Mosquito attractant formulations and uses thereof

Abstract

There is herein provided a mosquito attractant formulation comprising: (a) an aldehyde component; (b) a first monoterpene component; and (c) a second monoterpene component, wherein components (a) to (c) are defined in the description provided herein, 5 and products, uses and methods relating to the same. There is also herein provided the use of the compound HMBPP as a phagostimulant, and products and methods relating to the same.

Claims

1. A combination product or kit-of parts comprising: (a) (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP); (b) an effective amount of a mosquito and/or parasite control agent; and optionally (c) a food source.

2. The combination product or kit-of parts of claim 1, wherein the product or kit-of-parts comprises an effective amount of the mosquito control agent, wherein the mosquito control agent is capable of affecting a mosquito population or the ability of a mosquito to act as a vector for a parasite.

3. The combination product or kit-of parts of claim 2, wherein the mosquito control agent is a mosquito killing agent.

4. The combination product or kit-of parts of claim 2, wherein the mosquito control agent is a chemical compound which is capable of inducing death in mosquitoes upon consumption of an effective amount thereof.

5. The combination product or kit-of parts of claim 2, wherein the mosquito control agent is a mosquito pathogen which is capable of affecting mosquito populations.

6. The combination product or kit-of parts of claim 2, wherein the mosquito control agent is a mosquito pathogen and the mosquito pathogen is a bacterial agent.

7. The combination product or kit-of parts of claim 6, wherein the bacterial agent is Bacillus thurgiensis.

8. The combination product or kit-of parts of claim 2, wherein the mosquito control agent is mosquito pathogens which are capable of affecting the ability of a mosquito to take on, carry and/or transmit a parasite.

9. The combination product or kit-of parts of claim 2, wherein the product or kit-of-parts comprises an effective amount of a parasite control agent, wherein the parasite control agent is capable of killing a parasite within a mosquito.

10. The combination product or kit-of parts of claim 9, wherein the parasite control agent is commensal bacteria that have been genetically modified.

11. The combination product or kit-of parts of claim 1, wherein the product or kit-of-parts further comprises a mosquito attractant.

12. The combination product or kit-of parts of claim 1, wherein the product or kit-of-parts comprises at least 1% of HMBPP by weight of the product or kit-of-parts.

13. The combination product of kit-of-parts of claim 1, wherein the food source comprises carbohydrates, lipids, proteins, vitamins, minerals, electrolytes and/or hydration.

14. The combination product or kit-of-parts of claim 13, wherein the food source comprises proteins and/or lipids.

15. The combination product or kit of parts of claim 13, wherein the food source is an aqueous sugar solution.

16. A mosquito control product comprising: (a) a mosquito trapping and/or killing device; (b) (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP); and (c) a food source, and/or an effective amount of a mosquito and/or a parasite control agent.

17. The mosquito trapping and/or killing device of claim 16, wherein the combination product comprises an effective amount of a mosquito control agent, wherein the mosquito control agent is capable of affecting a mosquito population or the ability of a mosquito to act as a vector for a parasite.

18. The mosquito trapping and/or killing device of claim 17, wherein the mosquito control agent is a mosquito killing agent.

19. The mosquito trapping and/or killing device of claim 17, wherein the mosquito control agent is a chemical compound which is capable of inducing death in mosquitoes upon consumption of an effective amount thereof.

20. The mosquito trapping and/or killing device of claim 17, wherein the mosquito control agent is a mosquito pathogen which is capable of affecting mosquito populations.

21. The mosquito trapping and/or killing device of claim 17, wherein the mosquito control agent is a mosquito pathogen and the mosquito pathogen is a bacterial agent.

22. The mosquito trapping and/or killing device of claim 21, wherein the bacterial agent is Bacillus thurgiensis.

23. The mosquito trapping and/or killing device of claim 17, wherein the mosquito control agent is mosquito pathogens which are capable of affecting the ability of a mosquito to take on, carry and/or transmit a parasite.

24. The mosquito trapping and/or killing device of claim 17, wherein the combination product comprises an effective amount of a parasite control agent, wherein the parasite control agent is capable of killing a parasite within a mosquito.

25. The mosquito trapping and/or killing device of claim 24, wherein the parasite control agent is commensal bacteria that have been genetically modified.

26. The mosquito trapping and/or killing device of claim 16, wherein the product further comprises a mosquito attractant.

27. The mosquito trapping and/or killing device of claim 16, wherein the product comprises at least 1% of HMBPP by weight of the product.

28. The mosquito trapping and/or killing device of claim 16, wherein the food source comprises carbohydrates, lipids, proteins, vitamins, minerals, electrolytes and/or hydration.

29. The mosquito trapping and/or killing device of claim 28, wherein the food source comprises proteins and/or lipids.

30. The mosquito trapping and/or killing device of claim 28 wherein the food source is an aqueous sugar solution.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1A: Mosquito preference (lower panel) for red blood cells (RBCs), serum and glucose with PABA in the presence or absence of HMBPP (h) in a dual choice attraction assay (upper panel) [RBCs: χ21=29.11, p<0.001; Serum: χ21=0.26, p=0.60; Glucose with PABA: χ21=0.34, p=0.56].

(2) FIGS. 1B to 1F: Feeding proportions of mosquitoes allowed to feed for 90 s on: (B) RBCs, HMBPP-supplemented RBCs (hRBCs), asexual trophozoite-, or gametocyte-infected RBCs (tiRBCs and giRBCs, respectively) [hRBCs: z=4.25, p<0.001; tiRBCs: z=4.59, p<0.001; giRBCs: z=4.60, p<0.001]; (C) RBCs supplemented with different concentrations of HMBPP; (D) dilutions in RBCs of gametocyte culture supernatants (control RBCs, —) and FR-900098-treated tRBCs supplemented with IPP, .box-tangle-solidup.; χ21=1.21, p=0.42).

(3) FIGS. 1E and 1F: Percentage of mosquitoes landing, probing and initiating feeding [Persistency (%)] within five minutes on: (E) serum, with and without HMBPP; or (F), physiological salt solution, with or without HMBPP (h-NaCl), as well as IPP-supplemented NaCl (ipp-NaCl) [hSerum: χ21=10.62, p=0.001; hNaCl: χ21=7.93, p=0.005; ipp-Nacl: χ21=1.01, p=0.55]. Error bars, ±SE; asterisks denote significant differences (*p<0.05, **p<0.01, ***p<0.001; n.s., non-significant, n=180).

(4) FIG. 2A: Attraction of mosquitoes to red blood cells (RBCs), RBCs plus media (M) or HMBPP-supplemented RBCs (hRBCs) was assessed in comparison to RBCs, hRBCs, gametocyte culture supernatant-supplemented RBCs (giRBCs) and isopentenyl pyrophosphate-supplemented RBCs (ippRBCs) and their headspace extracts, with or without CO.sub.2 in the wind tunnel assay, as indicated [RBCs/RBCs: χ21=0.25, p=0.62; RBCs/hRBCs: χ21=79.14, p<0.001; RBCs+M/giRBCs: χ21=17.53, p<0.001; RBCs/ippRBCs: χ21=0.49, p=0.48; RBCs+CO.sub.2/hRBCs: χ21=41.82, p<0.001; hRBCshs/hRBCs: χ21=6.79, p=0.009; giRBCshs/hRBCs: χ21=14.70, p=0.001; hRBCshs+CO.sub.2/hRBCs: χ21=0.39, p=0.53; giRBCshs+CO2/hRBCs: χ21=1.93, p=0.16].

(5) FIG. 2B: CO.sub.2 discharge from RBC samples with or without HMBPP [t=11.28, df=1, p=0.008].

(6) FIGS. 2C and 2D: GC-MS analyses of headspace extracts from RBCs and hRBCs [χ21=19.96, p<0.001; χ21=27.12, p<0.001].

(7) FIG. 2E: A synthetic volatile blend (Blend), consisting of the compounds identified as enhanced within the headspace of hRBCs, was tested against both hRBCs and RBCs, with or without CO.sub.2. [Blend (10 μl)+CO.sub.2/hRBCs: χ21=0.26, p=0.28; Blend (5 μl)+CO.sub.2/hRBCs: χ21=7.28, p=0.007; Blend (1 μl)+CO.sub.2/hRBCs: χ21=16.56, p<0.001; Blend (10 μl)/RBCs: χ21=9.87, p=0.002; Blend (10 μl)+CO.sub.2/RBCs: χ21=17.39, p<0.001]. Error bars, ±SE; asterisks denote significant differences (*p<0.05, **p<0.01, ***p<0.001; n.s., non-significant, n=90).

(8) FIG. 3A: The relationship between fecundity (number of eggs per mosquito) and blood meal size (haematin) [treatment*hematin: χ21=6.27, p=0.01; treatment: χ21=0.19, p=0.66] of individual mosquitoes (.square-solid. hRBCs • RBCs).

(9) FIG. 3B: Survival of mosquitoes monitored daily post-feeding until natural death [Cox hazard proportional model, χ21=0.71, p=0.39]. This experiment was performed in triplicate (random effect, in each replicate n=210).

(10) FIG. 3C: Average meal size was determined by the amount of haematin excreted and normalised individually to wing length [χ21=5.67, p=0.02].

(11) FIG. 3D: Average number of oocysts per midgut at 10 dpi [deviance=33.13, df=1, p<0.001].

(12) FIG. 3E: Sporozoite infection (% prevalence in salivary glands of infected mosquitoes) at 14 dpi [χ21=8.84, p=0.002]. Error bars, ±SE; asterisks denote significant differences (*p<0.05, **p<0.01, ***p<0.001).

(13) FIG. 3F: Schematic model of proposed HMBPP effects within the malaria parasite transmission cycle: (1) HMBPP produced by P. falciparum in an infected human (2) induces an increased release of (3) a volatile blend consisting of CO.sub.2, aldehydes and monoterpenes from RBCs. This causes (4) an increased attraction of female mosquitoes. (5) The female blood ingestion size and biting rate increases, while its fecundity and survival does not change. (6) The behavioural and physiological effects are reflected in the mosquito transcriptional profile, in which genes involved in immune responses, oogenesis neuronal synapses as well as translation, are affected. (7) The parasite fitness reward is reflected in an increased infection intensity and prevalence (8) that increases the probability of parasite transmission to another human.

EXAMPLES

(14) The present invention is further illustrated by way of the following examples, which are not intended to be limiting on the overall scope of the invention but which may define certain features and embodiments thereof.

(15) General Experimental Methods

(16) Ethics:

(17) Human blood (type 0) was provided in citrate-phosphate-dextrose-adenine anti-coagulant/preservative, and serum (type AB) was obtained from the Blood Transfusion Service at Karolinska Hospital, Solna, Sweden in accordance with the Declaration of Helsinki and approved by the Ethical Review Board in Stockholm (2011/850-32). The Swedish Work Environment Authority, Stockholm, Sweden (Dnr SU FV-2.10.2-2905-13/31-01-2017) approved the class 3 biological agent laboratory and practices, including insectary design and equipment to work with P. falciparum infected mosquitoes. The ordinances are mainly based on the EU directive 2000/54/EC on the protection of workers from risks related to exposure to biological agents at work.

(18) Materials:

(19) (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate, HMBPP, and isopentenyl pyrophosphate, IPP, were purchased from Sigma Aldrich (St. Louis, Mo., USA) and Isoprenoids (Tampa, USA, LC), respectively. (2E)-2-methylbut-2-ene-1,4-diol was synthesised and structurally verified by nuclear resonance spectroscopy at Royal Institute of Technology, Stockholm. The anti-coagulant/preservative, citrate-phosphate-dextrose-adenine was from Vacuette (Greiner Bio-One Kremsmünster, Austria). Giemsa stain was purchased from Sigma Aldrich (St. Louis, Mo., USA), and the RPM11640 medium from Gibco (NY, USA). Drierite was from WA Hammond Drierite (Xenia, Ohio, USA) and Ascarite from Thomas Scientific, (Swedesboro, N.J., USA). DNasel was from Fermentas (Vilnius, Lithuania).

(20) Mosquito Rearing and Blood Feeding:

(21) Mosquitoes used in this study were from the laboratory colony of A. gambiae sensu lato (Keele line, University of Glasgow), which was produced by balanced interbreeding of four A. gambiae s.s. strains (H. Hurd, et al. Evolution, 59, 2560-2572 (2005)). Larvae were reared under standard insectary conditions (27±1° C., 70% humidity, 12 h light: 12 h dark cycle) and fed on TetraMin fish flakes (Tetra ltd., Germany). Pupae were transferred into holding cages for emergence. Emerged adults were fed ad libitum on 5% glucose solution, supplemented with 0.05% (w/v) 4-aminobenzoic acid (PABA, Sigma-Aldrich), through soaked filters on top of the 2 ml tubes and with soaked filter pads inside cages. Red blood cells (RBCs) were washed with Roswell Park Memorial Institute (RPMI) medium and stored in RPMI at 50% haematocrit at 4° C. until use. To measure the effect of HMBPP on mosquito fitness, RBCs stored in RPMI were centrifuged at 2500×g for 5 min followed by replacement of the medium with AB serum for a final haematocrit of 40%. HMBPP (stock concentration at 4 mM in Nanopure water, stored at −80° C.) was diluted to 10 μM in 1 ml RBC suspension, and the corresponding volume of Nanopure water was added to the control RBCs. All experiments were, unless otherwise stated, conducted on 5-7 days post-emergence female mosquitoes maintained in separated cages (approximately 210 individuals per cohort) and fed RBCs either with or without HMBPP for 30 min. All experiments were performed in triplicates.

(22) Plasmodium falciparum Culture and Strain:

(23) A strain of the human malaria parasite P. falciparum, denoted NF54 (T. Ponnudurai, et al., Trop Geogr Med, 33, 50-54 (1981); kindly donated by Klays Berzins, Stockholm University), known to produce gametocytes infectious to mosquitoes (D. Walliker, paper presented at the Symposia on Molecular and Cellular Biology New Series 42, UCLA (1987)), was used for all experimental infections. Parasites were cultured in vitro according to standard protocols (W. Trager and J. B. Jensen, Science, 193, 673-675 (1976)) at 5% haematocrit in complete RPMI-1640 medium with 10% human serum under a gas environment of 1% O.sub.2, 3% CO.sub.2 and 96% N.sub.2 (Labline, Goteborg, Sweden). The culture medium was replaced daily and parasitaemia assessed using Giemsa staining every two days (ibid.). When the parasitaemia reached ˜6% (approximately 2 days after initiating the culture), cultures were diluted with a freshly made 5% haematocrit mixture of RBCs/complete medium. Asexual, synchronised trophozoites were fed to mosquitoes at 5% parasitaemia.

(24) Mosquito Infectious Feed:

(25) Plasmodium falciparum gametocyte cultures (NF54 strain) were set up at 0.5-0.7% parasitaemia, 6% haematocrit in complete RPMI medium, according to standard procedures (R. Carter, et al. Methods Mol Biol, 21, 67-88 (1993)). For each infectious feeding, a mixture of gametocytes from 14- and 17-day gametocyte cultures were used (ibid.). On the day of experimental infection, uninfected RBCs were supplemented with P. falciparum gametocytes to a final gametocytaemia of approximately 3%, which is known to generate good infection prevalence (>50% in A. gambiae s.s. mosquitoes; ibid.). To measure the effect of HMBPP on parasite infectiousness, two groups of approximately 210 mosquitoes were fed on mature gametocyte-infected red blood cells (giRBCs) with or without HMBPP using a membrane feeder, as previously described (ibid.). Gametocyte density was kept equal in control and treatments. To give the final gametocytaemia of 3% in each tubes, on the day of feed, the final mature gametocytaemia was calculated in cultures and then the volume of gametocyte inoculum was estimated per 1 ml of blood. The gametocyte inoculum was prepared using an equal volume (˜400 μl) of the gametocyte mixture was added to 1 ml of each of the blood treatments (with/without HMBPP). Adding a fixed volume of well-mixed gametocyte culture to a fixed volume of each blood treatment, it was ensured that parasite density (no. gametocytes per ml) was constant in both groups.

(26) Dual Choice:

(27) Initial experiments were performed using a Y-tube olfactometer (length of the central cylinder and two arms: 25/15/15 cm respectively, inner diameter: 5 cm; see FIG. 1a). For each experiment, 180 female mosquitoes were individually placed in a release chamber and flown one by one. Females flew towards the upwind end (airflow 3 L min.sup.−1) and entered one of two trapping chambers through which odour released from nearby membrane feeders passed (5 cm distance to each trapping chamber). Three experiments were conducted. In the first experiment, the membrane feeders were loaded with RBCs with or without HMBPP, while in the second they were loaded with serum with or without HMBPP. The third experiment was performed with a 5% glucose with 0.05% PABA solution with or without HMBPP. To control for possible spatial effects, the location of treatment at each olfactometer arm was switched every 30 minutes (˜15 mosquitoes flied one by one in each 30 minutes). Each replacement (feeder rotation with fresh RBCs) was counted as an experimental replicate (experimental block). Mosquitoes reaching any of the trapping chambers were considered to have made a choice. Each experiment was repeated six times with in total 180 mosquitoes per treatment.

(28) Feeding Proportion Experiments:

(29) Fifty mosquitoes were isolated in three separate cages (8 cm internal diameter×10 cm in height) covered with netting, and fed on 1 ml of control RBCs, hRBCs or giRBCs, respectively. Each group was allowed to feed on its own separated membrane feeder for 90 s, the average time for Anopheles mosquito engorgement (D. D. Chadee and J. C. Beier, Ann Trop Med Parasitol, 89, 531-540 (1995)). For each group, the number of fully fed mosquitoes compared to unfed controls was immediately recorded. In another set of feeding experiments, serum or a 0.9% physiological saline solution (NaCl, pH 7.4), either with or without 10 μM HMBPP, were offered to mosquitoes during a 5 min feeding window, and the number of fed mosquitoes recorded every minute for each cage separately. In the saline experiment, an additional treatment of 10 μM IPP in NaCl was included.

(30) Fosmidomycin Assay:

(31) The fosmidomycin assay was carried out as previously described (12). 48 hours (h) after the initiation of antibiotic treatment (FR-900098, the N-acetyl analogue to fosmidomycin), asexual parasite cultures were rescued by supplementation with 200 μM IPP for 48 h of continued passaging. Media from the rescued culture were collected and offered to mosquitoes, in comparison with media from control parasites (without antibiotic treatment). Fifty mosquitoes, isolated in two separate cages, were offered different treatments. Each group was allowed to feed on its own membrane feeder for 90 s.

(32) Volatile Collection and GC-MS Analysis:

(33) RBCs (1 ml) were transferred to 4 ml glass vials closed with auto-sampler screw caps, and equilibrated in a thermostat at 38° C. for 15 min. Then, 2.5 μl of HMBPP or Nanopure water were added to the RBCs samples (final concentration: 10 μM HMBPP per sample). Volatiles were collected from the headspace using solid phase micro-extraction (SPME; Supelco, Bellmonte, Pa., USA) (M. J. Yang, el al., J. Chem. Educ., 74, 1130-1132 (1997) and A. K. Borg-Karlson and R. Mozuraitis, Naturforsch., C. 51, 599-602 (1996)). Prior to headspace exposure, the polydimethylsiloxane/divinylbenzene/carboxene-coated SPME fiber (grey) was conditioned for 5 min at 250° C. in a gas chromatograph (GC) injector (6890, Agilent Technologies, Santa Clara, Calif., USA). Volatiles were collected for 30 min (n=5). The volatiles on SPME fiber were then desorbed in the injector (splitless mode, 0.5 min, 225° C.) of a combined Agilent 7890 GC and 5977 mass spectrometer (MS). Helium was used as the carrier gas with a constant flow of 35 cm min.sup.−1. The GC was equipped with a DB-wax coated fused silica capillary column (J&W Scientific, Folsom, Calif., USA; 60 m×0.25 mm i.d., d.f.=0.25 μm). The GC oven temperature was programmed from 30° C. (3 min hold) to 225° C., increasing at 8° C. min.sup.−1, then isothermal for 10 min. Electron ionisation mass spectra were determined at 70 eV with the ion source kept at 230° C. and mass spectra were obtained at mass to charge ratio (m/z) 29-400. Chromatographic profiles of volatiles, which were sampled from RBCs and hRBCs, were compared. The compounds that were more abundant in the hRBC samples were identified according to their retention indices (Retention Index) and mass spectra in comparison with the NIST library (Agilent Technologies) and authentic standards. A similar protocol was used for the quantification of CO.sub.2, with the mass spectral data and GC retention times being compared with emission from frozen CO.sub.2. The relative amounts of CO.sub.2 in arbitrary units were determined from the areas of the respective chromatogram peaks using single ion display mode at m/z 44±0.5 Da. The amounts of identified compounds were quantified by the use of extracted ion chromatograms, since they are very minute. The compounds identified were the only volatiles identified in the SPME samples. The estimations were based on the very same quantifications given in ‘arbitrary area units’. The SPME volatile collections were made under very controlled conditions: the same volume of ‘RBCs’ and the very same temperature, which gives the same % RH in the vials, and under these very same conditions the affinities of the volatiles to the SPME fiber coating are the same.

(34) Infrared-Gas Analysis (IRGA) of RBC CO.sub.2 Production:

(35) RBCs from a single donor (500 μl) were pipetted into pre-weighed aluminium cups, and then distilled water (control) or 10 μM HMBPP was added after 30 s, after which measurement began. T otal sample preparation time was 90 seconds (s). The analysis of RBCs and hRBCs was performed, in parallel and in triplicate, on two consecutive days. Total CO.sub.2 production was measured with a Li-7000 CO.sub.2 analyser (LiCor, Lincoln, Nebr., USA) connected to a flow-through differential mode respirometry system (Sable Systems, Las Vegas, Nev., USA). Two separate lines of air were scrubbed of H.sub.2O and CO.sub.2 using drierite and ascarite, respectively, at 60 ml min-1 using SS4 sub-samplers (Sable Systems). Two cylindrical glass respirometry chambers (volume 10 ml; of which an empty chamber served as the baseline) were connected to a MUX multiplexer (Sable Systems) at 37° C. Preliminary tests were performed to ensure that the incurrent air temperature flowing through the respirometry chamber was stabilised with the surrounding ambient temperature (37° C.). The multiplexer was programmed to measure the empty chamber baseline for 10 s, the chamber containing the RBC sample for 5 minutes (min) and then the empty chamber for 1 min. The sampling interval was 1 Hz. The RBC and hRBC samples were weighed (±0.1 mg, AM100; Mettler TOLEDO, Columbia, Ohio, USA) post-assay. Respirometry data were baseline corrected and converted to ml CO.sub.2 h.sup.−1 by the acquisition and analysis software Expedata version 1.7.30 (Sable Systems, Las Vegas, Nev., USA). Prior to measurements, the CO.sub.2 analyser was calibrated with 1000 ppm CO.sub.2 in nitrogen.

(36) Olfactometer Bioassay:

(37) Behavioural experiments were performed using different odour sources with and without pulsed CO.sub.2 stimuli in a Plexiglas Y shaped olfactometer (9.5 cm i.d.×length of the central cylinder and two arms: 120/120/120 cm respectively), illuminated from above at 280 lx. A charcoal-filtered and humidified air stream (25±2° C., RH 65±2%) passed through stainless steel mesh screens at 30 cm s.sup.−1 to generate a laminar flow in the olfactometer. Odors were introduced into the air stream of each arm at the downwind end. Synthetic air (79.1% N.sub.2, 20.9% O.sub.2; Strandmöllen, Ljungby, Sweden) was pumped through an activated charcoal filter, humidified and then split between the two treatments. Odor sources consisted of either RBC-based treatments or headspace extracts. RBC-based treatments included RBCs alone, hRBCs, giRBCs or ippRBCs (RBCs supplemented with HMBPP, gametocytes or IPP), as well as RBCs with supernatants from gametocyte-infected RBCs, or RBCs supplemented with the synthetic odour blend. Headspace extracts collected from hRBCs and giRBCs were used. Treatments including RBCs were heated to 37° C. prior to being pumped (150 ml min.sup.−1) into the airstream of each arm separately. The blend and headspace extracts were delivered to each arm via a wick dispenser. Carbon dioxide (Strandmöllen) was introduced into a plume generator that was placed behind the metal screens to create a homogenous-pulsed flow, which was regulated by a stimulus controller (SEC-2/b, Syntech, Löptin Germany). In treatments including 5 ppm CO.sub.2, pure CO.sub.2 was mixed with synthetic air at 1.5 l min.sup.−1 to generate the homogenised pulses in the selected arm. A CO.sub.2 analyser (LI-COR Biosciences, Nebraska, USA) was used to measure the concentration of CO.sub.2 and track the pulsed stimuli at the downwind end of the olfactometer. Female mosquitoes were stored individually in 7×9 cm internal diameter release cages in the bioassay chamber for 1 h before experiments. A release cage was placed at the downwind end of the wind tunnel. After 1 min acclimatisation, the cage door was opened and the female mosquito was allowed 90 s to fly toward either of the provided odour sources. Only females that reached the upwind capture cages were considered to have made a choice. Each experiment was repeated using 90 flown individual mosquitoes per condition (˜15 mosquitoes flown one by one in each 30 minutes).

(38) Statistical Analysis:

(39) GLMM statistical modelling was used to corroborate the validity of results based on the whole data set by including the effect of replications (experimental blocks), including weighting for multiple replications. In all analyses, the effect of the main experimental effects (e.g. treatment) was investigated while controlling for variation in experimental replication (random variable). For all results, the significance of all explanatory effects were evaluated by using likelihood ratio test (LRT). Analyses were performed using R statistical software. In all analyses, HMBPP supplementation of RBCs was investigated as the primary effect of interest. Generalised Linear Mixed Models (GLMM, R statistical software v. 3.1.1) assuming a binomial distribution were used to test the effect of HMBPP on the binary response variable of dual choice in the attraction, feeding rate, oviposition, and oocyst and sporozoite infection prevalence assays (Logistic regression models, absent or present; Ime4 package, glmer, R, v. 3.1.1). Logistic regression is a powerful statistical method for binomial outcome (take the value 0 or 1) with one or more explanatory variables. In this study, we included at least two variables: 1-Treatment (main effect) and 2-Experimental blocks (random effect). In all analyses, Treatment (e.g. blood with/without HMBPP) was investigated as the main effect of interest (M. Crawley, The R book, John Wiley & Sons Ltd., (2007)). A similar approach was used to test for variation in oocyst intensity between different experimental treatments. Given the highly over-dispersed nature of parasite abundance data, negative binomial distribution was assumed in these GLMMs (glmmADMB, nlme package, R, v. 3.1.1). For blood meal size, oviposition rate, fecundity and survival, a backwards elimination approach was used to test for the significance of all fixed effects (HMBPP treatment, body size) and their interactions, while controlling for random variation within each replicate. Mosquito body size was fitted as an additional fixed explanatory variable in all cases due to considerable influence in variation of mosquito feeding and fitness parameters (E. O. Lyimo and W. Takken, Med Vet Entomol, 7, 328-332 (1993)). Survival analysis was conducted using the Cox proportional hazards model in the R statistical software (v.3.1.1) to assess whether mosquito longevity (days until death) varied between experimental treatments. In this analysis, a frailty function was used to integrate the random effect of replicates into the Cox model with HMBPP treatment and mosquito wing length fitted as fixed effects. From this maximal model, non-significant terms were sequentially removed through backward elimination to reach the minimal statistically significant model (B. M. Bolker, Ecological models and data in R, Princeton University Press (2008)). Analysis was restricted to estimating variation in mosquito survival up to 14 dpi and death, sequentially. The difference between means of CO.sub.2 emission from treated versus control samples was analysed with a paired Student t-test using the IBM SPSS statistics 20.0 (IBM SPSS Inc., Chicago, Ill., USA) statistical software package. Experiment replication was treated as a random variable in all statistical mixed models. All data conformed to the assumptions of the test (normality and error homogeneity). In all mixed models, a maximal model was built that included fixed effects plus the random effects of the experimental replicates.

(40) Following the above-mentioned general experimental methods, the following example experiments were conducted.

Example 1—Direct and Indirect Effects of HMBPP on Mosquito Attraction and Feeding

(41) In a dual choice attraction bioassay, 95% of the host-seeking mosquitoes chose HMBPP-supplemented (hRBCs) over RBCs, suggesting the involvement of airborne factors derived from hRBCs. HMBPP-supplemented serum or glucose solution (5%) containing para-aminobenzoic acid (PABA 0.05%) did not, however, increase attraction, which indicates that the attraction is an RBC dependent and indirect effect (see FIG. 1A). The diol, (2E)-2-methylbut-2-ene-1,4-diol, a putative volatile form of HMBPP, had no effect on the attraction of mosquitoes to RBCs, indicating that the phosphate groups are required for the activity of HMBPP.

(42) RBC feeding rates were compared with hRBCs, P. falciparum asexual trophozoite- and gametocyte-infected RBCs (tiRBCs and giRBCs, respectively). The proportion of females fed more than doubled when hRBCs, tiRBCs or giRBCs were provided, respectively (see FIG. 1B). The amount of HMBPP released in the medium of giRBCs was investigated and found to be sufficient to stimulate mosquito blood feeding. The proportion of mosquitoes feeding on RBCs supplemented with HMBPP (hRBCs) or supernatants from giRBCs, respectively, were compared over a wide range of concentrations (see FIGS. 10 and D). This confirmed that 10 μM HMBPP, used in the experiments the results of which are shown in FIGS. 1A and B, corresponds to the concentration present in the undiluted supernatant from giRBCs, and also that substantially lower doses are sufficient to trigger enhanced mosquito feeding (see FIG. 10). Moreover, treatment of tiRBCs with a fosmidomycin derivative to block HMBPP synthesis (E. Yeh and J. L. DeRisi, PLoS Biol 9, e1001138 (2011)) resulted in a reduction of feeding to control levels, despite supplementation with the universal isoprenoid and downstream metabolite IPP (see FIG. 1D, triangle). Thus, it can be sene that HMBPP released from parasite-infected RBCs is a key metabolite for triggering mosquito feeding stimulation.

(43) To further decipher the phagostimulatory action of HMBPP, cell-free meals were provided to mosquitoes and the percentage of females that landed and initiated probing and feeding (referred to as percent persistency within 5 min) was examined (see FIGS. 1E and F). Approximately 80% of the mosquitoes displayed behavioral persistence when provided HMBPP-supplemented serum compared to 20% of those provided with serum alone (see FIG. 1E). Supplementation of a physiological salt solution with HMBPP generated a similar effect (see FIG. 1F). Hence, it can be seen that HMBPP acts as a phagostimulant that is neither dependent on factors from RBCs, nor from the serum. IPP displayed no phagostimulatory effect, suggesting a direct activity of HMBPP (see FIG. 1F).

Example 2—Attraction of Mosquitoes to Red Blood Cells (RBCs), RBCs Plus Media (M) or HMBPP-Supplemented RBCs (hRBCs)

(44) Using a larger Y-tube olfactometer, it was confirmed that hRBCs could mimic the attraction of mosquitoes to giRBCs, while IPP supplemented RBCs (ippRBCs) did not (see FIG. 2A). It was therefore shown that HMBPP indirectly stimulates attraction via the release of volatiles from RBCs, and acts directly as a feeding stimulant, whereas the structurally similar downstream IPP confers neither of these effects (M. E. Smalley, et al., Trans R Soc Trop Med Hyg, 75, 318-319 (1981) and A. L. Ouedraogo, et al., Acta Trop, 105, 28-34 (2008)). These findings all point to parasite-derived HMBPP modulatory effects within the malaria-infected human host affecting the blood seeking and feeding behaviors of its vector, the Anopheline mosquito.

(45) The volatiles released from hRBC, giRBC and RBCs were investigated. Since carbon dioxide (CO.sub.2), emitted from vertebrates through skin and breath, is a key activator and attractant for host-seeking anopheline mosquitoes (G. W. Frame, et al., J Invest Dermatol, 59, 155-159 (1972) and B. A. Omondi, et al., J Exp Biol., 218, 2482-2488 (2015)), CO.sub.2 emission was quantified. Through combined gas chromatography-mass spectrometry (GC-MS) (G. W. Frame, et al., J Invest Dermatol, 59, 155-159 (1972)) and quantitative respirometry CO.sub.2 release was estimated. The amount of CO.sub.2 released into the gas headspace above hRBC suspensions increased by 16% compared to RBCs alone (see FIG. 2B), an increase that is thought to be sufficient to promote mosquito attraction (B. A. Omondi, et al., J Exp Biol., 218, 2482-2488 (2015) and T. Dekker, et al., J Exp Biol, 208, 2963-2972 (2005)). However, carbon dioxide supplementation of RBCs not sufficient to reproduce the full mosquito attraction to hRBCs (see FIG. 2A), indicating the involvement of additional volatiles.

(46) The headspace above hRBC and giRBC suspensions was collected and the behavioral response of mosquitoes in the presence or absence of CO.sub.2 was assayed. The response to headspace extracts of hRBCs and giRBCs, in the presence but not in the absence of CO.sub.2, fully reproduced that of hRBCs (see FIG. 2A).

(47) Solid-phase microextraction (SPME) and GC-MS analyses of the headspace from hRBCs identified an increase in aldehydes (C8-C10:al; 1.7-to-5.2 fold) and monoterpenes (α- and ß-pinene, limonene; 1.2-to-1.6 fold) compared to that of the headspace of RBCs (see FIGS. 2C and D).

(48) The proportions of compounds identified in the headspace for hRBCs are as follows.

(49) TABLE-US-00011 Compound % by weight (of the combined compounds) α-pinene 7 β-pinene 5 limonene 18 octanal 23 nonanal 35 decanal 12

(50) A synthetic blend of these volatiles with CO.sub.2, at their enhanced natural emission rates and ratios, was able to reproduce the behavioral attraction of A. gambiae s.l. to hRBCs in a dose dependent manner (see FIG. 2E). The synthetic blend also attracted females more strongly than that of RBCs alone, both with and without CO.sub.2 (see FIG. 2E).

Example 3—Effects of HMBPP on Mosquito Fitness and Parasite Transmission Success and Schematic Model

(51) Females were fed on red blood cells (RBCs) or HMBPP-supplemented RBCs (hRBCs). The presence of HMBPP significantly increased the size of mosquito blood meals (see FIG. 3C), which was also shown to be independent of mosquito body size. Previous studies have indicated that the parasite can cause an increase in the amount of blood imbibed by the mosquito, which could potentially increase nutrient gain and enhance reproductive capacity of the vector (H. Hurd, Annu Rev Entomol, 48, 141-161 (2003)). Neither mosquito fecundity, nor survival was affected by HMBPP, despite the prompted larger blood meals (see FIGS. 3B and 3A).

(52) Inhibition of the Plasmodium MEP pathway is lethal without continuous IPP addition (E. Yeh and J. L. DeRisi, PLoS Biol 9, e1001138 (2011)). This was tested by feeding mosquitoes with giRBCs with or without additional HMBPP and subsequently monitoring the burden of infection in the mosquito during the parasite sporogonic period. Equal gametocyte density was used in both treatments (3%). Oocyst prevalence (proportion of oocyst-carrying mosquitoes) among groups of females was not significantly different between treatments (close to 100% in both groups), whereas oocyst intensity (number of oocyst per midgut) was higher in mosquitoes fed on HMBPP-supplemented giRBCs (see FIG. 3D). The addition of HMBPP to blood meals also resulted in significantly higher sporozoite prevalence (proportion of sporozoite-carrying mosquitoes) and intensity (number of sporozoites per salivary gland; (see FIG. 3E). Taken together, these findings suggest that HMBPP per se has no obvious deleterious effects on vector fitness, but increases the mosquito susceptibility to Plasmodium infection: parasite prevalence (proportion of parasite-transmissible mosquitoes) and intensity (parasite loads).