Sexual attractiveness and reproductive performance in ageing females of a coccoid insect

Abstract

Female age strongly influences reproductive success in various animals. Males are, therefore, expected to show preferential responses to sexual signals released from females of ages suitable for reproduction. Here, however, we report an unexpected and contradictory effect of ageing on sexual attractiveness and reproductive performance in a coccoid insect: the pheromone-based attractiveness of females increased with ageing, though their reproductive performance was in rapid decline. Surprisingly, senescent females continued releasing relatively high amounts of pheromone and maintained their sexual attractiveness, even at ages when they had almost completely lost fertility, with reduced densities of endosymbionts to support their physiology. Our dataset suggests a potential sexual conflict within a pheromone communication system, where females benefit at males' expense through deceptive signals of fertility.

1. Introduction

A growing number of studies have demonstrated that parental ageing is a critical risk factor for reproduction in both humans and other animals (e.g. [1–3]). Invertebrate systems, including insects, are emerging models for understanding the mechanisms of underlying reproductive ageing and provide powerful tools for dissecting conserved molecular backgrounds [2,3]. However, the behavioural, ecological or evolutionary aspects of ageing in terms of reproduction remain unclear in most organisms.

Given that ageing of females is associated with reproductive performance, males are expected to be attracted to the sexual signals released by females at the ages suitable for reproduction to increase their reproductive fitness. Actually, sexually mature females often exhibit mechanical or chemical cues that are attractive to males [4]. However, the sexual attractiveness of females at advanced ages, and its relationship to reproduction, have been discussed in only a few studies (e.g. [5–7]), perhaps because, in nature, females mostly mate soon after maturation and cease reproduction in senescence [8].

Here, we found a clear but unexpected relationship between sexual attractiveness and reproductive performance with ageing in a pheromone communication system—a ubiquitous sexual signal in animals [9]—in a coccoid insect (mealybug, Planococcus kraunhiae). Because this insect is a plant-sap feeder and damages a variety of fruits, we recently developed a pheromone-based agrochemical, called a ‘mating disruptant’, that can interfere with its mating and lead to reduction in population densities [10]. This technique deprives the mealybugs of their chance for copulations and consequently generates many senescent females without mating experience in populations, which therefore offers a unique opportunity to examine the influences of ageing on sexual attractiveness and reproductive performance.

Female mealybugs are relatively long-lived and as adults can live for several months before copulation [11], although they die soon after mating and reproduction. We collected and analysed volatiles including a pheromone compound emitted by individual virgin females of various ages, and then examined their attractiveness to males in both laboratory and field bioassays. Moreover, we measured the females' fertility, as well as their body mass. In addition, we investigated the densities of endosymbionts, which are considered to play an essential role in metabolism of mealybugs and reflect their physiological status; mealybugs generally possess a bacteriome consisting of many bacteriocytes, which harbour endosymbionts belonging to the Betaproteobacteria (β-symbionts) and Gammaproteobacteria (γ-symbionts; e.g. [11,12]). The results demonstrated that senescent virgin females continued to release higher amounts of the pheromone than young fertile females and attracted males even after they had lost fertility.

2. Material and methods

(a). Insects

We used mealybugs of a stock colony established from individuals collected in a persimmon orchard (Fukuoka, Japan; 33.3° N, 130.3° E) and maintained in our laboratory for more than 10 years. They were placed individually in a tightly sealed laboratory dish (5.5 cm diameter) in a rearing room (16 L : 8 D; 23°C) and fed with a germinated broad bean plant. Age (in days) of adult females was counted from after the nymphs completed their final (third) ecdysis. To assess body mass, we measured weights of 16 different individuals at each age. To assess fecundity, each female was housed with a freshly emerged male, visually confirmed to copulate only once, and reared until her eggs were deposited (n = 3–28).

(b). Volatile collection and pheromone quantification

To collect volatiles including the pheromone compound emitted by female mealybugs, we placed a female individually into a 20 ml glass vial with food. For 1 day, air that passed through activated charcoal (15 g) was drawn over the female and then passed through 50 mg of an adsorbent (HayeSep Q; Alltech, Deerfield, IL, USA) at 0.1 l min⁻¹ using a vacuum pump. Volatiles were eluted from the adsorbent by using 0.1 ml of hexane to obtain a crude sample; n = 21 and 12 for virgin (1- to 77-day-old) and mated (1- to 14-day-old) females, respectively. The pheromone in the crude sample was quantified with a gas chromatograph (6890N; Agilent, Wilmington, DE, USA), using a calibration curve generated by the synthetic pheromone isolavandulyl butyrate [13]. Details of the analyses were described previously [14].

(c). Bioassays of sexual attractiveness of female volatiles

The behavioural responses of freshly emerged adult males (n = 23–26 per treatment), which were sexually mature, were assayed in a large glass-dish arena (20 cm diameter) under laboratory conditions (23°C; 50% relative humidity). Adult males were released into one side of the arena, and a piece of filter paper (5 mm²; Toyo Roshi Co., Tokyo, Japan), which was impregnated with one female equivalent (FE) of the crude volatile sample prepared from 3- to 70-day-old females as described above, was placed in the opposite side of the arena. The time taken for the male individual to locate and mount the volatile sources was measured as an index of sexual attractiveness. Each observation was performed for 15 min; two to five males per treatment that did not locate the sample were considered unresponsive and excluded from the data.

Sexual attractiveness of the volatiles emitted by virgin females was further investigated by means of a trap experiment in a field persimmon orchard (approx. 1000 m²) at NARO (Ibaraki, Japan; 36.0° N, 140.1° E). Wild males were captured with white delta-traps with a sticky board (12 × 22 cm), baited with 4 FE of the crude volatile sample impregnated into a piece of filter paper (1 cm²). A trap baited with a blank, food (broad bean plant) only, sample was prepared as a control. Traps were placed 1.5 m above the ground at 5 m intervals. From 4 to 10 August 2014, 12 replicates of the assay were performed, randomly changing the trap locations at each count. The volatile samples of the same age were pooled and supplied for the bioassays.

(d). Quantitative polymerase chain reaction and fluorescence in situ hybridization for analyses of symbiont densities

To measure the densities of endosymbionts that support mealybug physiology, we performed quantitative polymerase chain reaction (qPCR) of the groEL gene of symbionts and the ef1α gene of hosts (n = 25–30 per treatment). We also performed whole-mount fluorescence in situ hybridization (FISH) using oligonucleotide probes specific to the 16S rRNA sequences of the symbionts. We followed a previously reported method for these experiments [12].

(e). Statistical analysis

Data were analysed by using generalized linear models (GLMs) to assess the effects of age of adult females, with a Poisson distribution for count data (numbers of male captures and eggs), a binomial distribution for hatch rates or a Gaussian distribution for the other data. Calculations were performed by using the glm function in R v. 3.2.5 software [15]. The p-values in multiple comparisons were adjusted with false discovery rate-controlling procedures.

3. Results and discussion

Sex pheromone emission from virgin females increased rapidly for the first 30 days of the adult stage and then gradually decreased (figure 1a). However, pheromone emission remained at a substantial level for a long period—at least 50 days after the peak: the amount emitted by 70-day-old virgin females (approx. 27 ng d⁻¹) was equivalent to that emitted by 7-day-olds (approx. 22 ng d⁻¹; GLM likelihood ratio tests approximated with a χ² distribution, χ² = 1.45, d.f. = 1, p = 0.228). The pheromone was never released by those females who copulated at the beginning of their adult stage.

Figure 1.

Sexual signalling and male attractiveness in adult female mealybugs with ageing. (a) Pheromone emissions (mean ± s.e.m.) from virgin and mated females. (b) Box plots of time taken for males to locate and mount a volatile source including pheromone released from virgin females in a laboratory bioassay. (c) Number of wild males (mean + s.e.m.) attracted to traps baited with volatiles released from virgin females. Numbers in parentheses show sample sizes. Scores with different letters are significantly different.

In a glass arena under laboratory conditions, males showed significantly faster responses, indicating more positive responses, to the volatiles emitted by the 7- to 70-day-old females that intensively produced the pheromone than to those of 3-day-old females (figure 1b; χ² = 22.8–61.3, d.f. = 1, p < 0.001). No significant differences in response times were found among the volatiles emitted by females older than 7 days (χ² < 0.001, d.f. = 1, p = 0.976). Similarly, in a field trap experiment, significantly more males were attracted to the volatiles emitted by aged females (7- to 70-day-old) than to those emitted by young (3-day-old) females (figure 1c; χ² = 13.3–30.2, d.f. = 1, p < 0.001). A similar male preference for aged females was recently reported in a widow spider, although the specifics of female sexual signalling were not reported [5].

In contrast with these findings, maternal ageing had a negative effect on reproductive performance. More than 60% of virgin females lived for greater than 70 days without substantial loss of body mass (figure 2a), but egg numbers began to decrease significantly as early as 7 days (χ² = 67.4, d.f. = 1, p < 0.001) and had dropped to 43% in females aged 28 days and further to 8.3% in 70 days, with a corresponding decline of hatchability (figure 2b). Both β- and γ-endosymbiont densities in the females’ bacteriomes also decreased with ageing (figure 2c; χ² = 8.77–18.8, d.f. = 1, p < 0.01), indicating that there was a reduction in physiological support by these symbionts. The influence of maternal age on reproduction has also been reported in another mealybug, Planococcus citri [16].

Figure 2.

Physiological and reproductive performance of adult virgin females with ageing. (a) Body mass (mean ± s.e.m.) and survival rate. (b) Number of eggs (mean + s.e.m.), with their hatch rate (mean% ± s.e.m.), produced by females copulated at different ages. (c) Relative abundance (mean ± s.e.m.) of β- and γ-endosymbionts in females, as measured by qPCR. Bacteriomes harbouring symbionts of 7- and 70-day-old virgin females were visualized by FISH and are shown with red, green and blue indicating signals of β-symbiont, γ-symbiont and host nuclei, respectively. Numbers in parentheses show sample size. Scores with asterisks (*) are significantly decreased compared with those of females at the initial age.

Thus, our data clearly show the paradoxical effects of ageing on pheromone-based sexual attractiveness and reproductive performance. In theory, females increase sexual signalling effort when they remain unmated in response to elevating risk of reproductive failure [6,7]. Our dataset suggests a potential sexual conflict within a pheromone communication system, where females benefit at males' expense through deceptive signals of fertility. Alternatively, the cost of responding to the signals from older, less fertile females may not be as severe as the risk of being too discriminatory, and failing to respond to a receptive female. Particularly, in coccoid insects, males are dwarfed and fragile, with retrogressed mouth; handicapped by limited energy and a very short lifespan they must find the almost immobile females, which lack wings and often even legs [11,14,17].

Data accessibility

Data available from Dryad http://dx.doi.org/10.5061/dryad.406248v [18].

Authors' contributions

J.T. conceived and designed the study, performed the chemical/behavioural experiments and wrote the manuscript; M.T. participated in the design of the study, performed the physiological experiments and helped draft the manuscript. Both authors agreed to be held accountable for all aspects of the study and gave final approval for publication.

Competing interests

We declare we have no competing interests.

Funding

J.T. and M.T. were funded by grants-in-aid for scientific research from JSPS (nos 16K08103 and 26450476, respectively).