SEARCH

SEARCH BY CITATION

Keywords:

  • alternative reproductive tactics;
  • male dimorphism;
  • phenotypic plasticity;
  • sperm competition

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Data accessibility
  9. References
  10. Supporting Information
  1. Original sperm competition theory assumed that males trade expenditure on searching for mates for expenditure on the ejaculate and predicted that males should increase their expenditure on the ejaculate in response to increased risk of competition. A recent extension of this theory has modelled pre-copulatory expenditure in terms of direct contest competition and predicts that when the gains from marginal investment in weaponry are large, males might be expected to allocate resources to armaments even at the expense of the ejaculate.
  2. Here, we examine socially cued plasticity in allocation to pre- (body condition) and post-copulatory (testes mass) traits in a male dimorphic beetle, Onthophagus taurus, where major males fight for access to females and minor males obtain reproductive success via sperm competition. Both male morphs were either reared in social isolation or exposed to rivals during the period of sexual maturation following adult emergence.
  3. Testes mass was found to be insensitive to social cues of future mating competition for both major and minor males. Major males allocated more to body condition when exposed to rivals, a response expected for a species in which the outcome of dyadic contests strongly affects male reproductive success. In contrast, minor male allocation to condition was insensitive to social cues.
  4. Our data illustrate how socially cued plasticity in pre- and post-copulatory traits can depend on the relative importance of these episodes of selection for individual male fitness. In O. taurus dung beetles, males strategically adjusted the amount of resources they allocated to winning pre-copulatory contests over access to females. Strategic allocation to pre-copulatory contest competition did not come at a cost to male investment in sperm competition, suggesting that males may trade investment into contest competition against some other life-history trait, such as longevity. The lack of plasticity in testes size suggests that selection from sperm competition may be a relatively constant feature of this species mating system.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Data accessibility
  9. References
  10. Supporting Information

In polygamous mating systems, males will compete for access to multiple females, and females will mate with several males before producing offspring. Pre-copulatory competition among males for access to females is expected to favour the evolution of weapons or ornaments that affect a male's mating success (Darwin 1871; Andersson 1994), while multiple mating by females will generate post-copulatory sperm competition, which is expected to favour the evolution of traits that affect paternity (Parker 1970; Simmons 2001). Theoretical models of sperm competition assume that males have finite resources available for mating effort and ejaculate production and that ejaculate expenditure should increase with increased risk of sperm competition – the probability that a female will mate with more than one male (Parker & Pizzari 2010). Indeed, there is now a wealth of evidence to show that males are acutely sensitive to sperm competition risk and will increase their ejaculate expenditure when risks are elevated (Wedell, Gage & Parker 2002; Kelly & Jennions 2011). Early theory modelled male pre-copulatory expenditure as mate search, whereby male mating success increases linearly with expenditure. Many mating systems, however, are characterized by direct male contests over females. A recent extension to sperm competition theory has modelled pre-copulatory expenditure in terms of contest competition, in which the probability of mating can depend strongly on marginal increases in expenditure on armaments (Parker, Lessells & Simmons 2013). These models predict that expenditure on armaments should increase and expenditure on ejaculates should decrease, as the gains from marginal investment in armaments increase. Thus, where males directly contest access to females, increased levels of competition may be associated with increased allocation to pre-copulatory traits rather than to ejaculates.

The dung beetle Onthophagus taurus is an ideal model system with which to examine phenotypic plasticity in male allocation to both pre- and post-copulatory competition. Males exhibit morphological and behavioural dimorphisms that characterize alternative mating tactics. While major males develop horns used in direct trials of physical strength for the possession of breeding tunnels within which females nest, minor males have only rudimentary horns and sneak into breeding tunnels to mate with nesting females (Moczek & Emlen 2000). Thus, only major males contest access to females directly, while minors search for mating opportunities while avoiding pre-copulatory competition. Physical strength is condition dependent among major males, but not minor males (Knell & Simmons 2010), and during diadic contests, those with longer horns invariably win (Moczek & Emlen 2000). Thus, among major males, fitness rises sharply with their investment in pre-copulatory competition (Hunt & Simmons 2001) and the numbers of mates obtained (Simmons, Beveridge & Krauss 2004).

Nonetheless, females will mate with both major and minor males generating post-copulatory sexual selection. Sperm competition in this species conforms to a fair raffle (Tomkins & Simmons 2000; Simmons, Beveridge & Krauss 2004) and selects for increased testes size (Simmons & García-González 2008). Because minor males sneak copulations, they are always subject to sperm competition, so that variation in the population risk of sperm competition may have little effect on their allocation of resources to testes growth. Indeed, minor males typically allocate more to their testes than do major males (Simmons, Emlen & Tomkins 2007) and maintain allocation to testes growth even in the face of nutritional stress (Knell & Simmons 2010). In contrast, major males will be subject to varying risk of sperm competition depending on the number of competitors in the population and might be predicted to allocate more to testes growth under increased risk. However, if success in pre-copulatory contests contributes more to fitness than success in sperm competition, we might expect major males to allocate more to physical condition than to testes growth when faced with increased competition. Here, we test these predictions by examining phenotypic plasticity in allocation to body condition and testes growth when males are exposed to varying social environments during their post-eclosion pre-reproductive development.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Data accessibility
  9. References
  10. Supporting Information

Adult beetles were collected from a cattle pasture 60 km south of Perth, Western Australia. Beetles were maintained in mixed sex cultures for 1 week with constant access to fresh dung. We established 200 male–female pairs in individual breeding chambers (PVC piping, 25 cm in length and 6 cm in diameter, three-quarters filled with moist sand and topped with 250 ml of cow dung). Chambers were sieved after 1 week; broods were reburied in 5-L plastic boxes and incubated at 28 °C. Adult beetles emerged after 3–4 weeks and were used to produce an F2 generation following the same procedures.

Manipulating the Social Environment

As adult F2 beetles emerged from their brood masses, we assigned males to one of two social environments. The low-competition environment consisted of small chambers (7 × 7 × 5 cm) filled with moist sand and topped with 2 mL of fresh dung, each housing a single male. We assigned 20 horned (major) and 20 hornless (minor) males to the low-competition environments. The high-competition environments consisted of ten, 20-L chambers that were filled with 4·9 L of moist sand and 40 ml of fresh dung, each housing 20 males (12 minor males and eight major males, the approximate ratio of alternative phenotypes found in natural populations of O. taurus in Western Australia, Simmons, Tomkins & Hunt 1999). Thus, the density of males (four males per litre of sand) and the amount of dung (2 mL per male) were identical in both experimental treatments. We kept males in their social environments for 10 days, the time required for them to reach sexual maturity and fully develop their testes. At 10 days of age beetles were frozen.

We measured male size as the maximum width of the pronotum using digital calipers and weighed beetles to an accuracy of 0·01 mg. Beetles were then dissected, and their testes removed and weighed, again to an accuracy of 0·01 mg.

Analyses of Testes Mass and Condition

For our analyses, we randomly sampled two major and two minor males from each of the ten high-competition chambers, providing us with 20 of each morph to contrast with our sample of 20 major and 20 minor males that were reared in low-competition chambers. The chamber from which a beetle was sourced was initially included as a random effect in our analyses, but later removed if the model containing that effect had a higher AIC value than a model without it. To avoid part–whole correlations, we calculated soma mass as the weight of the beetle minus the weight of its testes. In our analyses of testes mass, we thus included soma mass as a covariate along with our predictor variables and all possible interactions. Condition is defined as the amount of resources available for allocation to life-history traits (Rowe & Houle 1996). We estimated condition as the weight of soma relative to body size. For O. taurus, this measure of condition is affected by resource availability and exhibits significant additive genetic variation (Kotiaho, Simmons & Tomkins 2001). Analyses of condition used soma mass as the dependent variable with pronotum width entered as a covariate along with our predictor variables and all possible interactions. All data were log10-transformed prior to statistical analyses, and non-significant interactions were removed from the final models. All analyses were performed in R version 3.0.1 (R_Core_Team 2013), using type 2 sums of squares as calculated by the function ‘anova’ from the package ‘car’ (Fox & Weisberg 2011). Mixed effects models were built with the function ‘lmer’ from the package ‘lme4’ (Bates, Maechler & Bolker 2013). Models were validated by examination of the standardized residuals, testing for heteroscedasticity (Fligner–Killeen test) and normality (Shapiro–Wilk test). Means are presented ± 1 SE. All data are deposited in the Dryad Digital Repository (Simmons & Buzatto 2014).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Data accessibility
  9. References
  10. Supporting Information

Testes Mass

The variance explained by the random effect of ‘chamber’ was extremely low (0·00004), and the model without this effect fit the data significantly better (∆AIC = 13·72). The interaction between morph and social environment was not significant (F1,72 = 0·20, P = 0·656) and so was removed. Examination of the residuals identified three individuals as outliers, all minor males (one from the low risk and two from the high risk treatments). After their removal, Fligner–Killeen test detected no heterogeneity of variances (median χ2 = 75·56, d.f. = 75, = 0·46), and the Shapiro–Wilk normality test detected no deviation from normality (W = 0·99, = 0·81). The final model revealed the expected increase in testes mass with soma mass (F1,73 = 45·39, P < 0·001) (Fig. 1). Minor males tended to have relatively larger testes for their soma mass, although not significantly so (F1,73 = 3·42, P = 0·068), and there was no effect of social environment on relative testes mass (F1,73 = 0·91, P = 0·343) (Fig. 2). Analysis including outliers returned similar results (Appendix S1, Supporting information); however, the residuals deviated from normality (W = 0·922, P < 0·001) so that their inclusion undermines the robustness of the analysis.

image

Figure 1. Testes allometry in the sample of dung beetles Onthophagus taurus reared under different social environments. The regression line is shown for log-transformed data. Untransformed values are provided in parenthesis. Three outliers are not shown (see 'Results').

Download figure to PowerPoint

image

Figure 2. Variation in testes mass after correcting for soma mass (least square means ± 1 SE) among major and minor male Onthophagus taurus reared under different social environments.

Download figure to PowerPoint

Condition

The variance explained by the random effect of ‘chamber’ was extremely low (0·0002), and the model without this effect fit the data significantly better (∆AIC = 14·43). Examination of the residuals identified three individuals as outliers, again, all minor males (two from the low risk and one from the high risk treatments). After their removal, Fligner–Killeen test detected no heterogeneity of variances (median χ2 = 64·18, d.f. = 59, = 0·30), and the Shapiro–Wilk normality test detected no deviation from normality (W = 0·98, = 0·27). After accounting for variation in soma mass due to body size (F1,72 = 308·99, P < 0·001) (Fig. 3), there was a significant effect of social environment on soma mass (F1,72 = 8·77, P = 0·004) and no significant effect of morph (F1,72 = 1·49, P = 0·226). The interaction effect between social environment and morph was also significant (F1,72 = 9·45, P = 0·003). Major males attained higher condition when in high-competition environments, while minor male condition was insensitive to the social environment (Fig. 4). Analysis including outliers returned similar results (Appendix S2); however, again the residuals deviated from normality (W = 0·868, P < 0·001) so that their inclusion undermines the robustness of the analysis.

image

Figure 3. The relationship between soma mass and body size (pronotum width) of dung beetles Onthophagus taurus reared under different social environments. The regression line is shown for log-transformed data. Untransformed values are provided in parenthesis. Three outliers are not shown (see 'Results').

Download figure to PowerPoint

image

Figure 4. Variation in male condition (least square mean ± 1 SE soma mass corrected for body size) among major and minor male Onthophagus taurus reared under different social environments.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Data accessibility
  9. References
  10. Supporting Information

Our manipulations of the pre-reproductive social environment of the dung beetle O. taurus revealed phenotypic plasticity in male allocation to body condition, but no phenotypic plasticity in male allocation to testes mass. Previous studies of the developmental responses made by males to variation in their social environment have found that males increase their expenditure on the ejaculate when exposed to rival males (Gage 1995; Stockley & Seal 2001; Firman, Klemme & Simmons 2013), as predicted by traditional sperm competition theory. However, such responses are not always found (Gay et al. 2009), and the magnitude of male responses to rival males can vary among populations (Yamane & Miyatake 2008; Firman, Klemme & Simmons 2013). New sperm competition models suggest that this variation in response may, at least in part, arise because of the relative importance of pre- and post-copulatory expenditure for male reproductive success among populations or species. When the gains from marginal investment in pre-copulatory contests among males are high, males are expected to increase their investment in pre-copulatory contest competition, potentially at the expense of ejaculate expenditure (Parker, Lessells & Simmons 2013). Our study lends empirical support to this expectation.

Male contest competition is a key feature of the mating system of O. taurus, where male expenditure on body size, weaponry, and strength strongly affects reproductive success (Hunt & Simmons 2001). However, only major males of this species fight for access to females. Accordingly, only major males were found to allocate more to body condition under elevated levels of competition. Minor males specialize on sperm competition by sneaking copulations with guarded females. We found that minor males tended to have relatively larger testes than major males, although not significantly so. Comparative studies across onthophagines have shown that sneaks often invest relatively more in testes mass than guards, but because the frequency of sneaks in Western Australian populations of O. taurus is so high, the asymmetry in sperm competition risk between major and minors is likely to be low, favouring maximal investment in testes mass for both morphs (Simmons, Tomkins & Hunt 1999; Simmons, Emlen & Tomkins 2007). Because minor males are always subject to sperm competition, we did not expect to see strategic adjustments in testes mass by this morph. Our data suggest that major male expenditure on testes mass is also insensitive to variation in the levels of competition, perhaps because major males divert available resources to increased body condition in anticipation of increased levels of pre-mating contest competition, rather than increased testes size (Parker, Lessells & Simmons 2013). Alternatively, it may be that sperm competition is a relatively constant selection pressure in these beetles, so that the evolution and maintenance of phenotypic plasticity in testes growth may not be favoured (Pigliucci 2005; Fusco & Minelli 2010).

Our finding of socially cued phenotypic plasticity in major male expenditure on pre-copulatory contest competition is consistent with a previous study of O. taurus in which females reared in competitive environments produced sons with larger horns for their body size than females reared in isolation (Buzatto, Tomkins & Simmons 2012). Thus, both maternal allocation to their son's adult phenotype and immediate environmental cues experienced by males appear to shape a given male's expenditure on pre-copulatory contest competition to suit his anticipated competitive environment. Phenotypic plasticity in pre-copulatory expenditure is in accord with recent sperm competition theory that incorporates specifically the dyadic contests that characterize mating competition among major males of this species (Parker, Lessells & Simmons 2013). However, Parker, Lessells & Simmons (2013) models predict that expenditure on armaments should come at a cost to expenditure on testes which was clearly not the case in these beetles; major males increased body condition while maintaining their expenditure on testes growth. This suggests that male contest competition and sperm competition may be equally important for male fitness and that increased investment in armament may trade off with some other life-history component such as male longevity. Indeed, several studies have shown how increased male expenditure on pre-and post-copulatory mating competition comes with a direct cost of reduced longevity (Hunt et al. 2004; Robinson et al. 2006; Bretman et al. 2013; Sentinella, Crean & Bonduriansky 2013).

It is interesting to speculate on the mechanism(s) by which males increase their body condition in response to social cues. Previous research has established that phenotypic variation in condition is determined by the availability of fresh dung on which these beetles feed (Kotiaho, Simmons & Tomkins 2001). Fresh dung was not limited in our study, but increased condition may have been achieved by an increase in feeding rate. As caloric intake can increase the production of reactive oxygen species that reduce lifespan (Gredilla & Barja 2005), animals are expected to balance their intake in relation to the demands placed upon them by their current environmental conditions (Simpson & Rubenstein 2013). Alternatively, or in addition, changes in resting and/or active metabolic rates in response to social cues could affect the assimilation of nutrients and their allocation to life-history traits (Kasumovic 2013).

Socially cued plasticity in male body condition and pre-copulatory mating effort is not unprecedented. For example, in mantids, Pseudomantis albofimbriata, social cues to sperm competition risk increase male development time, body condition and ejaculate expenditure (Allen et al. 2011). Male field crickets have been found to exhibit plasticity in their allocation to mating effort cued by the sound of calling rivals. In Teleogryllus commodus, males reared in highly competitive environments take longer to develop and emerge as larger, heavier adults that invest less calling effort over their lifespan than males reared in less competitive environments (Kasumovic et al. 2011). Male Gryllus integer reared in competitive environments had a faster growth rate and were less aggressive as adults than those reared in low-competition environments (DiRienzo, Pruitt & Hedrick 2012). In T. oceanicus, males reared in competitive environments not only emerge in better condition, but also invest more in reproductive tissues (accessory glands and testes) and ejaculate quality (Bailey, Gray & Zuk 2010; Gray & Simmons 2013). In contrast to dung beetles, these studies suggest that for crickets, social cues to future rivals result in a reduction in pre-copulatory mating effort (calling to attract females) and an increased allocation to ejaculate production. Such a pattern is expected where male competition is characterized more by mate search than direct contest competition (Parker, Lessells & Simmons 2013).

The relative responses to social cues in male allocation to pre- and post-copulatory traits will also depend on the patterns of sperm utilization. Sperm competition theory assumes that sperm utilization conforms to a fair raffle, which is not always the case. In redback spiders, Latrodectus hasselti, males mate only once as they are consumed by the female during copulation (Andrade 1996), and fertilization is strongly biased towards the first male to mate (Snow & Andrade 2005; Snow, Abdel-Mesih & Andrade 2006). Males can perceive the availability of females and density of rival males through pheromonal cues in the environment and emerge as smaller adults when they perceive a high density of females or rival males (Kasumovic & Andrade 2006). It is unknown how social cues affect ejaculate allocation in this species. However, early emergence allows smaller males to find females quickly and thereby monopolize paternity (Kasumovic & Andrade 2009).

In general, socially cued anticipatory plasticity is likely to prove a widespread mating strategy that includes more than just ejaculate expenditure (Kasumovic & Brooks 2011). Future studies of strategic male allocation should routinely measure both pre- and post-copulatory traits. By so doing, we will find answers to important questions regarding the potential trade-offs between these traits and their relative contributions to male fitness.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Data accessibility
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Data accessibility
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
fec12211-sup-0001-LaySummary.pdfPDF document49KLaySummary
fec12211-sup-0002-AppendixS1-S2.docxWord document153K

Appendix S1. Testes analysis with outliers included.

Appendix S2. Condition analysis with outliers included.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.