Fitness costs associated with chemical signaling

Abstract

The production, maintenance and transmission of chemical signals often entail costs. Costs can arise, for instance, if signal production depends on the availability of limited nutritional resources or if signal transmission leads to attraction of predators. Many species effectively reduce these costs by signaling at specific times or in certain contexts only. We previously reported that breeding burying beetle females facultatively adjust their pheromone emission in response to their social environment, emitting high amounts of their chemical signal in the presence of a male partner, but not when providing uniparental care. Here we present data showing that chemical signaling is costly, and that higher investments in signaling result in reduced clutch sizes, but not a shorter life span, in the burying beetle Nicrophorus vespilloides.

Keywords: :

• burying beetle
• chemical communication
• efficiency costs
• Nicrophorus
• strategic costs

This article refers to:

The study of animal communication, a central focus of behavioral ecology, almost always involves some consideration of signal costs. This is largely because it is widely held that the maintenance of honest communication requires that signaling must incur ‘strategic’ costs.¹ Although Zahavi’s handicap principle is still the most widely accepted explanation for signal reliability,^2-5 we have since learned that strategic costs are not the only force promoting signal evolution and maintenance, but rather, that there are other mechanisms that can enforce the honesty of a signal.^1,6,7 One example is indices, signals whose intensity is causally related to the quality being signaled, and which, due to physical or chemical constraints, cannot be faked.¹ However, even when a signal does not involve a strategic cost, it is not necessarily cost free to the signaler. Each signal will incur ‘efficacy’ costs, costs that the signaler has to pay to efficiently transmit the signal to the receiver.^1,8 Efficacy costs are likely to be affected by background noise, as signal intensity must be increased the more the environment becomes visually, acoustically or chemically cluttered.

Although the costs of chemical signaling are less well studied than those of other signaling modalities, and the distinction between strategic and efficacy costs has rarely been made for chemical signals, studies have found, not surprisingly, that chemical signals also incur costs.^9-13 Indirect evidence for costs of pheromone or allomone emission has already accumulated over a period of many years. Signalers in a wide array of species do not emit their chemical signals continuously, but rather, modulate their emission depending on the time,^14-16 location¹⁷ or social environment.^18,19 Such adjustment of signal quantity to communication context not only may improve the efficiency of information transfer due to better synchronization of sender and receiver, but likely reduce the physiological or ecological (i.e., attraction of parasites or predators) costs of signal production or transmission.^20,21 One example comes from a study of the fruit fly, Drosophila grimshawi, where males modulate pheromone production depending on the presence of a rival male.¹⁰ Males for whom pheromone deposition was artificially increased due to prolonged exposure to rival males suffered from a shorter life span than solitary males.

In a recent study, we found that female burying beetles Nicrophorus vespilloides are able to adjust their pheromone emission in relation to their social environment.²² Burying beetles are well known for their elaborate parental care.²³ During the period of parental care, female beetles communicate their breeding state via a chemical signal, methyl geranate,²⁴ but they only do so when breeding with a male partner (i.e., in the present of a receiver), but not when breeding alone.²² In the case of biparental care, the signal is important for the success of the breeding attempt because it helps the male to distinguish between his female partner and a non-breeding female intruder.^24,25 Any intruder, if not repelled, would almost certainly kill the entire brood.²⁶

Pheromone modulation allows burying beetle females to economize on signal costs, but the costs of methyl geranate production or transmission remain unclear. Ecological costs (i.e., the attraction of predators or parasitoids²⁰) are unlikely because the beetles are below ground in a crypt during breeding and signal emission. It is more likely that signal production results in metabolic costs. In addition, because the molecular structure of the chemical substance suggests that methyl geranate is an index,²⁴ the type of costs involved are likely efficacy costs, not strategic costs. In the present study, we evaluated the costs of chemical signaling by comparing fitness-relevant parameters (specifically, clutch size and longevity) of females breeding with a male partner (i.e., females that emit a high amount of methyl geranate) and solitary breeding females (i.e., females that emit a marginal amount of methyl geranate). Pairs of beetles (n = 15) and single females (n = 15) were permitted to breed in four consecutive reproductive bouts on 15-g carcasses. After 48 h, each female or pair was transferred to a new box along with its carcass. In the previous containers that contained the eggs, we placed small pieces of mouse carrion. Because larvae that hatch in the soil are attracted to carrion, this procedure enabled us to determine the number of hatched larvae and thus, the number of viable eggs laid. The boxes were checked for the presence of newly hatched larvae every 12 h. Once we observed larvae, we again transferred the beetles to a new box and placed 30 larvae on the carcass with the respective single or paired female. The old box was again provided with a small piece of carrion to collect newly hatched larvae. Once females are caring for larvae, they cease laying any additional eggs;²⁷ hence, clutch size was defined as the sum of eggs laid until the first larva hatched, specifically, the number of larvae that hatched from these eggs. To standardize the period of parental care and therefore the length of pheromone emission, pairs and singles were left to care for the larvae for a defined period of 72 h. After this time, the beetles were separated from the carcass and transferred to a new box. After being given a rest period of three days, beetles in both treatments were provided with a new carcass, and the cycle was repeated.

Twenty-four hours before each reproductive bout, single females were kept together with a male for one day to ensure their insemination. Hence, each single female was kept with a male partner for four days in their lifetime.

After the four reproductive bouts, females of both treatment groups were kept singly in small plastic containers filled with moist peat and fed mealworms twice a week until they died. In addition to the two treatment groups, we established a third group of beetles (n = 20) to investigate the effect of reproduction and parental care on longevity. In this group, females were kept singly throughout their lifetime and were never allowed to reproduce.

The clutch size of the beetles depended on reproductive bout (repeated measures ANOVA, F_{3, 75} = 20.71, p < 0.001; Figure 1). The clutch size in the first reproductive bout was smaller than those of the remaining three reproductive bouts (reproductive bout 1 vs. 2, 3, or 4, all p < 0.05). The family status of the females also had an effect on the number of eggs laid, with single females having significantly larger clutch sizes than paired females (F_{1, 25} = 7.91, p < 0.01; Figure 1). There was a significant interaction between reproductive bout and treatment in their effect on clutch size (F_{3, 75} = 5.28, p < 0.01). There was no difference in the number of eggs laid by single and paired females in the first two reproductive bouts (p > 0.05), but single females laid more eggs than females reproducing with a male partner in each of the last two reproductive bouts (each p < 0.05).

Figure 1. Clutch sizes (mean ± SE) of paired and single females in four subsequent reproductive bouts. Clutch size was determined by the number of hatched larvae.

There was no difference in survival between single breeding females and paired breeding females (Fig. 2). Moreover, reproduction and parental care had no effect on survival, as females reproducing four times on a carcass did not die sooner than females prevented from reproducing (Fig. 2).

Figure 2. Cumulative survival for females that had never bred on a carcass (n = 20), bred four times without (n = 15) or with a male partner (n = 15). There was no difference in survival rates between the three treatment groups (log rank test, ${\mathrm{\chi }}_2^2$ = 3.16, p = 0.206).

In contrast to the study by Johansson et al.,¹⁰ we did not find that a higher investment in pheromone production resulted in shorter life spans. However, our results suggest a reproductive cost of chemical signaling in burying beetles: paired females produced fewer eggs in subsequent reproductive bouts than single females. A reduced fecundity as a result of increased pheromone release was also found in female moths Lobesia botrana.²⁸ In the case of N. vespilloides females, a possible explanation might be that the production of pheromones and the production of hormones regulating egg production have to compete for the same precursors. In a number of species, juvenile hormone (JH) is required for vitellogenesis and oviposition,²⁹ and burying beetles are characterized by a large and rapid increase in JH III titer shortly after the discovery of a carcass before oviposition occurs.³⁰ Although direct evidence for a gonadotropic role of JH III in burying beetles is lacking, application of JH III to food-deprived N. orbicollis results in an increased clutch size.³¹ The pheromone methyl geranate and the hormone JH III are both terpenoids and hence, utilize the same precursors.³² The higher amount of methyl geranate released by paired females might have resulted in a shortfall of precursors for JH III at the beginning of a subsequent reproductive bout, leading to a reduced clutch size in comparison to single females. However, although this explanation seems plausible, there is an alternative interpretation for our results that cannot be excluded: the difference between our two treatment groups, paired and single females, is not only the amount of methyl geranate emission, but also the presence or absence of a male partner. We know from many studies that numerous copulations can result in costs to females.^33-36 Hence, a reduction in clutch size could be caused by a sexual conflict over mating frequency. Disentangling the costs of mating and the costs of pheromone emission on reproductive output must, however, await future study. We recommend that future investigations focus on the metabolic pathways of signal production as this would aid in identifying the exact nature of the strategic or efficacy costs of chemical signals.

Acknowledgments

We thank Scott K. Sakaluk for helpful comments on the manuscript. S.S. was supported by a Feodor Lynen Return Fellowship from the Alexander von Humboldt Foundation.