The human odorant receptor OR10A6 is tuned to the pheromone of the commensal fruit fly Drosophila melanogaster

Summary All living things speak chemistry. The challenge is to reveal the vocabulary, the odorants that enable communication across phylogenies and to translate them to physiological, behavioral, and ecological function. Olfactory receptors (ORs) interface animals with airborne odorants. Expression in heterologous cells makes it possible to interrogate single ORs and to identify cognate ligands. The cosmopolitan, anthropophilic strain of the vinegar fly Drosophila melanogaster depends on human resources and housing for survival. Curiously, humans sense the pheromone (Z)-4-undecenal (Z4-11Al) released by single fly females. A screening of all human ORs shows that the most highly expressed OR10A6 is tuned to Z4-11Al. Females of an ancestral African fly strain release a blend of Z4-11Al and Z4-9Al that produces a different aroma, which is how we distinguish these fly strains by nose. That flies and humans sense Z4-11Al via dedicated ORs shows how convergent evolution shapes communication channels between vertebrate and invertebrate animals.

Naturally, this invites the question -how do humans smell the scent of the fly? A range of human ORs is tuned to straight-chain aldehydes, which are commonly found in fruit and vegetable aromas (Schmiedeberg et al., 2007;Saraiva et al., 2019;Nara et al., 2011;Li et al., 2014;de March et al., 2015;Block 2018) and perception of Z4-11Al might be encoded by one or even several of these aldehyde-responsive ORs. We hence submitted Z4-11Al to an in vitro screening of all human ORs and their most frequent genetic variants, using heterologous expression in HEK-293 cells and a luminescence-based assay (Noe et al., , 2017b. This screening renders OR10A6 the single most responsive receptor for Z4-11Al. A subsequent olfactory panel test confirmed the results of an in vitro dose-response test of synthetic aldehyde analogs, showing that we discriminate between structurally related aldehydes and that our olfactory perception of Z4-11Al is remarkably sensitive and specific. The scent of the fly illustrates how chemical ecology research inspires the discovery of OR ligands and provides an account for convergent chemical communication across phylogenies.

Sensory evaluation of fly odor
Comparative chemical analysis of volatiles released by D. melanogaster male and female flies, followed by sensory evaluation of fly odor and synthetic compound by a professional wine panel, strongly suggest that Z4-11Al is the scent of the female fly (Lebreton et al., 2017;Becher et al., 2018).
To substantiate these findings, we compared females vs males painted with Z4-11Al. For this particular experiment, assessors were chosen according to their capacity to recognize the scent of synthetic Z4-11Al, at 10 ng formulated in water, during preliminary experiments. Assessors evaluated fly odor emanating from glass vials, which contained 10 males and 10 females, respectively, during 30 min, 1 h before the experiment. All assessors readily distinguished male and female vials. After adding 10 or 100 ng of Z4-11Al, discrimination was no longer significant ( Figure 1A).
We further asked whether females transmit Z4-11Al or its precursor Z7,Z11-27Hy ( Figure 2B) to males during mating. All assessors readily distinguished between vials impregnated with the odor of mated and unmated males, respectively ( Figure 1B) and all assessors recognized the scent of Z4-11Al in vials impregnated by mated males.
We next compared synthetic Z4-11Al with the scent of D. melanogaster females, of the cosmopolitan and the Zimbabwe strains, in water and wine, providing a rich odorant background (Figure 2A). A professional wine panel was employed for this test, with no previous experience with Z4-11Al. Vials, where five fly females had been kept and released before testing, and vials formulated with 10 ng Z4-11Al, were filled with water or wine, respectively. In vials with water, panelists found the odor of Z4-11Al to resemble ll OPEN ACCESS Headspace analysis of D. melanogaster confirmed that females of the cosmopolitan strain released Z4-11Al, whereas Z4-9Al was not detected in cosmopolitan fly effluvia collections. Zimbabwe females, on the other hand, produced Z4-9Al, in addition to Z4-11Al, at a 2.6 G 0.7-fold amount (n = 10) ( Figure 2C). Assessors participating in these experiments sensed 10 ng synthetic Z4-11Al formulated in water, according to a preliminary test. (A) All assessors distinguished the odor of 10 males vs 10 females emanating from glass vials, where flies had been kept during 30 min, 1 h before the onset of the experiment (n = 10 and 12 assessors in upper and lower panel, respectively; Chi2 = 10.0752, p = 0.001503 and Chi2 = 12.0, p = 0.005). Sex discrimination was no longer significant, when 10 and 100 ng Z4-11Al, respectively, was added to the male vials (Chi2 = 0.5952, p = 0.4404; Chi2 = 0.1777, p = 0.6733). (B) All assessors discriminated the odor of mated vs unmated males, emanating from glass vials, where flies had been kept during 30 min, 1 h before the experiment (n = 11 assessors; Chi2 = 10.9925, p = 0.0009). The control experiment with unmated males did not show differences (Chi2 = 1.0017, p = 0.3169).  (Olender et al., 2012(Olender et al., , 2013Trimmer and Mainland, 2017), showed by far the strongest response to 30 mmol/L of Z4-11Al beyond a 2s-threshold, in a screening of 616 human OR variants expressed in HEK-293 cells ( Figure 3A). Screening the aldehyde analog Z4-9Al at 100 mmol/L revealed two responding receptors beyond a 2s-threshold, OR2W1 and OR10A6 L 287 P, with comparable amplitudes ( Figure 3B). At 100 mmol/L Z4-11Al, OR2W1 showed an about six-fold lower response, compared with OR10A6 L 287 P ( Figure 3C), despite a slightly better surface expression of OR2W1 as compared with OR10A6 L 287 P, in both HEK-293 and NxG 108CC15 cells ( Figure S1).
A dose-response assay further confirmed that Z4-11Al was the most potent agonist for OR10A6 L 287 P, compared with the analog Z4-9Al and the positional isomer Z6-11Al ( Figure 4A). The EC 50 values for these three aldehydes on OR2W1 haplotypes were about 2-to 3-fold higher, throughout, compared with OR10A6 L 287 P (Table 1, Figure 4B). All other ORs that responded to Z4-11Al or Z4-9Al beyond a 2s-threshold in the screening experiments could not be validated in concentration-response assays ( Figures S2, S3, and S4), suggesting <2% false positives. Among the most abundant OR10A6 haplotypes (Olender et al., 2012), OR10A6 L 287 P was functional ( Figures 4A and S4).
For the ensuing panel test, 31 assessors were chosen at random. The odorant panel corroborated that we are more sensitive to Z4-11Al than to Z4-9Al or Z6-11Al ( Figure 4C). A significant number of panelists sensed Z4-11Al already at 1 ng/mL in water (0.006 mmol/L). In comparison, the response to Z4-9Al or Z6-11Al was not significant, at the amounts tested.
A low response threshold to Z4-11Al in vitro (Figures 3 and 4) corroborates our remarkable sensitivity to the female pheromone of cosmopolitan D. melanogaster, which is only a minor compound of fly headspace ( Figure 2C; Lebreton et al., 2017). Most panelists who discriminated Z4-11Al from control ( Figure 4C) perceived the aroma of the pure compound to be fruity and pleasant, whereas a fly female or synthetic Z4-11Al was found to disturb wine aroma (see also Becher et al., 2018).
According to the triangle test shown in Figure 4C, 21 of 31 panelists (68%) sensed Z4-11Al at 1 ng. In a large human population, only 35% individuals carry the functional haplotype OR10A6 L 287 P (Olender et al., 2012(Olender et al., , 2013, (Table 2) which is contradictory at first sight. We therefore genotyped 29 of the 31 panel members for OR10A6 and OR2W1. The functional variant OR10A6 L 287 P was found in 14 individuals (48.3%), and OR2W1 D 296 N was found in 26 individuals (89.7%), among the 29 panelists.
In a triangle test (3 vials), where 2 vials are the same, panelists are asked to pick the odd vial. The 29 anonymously genotyped panelists produced 20 correct answers ( Figure 4C shows results for 31 panelists).
Assuming that the 14 panelists carrying OR10A6 L 287 P picked the odd vial because they sensed Z4-11Al, and that every third of the remaining 15 panelists picks the odd vial by chance, we expect 19 correct answers, which is appreciably close to the 20 correct answers obtained with 1 ng Z4-11Al. The number of correct answers, from these 29 panelists, increased to 24 at 10 ng Z4-11Al. This might be owing to OR2W1 D 296 N, which shows a 43% higher EC50 value in response to Z4-11Al in the HEK assay, compared with OR10A6 L 287 P (Table 1). Figure 2. Sensory discrimination between cosmopolitan and Zimbabwe D. melanogaster females by humans, and production of Z4-11Al and Z4-9Al by these fly strains (A) Olfactory resemblance of 10 ng synthetic Z4-11Al and the odor of cosmopolitan or Zimbabwe female flies, in water and wine. Judges, who had not sensed synthetic Z4-11Al before, were asked whether or not the odors of two vials bear resemblance. Bars marked with asterisks are significantly different (n = 21 assessors; Chi2 = 6.9146, p = 0.0085; Chi2 = 10.5061, p = 0.0012; Chi2 = 4.9082, p = 0.0267; Chi2 = 0.0288, p = 0.8652, from left to right). (B) Z7,Z11-27Hy is the hydrocarbon precursor of the cosmopolitan D. melanogaster female pheromone Z4-11Al. Females of the Zimbabwe strain further produce Z5,Z9-27Hy and the corresponding oxidation product is Z4-9Al.

OPEN ACCESS
iScience 25, 105269, November 18, 2022 5 iScience Article Discrimination between Z4-11Al and a blend of Z4-9Al and Z4-11Al As we perceive Z4-11Al at the amounts produced by females, and not Z4-9Al, Zimbabwe and cosmopolitan flies should smell the same (Figures 2A and 2D; 4C). Females of these flies are expected to differ only with respect to odor intensity, not quality -unless a blend of Z4-11Al and Z4-9Al produces a different aroma than Z4-11Al alone. This is indeed the case. A triangle test involving 45 randomly selected panelists at SLU Alnarp shows a clear distinction between Z4-11Al and a 3:10-blend A B C Figure 3. Screening of 616 recombinant human IL-6-HaloTagâ-OR variants, with Z4-9Al and Z4-11Al, using an HEK-293 cell-based GloSensorä cAMP-luminescence assay The cDNA expression plasmid OR library is shown in Table S1. (A) OR10A6 L 287 P emerges as the sole OR responding to 30 mmol/L of Z4-11Al, beyond a 2s-threshold (red line). OR2W1 (dashed arrow) became activated only at higher concentrations (see Figures 3C and 4). (B) Both OR2W1 and OR10A6 L 287 P were activated by 100 mmol/L of Z4-9Al. Data for both screenings (relative luminescence units, RLU) were normalized to the signal amplitude of OR1A1 in response to 30 mmol/L R-(À)-carvone. OR families are color-coded and sorted in an ascending numerical order. The negative controls were cells transfected with a ''mock'' plasmid lacking any receptor coding region; false positives (FP) are indicated. (C) OR2W1 shows a significantly lower amplitude than OR10A6 L 287 P in response to Z4-11Al, but not in response to Z4-9Al. Data show mock control-subtracted raw data (luminescence units, LU) in response to 100 mmol/L of the respective aldehyde (mean G SD, n = 3), the asterisk shows a significant difference (paired two-tailed t-test; t = À4.14887, p = 0.0142).

Z4-11Al is a ligand for the highly expressed OR10A6
Human olfactory perception of Z4-11Al, the female pheromone of cosmopolitan D. melanogaster is highly sensitive and specific. An odorant panel sensed synthetic Z4-11Al at 1 ng/mL (0.006 mmol/L) and at subnanogram amounts released by single flies, and distinguished Z4-11Al from the structurally similar aldehydes Z4-9Al and Z6-14Al, or a blend of Z4-11Al and Z4-9Al. In vitro screening showed that the functional variant of OR10A6, which ranks among the most highly transcribed ORs in the olfactory epithelium (Saraiva et al., 2019;Verbeurgt et al., 2014), is most sensitively tuned to Z4-11Al. OR10A6 may also be tuned to other odorants, such as cyclamen aldehyde (Duroux et al., 2020), and further experiments are needed to map its entire agonist space.
OR2W1, the other receptor that showed a significant response to the Drosophila aldeydes in our OR screening, is a most broadly tuned human OR (  iScience Article A combination of highly selective and broadly tuned ORs is the basis for sensing a diverse odorant environment with only a limited number of ORs Block, 2018;Saraiva et al., 2019;Kurian et al., 2021). Theoretical and data-based models predict that mixed OR populations of different receptive ranges enable greater odor coverage (Alkasab et al., 2002;Fonollosa et al., 2012). OR2W1 may accordingly participate together with OR10A6 in enabling receptor activity patterns in response to aldehydes, even though the transcript levels of OR2W1 in the olfactory epithelium are low, in comparison with the highly expressed OR10A6 (Verbeurgt et al., 2014;Saraiva et al., 2019). Other human ORs with an affinity to various odor-active aldehydes (Nara et al., 2011;de March et al., 2015;Block, 2018) did not respond significantly to Z4-11Al and Z4-9Al.
Remarkably, the panel reliably discriminated between Z4-11Al and a blend of Z4-11Al and Z4-9Al. That the blend afforded an entirely different hedonic quality explains how we distinguish between cosmopolitan and Zimbawe flies. Response of OR2W1 to Z4-9Al and of OR10A6 to both compounds, Z4-9Al and Z4-11Al, suggests that input from 2 ORs produces a different perception. On the other hand, as OR10A6 was responsive to both compounds, it is even possible that modulation at the OR level encodes this blend discrimination. Processing of odorant interactions is not restricted to higher olfactory circuits, but occurs even peripherally, owing to synergistic and antagonistic responses of olfactory neurons to odor mixtures. Odorant interaction and encoding of mixtures at the OR level substantially extends the receptive range of ORs (Brann and Datta, 2020;De March et al., 2020;Inagaki et al., 2020;Xu et al., 2020).
It is further intriguing that we sense the small amounts of Z4-11Al released by single flies against the rich bouquet emerging from a glass of wine. Z4-11Al is only a minor compound among the volatiles released by Drosophila females (this study; Lebreton et al., 2017;Becher et al., 2018), whereas the bouquet of wine is overwhelmingly complex and comprises many volatiles at far larger amounts, including a suite of aldehydes (Swiegers et al., 2005;Cullere et al., 2007).
Single ORs and their key ligands play indeed a central role in olfactory object recognition, especially against heterogeneous backgrounds. Olfactory sensory neurons expressing high-affinity ORs with low activation thresholds have been shown to become activated early during a sniff and thus accentuate the response to behaviorally salient signals, whereas input from other ORs is temporarily tuned down (Wilson et al., 2017;Arneodo et al., 2018;Bolding and Franks, 2018;Dewan et al., 2018). The odorant panel attributed a pleasant, fruity aroma to Z4-11Al as a single compound. That the admixture of Z4-11Al to wine is perceived as unpleasant may accordingly be owing to a reduced or modulated perception of wine volatiles. Similarly, wine aroma is disturbed by larger amounts of (E)-2-decenal, a component of a hemipteran bug defensive secretion (Mohekar et al., 2017). Conversely, a suite of unsaturated, odor-active aldehydes from coriander, including (E,E)-2,4-undecadienal, had a deodorizing effect on the malodor of porc intestines (Kohara et al., 2006;Ikeura et al., 2010).
Taken together, our observations illustrate how a key compound contributes to olfactory perception via a single OR, in addition to combinatorial coding of odorant blends by arrays of several ORs (Mainland et al., 2014). Sensitivity is, in addition to ligand affinity, a function of OR expression in olfactory sensory neurons (van der Linden et al., 2020) and OR10A6 is among the most highly expressed ORs in our nose (  Highly abundant ORs are plausibly dedicated to odorants of critical physiological, behavioral or ecological function. This raises the question of what Z4-11Al may mean to us. Perception of the same odorant by insects and vertebrates is convergent, as the respective ORs share ligand affinity, but are built differently and lack a common phylogenetic root (Su et al., 2009;Bear et al., 2016). If the convergent evolution of the vertebrate and invertebrate olfactory systems reflects an underlying logic rather than shared developmental principles , it would follow that convergent perception of messenger chemicals points to a behavioral role.
What is the source of Z4-11Al in a human odorscape? Animals, plants, and associated microbes each release many hundreds of compounds and these volatile emissions change with age, phenology, and physiological state (Knudsen et al., 1993;El-Sayed, 2020;Lemfack et al., 2018;Ljunggren et al., 2019). Z4-11Al has not been searched for, synthetic standards are not available commercially, and we can therefore safely assume that the occurrence of Z4-11Al is only incompletely known.
Taken together, Z4-11Al is found in human food, it might even be produced by ourselves and could manifest food, social context, or both. A dual function of certain pyrazines as key food odorants and semiochemicals, selectively activating the same, single human OR, has recently been demonstrated (Marcinek et al., 2021). iScience Article Role of Z4-11Al in D. melanogaster The vinegar fly is our involuntarily domesticated animal, since it accompanied the human global expansion out of Africa. Cosmopolitan vinegar flies are associated with us on all continents and most climate zones, they are strictly anthropophilic, depend on our food and dwellings for survival and we share a taste for fermenting food (Lachaise and Silvain, 2004;Nielsen et al., 2017;Arguello et al., 2019). D. melanogaster females, not males, produce dienic hydrocarbons that give rise to monoenic aldehydes, which is why we smell the female flies (Everaerts et al., 2010;Lebreton et al., 2017).
The sibling species Drosophila simulans has also attained worldwide distribution in association with humans, but is, unlike D. melanogaster, not a strict commensal and more rarely found in households or buildings (Lachaise and Silvain, 2004). D. simulans females do not produce dienic hydrocarbons, which is a main element of the mating barrier between these species. The cuticular hydrocarbon Z7,Z11-27Hy promotes courtship in cosmopolitan D. melanogaster, and suppresses interspecific matings with D. simulans, owing to differential, species-specific coding of Z7,Z11-27Hy in neural circuits mediating reproductive behavior (Billeter et al., 2009;Billeter and Wolfner, 2018;Seeholzer et al., 2018;Sato and Yamamoto, 2020).
Cosmopolitan and African D. melanogaster strains also differ with respect to cuticular hydrocarbons. The female-specific desaturase gene desat2, which affords Z5,Z9-27Hy and Z4-9Al, is functional only in African and not in cosmopolitan flies (Dallerac et al., 2000;Grillet et al., 2012). This hydrocarbon polymorphism yields a distinctive aldehyde blend, which is how we differentiate the scent of these two fly strains. Species-specific differences in hydrocarbons align with corresponding aldehyde signatures, that entail behavioral consequences. Z4-11Al attracts D. melanogaster, but not males of the Zimbabwe strain, and has an antagonistic effect on upwind flight attraction in D. simulans. This underlines the role of female-produced volatile pheromones in long-range mate communication in Drosophila (Lebreton et al., 2017;Borrero-Echeverry et al., 2022).
Panel tests evaluating male and female fly odor unexpectedly discovered that Z4-11Al, in addition to its hydrocarbon precursor, is among the ''chemical words exchanged by Drosophila during courtship and mating'' (Jallon, 1984). At close range, Z4-11Al stimulates courtship in males (Borrero-Echeverry et al., 2022), whereas the transfer of Z4-11Al may be a factor in reducing courtship success of freshly mated males (Scott et al., 1988). That Z4-11Al by itself is attractive to females (Lebreton et al., 2017;Borrero-Echeverry et al., 2022), even points to an antagonistic interaction with a male-produced compound, such as 11-cisvaccenyl acetate.

Sensory drive and convergence
Convergent perception of Z4-11Al in humans and flies could be coincident or interconnected. ORs readily adapt to habitats and to dietary or social chemosensory niches, in insects and vertebrates alike (Bear et al., 2016;Hughes et al., 2018;Saraiva et al., 2019). Transcript variants of the fly receptor DmelOR69a (Robertson et al., 2003) are tuned to food odorants and the female pheromone, respectively, and are co-expressed in the same OSNs (Lebreton et al., 2017). These twin ORs yield a degree of freedom for the acquisition of new ligands, if only they match the food and mate-finding theme.
Habitat selection and specific mate recognition are tightly interconnected (Paterson, 1985;Endler, 1992;Boughman, 2002), and the interaction between natural and sexual selection has been shown to affect cuticular hydrocarbon composition and mate recognition in D. melanogaster (Blows, 2002). We are food and home to the flies, they depend on us for survival. A commensal lifestyle is expected to generate a sensory drive and select for odorants to mediate fly aggregation and premating communication -if these odorants are produced by the flies, and if they are, in addition, characteristic elements of human odor scenes. Convergent perception of Z4-11Al is reminiscent of dedicated olfactory channels for geosmin that alert flies and humans about the presence of mold, which is detrimental for all animals (Maga, 1987;Stensmyr et al., 2012).

Conclusion
Sensing the scent of a single fly is out of the ordinary, especially as the cue is the fly's sex pheromone . Yet, only the discovery that a most highly expressed human OR is tuned to this pheromone underlines the biological significance of this observation. Sensitive and specific perception encourages the hypothesis ll OPEN ACCESS 10 iScience 25, 105269, November 18, 2022 iScience Article that Z4-11Al is found in human habitats, where humans, domesticated animals, or shared food resources, including associated microorganisms, could be the source.
Ambient odorscapes contain countless chemicals of yet unknown activity. Our study highlights how the identification of key OR ligands leads to the discovery of messenger chemicals and delivers insights into how chemical communication interconnects species across phylogenies. Regrettably, we can barely speculate what the fly pheromone may mean to us and whether it signals food, social context, or both. Satisfying our curiosity is an excellent reason to pursue, as the vinegar fly continues to afford fundamental discoveries and studying fly sex perfumes may perhaps teach us about our own.

Limitations of the study
The significance of Z4-11Al for humans is yet unknown. Z4-11Al and close analogs are found in in food, and mediate communication between animals. In the search for sources in human environments, close attention must be paid to occurrence of trace amounts of Z4-11Al, in view of our sensitivity. Screening for other ligands, combined with structure-activity studies, will help to elucidate the behavioral relevance of the OR10A6 channel. Last but not least, we cannot entirely exclude that yet other ORs or OR variants may participate in the perception of Z4-11Al.

STAR+METHODS
Detailed methods are provided in the online version of this paper and include the following:

ACKNOWLEDGMENTS
We thank the members of the sensory panel (Systembolaget, Stockholm) and our colleagues at SLU Alnarp for evaluating the fly scent, as well as Jan-Eric Englund (SLU Alnarp) for statistical analysis. Three anonymous reviewers raised constructive questions, leading to additional experiments and substantial improvements. This study was supported by the Leibniz-Institut fü r Lebensmittel-Systembiologie (TU Mü nchen) and the Faculty of Landscape Architecture, Horticulture, and Crop Production Science (SLU, Alnarp).

Lead contact
Further information and requests for resources should be directed to and will be fulfilled by the Lead Contact, Peter Witzgall (peter.witzgall@slu.se).  Table S2 This paper N/A Primers -vector internal, see Table S3 This paper N/A Primers for site-directed mutagenesis, see Table S4 This paper N/A Primers for haplotype sequencing, see d Additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

EXPERIMENTAL MODEL AND SUBJECT DETAILS Insects
Cosmopolitan (Dalby) and Zimbabwe (S-29, Bloomington) strains of D. melanogaster were reared on a standard sugar-yeast-cornmeal diet at room temperature (25 G 2 C) and 50 G 5 rH under a 12:12-h L:D photoperiod. When preparing experiments, eclosing flies were collected every 4 h and sexed, according to the sex comb on the third segment of the male forelegs. Flies were tested when 3 days old. Presence of meconium was used as a distinguishing feature for virgin flies. Females and males were kept separately in 30-mL Plexiglas vials with fresh food.

Sensory panel
The Swedish Alcohol Retailing Monopoly (Systembolaget) continuously monitors product quality. A sensory panel of professional assessors at Systembolaget consisted of 9 women and 12 men, with an average age of 41.8 G 10.9 y. At SLU Alnarp, panel members were recruited from personnel at the Department of Plant Protection Biology and the Department of Biosystems and Technology, 18 women and 27 men, with an average age of 38.4 G 14.3 years.
Panel members were informed about experimental hypotheses and protocols, the scope of the study, and potential risks. Results cannot be traced to individual persons and the study is therefore exempt from ethical review. Informed consent was obtained from all subjects and the study was approved by a local ethics committee and the Legal Affairs Unit at the Swedish University of Agricultural Sciences.
For the OR screening assays, the following chemicals were used: Dulbecco's MEM medium (

Pheromone collection and chemical analysis
Sixty unmated 3-d-old cosmopolitan and Zimbabwe females (n = 9 and n = 10, respectively) were transferred to standard glass rearing vials (24.5 3 95 mm, borosilicate glass; Fisher Scientific, Sweden), which had been baked at 350 C overnight. After 24 h, the flies were removed and the vial was rinsed with 200 mL of hexane, containing 100 ng decanal as internal standard, in an ultrasonic water bath for 3 min. The solvent was transferred to 1.5 mL GC-MS vials with insert and condensed to ca. 5 mL in a fume hood.

OPEN ACCESS
iScience 25, 105269, November 18, 2022 iScience Article Two mL of the solvent rinses were analyzed by gas chromatography-mass spectrometry (GC-MS) (6890 GC and 5975 MS, Agilent, Santa Clara, CA, USA) on a fused silica capillary column (60 m 3 0.25 mm), coated with HP-5MS UI (d f = 0.25 mm; Agilent). Injections were made in splitless mode (30 s), at 275 C injector temperature. The GC oven was programmed from 50 to 250 C at 8 C/min (2 and 10 min hold, respectively) and a final temperature of 275 C, the mobile phase was helium (34 cm/s). The MS operated in scanning mode. Aldehydes were identified by direct comparison of mass spectra and retention data with synthetic standards.

OR expression and screening
Cell culture and transient DNA transfection Human embryonic kidney (HEK-293) cells were cultivated in Dulbecco's MEM medium (DMEM: w 3.7 g/L NaHCO3, w 4.5 g/L D-glucose, w/o L-glutamine, w/o Na-pyruvate) supplemented with 10% fetal bovine serum (FBS superior), 2 mM L-glutamine, 100 units/mL penicillin, and 100 mg/mL streptomycin in 10 cm cell culture dishes at 37 C, 5% CO2, and 100% humidity, as test cell systems for the functional expression of recombinant ORs (Geithe et al., 2015Noe et al., 2017a). One day before transfection, HEK-293 cells were transferred with a density of 12000 cells per well in white 96-well plates (Thermo Scientificä Nuncä F96 MicroWellä, white, #136102, Thermo Fisher Scientific, Waltham, USA). The transfection was done by the cationic lipid-transfection method using 100 ng OR plasmid-DNA, 50 ng olfactory G-protein Gaolf (Jones and Reed, 1989;Shirokova et al., 2005), 50 ng RTP1S (Saito et al., 2004), 50 ng Gg13 (Li et al., 2013), and 50 ng genetically modified cAMP-luciferase pGloSensorä-22F (Binkowski et al., 2009) (Promega, Madison, USA), each with ViaFectä Transfection Reagent. As negative control, transfection of an empty pFN210A-vector-plasmid (mock) was employed. As positive control OR1A1 was transfected on each plate. For concentration-response relations, each transfection was done in triplicate on the same 96-well plate. For receptor screening experiments, all 391 human OR wild-types plus 225 of their most frequent haplotypes (altogether 616 OR variants) were transfected in duplicates. The entire OR library, including official gene symbols, haplotypes, and sequence accession numbers is given in Table S1. The cells were taken into experiment 42 h post-transfection (Geithe et al., 2015Noe et al., 2017a).

cAMP luminescence assay
Cell culture media of the transfected HEK-293 cells in the 96-well plates was replaced 1 h prior to the luminescence measurement with physiological salt solution containing 140 mmol/L NaCl, 10 mmol/L HEPES, 5 mmol/L KCl, 1 mmol/L CaCl2, 10 mmol/L D-glucose and 2% D-luciferin, pH 7.4. After this incubation, basal luminescence signals for each well (three consecutive data points, 60 s intervals) were recorded with the GloMaxâ Discover Microplate Reader (Promega, Madison, USA) before odorant application. As positive control, 30 mmol/L (R)-(À)-carvone was applied on the OR1A1 transfected cells. Odorant stock solutions were prepared in DMSO, and diluted 1:1000 into the physiological salt solution containing 0.05% Pluronicâ PE 10500, as solvent mediator. Final DMSO concentration on the cells was 0.1%. Four min after odorants were applied to the cells, three consecutive data points at 60 s intervals were recorded for each well with the GloMaxâ Discover Microplate Reader.

Data analysis of cAMP luminescence measurements
OR library screenings. The raw luminescence data obtained from Sparkâ multimode microplate reader was transferred to Excel. Data points of basal level and data points after odorant application were each averaged. From each luminescence signal, the corresponding basal level was subtracted and afterwards normalized to the amplitude of the reference odorant-receptor pair (OR1A1 vs. 30mM R-(À)-carvone) on each 96-well plate. The normalized values for each receptor measured in duplicates were averaged and plotted alongside with the signal derived from mock-transfected cells. Signals above a 2s-threshold (average of all signals plus 2-times Standard Deviation) were considered as positive hits and objected to further analysis, such as concentration-response relations. False positives were defined as signals R 2s, which did not show a concentration-dependent activation in subsequent experiments.
Concentration-response relations. The raw luminescence data obtained from the GloMaxâ Discover Microplate Reader were processed as followed. For each well, the average of the three data points before odorant addition was subtracted from the average of the three data points after stimulation (Dsignal). Then, the corresponding mock of each substance/concentration was subtracted from each Dsignal value. All

Ethics statement
Participants were informed about the aim of the study, potential risks, the experimental protocol and they all provided formal consent. The study was conducted in accordance with the ethical principles for research involving human subjects developed in the Declaration of Helsinki (WMA). The study is exempt from ethical review, according to the Legal Affairs Unit at the Swedish University of Agricultural Sciences (SLU.ua.2022.2.2-3234).

QUANTIFICATION AND STATISTICAL ANALYSIS
Statistical tests for significance were calculated with a paired, two-tailed t-test and a Chi2-test, using Prism 9.3 (GraphPad) and ANOVA followed by Friedman's test (SAS). Statistical significance was defined as a p-value <0.05. p-values and sample sizes are given in the results section or the figure legends.