible interplay in the recognition of common ligands. Towards this goal their expression and co-localization in mosquito antennal sensilla have been examined and attempts were made to assess a combined functionality of ligand-matched OR/OBP pairs in cell-based activity assay systems. In order to localize OBP and OR mRNAs and proteins specific riboprobes for all selected OBPs and ORs have been synthesized. In addition, several high titer antisera have been generated and verified for specificity. Furthermore, new protocols for separate as well as combined whole mount fluorescence in situ hybridization (WM-FISH) and immuno-histochemistry (WM-FISH) have been established.

Through the use of these methods and confocal laser scanning microscopy, the antennal topography of OBP-transcribing support cells and OR-expressing olfactory sensory neurons (OSNs) have been visualized and significant differences in the number and antennal distribution of expressing cells have been detected, with noticeable broad expression of certain OBPs. Both sexes expressed the same ORs and OBPs; however, female antennae generally had considerably higher numbers of expressing cells, representing a possible basis for gender-specific differences in the ability to detect distinct odorants. Immunolocalization studies have confirmed and significantly extended the WM-FISH-results. Specifically, for the first time antisera against Anopheles plus-C and classical OBP-types visualized the respective proteins in the sensillum lymph of sensilla trichodea, thus corroborating their function in odorant detection. The co-localization of OBP-types and OBP/OR pairs in antennal sensilla has been examined by extensive two-color double WM-FISH experiments. Through pair-wise testing of all selected OBPs, complete, partial, and non-overlapping OBP expression patterns have been detected and co-expression of multiple OBP-types in single support cells was also established. Moreover, evidence was provided for expression of classical and plus-C OBP48 by different support cells in the same sensillum.

Overall, a complex OBP expression-mosaic was discovered indicating diverse sensilla types with overlapping OBP equipment, thus suggesting cooperation of distinct OBPs with different ORs or the existence of OBP-OBP heteromers with new ligand binding properties. In line with a functional relevance of ligand-matched OR-OBP pairs, double WM-FISH with combinations of OR2-OBP1 and OR2-OBP4 have demonstrated a close association of the expressing cells indicative for sensillar co-localization of the proteins. Finally, unambiguous assignment of OBP-OR co-expression, including determination of the sensillum type housing the proteins, was demonstrated by applying combined WM-FIHC and WM-FISH. Sensilla co-localization was identified for several other OR-OBP combinations, suggesting pairs, interplaying in odorant recognition and functionally testable in forthcoming studies.

Given the development and current availability of the robust and convenient insect cell-based platform for functional assessment of heterologously expressed mosquito ORs, which was described above, the testing of the interplay of ligand-matched OBP-OR pairs is now possible. Due to the late availability of the insect cell–based assay system, however, meaningful results from measurements with OR-expressing cells in the presence of co-localized and ligand-matched OBPs (i.e. OR2-OBP1, OR2-OBP4) have not been obtained yet, but are expected beyond the end of the project.

4. Functional properties of identified lead compounds

The capacity of extracts (essential oils) obtained from various aromatic plants to effect changes in female mosquito behaviour was examined through the combined use of a specific repellency assay involving A. gambiae females and gas chromatography-coupled electroantennograms (GC-EAG) recordings, which were used to identify the chemical nature of the bioactive compounds contained in active extracts. A total of 82 essential oils from different plant families, which were found to contain OBP binding components as described above, were tested using a specific assay, the warm body repellent assay that was developed within the Consortium. These essential oils were tested during the final hours of the scotophase in a randomized design, where each product was tested, in multiple cages, with a total of 250 A. gambiae females. Compared to an ethanol control sample, which did not display any mosquito repellent activity, a defined dose of the most widely used mosquito repellent DEET, which was much lower than the quantity normally applied to the human skin for protection against mosquitoes, caused a strong reduction in the number of mosquito landings to the warm body. In addition, 21 of the tested essential oils applied at the same dose to the warm body showed repellence no different from DEET, while a few additional ones showed an attraction trend for mosquitoes. Fractions of the repellent essential oils also showed DEET-like repellence for A. gambiae females.

Constituent compounds isolated from 5 strongly repellent essential oils subjected to further analyses were shown to be capable of inducing electroantennogram (EAG) responses from A. gambiae. These were identified using specific biochemical fractionation procedures coupled to electrophysiological testing on live mosquitoes [gas chromatography (GC)-linked electroantennogram (EAG) recordings]. Subsequently, their structures were determined through application of relevant methodologies (GC-mass spectrometry) that were available for this purpose. Through the use of additional behavioural tests involving the physiological (EAG) response-triggering constituents of each essential oil, several behaviour-modifying essential oil constituents and derivatives were identified. These included 3 compounds identified in the essential oils themselves and 5 structurally related compounds that showed DEET-like repellence when tested in the warm body assay at doses similar to DEET. Additional compounds were also identified, which apparently had an opposite type of effect, namely attraction to mosquitoes.

Besides dose responses for the most active repellents in relation to the widely used repellent standard DEET, the effects of binary and ternary mixtures of the active leads at total (cumulative) concentrations equal to those of the individual tested compounds were examined and certain mixtures were found to display repellent activities significantly superior to DEET. Moreover, the most active compound also showed spatial repellent effects against A. gambiae females in a wind tunnel that were similar to those of DEET.

Finally, the direct relationship between specific OBP-ligand interactions and physiological responses of the antenna was also demonstrated by RNAi-mediated inhibition experiments, which have shown that elimination of a cognate OBP in RNAi-injected mosquitoes abolishes the antennal electrophysiological responses to the relevant ligand.

The search for additional aromatic plant constituents with strong mosquito repellent activity continues as the majority of the essential oils that were shown to have significant mosquito repellent activities have yet to be subjected to further analyses of the type described above, which would reveal the identity of the constituents that are responsible for the repellent effects. As is also described below, the identified natural products displaying strong mosquito repellent activity are subjected to more extended analyses that involve biochemical studies and modeling of their interactions with various OBPs for the purpose of identifying modifications that may provide additional desirable properties to them. In fact, the same type of in vivo assessment has been carried out for (i) all synthetic compounds modeled by ENAROMATIC partners based on the structures of compounds displaying strong binding for various OBPs, and (ii) all compounds that were provided by our industrial collaborator and shown to possess both specific OBP binding and ligand activities against specific ORs. As a result of the additional electrophysiological and behavioral assays 9 additional compounds showing DEET-like repellence activities were identified.

5. OBP crystal structure-based discovery of novel repellents

Knowledge of the three-dimensional structure of target proteins provides a starting point for structure-based approaches to repellent/attractant discovery by defining the topographies and electrostatics of the complementary surfaces of ligands and their protein targets. This information can help both the computational and synthetic chemist to optimize compounds by building better interactions with the protein, resulting in improved potency and selectivity. Accordingly, the aim of this broad objective has been to crystallize and determine the three-dimensional structure of female specific OBPs to be used for the structure-based discovery of novel classes of repellents.

Eight classic OBPs and two C-plus OBPs, which contain additional domains, have been expressed in large quantities and purified for crystallization purposes.

High-resolution X-ray data were collected for 6 OBPs (including one, OBP1, whose crystal structure has been previously determined by another group). The five new OBPs included 3 members of the classic group (OBPs 4, 7, 9) and, for the first time, two members of the C-Plus group of OBPs (OBP47 and OBP48). The three-dimensional structures of the crystallized proteins were determined both for unliganded proteins (the apo forms of OBP4 and OBP47) and proteins containing a ligand in their binding pockets (OBP1-DEET, OBP4-1-NPN, OBP4-TC4, OBP4-TC9, OBP4-indole, OBP7-AZO, OBP7-lipid X, OBP9-lipid X and OBP48-PEG). The same studies also demonstrated that under certain conditions OBPs crystallize as dimers that may represent the physiological targets.

Of particular interest was the first high-resolution crystal structure of one of the classic OBPs, OBP1, in complex with the synthetic repellent DEET, which demonstrated experimentally for the first time that OBPs represent valuable molecular targets for repellent structure-based design. In general, the identification of different binding cavities, which varied in shape, position, and solvent accessibility, as well as the nature of the amino acids that form their boundaries provided a better understanding of the differences in the binding specificities of the various OBPs and thus specific hints related to the molecular basis of odorant detection.

Explicit molecular mechanics calculations were also undertaken subsequently in order to examine further:
(i) the interaction of various potential ligands with OBP1, the single OBP whose structure was known at the beginning of the project, as well as OBPs whose structures were determined by the ENAROMATIC project; and
(ii) predict three-dimensional structures for other OBPs by homology modelling approaches.

Thus, protein-ligand interactions were analyzed using both crystal structures (for OBPs 1, 4, 9, 47 and 48) and homology models built using appropriate computing programs (for OBPs 3, 5, 12, 20 and 22). Structural analyses of OBPs highlighted conserved residues comprising ligand-binding pockets. Homodimers and heterodimers were also analyzed with appropriate computational methods and interface residues involved in protein-protein interactions were predicted. To confirm experimentally the validity of the conclusions drawn from these models, the interactions between residues in the binding cavities of selected OBPs and functional groups and surfaces of bound ligands have been confirmed by site-specific mutagenesis. The mutated OBPs were analyzed for their physical properties and ligand binding characteristics and they experimentally validated the predicted functions of the investigated amino acids.

Additionally, comparative studies examining ligand binding to OBP1 and OBP48 homodimers were also performed. The theoretical docking results coincided with the experimentally determined positions of DEET in the OBP1 homodimer, while two ligand molecules were placed in the combined site of OBP48 homodimer located at the interface between the two OBP48 monomers.

6. Synthesis and testing of ligand derivatives

Ligand derivatives containing specific side groups relative to a 'parental' ligand of known binding specificity for two specific OBPs and two ORs have been tested in order to assess the importance of the modifications to the fit of a ligand into the binding pocket of the specific OBPs and ORs. Thus, based on the observation that in A. gambiae, a human sweat component, p-cresol, mediates electrophysiological activity in OR1 and that o-cresol produces OR2-mediated activity, these two volatile compounds were chosen initially for study. Techniques for preparing radiolabelled versions were optimised and then radioactive o- and p-cresols were prepared by catalytic reduction of the modified (halogenated) cresols with tritium gas. The crude labelled products were purified and their radiological activities determined. The stability of the compounds was also confirmed.

Attempts were made to show that expression of OR1 and OR2 together with OR7 (Orco) would enable binding to the tritiated ligands o- and p-cresols and indole. However, no binding could be detected. Since it is known that these compounds should interact with the ORs it was concluded that the failure must be due to either the nature of the radiolabelled analogs, the assay conditions or the inability of the receptors to adapt a conformation that can bind a ligand. Attempts to show binding of the radiolabelled ligands to OBP1 and OBP3, which are known to bind the non-radioactive ligands, were also unsuccessful.

7. Evaluation of lead compounds for cytotoxic effects

The aim of this objective was to undertake a safety evaluation of newly discovered olfactory disruptors through the use of neuronal network microelectrode array biosensors. Mammalian neuronal network microelectrode array (MEA) neurochips were used for this purpose to establish the lack of cytotoxicity of the selected ligands. Cultures of rodent cortex and midbrain cells cultured on the surface of MEA neurochips were set and analyses of neuronal network electrical activity patterns were determined upon acute application of two widely used repellents, DEET and EBAAP. These initial studies suggested that the two compounds could be used as reference compounds for newly discovered ligand leads as they cause no cytotoxic effects at the concentrations used for effecting their action. To expand the collection of standards, four pesticides, five compounds characterizing the acetylcholine receptor and four compounds characterizing the dopaminergic system, which are known to cause specific changes to neuronal cells were also analyzed and the electrical activity patterns of the responding cells established.

Subsequent testing of four most active of the newly discovered natural repellent compounds on the cortical networks was carried out using as standards the 15 compounds mentioned above. All four odorant compounds were found to cause cell activity loss at concentrations of 300 µM suggesting that none of the examined compounds are toxic unless ingested in considerable quantities.

The same repellent compounds have also been tested for acute cytotoxic effects on midbrain networks. Independently, the four compounds caused loss of activity of the cultures at concentrations lower than 200 µM. Mixtures of the same repellent compounds were also examined for acute cytotoxic effects on midbrain networks. The combination of repellent pairs, each at concentrations equal to those examined singly, induced no additional or synergistic effects compared to the single odorants. The combination of three odor compounds in a ratio of 1:1:1 shifted the EC50 produced enhanced acute effects relative to those of the single compounds.

Overall, the results of the acute experiments indicate that, if not applied orally, all new repellent compounds should be non-hazardous to mammals below concentrations of 100 µM. Caution is advised, however, because the data also suggests that these compounds could be toxic if taken orally at high concentrations. The toxicity of two of the new repellent compounds is comparable to those of DEET and EBAAP, while two other repellents appear to be safer to use than DEET and EBAAP.

The same repellent compounds were also examined for their chronic effects on midbrain and frontal cortex networks by repeated doses over 7 cellular divisions. Based on comparisons that involved 60 different activity parameters, two of the tested repellents were shown to induce activity changes in the midbrain network similar to those of DEET and should therefore be considered as slightly neurotoxic requiring caution in use, while the other two caused no impairment of the midbrain network activity.

For the frontal cortex MEA culture system, electrophysiological experiments to assess chronic neurotoxicity have been performed only with DEET (at 300 µM) and EBAAP (at 200 µM). The results suggest that these compounds taken orally could be toxic at high concentrations.

In conclusion, while caution is certainly advised for the use of the newly discovered repellent compounds, these compounds do not seen to be more toxic that the two widely used commercially available repellents DEET and EBAAP.

8. Field testing of the most promising disruptors

The disruption of host seeking behavior of female mosquitoes could provide new effective tools to reduce the incidence of human vector contact and disease transmission. The main objective of this work package is to evaluate the efficacy of new olfactory disruptor compounds on the behaviour of female Anopheles gambiae under (i) experimental hut and (ii) field conditions.

Accordingly, the first studies in a chosen site in Nigeria have been carried out that involved (i) the establishment of model huts to be used for the initial analyses of the effectiveness of the consortium's most promising lead compounds under semi-field conditions and (ii) an assessment of mosquito vector dynamics in the huts relative to the general geographic area where the huts are located.

Because normal village dwellings vary considerably in sizes, construction and amount of paraphernalia, which make mosquito collection difficult, experimental huts of standard shape and size are required for establishing initially mosquito behaviour under natural condition. For this reason, six experimental huts fitted with veranda and window exit-traps were built according to WHO guidelines and put into operation in 2010 in Nigeria where A. gambiae and malaria are endemic. Host seeking wild mosquitoes (anophelines and culicines) were able to enter and exit the hut via open eaves on three sides of each hut. The eave on the other side opened into a screened veranda that trapped the mosquitoes. Inside each hut a human volunteer slept overnight with the test odour disruptor repellent mounted in an odour-dispersing apparatus.

The first results established the identity and distribution of different Anopheles species and molecular form populations in the huts relative to the surrounding area. No significant variations in mosquito attraction to the position of different huts were found. Also, the experimental setup appeared to be ideal for studying mosquito behaviour and evaluating new vector control tools under natural conditions.

Ethical Issues

All compounds employed in field tests were thoroughly examined for their safety on humans and other mammalian species in accordance with section B4.1 of Annex I. As has been elaborated under section #6 above, these compounds have been evaluated for mammalian neurotoxicity using a highly sensitive neuronal network microelectrode array system for rodents and found to lack mammalian neurotoxicity.

As was also scheduled, none of the compounds tested for field efficacies as potential mosquito repellents came into contact with humans or nearby animals. Their effectiveness has been evaluated through the use of mosquito traps, which are normally used in disease endemic countries for the purpose of determining local mosquito population densities through counts of the numbers of trapped mosquitoes.

Potential Impact:


With almost 400 million clinical cases and approximately 1 million deaths annually (the majority of the victims being children under the age of five), malaria, which is transmitted to humans by the African malaria mosquito vector Anopheles gambiae, represents one of the most serious infectious diseases of the world, particularly for under-developed and developing countries. Additionally, other mosquito-born infectious diseases that had been eradicated in the industrialized world are now making a comeback due to climatic changes, increased frequency of air travel and high influxes of immigrants from third world countries. During the last two years, for example, repeated incidents of Plasmodium vivax-caused malaria transmitted by local Anopheline mosquitoes have been reported in Greece. In addition, West Nile virus encephalitis transmitted by local strains of Culex pipiens mosquitoes and, in 2012, Dengue virus haemorrhagic fever transmitted by local Aedes mosquitoes have also been reported.

The scientific impact: a new paradigm for prevention of mosquito-born infectious diseases established by ENAROMATIC

As detailed below, the strong collaborative interactions forged amongst the ENAROMATIC project participants, have produced results that demonstrated the feasibility of increasing the efficiency and impact of European research on mosquito olfaction both at the level of basic research and that of technology transfer. In turn, this outcome allowed ENAROMATIC to reach the goal of discovering and using specific volatiles as effective mosquito olfactory disruptors for control of transmission of mosquito-borne infectious diseases.

In the four years of its life, the ENAROMATIC project, which targeted the mosquito vector A. gambiae, achieved the development and deployment of rational, target-based bioassays that permitted the identification of powerful natural repellent compounds, which, when placed at a distance from human subjects, induced mosquitoes to exit rooms they had previously entered. More specifically, ENAROMATIC used subsets of A. gambiae odorant binding proteins (OBPs) and odorant receptors (ORs) expressed predominantly in female antennae as specific targets for identification of compounds of synthetic or natural origin capable of effecting changes in the normal responses of mosquitoes to odor plumes of human origin. For the OBPs, the screening assays allowed the identification of both synthetic chemicals and constituents of essential oils (EOs) capable of binding to them in vitro. With subsequent analyses of the OBP binding oils using specialized behavioural and electrophysological methods in conjunction with techniques for chemical compound identification, several EOs and individual constituents displaying strong repellent activities were identified. Thus, by using such rational approaches, which included parallel analyses for modulation of mosquito OR function, the rate of discovery of compounds with repellent activity was accelerated substantially relative to the random approaches employed by industry.

As indicated above, the approaches implemented by ENAROMATIC resulted in the identification of several essential oils and constituent compounds displaying strong repellent activities for A. gambiae mosquitoes. Four of the strongest natural repellents were tested for safety and found to be lack neurotoxic and cytotoxic activities; and field-testing in Africa showed two of them and blends thereof to be effective spatial repellents that matched and, in some cases, outperformed DEET, the most widely used olfaction standard, in effectiveness. Additionally, more than 150 EOs obtained from plant families different than those of the initial group, were examined and several were found to contain OBP binding and repellent activities of as yet unknown but easily identifiable consituents. Most importantly, the work of ENAROMATIC has shown that natural repellents developed against Anopleline mosquitoes were also very active against Culex mosquitoes that vector the West Nile, Japanese and St Louis encephalitis viruses, and even, but to a lesser extent, against sand flies, vectors of Leishmania parasites.

Last but certainly not least, the consortium's studies on OBP structure determination have confirmed that ligand- and structure-based modeling approaches could facilitate the design of novel repellents with enhanced affinity for OBPs that may be used in lower concentrations and be detected over longer distances and, therefore, be incorporated into release systems that deliver active ingredients into the air (spatial repellents) over significant periods of time. This is particularly important for endophilic mosquitoes like A. gambiae, which prefer indoor environments.

The socio-economic impact and wider societal implications of the project

For the mosquito malaria vector, the unique approach pursued by ENAROMATIC has resulted in the identification of disruptors of the host seeking behavior of female mosquitoes from natural sources and, thus, provided new tools to be employed in the effort to reduce the incidence of contact between the human host and the insect disease vector and a reduction in the spread of malaria transmission. Because malaria is such a pervasive plague in Africa, Central and South America and South East Asia and the control of the vector population such a formidable task, it was clear at the outset that the initial impact of the research would occur mainly in countries where malaria is currently endemic. This was also the reason why the field tests that assessed the effectiveness of the newly discovered natural repellent compounds were undertaken in an African country where the malaria mosquito vector and the disease are endemic. However, given the global warming caused by the climatic change and the consequent spread of malaria mosquito vectors in what used to be more temperate climatic zones, the need for malaria transmission control may in the foreseeable future become relevant to Europe and other countries where malaria had been previously eradicated. Moreover, the ENAROMATIC finding that at least some of the newly-discovered compounds of plant origin act as powerful repellents not only for A. gambiae but also for other hematophagous insect disease vectors such as female Culex mosquitoes (vectors of other infectious diseases including West Nile virus, filariasis, Japanese encephalitis and avian malaria) and female sandflies (responsible for the transmission of Leishmania infantum chagasi parasites) increases further the potential impact for the consortium's discoveries.

Main dissemination activities

For the scientific community outside the consortium boundaries and the public at large, caution was exercised to disseminate results with commercial potential after protection of patentable intellectual property had been ensured. For scientific results, the dissemination effort included:
(i) publication as normal scholarly literature in high impact scientific journals of general interest as well as insect-specific and applied science journals;
(ii) presentations at national and international conferences.

Importantly, ENAROMATIC, in collaboration with the European Commission and consultation with other EU-funded projects on malaria mosquito vector and transmission control, undertook the organization of the second conference on 'Malaria vector biology and transmission control', which took place in Crete, Greece in the summer of 2011 and became also the venue for the assessment of ENAROMATIC's activities by an international panel of experts that formed the advisory and evaluation board of the consortium and also contributed to the strategic orientation of future EU-funded vector and transmission research.

Finally, general information about the project and its aims and achievements was also presented to the public at large in the public access part of the ENAROMATIC website and also by giving interviews in mass communications media (radio, television, newspapers and magazines of general interest) and issuing relevant press releases for breakthroughs deemed to deserve wider publicity.

Intellectual property protection and exploitation

ENAROMATIC had developed an Intellectual Property (IP) Agreement, which contained all IP-related issues and general principles on approaches to be taken for matters relevant to this domain. This Agreement, which was signed by all partners and became part of the Consortium Agreement, anticipated the following:

- Identification of patentable innovations

Partners were to communicate on a confidential basis in order to evaluate the potential of new products or technologies for possible patent filing or other means of protecting intellectual property. The project policy was that patenting of results to be exploited should occur without undue delays that could prevent rapid publication of results and circulation of information among partners.

- Ownership of knowledge

As detailed in the model contract for Sixth Framework Programme (FP6), intellectual property belongs to the partner developing the concept and making the discovery. Protocols for joint ownership were also identified in the consortium agreement. Each partner identified and defined pre-existing intellectual property rights (IPR), which were available for the project and those that could not be made available on a royalty-free basis or had other specific restrictions for use by the consortium members. All partners also declared pre-existing know-how to be contributed to or introduced into the project.

- Protection of knowledge, access rights for further research and licensing options

The consortium Agreement anticipated access rights for use of the intellectual property for research activities on a royalty-free basis. Irrespective of ownership, there was an obvious imperative that all consortium partners who needed access to the developed technology should receive the necessary licences for research use from the owners. The IPR agreement provided for this, but did not include a blanket cross-license to all parties. In general, findings generated by one participant or one working group within the consortium that were required by another participant for conduct of their work under the project were made available on a royalty-free basis for intellectual property generated during the course of the contract, and, unless otherwise agreed before signing the contract, pre-existing know-how as well. Findings generated by other participants that were needed by a given participant in order to exploit knowledge, which he/she generated within the course of the contract, were royalty-free for intellectual property generated during the contract and non-discriminative conditions applied for intellectual property created outside the project during its duration.

- Use and dissemination of knowledge and exploitation of intellectual property

All partners who have contributed to the creation of intellectual property should share in fair and reasonable commercial returns should the intellectual property be exploited. Universities within the consortium might wish to see industrial collaborators of the consortium as the preferred route to commercialise, provided these collaborators have the interest, resources and base to undertake the commercialisation.

For the cases that the IP owner(s) sought to commercialise, the financial reward would follow inventorship (and hence ownership) with specific revenue sharing agreements made between the owners. The consortium did not advocate a wider revenue sharing obligation, as this has been judged to represent a major disincentive for commercial exploitation by the private sector, particularly SMEs. Other key points were that all pre-existing intellectual property at the start of the consortium remained in the ownership of the party that introduced it. There would be no obligation to license such existing intellectual property to any Consortium member, unless a clear need existed for this to happen in order for project objectives to be met. Similarly, the IPR agreement acknowledged that partners may be working in the same area outside of the Project and that independently generated intellectual property would not be part of the project intellectual property, though individual Consortium members might introduce such intellectual property at any time as they saw fit.

Decisions of whether and how to proceed for IP protection were to be made collectively by the inventors. For example, if newly identified compounds were already protected by a patent for a different or similar types of action, the inventors' claims would be based on the action of specific mixtures (blends with other natural compounds) and formulations containing the active compounds.

List of Websites: