Evolution of the Coeloconic Sensilla in the Peripheral Olfactory System of Drosophila Mojavensis

Evolution of the coeloconic sensilla in the peripheral olfactory system of Drosophila mojavensis

A thesis submitted to the

Graduate School

of the University of Cincinnati

in partial fulfillment of the

requirements for the degree of

Master of Science

in the Department of Biological Sciences

of the College of Arts and Sciences

Daniel Carl Nemeth

B.S. in Zoology

Miami University

May 2015

Committee Chair: S. M. Rollmann, Ph.D.

ABSTRACT

Distinct environmental conditions characterized by habitat, climate, resource availability, predation, and competition can result in the development of adaptive traits through natural selection. Divergence of adaptive traits can be seen in the olfactory systems of phytophagous insects with differing host plant preferences among conspecific populations. In this study, we examine the cactophilic fly, Drosophila mojavensis, which utilizes different host cacti across its range. Each cactus emits its own, unique composition of volatiles, which may serve as a primary sensory cue for host plant localization. Specifically, we study the olfactory system of these flies in attempts to elucidate mechanisms underlying the reported phenotypic divergence among populations. We observed variation in odor-evoked electrophysiological responses of acid and amine sensing olfactory sensory neurons (OSNs) of coeloconic sensilla across all sensillar types among populations. Additionally, we observed loss of expression an odorant receptor gene,

Or35a in all four populations. Expansion to other Drosophila species revealed comparable absences in OSN responses to OR35a agonists in D. virilis and D. arizonae, suggesting a potential similar loss in gene expression as seen in D. mojavensis. Furthermore, responses of ac3

OSNs significantly differed to sixteen of the seventeen odorants tested across the four drosophilid species. Collectively, our results highlight evolution of acid and amine sensing in coeloconic sensilla through peripheral changes observed in Drosophila olfactory systems.

iii

ACKNOWLEDGMENTS

The fruition of this thesis would have not been possible without the help and support of so many great people. First and foremost, I need to thank my brilliant advisor, Dr. Stephanie M.

Rollmann, who not only supported me as an educating mentor, but pushed and motivated me to accomplish much more than I thought I was capable of. The skills I have learned during my graduate career under her guidance, will be carried with me onto my future endeavors. I would also like to thank my two committee members, Dr. John E. Layne, for his expertise on electrophysiology and all his help with the statistical analysis of my project, and Dr. Ann M.

Ray, for her insightful critiques and assistance as we determined the full construct of my research project.

I would like to extend a thanks to all of my lab colleagues during my graduate work at the

University of Cincinnati, especially my lab mates Dr. Elizabeth Brown, Amber Crowley-Gall,

Vishnu Gomez, and Cody Patterson for their helpful discussions about my research, as well as their help with any fly work if I was out of town, as to not delay my data collection upon returning. Also, a special thanks to Dr. Byrappa Ammagarahalli Munishamappa, for his electrophysiological knowledge and help with the categorization of ac4 sensilla, as well as his company during the countless of hours of our antennal dissections.

Lastly, I would like to extend many thanks to my parents, Meg and Bill, my siblings and in-laws, Jen, Greg, Chris, April, and Steph as well as my relatives and close friends for their constant, loving support and encouragement throughout my graduate career and my life. I would not be where, nor who I am today without their help, and I cannot thank them enough.

TABLE OF CONTENTS

ABSTRACT ...... ii

ACKNOWLEDGMENTS ...... iv

TABLE OF CONTENTS ...... v

LIST OF TABLES & FIGURES ...... vi

1. INTRODUCTION ...... 1

2. MATERIALS & METHODS ...... 11

2.1 Fly stocks ...... 11

2.2 Electrophysiology ...... 11

2.2.1 Odorant stimuli ...... 11

2.2.2 Single sensillum recordings (SSR) ...... 12

2.3 Gene expression analysis: RT-PCR ...... 13

2.4 DNA sequencing of ionotropic receptor coding regions ...... 14

3. RESULTS ...... 15

3.1 Characterization of OSN response profiles for coeloconic sensilla...... 15

3.2 Electrophysiological differences among D. mojavensis populations ...... 16

3.3 Olfactory receptor gene expression of predicted ac3 sensilla orthologs...... 17

3.4 Population differences in olfactory receptor gene sequence ...... 18

3.5 Species differences in electrophysiological responses of the ac3 sensilla ...... 18

4. DISCUSSION ...... 20

5. REFERENCES ...... 29

6. FIGURE LEGENDS ...... 40

LIST OF TABLES & FIGURES

Table 1. List of chemical compounds used in electrophysiological recordings ...... 42

Table 2. RT-PCR Primers...... 43

Figure 1. Single sensillum recordings of D. mojavensis populations ...... 44

Figure 2. OSN classification with known D. melanogaster odor response spectra...... 45

Figure 3. Statistical differences in electrophysiological responses among the four D. mojavensis populations...... 46

Figure 4. Single sensillum recording from an ac3 sensillum...... 47

Figure 5. RT-PCR analysis of ionotropic receptor gene expression in all four D. mojavensis populations...... 48

Figure 6. (a) Amino acid sequence differences in IR75a among D. mojavensis populations...... 49

Figure 6. (b) Amino acid sequence differences in IR75bc_GI13611 among D. mojavensis populations ...... 50

Figure 7. (a, b) Single sensillum recordings of drosophilid species...... 51

1. INTRODUCTION

The process of environmental adaptation is a driving force in evolution, which encompasses the ability of an organism to survive in differing environments. Organisms that have colonized new environments may acquire specific adaptive traits that better suit them to local conditions.

Differences in environmental conditions such as habitats (terrestrial and aquatic), climate

(temperature and precipitation), resource availability (food and breeding sites), predation, and competition, can apply differing selection pressures, which may differentially adapt populations to different habitats (Schluter, 2001; Linn et al., 2003; Rundle and Nosil, 2005). This process may lead to morphological, physiological, and behavioral shifts, and divergence in traits between populations in natural conditions (Funk, 1998; Schluter, 2001; Fuse, 2017). These shifts have been documented in the lab through artificial environmental adaptation as well. For instance, a genetic line of the fly Drosophila melanogaster, dubbed ‘Dark fly’ because it has been reared in the dark for over 1500 generations, exhibited divergent morphological, physiological, and behavioral characteristics in its sensory system in response to living in complete darkness, aiding its ability to thrive under these environmental conditions (Fuse et al., 2014; Fuse, 2017). With the genetic resources available for D. melanogaster, ‘Dark fly’ not only presents a valuable genetic tool in our quest to better understanding the mechanisms underlying environmental adaptation in nature, but simply demonstrates the ability of how organisms are able to adapt.

Differential adaptation in response to distinct environmental conditions between populations with standing genetic variation can induce divergence within a given species complex (Schluter, 2001; Linn et al., 2003; Rundle and Nosil, 2005). To study natural examples of adaptive divergence, host plant interactions of phytophagous insects have shown to be favorable models (Via, 1999; Pophof et al., 2005; Date et al., 2013; Nishida, 2014). Host plant

recognition and preference behavior is driven by oviposition and feeding needs, and involves olfactory, visual, and gustatory cues (Kopp et al., 2008; Hussain et al., 2016). Insects can shift preference to an alternative host plant due to the various environmental factors mentioned earlier, which can lead to differential adaptation and divergent evolution of their sensory systems (Via,

1999; Schluter, 2001; Rundle and Nosil, 2005; Nishida, 2014). To illustrate this, consider two examples. The first occurs in the D. melanogaster species subgroup, which is made up of nine sibling species, the majority of which are generalists, feeding and ovipositing on rotting fruits and vegetation (Stensmyr et al., 2003a; Dekker et al., 2006; Kopp et al., 2008). One member of this group that is not a generalist is D. sechellia, a strict specialist on the ripe, unspoiled fruit of

Morinda citrifolia (Legal et al., 1994; Farine et al., 1996; Amlou et al., 1998; Dekker et al.,

2006; Mansourian and Stensmyr, 2015; Prieto-Godino et al., 2017). This host specialization of

D. sechellia is a behavioral shift from its sibling species of the melanogaster species subgroup because ripe morinda fruit contains high acid content, specifically octanoic acid, making it toxic to most insects, and subsequently avoided by generalists in the melanogaster subgroup, meanwhile D. sechellia has developed resistance to the toxicity of octanoic acid, enabling it to exploit this resource (Legal et al., 1994; Farine et al., 1996; Amlou et al., 1998; Jones, 2005).

Analysis of the headspace volatiles of morinda fruit revealed that octanoic acid constitutes over half of all volatiles, while a second acid, hexanoic acid, is the next most abundant compound at

19% (Legal et al., 1994; Farine et al., 1996; Amlou et al., 1998). Intriguingly, D. sechellia also exhibited much greater attraction to all acids tested when compared to D. melanogaster (Dekker et al., 2006). In the second example of adaptive divergence, the tephritid fly, Rhagoletis pomonella, consists of two populations in North America that have shifted host plant preference.

Within the last 150 years, apple-infesting populations have diverged from the ancestral

populations that fed on hawthorn (Linn et al., 2003, 2005; Olsson et al., 2006a, 2006b).

Behavioral studies on both R. pomonella populations have shown that they detect volatiles from both host plants, regardless of their natal host, yet still exhibit upwind flight towards their natal host plant volatiles. Non-host plant volatiles disrupt their upwind flight patterns, demonstrating the vital role olfactory preference plays in host plant localization (Linn et al., 2003, 2005; Olsson et al., 2006a, 2006b). The behavioral shifts of both D. sechellia and R. pomonella in identifying and exploiting host plants within their environments are models for understanding adaptive divergence and the subsequent evolution of novel characteristics in the olfactory system.

An environment can be characterized by its existing sensory information and a newly experienced environment often warrants novel sensory stimuli. Therefore, the process of environmental adaptation can manifest as alterations in sensory organs, and in the case of D. sechellia and R. pomonella, changes in olfactory sensitivity towards host-specific volatiles.

Indeed, adaptations in their olfactory systems can be seen in distinct morphological and physiological differences in peripheral olfactory machinery (Funk, 1998; Schluter, 2001;

Crowley-Gall et al., 2016; Fuse, 2017). Consider again the two examples above. A comparison of D. melanogaster and D. sechellia revealed that D. sechellia has morphologically distinct antennae, with shorter sensilla, a greater number of a particular sensillar type, and alterations to a specific olfactory sensory neuron (OSN) which increased its sensitivity to key odorants emitted from the morinda fruit, like methyl hexanoate and hexanoic acid, all mediating its ability to exploit this resource (Stensmyr et al., 2003a; Dekker at al., 2006; Mansourian and Stensmyr,

2015; Prieto-Godino et al., 2017). Likewise, the divergent olfactory preferences of R. pomonella were accompanied by sensory differences apparently related to their host plant. The host plants emit suites of odorants that are distinct from one another (Olsson et al., 2006a), and this is

reflected in the physiological differences among the R. pomonella populations in the sensitivity and firing patterns of their OSNs to the host volatiles (Olsson et al., 2006b). The behavioral and physiological shifts seen in the two R. pomonella populations, as adaptations to their respective host plants, has created an initial barrier to gene flow between the populations (Schluter, 2001;

Nosil et al., 2005; Rundle and Nosil, 2005). This is because mate choice in some phytophagous insects is centered around host plant choice, which therefore creates a premating barrier between the diverging populations, based primarily on their olfactory preference for the different host plant volatiles (Linn et al., 2003; Dekker et al., 2006; Olsson et al., 2006a, 2006b).

Detection and discrimination of odorants is accomplished by the olfactory system, which in drosophilids is composed of two olfactory organs: the third segment of the antenna, and the maxillary palp. Each organ is covered by tiny porous sensory hairs called sensilla, of which there are of three morphological types: basiconic, trichoid, and coeloconic sensilla; the third antennal segment possesses all three types, while the maxillary palp only contains basiconic sensilla

(Stocker, 1994; Reddy et al., 1997; Clyne et al., 1999; Yao et al., 2005; Benton et al., 2009).

Sensillar types are also distinct in that they are hypothesized to be tuned to different classes of chemical volatiles; basiconics are hypothesized to detect basic food odors such as alcohols, esters, and ketones, trichoids aid in mate recognition by detecting different classes of pheromone molecules, and coeloconics are tuned to the detection of acids and amines, signaling overly ripe fruit or bacterial degradation of protein-rich substrates, respectively (Stocker, 1994; van der Goes van Naters and Carlson, 2007; Kopp et al. 2008; Benton et al., 2009; Ai et al., 2010; Mansourian and Stensmyr, 2015). Sensilla house dendrites of up to four OSNs, depending on the sensillar type (Hallem and Carlson, 2004; Vosshall and Stocker, 2007; Abuin et al., 2011; Silbering et al.,

2011; Ai et al., 2013). Localized to the dendrites of OSNs are different transmembrane protein

complexes called olfactory receptors, which bind odorant molecules that enter the sensillar pores

(Stocker, 1994; Vosshall et al., 1999; Stensmyr et al., 2003b; Hallem and Carlson, 2004;

Vosshall and Stocker, 2007; Benton et al., 2009; Hussain et al., 2016; Prieto-Godino et al.,

2017). Once a receptor becomes bound and activated by an odorant ligand, it undergoes a conformational change, allowing selected cations to flow across and depolarize the neuronal membrane, which then propagates action potentials down the axons of the OSNs toward distinct glomeruli located within the antennal lobe, analogous to the olfactory bulb in mammals (Stocker,

1994; Clyne et al., 1999; Vosshall et al., 1999, 2000; Hallem and Carlson, 2004; Vosshall and

Stocker, 2007; Min et al., 2013; Rytz et al., 2013; Bell and Wilson, 2016). Glomeruli are laterally connected throughout the antennal lobe via local interneurons, which inhibit other glomeruli to filter out the “noise” and send a more refined signal to the higher centers of the brain. The glomeruli propagate this processed information by synapsing onto second order neurons called projection neurons which project their axons to higher brain centers, specifically the mushroom body and lateral horn of the protocerebrum (Stocker, 1994; Vosshall et al., 1999,

2000; Hallem and Carlson, 2004; Masse et al., 2009; Min et al., 2013; Rytz et al., 2013; Hussain et al., 2016). The OSNs innervating basiconic and trichoid sensilla generally express one odorant receptor (OR), along with a co-receptor, ORCO, whereas the OSNs innervating coeloconics do not express any ORs (with one exception, OR35a), but rather express a second family of olfactory receptors called ionotropic receptors (IRs) (Vosshall et al., 2000; Yao et al., 2005;

Couto et al., 2005; Vosshall and Stocker, 2007; Benton, 2007; Benton et al., 2009; Min et al.,

2013). These receptors are believed to be ancestrally derived from ionotropic glutamate receptors

(Rytz et al., 2013). Four functional classes of coeloconic sensilla (ac1 – ac4), which each house either two or three IR-expressing OSNs have been identified in D. melanogaster (Yao et al.,

2005). There are a total of sixteen Ir genes expressed in the antenna of D. melanogaster, ten of which are expressed in the OSNs that innervate the coeloconic sensilla and are odor-specific ligand receptors. Two of the IRs, Ir8a and Ir25a, are particularly broadly expressed, and act as co-receptors for acid-sensitive and amine-sensitive IRs, respectfully, and are necessary for proper receptor function (Benton et al., 2009; Abuin et al., 2011; Silbering et al., 2011; Ai et al.,

2013; Min et al., 2013; Rytz et al., 2013). The remaining four antennal Ir genes are not expressed in coeloconic sensilla but are found in two other sensory structures of the antenna, the sacculus and the arista. IR64a of the sacculus is co-expressed with IR8a and has been shown to form a functional olfactory receptor that mediates acid detection (Ai et al., 2010, 2013). The two other

IRs of the sacculus, both co-expressed with IR25a, IR93a and IR40a have demonstrated to be involved in a non-chemosensory modality and mediate hygrosensation (Knecht et al., 2016). The remaining aristal IR neuron, IR21a co-expressed with IR25a, has been hypothesized to be involved in thermosensation (Benton et al., 2009; Rytz et al., 2013).

Flies and other insects utilize olfaction for various behaviors such as localization of food and oviposition sites, toxin avoidance, and mate recognition (Stocker, 1994; Joseph et al., 2009;

Ai et al., 2010; Mansourian and Stensmyr, 2015; Kim et al., 2017). The IR olfactory subsystem is relatively conserved across all insects as these antennal expressed chemosensors are present in many species such as mosquitos, moths, and honey bees (Liu et al., 2010; Croset et al., 2010;

Bengtsson et al., 2012). The malaria mosquito, Anopheles gambiae utilizes its IR orthologs to detect different acids and amines like carboxylic acids and ammonia, which act as the primary olfactory cues when seeking out a new blood host (Smallegange et al., 2005; Qiu et al., 2006;

Liu et al., 2010; Croset et al., 2010). The ability to detect acids and amines is vital in drosophilids as well. The majority of species constituting the D. melanogaster subgroup are

generalists and feed on a wide variety of decaying fruit and plant matter (Stensmyr et al., 2003a;

Dekker et al., 2006; Kopp et al., 2008). D. melanogaster primarily feeds on yeast growing on fermenting fruit, which it locates with the help of its coeloconic sensilla by detecting pyruvic acid, a byproduct of the yeast induced alcohol fermentation process occurring within the fruit

(Carson and Stalker, 1951; Silbering et al., 2011; Stensmyr et al., 2012; Mansourian and

Stensmyr, 2015). Other acids, like acetic acid, which is produced by bacterial oxidation of ethanol, have been shown to promote oviposition and upwind flight, yet trigger behavioral avoidance when directly contacted by D. melanogaster (Amlou et al., 1998; Joseph et al., 2009;

Becher et al., 2010; Ai et al., 2010; Mansourian and Stensmyr, 2015; Kim et al., 2017). Acids are not the only class of compound released by bacteria in fermenting fruit, various amines are produced as well, which are essential molecules for various cellular functions, like regulation of ion channel function and cell proliferation (Min et al., 2013; Heby, 1981; Ramani et al., 2014;

Mansourian and Stensmyr, 2015; Hussain et al., 2016). Amines such as 1,4–diaminobutane, spermidine, and phenethylamine are emitted from the bacterial degradation of protein-containing organic substrates, which in regards to drosophilids may serve as an olfactory cue for the localization of nutrient-rich food and oviposition sites (Min et al., 2013; Mansourian and

Stensmyr, 2015; Hussain et al., 2016).

These various exploitations of the acid and amine detecting olfactory machinery seen across drosophilids can result in species-specific food and oviposition site preferences, which may be accompanied by changes to the sensory periphery that are reflected in physiological variations of IR-expressing OSNs. One mechanism that has been shown to influence olfactory preference is an alteration to the peripheral olfactory system. For example, D. sechellia has undergone shifts at the periphery, such as the recently found amino acid substitution in the

ligand-binding domain (LBD) of one of its acid-sensitive IRs, which contributed to increased sensitivity to hexanoic acid, a primary odorant of the morinda fruit (Legal et al., 1994; Farine et al., 1996; Amlou et al., 1998; Jones, 2005; Dekker at al., 2006; Prieto-Godino et al., 2017). This increased sensitivity to hexanoic acid may serve as a primary volatile cue for host plant localization, shaping the olfactory preference of D. sechellia to exploit this resource. This is an evolutionary shift in the acid-sensing olfactory pathway of D. sechellia because when compared to D. simulans, its most closely related species, D. simulans exhibits avoidance behavior when presented with hexanoic acid (Amlou et al., 1998; Jones, 2005). This difference in olfactory preference may have created a premating barrier between the last common ancestor of D. sechellia and D. simulans. And in general, a premating barrier decreases encounter rates between conspecific populations allowing them to become spatially separated, further restricting gene flow, potentially altering host adaptive traits in sensory systems, which can lead to reproductive isolation, and eventually the formation of new species (Rundle and Nosil, 2005; Schluter, 2001;

Schluter and Conte, 2009; Sobel et al., 2009; Date et al., 2013).

To gain further insight into the evolution of adaptive traits, it is most edifying to study a species with clear phenotypic divergence among conspecific populations. Drosophila mojavensis is an example of such, as it consists of four separate populations, each specializing on a different host plant. Endemic to the arid deserts of North America, D. mojavensis is a phytophagous insect, that uses the fermenting necrotic cactus tissue for food and oviposition sites (Fogleman and Abril, 1990; Newby and Etges, 1998). The four populations of D. mojavensis are largely isolated by geographical barriers which have allowed divergence in their odor-guided host plant preferences. The populations endemic to Baja California feed on pitaya agria (Stenocereus gummosus), organ pipe cactus (Stenocereus thurberi) is used by the mainland Sonora population,

the Santa Catalina Island population feeds on prickly pear cactus (Opuntia spp), and the population inhabiting the Mojave Desert exploits barrel cactus (Ferocactus cylindraceus)

(Newby and Etges, 1998; Smith et al., 2102; Date et al., 2013, 2017). Studies on the chemistry of the host plants have revealed that each species of cactus emits its own, unique suite of volatile odorants, which differ in the types and/or relative amounts of chemical compounds (Fogleman and Abril, 1990; Date et al., 2013, 2017). A principal component analysis of the headspace volatiles from cactus rots produced in the laboratory revealed that barrel cactus and prickly pear cactus were distinct, but the two Stenocereus cacti, agria and organ pipe, overlapped considerably (Date et al., 2013, 2017). Previous studies among cacti revealed acids as the most abundant in the headspace, in that butyric, acetic, and propionic acid are present in variable amounts amongst agria and organ pipe rots (Fogleman and Abril, 1990). A more recent study on the volatiles from naturally occurring field-rot cacti of organ pipe (S. thurberi) and prickly pear

(O. littoralis) revealed a trend with the types of volatiles emitted and the stage of rot, which is of particular note as D. mojavensis flies are more attracted to the beginning stages of rot in fermenting cactus tissue (Fogleman and Foster, 1989; Date et al., 2013). Both the naturally occurring “early rot” of organ pipe and the natural necrotic cactus tissue of prickly pear were dominated by carboxylic acids, butyric, hexanoic, octanoic, and nonanoic acid for organ pipe and isobutyric, butyric, and hexanoic acid for prickly pear, however the “late rot” of organ pipe only had one carboxylic acid present, palmitic acid (Wright and Setzer, 2014). The sources of these carboxylic acids are most likely the resulting byproducts of fermenting yeast and bacteria that are infecting the rots (Wright and Setzer, 2014; Mansourian and Stensmyr, 2015). In regards to barrel cacti, to date there is little evidence of these acids being emitted from rots, however there is less work on this species (Date et al., 2013; Rollmann unpublished). These chemical

differences between cactus rots provide variation in environmental signals, which may serve as a primary sensory cue for host plant identification in this system.

Insect olfactory systems have been studied extensively and more and more work has been conducted on the OSNs that express ORs, to understand the neural-sensory contributions to population divergence (Stensmyr et al., 2003a, 2012; Dekker at al., 2006; Olsson et al., 2006a,

2006b; Kopp et al., 2008; Crowley-Gall et al., 2016). Previous work on the olfactory system of

D. mojavensis populations has revealed divergence in the neurophysiological responses from entire populations of OSNs on the antenna to host plant volatiles through electroantennograms

(EAGs) (Date et al., 2013), and this work was further refined to reveal cell-specific differences in

OSN specificity, sensitivity, and abundance (Crowley-Gall et al., 2016). These studies have expanded knowledge of OR-based olfactory circuitry and its role in population divergence greatly. However, the importance of acid and amine content conveyed as a potential environmental cue, coupled with the IR-expressing OSNs detecting them as a prime contributor to population divergence, warrants examination. Here, we examine D. mojavensis to begin further teasing apart the intricacies of population divergence. Because of the differential volatile compositions of each host plant, particularly the acids and amines, this study examines the acid- and amine-sensitive IR-expressing OSNs housed within coeloconic sensilla to elucidate a potential mechanism that underlies the phenotypic divergence of D. mojavensis populations.

2. MATERIALS & METHODS

2.1 Fly Stocks

Drosophila mojavensis and Drosophila arizonae stocks were maintained on a standard banana cactus medium, while Drosophila melanogaster and Drosophila virilis stocks were raised on a standard cornmeal food medium. All stocks were reared at 25°C on a 12 h light/dark cycle. The

Mojave population (A997b: Providence Mountain, CA, USA), Baja population (SQ59a: San

Quintin, Baja California, MX), and Sonora population (OPNM9: Organ Pipe National

Monument, Arizona, USA) stocks were a gift from Dr. Bill Etges. The S. Catalina population

(stock number 15081-1352.22), D. arizonae (stock number 13081-1271.33), and D. virilis (stock number 15010-1051.87) were from the Drosophila Species Stock Center (San Diego, CA, USA).

The D. melanogaster stock (Canton-S: stock number 64349) was obtained from the Bloomington

Drosophila stock center (Bloomington, IN, USA).

2.2 Electrophysiology

2.2.1 Odorant stimuli

Chemical compounds were purchased from Sigma-Aldrich at the highest purity available (Table

-2 1) and were diluted in paraffin oil or sterilized H2O, as appropriate, in order to obtain a 10 dilution. Twenty microliters of odorant was loaded on a filter paper inserted into a Pasteur pipette (53/4", Fisherbrand, Waltham, MA, USA). A given odorant cartridge was used a maximum of two times during a given recording period. Diagnostic odorants were delivered first to determine sensillum identity based on D. melanogaster work (Silbering et al., 2011): 2- oxopentanoic acid and ethanolamine for ac1; 1,4-diaminobutane and pyridine for ac2; butyric acid, butyraldehyde, and hexanol for ac3; phenylacetaldehyde and phenethylamine for ac4.

2.2.2 Single sensillum recordings (SSR)

Female D. mojavensis, D.virilis and D. arizonae aged 9-12 days old, and D. melanogaster aged

2-10 days were used in the study (de Bruyne et al., 2009; Massie and Markow, 2005; Prieto-

Godino et al., 2017). A fly was restrained in a pipette tip with its third antennal segment protruding outside and held in place by a thin glass electrode. Humidified air flowed continuously over the antennae. A microscope with a 50x objective (Olympus BX40, Center

Valley, PA, USA) was used to visualize the coeloconic sensilla. Action potentials were obtained with electrolytically (1M KOH) sharpened tungsten microelectrodes (cat# M210, MicroProbes for Life Sciences, Gaithersburg, MD, USA). A reference electrode (also sharpened tungsten) was inserted into the eye and the recording electrode into the base of a sensillum. The signal from the recording electrode was amplified using an IDAC-4 (Syntech, Kirchzarten, Germany), and recorded and analyzed using Autospike software (Autospike v.3.9 Syntech, Kirchzarten,

Germany). The microscope and insect preparation were mounted on an anti-vibration table

(TMC, Peabody, MA, USA) and shielded by a Faraday cage. An odorant stimulus was applied in a 500 ms puff with inter-stimulus intervals of at least 30 s. A minimum of five recordings for each sensillum type and fly line were obtained. Action potentials were counted without distinguishing spike sizes due to difficulties in reliably distinguishing the spikes of individual neurons (Silbering et al., 2011). Responses were determined by the difference in number of action potentials between 0.5 s pre- and post- stimulation periods and the difference was calculated and subsequently doubled to obtain the odor-evoked spikes per second. Differences in electrophysiological responses between populations/species were determined using a one-way analysis of variance (ANOVA) and corrected for multiple testing by controlling false discovery rate (FDR of 0.05) (Benjamini et al., 2006). Minor odor-evoked responses (-15 spikes/s to 25

spikes/s) were excluded from the analyses. Hierarchical cluster analysis was performed using

MATLAB software (MathWorks, Natick, MA, USA). Euclidean geometry was used to determine the distance between pairs of observations in odorant response space. Clustering of the four D. mojavensis populations and D. melanogaster coeloconic sensilla was based on the responses of all coeloconic sensilla (ac1 – ac4) from all five drosophilids to 36 odorants, while the clustering of ac3 sensilla from four drosophilid species was based on their responses to 17 odorants, each odor-evoked OSN response comprising one data point in the multidimensional response space. These points were then grouped into discrete clusters based on their distances using Ward’s Clustering Criteria (similar to de Bruyne et al., 1999; Crowley-Gall et al., 2016).

D. melanogaster physiological data in the literature was estimated from various studies; majority were estimated from Figure 1 of Silbering et al., 2011; ac3 data from Figure 3 of Yao et al.,

2005, with absences of any remaining physiological data from Figure 1 of Prieto-Godino et al.,

2017.

2.3 Gene expression analysis: RT-PCR

Third antennal segments (500-600) from 10-11 day old female flies were dissected for each D. mojavensis population. Antennae were dissected directly into 700 µl of chilled Trizol reagent

(Invitrogen, Carlsbad, CA, USA) and either extracted immediately or stored at -80°C. The tissue was then homogenized, and chloroform was added followed by 15 s of vigorous shaking. After a

2 min 30 s incubation period at room temperature, the samples were spun at full speed in a centrifuge at 4°C for 15 min. The supernatant was removed and mixed with 70% ethanol and subsequently further isolated using the RNeasy Micro Kit (Qiagen, Hilden, Germany) according to manufacturer instructions. Two biological replicates were generated per population. cDNA was synthesized from 100 ng of RNA using an AccuScript High Fidelity 1st strand cDNA

Synthesis Kit (Agilent, Santa Clara, CA, USA) according to manufacturer instructions. A negative control (without reverse transcriptase (RT)) was also generated for each sample.

Samples prepared with cDNA template (RT) and negative control (no RT) were amplified with primers (Table 2) using standard PCR methods. Samples were also prepared with genomic DNA template to ensure the effectiveness of the primers. Gene expression was examined for the genes

Or35a, Ir75bc_GI13610, Ir75bc_GI13611, and the housekeeping gene, Actin 5C (Act5C).

Amplicons were separated by gel electrophoresis on a 2 % agarose gel and visualized with ethidium bromide. Each resulting gene amplicon was sequenced at the Cincinnati Children’s

Hospital Medical Center DNA Sequencing and Genotyping Core (Cincinnati, OH, USA) to confirm appropriate amplification.

2.4 DNA sequencing of ionotropic receptor coding regions

The coding region of the genes Ir75a and Ir75bc_GI13611 was amplified using primers that were designed with overlapping segments to span the entire coding region. DNA was isolated from entire bodies of 30 adult females flies for each D. mojavensis population using the

PUREGENE DNA Isolation Kit (Qiagen, Hilden, Germany) according to manufacturer instructions. DNA samples were amplified using standard PCR methods. PCR products were subsequently purified using the MinElute Reaction Cleanup Kit (Qiagen, Hilden, Germany) according to manufacturer instructions. The purified PCR products were sequenced at the

Cincinnati Children’s Hospital Medical Center DNA Sequencing and Genotyping Core

(Cincinnati, OH, USA). Sequence analysis and alignments were performed using Seqman Pro

Software Version 12.2.0 (DNASTAR, Madison, WI, USA). The assembled coding region sequences of each population were translated using the ExPASy Bioinformatics Resource Portal

(Translate tool, http://web.expasy.org/translate/) and aligned using GenomeNet (ClustalW,

http://www.genome.jp/tools/clustalw/). The S1 lobe, S2 lobe, and pore loop of the Ligand

Binding Domain (LBD) of each gene were determined from Prieto-Godino et al., 2016, 2017.

3. RESULTS

3.1 Characterization of OSN response profiles for coeloconic sensilla

Electrophysiological recordings from coeloconic sensilla on the third antennal segment of D. mojavensis were conducted using a suite of 47 chemical compounds (Table 1) for all four populations. This suite of compounds consisted of volatiles previously reported to be emitted from the fermenting host plants as well as known diagnostic odorants for D. melanogaster coeloconic sensilla (Fogelman and Abril, 1990; Yao et al., 2005; Benton et al., 2009; Silbering et al., 2011; Date et al., 2013). Four distinct antennal coeloconic sensillar types were identified (ac1

– ac4) and categorized based on their odor-evoked response properties (Figure 1). These types varied in the number and type of acids and amines eliciting responses, and in the strength of responses to the odorants.

Hierarchical cluster analysis of D. mojavensis OSN responses along with those of the data available from the existing D. melanogaster literature further support the presence of four homologous sensillar types (Figure 2; Yao et al., 2005; Silbering et al., 2011; Prieto-Godino et al., 2017). The OSN response properties within these types were qualitatively similar to those of

D. melanogaster (Yao et al., 2005; Benton et al., 2009; Silbering et al., 2011; Prieto-Godino et al., 2017), with a few exceptions. For instance, in ac2 sensilla, increased excitatory responses to butyric acid were observed relative to propionic and acetic acid (Silbering et al., 2011).

Additionally, the odor-evoked response properties of ac3 sensilla differed in several notable

ways. There was little to no response to hexyl acetate, (E)-2-hexenal, and hexanol, all of which elicit strong excitatory responses from OR35a-expressing OSNs located within D. melanogaster ac3 sensilla (Yao et al., 2005; Benton et al., 2009; Prieto-Godino et al., 2017). On the other hand, there was an excitatory response to 1,4-diaminobutane, similar to D. melanogaster, where it is dependent on OR35a (Yao et al., 2005; Prieto-Godino et al., 2017), as evidence by a lack of response to this odorant in an OR35a mutant line (Silbering et al., 2011). Also, responses were increased to spermidine, phenethylamine, and pyridine relative to D. melanogaster (Prieto-

Godino et al., 2017). Finally, ac4 sensilla showed diminished responses to hexanoic acid

(Silbering et al., 2011; Prieto-Godino et al., 2017).

3.2 Electrophysiological differences among D. mojavensis populations

OSN response profiles differed among the four D. mojavensis populations for all four sensillar types (Figure 1 – 3). Among populations and across sensillar types, the Baja and Sonora profiles for each type clustered closest together. The Mojave population profiles, on the other hand, stood apart from the other three populations, with the exception of ac2s (Figure 2). This can be attributed to a few factors. For instance, compared to the other populations, the Mojave population exhibited decreased responses to all acids in all sensillar types, with the exception of ac1 sensilla, where the Mojave population showed the greatest response to 2-oxopentanoic acid.

In ac2 sensilla, responses to amines by the Mojave and Catalina populations were generally greater than those of other two populations (with the exception of the response to phenethylamine from the Catalina population). Furthermore, the Sonora population demonstrated the greatest responses to various acids, specifically butyric acid (Figures 1, 3). Moreover, the ac3

OSN response spectrum of the Mojave population was overall decreased for the majority of odorants when compared to the other three populations and the Sonora population exhibited the

greatest sensitivity to propionic acid (Figures 1, 3). Finally, ac4 OSN responses of the Mojave population showed significant inhibitory responses to three odorants – acetic acid, 2- oxopentanoic acid, and (E)-2-hexanal when compared to the other three populations (Figures 1,

3).

3.3 Olfactory receptor gene expression of predicted ac3 sensilla orthologs

As noted above, the response profile of ac3 sensilla showed a marked difference to D. melanogaster in that it showed little to no response to odorants which are characteristic of

OR35a-expressing OSNs located within ac3 sensilla (Yao et al., 2005; Prieto-Godino et al.,

2017) (Figure 1). The ac3 sensilla contain two OSNs, one co-expressing IR75a, IR75b, and

IR75c and the second OR35a together with its OR co-receptor, ORCO and IR76b. IR76b is co- expressed in all four sensillar types and its role as a putative co-receptor and/or ligand detector remains to be determined (Abuin et al., 2011; Silbering et al., 2011). To investigate whether there is evidence for the loss of the neuron itself or a lack of Or35a expression, we revisited our single sensillum recordings and performed RT-PCR, respectively. Examination of our recordings revealed that it is unlikely that the limited response to OR35a diagnostic odorants is the result of a loss of the neuron itself because at least two different amplitudes were observed during electrophysiological recording, supporting the presence of at least 2 different OSNs within the sensillum (Figure 4). Therefore, we investigated whether the genes predicted to be expressed in ac3 sensilla, particularly Or35a, are in fact detected in the antenna of D. mojavensis. RNA was isolated from antennal tissue from each of the four D. mojavensis populations and RT-PCR was performed. Expression was examined for the Ir75b and Ir75c orthologs, namely Ir75bc_GI13610 and Ir75bc_GI13611, as well as Or35a (Flybase 2.0 at beta.flybase.org). Ir75a expression was not investigated because it is predicted to be expressed in both ac2 and ac3 sensilla. RT-PCR

results revealed that expression of Ir75bc_GI13610 and Ir75bc_GI13611 was found in all four populations. However, Or35a expression was not detected (Figure 5). The lack of Or35a expression is consistent with the absence of ac3 responses to known diagnostics of OR35a- expressing neurons (Yao et al., 2005; Benton et al., 2009; Mansourian and Stensmyr, 2015;

Prieto-Godino et al., 2017). Additionally, we wish to note that in a RNA-seq study examining D. mojavensis population differences in odorant receptor expression, Or35a was also not clearly detected (Crowley-Gall et al., 2016).

3.4 Population differences in olfactory receptor gene sequence

Population differences in ac3 responses exist to select acids and amines. These changes in ac3 response profiles among populations can be due to changes in the amino acid sequence of the receptor (Figure 1, 3). To tease apart the molecular mechanisms underlying the population differences in responses of ac3 sensilla, we sequenced the coding region of the genes Ir75a and

Ir75bc_GI13611. Eight amino acid substitutions in IR75a were found among populations. None were within the ligand binding domain (LBD) of this receptor, which consists of the S1 and S2 domains (Figure 5a; Benton et al., 2009; Prieto-Godino et al., 2016, 2017). In the case of

IR75bc_GI1361, nine substitutions were found among populations. Three were located in the S1 lobe of the LBD and a fourth located in the putative M1 transmembrane segment of the ion channel (Figure 5b; Benton et al., 2009; Prieto-Godino et al., 2016, 2017).

3.5 Species differences in electrophysiological responses of the ac3 sensilla.

To further investigate divergence in the ac3 sensilla response profiles across species, we examined three additional drosophilids: D. melanogaster, D. virilis, and D. arizonae. Divergence time estimates of these species relative to D. mojavensis are: D. melanogaster, 40 – 60, D. virilis,

20 – 35, and D. arizonae 1.7 – 3.1 MYA, respectively (Powell, 1997; Russo et al., 1995; Matskin and Eanes, 2003; Reed et al., 2007; Croset et al., 2010; Oliveira et al., 2012). D. arizonae, a cactophilic generalist, is also the sister species to D. mojavensis within the repleta group

(Powell, 1997; Durando et al., 2000; Matzkin and Eanes, 2003; Reed et al., 2007; Croset et al.,

2010). We conducted single sensillum recordings of ac3 sensilla and compared the responses among species to 17 chemical compounds (Figure 7), that either elicited statistically significant differences in ac3 responses among populations or that were known agonists (Figure 3; Benton et al., 2009; Silbering et al., 2011; Rytz et al., 2013; Prieto-Godino et al., 2017).

Variation in the sensitivity and specificity of sensillar responses were found within and across drosophilid species, as all odorants elicited significantly different response profiles except for phenylacetaldehyde, a known agonist for IR84a-expressing OSNs (Figure 6a; Silbering et al.,

2011). Comparisons between odor-evoked responses in our D. melanogaster (Canton-S) and D. melanogaster physiological data in the literature revealed several differences (Yao et al., 2005;

Silbering et al., 2011; Prieto-Godino et al., 2017). We observed a decrease in sensitivity to butyraldehyde, 1,4-diaminobutane, 2-oxopentanoic acid and an increase in response to acetic acid and phenethylamine (Figure 6a). Further examination of the drosophilid OSN response profiles revealed that D. virilis and D. arizonae exhibited minimal responses to OR35a- expressing OSN stimulating odorants (hexanol, (E)-2-hexenal and hexyl acetate), raising the possibility of a similar loss of this receptor in these species (Figure 6a). This awaits further study.

Moreover, the ac3 OSNs in D. virilis and D. arizonae exhibited a notable change in specificity to acetic acid (Figure 6a), further demonstrating the divergence within the ac3 sensillar type. The response profiles of D. virilis and D. arizonae are the most similar to each other and subsequently to the D. mojavensis cluster, particularly the Mojave population (Figure 6b).

Electrophysiological recordings of lab maintained D. melanogaster (Canton-S) clustered together with existing physiological data in the literature (Figure 6b; Yao et al., 2005; Silbering et al.,

2011; Prieto-Godino et al., 2017).

4. DISCUSSION

Variation in the odor-evoked electrophysiological responses of the coeloconic sensilla were observed between D. mojavensis populations across all sensillar types. The response profiles of the Mojave population were the least similar in clustering to the other three populations across coeloconic types. Relative to D. melanogaster, the response spectra of ac3 sensilla from all D. mojavensis populations exhibited a notable difference – little to no responses to known agonists of OR35a-expressing OSNs. This absence of ac3 responses to agonists of OR35a neurons is consistent with the lack of Or35a expression in all four D. mojavensis populations. Further deviations of ac3 sensilla OSN responses were observed across drosophilid species in sixteen of the seventeen odorants. Little to no response to OR35a stimulating odorants was also observed in

D. virilis and D. arizonae and species differed in their sensitivity to tested acids and amines.

This study examined the OSN response profiles of all four antennal coeloconic sensillar types from the four D. mojavensis populations and revealed electrophysiological differences among populations. For instance, differences in odor-evoked responses of ac3 sensilla to propionic acid were observed, with the Sonora population exhibiting a significantly greater excitatory response compared to the other three populations. Moreover, the Sonora population also exhibited increased excitatory responses to another propionic acid detecting OSN housed in ac2 sensilla. The Sonora population feeds and breeds on organ pipe cacti and studies have shown

that these rots, at times have released the greatest single concentration of propionic acid (44.7 mM compared to 39.6 mM in agria rots), albeit this is not a continued concentration as its average concentration over time is slightly less than that of agria (Fogleman and Abril, 1990).

Moreover, a second result emerging from this study is the response of ac2 sensilla OSNs of the

Sonora population to butyric acid, as it exhibited the greatest response among populations. This is reflected in the chemistry of its host plant as previous studies on the volatile oil obtained through hydrodistillation of early stages of rot, have shown that organ pipe is dominated by carboxylic acids of which butyric acid is the most prominent (Wright and Setzer, 2014). These increases in sensitivity in the Sonora population to odorants emitted from its host plant, may be acting as primary olfactory cues that this population utilizes to localize and then in turn, exploit organ pipe cactus as its host plant.

The association of the Sonora population with the natural ecology of its host plant was not the only reported one in this study, as the Mojave population reflects a similar pattern with its host plant, which coincides with previous work on this system as well. The OSNs of the Mojave population demonstrated a decreased response spectrum to acids tested in this study (except for

2-oxopentanoic acid in ac1). The overall decreased responses of Mojave OSNs to majority of odorants (especially the acids) is divergent compared to the other populations, which is illustrated in the cluster analyses, as the Mojave population is routinely the least similar in all four coeloconic types. Moreover, the diverged olfactory sensitivities of the Mojave population reported in this study are consistent with previous work on the D. mojavensis system where similar divergence patterns in neurophysiological responses were described (Date et al., 2013;

Crowley-Gall et al., 2016). It was shown by Crowley-Gall et al., 2016 that the Mojave population had greater sensitivity in its OR-expressing OSNs of basiconic sensilla to aromatics,

which are a primary component of barrel cactus rots. Likewise, the little to no evidence of volatile acids being released from their natural host (Date et al., 2013; Rollmann unpublished) correlates with the observed diminished OSN responses to acids. These consistent peripheral changes seen across neuronal types in the Mojave population are reflected in the odor profile of its host plant and may play a vital role in how the flies detect and discriminate these odorants acting as a primary sensory cue for their specific host preference of barrel cactus.

In addition to differences in odor-evoked responses among populations, we also observed differences in the specificity of D. mojavensis responses relative to D. melanogaster. In particular, a lack of excitatory responses in ac3 sensilla to OR35a agonists was found in D. mojavensis and expression of Or35a was not detected in the antenna. In D. melanogaster, OR35a along with ORCO is expressed in one of the two OSNs housed within an ac3 sensillum. This receptor is broadly tuned, responding to seven different chemical groups; lactones, sulfur compounds, aldehydes, ketones, aromatics, alcohols, and esters (Hallem and Carlson, 2006). The majority of the compounds that activate OR35a are 6-carbon chain aldehydes and alcohols, which are emitted from green plant tissue and unripe fruit and are known as a green leaf volatiles

(GLVs) (Stensmyr et al., 2003b; Hallem and Carlson, 2006; Mansourian and Stensmyr, 2015). It has been shown in D. melanogaster that activation of OR35a by these GLVs, such as 1-hexanol promotes behavioral aversion as a potential mechanism to avoid unripe fruits, as these are not preferred for food and oviposition sites (Stensmyr et al., 2003b). Additional ORs within the

Drosophila olfactory system, however, can also detect these same GLVs, such as OR7a which is expressed in the OSNs that innervate antennal basiconic sensilla (Vosshall et al., 2000; Hallem and Carlson, 2006). Therefore, it raises the question in the acid-sensing olfactory circuitry of D.

mojavensis, for the need of a broadly tuned OR in coeloconic OSNs and if this redundancy can subsequently result in the loss of the chemosensory gene.

This speculation is supported by several gene loss studies on Drosophila and other phytophagous insects, where accelerated rates in the loss of chemosensory genes between specialist and generalist species were observed (McBride, 2007; McBride and Arguello, 2007;

Drosophila 12 Genomes Consortium; Goldman-Huertas et al., 2015). When comparing the strict specialist of morinda fruit, D. sechellia to its sister species, D. simulans it has been hypothesized that changes in selective environmental pressures may be at play for the accelerated chemosensory gene loss observed in D. sechellia (McBride, 2007; McBride and Arguello, 2007;

Drosophila 12 Genomes Consortium). This is because with the minimal odorants associated with host plant localization underlying the behavioral shift to morinda fruit, the fly may essentially only need to detect those fundamental odorants from its host plant, negating the need to detect a broad array of environmental cues like the generalist, D. simulans which one would expect to detect many more diverse chemosensory stimuli as it feeds and breeds on a wider variety of substrates (Stensmyr et al., 2003a; Dekker at al., 2006; McBride, 2007; McBride and Arguello,

2007). Likewise, another study comparing the olfactory system of a yeast-feeding generalist, D. melanogaster to that of a derived herbivorous specialist, Scaptomyza flava, also highlighted the evolution and loss of chemosensory genes. It was found that S. flava, a distantly related drosophilid species, exhibited no behavioral attraction as well as weak EAG responses to yeast- derived volatiles, while D. melanogaster predictably exhibited the exact opposite (Goldman-

Huertas et al., 2015). In examining the olfactory circuitry of S. flava, this study found a deletion of OR gene orthologs that are associated with yeast-localization in D. melanogaster, which coincides with the behavioral and physiological shifts observed between the species, as these

deletions are hypothesized to be associated with the transition to herbivorous specialization in this species (Goldman-Huertas et al., 2015). As with D. sechellia and S. flava, D. mojavensis is specialist, in its case on cactus, suggesting that perhaps a similar chemoreceptor gene loss occurred in which response to differing environmental sensory stimuli may have shaped its peripheral olfactory system to be more sensitive to the acid component of the headspace of its host cacti.

Accompanying the lack of detection of Or35a expression in the antenna, another observation reported in this study is the maintained response to 1,4-diaminobutane across D. mojavensis populations. In D. melanogaster, this odorant normally elicits an excitatory response in OR35a-expressing OSNs, as evidence by a loss in ac3 OSN responses to this odorant in

OR35a mutants (Silbering et al., 2011). We don’t know why, D. mojavensis flies maintained an excitatory response to this odorant, considering they do not express OR35a. The response to 1,4- diaminobutane may be related to (a) a potential shift in the ligand binding affinity of other receptors expressed in the OSNs of ac3 sensilla, specifically a change in the specificity of the

IR75a/b/c-expressing neuron, (b) the specificity of a currently unknown receptor expressed in the neuron that previously expressed OR35a, and/or (c) something related to the putative co-receptor

IR76b, expressed in the OSNs of all four coeloconic types. Changes to IR76b alone would not explain its detection of the odorant because it is expressed in all four sensillar types, and ac1 and ac4 do not respond to the odorant. So, its involvement would have to be in conjunction with the unique IR pairing in ac3, as a given coeloconic OSN is proposed to consist of up to three IR subunits which together form a functional olfactory receptor (Abuin et al., 2011). Therefore, one might speculate that in the absence of OR35a, IR76b either combines with an unknown IR subunit to form a novel functioning olfactory receptor or that IR76b itself, has shifted to define

ligand-specificity in place of OR35a and now is responsible for 1,4-diaminobutane detection.

While highly speculative we cannot rule out this possibility.

To further resolve the underlying mechanisms, future work should (1) determine which neuron is responding to 1,4-diaminobutane through the use the empty IR decoder neuron (a molecular technique that utilizes UAS-GAL4 drivers to express specific IRs of interest in an

“empty neuron,” as to then measure and determine the electrophysiological response properties of the IR of interest (Grosjean et al., 2011)) to independently assess the responses of individual

D. mojavensis IRs by inserting them in the empty IR decoder neuron of D. melanogaster (i.e. expression of IR76b individually as well as with various combinations of the other antennal IRs),

(2) in situ hybridization to identify expression patterns and co-expression patterns of IRs per sensillar type, and (3) RNA-seq on antenna to assess which IRs are expressed and the extent to which novel chemosensory genes may be present. Still, the absence of OR35a in ac3, begs the question: what is being expressed in its place and are these OSNs responsible for the specificity and sensitivity changes that were observed? This question awaits future study.

To further delve into the evolution of ac3 OSN responses, we extended our analyses to include other drosophilid species, specifically D. melanogaster, D. virilis, and D. arizonae. Both ac3 OSNs of D. virilis and D. arizonae did not respond to OR35a agonists, raising the possibility that they too may lack this receptor. These species are a part of the “virilis-repleta” radiation; D. mojavensis and D. arizonae are sister species that diverged from each other no more than 3

MYA within the repleta species group, whereas D. virilis, a member of a more distant group, D. virilis species group diverged roughly 20 – 35 MYA. This recently diverged lineage of flies appeared within the subgroup Drosophila roughly 25 – 36 MYA (Powell, 1997; Durando et al.,

2000; Matskin and Eanes, 2003; Reed et al., 2007; Oliveira et al., 2012), and having multiple

species with various divergence times all demonstrating a potential absence in expression of

OR35a, may suggest that the single OR expressed in coeloconic sensilla was lost at the formation of the “virilis-repleta” radiation, however this warrants further study of additional species. One plausible explanation for the absence of OR35a can be seen in D. arizonae and D. mojavensis, which are both members of the repleta group. The evolution of the repleta group is well accompanied by the shift to using fermenting cactus tissue for food and oviposition sites from that of fermenting fruits of non-cactus plants (Oliveira et al., 2012), and as mentioned earlier, chemosensory gene loss is much more likely to occur within host specialists as opposed to generalists (McBride, 2007; McBride and Arguello, 2007; Goldman-Huertas et al., 2015).

Perhaps in the repleta group, it is possible to have chemosensory gene loss with host specialization, as this shift may have altered selection pressures on the peripheral olfactory system. However, in the fate of D. virilis, this explanation would not necessarily be the case because D. virilis, a generalist, is not a member of the repleta group, suggestive that the loss of

OR35a may predate the evolution of host plant specialization within the “virilis-repleta” radiation. But, in short various factors may be at play and the mechanisms underlying this gene loss remain to be identified. Future work is needed across species of the “virilis-repleta” radiation to examine more closely the evolution of OR35a and the extent to which it plays a role in ecological specialization.

Finally, differences in electrophysiological responses exist both within and between populations and the detailed mechanisms underlying these responses warrant further study. In the case of ac3 response profiles, we observed sensitivity changes to various amines across populations relative to D. melanogaster. Based on this result we sequenced the coding regions of

Ir75a and Ir75bc_GI13611 and, in brief we found multiple amino acid substitutions in both

genes of interest. A potential mechanism underlying these changes in sensitivity may be due to alterations to the amino acid sequence of the olfactory receptor as various studies have demonstrated that an amino acid substitution in a protein receptor can alter ligand binding affinity by transforming the structure and therefore the function of the receptor (Abaffy et al.,

2007; Keller et al., 2007; Rollmann et al., 2010; Prieto-Godino et al., 2017). For instance, a recent study on D. sechellia revealed a change in sensitivity of the ac3 OSNs to hexanoic acid

(Prieto-Godino et al., 2017). A single amino acid substitution within the LBD of the protein receptor IR75b resulted in the change in sensitivity to hexanoic acid. Upon examination, our results revealed eight different amino acid substitutions in IR75a, none of which were located within the LBD. The examination of IR75bc_GI13611 found nine total substitutions, four of which were located within the putative LBD. Future studies are needed to determine the contribution(s) of the four reported substitutions within the LBD of IR75bc_GI13611 to the observed shifts in ligand affinity, as well as the addition of IR75bc_GI13610 to determine its involvement in the increased sensitivity changes observed in ac3 OSNs of D. mojavensis populations.

To conclude, this study characterized OSN response profile differences across conspecific populations of D. mojavensis, as well as across drosophilid species. Various peripheral changes were observed in the Drosophila olfactory systems, and may reflect the plasticity of divergence in sensory systems of phytophagous insects in response to the differing selection pressures associated with environmental adaptation. In addition, this study provides further insight into the intricacies underlying how organisms adapt to new environments and how subsequent population divergence may facilitate reproductive isolation, preceding the evolution and formation of species. This study and the existing molecular, genetic, and neurophysiological

sensory knowledge available from Drosophila begins to provide a more complete picture of an olfactory system, on which future studies can further build and promote our understanding of the complex mechanisms by which Drosophila and other organisms adapt to their specific ecological habitats and how this may lead to the evolution and formation of new species.

5. REFERENCES

1. Schluter D. (2001). Ecology and the origin of species. Trends Ecol. Evol. 16: 372 – 380.

2. Linn C, Feder JL, Nojima S, Dambroski HR, Berlocher SH, and Roelofs W. (2003). Fruit

odor discrimination and sympatric host race formation in Rhagoletis. PNAS. 100(20):

11490 – 11493.

3. Rundle HD and Nosil P. (2005). Ecological speciation. Ecol. letters. 8: 336 – 352.

4. Funk D. (1998). Isolating a role for natural selection in speciation: host adaptation and

sexual isolation in Neochlamisus bebbianae leaf beetles. Evol. 52: 1744 – 1759.

5. Fuse N. (2017). Genome research elucidating environmental adaptation: Dark-fly project

as a case study. Gen. & Dev. 45: 97 – 102.

6. Fuse N, Kitamura T, Haramura T, Arikawa K, and Imafuku M. (2014). Evolution in the

Dark—Adaptation of Drosophila in the Laboratory. Springer. doi: 10.1007/978-4-431-

54147-9.

7. Via S. (1999). Reproductive isolation between sympatric races of pea aphids. I. gene flow

restriction and habitat choice. Evolution. 53(5): 1446 – 1457.

8. Pophof B, Stange G, and Abrell L. (2005). Volatile organic compounds as signals in a

plant-herbivore system: electrophysiological responses in olfactory sensilla of the moth

Cactoblastis cactorum. Chem. Senses. 30: 51 – 68.

9. Date P, Dweck HK, Stensmyer MC, Shann J, Hansson BS, and Rollmann SM. (2013).

Divergence in olfactory host preference in D. mojavensis in response to cactus host use.

PLoS ONE. 8(7): e70027.

10. Nishida R. (2014). Chemical ecology of insect-plant interactions: ecological significance

of plant secondary metabolites. Biosci. Biotech. Biochem. 78(1): 1 – 13.

11. Kopp A, Barmina O, Hamilton AM, Higgins L, McIntyre LM, and Jones CD. (2008).

Evolution of gene expression in the Drosophila olfactory system. Mol. Biol. Evol. 25(6):

1081 – 1092.

12. Hussain A, Zhang M, Üçpunar HK, Svensson T, Quillery E, Gompel N, Ignell R, and

Grunwald Kadow IC. (2016). Ionotropic chemosensory receptors mediate the taste and

smell of polyamines. PLoS Biol. 14(6): 1 – 30.

13. Stensmyr MC, Dekker T, and Hansson BS. (2003a). Evolution of the olfactory code in

the Drosophila melanogaster subgroup. Proc. R. Soc. Lond. B. 270(1531): 2333 – 2340.

14. Dekker T, Ibba I, Siju KP, Stensmyr MC, and Hansson BS. (2006). Olfactory shifts

parallel superspecialism for toxic fruit in Drosophila melanogaster sibling, D. sechellia.

Curr. Biol. 16: 101 – 109.

15. Legal L, Chappe B, and Jallon JM. (1994). Molecular basis of Morinda citrifolia (L.):

toxicity on Drosophila. J. Chem. Ecol. 20(8): 1931 – 1943.

16. Farine JP, Legal L, Moreteau B, and Le Quere JL. (1996). Volatile components of ripe

fruit of Morinda citrifolia and their effects on Drosophila. Phytochem. 41(2): 433 – 438.

17. Amlou M, Moreteau B, and David JR. (1998). Genetic analysis of Drosophila sechellia

specialization: oviposition behavior toward the major aliphatic acids of its host plant.

Behav. Genet. 28(6): 455 – 464.

18. Mansourian S and Stensmyr M. (2015). The chemical ecology of the fly. Neurobiol. 34:

95 – 102.

19. Prieto-Godino LL, Rytz R, Cruchet S, Bargeton B, Abuin L, Silbering AF, Ruta V, Dal

Peraro M, and Benton R. (2017). Evolution of acid-sensing olfactory circuits in

drosophilids. Neuron. 93: 1 – 16.

20. Jones CD. (2005). The genetics of adaptation in Drosophila sechellia. Genetica. 123: 137

– 145.

21. Linn C, Nojima S, and Roelofs W. (2005). Antagonist effects of non-host fruit volatiles

on discrimination of host fruit by Rhagoletis flies infesting apple (Malus pumila),

hawthorn (Crataegus spp.), and flowering dogwood (Cornus florida). Entomol. Exp.

Appl. 114(2): 97 – 105.

22. Olsson SB, Linn Jr CE, and Roelofs WL. (2006a). The chemosensory basis for

behavioral divergence involved in sympatric host shifts: I. characterizing olfactory

receptor neuron classes responding to key host volatiles. J. Comp. Physiol. A. 192: 279 –

288.

23. Olsson SB, Linn Jr CE, and Roelofs WL. (2006b). The chemosensory basis for

behavioral divergence involved in sympatric host shifts: II. Olfactory receptor neuron

sensitivity and temporal firing pattern to individual key host volatiles. J. Comp. Physiol.

A. 192: 289 – 300.

24. Crowley-Gall A, Date P, Han C, Rhodes N, Andolfatto P, Layne JE, and Rollmann SM.

(2016). Population differences in olfaction accompany host shift in Drosophila

mojavensis. Proc. R. Soc. B. 283: 1 – 9.

25. Nosil P, Vines TH, and Funk DJ. (2005). Perspective: reproductive isolation caused by

natural selection against immigrants from divergent habitats. Evolution. 59(4): 705 – 719.

26. Stocker RF. (1994). The organization of the chemosensory system in Drosophila

melanogaster: a review. Cell Tissue Res. 275: 3 – 26.

27. Reddy GV, Gupta B, Ray K, and Rodrigues V. (1997). Development of the Drosophila

olfactory sense organs utilizes cell-cell interactions as well as lineage. Development. 124:

703 – 712.

28. Clyne PJ, Warr CG, Freeman MR, Lessing D, Kim J, and Carlson JR. (1999). A novel

family of divergent seven-transmembrane proteins: candidate odorant receptors in

Drosophila. Neuron. 22: 327 – 338.

29. Yao CA, Ignell R, and Carlson JR. (2005). Chemosensory coding by neurons in the

coeloconic sensilla of the Drosophila antenna. J. Neurosci. 25: 8359 – 8367.

30. Benton R, Vannice KS, Gomez-Dia C, and Vosshall LB. (2009). Variant ionotropic

glutamate receptors as chemosensory receptors in Drosophila. Cell. 136: 149 – 162.

31. van der Goes van Naters W and Carlson JR. (2007). Receptors and Neurons for Fly

Odors in Drosophila. Curr. Biol. 17(7): 606 – 612.

32. Ai M, Min S, Grosjean Y, Leblanc C, Bell R, Benton R, and Suh GS. (2010). Acid

sensing by the Drosophila olfactory system. Nature. 468: 691 – 695.

33. Hallem EA and Carlson JR. (2004). The odor coding system of Drosophila. TRENDS

Genet. 20(9): 453 – 459.

34. Vosshall LB and Stocker RF. (2007). Molecular architecture of smell and taste in

Drosophila. Annu. Rev. Neurosci. 30: 505 – 533.

35. Abuin L, Bargeton B, Ulbrich MH, Isacoff EY, Kellenberger S, and Benton R. (2011).

Functional architecture of olfactory ionotropic glutamate receptors. Cell. 69: 44 – 60.

36. Silbering AF, Rytz R, Grosjean Y, Abuin L, Ramdya P, Jefferis G, and Benton R.

(2011). Complementary function and integrated wiring of the evolutionarily distinct

Drosophila olfactory subsystems. J. Neurosci. 31: 13357 – 13375.

37. Ai M, Blais S, Park JY, Min S, Neubert TA, and Suh GSB. (2013). Ionotropic glutamate

receptors IR64a and IR8a form a functional odorant receptor complex in vivo in

Drosophila. J. Neurosci. 33(26): 10741 – 10749.

38. Vosshall LB, Amrein H, Morozov PS, Rzhetsky A, and Axel R. (1999). A spatial map of

olfactory receptor expression in the Drosophila antenna. Cell. 96: 725 – 736.

39. Stensmyr MC, Giordano E, Balloi A, Angioy AM, and Hansson BS. (2003b). Novel

natural ligands for Drosophila olfactory receptor neurons. J. Exp. Biol. 206: 715 – 724.

40. Vosshall LB, Wong AM, and Axel R. (2000). An olfactory sensory map in the fly brain.

Cell. 102: 147 – 159.

41. Min S, Ai M, Shin SA, and Suh GSB. (2013). Dedicated olfactory neurons mediating

attraction behavior to ammonia and amines in Drosophila. PNAS. 110(14): E1321 –

E1329.

42. Rytz R, Croset V, and Benton R. (2013). Ionotropic receptors (IRs): chemosensory

ionotropic glutamate receptors in Drosophila and beyond. Insect Biochem. Mol. Biol. 43:

888 – 897.

43. Bell JS and Wilson RI. (2016). Behavior reveals selective summation and max pooling

among olfactory processing channels. Neuron. 91: 425 – 438.

44. Masse NY, Turner GC, and Jefferis GSXE. (2009). Olfactory information processing in

Drosophila. Curr. Biol. 19(16): 700 – 713.

45. Couto A, Alenius M, and Dickson BJ. (2005). Molecular, anatomical, and functional

organization of the Drosophila olfactory system. Curr. Biol. 15: 1535 – 1547.

46. Benton R. (2007). Sensitivity and specificity in Drosophila pheromone perception.

TRENDS Neurosci. 30(10): 512 – 519.

47. Knecht ZA, Silbering AF, Ni L, Klein M, Budelli G, Bell R, Abuin L, Ferrer AJ, Samuel

ADT, Benton R, and Garrity PA. (2016). Distinct combinations of variant ionotropic

glutamate receptors mediate thermosensation and hygrosensation in Drosophila.

Neurosci. 5: e17879.

48. Joseph RM, Devineni AV, King IF, and Heberlein U. (2009). Oviposition preference for

and positional avoidance of acetic acid provide a model for competing behavioral drives

in Drosophila. Proc. Natl. Acad. Sci. 106: 11352 – 11357.

49. Kim G, Huang JH, McMullen JG II, Newell PD, and Douglas AE. (2017). Physiological

response of insects to microbial products: insights from the interactions between

Drosophila and acetic acid. J. Insect Phys.

doi:http://dx.doi.org/10.1016/j.jinsphys.2017.05.005.

50. Liu C, Pitts RJ, Bohbot JD, Jones PL, Wang G, and Zwiebel LJ. (2010). Distinct

olfactory signaling mechanisms in the malaria vector mosquito Anopheles gambiae. PLoS

Biol. 8(8): e1000467.

51. Croset V, Rytz R, Cummins SF, Budd A, Brawand D, Kaessmann H, Gibson TJ, and

Benton R. (2010). Ancient protostome origin of chemosensory ionotropic glutamate

receptors and the evolution of insect taste and olfaction. PLoS Genet. 6(8): e1001064.

52. Bengtsson JM, Trona F, Montagne N, Anfora G, Ignell R, Witzgall P, and Jacquin-Joly

E. (2012). Putative chemosensory receptors of the codling moth, Cydia pomonella,

identified by antennal transcriptome analysis. PLoS One. 7: e31620.

53. Smallegange RC, Qiu YT, van Loon JJA, and Takken W. (2005). Synergism between

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

the malaria mosquito Anopheles gambiae sensu stricto (Diptera: Culicidae). Chem.

Senses. 30: 145 – 152.

54. Qiu YT, van Loon JJA, Takken W, Meijerink J, and Smid HM. (2006). Olfactory coding

in antennal neurons of the malaria mosquito, Anopheles gambiae. Chem. Senses. 31: 845

– 863.

55. Carson HL and Stalker HD. (1951). Natural breeding sites for some wild species of

Drosophila in the eastern United States. Ecol. 32(2): 317 – 330.

56. Stensmyr MC, Dweck HKM, Farhan A, Ibba I, Strutz A, Mukunda L, Linz J, Grabe V,

Steck K, Lavista-Llanos S, Wicher D, Sachse S, Knaden M, Becher PG, Seki Y, and

Hansson BS. (2012). A conserved dedicated olfactory circuit for detecting harmful

microbes in Drosophila. Cell. 151: 1345 – 1357.

57. Becher PG, Bengtsson M, Hansson BS, and Witzgall P. (2010). Flying the fly: long-range

flight behavior of Drosophila melanogaster to attractive odors. J. Chem. Ecol. 36: 599 –

607.

58. Heby O. (1981). Role of polyamines in the control of cell proliferation and

differentiation. Differentiation. 19: 1 – 20.

59. Ramani D, De Bandt JP, and Cynober L. (2014). Aliphatic polyamines in physiology and

disease. Clin. Nutr. 33(1): 14 – 22.

60. Schulter D and Conte GL. (2009). Genetics and ecological speciation. PNAS. 106(1):

9955 – 9962.

61. Sobel JM, Chen GF, Watt LR, and Schemske DW. (2009). The biology of speciation.

Evolution. 64: 295 – 315.

62. Fogleman JC and Abril JR. (1990). Ecological and evolutionary importance of host plant

chemistry. Ecol. Evol. Gen. Drosophila. pp. 121 – 143.

63. Newby BD and Etges WJ. (1998). Host preference among populations of Drosophila

mojavensis that use different host cacti. J. Insect. Behav. 11: 691 – 712.

64. Smith G, Lohse K, Etges WJ, and Ritchie MG. (2012). Model based comparisons of

phylogeographic scenarios resolve the intraspecific divergence of cactophilic Drosophila

mojavensis. Mol. Ecol. 21: 3293 – 3307.

65. Date P, Crowley-Gall A, Diefendorf AF, and Rollmann SM. (2017). Population

differences in host plant preference and the importance of yeast and plant substrate to

volatile composition. Eco. Evol. 00: 1 – 11.

66. Fogleman JC and Foster JLM. (1989). Microbial colonization of injured cactus tissue

(Stenocereus gummosus) and its relationship to the ecology of cactophilic Drosophila

mojavensis. Appl. Envir. Microbiol. 55(1): 100 – 105.

67. Wright CR and Setzer WN. (2014). Characterization of volatiles of necrotic Stenocereus

thurberi and Opuntia littoralis and toxicity and olfactory preference of Drosophila

melanogster, D. mojavensis wrigleyi, and D. mojavensis sonorensis to necrotic cactus

volatiles. Nat. Prod. Comm. 9(8): 1185 – 1192.

68. de Bruyne M, Smart R, Zammit E, and Warr CG. (2009). Functional and molecular

evolution of olfactory neurons and receptors for aliphatic esters across the Drosophila

genus. J. Comp. Phyiol. A. 196: 97 – 109.

69. Massie KR and Markow TA. (2005). Sympatry, allopatry and sexual isolation between

Drosophila mojavensis and D. arizonae. Hereditas. 142: 51 – 55.

70. Benjamini Y, Krieger K, and Yekutieli, D. (2006). Adaptive linear step-up procedures

that control the false discovery rate. Biometrika. 93: 491 – 507.

71. de Bruyne M, Clyne PJ, and Carlson JR. (1999). Odor coding in a model olfactory organ:

the Drosophila maxillary palp. J. Neurosci. 19(1): 4520 – 4532.

72. Prieto-Godino LL, Rytz R, Bargeton B, Abuin L, Arguello JR, Dal Peraro M, and Benton

R. (2016). Olfactory receptor pseudo-pseudogenes. Nature. 539: 93 – 97.

73. Powell JR. (1997). Progress and prospects in evolutionary biology: the Drosophila

model. Oxford University Press. ISBN: 0-19-507691-5.

74. Russo CA, Takezaki N, and Nei M. (1995). Molecular phylogeny and divergence

times of drosophilid species. Mol. Biol. Evol. 12: 391 – 404.

75. Matzkin LM and Eanes WF. (2003). Sequence variation of alcohol dehydrogenase (Adh)

paralogs in cactophilic Drosophila. Genetics. 163: 181 – 194.

76. Reed LK, Nyboer M, and Markow TA. (2007). Evolutionary relationships of Drosophila

mojavensis geographic host races and their sister species Drosophila arizonae. Mol

Ecol. 16: 1007 – 1022.

77. Oliveira DCSG, Almeida FC, O’Grady PM, Armella MA, DeSalle R, and Etges WJ.

(2012). Monophyly, divergence times, and evolution of host plant use inferred from a

revised phylogeny of the Drosophila repleta species group. Mol. Phy. Evol. 64: 533 –

544.

78. Durando CM, Baker RH, Etges WJ, Heed WB, Wasserman M, and DeSalle R. (2000).

Phylogenetic analysis of the repleta species group of the genus Drosophila using multiple

sources of characters. Mol. Phy. Evol. 16(2): 296 – 307.

79. Hallem EA and Carlson JR. (2006). Coding of odors by a receptor repertoire. Cell. 125:

143 – 160.

80. McBride CS. (2007). Rapid evolution of smell and taste receptor genes during host

specialization in Drosophila sechellia. PNAS. 104(12): 4996 – 5001.

81. McBride CS and Arguello JR. (2007). Five Drosophila genomes reveal nonneutral

evolution and the signature of host specialization in the chemoreceptor superfamily.

Genetics. 177: 1395 – 1416.

82. Drosophila 12 Genomes Consortium. (2007). Evolution of genes and genomes on the

Drosophila phylogeny. Nature. 450: 203 – 218.

83. Goldman-Huertas B, Mitchell RF, Lapoint RT, Faucher CP, Hildebrand JG, and

Whiteman NK. (2015). Evolution of herbivory in Drosophilidae linked to loss of

behaviors, antennal responses, odorant receptors, and ancestral diet. PNAS. 112(10): 3026

– 3031.

84. Grosjean Y, Rytz R, Farine JP, Abuin L, Cortot J, Jefferis GSXE, and Benton R. (2011).

An olfactory receptor for food-derived odours promotes male courtship in Drosophila.

Nature. 478: 236 – 242.

85. Abaffy T, Malhotra A, and Luetje CW. (2007). The molecular basis for ligand specificity

in a mouse olfactory receptor. J. Biol. Chem. 282(2): 1216 – 1224.

86. Keller A, Zhuang H, Chi Q, Vosshall LB, and Matsunami H. (2007). Genetic variation in

a human odorant receptor alters odour perception. Nature. 449: 468 – 473.

87. Rollmann SM, Wang P, Date P, West SA, Mackay TFC, and Anholt RRH. (2010).

Receptor polymorphisms and natural variation in olfactory behavior in Drosophila

melanogaster. Genetics. 186: 687 – 697.

6. Figure Legends

Figure 1. Single sensillum recordings of D. mojavensis populations. Odor response profiles of four antennal coeloconic sensillar types (ac1 – ac4) to select odorants (mean + SEM) are presented for all four D. mojavensis populations. The quantified summed responses of the OSNs innervating each sensillar type for the four populations are shown. The ordered list of odorants

(y-axis) is also found in Table 1.

Figure 2. OSN classification with known D. melanogaster odor response spectra.

Hierarchical cluster analysis was based on OSN response profiles to 36 odorants for the four D. mojavensis populations and known D. melanogaster spectra (Yao et al., 2005; Silbering et al.,

2011; Prieto-Godino et al., 2017).

Figure 3. Statistical differences in electrophysiological responses among the four D. mojavensis populations. Different letters depict significant population differences in odor- evoked responses for each coeloconic sensillar type as revealed by ANOVA and subsequent

Tukey HSD post-hoc analysis. ‘A’ designation was specified to the population with the greatest excitatory or inhibitory response.

Figure 4. Single sensillum recording from an ac3 sensillum. Representative trace of an extracellular recording of a spontaneously firing ac3 sensillum from the Catalina population. A

0.5 s trace is shown and the action potentials of two different neurons, labeled A and B are noted.

Figure 5. RT-PCR analysis of ionotropic receptor gene expression in all four D. mojavensis populations. Products were amplified and visualized on a 2.0% agarose gel electrophoresis.

Expression was examined for each gene and population for (a) the housekeeping gene Act5C, a positive control, (b) Ir75bc_GI13610, (c) Ir75bc_GI13611, and (d) Or35a. A negative control without reverse transcriptase (No RT) was also generated for each gene and population.

Figure 6. Amino acid sequence differences in ionotropic receptors among D. mojavensis populations. Alignment of amino acid sequences of each gene: (a) IR75a and (b)

IR75bc_GI13611. Amino acid substitutions are indicated by the grey boxes. The S1 and S2 lobes of the ligand-binding domain (LBD) are depicted by the blue bars and part of the pore loop (P) is depicted by the green bar in accordance with Benton et al., 2009; Prieto-Godino et al., 2016,

2017.

Figure 7. Single sensillum recordings of drosophilid species. (a) Odor response profiles of

OSNs in ac3 sensilla to the selected odorants (mean + SEM) for four Drosophila species. The summed odor-evoked responses are shown. For illustrative purposes, the electrophysiological response of the D. mojavensis Catalina population is depicted (Figure 1). (b) Classification of ac3 sensilla among species. The hierarchical cluster analysis is based on OSN response profiles to 17 odorants. 1Canton-S; 2physiological data from Yao et al., 2005; Silbering et al., 2011;

Prieto-Godino et al., 2017.

Table 1.

Table 1. List of chemical compounds used in electrophysiological recordings. Chemicals are grouped by chemical class and listed together with their CAS number and purity.

Chemical compounds CAS Number Purity (%) 2-pentylfuran 3777-69-3 ≥ 98 formamide 75-12-7 ≥ 99.5 6-methyl-5-hepten-2-one 110-93-0 ≥ 99 2-heptanone 110-43-0 ≥ 99 acetoin 513-86-0 ≥ 96 2-butanone 78-93-3 ≥ 99 acetone 67-64-1 ≥ 99.9 2-nonanone 821-55-6 ≥ 99 ethyl octanoate 106-32-1 ≥ 99 isoamyl butyrate 106-27-4 ≥ 98 n-propyl hexanoate 626-77-7 ≥ 98 phenethyl propionate 122-70-3 ≥ 98 isoamyl acetate 123-92-2 ≥ 97 ethyl butyrate 105-54-4 ≥ 99 propyl acetate 109-60-4 ≥ 99 ethyl hexanoate 123-66-0 ≥ 99 hexyl acetate 142-92-7 ≥ 99 benzonitrile 100-47-0 ≥ 99 acetophenone 98-86-2 ≥ 99 2-phenethyl acetate 103-45-7 ≥ 99 phenol 108-95-2 ≥ 99.5 2-methoxy-4-propylphenol 93-51-6 ≥ 98 p-cresol 106-44-5 ≥ 99 4-ethylguaiacol 2785-89-9 ≥ 98 ethyl benzoate 93-89-0 ≥ 99 benzaldehyde 100-52-7 ≥ 99.5 phenethylamine 64-04-0 ≥ 99 spermidine 124-20-9 ≥ 99 cadaverine 462-94-2 ≥ 97 benzyl cyanide 140-29-4 ≥ 98 pyridine 110-86-1 ≥ 99.8 1,4-diaminobutane 110-60-1 ≥ 99 ethanolamine 141-43-5 ≥ 99.5 (E)-2-hexenal 6728-26-3 ≥ 96 phenylacetaldehyde 122-78-1 ≥ 95 butyraldehyde 123-72-8 ≥ 99.5 isopropanol 67-63-0 ≥ 99.7 linalool 78-70-6 ≥ 97 hexanol 111-27-3 ≥ 99.9 2-oxopentanoic acid 1821-02-9 ≥ 98 isobutyric acid 79-31-2 ≥ 99 octanoic acid 124-07-2 ≥ 98 hexanoic acid 142-62-1 ≥ 99 valeric acid 109-52-4 ≥ 99 propionic acid 79-09-4 ≥ 99.5 acetic acid 64-19-7 ≥ 99.7 butyric acid 107-92-6 > 99

Table 2.

Table 2. RT-PCR Primers. List of primer pairs used in the RT-PCR of each gene and the expected amplicon size of the resulting product.

Figure 1.

Mojave Catalina Baja Sonora 2-pentylfuran formamide 6-methyl-5-hepten-2-one 2-heptanone acetoin 2-butanone acetone 2-nonanone ethyl octanoate isoamyl butyrate n-propyl hexanoate phenethyl propionate isoamyl acetate ethyl butyrate propyl acetate ethyl hexanoate hexyl acetate benzonitrile acetophenone 2-phenethyl acetate phenol 2-methoxy-4-propylphenol p-cresol 4-ethylguaiacol ethyl benzoate benzaldehyde phenethylamine spermidine cadaverine benzyl cyanide pyridine 1,4-diaminobutane ethanolamine (E)-2-hexenal phenylacetaldehyde butyraldehyde isopropanol linalool hexanol 2-oxopentanoic acid isobutyric acid octanoic acid hexanoic acid valeric acid propionic acid acetic acid butyric acid paraffin oil ac1 H2O blank 2-pentylfuran formamide 6-methyl-5-hepten-2-one 2-heptanone acetoin 2-butanone acetone 2-nonanone ethyl octanoate isoamyl butyrate n-propyl hexanoate phenethyl propionate isoamyl acetate ethyl butyrate propyl acetate ethyl hexanoate hexyl acetate benzonitrile acetophenone 2-phenethyl acetate phenol 2-methoxy-4-propylphenol p-cresol 4-ethylguaiacol ethyl benzoate benzaldehyde phenethylamine spermidine cadaverine benzyl cyanide pyridine 1,4-diaminobutane ethanolamine (E)-2-hexenal phenylacetaldehyde butyraldehyde isopropanol linalool hexanol 2-oxopentanoic acid isobutyric acid octanoic acid hexanoic acid valeric acid propionic acid acetic acid butyric acid paraffin oil ac2 H2O blank 2-pentylfuran formamide 6-methyl-5-hepten-2-one 2-heptanone acetoin 2-butanone acetone 2-nonanone ethyl octanoate isoamyl butyrate n-propyl hexanoate phenethyl propionate isoamyl acetate ethyl butyrate propyl acetate ethyl hexanoate hexyl acetate benzonitrile acetophenone 2-phenethyl acetate phenol 2-methoxy-4-propylphenol p-cresol 4-ethylguaiacol ethyl benzoate benzaldehyde phenethylamine spermidine cadaverine benzyl cyanide pyridine 1,4-diaminobutane ethanolamine (E)-2-hexenal phenylacetaldehyde butyraldehyde isopropanol linalool hexanol 2-oxopentanoic acid isobutyric acid octanoic acid hexanoic acid valeric acid propionic acid acetic acid butyric acid paraffin oil ac3 H2O blank 2-pentylfuran formamide 6-methyl-5-hepten-2-one 2-heptanone acetoin 2-butanone acetone 2-nonanone ethyl octanoate isoamyl butyrate n-propyl hexanoate phenethyl propionate isoamyl acetate ethyl butyrate propyl acetate ethyl hexanoate hexyl acetate benzonitrile acetophenone 2-phenethyl acetate phenol 2-methoxy-4-propylphenol p-cresol 4-ethylguaiacol ethyl benzoate benzaldehyde phenethylamine spermidine cadaverine benzyl cyanide pyridine 1,4-diaminobutane ethanolamine (E)-2-hexenal phenylacetaldehyde butyraldehyde isopropanol linalool hexanol 2-oxopentanoic acid isobutyric acid octanoic acid hexanoic acid valeric acid propionic acid acetic acid butyric acid paraffin oil ac4 H2O blank -50 0 50 100 150 200 -50 0 50 100 150 200 -50 0 50 100 150 200 -50 0 50 100 150 200 Spikes/sec 44

Figure 2.

Figure 3.

ac1 ac2 ac3 ac4 Mojave Catalina Baja Sonora Mojave Catalina Baja Sonora Mojave Catalina Baja Sonora Mojave Catalina Baja Sonora 2-pentylfuran- formamide- 6-methyl-5-hepten-2-one- 2-heptanone- acetoin- 2-butanone- acetone- 2-nonanone- A AB B AB ethyl octanoate- isoamyl butyrate- n-propyl hexanoate- phenethyl propionate- isoamyl acetate- ethyl butyrate- propyl acetate- ethyl hexanoate- hexyl acetate- benzonitrile- acetophenone- 2-phenethyl acetate- phenol- 2-methoxy-4-propylphenol- p-cresol- 4-ethylguaiacol- ethyl benzoate- benzaldehyde- phenethylamine- B A AB AB D C B A spermidine- A AB C BC A BC C AB cadaverine- B A B B benzyl cyanide- pyridine- A AB B B AB A C BC 1,4-diaminobutane- A B C C A A B AB ethanolamine- (E)-2-hexenal- A B B B phenylacetaldehyde- butyraldehyde- AB BC C A isopropanol- linalool- hexanol- B B A AB 2-oxopentanoic acid- A C BC AB AB B AB A B A A A A B B B isobutyric acid- B A A A octanoic acid- hexanoic acid- valeric acid- B A AB A propionic acid- AB B AB A B B B A acetic acid- AB A AB B A BC C B butyric acid- AB B AB A paraffin oil- H2O- blank-

Figure 4.

ac3: Catalina – spontaneous firing

2 mV

B A 0.1 s

Figure 5.

Sonora Baja Mojave Catalina (a) No RT RT No RT RT No RT RT No RT RT L Act5C 200bp (Control) 100bp

(b) No RT RT No RT RT No RT RT No RT RT L Ir75bc_GI13610 200bp

100bp

(d) No RT RT No RT RT No RT RT No RT RT L 300bp Or35a 200bp

Figure 6. (a)

Ir75a Mojave MQLLQLANFVLQNLLQSHISFIIFFHCWANNETLQFAQQIHQPHQQPIYYQFAHLRDWNW 60 Catalina MQLLQLANFVLQNLLQSRISFIIFFHCWANNETLQFAQQIHQPHQQPIYYQFAHLRDWNW Baja MQLLQLANFVLQNLLQSHISFIIFFHCWANNETLQFAQQIHQPHQQPIYYQFAHLRDWNW Sonora MQLLQLANFVLQNLLQSHISFIIFFHCWANNETLQFAQQIHQPHQQPIYYQFAHLRDWNW

61 EHLEQRYLDHAQPTLAIYVDLRCMRSRSLLAEASRARLYNQHYHWVLHDSSEAFSFYDFF 120 EHLEQRYLDHAQPTLAIYVDLRCMRSRSLLAEASRARLYNQHYHWVLHDSSEAFSFYDFF EHLEQRYLDHAQPTLAIYVDLRCMRSRSLLAEASRARLYNQHYHWVLHDSSEAFSFYDFF EHLEQRYLDHAQPTLAIYVDLRCMRSRSLLAEASRARLYNQHYHWVLHDSSEAFSFYDFF

121 RRSNISIDADVDYVKQHMPNSGDKDAVVYTVYDVYSNGNHIGGQLNMTVNYELSCNRSSC 180 RRSNISIDADVDYVKQHMPHSKDKDAVVYTVYDVYSNGNHIGGQLNMTVNYELSCNRSSC RRSNISIDADVDYVKQHMPNSGDKDAVVYTVYDVYSNGNHIGGRLNMTVNYELSCNRSSC RRSNISIDADVDYVKQHMPNSGDKDAVVYTVYDVYSNGNHIGGRLNMTVNYELSCNRSSC S1 181 DGIRYLSSLHLRSKYGNREQLTDVVLRVATVVTQRPITWPPEQLLQFLNQINDTHIDGIA 240 DGIRYLSSLHLRSKYGNREQLTDVVLRVATVVTQRPITWPPEQLLQFLNQINDTHIDGIA DGIRYLSSLHLRSKYGNREQLTDVVLRVATVVTQRPITWPPEQLLQFLNQINDTHIDGIA DGIRYLSSLHLRSKYGNREQLTDVVLRVATVVTQRPITWPPEQLLQFLNQINDTHIDGIA

241 RFGFQLSLILKDLLQCGMNFTFKDRWSYGDYNGGSVGAVVDETADIGSAPCLATPGRLHL 300 RFGFQLSLILKDLLQCGMNFTFKDRWSYGDYNGGSVGAVVDETADIGSAPCLATPGRLHL RFGFQLSLILKDLLQCGMNFTFKDRWSYGDYNGGSVGAVVDETADIGSAPCLATPGRLHL RFGFQLSLILKDLLQCGMNFTFKDRWSYGDYNGGSVGAVVDETADIGSAPCLATPGRLHL

301 LSAIIETGYFRSICLFRTPHNAGIRGDVFLQPFSALVWFLFAGILLLIALMLWLAFYMES 360 LSAIIETGYFRSICLFRTPHNAGIRGDVFLQPFSALVWFLFAGILLLIALMLWLAFYMES LSAIIETGYFRSICLFRTPHNAGIRGDVFLQPFSALVWFLFAGILLLIALMLWLAFYMES LSAIIETGYFRSICLFRTPHNAGIRGDVFLQPFSALVWFLFAGILLLIALMLWLAFYMES P 361 KRMHRSWRLEFVPSLLSTCLISFGAACIQSSALTPRSTGGRLAYFALFLISFIMYNYYTS 420 KRMHRSWRLEFVPSLLSTCLISFGAACIQSSALTPRSTGGRLAYFALFLISFIMYNYYTS KRMHRSWRLEFVPSLLSTCLISFGAACIQSSALTPRSTGGRLAYFALFLISFIMYNYYTS KRMHRSWRLEFVPSLLSTCLISFGAACIQSSALTPRSTGGRLAYFALFLISFIMYNYYTS S2 421 VVVSSLLSSPVKSKIQTVQQLAESSLTVGLEPLSFTKSYLNYTQRPEIHLLRKRKIEPQS 480 VVVSSLLSSPVKSKIQTVQQLAESSLTVGLEPLSFTKSYLNYTQRPEIHLLRKRKIEPQS VVVSSLLSSPVKSKIQTVQQLAESSLTVGLEPLSFTKSYLNYTQRPEIHLLRKRKIEPQS VVVSSLLSSPVKSKIQTVQQLAESSLTVGLEPLSFTKSYLNYTQRPEIHLLRKRKIEPQS

481 HNPELWLPAEQGILRVRDRPGYVFIFEASSGYEYVERYYTQQEICDLNEILFRPELRLYT 540 HNPELWLPAEQGILRVRDRPGYVFIFEASSGYEYVERYYTQQEICDLNEILFRPELRLYT HNPELWLPAEQGILRVRDRPGYVFIFEASSGYEYVERYYTQQEICDLNEILFRPELRLYT HNPELWLPAEQGILRVRDRPGYVFIFEASSGYEYVERYYTQQEICDLNEILFRPELRLYT

541 HLHRNSSYKELVRLRLLRVLETGIYRKHRTWWARMRLHCYSQNFVITVGMEYVAPLFFML 600 HLHRNSSYKELVRLRLLRVLETGIYRKHRTWWARMRLHCYSQNFVITVGMEYVAPLFFML HLHRNSSYKELVRLRLLRVLETGIYRKHRTWWARMRLHCYSQNFVITVGMEYVAPLFFML HLHRNSSYKELVRLRLLRVLETGIYRKHRTWWARMRLHCYSQNFVITVGMEYVAPLFFML

601 LCGYLLVVLLLLLELAWQRYVQRSD 625 LCGYLLVVVLLLLELAWQRYVQRSN LCGYLLVLLLLLLELAWQRYAQRSD LCGYLLVLLLLLLELAWQRYAQRSD

Figure 6. (b)

Ir75bc_GI13611 Mojave MLYTQYINLNHTNTLSEHLKDNLMEHNLVKPGIYLDINCRSSALFLSMANEKHLFKGRYH 60 Catalina MLYTQYINLNHTNTLSEHLQDNLMEHNLVKLGIYLDINCRSSALFLSMANEKNLFKGRYH Baja MLYTQYINLNHTNTLSEHLKDNLMEHNLVKLGIYLDINCRSSALVLSMANEKHLFKGRYH Sonora MLYTQYINLNHTNTLSEHLKDNLMEHNLVKLGIYLDINCRSSALVLSMANEKHLFKGRYH

61 WLIYDQDFNLTEVHGRFEETQLYIDTELTYVEPNPKRDSFILYDLYNKGKHLGAKLNMTV 120 WLIYDQDFNLTEVLGRFEETQLYIDTELTYVEPNPKRDSFVLYDLYNKGKHLGAKLNMTV WLIYDQDFNLTEVHGRFEETQLYIDTELTYVEPNPKRDSFILYDLYNKGKHLGAKLNMTV WLIYDQDFNLTEVHGRFEETQLYIDTELTYVEPNPKRDSFILYDLYNKGKHLGAKLNMTV S1 121 DRQIKCNDQRCELSRYLSDLHKGNRLQHRRYLTGLTLRINAVVTAIPANSPETQIKEFLS 180 DRQIKCNDQRCELSRYLSDLHKGNRLQHRRYLTGLTLRINAVVTAIPANSPESQIKEFLS DRQIKCNDQRCELSRYLSDLHKGNRLQHRRYLTGLTLRINAVVTAIPANSPETQIKEFLS DRQIKCNDQRCELSRYLSDLHKGNRLQHRRYLTGLTLRINAVVTAIPANSPETQIKEFLS

181 RQDDINNDSFARFGFQTHQVFKDLLDCNYTYIFRDRWSDSELTGGLMGDMRNQSVDITAT 240 RQDDINNDSFARFGFQTHQVFKDLLDCNYTYIFRDRWSDSELTGGLMGDIRNQSVDITAT RQDDINNDSFARFGFQTHQVFKDLLDCNYTYIFRDRWSDSELTGGLMGDMRNQSVDITAT RQDDINNDSFARFGFQTHQVFKDLLDCNYTYIFRDRWSDSELTGGLMGDMRNQSVDITAT

241 GFLFSSRRTKYFKMLSWHSSFRSTCMFLNPKSSAVELRISEFLQPFSATVWFIFGSLLLI 300 GFLFSSRRTKYFKMLSWHSAFRSTCMFLNPKSSAVELRISEFLQPFSATVWFIFGSLLLI GFLFSSRRTKYFKMLSWHSSFRSTCMFLNPKSSAVELRISEFLQPFSATVWFIFGSLLLI GFLFSSRRTKYFKMLSWHSSFRSTCMFLNPKSSAVELRISEFLQPFSATVWFIFGSLLLI P 301 AGLLLWMTFRLERRLNHIDIRPSLLTSCLLSFGAACIQGAWLLPRSTGGRMVFYAVMLLC 360 AGLLLWMTFRLERRLNDIDIRPSLLTSCLLSFGAACIQGAWLLPRSTGGRMVFYAVMLLC AGLLLWMTFRLERRLNHIDIRPSLLTSCLLSFGAACIQGAWLLPRSTGGRMVFYAVMLLC AGLLLWMTFRLERRLNHIDIRPSLLTSCLLSFGAACIQGAWLLPRSTGGRMVFYAVMLLC S2 361 FLLYNFYTSVVVSILLGEPPKSNIRTIQQLADSNLEVSVQPLIYTKVYIETSSYPDVRSL 420 FLLYNFYTSVVVSILLGEPPKSNIRTIQQLADSNLEVSVQPLIYTKVYIETSSYPDVRSL FLLYNFYTSVVVSILLGEPPKSNIRTIQQLADSNLEVSVQPLIYTKVYIETSSYPDVRSL FLLYNFYTSVVVSILLGEPPKSNIRTIQQLADSNLEVSVQPLIYTKVYIETSSYPDVRSL

421 HLNKILNSKRDNIWLPPDEGVRMVRNFPGFVYITEASSSYAFVRKHFLPHEICELNEILL 480 HLNKILNSKRDNIWLPPDEGVRMVRNFPGFVYITEASSSYAFVRKHFLPHEICELNEILL HLNKILNSKRDNIWLPPDEGVRMVRNFPGFVYITEASSSYAFVRKHFLPHEICELNEILL HLNKILNSKRDNIWLPPDEGVRMVRNFPGFVYITEASSSYAFVRKHFLPHEICELNEILL

481 RDETSAHTTVAINSSYAELFKQNYLRLLETGVHFKHFRYWVRNKLHCYDSNRGVVVGMDS 540 RDETSAHTTVAINSSYAELFKQNYLRLLETGVHFKHFRYWVRNKLHCYDSNRGVVVGMDS RDETSAHTTVAINSSYAELFKQNYLRLLETGVHFKHFRYWVRNKLHCYDSNRGVVVGMDS RDETSAHTTVAINSSYAELFKQNYLRLLETGVHFKHFRYWVRNKLHCYDSNRGVVVGMDS

541 AGPLFLLLICAYILCLFVLGLEILFHRRQQRAGRQ 575 AGPLFLLLICAYILCLFVLGLEILFHRRQQRAGRQ AGPLFLLLICAYILCLFVLGLEILFHRRQQRAGRQ AGPLFLLLICAYILCLFVLGLEILFHRRQQRAGRQ

Figure 7. (a, b)

(a) D. melanogaster D. virilis D. arizonae D. mojavensis (Catalina)

hexyl acetate A BC B C spermidine B B B A cadaverine B B B A phenethylamine A B B A pyridine C B B A 1,4-diaminobutane B B B A phenylacetaldehyde (E)-2-hexenal A BC B C butyraldehyde A B B B hexanol A B B B isobutyric acid BC C B A hexanoic acid B AB A AB valeric acid B B A AB propionic acid AB B AB A acetic acid C A B C 2-oxopentanoic acid C C B A butyric acid A B A A paraffin oil H20 blank

-50 0 50 100 150 200 -50 0 50 100 150 200 -50 0 50 100 150 200 -50 0 50 100 150 200

Spikes/sec

(b)