In most insects, the mate finding process is mediated by volatile sex pheromones, chemical signals that are released by one sex to attract the other (Wyatt, 2014). To guide the signal recipient effectively and reliably to the sender, sex pheromones need to be species-specific. Being species-specific, sex pheromones provide long-range precopulatory isolation even in sympatric species with overlapping mate recognition signals (Smadja and Butlin, 2009).

An excellent model system to study such processes is the parasitic wasp genus Nasonia (Ashmead, 1904). During the past decades, the genus has been frequently used to answer questions concerning all aspects of parasitic wasp biology (Werren and Loehlin, 2009; Werren et al., 2010; Lynch, 2015; Mair and Ruther, 2019; Wang et al., 2020). It consists of four species: N. vitripennis (Nv), N. giraulti (Ng), N. oneida (No) and N. longicornis (Nl; Darling and Werren, 1990; Raychoudhury et al., 2010; Werren et al., 2010). All species parasitize the pupae of various cyclorraphous fly species. The typical habitat of Nasonia wasps is in the nests of cavity-nesting birds. Nv is cosmopolitan and sympatric with Ng and No in eastern North America and with Nl in the west. In areas of sympatry, Nasonia species occur often microsympatrically, i.e., heterospecific offspring develop within the same host individuals (Darling and Werren, 1990; Grillenberger et al., 2009; Raychoudhury et al., 2010). Hence, interspecific interactions between males and females of sympatric species are likely to occur. Interspecific mating in Nasonia occurs in the lab with varying frequency but does not lead to hybrid offspring in most combinations of species due to Wolbachia-mediated cytoplasmic incompatibility (Breeuwer and Werren, 1990; Bordenstein et al., 2001; Giesbers et al., 2013). Hence, fertilized eggs of females that mated with a heterospecific male die or develop into male offspring due to the haplodiploid sex determination in Hymenoptera (Tram et al., 2006). The resulting all-son-offspring impose evolutionary costs on the respective females because the mating system of Nasonia is characterized by local mate competition and female-biased offspring sex ratios are common (Werren, 1983; Grillenberger et al., 2009; Ivens et al., 2009). Given that Nasonia females typically mate only once in their life, they depend on reliable information to reduce the probability of interspecific mating.

Males of all Nasonia species produce sex pheromones in their rectal vesicle (Abdel-Latief et al., 2008). Three pheromone components have been identified in Nasonia males: the two diastereomers (4R,5S)- and (4R,5R)-5-hydroxy-4-decanolide (RS and RR; Ruther et al., 2007) and 4-methylquinazoline (MQ) as a minor component (Ruther et al., 2008; Figure 1A). RS and MQ are produced by males of all Nasonia species as well as by males of the closely related species Trichomalopsis sarcophagae, whereas RR is specific for Nv. This suggests that the ancestral sex pheromone of Nasonia males consisted solely of RS and MQ and that the additional occurrence of RR is a derived state that evolved in the N. vitripennis lineage (Niehuis et al., 2013). RR has been shown to be synthesized from RS by short-chain dehydrogenases/reductases (SDRs; Niehuis et al., 2013; Ruther et al., 2016). The chemical identities of the pheromones have been established in all Nasonia species, but detailed analyses of the biological function are so far available only for the model organism Nv and, to a lesser extent, for Ng (Ruther et al., 2014; Mair and Ruther, 2019). Nv males eclose without pheromone, but pheromone titers increase within the first 2 days after eclosion and remain constant on day three (Ruther et al., 2007). Males deposit pheromone by dragging their abdominal tips over substrates, a behavior referred to as ‘abdomen dipping’ (Barrass, 1969; van den Assem et al., 1980). Nv males economize pheromone deposition by engaging in abdomen dipping more frequently after contact or mating with virgin females (Steiner and Ruther, 2009), i.e., if the chance of mating with (additional) females is high. The mating system of Nv is characterized by territorial males that monopolize parasitized hosts containing females about to emerge. Apart from showing aggressiveness towards male competitors (Leonard and Boake, 2006), dominant males often mark monopolized hosts to prevent emerging females from dispersing unmated (Mair and Ruther, 2018). Nv males are able to identify intact hosts (without exit holes) containing females about to emerge (Prazapati et al., 2022). It is unknown, however, whether the presence of those hosts also affects male marking.

Pheromone markings, abdomen extracts and synthetic blends of the ternary Nv pheromone (RS + MQ + RR) and the binary Ng pheromone (RS + MQ) all attract conspecific females (Niehuis et al., 2013). The exclusive occurrence of RR in Nv strongly suggests that this compound has a function in precopulatory isolation and reduces the risk of interspecific sexual interactions in sympatric Nasonia species both in eastern and western North America. The role of individual pheromone components, however, has so far only been studied in the eastern species pair Nv/Ng. In the initial study addressing this question (Niehuis et al., 2013), virgin females of both species were attracted to RS when presented alone (Figures 1B,C). The attractiveness of RS was synergized by the minor component MQ in both Nv and Ng, whereas MQ had no effect when presented alone. Likewise, RR was neither attractive for Nv nor for Ng virgin females when presented alone, but the addition of RR to the binary Ng blend (RS + MQ) synergized the attractiveness of this blend for Nv whereas this modification had no effect on Ng females. Hence, females of both species responded to the pheromone of heterospecific males, but Nv females preferred the ternary conspecific pheromone blend over the binary heterospecific one (Niehuis et al., 2013). The preference for the heterospecific pheromone blend in either species raised the question whether this is associated with evolutionary costs. In the case of Ng, this is unlikely, because Ng differs from its Nasonia congeners by showing a very high within-host-mating (WHM) rate, i.e., mating in Ng occurs already inside the host prior to female emergence (Drapeau and Werren, 1999). Given that mating makes Nv and Ng females unresponsive to the male sex pheromone (Ruther et al., 2010, 2014; Ruther and Hammerl, 2014) it is unlikely that the response to the ternary Nv pheromone blend has fitness consequences for them. In fact, females collected at the moment of emergence from the host were not even attracted to the conspecific pheromone (Ruther et al., 2014; Figure 1C). As for Nv, mating occurs almost exclusively outside the host within a few minutes after emergence of the females (Mair and Ruther, 2018) and thus, the attraction of females to the binary Ng blend might be problematic. However, a second study addressing the Nv/Ng pheromone interaction in more detail revealed that the attraction of virgin Nv females to the binary Ng blend is age-dependent (Ruther et al., 2014). Newly eclosed females and those collected at the moment of emergence from the host did not respond to the binary Ng blend but strongly preferred the conspecific ternary blend (Figure 1B). Only with increasing age (≥1d), when most females would have been mated and dispersed under natural conditions anyway (Mair and Ruther, 2018), do Nv females became less choosy and responsive to the heterospecific binary pheromone blend (Ruther et al., 2014). Nothing is known about the pheromone response and the role of individual pheromone components in Nl females. This species shows intermediate WHM rates (Drapeau and Werren, 1999; Trienens et al., 2021) and thus many females emerge as virgins and mate outside the host. Therefore, Nl females should discriminate more strongly against the ternary Nv pheromone blend than Ng females to avoid costly sexual interactions with Nv males in regions of sympatry.

The abdominal sex pheromone of Nv males is not only attractive for virgin females but also mediates site fidelity of the pheromone-marking males using their own pheromone markings (van den Assem et al., 1980). This behavior could also be interpreted as an economization of pheromone use because males have a way to stay near and return to their own pheromone markings rather than needing to deposit a new one. The minor component MQ has been shown to mediate this phenomenon in Nv (Ruther et al., 2011), whereas nothing is known about the male response to the pheromones in the other Nasonia species.

In the present study, we fill a knowledge gap about pheromone communication in the genus Nasonia by investigating chemical and behavioral aspects of the male Nl sex pheromone. We study the age dependence of pheromone titers in males, examine the context of pheromone deposition and elucidate the role of the different pheromone components in mediating the attractiveness of the pheromone for both females and males. We hypothesize that the sex pheromone of Nl males attracts virgin females and mediates site fidelity in males whereas the Nv-specific pheromone component RR contributes to precopulatory isolation between the sympatric species Nl and Nv.

Male pheromone titers were almost zero in newly eclosed males, increased significantly within the first 2 days after eclosion and remained constant on day three (median RS/MQ 0d: 0/0 ng; 1d: 14/0.1 ng; 2d: 302/0.4 ng; 3d: 318/0.4 ng; Figures 2A,B). In the behavioral bioassays, virgin Nl females were attracted to filter paper disks treated with purified abdomen extract or the binary conspecific blend consisting of synthetic RS + MQ (Figures 3A,D). However, they neither preferred RS nor MQ over the solvent controls when these chemicals were offered alone (Figures 3B,C). Virgin females preferred purified extract of conspecific males over extract of Nv males. They did not prefer the ternary heterospecific pheromone blend consisting of synthetic RS + MQ + RR when tested against the solvent control. When given the choice between the ternary hetero- and the binary conspecific pheromone blends, Nl females preferred the latter (Figures 3E–G). Mated Nl females were no longer attracted to the male sex pheromone (Figure 4A). However, females collected at the moment of emergence from the host were attracted to the conspecific binary pheromone blend indicating that most of them had emerged unmated (Figure 4B). This was confirmed by the analysis of the offspring’s sex, because following olfactometer bioassays most (84.4%) offspring-producing females had only sons, indicative of their virgin mating status.

Nl males were attracted to their own pheromone in the olfactometer tests. They preferred both the natural pheromone and the binary blend consisting of synthetic RS + MQ over the solvent control. When given the choice between RS or MQ alone and the solvent control, males preferred MQ but not RS indicating that MQ underlies the preference of males for their own pheromone (Figures 5A–D).

Naïve, 2-d-old males exhibited the typical abdomen dipping behavior that has been shown to serve pheromone deposition in Nv males (Steiner and Ruther, 2009). Both contact with a female and successful mating equally increased abdomen dipping frequency in Nl males (Figure 6A). In contrast, the presence of a host containing females about to emerge did not increase male abdomen dipping frequency (Figure 7A) and males showed this marking behavior almost exclusively in the vicinity of a host but hardly on a host itself (Figure 7B). Quantification of deposited pheromone after abdomen dipping revealed that 2-day-old, mated males deposited significantly more RS than naïve ones of the same age, while males with only female contact released intermediate amounts (median RS unmated: 1.5 ng; contact: 12 ng; mated: 176 ng; Figure 6B). As for the minor component MQ, deposited amounts showed a similar pattern (Figure 6C), however, statistical analysis did not reveal any significant differences between treatments (median MQ unmated: 0 ng; contact: 0.01 ng; mated: 0.2 ng).

The present study demonstrates that Nl uses the male abdominal sex pheromone in a similar way as previously described in the congeneric species Nv and Ng. Furthermore, it shows that the Nv-specific pheromone component RR mediates precopulatory isolation not only in the eastern North American species pair Nv/Ng but also in the western pair Nv/Nl. Like in Nv and Ng (Ruther et al., 2007, 2010, 2014), exclusively virgin Nl females were attracted to the conspecific male pheromone. The vast majority of Nl females emerged from the host as virgins showing that mating occurs predominantly outside the host and that the male pheromone is relevant for finding an adequate mating partner. In contrast to Nv and Ng (Niehuis et al., 2013), RS is not attractive for Nl females when presented alone. Rather, the presence of synergistic MQ is needed to attract Nl females. The addition of the Nv-specific pheromone component RR to the conspecific binary blend RS + MQ made Nl females unresponsive, and they preferred both natural and synthetic conspecific binary blends over the respective ternary heterospecific blends when exposed to them simultaneously. This demonstrates that Nl females discriminate more strongly against the Nv-specific pheromone component RR than previously reported for Ng (Niehuis et al., 2013). This might be explained by the lower WHM rate in Nl (Drapeau and Werren, 1999; Trienens et al., 2021) that makes encounters between virgin Nl females and Nv males in shared habitats more likely than in the eastern species pair Nv/Ng. The high WHM rate in Ng has been shown to be due to the males refraining from gnawing exit holes through which virgin females might escape. A recent study, however, has demonstrated that Ng shows much lower WHM rates when developing in microsympatry with Nv, because virgin Ng females emerge through exit holes gnawed by Nv males (Trienens et al., 2021). This suggests that, under conditions of microsympatry, Ng females are prone to get involved in costly sexual interactions with Nv males if they do not discriminate against the ternary pheromone blend produced by these sympatric heterospecific males. Therefore, it remains to be investigated whether Ng females having developed in microsympatry with Nv discriminate more strongly against the Nv-specific pheromone component RR than those having developed in the absence of Nv. The strong discrimination of Nl females against the Nv-specific pheromone component RR comes along with a low discriminatory ability shown by these females during courtship. In a previous study, Nl females accepted more than 80% Nv males as mates while vice versa Nv females accepted less than 30% of Nl males (Giesbers et al., 2013). This suggests that Nl females rely mainly on precopulatory isolation based on the male abdominal sex pheromone to avoid costly interspecific mating with Nv males, while Nv females use both the species-specific composition of the male pheromones (Niehuis et al., 2013; Ruther et al., 2014) as well as courtship-related cues (Giesbers et al., 2013) to decrease the risk of mismating with sympatric congeneric males.

Nl males economize both pheromone production and deposition. Like in Nv and Ng (Ruther et al., 2007, 2014), Nl males eclose without any pheromone in the rectal vesicle, and pheromone production starts only after eclosion. In Nv, newly eclosed males typically remain inside the host for approximately 1.5 days before emerging from the host and most males eclose and emerge before the females (protandry; Moynihan and Shuker, 2011). Assuming a similar scenario in Nl, emerging males should have produced sufficient pheromone to attract and arrest virgin females. Nl males show abdomen dipping increasingly if mating chances are high, i.e., after contact or successful mating with a female. Once an exit hole has been gnawed by protandrous Nasonia males, females may emerge from the host in rapid succession (Mair and Ruther, 2018). Hence, the presence of one female increases the probability of meeting others and pheromone deposition is likely to enable additional mating opportunities. This is supported by our TD analyses of deposited pheromone amounts: Mated Nl males deposited ca. two-thirds of their available pheromone which is probably a significant investment. The presence of hosts (without showing exit holes) containing females about to emerge is obviously not sufficient to trigger this investment. Pheromone markings remain attractive for virgin Nasonia females for a maximum of 2 h (Steiner and Ruther, 2009). Hence, males likely refrain from depositing higher pheromone amounts in the presence of female containing hosts, because they are unable to reliably estimate the time of female emergence when exit holes are missing. Dominant Nv males pheromone-marked monopolized hosts to prevent females from dispersing unmated. In these experiments, however, exit holes were already present (Mair and Ruther, 2018). Site fidelity at the sites of pheromone deposition is another strategy of economic pheromone use in Nasonia. As previously shown for Nv (Ruther et al., 2011), Nl males use the trace component MQ but not the major component RS to recognize the pheromone markings at which virgin females are likely to appear. In this context, 0.5 ng MQ was sufficient to mediate site fidelity in Nl males and to enhance responses of virgin Nl females. This falls into place with much lower MQ titers determined in Nl males (ca. 0.5 ng per male) when compared to values reported previously for Nv (ca. 3 ng per male; Ruther et al., 2008).

After this study, No remains the only Nasonia species to be investigated in detail with respect to responses of females and males to male sex pheromone. Given that No is sympatric with both Nv and Ng in eastern North America, No females likely encounter both the binary pheromone blend of Ng males and the ternary pheromone blend of Nv males, with the former blend being identical to that of conspecific No males. However, No has been reported to possess much lower titers of the binary pheromone blend (RS + MQ) than Ng (Diao et al., 2016) suggesting that No no longer uses the male abdominal sex pheromone in the context of mate finding. Further studies are needed to fully understand the role of male sex pheromones in precopulatory isolation of the three sympatric Nasonia species in eastern North America.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

JR designed the study, supervised AK, and wrote the first draft of the manuscript. AK performed the experiments and analyzed the data. JH synthesized and provided a sample of (4R,5S)-5-hydroxy-4-decanolide. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.