Socially-induced variation in physiological mediators ...

Hormones and Behavior 93 (2017) 39–46

Contents lists available at ScienceDirect

Hormones and Behavior journal homepage: www.elsevier.com/locate/yhbeh

Socially-induced variation in physiological mediators of parental care in a colonial bird Michaël Beaulieu a,⁎, André Ancel b,c, Olivier Chastel d, François Criscuolo b,c, Thierry Raclot b,c a

Zoological Institute & Museum, University of Greifswald, Johann-Sebastian-Bach-Str. 11/12, 17489 Greifswald, Germany Université de Strasbourg, CNRS, IPHC UMR 7178, F-67000 Strasbourg, France CNRS, UMR7178, 67087 Strasbourg, France d Centre d'Etudes Biologiques de Chizé (CEBC), UMR7372-CNRS/Univ. La Rochelle, F-79360, France b c

a r t i c l e

i n f o

Article history: Received 24 November 2016 Revised 11 March 2017 Accepted 19 March 2017 Available online xxxx Keywords: Antioxidants Ecophysiology Hormones Parental care Social environment Sociophysiology

a b s t r a c t Social facilitation of reproduction occurs in humans and animals, and may represent one of the bases of reproduction in groups. However, its underlying physiological mechanisms remain largely unexplored. Here, we found in a colonial bird, the Adélie penguin (Pygoscelis adeliae), that the number of parental interactions (nest relief ceremonies) performed by breeding individuals on the colony was positively related to prolactin levels in other breeding individuals exposed to these interactions (i.e. focal individuals). As prolactin is typically involved in the expression of parental behaviour in birds, this suggests that parental interactions by conspecifics represent social cues that might increase parental motivation in focal individuals. Moreover, parental interactions were not related to corticosterone levels in focal individuals, suggesting that these social cues were not stressful for penguins. However, social stimulation still had a cost for focal individuals, as it was negatively related to their antioxidant defences (a component of self-maintenance). As social stimulation was also positively related to prolactin levels, this highlights the fact that social stimulation acts on the trade-off between reproduction and selfmaintenance. For the first time, the results of the current study shed light on the physiological factors potentially underlying social facilitation of parental care. Importantly, they suggest that, even though social facilitation of parental care may increase breeding performance, it can also negatively affect other fitness components. © 2017 Elsevier Inc. All rights reserved.

1. Introduction Reproductive behaviour is socially facilitated when the reproductive behaviour of an individual augments the rate at which conspecific individuals perform the same behaviour or related behaviours (Zajonc, 1965). For instance, in humans, transition to parenthood increases after siblings, friends or colleagues have become parents (Balbo and Barban, 2014; Lois and Arránz Becker, 2013; Pink et al., 2014). Social facilitation of reproduction may therefore give the impression that the decision to reproduce spreads ‘contagiously’ within social groups. The most probable explanation for this phenomenon in humans is social learning, which is the process by which decisions are altered by new information obtained from observation of other individuals (Heyes, 1994). In that case, social learning provides potential new parents with information on the feasibility of becoming a parent, thereby minimizing the risk of taking the wrong decision. Social facilitation of reproduction is, however, not restricted to humans, as social stimulation can also hasten reproduction within animal groups (Danchin, 1988; Waas, 1988). ⁎ Corresponding author. E-mail address: [email protected] (M. Beaulieu).

http://dx.doi.org/10.1016/j.yhbeh.2017.03.007 0018-506X/© 2017 Elsevier Inc. All rights reserved.

Studies on social facilitation of reproduction have so far mostly focused on the decision to reproduce and become a parent but have not examined whether such a phenomenon occurred later when parents have to care for their offspring. This is important, as social facilitation of parental care may enhance the development and the survival of young within groups. Moreover, studies on social facilitation of reproduction have overlooked its underlying physiological mechanisms and its potential costs. Uncovering the physiological basis of social facilitation could be of particular importance in an evolutionary context, since we do not know if and how it could impact life history trade-offs. In mammals, oxytocin and prolactin represent the main hormones motivating parental behaviour, as they increase attraction, contact and protectiveness towards young (Rilling and Young, 2014). In birds, prolactin is also involved in the expression of parental behaviour, such as egg incubation, chick-brooding or chick-provisioning, and is also probably involved in alloparental behaviour (Angelier et al., 2016). Therefore, oxytocin and prolactin, directly acting on the central nervous system of parents, appear to be primarily related to their intrinsic motivation to provide parental care. However, the parental motivation triggered by these hormones needs to be weighted by the amount of resources individuals can invest into parental care. Stress hormones (cortisol and corticosterone) appear to assume this role, as they regulate resource

40

M. Beaulieu et al. / Hormones and Behavior 93 (2017) 39–46

allocation between self-maintenance and reproduction (Wingfield and Sapolsky, 2003). For instance, high cortisol levels in female rhesus macaques (Macaca mulatta) increase the probability of rejecting their young, and can annihilate the positive maternal effects associated with high oxytocin levels (Maestripieri et al., 2009). In birds, even though low corticosterone levels promote parental care, high corticosterone levels induce a decline in prolactin levels, so that high corticosterone levels coupled with low prolactin levels lead parents to decrease nest attendance and ultimately desert their nest (Angelier et al., 2009; Criscuolo et al., 2005; Spée et al., 2010). Such negative effects of stress hormones on oxytocin and prolactin may explain why stress hormones are typically low and unresponsive to perturbations when parental care needs to be provided to offspring, while the opposite is true at the onset of breeding when energy needs to be mobilized to promote reproduction (Krause et al., 2015, 2016). Social facilitation of parental care could then be triggered in species breeding in groups if the hormonal status of an individual (thereafter, focal individual) is modulated by its exposure to the reproductive behaviour of conspecifics. In that case, focal individuals exposed to the reproductive behaviour of conspecifics would be expected to exhibit increased oxytocin and prolactin levels but low and constant levels of stress hormones, which may allow them to enhance their current reproductive performance. Prioritizing reproduction, however, is expected to decrease resource allocation to self-maintenance, which could be assessed through the measurement of physiological markers mediating homeostasis balance. For instance, reproduction may lead to a deficit in antioxidant defences, thereby increasing the risk of exposure to high levels of oxidative damage (Alonso-Álvarez et al., 2004). This shift in the oxidative balance of reproductive individuals is therefore thought to reflect how reproducing individuals prioritize reproduction relative to self-maintenance (Beaulieu et al., 2015). Thus, social facilitation of parental care is likely to be mediated not only by hormonal levels, but also by variation in oxidative status (i.e. low antioxidant defences and high oxidative damage). The social conditions experienced by reproducing individuals have to be considered in view of the environmental conditions under which reproduction occurs. Indeed, environmental conditions can alter reproductive behaviour (Winkler et al., 2002), presumably because they determine the overall level of resources to allocate between reproduction and self-maintenance. Accordingly, environmental conditions can alter the afore-mentioned physiological markers mediating reproductive behaviours (Angelier et al., 2016; Beaulieu and Costantini, 2014; Wingfield, 2013). Reproductive behaviour is therefore likely to depend on the interaction between environmental conditions and social stimulation. As favourable environmental conditions allow parents to maximize their investment into reproduction, the effects of social stimulation on parental investment are likely to be weak under such conditions. In contrast, under poor environmental conditions, parental investment should be far from its upper limit, so that it could still be enhanced by social stimulation. Under the assumption that social stimulation still occurs despite poor environmental conditions and that high parental investment enhances reproductive success, an evolutionary role for social facilitation may be to buffer the negative impact of environmental conditions on parental investment. Alternatively, reproduction failure is likely to be precipitated under extremely poor environmental conditions that reduce not only resource availability but also social stimulation, thereby preventing the occurrence of social facilitation of reproduction. Here, we examined whether the exposure of focal individuals to parental interactions performed by conspecifics altered physiological mediators of parental care (hormonal levels, oxidative status). We conducted our study in a colonial Antarctic seabird, the Adélie penguin (Pygoscelis adeliae), in two years characterized by different environmental conditions: the first year (2006–07) was characterized by an early sea-ice retreat before summer compared with the second year (2007–08). This resulted in more favourable conditions when penguins reproduced in summer 2006–07, as proxies of food availability were

enhanced (Beaulieu et al., 2010a). Importantly, we have already shown in Adélie penguins that low prolactin levels and high corticosterone levels deteriorate behavioural traits involved in parental care (e.g. incubation commitment, foraging behaviour) (Cottin et al., 2014; Spée et al., 2010; Thierry et al., 2013), and that antioxidant defences appear to be involved in self-maintenance mechanisms during reproduction (Beaulieu et al., 2011). Therefore, if social facilitation of parental care occurs in this species, we expect high exposure to parental interactions to be related to elevated prolactin levels, to low corticosterone levels, and to reduced antioxidant defences (and possibly to high oxidative damage). We also expect the effects of social stimulation on physiological mediators of parental care to be exacerbated during the second year, when environmental conditions were less favourable. 2. Methods 2.1. Fieldwork Our study was conducted in Dumont d'Urville, Antarctica (66°40′S, 140°01′E) and was approved by the Ethic Committee of the French Polar Institute Paul-Emile Victor, as part of the Program 137. In austral summers 2006–07 and 2007–08, 92 and 96 penguins, respectively, were selected within a colony of ca. 150 m2 and containing a total of N500 individuals (see picture in Supplementary material 1). A few days before egg laying (early November), the birds were individually identified with a symbol painted on their chest with waterproof Nyanzol-D, and with a pseudo passive transponder (31.2 × 3.8 mm, 0.8 g, Texas Instruments TIRIS, Dallas, TX, U.S.A.) implanted subcutaneously. Sex was determined a posteriori by using a combination of parameters including cloacal inspection, copulatory position and incubation routine (Kerry et al., 1993). In Adélie penguins, after egg hatching (mid- to late December), one parent stays on the nest to provide protection to its offspring, while the other forages at sea to bring food back to the nest. Throughout the brooding stage, males and females regularly switch parental duties during nest relief ceremonies, when one parent comes back from the sea while the other one is about to leave the nest to forage at sea. In our study, during the early brooding stage (late December to early January), the colony was observed from a distance (ca. 20 m) every hour each day to monitor the nest relief ceremonies of identified birds that were still rearing chicks (n = 84 in 2006–07 and n = 90 in 2007–08). The time of nest relief was recorded. Among these identified birds, 94 individuals (n = 47 in both seasons) were captured once and bled immediately after being relieved by their partner (i.e. after they took part in their own nest relief ceremony). We captured focal penguins after they were relieved by their partner to avoid capturing them while they were still brooding their young chicks on the nest. Blood was collected from the wing vein with a heparinized syringe. After centrifugation, plasma samples were stored at −20 °C. Because the capture and the restraint constitute an acute stress that may influence physiological markers in blood (Angelier et al., 2013), we minimized the stress of penguins by covering their head with a hood, and by collecting blood as quickly as possible (mean ± SE = 267 ± 7 s). 2.2. Social stimulation The number of nest relief ceremonies involving identified penguins and occurring within the hour preceding blood sampling was used as an index of social stimulation received by focal penguins from other active parents on the colony. During nest relief ceremonies, both partners typically stand and wave their necks back and forth while uttering a loud cackling call (Müller-Schwarze and MüllerSchwarze, 1980). Given the small size of the colony (ca. 150 m2 ), we assumed that these loud mutual displays associated with nest relief ceremonies were audible by all penguins breeding on the colony, irrespective of the location of their nest within the colony (see


picture in Supplementary material 1). We also assumed that nest relief ceremonies might represent social cues affecting physiological mediators of parental care in focal individuals, as within pairs, loud mutual displays have been described as reflecting the readiness of parents to switch duties at the nest. Indeed, the duration of nest relief ceremonies decreases with the number of loud mutual displays performed at the beginning of these ceremonies (Müller-Schwarze and Müller-Schwarze, 1980), suggesting that loud mutual displays can enhance parental motivation. As the colony was observed every hour in our study, we used this time interval to estimate the level of social stimulation received by focal penguins. Moreover, previous studies have shown that the physiological parameters considered here could vary in the blood of birds within such a period of time (Angelier et al., 2013; Haussmann et al., 2012; Treidel et al., 2013). 2.3. Laboratory analyses Plasma prolactin concentrations were determined at the CEBC (France) by an heterologous radioimmunoassay already validated for Adélie penguins, and corticosterone levels were measured at the IPHC (France) by a quantitative sandwich immunoassay technique (Assay Pro, AssayMax Corticosterone ELISA Kit), as already reported for this species (Cottin et al., 2014; Spée et al., 2010). We also measured two markers of oxidative status in plasma samples: (1) total antioxidant capacity as measured by the OXY-adsorbent test (Diacron International, Grosseto, Italy), and (2) hydroperoxide concentration as measured by the d-ROM test (Diacron International, Grosseto, Italy). The OXY-adsorbent test quantifies the ability of plasma to oppose the massive oxidative action of hypochlorous acid through different antioxidant compounds (Costantini, 2011). The d-ROM test measures the concentration of hydroperoxides, which derive from the oxidation of fatty acids, proteins and nucleic acids (Costantini, 2016). All tests have already been used successfully in plasma samples of Adélie penguins (Beaulieu et al., 2010b; Spée et al., 2010). Intra- and inter-assay coefficients of variations for all tests were between 4 and 9%. 2.4. Statistical analyses As focal penguins were exposed to the nest relief ceremonies of conspecifics but also took part in their own nest relief ceremony before being bled, we considered total social stimulation, which included the nest relief ceremonies performed by conspecifics (ranging between 0 and 6) and the nest relief ceremony performed by the focal penguin (equal to 1). Because 27 penguins were monitored in both years, and because some males and females belonged to the same nests, we examined the effects of social stimulation on physiological traits by using general linear mixed models (GLMM) with the factor “year” as a repeated factor and individual nested within the nest as a random factor. These models included prolactin (log-transformed), corticosterone (square-root transformed), antioxidant defences or oxidative damage as dependent variables, and social stimulation, year and sex as independent variables. With a similar GLMM, we also examined whether the social stimulation differed between years and sexes. We used the longitudinal data from the 27 individuals that we sampled in both years (n = 14 females, n = 13 males) to examine whether, at the intra-individual level, inter-annual variation in social stimulation was reflected by variation in physiological mediators of parental care. Towards this end, we used GLMM with nest as a random factor, interannual variation in social stimulation (Δ social stimulation = social stimulation2007–08 − social stimulation2006–07) and sex as fixed factors, and inter-annual variation in physiological parameters (Δ physiological parameter = physiological parameter2007–08 − physiological parameter2006–07) as dependent variable. Δ social stimulation ranged between − 3 and + 3 nest relief ceremonies in general, but only between − 1 and +1 for most focal individuals (19 out of 27 individuals).

41

All interactions were included in all models. Multiple comparisons were conducted with Bonferroni post-hoc tests. We also estimated effects sizes (η2p) for each fixed factor by using general linear models including the same fixed factors as in the general linear mixed models described above. Analyses were conducted using SPSS 22.00 (SPSS Inc., Chicago, Ill., USA). Results are expressed as means ± SE. 3. Results Male and female penguins were exposed to the same level of social stimulation (mean ± SE = 2.7 ± 0.1 nest relief ceremonies) in both years (year: F1, 34 = 0.01, P = 0.980, η2p = 0.00, sex: F1, 42 = 0.02, P = 0.882, η2p = 0.00, year ∗ sex: F1, 34 = 1.51, P = 0.228, η2p = 0.02). Social stimulation was positively related to prolactin levels in focal individuals (Table 1, Fig. 1a). Accordingly, longitudinal data indicated that inter-annual variation in social stimulation tended to be positively related to inter-annual variation in prolactin levels (although this was not significant; Table 2, Fig. 2a). In both cases (all and longitudinal data), effect size values suggested moderate to large effects of social stimulation on prolactin levels (Tables 1 and 2). In contrast to prolactin levels, corticosterone levels remained unaffected by social stimulation (Table 1, Fig. 1b) or by inter-annual variation in social stimulation (Table 2, Fig. 2b). The relationships between social stimulation and hormonal levels did not differ between males and females, irrespective of the year when they were monitored (Table 1). However, males showed lower levels of prolactin than females (2006–07: P = 0.061, 2007–08: P = 0.001; Table 1, Fig. 3a), and both males and females showed slightly higher corticosterone levels in 2007–08 than in 2006–07 (Table 1, Fig. 3b). Similar to prolactin levels, we found that social stimulation was related to the antioxidant defences of focal penguins (Table 1). However, in contrast to prolactin levels, high levels of social stimulation were associated with low antioxidant defences irrespective of any other factor (Table 1, Fig. 1c). Accordingly, longitudinal data indicated that inter-annual variation in individual social stimulation tended to be negatively related to inter-annual variation in individual prolactin levels (Table 2, Fig. 2a), although this was only significant in females (females: F1, 12 = 7.82, P = 0.016, η2p = 0.39; males: F1, 11 = 0.62, P = 0.446, η2p = 0.05). In both cases (all and longitudinal data), effect size values suggested moderate to large effects of social stimulation on antioxidant defences (Tables 1 and 2). In contrast to antioxidant defences, the levels of oxidative damage in penguins were not significantly related to social stimulation (Table 1, Fig. 1d) but differed between years, with penguins showing higher levels of oxidative damage in 2006–07 than in 2007–08 (Table 1, Fig. 2d). A significant interaction between year and sex indicated that oxidative damage in males and females varied differently in relation to environmental conditions (Table 1). However, post-hoc tests failed to identify significant differences between groups (all P N 0.9). 4. Discussion We examined here, in a colonial bird, whether variation in exposure to parental interactions performed by conspecifics (i.e. social stimulation) was reflected by variation in physiological mediators of parental care in focal individuals. We found that focal individuals exposed to higher levels of social stimulation exhibited higher prolactin levels and lower antioxidant defences. However, their levels of corticosterone and oxidative damage remained unrelated to social stimulation. To our knowledge, these results represent the first evidence showing that inter-individual differences in social stimulation is mirrored by interindividual differences in terms of physiological mediators of parental care. We also found corresponding trends at the intra-individual level, with higher exposure to social stimulation tending to increase prolactin levels and to decrease antioxidant defences in the same focal individuals followed over two consecutive years. The fact that we only found non-

42


Table 1 Results of general linear mixed models examining the effect of social stimulation on hormonal levels (prolactin, corticosterone) and markers of oxidative status (antioxidant defences and oxidative damage) in brooding Adélie penguins. Additional factors (sex and year) were included in the models. Significant results are represented in bold. Effects sizes (η2p) were estimated by general linear models with the same fixed factors as in the general linear mixed models.

Social stimulation Sex Year Social stimulation ∗ sex Social stimulation ∗ year Sex ∗ year Social stimulation ∗ sex ∗ year

Prolactin

Corticosterone

Antioxidant defences

Oxidative damage

F1, 86 = 8.83, P = 0.004, η2p = 0.09 F1, 80 = 2.73, P = 0.103, η2p = 0.00 F1, 75 = 0.55, P = 0.460, ηp2 = 0.00 F1, 86 = 1.54, P = 0.219 η2p = 0.02 F1, 85 = 0.04, P = 0.836 η2p = 0.00 F1, 75 = 4.47, P = 0.038 η2p = 0.05 F1, 85 = 3.26, P = 0.075 η2p = 0.04




significant trends in this case is probably due to the small intra-individual variation in exposure to social stimulation between years, with most individuals experiencing almost the same level of social stimulation in the two years we considered. 4.1. Nest relief ceremonies as social cues In colonial animals, reproduction necessarily implies repeated obligate exposure to conspecifics. For instance, in Adélie penguins, except during the courtship period, reproducing individuals are subjected to more stimulation from conspecifics than from their own partner (which forages at sea except during short nest relief ceremonies).

Colonial animals are therefore likely to have evolved mechanisms allowing them not to consider their neighbours as a constant source of stress, which would result in chronic stress and alter their fitness (McEwen and Wingfield, 2011). One of these mechanisms may be the low responsiveness of stress hormones to perturbations during the parental stage of breeding (Krause et al., 2016, 2015). The fact that we conducted our study during this breeding stage is likely to explain why we did not find any relationship between social stimulation and corticosterone levels in Adélie penguins. However, during the parental stage, king penguins (Aptenodytes patagonicus) still show elevated corticosterone levels when bird density and social interactions increase within the colony (Viblanc et al., 2014). This difference between Adélie and king

Fig. 1. Relationships between social stimulation (number of nest relief ceremonies occurring on the colony within the hour preceding blood sampling) and physiological mediators of parental care in Adélie penguins. Results are presented for hormonal levels (prolactin (a), corticosterone (b)) and markers of oxidative status (antioxidant defences (c), oxidative damage (d)). Solid lines represent regression lines for physiological parameters significantly related to social stimulation.


43

Table 2 Results of general linear mixed models examining the effect of inter-annual variation in social stimulation on inter-annual variation in hormonal levels (prolactin, corticosterone) and markers of oxidative status (antioxidant defences and oxidative damage) in brooding Adélie penguins. The factor ‘sex’ was also included in the models. Significant results are represented in bold. Effects sizes (η2p) were estimated by general linear models with the same fixed factors as in the general linear mixed models.

Δ Social stimulation Sex Δ Social stimulation ∗ sex

Δ Prolactin

Δ Corticosterone

Δ Antioxidant defences

Δ Oxidative damage

F1, 23 = 2.98, P = 0.098, η2p = 0.12 F1, 23 = 0.09, P = 0.770, η2p = 0.00 F1, 23 = 0.87, P = 0.360 ηp2 = 0.04

F1, 8 = 0.84, P = 0.386, η2p = 0.00 F1, 8 = 4.50, P = 0.068, η2p = 0.01 F1, 11 = 0.01, P = 0.959 ηp2 = 0.02

F1, 14 = 3.16, P = 0.097, η2p = 0.07 F1, 10 = 4.65, P = 0.057, η2p = 0.10 F1, 21 = 5.05, P = 0.035 ηp2 = 0.21

F1, 18 = 0.09, P = 0.774, η2p = 0.00 F1, 10 = 2.87, P = 0.121, η2p = 0.08 F1, 23 = 3.20, P = 0.087 ηp2 = 0.12

penguins may be due to differences in terms of density. Indeed, bird density is higher in colonies of king penguins (especially in the centre of the colony, 1.6–2.3 bird/m 2) than in colonies of Adélie penguins (where density is more constant across the colony, 0.5 and 1.5 nest/m 2 ) (Chamaille-Jammes et al., 2000; LaRue et al., 2014; Lengagne et al., 1999; Woehler and Riddle, 1998). The fact that king penguins still breed in dense and stressful areas within the colony suggest that they still benefit from them. One of these benefits may be due to elevated prolactin levels associated with higher social stimulation, as we found in Adélie penguins. During the early brooding stage, Adélie penguins are vigilant and monitor their surroundings for about half of their time budget on the nest (Thierry et al., 2014). It is therefore probable that the visual and acoustic exposure of focal penguins to the nest relief ceremonies of conspecifics underlies the relationship between social stimulation and prolactin levels that we found. Even though visual cues might be involved, loud mutual displays performed by parents during nest relief ceremonies and that easily propagate across the colony appear more likely to

be detected by focal penguins. Moreover, within pairs, loud mutual displays have been described as reflecting the readiness of parents to switch duties (Müller-Schwarze and Müller-Schwarze, 1980). This observation along with our results suggests that loud mutual displays during nest relief ceremonies contain information that might modulate parental behaviour within pairs and across the colony. However, the behaviour of conspecific chicks during nest relief ceremonies may also be part of the social stimulators acting on prolactin levels in focal penguins. Indeed, during nest relief ceremonies, the chicks usually become agitated, beg for food and can even perform an immature version of the adult loud mutual display (Spurr, 1975). As parent-offspring recognition is effective only at the end of the brooding stage (Davis and McCaffrey, 1989), it is probable that the calls of unidentified conspecific chicks could affect physiological mediators of parental care in non-discriminating focal adults. Similarly, it is also possible that the behaviour of the chicks of focal individuals underlies the relationship between social stimulation and prolactin levels in focal adults. Indeed, chicks may react to the calls of conspecifics adults (as they do later during the

Fig. 2. Relationship between inter-annual variation in social stimulation (number of nest relief ceremonies occurring on the colony within the hour preceding blood sampling) and interannual variation in physiological mediators of parental care in Adélie penguins. Results are presented for hormonal levels (prolactin (a), corticosterone (b)) and markers of oxidative status (antioxidant defences (c), oxidative damage (d)). White symbols represent females, dark grey symbols males, and light grey symbols males and females together. Only inter-annual variation in antioxidant defences in females was significantly related to inter-annual variation in social stimulation. The solid line represents the regression lines for this negative relationship. Dashed lines represent regression lines for non-significant trends (0.05 b P b 0.1) for all penguins (males and females).

44


Fig. 3. Physiological mediators of parental care in focal Adélie penguins. Results are represented by box plots (box: median and inter-quartile range; whiskers: 10% and 90% quantiles) for hormonal levels (prolactin (a), corticosterone (b)) and markers of oxidative status (antioxidant defences (c), oxidative damage (d)) in females (F, white bars) and males (M, grey bars) in two breeding seasons (2006–07 (indicated with 2007) and 2007–08 (indicated with 2008)). Different letters above boxes indicate significant differences between groups.

crèche stage; (Beaulieu et al., 2009)), which in turn may affect the physiology of their parents. Finally, it is also possible that focal penguins exposed to social stimulation before being relieved interact more with their partner during their own nest relief ceremony, which in turn may lead to the physiological differences that we observed. It is improbable, however, that the behaviour of the partners of focal penguins could affect the physiology of focal penguins. Indeed, in the colony that we studied, nests are very close to the sea (see picture in the Supplementary material 1), and birds returning from a foraging trip typically need b10 min to reach their nest. Consequently, partners were unlikely to be exposed to the social stimulation occurring on the colony within the hour preceding the relief. Clearly, investigating each link in the chain of events following nest relief ceremonies in Adélie penguins should be the target of future studies if we want to understand the causal relationships underlying our results. Assuming that focal individuals directly monitor their conspecifics, observing individuals coming back from the sea to relieve their partner and feed chicks may provide them with information on the feasibility of investing in reproduction. Indeed, the return of penguins from the sea to relieve their partner and feed offspring is likely to indicate that conditions at sea are favourable for food acquisition. If so, nest relief ceremony is likely to represent public information, “a form of indirect social information […] used by individuals to estimate the quality of environmental parameters” (Valone, 2007). As penguins cannot directly gather personal information about oceanic conditions from their nests, they are likely to exploit inadvertent social information on the colony, with a high frequency of nest relief ceremonies likely indicating favourable environmental conditions at sea and the possibility to increase parental investment safely. This may explain why in our study a high frequency of nest relief ceremonies was associated with elevated prolactin levels in focal individuals. Eavesdropping nest relief ceremonies from neighbours may therefore allow penguins to assess the risk-benefit ratio associated with increased parental investment (Danchin, 2004). This explanation is somewhat reminiscent of the hypothesis according to

which social learning in humans provides potential new parents with information on the feasibility of becoming a parent (Balbo and Barban, 2014; Lois and Arránz Becker, 2013; Pink et al., 2014). 4.2. Prolactin as a mediator of social facilitation of parental care Compared with individuals for which prolactin was experimentally inhibited with bromocriptine, control Adélie penguins show higher incubation commitment, are less likely to abandon their nest and increase their diving effort (Cottin et al., 2014; Thierry et al., 2013). This suggests that socially-enhanced prolactin levels are likely to at least preserve and at most enhance the reproductive performance of focal parents. As social stimulation is expected to be more frequent in large colonies, its positive effects on prolactin secretion may add up to other benefits related to colonial breeding (e.g. dilution effects, anti-predator defence, foraging efficiency and finding mates) and may contribute to Allee effects in animal colonies (i.e. positive relationship between population density and individual fitness) (Schippers et al., 2011). This greatly enlarges the scope of the social environment of breeding individuals, usually restricted to the partner and the offspring of focal individuals (Royle et al., 2014). One key fundamental consequence is that a colony should not be considered just as an accumulation of breeding pairs but as a social group where individuals can influence and benefit from each other. As all breeding penguins on the colony can equally be producers and receivers of nest relief ceremonies, reciprocal altruism may explain how the positive effects of social stimulation on prolactin levels have evolved in this species. Moreover, if producers and receivers share some degree of genetic relatedness (which remains to be examined in Adélie penguins), kin selection may also underlie these effects (Bourke, 2007; Valone, 2007). As such, these effects appear comparable to social facilitation of parental care in helpers of cooperative animal species (Brown and Vleck, 1998; Schoech et al., 1996). Environmental conditions did not affect the levels of social stimulation experienced by focal penguins in our study. Moreover, in contrast


to corticosterone levels (insensitive to social stimulation), prolactin levels remained unaffected by environmental conditions. This suggests that constant social stimulation may have counterbalanced the effects of unfavourable environmental conditions on parental investment through its effects on prolactin levels. To some extent, social stimulation may therefore act as a buffer between environmental conditions and reproductive performance, tending to reduce variability among individuals and across years. This may explain why the reproductive success of penguins was similar in the two years we considered in our study, despite strong differences in terms of environmental conditions (Beaulieu et al., 2010a). 4.3. Altered oxidative balance as a cost of social stimulation A side effect of social stimulation that we observed in Adélie penguins was an altered oxidative balance. Indeed, the exposure of focal individuals to a high number of nest relief ceremonies was related to low antioxidant defences. Previous studies have shown that the social environment of animals could affect their oxidative status, particularly in species with a hierarchical social structure (Beaulieu et al., 2014; Georgiev et al., 2015). However, in these studies, the link between social rank and oxidative balance may not only be explained by differences in terms of social stimulation but also by other factors related to social rank (e.g. access to resources, reproduction rate, aggressiveness). Here, we show in a species with no social hierarchical structure that social stimulation and antioxidant defences were negatively correlated. The fact that we did not find any significant relationship between social stimulation and oxidative damage suggests that the effects of decreased antioxidant defences on oxidative damage may be either buffered by additional mechanisms (e.g. reparation processes) or postponed (for instance, if conditions become energetically more demanding while fasting on the nest or foraging at sea (Schull et al., 2016)). The precise mechanisms responsible for the negative relationship between social stimulation and antioxidant defences that we found remain unknown. However, given that social stimulation was positively related to prolactin levels in our study, it is tempting to speculate that increased prolactin levels in focal penguins exposed to high levels of social stimulation were responsible for their simultaneously low antioxidant defences. Indeed, pro-oxidant effects of prolactin have been described in previous studies, where the use of a prolactin inhibitor (bromocriptine) decreases the production of free radicals in treated rodents (Hilfiker-Kleiner et al., 2007; Yoshikawa et al., 1994). Clearly, further experimental studies are needed to examine more precisely the interplay between social stimulation, prolactin levels and oxidative status. 4.4. Conclusion and perspectives Overall, our study emphasizes the importance of social stimulation on physiological mediators of parental care in animals reproducing in colonies, and opens exciting avenues for future research. For instance, as Adélie penguins can breed syntopically with other penguins, such as Gentoo (Pygoscelis papua) and Chinstrap penguins (Pygoscelis antarcticus) (Trivelpiece et al., 1987), it is possible that they can also eavesdrop information from these closely-related species. Thus, physiological mediators of parental care could not only be modulated by conspecifics but also by syntopic individuals from other species. If so, this would give a novel perspective on the social environment of focal breeding individuals by expanding it from the intra- to the interspecific level. Moreover, as prolactin levels have been described as depending on individual characteristics such as age or breeding experience in other seabird species (e.g. old birds have higher prolactin levels) (Angelier et al., 2006), it is likely that physiological mediators of parental care vary more easily with social stimulation in certain population categories than in others (e.g. in young birds with low prolactin levels). We could not examine the effects of age in the current study, as we

45

worked with animals of unknown age. This highlights the need for more studies examining the effects of social stimulation on physiological mediators of parental care at the individual level and using experimental approaches. For instance, in our observational study, intraindividual variation in social stimulation was too small to clearly confirm the results that we found at the inter-individual level (although it showed the same trends). Using playbacks of nest relief ceremonies on the colony would therefore be worthwhile in future studies, as it would allow us to expose the same penguins to a broader range of social stimulation. Finally, the fact that social stimulation was related to oxidative status provides a novel framework to study the relationship between social conditions and ageing, through other ageing-associated processes like telomeres (Blackburn and Epel, 2012; Bourke, 2007). Measuring prolactin levels, oxidative status and telomere length longitudinally, along with reproductive success and survival in colonies of different sizes and where the levels of social stimulation vary would therefore be another interesting next step to examine the long-term effects of socially-induced variation in physiological mediators of parental care. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yhbeh.2017.03.007. Authors' contribution M.B. designed this research, conducted fieldwork and laboratory analyses, and wrote the article. A.A. and T.R. conducted fieldwork. F.C., O.C. and T.R. took part to laboratory measurements. All authors contributed to the writing of the article. Competing interests We have no competing interests. Funding At the time of the study, MB was funded by a PhD fellowship from the French Ministry of Research, and was funded by the University of Greifswald when the article was written. The French Polar Institute Paul-Emile Victor provided financial and logistical support in Adélie Land. Acknowledgements We are grateful to Antoine Dervaux, David Lazin and Anne-Mathilde Thierry for their help in the field in Dumont d'Urville, to Marion Spée for measuring corticosterone levels at the IPHC, and to Charline Parenteau for measuring prolactin levels at the CEBC. We would like to dedicate this article to Guillaume Bouteloup, who overwintered as an ornithologist in Adélie Land as a member of the 55th mission. References Alonso-Álvarez, C., Bertrand, S., Devevey, G., Prost, J., Faivre, B., Sorci, G., 2004. Increased susceptibility to oxidative stress as a proximate cost of reproduction. Ecol. Lett. 7: 363–368. http://dx.doi.org/10.1111/j.1461-0248.2004.00594.x. Angelier, F., Shaffer, S.A., Weimerskirch, H., Chastel, O., 2006. Effect of age, breeding experience and senescence on corticosterone and prolactin levels in a long-lived seabird: the wandering albatross. Gen. Comp. Endocrinol. 149:1–9. http://dx.doi.org/10.1016/ j.ygcen.2006.04.006. Angelier, F., Clément-Chastel, C., Welcker, J., Gabrielsen, G.W., Chastel, O., 2009. How does corticosterone affect parental behaviour and reproductive success? A study of prolactin in black-legged kittiwakes. Funct. Ecol. 23:784–793. http://dx.doi.org/10.1111/j. 1365-2435.2009.01545.x. Angelier, F., Wingfield, J.C., Trouvé, C., de Grissac, S., Chastel, O., 2013. Modulation of the prolactin and the corticosterone stress responses: do they tell the same story in a long-lived bird, the Cape petrel? Gen. Comp. Endocrinol. 182:7–15. http://dx.doi. org/10.1016/j.ygcen.2012.10.008. Angelier, F., Wingfield, J.C., Tartu, S., Chastel, O., 2016. Does prolactin mediate parental and life-history decisions in response to environmental conditions in birds? A review. Horm. Behav. 77:18–29. http://dx.doi.org/10.1016/j.yhbeh.2015.07.014.

46


Balbo, N., Barban, N., 2014. Does fertility behavior spread among friends? Am. Sociol. Rev. 79:412–431. http://dx.doi.org/10.1177/0003122414531596. Beaulieu, M., Costantini, D., 2014. Biomarkers of oxidative status: missing tools in conservation physiology. Conserv. Physiol. 2, cou14. Beaulieu, M., Thierry, A.-M., Maho, Y., Ropert-Coudert, Y., Ancel, A., 2009. Alloparental feeding in Adélie penguins: why is it uncommon? J. Ornithol. 150:637–643. http:// dx.doi.org/10.1007/s10336-009-0386-2. Beaulieu, M., Dervaux, A., Thierry, A.-M., Lazin, D., Le Maho, Y., Ropert-Coudert, Y., Spée, M., Raclot, T., Ancel, A., 2010a. When sea-ice clock is ahead of Adélie penguins' clock. Funct. Ecol. 24:93–102. http://dx.doi.org/10.1111/j.1365-2435.2009.01638.x. Beaulieu, M., Ropert-Coudert, Y., Le Maho, Y., Ancel, A., Criscuolo, F., 2010b. Foraging in an oxidative environment: relationship between delta13C values and oxidative status in Adélie penguins. Proc. R. Soc. B 277:1087–1092. http://dx.doi.org/10.1098/rspb.2009. 1881. Beaulieu, M., Reichert, S., Le Maho, Y., Ancel, A., Criscuolo, F., 2011. Oxidative status and telomere length in a long-lived bird facing a costly reproductive event. Funct. Ecol. 25:577–585. http://dx.doi.org/10.1111/j.1365-2435.2010.01825.x. Beaulieu, M., Mboumba, S., Willaume, E., Kappeler, P.M., Charpentier, M.J.E., 2014. The oxidative cost of unstable social dominance. J. Exp. Biol. 217:2629–2632. http://dx.doi. org/10.1242/jeb.104851. Beaulieu, M., Geiger, R.E., Reim, E., Zielke, L., Fischer, K., 2015. Reproduction alters oxidative status when it is traded-off against longevity. Evolution 69, 1786–1796. Blackburn, E.H., Epel, E.S., 2012. Telomeres and adversity: too toxic to ignore. Nature 490: 169–171. http://dx.doi.org/10.1038/490169a. Bourke, A.F.G., 2007. Kin selection and the evolutionary theory of aging. Annu. Rev. Ecol. Evol. Syst. 38:103–128. http://dx.doi.org/10.1146/annurev.ecolsys.38.091206. 095528. Brown, J.L., Vleck, C.M., 1998. Prolactin and helping in birds: has natural selection strengthened helping behavior? Behav. Ecol. 9:541–545. http://dx.doi.org/10.1093/ beheco/9.6.541. Chamaille-Jammes, S., Guinet, C., Nicoleau, F., Argentier, M., 2000. A method to assess population changes in king penguins: the use of a geographical information system to estimate area-population relationships. Polar Biol. 23, 545–549. Costantini, D., 2011. On the measurement of circulating antioxidant capacity and the nightmare of uric acid. Methods Ecol. Evol. 2:321–325. http://dx.doi.org/10.1111/j. 2041-210X.2010.00080.x. Costantini, D., 2016. Oxidative stress ecology and the d-ROMs test: facts, misfacts and an appraisal of a decade's work. Behav. Ecol. Sociobiol. 70, 809–820. Cottin, M., Chastel, O., Kato, A., Debin, M., Takahashi, A., Ropert-Coudert, Y., Raclot, T., 2014. Decreasing prolactin levels leads to a lower diving effort but does not affect breeding success in Adélie penguins. Horm. Behav. 65:134–141. http://dx.doi.org/ 10.1016/j.yhbeh.2013.12.001. Criscuolo, F., Chastel, O., Bertile, F., Gabrielsen, G.W., Le Maho, Y., Raclot, T., 2005. Corticosterone alone does not trigger a short term behavioural shift in incubating female common eiders Somateria mollissima, but does modify long term reproductive success. J. Avian Biol. 36:306–312. http://dx.doi.org/10.1111/j.0908-8857.2005.03371.x. Danchin, E., 1988. Social interactions in kittiwake colonies: social facilitation and/or favourable social environment. Anim. Behav. 36:443–451. http://dx.doi.org/10. 1016/S0003-3472(88)80014-9. Danchin, É., 2004. Public information: from nosy neighbors to cultural evolution. Science 305:487–491. http://dx.doi.org/10.1126/science.1098254. Davis, L.S., McCaffrey, F., 1989. Recognition and parental investment in Adélie penguins. Emu 89, 155–158. Georgiev, A.V., Muehlenbein, M.P., Prall, S.P., Thompson, M.E., Maestripieri, D., 2015. Male quality, dominance rank, and mating success in free-ranging rhesus macaques. Behav. Ecol. 26:763–772. http://dx.doi.org/10.1093/beheco/arv008. Haussmann, M.F., Longenecker, A.S., Marchetto, N.M., Juliano, S.A., Bowden, R.M., 2012. Embryonic exposure to corticosterone modifies the juvenile stress response, oxidative stress and telomere length. Proc. R. Soc. B 279:1447–1456. http://dx.doi.org/10. 1098/rspb.2011.1913. Heyes, C.M., 1994. Social learning in animals: categories and mechanisms. Biol. Rev. 69: 207–231. http://dx.doi.org/10.1111/j.1469-185X.1994.tb01506.x. Hilfiker-Kleiner, D., Kaminski, K., Podewski, E., Bonda, T., Schaefer, A., Sliwa, K., Forster, O., Quint, A., Landmesser, U., Doerries, C., Luchtefeld, M., Poli, V., Schneider, M.D., Balligand, J.L., Desjardins, F., Ansari, A., Struman, I., Nguyen, N.Q.N., Zschemisch, N.H., Klein, G., Heusch, G., Schulz, R., Hilfiker, A., Drexler, H., 2007. A cathepsin Dcleaved 16 kDa form of prolactin mediates postpartum cardiomyopathy. Cell 128: 589–600. http://dx.doi.org/10.1016/j.cell.2006.12.036. Kerry, K.R., Clarke, J.R., Else, G.D., 1993. Identification of sex of Adelie penguins from observations of incubating birds. Wildl. Res. 20:725–732. http://dx.doi.org/10.1071/ WR9930725. Krause, J.S., Mcguigan, M.A., Bishop, V.R., Wingfield, J.C., Meddle, S.L., 2015. Decreases in mineralocorticoid but not glucocorticoid receptor mRNA expression during the short Arctic breeding season in free-living Gambel's white-crowned sparrow (Zonotrichia leucophrys gambelii). J. Neuroendocrinol. 27:66–75. http://dx.doi.org/ 10.1111/jne.12237. Krause, J.S., Pérez, J.H., Chmura, H.E., Meddle, S.L., Hunt, K.E., Gough, L., Boelman, N., Wingfield, J.C., 2016. The stress response is attenuated during inclement weather in parental, but not in pre-parental, Lapland longspurs (Calcarius lapponicus) breeding in the Low Arctic. Horm. Behav. 83:68–74. http://dx.doi.org/10.1016/j.yhbeh.2016. 05.018.

LaRue, M.A., Lynch, H.J., Lyver, P.O.B., Barton, K., Ainley, D.G., Pollard, A., Fraser, W.R., Ballard, G., 2014. A method for estimating colony sizes of Adélie penguins using remote sensing imagery. Polar Biol. 37:507–517. http://dx.doi.org/10.1007/s00300014-1451-8. Lengagne, T., Aubin, T., Jouventin, P., Lauga, J., 1999. Acoustic communication in a king penguin colony: importance of bird location within the colony and of the body position of the listener. Polar Biol. 21:262–268. http://dx.doi.org/10.1007/ s003000050361. Lois, D., Arránz Becker, O., 2013. Is fertility contagious? Using panel data to disentangle mechanisms of social network influences on fertility decisions. Adv. Life Course Res. 21:123–134. http://dx.doi.org/10.1016/j.alcr.2013.10.001. Maestripieri, D., Hoffman, C.L., Anderson, G.M., Carter, C.S., Higley, J.D., 2009. Mother-infant interactions in free-ranging rhesus macaques: relationships between physiological and behavioral variables. Physiol. Behav. 96:613–619. http://dx.doi.org/10.1016/j. physbeh.2008.12.016. McEwen, B.S., Wingfield, J.C., 2011. What's in a name? Integrating homeostasis, allostasis and stress. Horm. Behav. 57:105–111. http://dx.doi.org/10.1016/j.yhbeh.2009.09.011. What. Müller-Schwarze, D., Müller-Schwarze, C., 1980. Display rate and speed of nest relief in Antarctic Pygoscelid penguins. Auk 97, 825–831. Pink, S., Leopold, T., Engelhardt, H., 2014. Fertility and social interaction at the workplace: does childbearing spread among colleagues? Adv. Life Course Res. 21:113–122. http://dx.doi.org/10.1016/j.alcr.2013.12.001. Rilling, J.K., Young, L.J., 2014. The biology of mammalian parenting and its effect on offspring social development. Science 345:771–776. http://dx.doi.org/10.1126/science. 1252723. Royle, N.J., Russell, A.F., Wilson, A.J., 2014. The evolution of flexible parenting. Science 345: 776–781. http://dx.doi.org/10.1126/science.1253294. Schippers, P., Stienen, E.W.M., Schotman, A.G.M., Snep, R.P.H., Slim, P.A., 2011. The consequences of being colonial: Allee effects in metapopulations of seabirds. Ecol. Model. 222:3061–3070. http://dx.doi.org/10.1016/j.ecolmodel.2011.05.022. Schoech, S.J., Mumme, R.L., Wingfield, J.C., 1996. Prolactin and helping behaviour in the cooperatively breeding Florida scrub-jay, Aphelocoma c. coerulescens. Anim. Behav. 52:445–456. http://dx.doi.org/10.1006/anbe.1996.0189. Schull, Q., Viblanc, V.A., Stier, A., Saadaoudi, H., Lefol, E., Criscuolo, F., Bize, P., Robin, J.-P., 2016. The oxidative debt of fasting: evidence for short to medium-term costs of advanced fasting in adult king penguins. J. Exp. Biol. 219, 3284–3293. Spée, M., Beaulieu, M., Dervaux, A., Chastel, O., Le Maho, Y., Raclot, T., 2010. Should I stay or should I go? Hormonal control of nest abandonment in a long-lived bird, the Adélie penguin. Horm. Behav. 58:762–768. http://dx.doi.org/10.1016/j.yhbeh.2010. 07.011. Spurr, E.B., 1975. Behavior of the Adelie penguin chick. Condor 77, 272–280. Thierry, A.M., Brajon, S., Massemin, S., Handrich, Y., Chastel, O., Raclot, T., 2013. Decreased prolactin levels reduce parental commitment, egg temperatures, and breeding success of incubating male Adélie penguins. Horm. Behav. 64:737–747. http://dx.doi. org/10.1016/j.yhbeh.2013.06.003. Thierry, A.-M., Brajon, S., Spee, M., Raclot, T., 2014. Differential effects of increased corticosterone on behavior at the nest and reproductive output of chick-rearing Adélie penguins. Behav. Ecol. Sociobiol. 68:721–732. http://dx.doi.org/10.1007/s00265-0141685-z. Treidel, L.A., Whitley, B.N., Benowitz-Fredericks, Z.M., Haussmann, M.F., 2013. Prenatal exposure to testosterone impairs oxidative damage repair efficiency in the domestic chicken (Gallus gallus). Biol. Lett. 9:20130684. http://dx.doi.org/10.1098/rsbl.2013. 0684. Trivelpiece, W.Z., Trivelpiece, S.G., Volkman, N.J., 1987. Ecological segregation of Adélie, gentoo, and chinstrap penguins at King George Island, Antarctica. Ecology 68, 351–361. Valone, T.J., 2007. From eavesdropping on performance to copying the behavior of others: a review of public information use. Behav. Ecol. Sociobiol. 62:1–14. http://dx.doi.org/ 10.1007/s00265-007-0439-6. Viblanc, V.A., Gineste, B., Stier, A., Robin, J.P., Groscolas, R., 2014. Stress hormones in relation to breeding status and territory location in colonial king penguin: a role for social density? Oecologia 175:763–772. http://dx.doi.org/10.1007/s00442-014-2942-6. Waas, J.R., 1988. Acoustic displays facilitate courtship in little blue penguins, Eudyptula minor. Anim. Behav. 36:366–371. http://dx.doi.org/10.1016/S0003-3472(88)80007-1. Wingfield, J.C., 2013. Ecological processes and the ecology of stress: the impacts of abiotic environmental factors. Funct. Ecol. 27:37–44. http://dx.doi.org/10.1111/1365-2435. 12039. Wingfield, J.C., Sapolsky, R.M., 2003. Reproduction and resistance to stress: when and how. J. Neuroendocrinol. 15:711–724. http://dx.doi.org/10.1046/j.1365-2826.2003. 01033.x. Winkler, D.W., Dunn, P.O., McCulloch, C.E., 2002. Predicting the effects of climate change on avian life-history traits. Proc. Natl. Acad. Sci. U. S. A. 99:13595–13599. http://dx. doi.org/10.1073/pnas.212251999. Woehler, E.J., Riddle, M.J., 1998. Spatial relationships of Adélie penguin colonies: implications for assessing population changes from remote imagery. Antarct. Sci. 10, 449–454. Yoshikawa, T., Minamiyama, Y., Naito, Y., Kondo, M., 1994. Antioxidant properties of bromocriptine, a dopamine agonist. J. Neurochem. 62, 1034–1038. Zajonc, R.B., 1965. Social facilitation. Science http://dx.doi.org/10.1126/science.149.3681. 269.