The nature and nurture of female receptivity
Gorter, Jenneke Anne
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Gorter, J. A. (2018). The nature and nurture of female receptivity: A study in Drosophila melanogaster. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
The nature and nurture of female receptivity
a study in Drosophila melanogaster
This research has been carried out in the Evolutionary Genetics, Behaviour and Development (EGDB) group at the Groningen Institute for Evolutionary Life Sciences (GELIFES) according to the requirements of the Graduate School of Science (Faculty of Science and Engineering, University of Groningen, The Netherlands).
This research was supported by a Neuroscience Research School BCN/NWO Graduate Program grant (reference 022.OO4.OO8). Financial support for the printing of this thesis was kindly provided by the University of Groningen and the Faculty of Science and Engineering. Cover design, figures and lay-out: Jenke A. Gorter and Jean-Christophe Billeter
Printed by: Gildeprint, Enschede
ISBN: 978-94-034-0455-4 (printed version)
The nature and nurture of female
receptivity
A study in Drosophila melanogaster
PhD Thesis
to obtain the degree of PhD at the University of Groningen
on the authority of the Rector Magnificus Prof. E. Sterken
and in accordance with the decision by the College of Deans. This thesis will be defended in public on
Friday 23 March 2018 at 16:15 hours
by
Jenneke Anne Gorter
born on 27 February 1990 in Dordrecht
Supervisor
Prof. J.-C. Billeter
Assessment committee
Prof. B. Wertheim Prof. A.T. Groot Prof. T. Chapman
Table of contents
Chapter 1: General Introduction 7
Chapter 2: A method to test the effect of environmental cues on mating behaviour
in Drosophila melanogaster 23
Chapter 3: The nutritional and hedonic value of food modulate sexual receptivity in Drosophila melanogaster females 39
Chapter 4: Immediate social context and early social experience differentially modulate female sexual receptivity in Drosophila melanogaster 59
Chapter 5: The odorant receptor Or47b promotes sexual receptivity in mated Drosophila melanogaster females 83
Chapter 6: Different genes influence virgin and mated female receptivity in Drosophila melanogaster: a genome wide association study 107
Chapter 7: Synthesis 137
References 147
Nederlandse samenvatting 177
Personal information 183
General introduction
Introduction
Sexual reproduction requires cooperation between a male and a female to produce viable offspring. The likelihood to engage in reproduction differs per individual based on genetic background, which leads, for example, to different levels of sexual activity or sensitivity to cues from the other sex. The environment the pair meets in also impacts the individuals’ likelihood of reproduction. An environment with high nutritional quality is a good place for reproduction due to available resources for egg production and offspring rearing. In addition to food, social interactions beyond the pair can facilitate reproduction, since living in a group reduces the individual risk of predation, increases offspring survival and provides higher mate choice. Therefore, individuals have likely evolved mechanisms to couple sex drive to the environment. Lastly, prior experiences of each partner might modulate their sexual activity. For example, after many rejections males might reduce investment in courtship displays, whereas with prior successful mating experience a female might be less receptive to mate again. Reproduction is thus not only an interaction between a male and a female influenced by their genomes, but is also majorly impacted by their environment and prior experience.
Sexual reproduction
Different organisms have different mating systems; a description of the way sexual interactions are structured around reproduction. Mating systems include monogamy (partner bond between one male and one female) and polygamy (several mating partners per individual). Polygamous mating systems are polygyny (partner bond between one male and several females), polyandry (partner bond between one female and several males), polygynandry (both sexes have several partners) and promiscuity (no partner bond, frequent different partners, often implying indiscriminate mate choice). Males are commonly seen as the promiscuous sex and females are depicted as choosy (Bateman, 1948). This is for example illustrated by the higher occurrence of polygynous pair bonds in human populations than polyandrous pair bonds (Archetti, 2013). However, development in paternity testing is revealing that female promiscuity, or polyandry, is more widespread than previously thought in the animal kingdom and might be the rule rather than the exception (Holman and Kokko, 2013; Parker and Birkhead, 2013; Taylor et al., 2014). The same realisation is starting to emerge in the model organism Drosophila melanogaster. For a long time, female fruit flies were thought to accept a low number of mates and go for long periods of time without mating, but it is becoming clear that D. melanogaster is a promiscuous species in which both sexes frequently mate with different individuals within short timeframes (Giardina et al., 2017; Imhof et al., 1998; Ochando et al., 1996).
9 Ch ap te r 1
Drosophila melanogaster male courtship and female receptivity The mating of Drosophila melanogaster consists of males performing an elaborate courtship
sequence towards females (Spieth, 1974). This male display consists of the following
behaviours: orienting, tapping, following, wing vibration, licking genitalia, mounting and copulation attempt ((Bastock and Manning, 1955; Spieth, 1974), illustrated in figure 1). When subjected to male courtship, females can accept or reject the males’ interest with rejection signals (Hall, 1994). Rejection is signalled through wing fluttering, decamping, fending, kicking and full ovipositor extrusion (Bastock and Manning, 1955; Dukas and Scott, 2015; Hall, 1994; Markow and Hanson, 1981). Receptivity to mating is signalled by the females slowing down (Bussell et al., 2014; Fabre et al., 2012; Markow and Hanson, 1981; Tompkins et al., 1982) and increased abdominal preening, partial ovipositor extrusion and droplet emission, which have been suggested to function to spread a chemical cue signalling willingness to the mate (Lasbleiz et al., 2006). Next, females spread their wings to allow males to mount and finally open the vaginal plates to accept a copulation attempt (Bastock and Manning, 1955). After a copulation attempt is accepted and mating has taken place, the male and female go their separate ways.
Figure 1: Male courtship and female receptivity Schematic overview of the different components of the male
courtship sequence and the observed female cues of receptiveness. The first steps in the male courtship sequence are often repeated several times, before the first female cues of receptiveness start to appear.
Sexual conflict over female multiple mating
After a first successful mating, both sexes can, in theory, move on to the next sexual partner. However, when females remate this diminishes the chances of fertilisation for the first male (Bateman, 1948; Lefevre and Jonsson, 1962) as the second male’s sperm fertilises about 80% of the offspring (Clark et al., 1995). Males have evolved countermeasures to increase their individual reproductive success. The first of such countermeasures is the adjustments of the ejaculate size when a male senses rival males in his environment (Garbaczewska et al., 2013; Sirot et al., 2011; Wigby et al., 2009). Since the sperm of different males compete within the female reproductive tract for fertilisation (Lefevre and Jonsson, 1962; Manier et al., 2010; Parker and Pizzari, 2010), males with bigger ejaculates might gain an advantage as they transfer more sperm to compete with that of other males (Letsinger and Gromko, 1985). Not only the size, but also the composition of the ejaculate is adjusted in the presence of competitor males (Wigby et al., 2009). By increasing the proportion of peptides that decrease
female receptivity like Sex peptide (Sp,(Chapman et al., 2003; Chen et al., 1988; Liu and Kubli, 2003)), discussed in more detail in section “A decrease in female receptivity after mating”, males might decrease the chances of being outcompeted by consecutive males. Another form of countermeasure is mate guarding, chemically or physically. Chemical mate guarding is achieved by transferring pheromones, such as cis-Vaccenyl Acetate (cVA) and 7-Tricosene (7-T), that make mated females less attractive to other males (Jallon, 1984; Laturney and Billeter, 2016; Yew et al., 2009; Zawistowski and Rollin, 1986). Physical mate guarding can be observed as a mating plug that might make it physically more difficult to remate, as it does in butterflies (Kawahara et al., 2017). More importantly, the mating plug in D. melanogaster serves to assure sperm retention (Avila et al., 2015; Lung and Wolfner, 2001) and contains compounds that make females less attractive (Guiraudie-Capraz et al., 2007) and less receptive (Bretman et al., 2010).
These male adaptations make it more difficult for a female to gain consecutive matings. Females have, however, evolved countermeasures to regain control over reproduction. For example, females eject the mating plug, part of the ejaculate and the associated pheromones and possibly peptides (Laturney and Billeter, 2016; Lee et al., 2015; Lüpold, 2013; Manier et al., 2010), recovering her attractiveness and making her more accessible for following mates. These illustrations show that male and female D. melanogaster are in sexual conflict over the level of female receptivity after the first mating (post-mating receptivity).
Female multiple mating is determined by her receptivity
Why females remate with several males is a controversial topic. For a virgin female the reason to mate is intuitive, it serves to ensure offspring production and therefore her fitness (Kokko and Mappes, 2005). Additionally, wild-caught virgin D. melanogaster females have a lower lifespan than mated ones, so mating might even provide direct health benefits or alternatively there might be a cost of staying virgin (Markow, 2011). Since female fitness is not constrained by sperm availability (a single male ejaculate has more sperm than a female has eggs), but by egg production, a female does not “need” to remate for several days until her sperm storage are exhausted (Bateman, 1948; Lefevre and Jonsson, 1962). Additionally, several costs to mating have been documented. Mated females suffer decreased immunity caused by male seminal fluid peptides (Chapman et al., 1995; Fedorka et al., 2007; Schwenke and Lazzaro, 2017; Short et al., 2012), physical harm (Kamimura, 2007) and decreased lifespan and lifetime fecundity (Kuijper et al., 2006). However, multiple matings have also been proposed to serve different benefits like ensuring fecundity, producing genetically diverse offspring (Billeter et al., 2012; Jennions and Petrie, 2000), trading up for a better quality or adapted male (Bleu et al., 2012; Jennions and Petrie, 2000; Long et al., 2010; Seeley and Dukas, 2011) and to receive seminal peptides that increase egg production and the fitness of the offspring (Fricke et al., 2010; Herndon and Wolfner, 1995; Priest et al., 2008). Another reason for remating might be “convenience polyandry”, whereby a female does not
11 Ch ap te r 1
remate to gain direct benefits, but to reduce harassment by males as mating typically reduces this for a short time after mating (Newport and Gromko, 1984; Rowe, 1992).
Taking the costs and benefits into account, it is suggested that females should evolve an intermediate copulation number to balance the costs and benefits and maximise their fitness (Arnqvist and Nilsson, 2000). However, number of copulations is not a direct female trait and females cannot be expected to predict or keep track of the number of mate encounters (Kokko and Mappes, 2012). Therefore, selection must act on a female sexual trait that can indirectly determine the number of copulations. This trait is suggested to be a female’s acceptance threshold or level of receptivity (Kokko and Mappes, 2012). When a female has a high level of receptivity, she is more likely to accept than reject a male each time she encounters one and, therefore, will end up with a higher number of copulations during her lifetime than the suggested intermediate number (Kokko and Mappes, 2012). This theory provides an explanation of why females mate more often than necessary to maximise fitness, but it assumes a fixed level of receptivity over a female’s lifetime (Kokko and Mappes, 2012). However, the level of receptivity is known to be plastic and influenced by other factors like a female’s mating state (Kokko and Mappes, 2005), with mated females displaying lower levels of receptivity than virgin females. Female mating is thus instructed by her sexual receptivity and the level of receptivity can be adapted to mating state.
Neuronal circuitry of virgin receptivity
Research on the mechanisms underlying a virgin female’s sexual receptivity has revealed sensory pathways involved in detecting and processing different male courtship signals. The most important male sexual signal for a virgin female appears to be the quality of the courtship song (Dickson, 2008; Markow, 1987; Rybak et al., 2002). A second main stimulating medium for females during courtship are pheromones (Billeter et al., 2009). The identified elements of the neuronal circuitry regulating virgin receptivity are based on sensing and responding to these courtship signals. Females express an odorant receptor called Or67d (Bartelt et al., 1985; Kurtovic et al., 2007) that detects a stimulatory male pheromone, 11-cis-vaccenyl acetate (cVA) (Bartelt et al., 1985; Kurtovic et al., 2007). Females lacking this receptor are less receptive than wild-type females (Kurtovic et al., 2007). Further sensory processing of this pheromone as well as the courtship song is achieved by neuron clusters in the posterior dorsal protocerebrum in the central brain that gets activated when female are
exposed to these male signals (Zhou et al., 2014). Inactivationof these neurons via tetanus
toxin expression leads to decreased copulation by females (Zhou et al., 2014). After detection of male cues, receptive females show pausing behaviour preceding a copulation (Bussell et al., 2014). This pausing behaviour is coordinated by Apterous-expressing neurons in the brain (Aranha et al., 2017; Ringo et al., 1991) and abdominal-B-expressing neurons in the abdominal ganglion, ventral nerve cord and reproductive tract. Together, these accounts show that the main mechanism for virgin receptivity is based on how females determine male
courtship cues and show acceptance behaviour, suggesting increased receptivity in direct response to a male encounter and little involvement of any other factors like environmental richness.
A decrease in female receptivity after mating
Females undergo behavioural changes after mating known as the post-mating response, which results in changes in diet preference (Ribeiro and Dickson, 2010), increased oviposition (Heifetz et al., 2005; Herndon and Wolfner, 1995; Ram and Wolfner, 2007) and decreased receptivity towards courting males (Manning, 1967). Females signal this decrease in receptivity to males with full ovipositor extrusion, a rejection behaviour exclusive to mated females (Manning, 1967). The change in receptivity has been subdivided into a short-term or “copulation” effect, from directly after mating up to 48h, and a long-term or “sperm” effect, from 24h to 10 days after mating (Manning, 1967).
The short-term effect has been linked to substances produced and secreted by the accessory glands of Drosophila males (Chen et al., 1988; Kalb et al., 1993). Females mated with males missing the main accessory gland cells show increased remating at 24h as compared to mated with wild-type males (Kalb et al., 1993), accessary gland peptides must thus be involved in the reduction of receptivity shortly after mating. Among those peptides is Sex peptide (Sp), which has the ability to decrease remating demonstrated by artificial injection in the abdominal cavity or ectopic expression by means of transgenes in females (Aigaki et al., 1991; Chen et al., 1988). Furthermore, Sp has been weakly implicated in the short-term response at 24h (Chapman et al., 2003; Liu and Kubli, 2003). However, due to this weak response, it is debated whether Sp is the actor of the short-term decrease in receptivity. Another factor able to elicit a short-term decrease by artificial injection into the female abdomen is the ejaculatory duct peptide DUB99B (Saudan et al., 2002). Whether this peptide is actually involved in the short-term effect is unclear as accounts of females mated with males that lack accessory glands, but still produce normal amounts of DUB99B, do not show the characteristic decrease in post-mating receptivity, suggesting not only that DUB99B is not sufficient but also that accessory gland proteins are necessary for this effect (Rexhepaj et al., 2003; Xue and Noll, 2000). A third factor involved in the short-term decrease of receptivity is the mating plug protein PEBII produced in the ejaculatory bulb, which directly decreases female receptivity within the first 4h after mating (Bretman et al., 2010). Even though PEBII is a mating plug protein and mutant males produce smaller plugs, this does not lead to changes in fecundity (Bretman et al., 2010). This suggest that the effect of PEBII is not due to its function as mating plug protein, but it might work by acting on the female receptivity pathway.
For the long-term response of mating on female receptivity, both sperm and the accessory gland peptide Sp are necessary (Kalb et al., 1993; Peng et al., 2005). Sp elicits a reduction in female receptivity after ectopic expression or artificial injection (Aigaki et al., 1991; Chen et al., 1988), but the proof of Sp’s involvement in the long-term effect comes from knockdown studies with RNA interference and genetic mutations where males without
13 Ch ap te r 1
Sp are unable to elicit a long-term decrease in female receptivity (Chapman et al., 2003; Liu and Kubli, 2003). For the long-term effect in receptivity reduction, Sp requires association with sperm (Kalb et al., 1993). Sp is bound to and gradually released from sperm, due to cleavage at the N-terminal end of the peptide (Peng et al., 2005). To assure this effect of Sp, the picture is more complex with several other male accessory gland peptides involved which, for example, stabilize the Sp-sperm bond, ensure proper transfer of sperm and localisation into the sperm storage organs (Ram and Wolfner, 2007; 2009; Sirot et al., 2009; Sitnik et al., 2016).
The effect of Sp on female receptivity is mediated by the G-protein coupled Sex peptide receptor (SPR), located in the female reproductive tract as well as throughout the central nervous system (Yapici et al., 2008). A small number of SPR sensory neurons on each side of the uterus are necessary for sensing Sp and to elicit a decrease in female receptivity (Häsemeyer et al., 2009; Yang et al., 2009). These neurons relay information onto neurons in the abdominal ganglion (SAG neurons, (Feng et al., 2014; Rezával et al., 2012)), which in turn project to the dorsal protocerebrum in the central nervous system which is an area known to be involved in female receptivity (Feng et al., 2014). Inhibition of the SAG neurons by Sp sensory neurons signals mated state to the protocerebrum and decreases female receptivity, while hyperactivation leads to virgin-like female receptivity (Feng et al., 2014). Additionally, the post-mating response can be induced by Sp acting directly on neurons in the central nervous system in the absence of SPR, but only with ectopic neuronal expression of Sp or a leaky blood-brain-barrier (Haussmann et al., 2013). This suggest that Sp also acts directly on the central nervous system, bypassing the SPR pathway.
Reduced female receptivity after mating is not final, fixed or general. Females can counteract these effects by ejecting the mating plug including the short-term factors as well as the ejaculate, potentially including a portion of the factors for the long-term effect (Laturney and Billeter, 2016; Lee et al., 2015; Manier et al., 2010). Sperm ejection can therefore restore females’ ability to remate. Next to that, not all strains of D. melanogaster show the same level of post-mating response, some strains, like wild-type Canton-S, stay more receptive, both short- and long-term, even though they receive the same factors as less receptive strains (Denis et al., 2017).
Mechanisms of post-mating receptivity, when it does occur
Beyond the effect of male peptides inducing decreases in receptivity as discussed above, the mechanisms and modulators of female post-mating receptivity are not well understood. One potential contributing factor to this incertitude are the type of behavioural assays employed by the field that studies it. Female post-mating receptivity has mostly been explored using an assay in which the female is provided with an initial opportunity to mate, followed by opportunities to remate every 24h or 48h for a fixed amount of time (30 min- 2h). In between mating sessions, the female is isolated on egg laying substrate. This confinement assay has revealed that female post-mating receptivity, quantified as the time to remating (in days) or
the likelihood of remating at a certain time point, has a genetic basis that can be selected for (Fukui and Gromko, 1991a; 1991b; 1991c; Gromko and Newport, 1988a; 1988b; Pyle and Gromko, 1981) and that sperm presence in the female sperm storage organs is a main determinant of remating (Letsinger and Gromko, 1985; Newport and Gromko, 1984). Females that receive more sperm are less likely to remate and take longer if they do and, similarly, females mated with sperm-depleted males are more willing to remate (Lefevre and Jonsson, 1962). In accordance, studies investigating the genetics underlying post-mating receptivity have identified genes associated with sperm storage and immunity, which is also affected by male seminal peptides, as well as odorant binding proteins possibly facilitating interactions between males and females (Giardina et al., 2011; Lawniczak and Begun, 2004). This confinement assay, therefore, shows an effect of sperm presence in female storage and suggests that female receptivity is more reflective of a reflex to substances transferred by the male rather than a malleable response.
However, the effect of sperm presence can be modulated and may be a result specifically observed in the confinement assay. For example, the availability of food for females in this assay can increase her willingness to remate even before her sperm storage is depleted (Harshman et al., 1988). Furthermore, this assay suggests that females rarely remate or at least take several days to do so, which is in conflict with the finding that females in the wild mate at least once per day (Giardina et al., 2017) and often carry sperm from several males (Imhof et al., 1998; Ochando et al., 1996). Other assays have, therefore, tried to mimic natural variation by continuously housing mating pairs for longer periods of time, up to 48h (Krupp et al., 2008; 2013; Lefevre and Jonsson, 1962; Newport and Gromko, 1984; Smith et al., 2017; van Vianen and Bijlsma, 1993). In such continuous assays, females remate more often with an average of up to 6 times per 24h depending on the strain and context (Billeter et al., 2012; Krupp et al., 2008; 2013). Remating in this assay does not depend on the amount of sperm left in female storage (Newport and Gromko, 1984). The continuous assay has been suggested to result in more harassment for the female and less opportunity to decamp, and thus reject the male (Newport and Gromko, 1984). This is used to explain why a higher post-mating receptivity is observed compared to the confinement assay. However, 50 percent of females remate within 6h after the first mating (Smith et al., 2017; van Vianen and Bijlsma, 1993) both when continuously housed with a second male and when the second male is introduced 3h later (van Vianen and Bijlsma, 1993), suggesting that the effect is due to an internal change in receptivity of the female rather than continuous inability to reject the male. Additionally, the level of post-mating receptivity can be largely explained by female rather than male genotype, as 47 per cent of the variance in post-mating frequency depends on the strain of the female and only 11 on that of the male (Billeter et al., 2012), giving difference in male courtship and seminal peptides less impact than differences in female willingness. Thus, post-mating receptivity depends on female genetics and different factors can affect it depending on context. What the genetic differences and modulators are that determine mated female receptivity are, however, still largely unidentified.
15 Ch ap te r 1
Environmental influences on female receptivity
Since high receptivity is a costly phenotype due to resource allocation towards egg-production as well as the costs incurred during mating, female sexual receptivity is likely to be influenced by environmental factors. Females might increase receptivity when they receive cues that resources are plentiful. Food is a likely candidate to influence receptivity as females are dependent on this external resource to produce eggs and viable offspring (Becher et al., 2012; Bownes et al., 1988; Lee et al., 2008; Terashima, 2004). Indeed, the availability of food (determined in a confinement assay) influences post-mating, but not virgin, receptivity (Harshman et al., 1988). The modulation of receptivity connected to food availability only occurs when females still have sperm in storage. Female receptivity increases by sperm-depletion and this is not further affected by food availability (Harshman et al., 1988). Additionally, females fed on high nutritional diets, protein levels or overall food content, increase remating measured as time to remating in days or occurrence of remating over the whole lifespan with daily exposure (Chapman and Partridge, 1996; Schultzhaus and Carney, 2017). Food can thus modulate female receptivity, but the mechanisms underlying this modulation of receptivity are unknown.
Another factor that influences female receptivity is social context. There are several reasons why females can be expected to increase sexual receptivity in a dense social context, an environment with many individuals of the same species. The first reasons is that there are more males for the female to choose from providing the possibility to produce more diverse offspring (Billeter et al., 2012; Jennions and Petrie, 2000). Second, females prefer to lay their eggs communally (Duménil et al., 2016; Lin et al., 2015; Lof et al., 2009; Wertheim et al., 2002a; Yang et al., 2008), because higher density of adults and larvae keeps the fungal growth to a minimum which would otherwise decrease larval development (Wertheim et al., 2002b). Third, a female might experience competition with other females for egg laying substrate or available mates, especially after mating, which could be reflected in an increase in aggression as well as higher receptivity to increase her chances of getting the best mates. Indeed, it is shown that females increase aggression towards other females after mating (Bath et al., 2017), but whether this is further increased in higher density is unclear. In regards to sexual receptivity, females have indeed achieve mating faster when tested in big group as opposed to single pairs (Ellis and Kessler, 1975; Laturney and Billeter, 2016), post-mating receptivity can increase in response to higher density (Harshman et al., 1988) and females caught inside a high density winery have a higher paternity estimations as opposed to low density woodlands outside (Marks et al., 1988). Lastly, more genetic diversity in the social context during mating increases the number of copulations in 24h of mature females, which is blocked when females are unable to sense their environment through classical odorant receptors (Billeter et al., 2012; Krupp et al., 2008). However, how social density instructs female receptivity is still to be determined.
Is post-mating receptivity a return to virgin receptivity?
It is clear that virgin females are much more sexually receptive than mated females. Mated female receptivity is often interpreted as the return of virgin-like state after several days (Kalb et al., 1993; Peng et al., 2005), but is it the same receptivity? Theory predicts that virgin and mated females should not use the same rules when it comes to receptivity due to a different balance in the cost-benefit of mating as previously discussed (Arnqvist and Nilsson, 2000; Jennions and Petrie, 2000; Laturney and Billeter, 2014). However, they could use the same mechanisms, but perhaps with emphasize on different cues.
For virgin receptivity, most neuronal pathways identified are involved in the response a female shows towards the male’s courtship advances (Aranha et al., 2017; Bussell et al., 2014; Kurtovic et al., 2007; Zhou et al., 2014). Even though a mated female similarly responds to the courtship signals (song and cVA) and decreases movement before mating, some of the accounts on virgin receptivity show that the same manipulations do not impact post-mating receptivity, measured as mated female full ovipositor extrusion (Aranha et al., 2017; Bussell et al., 2014). This lack of effect of the virgin mechanism manipulations on post-mating receptivity suggests that even though mated females signal receptivity in a similar manner, this behaviour might be invoked differently in virgin versus mated females. Therefore, virgin and post-mating receptivity might not share the same mechanisms. Most of the known mechanisms of mated female receptivity are involved in sensing or dealing with male compounds transferred during mating (Chapman et al., 2003; Letsinger and Gromko, 1985; Liu and Kubli, 2003), which is a specific challenge for mated females. However, this challenge could be a new factor feeding into the same mechanisms for receptivity and, therefore, does not provide any indication whether virgin and mated females use the same mechanisms.
Reviewing studies that have both reported virgin and post-mating receptivity suggests that these two processes rely on different mechanisms. First, selection experiments based on female remating speed (assessed with a confinement assay) show that the resulting virgin mating latency is either uncorrelated, positively or negatively correlated with remating latency (Gromko and Newport, 1988b; Pyle and Gromko, 1981). This shows that genetic selection for fast post-mating receptivity does not select for either fast or slow virgin receptivity suggesting different genetic architectures for virgin and mated female receptivity. Second, an account of female mating in a continuous assay also shows that time to first remating is uncorrelated with virgin mating latency (van Vianen and Bijlsma, 1993). Third, an attempt to correlate variation in 10 candidate genes to variation in several measures of female sexual behaviour, including mating latency and remating, finds variation in some of the same genes correlated to both virgin and post-mating receptivity, but never the exact same variation site (Giardina et al., 2011). These examples suggest a different genetic background for the two behaviours.
Research has focussed on mechanisms of female receptivity as virgins, the switch between virgin to mated state facilitated by Sp and what factors explain an earlier return of
17 Ch ap te r 1
receptivity in mated females. However, post-mating receptivity might not be the return of virgin receptivity, but rather a different phenomenon.
Thesis overview
Females are more promiscuous than was previously assumed which suggests that female receptivity is a more nuanced behaviour than a simple on-or-off state where virgin females are receptive and mated females are not. Female receptivity starts with mature virgin females with their specific “decision” rules about whom to mate with and setting their level of receptivity. After a first mating the females’ receptivity levels change due to the post-mating response. As mated females are expected to have different costs and benefits determining their receptivity, this predicts that post-mating receptivity is a different trait than virgin receptivity. Here, my main aim is to understand what factors influence female virgin and post-mating receptivity and how these modulators are signalled and sensed. To investigate these factors, I used a continuous mating assay as it more closely mimics the natural levels of female receptivity and allows for continuous manipulation of the environment in which both virgin (latency to virgin mating) and post-mating (latency to first remating and number of copulations in 24h) receptivity are assessed in one assay. The environmental manipulations are achieved by methods described in chapter 2, including the use of an airpump system to supply specific odours and manipulation of the components present in the food substrate or mating area to quantify the effect on female mating behaviour.
First, I focus on the influence of environment on female receptivity. The effect of food, which specific nutrient, and how these are sensed to influence female receptivity are tested in chapter 3. For testing nutritional cues, the airpump system mentioned above is used as well as different compositions of the food substrates. To determine how specific food components are sensed, genetic mutants for sensory modalities and knockdown of sensory neurons by use of the Gal4-UAS system (explained in box 1) are tested in the different environmental conditions. Next, a similar approach is taken for the social environment in chapter 4. The social environment is manipulated by testing different group sizes and the airpump system is employed to determine which signals are necessary to sense group size affecting female receptivity. Additionally, female receptivity is tested after manipulation of the social raising environment as experience can modulate sexual behaviours. These chapters thus cover extrinsic cues involved in female sexual receptivity.
Second, several intrinsic cues of sexual receptivity are investigated. A candidate gene approach is taken for odorant receptors detecting fly odours, namely Or47b and Or88a, in chapter 5. These two candidate genes are explored through genetic mutants and manipulation of the cells these odorant receptors are normally expressed in through the Gal4-UAS system. For the brains of one specific Gal4-Gal4-UAS manipulation, the olfactory regions associated to these odorant receptor neurons are analysed with immunohistochemistry and volumetric analysis. Last, in chapter 6, a genome wide association (GWA) approach is taken to identify new candidate genes and tissues of interest for both virgin and mated female
receptivity. For mated female receptivity a follow-up RNA interference study is performed for a subset of identified genes by use of the Gal4-UAS system. This covers some of the intrinsic factors involved in female receptivity. Altogether, this thesis provides new insights and areas for further exploration of both the nurture and nature of female sexual receptivity.
19 Ch ap te r 1
Box 1: Gal4-UAS binary system
In this thesis, the involvement of specific genes and tissues are questioned for their influence on female sexual receptivity. Next to mutations in the genes of interest, a binary system is used for targeted manipulation. This binary system is the Gal4-UAS system (Brand and Perrimon, 1993). Genetically modified fly stocks are generated to express a yeast (Saccaromyces cerevisiae) derived transcription activator protein, Gal4. Gal4 is controlled by a D. melanogaster promotor, also known as a driver, to ensure that Gal4 is only expressed in tissue in which this driver is activated. Gal4 drivers can have wide expression patterns, by use of a promotor region of a protein expressed in all neuronal tissue for example, or they can have very specified expression, in cells only expressing a very specific protein. On its own Gal4 has little effect in D. melanogaster tissue as it is not endogenous. One of the main targets of Gal4 is a cis-regulatory site (DNA sequence) called Upstream Activating Sequence (UAS). When Gal4 binds to this regulatory site, the DNA sequence UAS regulates can be transcribed. A second set of genetically modified fly stocks, therefore, have an UAS site inserted into the genome followed by a transgene. This transgene can be any genetic sequence ranging from reporter proteins to proteins that manipulate the electric activity of neurons. As for Gal4, on its own the UAS transgene has little effect as it is not recognized by endogenous D. melanogaster transcription activators. Then, to manipulate or visualize target tissue, the appropriate Gal4 line is crossed to the UAS line that serves the manipulation’s purpose. The offspring of this cross has both transgenes in its genome and the UAS transgene is expressed in the tissue of interest. Here, this system is used to manipulate the activity and development of odorant receptor tissue, to localize rescue of mutations to specific cells and to inhibit candidate gene expression with RNA interference in all brain tissue or a specific brain area. Female offspring harbouring both transgenes are tested for their sexual receptivity as well as, for some specific hypotheses, male offspring for their sexual activity.
Figure 2: Gal4-UAS binary system. A schematic overview of the Gal4-UAS binary system to drive transgenes in
specific tissue. The Gal4 and UAS transgene complexes are kept in separate F0 stocks. Only after a stock for Gal4 is crossed with a stock for UAS the two transgenes occur in the F1 population and the transgene of interest is expressed in the target tissue. This system is used throughout this thesis, where the F1 offspring is tested for specific questions to do with female sexual receptivity.
A method to test the effect of environmental
cues on mating behaviour in
Drosophila
melanogaster
Jenke A. Gorter and
Jean-Christophe Billeter
Abstract
An individual’s sexual drive is influenced by genotype, experience and environmental conditions. How these factors interact to modulate sexual behaviours remains poorly understood. In Drosophila melanogaster, environmental cues, such as food availability, affect mating activity offering a tractable system to investigate the mechanisms modulating sexual behaviour. In D. melanogaster, environmental cues are often sensed via the chemosensory gustatory and olfactory systems. Here, we present a method to test the effect of environmental chemical cues on mating behaviour. The assay consists of a small mating arena containing food medium and a mating couple. The mating frequency for each couple is continuously monitored for 24 h. Here we present the applicability of this assay to test environmental compounds from an external source through a pressurized air system as well as manipulation of the environmental components directly in the mating arena. The use of a pressurized air system is especially useful to test the effect of very volatile compounds, while manipulating components directly in the mating arena can be of value to ascertain a compound’s presence. This assay can be adapted to answer questions about the influence of genetic and environmental cues on mating behaviour and fecundity as well as other male and female reproductive behaviours.
Keywords
Drosophila melanogaster, female sexual receptivity, reproductive behaviour, environmental cues, yeast, nutrition, mating behaviour, olfaction.
Video link
25 Ch ap te r 2 Introduction
Reproductive behaviours typically have high energy costs, especially for females, who produce larger gametes than males and must carefully chose the conditions to raise their developing offspring. Because of the energy cost, it is not surprising that reproduction is connected to nutritional conditions. This is true in most, if not all, animals including mammals, whose puberty can be delayed by malnutrition, and whose sexual drive can be negatively affected by food-restriction (Hileman et al., 2000).
The reproduction of the genetic model organism Drosophila melanogaster is also affected by nutritional conditions. Males court at higher level in the presence of food volatiles (Grosjean et al., 2011), and females are more sexually receptive in the presence of yeast, a major nutrient for egg production and offspring survival (Harshman et al., 1988). This evolutionary conserved reproductive response to food offers the opportunity to study mechanisms that connect environmental food availability to sexual reproduction in a genetically tractable and time-efficient organism. Indeed, work in D. melanogaster has implicated the insulin pathway as an important regulator of the connection between food and mating behaviour (Wigby et al., 2011). It has also shown that the act of mating itself changes the food preference of females as well as the associated chemosensory neurons (Hussain et al., 2016; Ribeiro, 2013; Walker et al., 2015).
It is clear that food cues affect reproductive behaviours in D. melanogaster. These effects seem to mainly affect females, specifically those who have already mated (Gorter et al., 2016). However, to test these acute effects of environmental conditions the assay classically used for female mating behaviour might not be very suitable due to the long interruptions between mating episodes. In the classic remating assay, a virgin female first mates with a male, and is immediately isolated and presented with a new male 24 to 48 h later. This classic assay has been used with great success to identify components of the male ejaculate that modify the female behaviour and the female response (Haussmann et al., 2013; Häsemeyer et al., 2009; Liu and Kubli, 2003; Ram and Wolfner, 2007; Rezával et al., 2012; Yang et al., 2009; Yapici et al., 2008). The continuous mating assay demonstrated here, is therefore, an addition to classic mating assays that can be used to study the acute effect of environmental conditions on reproductive behaviours.
Using the continuous assay for mating behaviour that is explained here, we previously showed that a pair of flies exposed to yeast will remate several times over a 24 h observation period (Billeter et al., 2012; Gorter et al., 2016; Krupp et al., 2008; 2013), while flies not exposed to food will only remate once (Gorter et al., 2016). This finding can be puzzling in the light of a large portion of the D. melanogaster literature indicating that females do not remate for several days after an initial mating (reviewed in references (Avila et al., 2011; Laturney and Billeter, 2014)). However, this discrepancy can easily be explained by assay conditions, where a female is isolated for one to several days before a new mating opportunity is provided. If the pair does not mate in this hour-long observation period, the female is characterized as not receptive. Moreover, the high mating frequency should not be surprising given that the data from wild-caught flies show that females contain sperm from 4
to 6 males in their storage organs; thus indicating that females naturally remate several times (Imhof et al., 1998; Ochando et al., 1996).
Here, we demonstrate the use of this continuous mating assay to unravel how flies gather and combine information about environmental conditions to modulate mating frequency. This assay allows one to test a relatively large number of mating couples for genetic studies and to test the influence of volatile and non-volatile environmental cues. The assay typically runs for 24 h, but can be extended to 48 h, allowing the testing of cycling environmental cues such as the light-dark (LD) cycle. We demonstrate this assay by testing the influence of volatile cues from a yeast culture within a pressurized air system in combination with the availability of non-volatile yeast nutrient in the food substrate.
The pressurized air system continuously pumps volatile cues into a mating arena that contains a food substrate and a test couple (whose mating behaviour is monitored). To further determine the specifics through which yeast influences mating, we test a major volatile
compound of yeast, namely acetic acid(Becher et al., 2012), in combination with an amino
acid content that corresponds to that of yeast in the food substrate, in the form of peptone (amino acids derived from enzymatic digestion of animal proteins). Together these experiments demonstrate how the effect of environmental cues on the mating behaviour of D. melanogaster can be tested with this assay.
Protocol
1. Environmentally controlled mating box
1.1) To ensure a controlled and easy to clean test area, setup a stainless-steel kitchen cabinet of 120 cm x 64 cm x 85 cm as illustrated in figure 1A.
1.1.1) Drill one hole at the back of the cabinet just below the ceiling and four sets of four holes into the sides, each with a diameter of 2 cm. Drill the first two sets of four holes, on each side of the box at a height of 7 cm from the bottom of the box and with 12.5 cm in between holes. Drill the other two sets on each side of the box at a height of 35 cm from the bottom.
Note: The four sets of four holes are used for camera power cables and air pump tubing to enter and exit the cabinet. The hole at the back is used for the power cables of a light board.
1.1.2) Build a light board with 18 rows of 40 alternating white and red light-emitting diodes (LED) with 2.5 cm space between each light. Mount the white and red LEDs in a circuit with power supply. Connect each LED in series with a resistor of 560 Ω, 0.25W and 5% tolerance.
Note: The emission wavelengths of the red light spanned from 590-661 nm with a sharp peak at 627 nm. The resulting light intensity in the experimental area is approximately 900 lux with both lights on and 90 lux with only the red lights on; this was measured using a smartphone light meter app.
27 Ch ap te r 2
1.1.3) Attach the light board to the top of the stainless-steel cabinet and pass the power cables through the hole at the back.
1.1.4) Connect the adapter of the white LEDs to a power control timer to allow switching off the white LED during the dark phase of the experiment. Connect the adapter of the red light, which flies are blind to (Montell, 2012), to a regular power supply to keep them on for the whole duration of the experiment.
1.1.5) Fix one 110-cm and two 54-cm long metal brackets, each with a width of 0.5 cm, to the inner sides of the box at a height 50 cm from the bottom of the box. Place a frosted glass diffusion plate (dimensions of 119.0 cm x 54.5 cm x 0.5 cm) on these brackets.
1.1.6) Add three layers of filter paper (120 cm x 50 cm) in between the light board and glass plate to diffuse the light and limit glare on the surface of the mating arenas (described in section 4). Pin two filter paper sheets together on their long edge on 120-cm long wooden rods using magnets (on the side), to stick them to the insides of the metal cabinet; perform this action 3 times.
1.1.7) Attach four fans at the sides of the box to create an air stream that continuously vents the mating box. Attach the first set of 8 cm fans between the light board and glass plate, with the inlet of air at the left side and the exhaust at the right side of the box, to minimize the building-up of heat generated by the light board. 1.1.8) Attach the second set of 12-cm fans 25 cm above the bottom of the cabinet to create an outward air flow that vents the inside of the cabinet and cools it to a stable 26 °C. Attach the fans at the exhaust side to a suction hose and lead the airstream out of the room to prevent re-cycling of the air in the cabinet.
1.2) Set up two stands (approximately 48 cm tall) mounted with two clamps, one at 28 cm and one at 30 cm from the base of the stand. Fix a webcam onto each of the clamps. Connect the 4 cameras to a computer running the monitoring software.
1.3) Place A4 sheets underneath each webcam. Use unprinted white sheets or sheets with pre-numbered grid of 7 by 5 squares with 4-cm axes to accommodate mating arenas (described in section 4).
Note: An HD webcam camera with 78-degree wide-angle view and 5-million-pixel resolution can cover an area of 21 cm x 30 cm corresponding to an A4 sheet and monitor between 20 and 35 mating arenas.
2. Fly rearing and collection
2.1) Place 20 male and 20 female wild-type Canton-S flies into fly rearing bottles containing 45 mL rich fly food medium (see section 3) for three to four days. Transfer the same adults three times by first tapping them down and then into fresh bottles.
2.1.1) Place the bottles in an incubator at 25 °C, and 12 h: 12 h light-dark cycle with lights on at 09:00 (Zeitgeber time (ZT) 0). A new generation will appear about 10 days later.
2.2) Anesthetize the resulting newly eclosed flies on carbon dioxide pads for no longer than 5 minutes and collect them into fly food vials using a paint brush.
2.2.1) Collect virgin (newly eclosed) females and virgin males from the wild-type Canton-S stock bottles into 2.5 cm x 9.5 cm fly rearing vials with 6.5 mL of rich fly food medium.
2.3) Age the flies in same sex groups of 20 flies each in fly rearing vials for 5 to 8 days at 25 C and 12 h: 12 h light-dark cycle and lights on at 09:00 (ZT 0).
2.4) Transfer the flies to fresh fly rearing vials on the day before the experiment. 3. Food medium preparation
3.1) Prepare 1 L of rich fly medium as follows.
3.1.1) Pour 1 L of tap water in a 2 L glass beaker with a magnetic stir bar and put the beaker on a magnetic hot plate. Keep the stirring off and turn the heating up to 300
C until boiling temperature is reached.
Note: During the long boiling time in the following steps a proportion of water will evaporate, but together with the added ingredients this protocol results in 1 L of rich fly medium when prepared at room temperature of approximately C.
3.1.2) Turn on the stirring to 500 rounds per minute (rpm) and add the following ingredients to the boiling water: 10 g agar, 30 g glucose, 15 g sucrose, 15 g cornmeal, 10 g wheat germ, 10 g soy flour, 30 g molasses, 35 g active dry yeast. Wait for the yeast to foam vigorously, then turn down the hot plate temperature to 120 C.
3.1.3) After 10 min turn the hot plate down to 30 C and let the mixture stir until cooled to 48 C. Monitor temperature by inserting a thermometer directly into the food.
3.1.4) Dissolve 2 g of p-hydroxy-benzoic acid methyl ester (tegosept, 100%) into 10 mL of 96% ethanol. Add this and 5 mL of 1 M propionic acid to the mixture. Stir for 3 min.
3.1.5) Pour the fly food medium into the arenas (described in section 4) to create a 0.3 cm thick layer at the bottom of the arena.
3.1.5.1) Use a 200 mL glass beaker for pouring. When exact quantities are important, use a 10 mL serological pipette.
3.2) Prepare fly medium minus yeast exactly as described in step 3.1.1 until 3.1.5, but leave out the yeast in step 3.1.2.
3.3) Prepare medium with agar with or without peptone by mixing 10 g agar and 35 g peptone in 1 L of boiling water and perform steps 3.1.4 to 3.1.5.
4. Mating arena preparation
4.1) Pierce a hole approximately 0.3 cm in diameter on the upper side of a 3.5 cm x 1.0 cm plastic Petri dish using a heated preparation needle (heated to redness in a Bunsen burner). Alternatively, use a soldering iron.
29 Ch ap te r 2
4.2) When preparing food medium with odorous compounds, first pipette 30 µL (1% of the final food medium, e.g. Acetic Acid glacial 100%) of desired compound into the dish for half of the experimental dishes. Leave the other half of the dishes empty for comparison.
Note: With the set-up described here, a maximum of 140 arenas can be tested at once. 4.3) Using a 5 mL serological pipette, pour 3 mL of food medium at the bottom of the dish on top of the desired compound. Cover it with a cheese cloth to prevent contamination, and leave the medium to solidify for approximately 1 h at room temperature.
4.4) Place a lid on the dishes and tape each shut at two sides. Prepare small paraffin film plugs to cover the holes of the dishes by rolling pieces of paraffin film into 0.2-cm thick rolls and then cut them into 0.5-cm segments.
5. Yeast culture for odour
5.1) Grow dry active yeast on yeast extract peptone dextrose (YPD) agar in a 14.0 cm x 2.06 cm Petri dish. Wear gloves to prevent contamination in this step.
5.1.1) Prepare YPD agar plates by adding 10 g yeast extract, 20 g peptone, 22 g glucose (0(+)-glucose monohydrate) and 15 g agar (pure) to 1 L of boiling ultrapure water. Layer the bottom of the Petri dish once everything is dissolved and store upside down in the refrigerator at 4 C, for up to 2 months.
5.1.2) Sprinkle a few grains of dried yeast on a YPD medium plate, let them dissolve. Then streak the medium plate using a sterile loop. Store the plate in a 30 C incubator overnight. Afterwards, store the culture in the refrigerator for no longer than 1 week.
5.2) Prepare YPD liquid medium in 1 L bottles by adding 10 g yeast extract, 20 g peptone, 22 g glucose (0(+)-glucose monohydrate) 1 L ultrapure water, stir until all solids are dissolved and devide the 1 L into two bottle of equal volume to serve as control and experimental bottle.
5.2.1) Autoclave for 25 min at 120 C and 1 bar pressure. Afterwards, store the bottles at 4 C for up to 2 months until use.
5.3) Fit open bottle caps (4.5 cm) with a 0.32 cm thick silicone septum.
5.3.1) Cut two small holes in the septum to snugly fit barbed bulkhead fittings. Attach the small polyvinyl Chloride (PVC) tubing (diameters: outer 0.8 cm and inner 0.5 cm) to both outlets that exit the bottle and to only one of the inlets entering the bottle. See figure 1B for illustration.
5.3.2) Wrap the fitted caps and tubing in aluminum foil and autoclave for 25 min at 120 C and 1 bar pressure.
5.4) Wear gloves and work next to a Bunsen burner flame to protect against contamination in this step. Dip a sterile 100 µL pipette tip into one of the yeast colonies from the YPD agar plate (described in 5.1) and drop it into the autoclaved YPD liquid medium bottle.
5.4.1) Cap this yeast-inoculated YPD liquid medium bottle as well as a YPD medium control bottle (yeast not added) with autoclaved caps mounted with in- and outlet (described in step 5.3). Put both bottles on separate magnetic plates and stir at
100 rpm at room temperature for 24 h before the start of the experiment to allow the yeast culture to grow.
5.4.2) Connect the inlets of both bottles to pressurized air to supply the yeast culture with air. Make sure to connect the outlet of the experimental yeast bottle to a tube venting the yeast smell out of the experimental room to prevent interference with the experiment.
6. Air pump set-up
6.1) Attach the large PVC tubing (diameters: outer 1.2 cm and inner 0.9 cm) to a pressurized air supply and lead it through two 1 L glass Erlenmeyer flasks filled with activated charcoal to the 800 mL line to purify the air. Use either pressurized air commonly supplied in labs as air supply, or connect the tubes to an air pump (pressurized air was used here).
Note: Tubing material should be selected based on the chemical properties of the volatile, and tested to prevent the volatile from sticking to the lining of the tubing (e.g. polytetrafluoroethylene, nylon or stainless steel).
6.2) Make two air splitters from 15 mL tubes and three 1000 µL pipette tips each.
6.2.1) Make three holes of ~ 1 cm diameter. First burn two holes, using a heated preparation needle (heated red in a Bunsen burner), adjacent to each other just below the lid of the 15 mL tube. Then make the third hole by removing the bottom of the tube.
6.2.2) Glue the 1000 µL pipette tips into the holes with the narrow end pointing outward. Cut the end of the pipette tips to allow greater air flow.
6.3) Connect the large PVC tubing from the outlet of the charcoal-filled Erlenmeyer flask to the pipette tip at the bottom of the 15 mL tube. Add small PVC tubing outlets to the two horizontal pipette tips and lead them towards a control and an experimental bottle.
6.4) To prevent contamination of the YPD medium with microorganisms, attach the small tubing to a sterile syringe filter (0.45-µm pore size) with a plastic push on bulkhead tubing connector going towards the filter, and a screw on plastic bulkhead connector leaving the filter. Then, attach the tubing to the inlet of the YPD culture bottle (see figure 1B).
6.5) To prevent airborne yeast from traveling from the culture flask into the experimental arena, attach a glass tube (6.5 cm long, outer diameter = 0.5 cm and inner diameter = 0.3 cm) to the outlet of each YPD culture bottle using small tubing. Attach PVC tubing to the other side of the glass tube and lead this towards the lower holes (drilled at each side) of the experiment box.
6.5.1) Fill the tube with glass fiber and autoclave it before use.
6.6) Add another 15 mL tube splitter (described in section 6.2) to the small PVC tubing, at each side of the experimental box, to get two tubes running into the box at both the experimental side (closest to exhaust of fan air stream, right) and the control side (closest to the inlet of the fan air stream, left).
6.7) Prepare 8x 10 mL serological pipettes each with 10 outlets to test 80 mating couples at the same time, i.e. 40 for each air condition.
31 Ch ap te r 2
6.7.1) Burn 10 holes of ~ 0.8 cm diameter, 2 cm apart into the pipette.
6.7.2) Cut the outer part of a 1 mL syringe into one small (2.5 cm) and one big (5 cm) outlet.
6.7.3) Glue these outlets into the holes with hot glue.
6.7.4) Wrap a small band of plastic paraffin film around the end of the outlets and attach a 1000 µL pipette tip. Diameter of the tip opening is 0.1 cm. Use clean tips for each experiment.
6.8) Attach two serological pipettes, using a T-splitter with outer diameter ≥ 0.5 cm and short pieces of small PVC tubing, to each of the two outlets at both sides. Tape the pipettes flat on the white paper sheet (under the cameras in the steel box).
6.9) Using an air flow monitor, set the air flow so that air velocity at the exit of the 1000 µL pipette tip is 0.5 m/s. This corresponds to an air flow of 0.0017 L/s per tip.
7. Monitoring of mating behaviour
7.1) Use a mouth pipette (as described in (Ejima and Griffith, 2011)) to place one experimental female into a small Petri dish (described in section 4) at 15:00 o’clock (ZT 6) and give her 1 h to acclimatize to the mating arena.
7.2) Set up the experimental box (described in section 1) as follows:
7.2.1) Turn on the lights, i.e. white light LEDs on a 12h :12 h light-dark cycle connected to a timer that switches on the light at 09:00 (ZT0) and continuous red light LEDs to allow monitoring of the flies during the dark phase of the experiment. Turn on the fans to limit heating up of the cabinet by the light source and to ensure excess odours are vented outside the test area.
7.2.2) Connect webcam cameras to a computer and start them with the monitoring software for picture monitoring.
7.2.3) For each camera, set the focus, brightness, and zoom in the monitoring software.
7.2.3.1) Right-click on the camera screen, open “camera properties,” and unclick “automatic focus.” Adjust the “focus” to clarify of the grid or any written words on the paper sheet. If necessary, change the “brightness” and “zoom.”
7.2.4) Set the program of the monitoring software to capture 1 picture every 2 min. Right-click on each camera screen and choose “edit camera” and then the “Actions” option. Click to start the actions “At regular intervals” and change the time to “2 minutes.” Choose “Take Photo” for select actions to perform and finally click on “ok.”
7.2.5) Right-click on each camera screen and select “start monitoring.”
7.3) After 1 h (at ZT 7), transfer a wild-type male to the Petri dish using the mouth pipette, place the dish on the A4 paper sheets under the webcam cameras, and click “start monitoring” for 24 h. For the air pump experiment, place the dish in such a way that a pipette outlet is connected to the entrance hole of the mating arena.
7.4) To analyse the mating behaviour of the couple, do the following.
7.4.1) Select and open all pictures in an image viewing software and page through them in chronological order.
7.4.2) Write down the date, experiment number, dish number, and start time into a spreadsheet within the same row. Take the start time of each arena from the moment that it is placed under the webcam camera. Record the time stamp from the pictures. 7.4.3) Mark the start time of each copulation in the same row into the spreadsheet. Count a mating as an incident when the male has mounted the female and the couple remains moderately stationary and in the same posture for at least five consecutive frames (10 min).
Note: This criterion is based on the reported length of copulation, ranging from 12 to 27 min in D. melanogaster, and the observation that copulations of 10 min and over are fertile (Bretman et al., 2009; Crickmore and Vosshall, 2013).
7.4.4) Count the number of copulations for each row in the spreadsheet to determine the mating frequency. Alternatively, subtract the start time of the experiment from the time of first mating for each row as a measure of mating latency, or subtract the time of the first mating from the time of the second mating as a measure of remating latency.
7.4.4.1) To calculate the remating latency, make sure to define the dates of the first and the second mating as consecutive days in the spreadsheet software.
7.4.5) Analyse the data with mixed effects models, assuming normal distribution of the data and including the date of the experiment as a random factor, using a statistical software (see the table of materials) to determine the statistical significance of the independent variables—food medium, air type, and interaction—
as previously described5.
7.4.6) Select the best explaining model by performing the backwards elimination of non-significant independent variables using log-likelihood ratio tests and the associated Akaike information. After running the model, visually inspect the data residuals to confirm normality. Confirm the homogeneity of variances using the Levene’s test. In the case of unequal homogeneity, square-root transform the data.
Representative results
Using this continuous assay, mating behaviour, and mating frequency in specific, can be determined under experimental environmental conditions. To control environmental conditions, we transformed a stainless-steel kitchen cabinet into a test area, with its own light source and diffusion, which ensures a high abundance of light and a minimum amount of glare from the top of the mating arenas (figure 1A). The inner test area is completely encased by stainless steel and glass, which allows for cleaning with organic solvents, such as hexane or ethanol. Additionally, the cabinet is equipped with holes that act as inlets for tubing,
33 Ch ap te r 2
bringing volatile cues from the pressurized air system (see figure 1A and 1B). The pressurized air system, adjusted for yeast odours, consists of an airflow guided through a liquid yeast culture before entering the test arenas through 4 pipette splitters with 10 outlets each (figure 1C). The whole system is airtight and fitted with several particle filters, both before and after entering the yeast culture, to minimize contamination with confounding odours (figure 1B).
Figure 1: Diagram of the experimental box and pressurized air system with yeast (A) Schematic illustration of
the environmentally controlled mating box described in section 1. Description of the annotated numbers and arrows: 1. light board with alternating white and red lights; 2. small fan; 3. 3 layers of filter paper, each layer consisting of two filter-paper sheets; 4. glass diffusion plate resting on brackets attached to 3 sides of the box; 5. big fan; 6. holes for tubing and cables; 7. experimental area; large arrow, 50 cm to the glass plate; middle arrow, 35 cm height for the cable holes; and small arrow, 7 cm height for the tubing holes. (B) Schematic illustration of the liquid yeast culture with airflow, as described in sections 5, 6.4, and 6.5. Description of the annotated numbers: 1. disposable filter unit; 2. cap with silicone septum and out- and inlets; 3. liquid medium; and 4. glass tube with glass fiber. (C) Schematic illustration of the air outlets as described in section 6.7. Description of the annotated numbers: 1. serological pipette; 2. tubing cut from 1-mL syringe, and 3. 1,000 µL pipette tip.
A
B
C
= air flow
= air flow
1
2
3
4
5
6
7
3
2
1
4
1
2 3
To demonstrate the use of this assay, we tested whether volatile cues from a yeast culture can influence mating behaviour. Air was bubbled through a liquid yeast culture for 24 h, and the air outlets were placed in the entrance of each mating arena (see figure 2A). Half of the mating arenas contained fly food with yeast (Food + yeast), and the other half contained fly food without yeast added (Food – yeast). A wild-type male and female were exposed to the odours coming from the external yeast culture, and their mating frequency was recorded. To determine which variables are necessary to explain the graphed results, we ran mixed-effects models, either including or excluding the independent variables of food medium, yeast air, and an interaction of the two. The data in figure 2B is best represented by a model including the independent variables of food medium (p=0.001) and yeast air (p=0.061), but there is no explaining interaction effect. Even though the yeast air variable is not significant in this full data set, it is necessary to explain the results. Analysis of yeast air separated for food medium shows that a mating couple does not respond to yeast odours when there is no yeast present in the food medium (air: p=0.992), but they do increase their mating frequency in yeast air when yeast is also added to the food medium (air: p=0.018). Together, these results demonstrate the applicability of the pressurized air system to test the influence of environmental odours in combination with food medium conditions.
Figure 2: Yeast odour increases female receptivity in the presence of yeast in the food substrate (A) Schematic
illustration of a mating arena with one male and one female and a pipette tip from the air outlet in Figure 1C entering through the entrance hole. (B) Graphical presentation of the response in mating frequency of a Canton-S mating couple to yeast odour with and without yeast in the fly food medium (Food – yeast: medium air n=12, yeast air n=13 and Food + yeast: medium air n=24, yeast air n=23). Line graph with SEM error bars and statistical output of mixed effects models with air as the independent variable and the date as a random variable for each food medium independently. The main statistical model includes food (p=0.001) and yeast air (p=0.061). Adapted from (Gorter et al., 2016).
