Micropyle number is associated with elevated female promiscuity in Lepidoptera

Abstract

In the majority of insects, sperm fertilize the egg via a narrow canal through the outer chorion called the micropyle. Despite having this one primary function, there is considerable unexplained variation in the location, arrangement and number of micropyles within and between species. Here, we examined the relationship between micropyle number and female mating pattern through a comparative analysis across Lepidoptera. Three functional hypotheses could explain profound micropylar variation: (i) increasing micropyle number reduces the risk of infertility through sperm limitation in species that mate infrequently; (ii) decreasing micropyle number reduces the risk of pathological polyspermy in species that mate more frequently; and (iii) increasing micropyle number allows females to exert greater control over fertilization within the context of post-copulatory sexual selection, which will be more intense in promiscuous species. Micropyle number was positively related to the degree of female promiscuity as measured by spermatophore count, regardless of phylogenetic signal, supporting the hypothesis that micropyle number is shaped by post-copulatory sexual selection. We discuss this finding in the context of cryptic female choice, sperm limitation and physiological polyspermy.

1. Introduction

Micropyles (from the Greek mikros, small, pulé, gate) are small openings that allow male gametes to enter and fertilize the ovum in a wide diversity of taxa, including insects, fishes, cephalopods and plants [1]. Among insect orders, micropyles exhibit considerable variation in position, arrangement and number. For example in some species, such as Drosophila spp. [2], micropyles protrude from the egg chorion on ‘stalks’ (micropylar processes), whereas others are located in micropylar pits as in some Lepidoptera (e.g. [3]), while others are superficial [1,2]. Within the Heteroptera, variation in micropyle number is extensive: 0–70 [4], while in the Lepidoptera there are between 1 and 20 [1], with some evidence of intraspecific variation [5,6]. Despite such large and obvious differences between species, few authors have attempted to seek functional explanations for this variation.

Here, we use a comparative approach to investigate variation in micropyle number, testing between three hypotheses associated with female mating pattern. If micropyles only act to facilitate fertilization success, we predict more micropyles in those species at greater risk of fertilization failure due to sperm limitation. This could occur in populations that have a strongly female-biased operational sex ratio [7] and/or in populations in which females mate infrequently. For example, female Drosophila pseudoobscura that copulate only once appear to have insufficient viable sperm stores to maintain fertility [8]. Hence, if greater micropyle number increases fertilization success we predict a negative association with the likelihood of female promiscuity. Similarly, if micropyle number functions to mitigate against pathological polyspermy (embryonic failure due to more than one sperm entering the oocyte cytoplasm [9]), then we predict a negative association between micropyle number and female mating frequency, such that promiscuous species at greater risk of pathological polyspermy have fewer micropyles. (It should be noted that our approach cannot distinguish between these two hypotheses.) By contrast, if micropyles are shaped by post-copulatory sexual selection, then greater micropyle number is predicted to be positively associated with promiscuous mating patterns. In species where ejaculates from more than one male compete to fertilize a female's set of eggs cryptic female choice can operate to bias fertilization success in favour of particular male traits [10]. In species with internal fertilization, this can manifest itself as a number of male–female interactions [11] including those at the sperm–egg interface [12]. Thus, it is possible that variation in micropyle number could be driven by post-copulatory sexual selection if females are able to use these structures to exert control over fertilizations. However, to our knowledge, no studies have yet examined this novel hypothesis.

We therefore compare micropyle number against the extent of female promiscuity, using lepidopteran species that vary greatly in both female mating pattern and egg micropyle number. The Lepidoptera are especially suitable for this study because mating pattern can be quantified from spermatophore counts which persist within the female bursa copulatrix [13]. We hypothesize that variation in micropyle number functions to: (i) reduce the risk of sperm limitation and egg infertility, (ii) reduce the risk of pathological polyspermy or (iii) allow greater control over paternity.

2. Material and methods

(a). Data collation

Species-specific average micropyle number and spermatophore count (number of spermatophores recovered from the bursa copulatrix) were collated from the literature alongside egg size (a potential covariate; [14]) for 56 species of Lepidoptera from 15 families (25 butterflies and 31 moths). Lepidopteran eggs fall broadly into two shapes: fusiform (butterflies) and flat/round (moths). Thus, volume was approximated using the formula for a prolate ellipsoid ((½ egg length × ½ egg width² × π) 4/3) for butterfly eggs and half-oblate spheroids for the moth eggs (((½ egg length × ½ egg width² × π) 4/3)/2). In particular, estimates of species-level promiscuity were gained primarily from field studies (as opposed to laboratory-based studies) which reported spermatophore count (for a discussion on using this method, see [13]).

(b). Statistical analyses

We used a phylogenetic generalized least-squares regression (PGLM) [15,16] between mean micropyle number and spermatophore count. The pglmEstLambda function of the ‘CAIC’ package was used to identify the maximum-likelihood value of λ [15,17,18], which measures the degree to which the matrix follows a Brownian model; λ can vary between 0 (no phylogenetic autocorrelation) and 1 (complete phylogenetic autocorrelation). We present results from the PGLM along with the ordinary least squares for comparison [19], where λ = 0, the resulting model is equivalent to a standard linear model. Analysis was carried out using R code kindly provided by R. P. Freckleton (University of Sheffield). We used butterfly phylogenies available on the Tree of Life Web Project [20] with branch lengths set to one. All analyses were run in R version 2.15.2 [21].

3. Results and discussion

Species-specific micropyle number varied from 1 to 15 (mean 4.06 ± s.e.m. 0.43) across the Lepidoptera sampled (table 1). We found no evidence that micropyle number was associated with risk of sperm limitation and infertility, or that fewer micropyles were associated with a likely increased risk of polyspermy. Rather, micropyle number was positively correlated with our estimate of female promiscuity (table 2). Micropyle number was positively related to spermatophore count in corrected and non-corrected PGLS (table 2 and figure 1).

Table 1.

Average micropyle number across lepidopteran families.

family	number of species	average micropyle number
Arctiidae	1	4 (4–6)
Erebidae	1	2
Gelechiidae	1	3
Heliothinae	1	3 (3,4)
Lycaenidae	2	3.5
Noctuidae	14	5.2
Notodontidae	2	10.5
Nymphalidae	18	3.6
Papillonidae	1	7
Pieridae	2	1.5
Pyralidae	4	2
Saturniidae	1	7
Sphingidae	1	1
Tortricidae	2	1

Table 2.

Results of PGLM model (N = 39) for the relationship between micropyle number and spermatophore count. For each model, the β ± s.e., t- and p-values are presented. In addition, Pagel's λ [15] is calculated.

	non-phylogenetically corrected			phylogenetically corrected
parameter	β ± s.e.	t	p	β ± s.e.	λ	t	p
intercept	−0.08 ± 1.19	−0.07	0.946	1.09 ± 1.54	1.00	0.70	0.485
egg volume	1.86 ± 1.60	1.16	0.252	1.14 ± 1.49		1.76	0.449
spermatophore count	2.30 ± 0.55	4.20	<0.001	1.09 ± 0.45		2.40	0.022

Figure 1.

The relationship between micropyle number and spermatophore count in corrected (solid line) and non-corrected (dashed line) PGLS.

At a functional level, more micropyles would suggest greater potential for multiple sperm entry into the egg. This raises two questions: (i) why allow multiple sperm to enter the egg? And (ii) why make this easier in species with greater female promiscuity? Physiological polyspermy is widespread in nature, being the norm in Urodeles and birds but also reported in other taxa [9,22] including Lepidoptera [23]. In physiological polyspermy, several sperm enter the egg but only one fuses with the female pronucleus. The remaining supernumerary sperm nuclei degenerate [2]. Why physiological polyspermy occurs only in some taxa is unclear, although a recent study by Hemmings & Birkhead [24] indicates that polyspermy is essential for early embryonic development in both domestic fowl and the zebra finch (Taeniopygia guttata). Physiological polyspermy enables the intriguing possibility of mate choice within an egg cell [25]. In the polyspermic ctenophore (Beroe ovata), the female pronucleus migrates among male pronuclei within the egg before fusing with one [26]. Thus, the presence of multiple micropyles could increase the opportunity for post-copulatory female choice within the egg environment. Such mechanisms are likely to be most relevant for polyandrous species where selection has acted on mating pattern to increase the opportunity for sperm choice.

Alternatively, more micropyles may represent a bet-hedging strategy for the female where sperm numbers are limited. Although sperm are cheaper to produce than eggs, they still involve a reproductive cost. When sperm competition is high, males can allocate their ejaculates prudently, resulting in sperm limitation for females [27]. Thus, the presence of a greater number of micropyles may represent an evolved mechanism to counter male traits that incidentally lower female fitness.

Lastly, the micropyle opening is only one component of micropyles. The micropylar openings lead to canaliculi, minute ducts through the chorion. In some species, these canals show complex structuring; for example, Bombyx mori have a single external micropyle which branches to three to five canaliculi that lead to the chorion [28]. Given that the number of sperm entering eggs in B. mori varies from 1 to 11, Kawaguchi et al. [6] proposed that the number of canals is related to the degree of polyspermy. Such diversity in internal structuring of the micropyles suggests a greater degree of complexity to their function than has been considered previously and a possible role in polyspermy in insects.

4. Conclusion

This is the first study to show that micropylar variation is in part driven by the degree of female promiscuity. Micropyles allow sperm entry into the egg; hence more micropyles should aid sperm entry in to the egg, reducing the likelihood of infertility, while at the same time increasing the likelihood of physiological polyspermy. Whether physiological polyspermy benefits early embryogenesis in insects as it appears to do in birds [24] and/or offers an alternative site for cryptic female choice [25] requires further study.

Ethics

Ethical approval was granted from University of Lincoln's College of Science Ethics Committee (COSREC37).

Data accessibility

Data are available on the Dryad Digital Repository: http://dx.doi.org/10.5061/dryad.2c5m1 [29].

Authors' contributions

G.I. and P.E.E. designed the study; G.I. collated and analysed the data; G.I. and P.E.E. drafted the manuscript. M.J.G.G. shared his butterfly sperm dataset, contributed to the study design and helped draft the manuscript. All authors gave final approval for publication and agreed to be accountable for all aspects of the content therein.

Competing interests

We have no competing interests.

Funding

The study was funded by the University of Lincoln Back to Science Fellowship (G.I.).