Abstract
The conservation of the mosquito indolergic receptors across the Culicinae and Anophelinae mosquito lineages, which spans 200 million years of evolution, is a testament to the central role of indolic compounds in the biology of these insects. Indole and skatole have been associated with the detection of oviposition sites and animal hosts. To evaluate the potential ecological role of these two compounds, we have used a pharmacological approach to characterize homologs of the indolergic receptors Or2 and Or10 in the non-hematophagous elephant mosquito Toxorhynchites amboinensis. We provide evidence that both receptors are narrowly tuned to indole and skatole like their counterparts from hematophagous mosquitoes. These findings indicate that indole and skatole are operating in a non-animal-host seeking context in Toxorhynchites and underscore the importance of understanding their roles in hematophagous mosquitoes.
1. Introduction
It is well-established that resource-locating mosquito behaviors are mainly mediated by olfactory signals (Takken and Knols, 1999; Zwiebel and Takken, 2004). However, only a few relevant animal/plant hosts and oviposition odorants have been identified (Davis and Bowen, 1994). In this regard, the ecological roles of indole and its close analog skatole (3-methylindole), two nitrogen-containing aromatic compounds, are complex (Figure 1A). Indole and skatole alone or as a mixture, have been proposed to act as oviposition attractants in Aedes aegypti (Baak-Baak et al., 2013), Culex spp. (Beehler et al., 1994; Blackwell et al., 1993; Du and Millar, 1999; Mboera and Takken, 1999; Millar et al., 1994; 1992; Mordue et al., 1992) and Anopheles gambiae (Lindh et al., 2008). These two compounds, products of the metabolic activity of the microflora, are present in significant amount in mammalian waste products (Garner et al., 2007; Yokoyama and Carlson, 1979), human skin (Bernier et al., 2000; 2002) and human sweat (Meijerink et al., 2000), which indicate they may also act as animal-host attractants (Cork, 1996). Indole is a ubiquitous component of flower scents of many plant families (Knudsen et al., 2006) and it has been identified from host plants of Ae. aegypti and An. gambiae (Nyasembe et al., 2018).
Both indole and skatole are electrophysiologically active compounds detected by the antennae of Ae. aegypti (Siju et al., 2010), C. quinquefasciatus (Du and Millar, 1999) and An. gambiae (Blackwell and Johnson, 2000; Meijerink et al., 2000; 2001; Qiu et al., 2006). Two highly conserved odorant receptors (ORs), OR2 and OR10, are highly sensitive and selective towards indole and skatole, respectively in Ae. aegypti (Bohbot et al., 2011), C. quinquefasciatus (Hughes et al., 2010; Pelletier et al., 2010) and An. gambiae (Carey et al., 2010; Wang et al., 2010). The Or2 and Or10 genes are expressed in the antennae of adult male and female mosquitoes, while Or2 is only expressed in the antennae of larvae along with a third paralog named Or9. The functional conservation of these receptors across the two mosquito subfamilies in both sexes and in larvae suggest that indole and skatole play important and multiple roles in the biology of these insects (Figure 1A) (Cork, 1996; Nyasembe et al., 2018).
Our objective was to explore the role of these two compounds in the context of animal-host selection by functionally characterizing candidate homologs of the indole and skatole receptor genes in the non-hematophagous elephant mosquito Toxorhynchites amboinensis. Lack of functional conservation would argue the case for a role of the OR2-indole and OR10-skatole pairs in animal-host seeking in hematophagous insects. Our study shows that T. amboinensis OR2 (TaOR2) and OR10 (TaOR10) share high sequence identity with their Aedes counterparts in support of a highly conserved role in mosquitoes outside animal-host seeking. We provide pharmacological evidence that the elephant mosquito TaOR2 and TaOR10 are indole and skatole receptors operating in a non-animal host context, including oviposition site selection and/or plant-host-seeking.
2. Materials and Methods
2.1. Cloning TaOr genes and sequence analyses
Cloning and sequencing of TaORco was described elsewhere (Dekel et al., 2016a). TaOr2 and TaOr10 were custom-synthesized (Bio Basic Inc., Markham Ontario, Canada), subcloned into into the pENTRTM vector using the GatewayR directional cloning system (Invitrogen Corp., Carlsbad, CA, USA) and subcloned into the Xenopus laevis expression destination vector pSP64t RFA. Plasmid purification was carried out using the The ZR Plasmid MiniprepTM-Classic (Zymo Research, Irvine, CA, USA) and sequenced by Macrogen Europe (Amsterdam, the Netherland). DNA and amino-acid sequences for TaOr2, TaOr10 and TaORco have previously been published (Zhou et al., 2014) and can be accessed at Figshare (https://figshare.com/articles/Transcriptome_assembly_of_T_ambionsis/2182684/2, last accessed on Dec 26, 2018).
Amino-acid sequence alignments (Supplementary Figure 1) were executed using MAFFT version 7 (Nakamura et al., n.d.). Phylogenetic analysis was performed using the neighbor-joining statistical function and 10,000 Bootstrap replications of the MEGA 7 software (Kumar et al., 2016).
2.2. Chemical reagents
For establishing the tuning curve, we used 30 odorants, including 19 compounds from Sigma-Aldrich (Milwaukee, WI, USA), including 1-hepten-3-ol (CAS 4938-52-7), 3-methylbutanol (CAS 123-51-3), E-2-hexen-1-al (CAS 6728-26-3), heptaldehyde (CAS 111-71-7), octanal (CAS 124-13-0)c,etpatreop(CylA-aS 109-60-4), 3-octanone (CAS 106-68-3), 6-methyl-5-hepten-2-one (CAS 110-93-0), 2,4,5-trimethylthiazole (CAS 13623-11-5), diallyl-sulfide (CAS 2179-57-9), benzaldehyde (CAS 100-52-7), indole (CAS 83-34-1), histamine (CAS 51-45-6), (+)-limonene oxide (CAS 203719-54-4), geranyl-acetate (CAS 105-87-3), (+)- fenchone (CAS 4695-62-9), 2-oxopentanoic acid (CAS 1821-02-9), (±)-1-octen-3-ol (CAS 3391-86-4) and 3-methylindole (CAS 83-34-1); 7 compounds from Merck (Darmstadt, Germany), including methyloctanoate (CAS 111-11-5), ethyl-hexanoate (CAS 123-66-0), 2-heptanone (CAS 110-43-0), dimethyl-sulfide (CAS 2179-57-9), tryptamine (CAS 61-54-1), octanoic-acid (CAS 124-07-2) and D-glucuronolactone (CAS 32449-92-6); 2 compounds from Acros Organics (Thermo Fisher Scientific, Waltham, MA, USA), including methyl salicylate (CAS 119-36-8) and octopamine (CAS 770-05-8); and 2 compounds from Alfa-Aesar (Ward Hill, MA, USA), including L-lactic acid (CAS 79-33-4) and δ-Decalactone (CAS 705-86-2).
2.3. Two-electrode voltage clamp electrophysiological recording of Xenopus oocytes expressing TaOR2, TaOR10 and TaORco
The methodologies and protocols used in this study have been described elsewhere (Bohbot and Dickens, 2009). Briefly, TaOr2, TaOr10 and TaORco cRNA were synthesized using linearized pSP64tRFA expression vectors as template for in vitro transcription according to the instructions of the mMESSAGE mMACHINE® SP6 Transcription Kit (ThermoFisher Scientific). Stage V-VI oocytes were manually separated and enzymatically defolliculated using a 1 mg/mL collagenase (Sigma-Aldrich, Milwaukee, WI, USA) solution (calcium-free ND96 buffer, [pH 7.6]) for 40-50 min at 18 °C. Oocytes were then successively washed in calcium-free ND96 and gentamycin-supplemented (10 mg/mL, Sigma-Aldrich, Milwaukee, WI, USA) calcium-free ND96. Oocytes were then washed and incubated in ND96 buffer supplemented with cal- cium (0.1 M), 5% heat-inactivated horse serum (ThermoFisher Scientific), 50 mg/ml tetracycline (Carl Roth GmbH), 100 mg/ml streptomycin (Sigma-Aldrich, Milwaukee, WI, USA) and 550 mg/ml sodium pyruvate (Sigma-Aldrich, Milwaukee, WI, USA) for four to five days. Oocytes were injected with 27.6 nL (27.6 ng of each cRNA) of RNA using the Nanoliter 2010 injector (World Precision Instruments, Inc., Sarasota, FL, USA). Odorant-induced currents of oocytes expressing TaOr2/10 and TaORco were recorded using the two microelectrode voltage-clamp technique (TEVC). The OC-725C oocyte clamp (Warner Instruments, LLC, Hamden, CT, USA) maintained a −80 mV holding potential.
For the establishment of concentration-response curves, oocytes were exposed to to indole or skatole alone 10−10 M to 10−4 M). Data acquisition and analysis were carried out with the Digidata 1550 A digitizer and pCLAMP10 software (Molecular Devices, Sunnyvale, CA, USA).
The tuning curve was generated using a panel of 30 odorants including indole, skatole and other compounds known to elicit physiological or behavioral responses in mosquitoes. All chemicals used were administered at 90 nM, which approximates the EC50 of indole and skatole. All the data analyses were performed using GraphPad Prism 7 (GraphPad Software Inc., La Jolla, CA, USA).
3. Results
3.1. TaOR2 and TaOR10 are highly conserved
The antennae of Toxorhynchites amboinensis express two highly conserved indolergic receptor homologs (TaOR2 & TaOR10). Multiple sequence alignments and phylogenic analyses show that they share about 80% amino-acid identity with their Aedes counterparts (Figure 1b). TaOR2 and AaOR2 encode 376 amino-acid proteins sharing 82.67% overall sequence identity. TaOR10 and AaOR10 encode 375 amino-acid proteins, which share 78.34% amino-acid identity. The OR10 alignment requires only one gap to be introduced (Supplementary Figure 1). Amino-acid divergence is evenly distributed throughout the peptide sequence. Both TaOR2 and TaOR10 grouped with their Aedes counterparts supported by bootstrap support with values above 95%.
3.2. TaOR2 and TaOR10 are highly sensitive to indolic compounds
Our next question was to investigate whether the observed sequence conservation determined functional orthology. To do so we expressed both receptors in the frog Xenopus oocyte system for pharmacological characterization.
Because Aedes and Anopheles OR2 are highly sensitive to indole (Bohbot et al., 2011; Wang et al., 2010), we challenged TaOR2 with ten-fold increasing concentrations of this odorant and skatole, a methylated analog of the former. Like its Aedes counterparts, TaOR2 proved to be a challenging receptor to express in Xenopus oocytes as it consistently generates comparatively lower currents than other mosquito ORs such as OR10 or OR8 (Figure 2A). The resulting concentration-response relationships derived from the maximum amplitudes elicited by each tested concentrations provided EC50 values of 88 nM for indole and 1,380 nM for skatole (Figure 2B). Indole was about 15 times more potent than skatole and the indole dynamic concentration range occurred in the nanomolar range.
We applied the same approach to TaOR10 informed by the sensitivity of AaOR10 towards skatole (Bohbot and Dickens, 2012). Indeed, AaOR10 dynamically responds to skatole in the nM range while it is much less sensitive to indole (mM range). TaOR10 consistently generated robust and larger currents than the TaOR2 paralog (Figure 2A). The relative potency of indole and skatole was however reversed with TaOR10 being 105 times more sensitive to skatole (EC50 = 87 nM) than to indole (EC50 = 9,163 nM) (Figure 2B).
3.3. TaOR2 and TaOR10 are narrowly tuned to indole and skatole, respectively
We further tested the odorant selectivity of these two receptors using a panel of 30 compounds belonging to diverse chemical classes, including alcohols, aldehydes, esthers, ketones, sulfur compounds, aromatics, amines, terpenes, carboxylic acids and lactones (Dekel et al., 2016b) (Figure 3B). In order to avoid receptor adaptation, antagonist effects and technical artefacts such as broad molecular receptive ranges (Bohbot and Pitts, 2015) associated with high chemical concentrations (Bohbot and Pitts, 2015), the screens were carried out at low 90 nM odorant concentration, which nearly corresponds to the EC50 values of the TaOR2-indole and TaOR10-skatole pairs. We controlled for possible position effects by administering odorant sets in reverse order (Supplementary Figure 2). At this concentration, indole and skatole elicited the strongest responses for their respective cognate receptors (Figure 3A). We did not observe any modulation of receptor activity in response to the cognate ligands at the end of the recording sessions.
Overall, TaOR2 exhibited a medium response profile with a kurtosis value of 19 (Figure 3B). TaOR10 was narrowly tuned to skatole with a maximal kurtosis value of 30. These findings confirm the selectivity of these two receptors for skatole and indole, respectively. The activation of TaOR2 elicited by indole was 2.7 times greater than the next most potent odorant ((±)-1-octen-3-ol) and 3.8 times greater than the third most active odorant (methyl-octanoate). The activation of TaOR10 by skatole was 23.1 times higher than the next most activating odorant (3-octanone) and 29 times greater than indole.
4. Discussion
It is unclear whether indole and skatole act as oviposition or as animal-host (Cork, 1996; Millar et al., 1992). To complicate the matter, these benzenoid compounds participate to the flower scent of plants (Jürgens et al., 2010; Smith and Meeuse, 1966)1 and are preferred by Ae. aegypti in the context of plant-host attraction (Nyasembe et al., 2018). To explore the ecological role of indole and skatole in the biology of blood-feeding mosquitoes, i.e., to determine whether these receptors may operate in a non-animal host seeking context, we have pharmacologically tested the responses of the OR2 and OR10 homologs from the strict-nectar feeding mosquito T. amboinensis. Our criteria for determining functional orthology included both pharmacological sensitivity and selectivity properties. On both accounts, we provide evidence that TaOR2 and TaOR10 are functional orthologs of their counterparts from blood-feeding mosquito species, as they are highly sensitive (nanomolar range EC50 values) and narrowly tuned (kurtosis values) to indole and skatole, respectively.
While, the biological significance between these two receptors in terms of tuning breadth remains unresolved, the nanomolar level sensitivity of these receptors do suggest cognate relationships between these pairs. TaOR2 and TaOR10 exhibit high-level of sensitivity (nM range) among T. amboinensis ORs. For comparison, the EC50 value for (R)-1-octen-3-ol in relation to TaOR8 is 401 nM (Dekel et al., 2016a), about 4 times higher than TaOR2-indole and TaOR10-skatole interactions. The high sensitivity of TaOR2 and TaOR10 towards indole and skatole underscores their fundamental ecological significance.
TaOR10 exhibited outstanding specificity towards skatole, second to OR8 towards (R)-1-octen-3-ol (Bohbot and Dickens, 2009; Dekel et al., 2016a). Using a panel of 29 compounds identical to the ones tested here but excluding skatole, we find that the kurtosis value (k = 29) for TaOR8 was maximal akin to TaOR10. (R)-1-octen-3-ol was 30.3 times more potent than the next most activating odorant. Such ligand specificity is suggestive of the ecological significance of this odorant. Additionally, such a high degree of specificity may reflect an adaptation for high signal to noise ratio at the peripheral level (Lu et al., 2007).
We have provided evidence that TaOR2 and TaOR10 share the same function as their counterparts from the blood-feeding mosquitoes Ae. aegypti, An. gambiae (Bohbot et al., 2011; Bohbot and Dickens, 2012; Wang et al., 2010) and Culex quinquefasciatus (Hughes et al., 2010; Pelletier et al., 2010). In Toxorhynchites sp., these receptors may mediate oviposition site selection (Collins and Blackwell, 2002) and exhibit sensitive physiological responses (Collins and Blackwell, 1998). Our findings exclude the role of indole and skatole in animal-host seeking as far as T. amboinensis is concerned and underscore the need to decipher the role(s) of these compounds in blood-feeding mosquitoes using detailed behavioral studies. The unusual sequence and functional conservations of OR2 and OR10 during mosquito evolution reflect the importance of indole and skatole to mosquito ecology and behavior. Indeed, not only do adult detect indole but the larva antennae of Ae. aegypti also express Or2, suggesting a separate ecological role of this compound in aquatic environments. The prevalence and abundance of indole and skatole present us with a challenge that is to understand their potential role in foraging, mate searching, habitat finding and oviposition site seeking. In addition, odorants can elicit different activities from different mosquito species (Xu et al., 2015), which means that indole and skatole may operate in different contexts within and between species.
The conservation and central role of the mosquito-specific Or2 and Or10 genes may be leveraged for the development of future mosquito control agents, including receptor (agonists and antagonists) and behavioral modulators (repellents and attractants). However, comprehensive behavioral studies are wanting to develop such tools for vector population control and personal protection.
Supplemental information
Supplementary Figure 1. (A) Amino-acid sequence alignment of OR2 and OR10 between Toxorhynchites amboinensis and Aedes aegypti. (B) Percentage of amino-acid sequence identity.
Supplemental Figure 2. The order of odorant administration does not affect the relative receptor activity. Representative current traces elicited by 90 nM of indole, skatole, (±)-1-octen-3-ol (1), heptaldehyde (2), propyl-acetate (3), 3-octanone (4), benzaldehyde (5) and octopamine (6) recorded from Xenopus oocytes co-expressing the TaOr2 and the TaORco receptor complex.
Acknowledgements
The authors They are grateful to Prof. Eitan Reuveny and Dr. Izhar Karbat from the Weizmann Institue of Science for their help with the frog oocytes. This work was supported by the by the Israel Science Foundation [grant number 1990/16].