Ontogeny of sexual dimorphism via tissue duplication in an ostracod (Crustacea)

EVOLUTION & DEVELOPMENT

11:2, 233 –243 (2009)

DOI: 10.1111/j.1525-142X.2009.00323.x

Ontogeny of sexual dimorphism via tissue duplication in an ostracod (Crustacea) Ajna S. Rivera,a, and Todd H. Oakleyb a

Marine Science Institute, University of California at Santa Barbara, Santa Barbara, CA 93106, USA Department of Ecology Evolution and Marine Biology, University of California at Santa Barbara, Santa Barbara, CA 93106, USA b

Author for correspondence (email: [email protected])

SUMMARY The adaptive significance of specific sexual dimorphism is well studied. However, the evolutionary history and ontogenic origins of the dimorphism are often unknown. As dimorphism represents two phenotypes generated from relatively similar genotypes, it is of interest to understand both its evolutionary and developmental/genetic underpinnings. Here, we present the first ontogenetic examination of the eyes of philomedid ostracods (Crustacea), which exhibit extremely sexually dimorphic lateral eyes. Adult male philomedids have large compound lateral eyes, whereas females have rudimentary lateral eyes. First, we show that eye dimorphism is unlikely to be due to additional genes present on a male-specific chromosome because karyotype analysis suggests philomedids are XX/XO. We then examine the ontogeny of eye development and find that in at least two

species of Euphilomedes, this dimorphism is not generated solely by differences in tissue growth rates, as has been commonly shown for sexually dimorphic characters of other species. Instead, the dimorphism appears to arise during development via tissue duplication, where a single tissue becomes two, perhaps with different developmental potentials. The second eye field is only observed in male Euphilomedes, producing most of the adult eye tissue. We point out that tissue duplication is a developmental process with evolutionary implications because novel characters could evolve via alternative modification of the duplicated fields, analogous to the origin of new genes by gene duplication and alternative modification. Depending on the evolutionary history of the duplicated field, it may have either facilitated or directly caused the observed sexual dimorphism of philomedid ostracods.

INTRODUCTION

tree bark, suggesting resource partitioning between sexes (Ligon 1968). Other examples of resource partitioning driving sexual dimorphism include bill differences in hummingbirds (Temeles et al. 2000) and size differences in sea snakes (Shine et al. 2002). While the adaptive cause of these traits is apparent, the developmental mechanisms producing them are largely unknown. There are several developmental mechanisms by which sexual dimorphism can arise. Generally, a single bipotential tissue responds differentially to male and female growth signals. An organ can develop different features in males and females, for example pigmentation or ornamentation patterns (Kopp et al. 2000; Jawor and Breitwisch 2003). An extreme example of this is the mammalian gonad which begins as a bipotential tissue and then differentiates based on downstream sex-chromosome genes (Estrada et al. 2003; Brennan and Capel 2004). A bipotential tissue can also be differentially regulated in terms of tissue growth or remodeling; in horned beetles cell proliferation and female-specific retraction lead to the dimorphic expression of horns (Moczek and Nagy 2005; Moczek, Cruickshank et al. 2006; Moczek et al. 2006).

Sexual dimorphism is widespread throughout Bilateria, including both primary reproductive differences in sex structures (i.e. gonads and genitalia) and secondary sexual characteristics such as size, behavior, and decoration. Of interest to biologists are both evolutionary mechanisms (why sexual dimorphism exists) and developmental mechanisms (how sexual dimorphism is manifested). An evolutionary mechanism commonly leading to sexual dimorphism is sexual selection (Andersson 1994). For example, in snakes, males of species that undergo combat for mate choice are generally larger than females, while in noncombative species, the females are typically larger (Shine 1978). Other examples of overtly sexually selected dimorphic characters include plumage in birds (Dunn et al. 2001) and the complex mating dances of Drosophila (Billeter et al. 2006). But sexual selection is not the only mechanism driving sexual dimorphism. Males and females of Dendrocopus woodpecker species have different beak shapes and also occupy separate environmental niches, eating insects from different layers of & 2009 The Author(s) Journal compilation & 2009 Wiley Periodicals, Inc.

233

234

EVOLUTION & DEVELOPMENT

Vol. 11, No. 2, March--April 2009

Fig. 1. (A, B) Lateral view of adult animals with half of carapace removed. Anterior is to the left. One lateral eye is visible (arrows). The female eye (A) is small and rudimentary without ommatidia and is only faintly pigmented. The male eye (B) is much larger, dominating the head of the animal with clearly visible ommatidia and a small rudiment (arrowhead) (C, D, E). Lateral view of instar IV male eyes from Euphilomedes longiseta (C) and Euphilomedes morini (D) and instar V Euphilomedes carcharodonta. Arrowheads denote rudiments. Stage matched lateral eyes from these three species are indistinguishable. Panel (E) depicts the development of a male compound eye. Pigment was photobleached under UV to allow a better view of the morphology. All rudiment pigmentation was removed in the photobleaching. The largest lenses and darkest pigmentation are at the distal end of the ommatidial field. Growth appears to occur in the direction of the curved arrow, as suggested by the presence of smaller lenses and fainter pigmentation. A putative morphogenic front is marked with a red arrowhead. (F, G) Eyes of late stage embryos (carapace fully formed) of the cylidroleberid Postasterope barnesi (F) and E. carcharodonta (G). Ommatidia are forming in the cylindroleberid embryo (arrows) but are absent in E. carcharodonta. (H, I) Ommatidial structure of E. morini. The schematic is based on panel (H), DAPI nuclear staining, and previous data on ostracod eyes (Land and Nilsson 1990). The distal portion of the ommatidia is the two lenses, which are clear and highly autofluorescent. These overlie two pigment cells (P) and the crystalline cone cells (C). Three retinular cells (R) were clearly visible, but six to eight have been reported in previous literature (Huvard 1990; Land and Nilsson 1990). A singe cone cell nucleus (C) was visible; the other may have been lost during preparation. Scale bar: (A, B) 500 mm, (C, D, F) 40 mm, (F) 55 mm (G, H, I) 15 mm. DAPI, 40 ,6-diamidino-2-phenylindole.

While these two mechanisms for producing dimorphism are the most commonly cited, there is at least one other developmental mechanism that could produce dimorphism, tissue duplication. In contrast to the single dimorphic tissues that arise in the above examples, a duplicated tissue can form two distinct organs. Tissue duplications can arise during development by ectopic expression (heterotopy) or by the fission of a homogeneous field of cells (field splitting). These duplicated tissues can then follow different developmental trajectories. While tissue duplication is not generally associated with sexual dimorphism, if the process were to occur only in one sex, a dimorphic character would be the obvious result. Tissue duplication is already known as a developmental mechanism. One example is seen in the eye development of Drosophila and vertebrates. In these animals, the eye anlage begins as a single field that splits into two fields, which then migrate to lateral positions (Green et al. 1993; Li et al. 1997). Tissue duplication as a developmental mechanism for producing sexual dimorphism has not yet been described, as sexual dimorphism is more often studied in terms of size differences, rather than in terms of presence or absence of characters (Badyaev 2002).

In order to consider the different ontogenic mechanisms of sexual dimorphism, we must study species with differences in the presence/absence of characters in males and females. Here we will examine such animals, representatives from the philomedid ostracod genus Euphilomedes. Many philomedid ostracods show dramatic sexual dimorphism with regards to eye morphology (Kornicker 1992). Males generally have multifaceted compound eyes composed of large and obvious ommatidia. Females of dimorphic species, however, have no lateral eyes or only rudimentary lateral eyes (Fig. 1, A and B). This dimorphism is thought to be related to the difference in lifestyle between the two sexes. Males have pelagic phases where they swim in the water column, presumably to forage or for mating purposes, while females and juvenile males primarily burrow in the sand (Fage 1933). These differences in eye morphology could be directly due to genetic differences in males and females resulting from heterogamous male sex chromosomes (e.g., XX/XY). If this were the case, genes on the male-only chromosome could contribute to eye morphology. On the other hand, if males are XO, there are no male-specific genes. In this case, all sexual dimorphism would be under the control of higher-level sex

Rivera and Oakley determination pathway genes. Changes in gene expression levels, rather than male-specific genes, would then be the likely molecular causes of eye dimorphism. This scenario would be particularly interesting as presence or absence of eyes must then occur in extremely similar genetic backgrounds. A few karyotypes of myodocopid ostracods have been determined previously. Karyotypes from 12 cypridinids have been examined including members of the genera Gigantocypris, Vargula, and Conchoecia. All of these species have XX/XO sex determination, but no one has published karyotype data for any philomedid (Moguilevsky 1985; Moguilevsky and Whatley, 1988; Moguilevsky, 1990, 1992, 1995). Euphilomedes are generally subtidal philomedid ostracods, commonly found along the Eastern Pacific coast of North America (Smith 1952; Poulson 1962; Kornicker and Harrison-Nelson 1997; Bergen et al. 2001). While Kornicker and Harrison-Nelson (1997) did not observe any lateral eyes in females, we observe small rudimentary eyes in females. These eyes are small compared with the male eyes of corresponding stages and are relatively lightly pigmented when viewed alive or preserved in glycerol. This pigment washes out upon fixation in ethanol, the traditional method of preservation, leading to a potential for incorrect diagnosis of eye absence in specimens so treated. Ostracods of the genus Euphilomedes undergo five larval stages before their final molt to become sexually mature adults (Hiruta 1980). All of these stages generally resemble the adult, with a full complement of limbs and major features (Hiruta 1980). A notable exception to this is the development of the lateral compound eyes (Kornicker and HarrisonNelson 1997). Unlike other ostracods whose eyes develop as embryos (Cohen 1983), the eyes of male Euphilomedes are without obvious ommatidia until late in juvenile ontogeny. Kornicker and Harrison-Nelson (1997) found that juvenile male Euphilomedes do have small lateral eyes by the third instar, but the cells in these are indistinct. Kornicker and Harrison-Nelson’s (1997) description of the eyes of juvenile Euphilomedes leads to the conclusion that ommatidial development in males occurs primarily between instar V and the adult. We propose three nonexclusive ontogenic hypotheses to account for this and the observed differences between male and female eyes: (1) Female and male eyes are inherently different and are formed by developmentally distinct pathways from the same tissue (bipotential differentiation). (2) Female eyes undergo less growth than male eyes (bipotential growth). (3) Male and female eyes are terminally arrested. Male eyes are able to escape this fate by making a second eye-field via tissue duplication, which will develop into the ommatidial field. To test these three possibilities, we examined eyes from male and female ostracods at all stages from embryo to adult. Comparisons were made based on gross morphology, pigmentation patterns, and tissue organization. While both sexes

Ontogeny of sexual dimorphism

235

begin with a single eye field, later male eye development occurs via two fields, wherein one field produces the ommatidia and the other a rudimentary structure. This second, ommatidial, field is never observed in females who have only one field that may be homologous to the male rudimentary field. This suggests that Euphilomedes dimorphism develops via a field duplication mechanism unique to males, a novel mechanism for producing a sexually dimorphic character.

METHODS Collection Our original studies were conducted on Euphilomedes carcharodonta, a species found abundantly off Pillar Point in Half Moon Bay, CA (Kornicker and Harrison-Nelson 1997). We subsequently found near our laboratory in Goleta, CA a large population of Euphilomedes morini, a species very similar to E. morini. Because of the similarity of the species and the relative ease of collection of E. morini, we used E. morini to supplement our earlier studies on E. carcharodonta. We also include in our examinations, a juvenile male specimen of Euphilomedes longiseta which is another sexually dimorphic Euphilomedes species that we have found only rarely. All pictures are of E. carcharodonta unless otherwise noted. We collected E. carcharodonta, E. morini, and Postasterope barnesi in California off Pillar Point in Half Moon Bay at 37129.6 0 N; 122129.40 W, off Coal Oil Point in Goleta at 34124.20 N; 119152.4 0 W, and from the Goleta and Santa Barbara piers at 34124.6 0 N; 119149.4 0 W and 34124.40 N; 119140.5 0 W. E. longiseta were collected from the Ocean Beach pier in San Diego, California at 32145.0 0 N; 117115.30 W. Pillar Point ostracods were collected by hand-netting sediment past the tidal zone. Coal Oil Point ostracods were collected by epibenthic sled at approximately 20 m depth. All pier collections were done using an Eckman grab (Wildco, Buffalo, NY, USA). All E. carcharodonta shown were collected from Pillar Point except the instar V female in Fig. 2, C, which was found off Coal Oil Point. All E. morini shown were collected from the Goleta and Santa Barbara piers. The P. barnesi shown was collected from the Goleta pier.

Staging and sex determination Instar stages were determined primarily by furcal claw number as described previously (Kornicker and Harrison-Nelson 1997). To distinguish between stages with the same numbers of claws, we checked carapace size and shape, presence of large eyes (diagnostic for adult males), and presence of eggs or embryos (diagnostic for adult females), as well as examining the number of bells on the seventh limb bristles (Kornicker and Harrison-Nelson 1997). We distinguished sexes of juveniles by the morphology of the endopodite of the second antennae: the endopodite of males has three joints or, in instars II and III, three regions separated by constrictions, while that of females has two joints and ends in a long terminal bristle (ibid).

Karyotyping Karyotyping was performed by modifying the Grozeva and Nokkala Giemsa staining protocol (Grozeva and Nokkala 1996).

236

EVOLUTION & DEVELOPMENT

Vol. 11, No. 2, March--April 2009 The head, gut, and furca were dissected off male E. morini and bathed in a 0.01% colchicine solution for 2 h. These were transferred to a hypotonic sodium citrate solution for 1 h then fixed quickly in 3:3:4 water:acetic acid:ethanol on a clean slide. The tissue was pulled apart with sharp needles. A second fix of 1:1 ethanol:acetic acid was added directly to the slide immediately followed by a rinse in acetic acid. Slides were dried overnight and stained with 1:10 Geimsa stain before being mounted on the slide with Permount. Karyotypes were examined with a  100 oil immersion lens.

Dissection and imaging Ostracods were dissected in 100% glycerol using tungsten needles. Images were taken under brightfield on an Olympus BX61microscope with an Olympus Microfire digital camera (Melville, NY, USA) using PictureFrame software. For further analysis, we squashed eyes in glycerol or Fluoromount (Electron Microscopy Sciences, Hatfield, PA, USA) under a glass coverslip and examined at  40. Some male eyes (Fig. 1, C and D) were mounted under a raised coverslip (using silicone glue to hold the coverslip over the specimen) to facilitate higher magnification without tissue squashing. Sectioning was done on a Leica CM1900 cryostat (Solms, Germany) using manufacturer’s recommendations. Sections were taken at 8–15 mm thickness. To assess the cellular composition of the tissues, 4 0 ,6-diamidino-2-phenylindole (DAPI, Invitrogen, San Diego, CA, USA) was added to Fluoromount at a final concentration of 43.6 mM (20 mg/ml). Final images represent montages of two to six focal planes combined with Adobe Photoshop software and/or with Helicon software (Helicon Soft, Kharkov, Ukraine).

Statistics

Fig. 2. Side views of juvenile and adult lateral eyes with females in the left column and males in the right, panels are not to scale. Large images are tissue squashes and insets are wholemount. Each inset is from the same animal as each tissue squash except (H). Stages progress from youngest at the top to oldest at the bottom. Arrows mark the rudiment in adult males (J) and the prospective rudiment in juvenile males (F, G, H). At instar II, females and males have a single field of pigmented cells (A, F). At instar III, females (B) have eye morphology very similar to females at instars IV (C), V (D), and adult (E) stages. Instar III males (F) have two fields of pigmented tissue with the more lightly pigmented field (arrow) slightly larger. By instar IV (H), the two fields are roughly the same size. At instar V (I), the more lightly pigmented field (arrow) is much larger and has begun to develop ommatidia, as evidenced by the presence of lenses (white arrow). Note that the orientation of the wholemount instar V eye is different from the orientation of the other eyes, the eye tissue curves under the focal plane putting the distal rudiment on the left. The adult’s eye morphology (J) is somewhat obscured due to the thickness of the tissue, the rudiment (arrow) is obvious at the distal tip, however. Numbers in each panel are the numbers of Euphilomedes carcharodonta and Euphilomedes morini individuals examined. All pictures are E. carcharodonta. For scale, see Fig. 5.

Eyes were measured on their long axis with Adobe Illustrator, using images taken at  10 magnification on an Olympus BX61 microscope. We tested the null hypothesis that eyes of two stages were the same size using Welch’s t-test (Welch 1947). Rejection of the null hypothesis indicates significant eye growth occurred between stages.

RESULTS E. carcharodonta has paired lateral eyes that are attached at their proximal ends to the head posterior to the second antennae (Fig. 2). In juvenile stages, these ‘‘eye-flaps’’ appear as unpigmented tissue with a central region of red-pigmented tissue. We examined the embryos, juveniles, and adults of male and female E. carcharodonta and the extremely similar E. morini (Fig. 1D). From this, we were able to see both the ontogenic progression of eye development and the dimorphism between males and females. We did not observe any differences in male and female morphology until mid-juvenile stages. Once manifested, these differences were maintained throughout development with the most dramatic differences in the adults (Fig. 3). Eye development appears to be slow throughout embryonic and

Rivera and Oakley

Ontogeny of sexual dimorphism

237

laterally on each side of the head and comprise four pigment spots. Unlike ostracods from both of the other myodocopid superfamilies (Cylidroleberids and Cyprinidids) (Fig. 1F; Wakayama, 2007), embryonic eyes did not significantly grow or differentiate between the time they were first observed and hatching. All observed embryos exhibited the same morphology so it was concluded that there are no sex-specific differences at this stage.

Instar II Fig. 3. Sections of adult Euphilomedes morini eyes. Panels (A, C, and E) show DAPI stained nuclei, (E) is a wholemount. Panels (B and D) are brightfield photos. Male eyes have a clear dorsoventral and distomedial differentiation. The distalmost portion of the eye is the rudiment. It is composed of a number of densely packed cells without obvious organization. Medially, the bulk of the eye is made up of ommatidia. Five are visible in panels (A and B). These comprise, from dorsal to ventral, a lens with two associated pigment cells, a cone, and a number of retinular cells with cell bodies packed tightly under the cone cells. Female eyes most closely resemble the male rudimentFthey are somewhat dense assemblages of cells lacking clear organization. A section through a female eye (C) shows a random scattering of these cells. A wholemount eye (E) shows the full complement of cells. Scale bar: 35 mm (A, B), 24 mm (C, D, E).

juvenile stages, with the most rapid growth in males occurring between instar V and adult stages (Table 1). Some ommatidial and lens development occurred between male instars IV and V. Two main differences were seen comparing the development of male and female eyes; male eyes undergo more growth, especially in later stages (Table 1), and have two separate fields of pigmentation, whereas females have only one.

Embryo We could not determine the sexes of embryos, but we observed embryos of multiple clutches (4100 embryos). The embryonic eyes appear after limbs have begun to develop distinct morphologies and the carapace has grown to cover the dorsal half of the embryo. The eyes are located dorso-

Table 1. Eye size measurements throughout development Males Average n size (mm) Instar Instar Instar Instar Adult

II III IV V

2 5 2 3 4

22 47 50 60 150

Pgrowth N/A 0.025 0.39 0.005 2.210E-4

Females

Field size Pomm Average ratio average growth n size (mm) N/A 1.7 2.7 5.9 8.1

N/A N/A 0.92 0.09 0.001

1 3 3 1 3

9 44 41 38 52

We do not include data from instar I as we could not distinguish sexes at this stage. Eye morphology appears very similar to instar II. We did not observe any differences in male and female eyes at instar II (Fig. 2, A vs. F). Their lateral eyes consist of a rounded flap of tissue connected proximally to the head by a narrow clear sclerotized tissue. Granular, bright red, pigmentation is observed in the center of the eye-flap. Compared with later stages, this pigmentation is sparse and takes up little of the total lateral eye anlage. A pigmented spot is visible just medial to the point of eye attachment on the main body of the animal. The significance of this spot is unknown, but it appears in all examined animals (male and female; data not shown).

Instar III At instar III, the female eye is already the elongated shape it will keep through adulthood. It is attached to the body at its wide base and tapers off towards its distal end. The pigmented tissue is roughly spherical and surrounded by unpigmented eye-flap tissue, occupying roughly half of the eye flap (Fig. 2B). In the tissue squash, lines of pigmented tissue are visible between the spherical pigmented tissue and the body. The male instar III eye is already quite distinct from the female eye. While approximately the same size as the female eye, the shape and pigmentation patterns differ even at this early stage. While the female eye-flap is elongated and tapering, the male eye-flap is rounder and thicker. Like the female eye, the male instar III eye is composed of pigmented tissue surrounded by unpigmented eye-flap tissue. The pigmented region of the eye anlage comprises two zones. One of these is heavily pigmented while the other is more lightly pigmented (Fig. 2G, inset). In the tissue squash, the darker region appears to contain many more large aggregations of pigment granules than the lighter region. The darker region occupies a discrete portion of the eye-flap, possibly surrounded by a thin membrane. The lighter region is more amorphous and has smaller pigment granule aggregations, which are more dispersed (Fig. 2G). At this stage, it is not clear which of these two zones most closely resembles the female pigmented tissue.

238

EVOLUTION & DEVELOPMENT

Vol. 11, No. 2, March--April 2009

Instar IV The female eye at instar IV is almost identical to the female eye at instar III. The pigmented portion of the eye is still composed of a single tissue terminating proximally in thin tracks of pigment granules. The eye-flap retains the elongated shape of the female instar III eye and the pigmented portion is in the same approximate position within this tissue (Fig. 2C). Like the female eye, the male instar IV eye does not appear vastly different from the male instar III eye. The eye-flap tissue is less elongated than the female eye-flap and the pigmented region is larger and composed of two separate zones. The more proximal zone is heavily pigmented and appears to correspond to the heavily pigmented zone in the instar III male eye. The distal zone is more lightly pigmented and corresponds to the lightly pigmented region in the earlier stage. The two zones are now approximately the same size (Fig. 2H, inset). This suggests that the darkly pigmented zone has grown relative to the lightly pigmented zone (Table 1). After squashing the tissues, it is apparent that the darkly pigmented zone is made up of dense pigment granule aggregations. The darkness of this zone appears to be caused by its thickness relative to the other zoneFthat is it comprises more tissue in the z-axis of the slide. This darker zone still appears to be discreet; while the lighter zone has more amorphous boundaries (Fig. 2H).

Fig. 4. Line drawings showing relative sizes of eyes and eye fields in Euphilomedes carcharodonta females (left) and males (right). Eyes from younger specimens are at the top. Note that the orientation of the instar V eye is different from the orientation of the other eyes. In this specimen, the eye tissue curves under the plane of the drawing. This results in a view of the anterior of the eye with the distal rudiment to the left of the ommatidial field. The line indicates where the eye tissue overlaps. Scale bar: 30 mm, 60 mm for male adult eye.

Instar V

Adult

At this stage, the difference between the male and female eyes is dramatic. While the female eye looks very similar to the female instar IV and III eyes (Fig. 2D), the male eye has undergone much differentiation (Fig. 2I). The pigmented portion of the male eye now occupies most of the eye-flap space. The darkly pigmented zone is now an order of magnitude larger than the lightly pigmented zone (Fig. 2I, inset). Further differentiation includes the development of transparent cuticular lenses. These are the lens portion of the crystalline cones, which are composed of two abutting cells (Land and Nilsson 1990). From an oblique angle, these appear as a clear oblong ellipse, with the line of abutment down the center (Figs. 1 and 2, white arrows). Each pair is associated with an ommatidium. At instar V, they are largest at the distal portion of the eye-flap and smaller at the more medial portion (Figs. 1E and 2I, white arrows). This suggests that eye development occurs in a distal to medial progression. Supporting this is the observation that pigment granules at the distal edge of the eye flap are darker and denser than towards the proximal portion. The female instar V pigmented tissue most closely resembles the smaller lightly pigmented zone of the male. These two regions are made up of small to medium-size aggregations of pigment granules and lack large lenses (Fig. 2D).

The female adult eye retains the characters of the juvenile eyes. It has grown in size, but has not undergone further obvious differentiation (Figs. 2 and 4, left column, Table 1). The eye flap is still elongated with a thick base. The pigmented portion of the eye is approximately in the middle of the eye flap and is surrounded by a large amount of unpigmented tissue. The pigmented portion is spherical, and small tracks of granules are apparent between this tissue and the main body of the animal (Fig. 2E). While a few small lenses were observed in adult females, nuclear staining with DAPI shows that the female eye does not have any obvious ommatidial morphology. Rather, cells are regularly distributed in the eye tissue. The male adult eye has undergone a striking transformation from the instar III/IV eye. Whereas the instar III/IV eye is composed of eye-flap tissue surrounding two more or less equally sized pigmented tissues, the adult eye is composed of one large field of ommatidia and a small distal rudimentary pigmented field, all surrounded by a thin layer of unpigmented tissue. At this stage, lenses are large and cover the surface of the ommatidial field. The distally located rudiment field does not have any associated lenses (Fig. 2J). Sectioning and DAPI staining of male eyes shows that the each pair of lenses

Rivera and Oakley

Fig. 5. Karyotype of a male cell. Trunk tissue including testes from three animals was used. Nine cells had a full complement of chromosomes. Each of these have 25 chromosomes, representing 2  12 autosomes and a single X chromosome, consistent with the hypothesis that males are XO.

is associated with two cone cells and a number of rhabdomeric cells. The ommatidial field has 33 of these ommatidial assemblages (n 5 3; see also Kornicker and Harrison-Nelson 1997) while the smaller field has none. The smaller field is made up of many cells without obvious organization. Because of this striking difference in the two pigmented zones of the adult male eye, we refer to the large region with ommatidia as the ommatidial zone, or compound eye, and the smaller region as the rudimentary zone. The pigmented portion of the female eye is also referred to as rudimentary, though the homology of the male and female rudimentary fields is unknown. However, the similar pigmentation patterns, lack of obvious ommatidia or lenses, lack of organization, slow growth throughout eye development, and appearance in ontogeny suggests homology between these two regions.

Ontogeny of sexual dimorphism

239

sexual dimorphism arising during ontogeny from a duplicated developmental field in males. This second field is present only in the males and forms ommatidia, resulting in an extremely dimorphic organ in males versus females. E. carcharodonta and E. morini adult males have large compound eyes containing both an ommatidial field, comparable to the ommatidial field of other ostracods, and a small distal rudiment. A combination of compound eye and rudiment in a single individual has not been previously described in this or other species. Female lateral ‘‘eyes’’ are small and of unknown function (Fig. 1). We find that female eyes develop as a slowgrowing tissue containing a single pigmented zone (Figs. 2 and 4, left columns, Table 1). Male eyes also begin as a single field, with no ommatidia formed by the end of embryogenesis, unlike other myodocopid eyes which undergo extensive embryonic development (Fig. 1; Wakayama, 2007). However, by instar III, male eyes comprise two pigmented zonesFone of which grows relatively quickly in late instars and contributes to the main compound eye, the other forms the distal rudiment (Figs. 2 and 4, right columns, Table 1). Based on the new data presented here, we discuss below three hypotheses: (1) The female eye is homologous to the male rudiment. (2) Euphilomedes dimorphism is due to differences in gene expression. (3) The male ommatidial field arises via field splitting, the fissure of a single field into two fields with distinct developmental potentials. We then discuss the significance of field splitting with regards to natural history and alternative hypotheses for the evolutionary origins of field splitting and the sexual dimorphism.

Karyotyping We found that female and embryonic tissue was not suitable for karyotyping with our procedure. As E. morini males are more abundant than E. carcharodonta males in the environs of the laboratory, three E. morini males were examined and nine cells had a full complement of easily differentiable chromosomes. If males are heterogamous, they should have an even number of chromosomes. However, we found that each cell had 25 chromosomes, suggesting that males are not heterogamous, but rather 12A;XO (Fig. 5). This agrees with previous analyses on representatives from other superfamilies of myodocopid crustaceans (Moguilevsky 1985; Moguilevsky and Whatley 1988; Moguilevsky 1990, 1992, 1995).

DISCUSSION Here we propose a novel mechanism for the development of a sexually dimorphic trait. Although most studies on sexual dimorphism focus on differences in tissue growth rates (e.g. Hens 2005; Moczek and Nagy 2005) or on ornamentation/ pigmentation differences of homologous tissues (e.g. Kopp et al. 2000; Jawor and Breitwisch 2003), we report here a

Ontogeny of philomedid eye dimorphism We hypothesize that the female eye is homologous to the male rudiment, based on gross morphology, tissue structure, and developmental timing. Both the female eye and male rudiment are made up of a seemingly homogenous field of cells (Fig. 3), and both appear at the same time in ontogenyFbefore the completion of embryogenesis, but after limb bud formation. This is roughly the same developmental time that pigmented eyes appear in another myodocopid ostracod from a different family (Wakayama 2007). In other ostracods, however, eyes develop ommatidia before hatching (instar I) while in E. carcharodonta and E. morini, they do not. The ommatidial field appears later in ontogeny in Euphilomedes males and undergoes growth and differentiation similar to that seen in other myodocopids (ibid). Thus, it is probable that the male and female rudiments are homologous. The large differences between male and female eye ontogeny and adult morphology could arise via additional genes in males (i.e., eye-genes on a male specific sex chromosome) or because of differences in eye-gene expression. We hypothesized that the regulation of dimorphism in eyes is in the same genetic background in both males and females, as

240

EVOLUTION & DEVELOPMENT

Vol. 11, No. 2, March--April 2009

sex-determination in other myodocopid ostracods is XX/XO (Moguilevsky 1985; Moguilevsky and Whatley, 1988; Moguilevsky, 1990; Moguilevsky, 1992; Moguilevsky, 1995). We found that males have 25 chromosomes, consistent with the hypothesis that males are 12A;XO and do not have a malespecific chromosome. That is, male and female development occurs in the same genetic background, with early dosage effects of X-linked genes likely to control the initiation of all aspects sexual dimorphism. Thus, we propose that the dimorphism is due to differences in gene expression in males and females. Candidate genes include those conserved in eye development across phyla and whose function is well understood in insects (e.g. pax6, wingless, hedgehog). Most insects examined, including Drosophila, Tribolium, and Schistocerca, pattern their ommatidial fields with a wave of differentiation very similar to that seen in crustaceans (Wolff and Ready 1993; Harzsch et al. 1999; Friedrich and Benzer 2000; Hafner and Tokarski 2001; Duman-Scheel et al. 2002). Our data from ostracods suggests that the male ommatidial fields are patterned similarly, with differentiation occuring in a distal to medial progression. As photoreceptor structure and molecular studies suggest that crustaceans and insects comprise a single monophyletic group (Giribet and Ribera 2000; Richter 2002; Dunn et al. 2008), we hypothesize that similar genes will be involved in lateral eye development in model insects and in ostracod crustaceans. If this is the case, Euphilomedes gives us the opportunity to study how changes in gene expression can lead to vastly different eye morphologies. Third, we propose that the male ommatidial tissue arises during ontogeny via splitting of the eye field during larval development. It remains possible that females also duplicate the eye field, but lose one duplicate between instar II and III. However, we do not see any evidence for this, despite examination of embryos and all instars. The pigmented field duplication that we observe in the male ostracods could arise either by ectopic expression of the field (heterotopy) or by a fissure of the eye field (field splitting). While heterotopy is a possibility, we favor field splitting for two reasons: first, field splitting occurs in eye development in other arthropods (e.g., Heming 1982; Green et al. 1993) and second, the two pigmented fields observed in male E. carcharodonta and E. morini are closely abutting. Therefore, we favor the field splitting hypothesis, and focus the remaining discussion on this mechanism, although much of the discussion would also apply to field duplication in general, including ectopic expression (heterotopy).

Significance of field splitting Field splitting is not only a developmental mechanism but also an evolutionary one. Over evolutionary time, it may allow the origin of novel characters when different organs evolve from the two tissues resulting from the splitting of an

ancestral field. As such, field splitting has intellectual parallels with gene duplication, speciation, and vicariance biogeography, all of which may be mechanisms that allow for increased biological diversity during evolution and may be put under the general heading of ‘‘furcation,’’ referring to the splitting of an evolutionary lineage into two or more new lineages (Oakley et al. 2007). One example of the furcation of a tissue is seen in insect eyes, which also likely originated by the developmental mechanism of field splitting. While most insects have a lateral eye field that undergoes ommatidial development throughout embryonic and juvenile stages, some holometabolous insects develop separate larval and adult lateral eyes (reviewed in Friedrich 2006). The larval and adult eyes both develop from a contiguous and homogeneous field of undifferentiated cells. Structural and gene expression data shows that both eye types can be considered homologous to other arthropod compound eyes, which develop continuously. In other words, there was apparently a furcation event during evolution of the lineage leading to holometabolous insects. In this way, a single ancestral compound eye field split to form two distinct fields, each of which underwent separate evolutionary trajectories (Friedrich 2006). Larval and adult eyes of these insects may thus be considered in-paralogs (Koonin 2001) relative to other arthropod compound eyes. Since the ecology of adult and larval insects is very different, it follows that different evolutionary forces shaped the evolution of each paralogous eye. We conclude that an analogous furcation event and divergence of function may have also occurred in the lineage leading to Euphilomedes, producing the two eye structures seen in males. Like holometabolous insect larvae, E. carcharodonta and E. morini juveniles have simple eyes (although function of the ostracod rudimentary eyes has not yet been examined). The similarity of these juvenile eyes with the early embryonic eyes of other myodocopid ostracods (Wakayama 2007) suggests that they are neotenous. The pre-ommatidial tissue that arises in mid-juvenile male E. carcharodonta and E. morini may be seen as similar to the adult holometabolous insect eyesFespecially in that the ‘‘larval eyes’’ (rudiments) are retained in adults, suggesting a field splitting event in both development and evolution. In a further parallel to holometabolous insects, the paralogous eye structures of E. carcharodonta and E. morini mainly exist in different environments with presumably different evolutionary pressures. Juveniles of both sexes in E. carcharodonta and E. morini are apparently mainly confined to living in the sediment, primarily interstitially. Presumably, large compound eyes would be of minimal use for the interstitial juveniles and adult females. In E. carcharodonta and E. morini, like several other philomedids, adult males are the only swimming stage. Like many other ostracods including bioluminescent Cypridinidae (Morin and Cohen 1988; Cohen 1989), philomedid males actively swim about 2 h after sunset (Fage 1933; T. H.

Rivera and Oakley Oakley, unpublished data). Mating probably occurs in the water column, with females swimming up from the sediment. Some females lose, and probably metabolize their flight muscles after mating (Kornicker 1993, pp. 65–66), similar to some ant queens that consume their wings after a nuptial flight, supporting the idea that female E. carcharodonta and E. morini only swim for mating purposes. The planktonic adult males therefore may have a greater need for eyes than juveniles or females to evade potential predators and/or to locate potential mates. These proposed adaptive mechanisms for the evolution of rudimentary eyes in female adults and juveniles versus compound eyes in male adults of E. carcharodonta and E. morini are similar to the resource partitioning adaptations seen in other animals. Unlike bill shape and body-size differences, however, Euphilomedes dimorphism is based on the presence/absence of a character (the ommatidial field), not simply differential growth of a homologous tissue.

Evolutionary origins of philomidid sexual dimorphism We present two evolutionary hypotheses for the relationship between field splitting and dimorphism. In the first, field splitting occurs before dimorphism and in the second, the two evolve simultaneously. The first possibility is that the females have lost the ommatidial field. In this case, the philomedid ancestor would have resembled the male E. carcharodonta and E. moriniFhaving both ommatidial and rudimentary fields. The ommatidial field would have been lost only in females, either through loss of the ability to duplicate the tissue in development or through developmental arrest of the eye anlage before both fields formed. In this scenario the eye field furcation in the philomedid lineage occurred before the advent of dimorphic eyes (Fig. 6, Hypothesis 1). The selective advantage of having two eye fields is unclear. Without more data on the possible function of the rudiment field, we hypothesize that the initial mutation causing furcation was a neutral event. This furcation then allowed for female-specific eye reduction without a loss of the entire eye field, something that may be otherwise impossible due to developmental constraints. The second possibility is that dimorphism originated when males underwent eye tissue furcation. In this case, the philomedid ancestor would have resembled female E. carcharodonta and E. morini, having only a rudimentary field. The ancestral myodocopid very likely had ommatidia (Oakley and Cunningham 2002), so in this hypothesis the philomedid ancestor underwent a reduction of the eye field to produce a rudiment. Philomedid males later regained an ommatidial field, possibly by a redeployment of a suppressed eye-development program on a duplicated eye field (Fig. 6, Hypothesis 2). Here, furcation and dimorphism arose simultaneously. In

Ontogeny of sexual dimorphism

241

Fig. 6. (A) Two evolutionary hypotheses for the presence of dimorphic eyes in Euphilomedes carcharodonta. Hypothesis 1: Field splitting occurred between the ancestral myodocopid and the ancestral philomedid, giving the ancestral philomedid rudimentary and ommatidial fields. The ommatidial field was lost or reduced in the females of some philomedid lineages. Hypothesis 2: An eye reduction occurred in the lineage leading to Philomedinae resulting in a philomedid ancestor with reduced eyes. This reduced (rudimentary) field was then split in development in some male philomedids. One of the two resultant fields then redeployed the eye developmental pathway to form a new ommatidial field. (B) A phylogenetic tree redrawn from Oakley (2005) of the myodocopid ostracods with eye-types of families shown. The myodocopid ancestor had ommatidia (Oakley and Cunningham 2002), but probably no rudimentary field, as rudimentary fields in species outside Philomedinae have not been reported and some philomedids do not have a rudimentary field (T. H. Oakley, unpublished observations).

this case, we hypothesize that the original loss of eyes in the ancestral philomedid could have occurred because of a change in niche. For example, ostracod eyes are generally reduced in deep-sea species, where there is little light to make a large compound eye beneficial (Kornicker 1992). If the ancestral philomedid were a deep-sea species with reduced eyes, a later furcation in males could have provided a tissue for the redeployment of the compound eye developmental pathway. This possibility is attractive as there exist multiple species of deep-sea Philomedinae, including one deep sea genus (Igene) in which all known species lack lateral eyes in both males and females (Kornicker 1975, 1992, 1995). However, there is not enough data currently to distinguish between our two hypotheses on evolutionary origin. To do this, a well-supported phylogeny and ancestral trait reconstruction of compound eye history in philomedids is needed. These future studies of sexual dimorphism in Philomedinae will give insight into furcation by field splitting in both evolutionary and developmental contexts. Specific evolutionary questions could be resolved by looking at the development of other Philomedinae. This will show whether field splitting is a common mechanism for all Philomedinae or if it is an ancestral mechanism for redeploying a repressed eye-development cassette. Such a finding will have implications for understanding the evolution of novel characters in general.

242

EVOLUTION & DEVELOPMENT

Vol. 11, No. 2, March--April 2009

Further, a more detailed developmental understanding of the relatively simple field-splitting event in male E. carcharodonta and E. morini ontogeny will aid in understanding more complex examples, such as somitogenesis in vertebrates. The ontogeny of a specific dimorphic character can give insight into the evolution of that character (Kopp et al. 2000; Emlen et al. 2005). In our case, the ontogeny of a tissue gives a mechanism for the presence/absence of ommatidia within a species that may be extended to the presence/absence of ommatidia across a family. Whether the philomedid ancestor had two fields in the eye anlage or one, and whether it had ommatidia or only a rudimentary field are questions that remain open. In any case, it is clear that further study of this will lead to a better understanding of both field splitting as a means of creating or facilitating sexual dimorphism in a single species as well as a potential evolutionary mechanism for creating novelty. Acknowledgments We thank T. Schwander for aid in the karyotype analysis. We also thank T. Long, M. Friedrich, and members of the Oakley, Mazer, and Rice labs for valuable comments.

REFERENCES Badyaev, A. V. 2002. Growing apart: an ontogenetic perspective on the evolution of sexual size dimorphism. Trends Ecol. Evol. 17: 369–378. Bergen, M., et al. 2001. Relationship between depth, sediment, latitude, and the structure of benthic infaunal assemblages on the mainland shelf of southern California. Mar. Biol. 138: 637–647. Billeter, J. C., Rideout, E. J., Dornan, A. J., and Goodwin, S. F. 2006. Control of male sexual behavior in Drosophila by the sex determination pathway. Curr. Biol. 16: R766–R776. Brennan, J., and Capel, B. 2004. One tissue, two fates: molecular genetic events that underlie testis versus ovary development. Nat. Rev. Genet. 5: 509–521. Cohen, A. 1983. Rearing and postembryonic development of the myodocopid ostracode Skogsbergia lerneri from Coral Reefs of Belize and the Bahamas. J. Crustacean Biol. 3: 235–256. Cohen, A. C. 1989. Comparison of myodocopid ostracodes in two zones of the Belize barrier reef near Carrie Bow Cay with changes in distribution 1978–1981. Bull. Mar. Sci. 45: 316–337. Duman-Scheel, M., Pirkl, N., and Patel, N. H. 2002. Analysis of the expression pattern of Mysidium columbiae wingless provides evidence for conserved mesodermal and retinal patterning processes among insects and crustaceans. Dev. Genes Evol. 212: 114–123. Dunn, C. W., et al. 2008. Broad phylogenomic sampling improves resolution of the animal tree of life. Nature 452: 745–U5. Dunn, P. O., Whittingham, L. A., and Pitcher, T. E. 2001. Mating systems, sperm competition, and the evolution of sexual dimorphism in birds. Evolution 55: 161–175. Estrada, B., Casares, F., and Sanchez-Herrero, E. 2003. Development of the genitalia in Drosophila melanogaster. Differentiation 71: 299–310. Fage, L. 1933. Peˆches planctoniques a` la lumie`re effectue´es a` Banyuls-surMer et a` Concarneau. 3. Crustaces. Arch. Zool. exp. ge´n. 76: 105–248. Friedrich, M. 2006. Continuity versus split and reconstitution: exploring the molecular developmental corollaries of insect eye primordium evolution. Dev. Biol. 299: 310–329. Friedrich, M., and Benzer, S. 2000. Divergent decapentaplegic expression patterns in compound eye development and the evolution of insect metamorphosis. J. Exp. Zool. 288: 39–55.

Giribet, G., and Ribera, C. 2000. A review of arthropod phylogeny: new data based on ribosomal DNA sequences and direct character optimization. Cladistics-Int. J. Willi Hennig Soc. 16: 204–231. Green, P., Hartenstein, A., and Hartenstein, V. 1993. The embryonic development of the Drosophila visual system. Cell Tissue Res. 273: 583–598. Grozeva, S., and Nokkala, S. 1996. Chromosomes and their meiotic behavior in two families of the primitive infraorder Dipsocoromorpha (Heteroptera). Hereditas 125: 31–36. Hafner, G. S., and Tokarski, T. R. 2001. Retinal development in the lobster Homarus americanusFcomparison with compound eyes of insects and other crustaceans. Cell Tissue Res. 305: 147–158. Harzsch, S., Benton, J., Dawirs, R. R., and Beltz, B. 1999. A new look at embryonic development of the visual system in decapod crustaceans: Neuropil formation, neurogenesis, and apoptotic cell death. J. Neurobiol. 39: 294–306. Heming, B. S. 1982. Structure and development of the larval visual-system in embryos of Lytta-Viridana Leconte (Coleoptera, Meloidae). J. Morphol. 172: 23–43. Hens, S. M. 2005. Ontogeny of craniofacial sexual dimorphism in the orangutan (Pongo pygmaeus). I: face and palate. American Journal of Primatology 65: 149–166. Hiruta, S. 1980. Morphology of the larval stages of Vargula hilgendorfii (G. W. Mu¨ller) and Euphilomedes nipponica Hiruta from Japan (Ostracoda: Myodocopina). J. Hokkaido Univ. Educ. 30: 145–167. Huvard, A. L. 1990. The ultrastructure of the compound eye of two species of marine ostracods (Crustacea: Ostracoda: Cypridinidae). Acta Zool. 71: 217–224. Jawor, J. M., and Breitwisch, R. 2003. Melanin ornaments, honesty, and sexual selection. Auk 120: 249–265. Koonin, E. V. 2001. An apology for orthologsFor brave new memes. Genome Biol. 2: COMMENT1005.1-1005.2 Kopp, A., Duncan, I., and Carroll, S. B. 2000. Genetic control and evolution of sexually dimorphic characters in Drosophila. Nature 408: 553–559. Kornicker, L. S. 1975. Antarctic Ostracoda (Myodocopina). Smithsonian Contrib. Zool. 163: 1–389. Kornicker, L. S. 1992. Myodocopid ostracoda of the Benthedi Expedition, 1977, to the NE Mozambique channel, Indian Ocean. Smithsonian Contrib. Zool. 531: 1–243. Kornicker, L. S. 1995. Ostracoda (Myodocopina) of the SE Australian continental slope, Part 2. Smithsonian Contrib. Zool. 97: 1–97. Kornicker, L. S., and Harrison-Nelson, E. 1997. Myodocopid ostracoda of Pillar Point Harbor, Half Moon Bay, California. Smithsonian Contrib. Zool. 593: 1–50. Land, M. F., and Nilsson, D. E. 1990. Observations on the compound eyes of the deep-sea ostracod Macrocypridina castanea. J. Exp. Biol. 148: 221– 234. Ligon, J. D. 1968. Sexual differences in foraging behavior in 2 species of Dendrocopos Woodpeckers. Auk 85: 203–215. Morin, J. G., and Cohen, A. C. 1988. 2 New Luminescent Ostracodes of the Vargula (Myodocopida, Cypridinidae) from the San Blas Region of Panama. J. Crustacean Biol. 8: 620–638. Moczek, A. P., Andrews, J., Kijimoto, T., Yerushalmi, Y., and Rose, D. J. 2007. Emerging model systems in evo-devo: horned beetles and the origins of diversity. Evol. Dev. 9: 323–8. Moczek, A. P., and Nagy, L. M. 2005. Diverse developmental mechanisms contribute to different levels of diversity in horned beetles. Evol. & Dev. 7: 175–185. Moczek, A. P., Rose, D., Sewell, W., and Kesselring, B. R. 2006. Conservation, innovation, and the evolution of horned beetle diversity. Dev. Gen. and Evol. 216: 655–665. Moguilevsky, A. 1985. Cytogenetic studies on marine ostracods: the karyotype of Gigantocypris muelleri Skogsberg, 1920 (Ostracoda, Myodocopida). J. Micropalaeontol. 4: 159–164. Moguilevsky, A., and Whatley, R. C. 1988. Cytogenetic studies on marine myodocopid Ostracoda; the karyotypes of Gigantocypris dracontovalis Cannon 1940 and Macrocypridina castanea (Brady, 1897). Dev. Palaeontol. and Stratigr. 11: 293–300. Moguilevsky, A. 1990. Cytogenetic studies on marine myodocopid Ostracoda: the karyotypes of some species of Vargula Skogsberg, 1920.

Rivera and Oakley In R. Whatley and C. Maybury (eds.). Ostracoda and Global Events. Chapman and Hall, London. pp. 187–203. Oakley, T. H. 2005. Myodocopa (Crustacea: Ostracoda) as models for evolutionary studies of light and vision: multiple origins of bioluminescence and extreme sexual dimorphism. Hydrobiologia 538: 179–192. Oakley, T. H., and Cunningham, C. W. 2002. Molecular phylogenetic evidence for the independent evolutionary origin of an arthropod compound eye. Proc. Natl. Acad. Sci., USA 99: 1426–1430. Oakley, T. H., Plachetzki, D. C., and Rivera, A. S. 2007. Furcation, fieldsplitting, and the evolutionary origins of novelty in arthropod photoreceptors. Arthropod Struct. Dev. 36: 386–400. Richter, S. 2002. The tetraconata concept: hexapod-crustacean relationships and the phylogeny of Crustacea. Organisms Divers. Evol. 2: 217–237. Shine, R. 1978. Sexual size dimorphism and male combat in snakes. Oecologia 33: 269–277. Shine, R., Reed, R. N., Shetty, S., and Cogger, H. G. 2002. Relationships between sexual dimorphism and niche partitioning within a clade of seasnakes (Laticaudinae). Oecologia 133: 45–53.

Ontogeny of sexual dimorphism

243

Smith, V. Z. 1952. Further Ostracoda of the Vancouver Island Region. J. Fish. Res. Board Can. 9: 16–41. Temeles, E. J., Pan, I. L., Brennan, J. L., and Horwitt, J. N. 2000. Evidence for ecological causation of sexual dimorphism in a hummingbird. Science 289: 441–443. Vannier, J., and Abe, K. 1992. Recent and early Paleozoic myodocope ostracodesFfunctional morphology, phylogeny, distribution and lifestyles. Palaeontology 35: 485–517. Wakayama, N. 2007. Embryonic development clarifies polyphyly in ostracod crustaceans. J. Zool. 273: 406–413. Welch, B. L. 1947. The generalization of students problem when several different population variances are involved. Biometrika 34: 28–35. Whatley, R. C., Nicholas, T., Moguilevsky, A., and Coxill, D. 1995. Ostracoda from the South west Atlantic; Part I, The Falkland Islands. Rev. Espan. de Micropaleontol. 27: 17–38. Wolff, T., and Ready, D. F. 1993. Pattern formation in the Drosophila retina. In M. Bate and A. Martinez-Arias (eds.). The Development of Drosophila Melanogaster. Cold Spring Harbor Laboratory Press, Woodbury, New York, pp. 1277–1325.