(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .
Failure to mate enhances investment in behaviors that may promote mating reward and impairs the ability to cope with stressors via a subpopulation of Neuropeptide F receptor neurons [1]
['Julia Ryvkin', 'The Mina', 'Everard Goodman Faculty Of Life Sciences', 'The Leslie', 'Susan Gonda Multidisciplinary Brain Research Center', 'The Nanotechnology Institute', 'Bar-Ilan University', 'Ramat Gan', 'Liora Omesi', 'Yong-Kyu Kim']
Date: 2024-02
Living in dynamic environments such as the social domain, where interaction with others determines the reproductive success of individuals, requires the ability to recognize opportunities to obtain natural rewards and cope with challenges that are associated with achieving them. As such, actions that promote survival and reproduction are reinforced by the brain reward system, whereas coping with the challenges associated with obtaining these rewards is mediated by stress-response pathways, the activation of which can impair health and shorten lifespan. While much research has been devoted to understanding mechanisms underlying the way by which natural rewards are processed by the reward system, less attention has been given to the consequences of failure to obtain a desirable reward. As a model system to study the impact of failure to obtain a natural reward, we used the well-established courtship suppression paradigm in Drosophila melanogaster as means to induce repeated failures to obtain sexual reward in male flies. We discovered that beyond the known reduction in courtship actions caused by interaction with non-receptive females, repeated failures to mate induce a stress response characterized by persistent motivation to obtain the sexual reward, reduced male-male social interaction, and enhanced aggression. This frustrative-like state caused by the conflict between high motivation to obtain sexual reward and the inability to fulfill their mating drive impairs the capacity of rejected males to tolerate stressors such as starvation and oxidative stress. We further show that sensitivity to starvation and enhanced social arousal is mediated by the disinhibition of a small population of neurons that express receptors for the fly homologue of neuropeptide Y. Our findings demonstrate for the first time the existence of social stress in flies and offers a framework to study mechanisms underlying the crosstalk between reward, stress, and reproduction in a simple nervous system that is highly amenable to genetic manipulation.
In this study we investigated the effects of failure to obtain reward on the behavioral actions and physiology of male flies. We exposed Drosophila males to repeated sexual encounters with non-receptive females that rejected their courtship efforts and tested the effect on their behavioral responses using a collection of behavioral paradigms. These responses encompass alterations in social behavior, increased aggression, heightened motivation to mate, and a reduced capacity to cope with stressors. We further show that the high motivational state and sensitivity to stress is mediated by the disinhibition of a small population of neurons that express receptors for the fly homologue of neuropeptide Y.
We previously showed that successful mating, and, more specifically, ejaculation, is rewarding for male fruit flies and reduces the motivation to consume drug reward [ 43 , 77 , 78 ]. While most research focuses on mechanisms that encode sexual reward [ 43 , 60 , 75 , 77 , 79 – 84 ], it remains unknown whether failure to obtain sexual reward is simply a lack of reward or is perceived as a stressor. Here we used the courtship suppression paradigm in Drosophila [ 45 , 85 – 87 ] to model the effects of failures to obtain sexual reward on different aspects of male behavior. We discovered that failure to mate induces a stress-like response characterized by a larger investment in actions that can enhance the odds of obtaining sexual reward while, at the same time, reduces the ability of rejected males to endure starvation and oxidative stressors via the disinhibition of a small subset of NPF-receptor neurons.
Similar responses to social stress and reward-seeking behaviors can be seen in a variety of animals, suggesting that the central systems facilitating survival and reproduction originated early in evolution and that similar ancient basic building blocks mediate these processes [ 41 , 42 ]. In agreement with this concept, we and others showed that Drosophila melanogaster can adjust its behavior and physiology to various social conditions [ 43 – 55 ] and that the brains of mammals and fruit flies share similar principles in encoding stress and reward [ 43 , 56 – 58 ]. For example, the fly homologue of the NPY signaling system (i.e., NPY and its receptor) functions in processing natural and drug rewards, decreases aggressive behaviors and suppresses responses to aversive stimuli such as harsh physical environments [ 43 , 59 – 69 ]. Moreover, similar to the essential role that NPY\Y1 signaling plays in various processes affecting health and lifespan in mammals [ 70 – 74 ], activation of neuropeptide F (NPF) neurons in male flies leads to decreased resistance to starvation and decreased lifespan [ 75 , 76 ].
There is considerable anatomical overlap between brain regions that process reward and stress stimuli, as well as opposing functionality: exposure to natural rewards buffers the effect of stressors, while stressors such as social defeat can alter sensitivity to reward and increase the rewarding value of certain addictive drugs [ 27 , 32 – 38 ]. An example of such opposing functions is seen in rodents where corticotropin-releasing factor (CRF) increases alcohol intake, and the binding of Neuropeptide Y (NPY) to NPY receptor Y1 on CRF-positive neurons within the bed nucleus of the stria terminalis (BNST) inhibits binge alcohol drinking [ 34 , 39 , 40 ].
It has been appreciated for quite some time that stress-response mechanisms allow animals to cope successfully with challenges encountered when living in social groups [ 8 , 11 – 14 ]. For instance, high levels of stress hormones promote attentive care during maternal behavior in humans [ 15 ], and the CRF-Receptor2 and its ligand Urocortin-3 are necessary for coping with social engagement in mice [ 16 – 19 ]. Moreover, high levels of CRF are correlated with increased motivation to obtain natural rewards, as documented in individual cichlid fish that ascend in the hierarchy of their social group [ 20 ]. While these are examples of the ways by which stress response mechanisms improve the ability of individuals to cope with social challenges, some types of social stress can induce social defeat, increase drug consumption [ 21 – 26 ], impair health and shorten lifespan [ 27 – 31 ].
Living in a social environment involves diverse types of interactions between members of the same species, the outcomes of which affect health, survival, and reproductive success [ 1 ]. Coping with the challenges and opportunities associated with this dynamic environment requires individuals to rapidly process multiple sensory inputs, integrate this information with their own internal state and respond appropriately to various social encounters [ 1 – 7 ]. Encounters that hold opportunities to secure resources, mating partners, and a higher social status are considered rewarding and are, therefore, reinforced by the brain reward systems, whereas the failure to obtain such rewards due to high competition, lack of competence, or aggressive encounters with competitors are perceived as stressors [ 8 – 10 ].
Lastly, activation of a small neuronal population consisting of 22–26 cells that co-express the neuropeptide Tachykinin (NPFR TK , Fig 5C and S1 Movie ), induced starvation sensitivity that was similar in its extent to the activation of the entire NPFR population ( Fig 5D ). The effects do not depend on the TK neuropeptide itself, as its knockdown did not affect starvation rates ( S8A and S8B Fig ), suggesting that activation by itself is responsible for the enhanced sensitivity. Remarkably, acute activation of these neurons mimics also the heightened arousal and some of the aggression-related features associated with the frustrative-like state. This included high ratios of male-male touch and chase events ( Fig 5E and 5F ), higher frequency of chase events ( Fig 5I ), higher persistence reflected by longer duration of touch and chase actions ( Fig 5G and 5H ) and as a result, reduced formation of social clusters ( S8C Fig ). In addition, the activation of NPFR TK neurons resulted in increased coordination between pairs of flies as seen by lower values of features that measure relative changes in angle and speed between two close individuals (absanglefrom1to2, absphidiff and absthetadiff; S8C Fig ), presumably suggesting that flies engage more persistently with others when interacting. In many cases the high arousal and persistence resulted in the formation of long chains containing 4–8 flies ( S8C Fig ). Given the overlap between some NPFR TK neurons and Dh44 or NPF neurons ( S8D and S8E Fig ), the activation of which does not promote starvation sensitivity or enhanced social arousal, we suspect the rest of NPFR TK neurons (12–16 cells which are negative to NPF and DH44), to regulate both sensitivity to starvation and the observed social responses ( Fig 5J and S8D–S8F Fig ). Taken together, these results suggest that the disinhibition of a small subset of NPFR neurons in rejected males promotes behavioral responses that may serve to increase the odds of obtaining a sexual reward, and at the same time impair the ability of male flies to resist starvation stress.
A. A portion of NPFR neurons co-express the neuropeptide DH44. Colocalization of DH44 using rabbit anti DH44 (magenta) in GFP expressing NPFR neurons. B. No effect for activation of Dh44 neurons on starvation resistance. p>0.05, n = 56 for all cohorts. C. Shared NPFR TK neurons as visualized using genetic interaction between NPFR and TK drivers: NPFRG4;+, TK-LexA;+, +;LexAop-FlpL, +;UAS<dsFRT>cs-Chrimson-mVenus flies. Green marks NPFR TK neurons, magenta marks nc-82. D. Activation of NPFR TK neurons enhances sensitivity to starvation (NPFR TK , red, n = 51) compared to genetic control flies (TK-LexA;LexAop-FLPL, black, n = 44; and NPFR G4;UAS<dsFRT>csChrimson-mVenus, gray, n = 51). *p<0.05, **p<0.01. E-I. Analysis of male-male social interaction during optogenetic activation of NPFR TK neurons compared to genetic controls using the FlyBowl system. Activation of NPFR TK neurons induces higher rates of touch behavior (E), chase events (F), longer duration of touch events (G), longer duration of chase events (H), and higher frequency of chase as reflected by linger inter-bout intervals between chase events (I). n = 13 for NPFR TK (red). n = 13 TK-LexA;LexAop-FLPL (black), and n = 12 NPFRG4;UAS<dsFRT>csChrimson-mVenus (gray). *p<0.05, **p<0.01, ***p<0.001. ANOVA or Kruskal-Wallis with post-hoc Tukey’s or Dunn’s test, and FDR correction for multiple comparisons were performed. J. Schematic representation of brain NPFR neurons that intersect with TK, Ilp2, DH44, and NPF neurons. K. Summary of main findings according to which deprivation of sexual reward induces a stress-like response and increases sensitivity to subsequent acute stressors. This is mediated by a neuronal response: Sexual deprivation decreases NPF signaling, thereby disinhibits NPFR neurons and induces a dynamin-independent activity, which increases sensitivity to starvation.
We next sought to determine which subset of the 100 NPFR neurons is responsible for the enhanced sensitivity to starvation. Given the possibility that sensitivity to starvation is mediated via neuropeptide release, we focused our attention to neuropeptide expressing neurons with shared expression with NPFR such as NPF [ 106 ], Dh44 [ 61 ], and Tachykinin [ 107 ]. Activation of NPFR NPF mutual cells (P1 and L1-l neurons as in Shao et al., [ 106 ]) did not affect either the sensitivity to starvation or male-male social interactions ( S6C and S6D Fig ), suggesting that NPF-expressing NPFR neurons, which presumably have the capacity for autoinhibition, are not responsible for these effects. Next, we tested Dh44 expressing neurons, as we previously documented similarity between Dh44 and NPFR transcriptional programs [ 61 ]. We first validated the overlap between the two populations and discovered that all six pars intercerebralis (PI) Dh44 neurons are NPFR neurons as well ( Fig 5A ). Activating Dh44 neurons did not affect sensitivity to starvation, nor did it lead to apparent effects on male-male behavioral responses (Figs 5B and S7A ). In addition, knocking down the expression of Dh44 in NPFR neurons did not change the sensitivity to starvation ( S7B Fig ). Altogether, these findings suggest that although Dh44 neurons are the functional homologue of mammalian CRF, Dh44 signaling does not mediate the sexual reward deprivation stress response in flies.
To assess whether the activation of NPFR neurons can also mediates other behavioral phenotypes observed in rejected males, we assayed the behavior of NPFR>csChrimson flies during optogenetic activation in FlyBowl arenas. Interestingly, we observed minor differences in the behavior of the experimental flies compared to genetic controls ( S6B Fig ), indicating that disinhibition of all NPFR neurons is sufficient to increase sensitivity to starvation, but not to induce changes in male-male social interactions.
The sensitivity of rejected males to acute stressors implies the existence of a link between the inability to obtain rewards and the stress response. Given the causal link between sexual rejection, NPF/R signaling and ethanol consumption in flies [ 43 , 63 ], and the role of the mammalian NPY in mediating the crosstalk between reward and stress by inhibiting downstream neurons [ 34 ], we postulated that rejection leads to disinhibition of NPFR neurons, which in turn sensitizes males to starvation. To test this hypothesis, we knocked down NPF-receptor in NPFR neurons ( S6A Fig ) and compared the sensitivity of naïve-single, rejected, and mated male flies to starvation stress. If sexual rejection reduces NPF signaling, which in turn disinhibits NPFR neurons, mated male flies should exhibit similar responses as rejected male flies. Our findings show that this manipulation abrogated the differences between mated and rejected cohorts ( Fig 4D ), implying that the release of NPF and binding to its receptor on NPF-receptor neurons (NPFR) is necessary for the resistance of mated males to starvation stress, and presumably that NPFR neurons are disinhibited in rejected males, resulting in enhanced sensitivity. To further test this conclusion, we next attempted to mimic rejection in naïve-single males using optogenetic activation of NPFR neurons (three, 5 minutes long activation/day). And indeed, this short activation protocol significantly increased their sensitivity to starvation compared to control flies ( Fig 4E ), further demonstrating that the disinhibition of NPFR in rejected males induces sensitivity to starvation stress. Interestingly, neurotransmitter release may not be required for starvation sensitivity triggered by the activation of NPFR neurons, as the same NPFR neurons activation protocol (three 5 minutes/day) and simultaneous inhibition of their synaptic transmission (using temperature sensitive Shibire), yielded similar levels of sensitivity to starvation as that of activation of NPFR neurons alone ( Fig 4F ). These results suggest that sexual-deprivation-induced sensitivity to starvation may involve neuropeptide based signaling and not neurotransmitter release that depend on functional dynamin.
The sensitivity of rejected males to starvation was not due to the increased activity of the rejected cohort, as there was no difference in the circadian activity and sleep patterns between rejected males and the other two cohorts ( S4B–S4D Fig ). It also implies that the enhanced activity in the FlyBowl experiments does not reflect an inherent increase in the activity of individual flies, but rather an emergent property of their response to the presence of other males. Additionally, no difference was documented in body triglycerides (TAGs), glucose levels, or body weight between the cohorts ( S5A–S5D Fig ), suggesting that rejected male flies do not suffer from an energetic deficit that increases their sensitivity to starvation. Although targeted metabolite analysis of head tissues using liquid chromatography–mass spectrometry (LC-MS) revealed a unique profile for each cohort ( S5E Fig ), we did not document any significant accumulation or depletion of metabolites such as oxidative agents or in antioxidant activity and glucose levels. Thus, sexual rejection decreases the ability of males to cope with additional stressors, but this is not simply explained by metabolic deficits.
A. Schematic representation of courtship conditioning: control males were introduced to either virgin, sexually receptive or sexually non-receptive females. As a result, males were either mated or rejected. The third cohort consisted of control single housed males that did not experience any social or sexual event (naïve-single). Encounters with females were repeated 3x a day for two days. B . Starvation resistance assay: rejected males (red, n = 91) compared to mated (blue, n = 102) and single housed (green, n = 110) males, ***p< 2E-16; mated vs single males, *p<0.05. C. Resistance to oxidative stress (20mM Paraquat): rejected males (red, n = 50) compared to mated (blue, n = 53) p<0.01** and naïve-single (green, n = 54) males ***p<0.001. Pairwise log-rank with FDR correction for multiple comparisons was performed in E, F. D. Starvation resistance assayed on NPFR>RNAi mated (blue, n = 68), rejected (red, n = 68) and naïve-single (green, n = 63) male flies. *p<0.05 rejected vs naïve-single, **p<0.01 mated vs naïve-single. Pairwise log-rank with FDR correction E. Activation of NPFR neurons promotes sensitivity to starvation. NPFR>csChrimson flies underwent three times a day (5 minutes each) optogenetic activation for two days (pink), NPFR>csChrimson w/o activation (grey) serve as controls. Starvation resistance of experimental (n = 55) and control flies (n = 72) was assayed. Log-rank test was performed, ***p< 0.001. F. Male flies expressing UAS-csChrimson and UAS-Shibire ts in NPFR neurons were subjected to three 5 min long optogenetic activations for three days, and their synaptic signaling was blocked at 28–29°C (light+ heat, orange, n = 52). Positive control males (light+cold, pink, n = 52), synaptic release block control (dark+heat, light gray, n = 52), negative control (dark+cold, dark gray, n = 50). Experimental and positive control flies showed no significant difference in resistance to starvation (p>0.05). Both experimental and positive control flies were significantly more sensitive to starvation than ‘dark+heat’, and ‘dark+cold’ flies (**p<0.01). Pairwise log-rank test with FDR correction for multiple comparisons was performed.
Next, we assessed whether there is a cost for this “frustration-like” state by testing the ability of rejected males to endure starvation and oxidative stress. Rejected, mated and naïve-single males were conditioned over the course of two days, and each cohort was subsequently exposed to starvation (1% agarose) or oxidative stress (20mM paraquat), and the rate of survival over time was documented ( Fig 4A ). If deprivation of sexual reward is a stressor, rejected male flies should be more sensitive to other stressors. Indeed, rejected males exhibited higher sensitivity to both starvation and oxidative stress compared to controls ( Fig 4B and 4C ). While the survival of mated and naïve-single cohorts declined to 50% after 22–24h of starvation, in the rejected male flies this was reached after less than 18h ( Fig 4B ). Exposure to paraquat led to a 50% decline in survival after more than 17h for both the mated and naïve-single cohorts vs. less than 14 hours for rejected males ( Fig 4C ). This demonstrates that sexual rejection promotes sensitivity to starvation and oxidative stress, further suggesting that multiple rejection events serve as a stressor, compromising the ability of male flies to cope with other stressors. Importantly, complete deprivation of both social and sexual interactions (i.e., naïve-single cohort) did not result in the same effect as active rejection, suggesting that the deprivation of an expected sexual reward, but not the lack of mating, increases sensitivity to additional stressors. The overall lifespan of rejected males was not affected by sexual deprivation ( S4A Fig ).
To further explore the effects of rejection on mating drive, we performed a detailed analysis of the action selection exhibited by males towards mated females during the first day of the training phase that was used to generate the rejected cohort (three one-hour training interactions spaced by one-hour resting intervals). As controls we used in each training session virgin males that interacted with virgin females ( S3A Fig ). The behavior of both cohorts during the first 10 min of each session was manually analyzed. In line with previous studies, the rejected cohort displayed marked courtship suppression, reflected by a reduction in the overall time spent courting in all 3 sessions ( S3B Fig ) [ 45 , 85 , 104 ]. Yet, the overall number of males exhibiting courtship action did not decline over the course of two days ( S3C Fig ), and interestingly, other aspects of their courtship, such as the overall number of licking actions and number of copulation attempts, were no less vigorous ( S3B Fig ). Notably, although the rejected males depicted longer latency to court mated females during the first encounter (consistent with an innate aversion to the male pheromone cVA [ 104 , 105 ]), they overcame this aversion in subsequent sessions and initiated courtship at the same time as males that courted virgin females ( S3B Fig ). Another feature that may reflect their surprising persistence is the duration of copulation attempts, which was 6 times longer compared to the controls ( Fig 3F ). This finding is suggestive of a conflict between high motivation to mate and repeated inability to fulfill this mating drive. Taken together, the behavioral responses of rejected males towards male and female flies imply that repeated failures to mate induce a “frustration-like” state where rejected males experience a conflict between their high motivation and inability to fulfill their mating drive.
A.-C. Aggression display (number of lunges) was compared between pairs of rejected and mated male flies (n = 16, statistical significance determined by T-test, *** p< 0.005 (A), and mixed pairs (n = 12) (B-C). The log 2 ratio between the number of lunges in rejected and mated flies was calculated for each pair, and then a one-sample T-test was performed to test whether the mean ratio was significantly different than 0, *** p<0.005. Data is presented as the mean ± SEM. D. Duration of copulation in rejected vs. naïve-single male flies. Statistical significance was determined by T-test, *** p < 0.001. Data is presented as mean ± SEM, n = 25. E. Relative transcript levels of candidate genes expressed in abdomens of rejected and naïve-single male flies were quantified by qRT-PCR, n = 6 independent experiments of 15–20 fly heads and abdomen. Statistical significance was determined by Student’s T-test with Bonferroni correction for multiple comparisons. ** p <0.01, *** p < 0.005. F. Virgin male flies were exposed to either mated or virgin females for three 1h sessions, and their behavior was recorded. At the end of each session, the females and males from the control cohort (Naive) were removed, and the males that experienced rejection were kept isolated in narrow glass vials for 1h. Mean bout duration of their copulation attempts was measured in the first (I), second (II), and third (III) sessions. Student’s t-test or Mann-Whitney was performed with FDR correction for multiple comparisons. *p<0.05, ***p<0.001. n = 41.
We compared the aggression levels of rejected and mated male flies, as both cohorts experienced interaction with females, unlike naïve-single flies, which were socially isolated, a condition known by itself to promote aggression. We discovered that when paired together, rejected male flies exhibited significantly higher displays of aggression in comparison to pairs of mated male flies ( Fig 3A ). In mixed pairs, the rejected male exhibited a far greater number of lunges compared to its mated opponent ( Fig 3B and 3C ), indicating that failure to mate does not induce a loser-like state. To extend this analysis to mating-related actions, we compared the mating duration of rejected males to that of control naïve/virgin male flies, as the investment in mating of the mated cohort is shaped by their previous mating events. When allowed to mate with virgin female flies, the copulation duration of rejected male flies was 25% longer (3.5 minutes longer) than the control males ( Fig 3D ). This was accompanied by an increase in the expression levels of genes associated with reproductive success compared to control males, including a two-fold increase in the transcript levels of Sex-Peptide (Acp70A) and Acp-63, both of which facilitate post-mating responses and fertility in female flies ( Fig 3E ) [ 103 ]. Taken together, these findings suggest that repeated events of sexual rejection promote a high motivational state rather than a loser-like state.
The similarity between some of the behavioral features exhibited by sexually rejected male flies to stress responses in other animals, including social avoidance, enhanced activity/arousal, and increased consumption of drug rewards [ 43 ], suggests that failure to mate induces a stress response in male flies. This prompted us to investigate whether failure to mate induces the loser-like state observed in socially defeated animals [ 16 , 91 ] or, rather, a high motivation to obtain a sexual reward, as described in several species upon the omission of an expected reward [ 92 – 102 ]. To discriminate between these two options, we analyzed the behavioral responses of rejected males in aggressive and mating encounters. If rejection promotes a loser-like state, rejected males are expected to exhibit low levels of aggression towards other male flies and reduced investment in mating. Conversely, if it promotes a high motivational state, rejected males should display enhanced aggression and higher investment in mating-related behaviors.
The behavioral signatures of the three cohorts across all 60 parameters (kinetic features, velocities, distances and angles between flies, scores for 8 behaviors, their frequencies as well as network features), are summarized in a scatter plot of normalized differences and are divided into 4 main categories: activity, interaction, coordination between individuals and social clustering, i.e., the aggregation of males to form clusters ( S2 Fig ). The existing similarities and differences between the three cohorts indicate that sexual rejection induces a discrete behavioral state that differs from successful mating and lack of mating and point to sexual rejection as the major contributor to reduced social interaction.
A. Illustration of network parameters. Density of networks represents how saturated they are compared to the maximum possible. Strength is proportional to vertex size (high in red individual). B-E. Social network analysis of groups composed of mated (blue) naïve-single (green) and rejected (red). Network density, and strength calculated by network weights according to duration (B-C) or number of interactions (D-E). Kruskal–Wallis test followed by Wilcoxon signed-rank test and FDR correction for multiple tests *p < 0.05, **p < 0.01, ***p < 0.001, Box plot with median IQR. F. Rejected male flies maintain large distances between flies along as measured by anglesub which is the Maximum total angle of animal’s field of view (fov) occluded by another animal, statistical significance was tested on averages across time. one-way ANOVA followed by Tukey’s and FDR correction for multiple tests *p < 0.05, **p < 0.01, ***p < 0.001. Error bars signify SEM. G. Average number of flies close to a focal fly. n = 15, 8, 15 for mated, naïve and rejected respectively. One-way ANOVA followed by Tukey’s and FDR correction for multiple tests *p < 0.05, **p < 0.01, ***p < 0.001. Error bars signify SEM.
To extend the analysis of their social behavior, we next investigated the network structures of the various fly cohorts, by analyzing features that describe the properties of their respective social networks and the individuals within them, including the density of the network and the strength of individuals within it ( Fig 2A ) [ 88 ]. Interaction events were defined using the following parameters: focal fly can see the other fly (requirement: at least 2 seconds in which the focal fly was at a distance of >8mm from the other fly and in which the visual field of view of the focal fly was occupied by the other fly (angle subtended>0); and network weights, i.e., the overall duration of interactions (emphasizing long-lasting interactions) or overall number of interactions (emphasizing short interactions) between each pair of flies ( Fig 2B–2E ). Analysis by duration indicated that the social networks of rejected males are less dense ( Fig 2B ) and that the average strength of individual flies in these networks is lower compared to mated and naive male flies ( Fig 2C ). Likewise, analyzing network features by number of interactions identified reduced density and low strength of the individuals than naive males ( Fig 2D and 2E ). These findings suggest that rejection promotes the formation of sparser groups containing individuals with reduced social interaction. The reduced network density of rejected male flies prompted us to compare the average distance between individuals of each group along the experiment, as indicated by the average field of view occluded by another fly (anglesub), a feature that increases as the distance between individuals decreases ( Fig 2F ). This analysis revealed that rejected male flies maintained significantly low values of anglesub, suggesting that rejected males maintain long distances between one another across the experiment. This finding is also supported by a reduced number of flies near a focal rejected fly compared to mated and naive male flies ( Fig 2G ). Given that rejected male flies exhibit similar activity levels as naive male flies, and presumably a higher probability to encounter other flies, their reduced network density and high inter-individual distance suggest that rejected individuals actively avoid social interactions with other flies, resulting in low-density groups.
Examining features that reflect activity levels, such as time spent walking, number of turns, time spent stopping, and average speed, revealed similar activity levels in rejected and naïve-single males, which were significantly higher than those in mated males ( Fig 1C–1F ). Analysis of specific social behaviors indicated an overall reduction in the degree of social interaction in rejected males than in naïve-single males ( Fig 1G–1J ). This included a lower number of approaches towards other males ( Fig 1G ), low levels of male-male chase behavior ( Fig 1H ), reduced formation of chains composed of multiple chasing males ( Fig 1I ), and fewer close touch encounters ( Fig 1J ). Taken together, the results indicate that rejected males are behaviorally distinct from both naïve and mated flies, marked by reduced levels of some behaviors associated with social interaction.
A. Schematic representation of sexual experience and behavioral analysis. Rejected males experienced 1h long interaction with non-receptive female flies, 3 times a day for 4 days. Mated males experienced 1h long interactions with a receptive virgin female, 3 times a day for 4 days. Male flies with no sexual experience were kept in social isolation (naïve-single). The social group interaction of the 3 cohorts was recorded for 30 min using the FlyBowl system. B. List of kinetic features, 8 behaviors and network parameters produced using the FlyBowl system. C-J Average percentage of time mated (blue) naive (green) and rejected (red) took to perform walk (C), turn (D), stop (E), change orientation (F), approach (G), chase (H), chain (I) and touch (J) behaviors. n = 15, 8, 15 for mated, naïve-single and rejected respectively. One-way ANOVA followed by Tukey’s and FDR correction for multiple tests *p < 0.05, **p < 0.01, ***p < 0.001. Error bars signify SEM.
To explore the way by which failure to obtain rewards affects behavior, and to ask whether failure to obtain reward is simply a lack of reward or is perceived as a challenge, we used the courtship suppression paradigm to induce repeated events of sexual rejection and assayed various aspects of male behavior. As a starting point, we employed the agnostic nature of the FlyBowl system to compare multiple behavioral features of male flies that experienced repeated failures to mate over the course of 4 days (“rejected”), to that of controls that experienced successful mating (“mated”), or lack of any sexual and social experience (“naïve-single”). Following the experience phase, 10 flies of each cohort were placed in circular arenas, and their behavior was recorded for 30 min and analyzed using the FlyBowl suite of tracking and behavior analysis softwares [ 88 – 90 ] ( Fig 1A ). The tracking data obtained was used to calculate various kinetic features, including velocities, distances and angles between flies and their relative differences across time, and to train 8 types of behavior classifiers using the Janelia Automatic Animal Behavior Annotator (JAABA) (Figs 1B and S1 , 88 ].
Discussion
In this study, we discovered that the NPF/NPFR system serves as a junction that integrates the crosstalk between reward and stress, and showed for the first time that Drosophila males perceive failures to obtain sexual reward as social stress. We used a collection of behavioral paradigms to explore responses altered by sexual interaction and discovered that repeated events of failure to mate lead to complex behavioral and physiological responses, encoded in part by the disinhibition of a subset of NPFR neurons. This includes avoiding interaction with other male flies and, at the same time, competing over mating partners via increased aggression and prolonged copulation (known as mate guarding); the latter is strengthened by the increased production of certain seminal fluid proteins that facilitate stronger post-mating responses in female flies. The regulation of ejaculate composition and mating duration were previously described in Drosophila males in response to perceived competition with rival male flies as a strategy to transfer a higher amount of accessory gland proteins (Acps) to intensify the females’ post-mating responses [47,108,109]. Although rejected males were not exposed directly to other male flies, the observed extension of copulation events and increased expression of sex-peptide suggest that failure to mate induces responses like those following the perception of competing male flies, presumably by evaluating the quality of their sexual interaction with female flies. It remains to be tested whether the enhanced investment in mating-related elements promotes mating success.
While most studies focus on mechanisms that encode reward on a scale of zero to one, our findings demonstrate the existence of negative values, where failure to obtain rewards is different than its lack, as shown by the clear differences between rejected and virgin males who never experienced mating or rejection. This provides a conceptual framework for investigating mechanisms that regulate deprivation or omission of an expected reward which is located at the negative part of the scale, when organisms fail to obtain a reward they expect to receive despite signals for its presentation [92]. This condition is known to induce a frustrative-like state, characterized by increased motivation to obtain the reward [93,95,99], increased levels of arousal [99,100], agitation [94], grooming [92,100], stress-associated behaviors [93,94], drug consumption [95,101], locomotion [94], and aggression [95]. The use of courtship suppression to deprive male flies of the inherent expectation of a sexual reward provides a possible model for frustration-like stress responses in Drosophila. While courtship suppression is associated with reduced courtship and presumably a defeat-like state, we discovered that rejected males are rather persistent in their attempts to obtain a sexual reward. This finding is not completely surprising considering the innate nature of mating motivation and the presence of female aphrodisiac pheromones. It can also be attributed to measuring various courtship actions rather than to the overall percentage of time spent courting, which is the usual indicator for the quantification of courtship. In response to repeated failures to mate, rejected males exhibit features characteristic of a frustration-like state, such as persistent mating actions (some of which are elongated), increased arousal, increased aggression, as well as longer mating duration upon successful mating encounters, all of which are reminiscent of a high motivational state.
The presumed discrepancy in the experience of rejected males where their innate expectation to obtain mating reward when perceiving female associated cues is unmet, resembles a well-studied neuronal mechanism known as error predication. This associative learning-based mechanism was largely studied in mammals and serves to encode a discrepancy between expected and actual rewards by sub-second-fast dopaminergic signals [110–112]. A recent model suggests that the Drosophila dopaminergic system harbors the capacity to encode for error prediction in a similar fashion to its mammalian counterparts [113]. Although the sexual rejection paradigm used in this study does not completely fit the error prediction hypothesis, as the tested males did not learn to associate females with mating reward, it may still be a useful paradigm to study error prediction in flies. This idea is strengthened by the functional interaction between NPF expressing neurons and subsets of dopaminergic neurons that express receptors for NPF [60,62] and the established role of dopaminergic neurons in encoding the negative valence of sexual rejection [60,87]. An interesting avenue for future research will be to assess the behavioral and physiological trajectories of male flies that experienced successful mating before undergoing sexual rejection.
While this motivational state may assist males in coping with social challenges associated with failure to mate, it is perceived as a stressful experience manifested by temporal costs in the form of sensitivity to acute stressors. This is the first example in flies that not obtaining a sexual reward is not simply a case of reward deprivation but also a stressful experience. But what mechanism governs the behavioral and physiological responses to deprivation of sexual reward? In search of mechanisms that could explain the behavioral and physiological responses to the failure of obtaining a sexual reward, we examined three possible directions: (1) sleep and activity (2) energetic costs or modulation of metabolic pathways, and (3) disinhibition of NPF target neurons.
Regarding the possibility that high motivational state is associated with high energetic costs, and although Gendron et al. [75] showed that exposure to female pheromones lowers body triglyceride levels in males, we did not detect metabolic changes that could explain sensitivity to starvation or oxidative stress. Still, our findings that some metabolites, such as 5-Aminolevulinic acid, are enriched in rejected flies may indicate a reduction in Heme synthesis and, consequently, an elevation in protoporphyrin, as well as a possible reduction in heme oxygenase (HO). While there is some evidence that protoporphyrin can act as an antioxidant [114,115], it mainly functions as a pro-oxidative agent [116,117]. HO is a rate-limiting enzyme that degrades heme into biliverdin, carbon monoxide (CO), and iron [118]. In Drosophila, the ho gene is expressed in different brain tissues, including in the optic lobe, central brain, and glial cells, and plays an important role in cell survival and protection against paraquat-induced oxidative stress [119]. This is consistent with rejected males experiencing increased sensitivity to oxidative stress, reflected in their heightened sensitivity to paraquat. However, this does not explain why rejected males were also more sensitive to starvation. Five of the metabolites were specifically enriched or depleted in rejected compared to both mated and naïve-single cohorts: glycine, tryptophan, 5-aminolevulinic acid (5ALA), acetyl-glutamine, and stearic acid (S5E Fig and S1 Table). Though tryptophan can be converted to serotonin or tryptamine, and most of the tryptophan metabolizes to kynurenine pathway (KP) metabolites, which contribute to a shorter lifespan [120–123], we did not document any significant accumulation or depletion of metabolites in the KP, or a difference in the abundance of serotonin (S5E Fig). Similarly, no difference was observed in oxidative agents or in antioxidant activity and glucose levels, though trehalose (the main circulating sugar in flies) levels declined in naïve-single males compared to mated and rejected males, indicating a possible change in insulin signaling [124–127].
After eliminating the contribution of sleep or metabolic deficit to the inability of rejected males to cope with stress, we demonstrated that their sensitivity to starvation stress is caused by the disinhibition of NPF target neurons, mimicked in our experiments by optogenetic activation of NPFR neurons as well as knock-down (KD) of npfr. We mapped the relevant neurons to a subset of 22–26 NPFRTK neurons, the activation of which is sufficient to induce sensitivity to starvation and some of the behavioral phenotypes observed in rejected males. The identified NPFR sub-population can be divided into smaller known subpopulations: six of these neurons express the CRF-like DH44, and two pairs of NPFRTK neurons, L1-l and P1, colocalize with NPF-expressing neurons. While CRF neurons in mammals mediate reward-seeking behaviors and responses to social stress [13,14,128–134] and the fly Dh44 neurons are known to facilitate aggressive behaviors [135], we did not find evidence to support their role in regulating responses to social stress. We concluded that the activation of a subset of 12–16 neurons co-expressing the Tachykinin neuropeptide governs both starvation sensitivity and enhanced arousal and some features of aggressive behaviors. It should be noted that the effects of optogenetic activation of NPFRTK neurons was tested using the FlyBowl setup, a context that does not allow for the expression of lunging behavior (due to low ceiling) but do support the expression of other forms of aggression such as chases and wing threats. Therefore, the activity of the NPF\NPFR circuit alone is sufficient to facilitate a stress-like response caused by the deprivation of a reward and increase subsequent sensitivity to acute stressors.
While the subset of NPFR neurons that mediate the behavioral and physiological response shares expression with Tachykinin [61], the TK neuropeptide itself does not play a role in modulating the sensitivity to starvation upon sexual rejection. In addition, while Wohl et al., [136] found that TK supports cholinergic synaptic transmission, our results suggest that the increased sensitivity to starvation is mediated by a dynamin-independent signaling mechanism, possibly bulk endocytosis for vesicle retrieval or neuropeptide release, which commonly does not involve synaptic vesicle reuptake. Further dissection of the NPFR and NPFRTK neuronal subpopulations is needed to better understand the role of each cell type in response to sexual reward deprivation.
This is the first documentation of a frustration-like stress response in Drosophila, which is caused by the innate expectation of a natural reward that is not achieved. Our findings offer a paradigm shift in the perception that rejected males refrain from obtaining a sexual reward by showing that their motivation to copulate increases, not declines, and thus can serve as a conceptual framework to study the interplay between reward-seeking behaviors such as ethanol consumption, stress, and addiction. The bi-directional role of the NPF system in mediating reward and suppressing response to stressors [59] and the remarkable similarity to mammalian NPY [137] suggest that it is possible to simplify the complexity of studying the crosstalk between reward, stress, and reproduction in mammals by stripping it down to its most fundamental building blocks in fruit flies.
[END]
---
[1] Url:
https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1011054
Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.
via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/
