Morphological evolution in seaurchin development: hybridsprovide insights into the paceof evolutionMaria Byrne1* and Janice Voltzow2

SummaryHybridisations between related species with divergentontogenies can provide insights into the bases forevolutionary change in development. One example ofsuch hybridisations involves sea urchin species thatexhibit either standard larval (pluteal) stagesor those thatdevelop directly from embryo to adult without an inter-vening feeding larval stage. In such crosses, plutealfeatures were found to be restored in fertilisations of theeggs of some direct developing sea urchins (Heliocidariserythrogramma) with the sperm of closely (Heliocidaristuberculata) and distantly (Pseudoboletia maculata)related species with feeding larvae. Such results can beargued to support the punctuated equilibrium model—conservation in pluteal regulatory systems and a com-paratively rapid switch to direct development in evolu-tion.(1,2) Generation of hybrids between distantly relateddirect developers may, however, indicate evolutionaryconvergence. The ‘rescue’ of pluteal features by paternalgenomes may require maternal factors from H. erythro-gramma because the larva of this species has plutealfeatures. In contrast, pluteal featureswere not restored inhybridisations with the eggs of Holopneustes purpur-escens, which lackspluteal features.Howmuchofplutealdevelopment can be lost before it cannot be rescued insuch crosses? The answer awaits hybridisations amongindirect and direct developing sea urchins differing indevelopmental phenotype, in parallel with investigationsof the genetic programs involved. BioEssays 26:343–347, 2004. � 2004 Wiley Periodicals, Inc.

Introduction

Regulatory gene systems that control development can

remain stable over long evolutionary periods, undergo gradual

change, or change in a comparatively short time frame in a

punctuated event. Determining the relative influence of

different paces of change in the generation of morphological

developmental diversity is a major challenge for understand-

ing the evolution of development (Evo-Devo). Morphological

evolution is assumed to be driven by changes in the underlying

mechanisms of gene regulation, which in turn are considered

to be the driving force underlyingmany speciation events.(3) In

principle, insights into the relative importance of different

paces of evolutionary change can be generated by hybridisa-

tion experiments between species with divergent develop-

mental phenotypes.(1,2) In a series of hybridisations between

closely related and distantly related species of sea urchin

with different larval phenotypes (Figs. 1 and 2), Raff and

colleagues(1,2) assess the relative importance of gradual

versus punctuated evolution in development. They address

two hypotheses: (1) that gene regulatory systems evolve in a

punctuated manner during periods of rapid morphological

change and are relatively stable over long periods of slow

morphological evolution and (2) that regulatory pathways in

species that have independently evolved direct development

are convergent.

Indirect development that includes a planktotrophic echi-

nopluteus is viewed as the primitive condition for sea urchins.

Over the course of their evolution, numerous species of sea

urchins have independently evolved a non-feeding, directly

developing larva.(4) Theseaurchins usedby theRaff teamasa

model system for Evo-Devo include Heliocidaris tuberculata,

an indirect developer with a typical feeding echinopluteus,

and Heliocidaris erythrogramma, a direct developer that has

evolved a highly modified non-feeding ovoid larva (Figs. 1 and

2). These echinometrids are phylogenetically distant to the

other Australian species used for the hybridisation study, the

toxopneustid Pseudoboletia maculata, an indirect developer,

and the temnopleurid Holopneustes purpurescens, a direct

developer.

Hybridisation experiments

Fertilisation of the eggs of the direct developer Heliocidaris

erythrogramma (He) with the spermof either of the two indirect

BioEssays 26:343–347, � 2004 Wiley Periodicals, Inc. BioEssays 26.4 343

1Department of Anatomy and Histology, University of Sydney, NSW

2006, Australia.2Department of Biology, University of Scranton, Scranton, PA 18510-

4625, USA.

*Correspondence to: M. Byrne, Department of Anatomy and Histology,

F13, University of Sydney, NSW 2006, Australia.

E-mail: mbyrne@anatomy.usyd.edu.au

DOI 10.1002/bies.20024

Published online in Wiley InterScience (www.interscience.wiley.com).

What the papers say

developers, the closely related Heliocidaris tuberculata (Ht)

(4Myadistant) or the distantly relatedPseudoboletiamaculata

(Pm) (40 Mya distant) generated hybrid larvae with pluteal

features (Fig. 2). The hybrids possessed larval arms

supported by fenestrated skeletal rods, a larval gut and a

complete ciliary band. The paternal genomes ofH. tuberculata

and P. maculata apparently ‘restored’ features of indirect

development in the hybrids.(1) Development of some of these

hybrids was harmonious to metamorphosis and the juvenile

stage. Gene expression and immunocytochemical studies

with the He�Ht hybrids (maternal parent named first) show

that the oral and aboral larval territories demarcated by the

Figure 1. Larval phenotypesof the speciesused for thehybridisation trials byRaff et al.(1,2)A:Feedingechinopluteus larvaofHeliocidaristuberculata. Pseudoboletia maculata also has an echinopluteus. B: Heliocidaris erythrogramma has a non-feeding simple larva with a

prominent ciliary band.C:The larva ofHolopneustes purpurescens lacks pluteal features. A andB fromByrneM, Emlet RB, Cerra A. 2001.

Acta Zool 82:189–199 with permission of Blackwell Press, C, courtesy M. Ricketts. Scales: A 65 mm, B 15 mm, C 90 mm.

Figure 2. Results of conspecific and heterospe-

cific crosses among distantly and closely related

sea urchin species with different modes of devel-

opment as reported by Raff et al.(1,2) Figures for

conspecific crosses after 2,4,14,15, heterospecific

crosses after 1. Diagrams not to scale.

What the papers say

344 BioEssays 26.4

ciliary band in a feeding larva are restored.(1,2,5) Larval

territories, however, are not restored in the He�Pm hybrids,

despite the presence of the ciliary band.

In similar experiments in which the eggs of another direct

developer, Holopneustes purpurescens (Hp), were fertilised

by the sperm of Heliocidaris tuberculata or Pseudoboletia

maculata (60–70 Mya distant from H. purpurescens) pluteal

characters were not restored, however (Fig. 2). In this case,

the hybrids displayed very variable abnormal larval pheno-

types. To achieve hybridisation, however, excess sperm had

been required and so polyspermy may have influenced this

result.

Reciprocal crosses between the two direct developers

Heliocidaris erythrogramma and Holopneustes purpurescens

generated hybrids that had the simple ovoid phenotype typical

of direct development (Fig. 2). There were no signs of pluteal

characters. The eggs of the indirect developers could not be

used to generate hybrids with sperm of direct developers

because the embryos arrested development. In the Ht�Pm

crosses, hybrid echinoplutei were generated, as in other

hybridisation studies.(6,7) The reciprocal fertilisation (Pm�Ht)

was not successful. Gametic compatibility varies considerably

between sea urchin species and this is frequently a limiting

factor for hybridisation experiments.(8)

Punctuated equilibrium, conservation

and larval axis

So what is one to make of these results? Raff et al.(1) use the

hybrid phenotypes to address the two hypotheses detailed

above. The success of the crosses between the eggs of the

direct developerHeliocidaris erythrogramma and the sperm of

the two indirect developers indicates a remarkable conserva-

tion of the control pathways underlying pluteus formation in

phylogenetically distant indirect developers. The sperm of

both species were equally capable of ‘restoring’ pluteal fea-

tures in the hybrids (Fig. 2). The data support the punctuated

equilibrium model of long (at least 40 Mya) conservation of

regulatory systems to make the pluteus, an intricate larval

form, and a comparatively rapid (�4 Mya) switch to a simple

larval form by H. erythrogramma. That the two indirect

developers appeared equally able to form hybrid pluteal

morphs with H. erythrogramma indicates that developmental

drift between the pluteal programs of H. tuberculata and

Pseudoboletia maculata has been slow despite their 40 Mya

distance. Stasis or slow developmental system drift in the

regulation of pluteus formation was also indicated by produc-

tion of hybrid echinoplutei in the Ht�Pm cross.

The inability of the sperm of the indirect developers to

restore pluteal features in hybrids with Holopneustes purpur-

escens (Fig. 2) indicates that the developmental program of

this species may be toomodified to allow ancestral characters

to be restored. To test the punctuated equilibrium model with

H. purpurescens, we need to determine the larval phenotype

that would be generated by hybridisation between this direct

developer and a closely related indirect developer. Most

temnopleurids are indirect developers(9) and several species

could be used in hybridisation experiments with H. purpur-

escens. Direct development apparently evolved only once in

temnopleurids, 4–7 Mya ag.(9)

The second hypothesis of convergence in evolution of

direct development is supported by the harmonious develop-

ment of simple larvae in reciprocal crosses between Helioci-

daris erythrogramma andHolopneustes purpurescens (Fig. 2)

A pluteal phenotype was not generated. These results, in

species whose families diverged 60–70 Mya ago, indicate

convergence in morphological evolution associated with the

shift to direct development and a likely convergence in

the underlying genetic regulatory mechanisms. The result of

harmonious direct development in hybrids of the distantly

related species indicate that parallelisms in morphology might

be underlain by parallelisms in genetic regulatory mechan-

isms. Genotype and phenotype might be linked in achieving a

solution to the overarching problem—selection to reduce the

vulnerable planktonic period by deleting the feeding stage.

Data on expression of skeletogenic genes indicate that

convergent regulatory gene expression among direct devel-

oping sea urchins may be more common than previously

thought.(10)

Radialisation and death of embryos generated by fertilisa-

tion of the small eggs of the indirect developers Heliocidaris

tuberculata andPseudoboletia maculatawith the sperm of the

direct developers Heliocidaris erythrogramma and Holop-

neustes purpurescens (Fig. 2) is interpreted to reflect inability

of the embryo to specify the oral-aboral embryonic axis.(1) This

axis is maternally determined in H. erythrogramma but, in the

indirect developers, is formed after fertilisation.(11,12) The oral-

aboral axis of H. purpurescens, however, is influenced by the

sperm entry point and so is not maternally determined.(13)

How much of pluteal development can be lost

before it cannot be rescued?

The contrasting results of the restoration of characters of

feeding larvae in hybrids with the eggs of Heliocidaris

erythrogramma and the inability to achieve hybrids with the

eggs of Holopneustes purpurescens is an intriguing aspect of

this research. The results prompt the question as to how far

echinoid species can go down the path of direct development

before hybrid larvae with pluteal characters cannot be gene-

rated. The phenotypic diversity of the lecithotrophic larvae of

sea urchins is evidence of morphological evolution and shows

differences in the extent of loss of pluteal characters (Fig. 3).

The larva of H. erythrogramma differs from that of H.

purpurescens in the possession of vestigial pluteal charac-

ters.(14) Although H. erythrogramma has the overall simple

larval phenotype of direct developers, it is one of the few direct

developers to retain a ciliary band (Fig. 1). Based on its

What the papers say

BioEssays 26.4 345

morphology and anatomical position, the ciliary band can be

interpreted as a combination of the feeding ciliary band and

the epaulettes of the feeding echinopluteus.(15) The role of

the ciliary band in swimming in the larve is similar to that of

epaulettes in plutei, indicating selection to retain this structure

despite loss of its role in feeding. In addition, the larva of H.

erythrogramma has a reduced pluteal skeleton that, in the

absence of larval arms, appears to be vestigial. Although the

restoration of pluteal characters in the He�Ht and He�Pm

hybrids is suggested to be due to dominance of the paternal

genome,(1,2) it also seems likely that the eggs of H.

erythrogramma contain factors required for production of a

pluteus-like hybrid.

In contrast to Heliocidaris erythrogramma, the larva of

Holopneustes purpurescens lacks all vestiges of the pluteus

(Figs. 1 and 3). Its larval phenotype is among the most

modified seen in direct developing echinoids. The gastrula

gives rise promptly to a ‘vestibular larva’, so-called for its

prominent vestibule.(13) The development of H. purpurescens

may be too modified to have feeding larval features restored

through hybridisation. To explore this suggestion, we need to

determine whether pluteal features can be restored by the

genome of an indirect developing temnopleurid. As suggested

above, this experiment is required to test the punctuated

equilibrium model of evolution of regulatory gene system in

sea urchins.

Interestingly, the ciliary band of Heliocidaris erythrogram-

ma did not develop in the He�Hp hybrids (Fig. 2). The

paternal genome of the morphologically simple direct devel-

oper Holopneustes purpurescens apparently knocked out

expression of the band. This indicates that paternal dom-

inance and/or suppression of maternal regulatory factors that

may be required for ciliary band production.

Phylogenetic distance has not prevented the sperm of a

toxopneustid echinoid (Pm) from restoring pluteal features in

hybrids with eggs of an echinometrid (He). Thus far,

successful hybrids have been produced between echinoids

within a 60 Mya divergence time.(1) Within the limits imposed

by sea urchin fertilisation biology, what is the greatest

phylogenetic distance over which hybridisations will restore

pluteal features? How simple can the phenotype of echinoid

larvae be before hybridisation experiments fail to produce

‘intermediate’ pluteal morphs? The diversity of echinoid larval

forms shows the variation in expression of pluteal characters

and identifies candidate species for future hybridisation trials

(Fig. 3). The larval types include typical feeding echinoplutei,

facultatively feeding plutei, non-feeding reduced plutei and

larvae that completely lack pluteal characters. The direct

developing phenotypes include larvae that have (1) arms,

larval skeleton and ciliary bands, (2) arms and skeleton, and

(3) skeleton and ciliary bands. Considering the phylogenetic

distance, the cidaroid Phyllacanthus imperialis may be too

remote from the euechinoid genera (250 Mya) for hybridisa-

tion, but crosses with Phyllacanthus parvispinus may be

possible and would undoubtedly be of interest.

It may be possible to generate hybrid larvae by crossing

direct and indirect developers along gradients of phenotypic

simplification within the limits imposed by phylogenetic

distance. A diversity of larval phenotypes is available for

further exploration of the pace of evolutionary change and the

underlying genetic mechanisms regulating the evolution of

development. It will be important to conduct hybridisation

studies in parallel with investigation of the genetic systems

involved. This goal may not be so difficult to achieve once

the sea urchin genome is sequenced, given the power of

gene array technologies. As concluded by Raff et al.,(1)

comparative investigation of genetic expression is ultimately

needed to test hypotheses of punctuated evolution and

convergence of genetic regulatory systems.

References1. Raff EC, Popodi EM, Kauffman JS, Sly BJ, Turner FR, Morris VB, Raff RA.

2003. Regulatory punctuated equilibrium and convergence in the

evolution of developmental pathways in direct-developing sea urchins.

Evol Dev 5:478–493.

Figure 3. Phenotypes of sea urchin larvae including feeding echinoplutei, faculatively feeding plutei, reduced non-feeding plutei with

various pluteal characters and larvae that lack pluteal characters (after 1,4,13–21). This figure illustrates the morphological expression of

pluteal features rather than evolutionary or developmental sequences.

What the papers say

346 BioEssays 26.4

2. Raff EC, Popodi EM, Sly BJ, Turner FR, Villinski JT, Raff RA. 1999. A

novel ontogenetic pathway in hybrid embryos between species with

different modes of development. Development 126:1937–1945.

3. Raff RA, Kaufmann TC. 1983. Embryos, Genes and Evolution. New York:

Macmillan.

4. Wray GA. 1996. Parallel evolution of nonfeeding larvae in echinoids. Syst

Biol 45:308–322.

5. Nielsen MG, Wilson KA, Raff EC, Raff RA. 2000. Novel gene expression

patterns in hybrid embryos between species with different modes of

development. Evol Dev 2:133–144.

6. Hinegardner RT. 1975. Morphology and genetics of the sea urchin

embryo. Amer Zool 15:679–689.

7. McClary DJ, Sewell MA. 2003. Hybridisation in the sea: gametic and

developmental constraints on fertilisation in sympatric species of Pseu-

dechinus (Echinodermata:Echinoidea). J Exp Mar Biol Ecol 284:51–70.

8. Zigler KS, Raff EC, Popodi E, Raff RA, Lessios HA. 2003. Adaptive

evolution of bindin in the genus Heliocidaris is correlated with the shift to

direct development. Evolution 57:2293–2302.

9. Jeffrey CH, Emlet RB, Littlewood DTJ. 2003. Phylogeny and evolution of

developmental mode in temnopleurid echinoids. Mol Phylog Evol 28:

99–118.

10. Wray GA. 2002. Do convergent developmental mechanisms underlie

convergent phenotypes? Brain Behav Evol 59:327–336.

11. Henry JJ, Raff RA. 1990. Evolutionary change in the process of

dorsoventral axis determination in the direct developing sea urchin,

Heliocidaris erythrogramma. Dev Biol 141:55–69.

12. Goldstein B, Freeman G. 1997. Axis specification in animal development.

BioEssays 19:105–116.

13. Morris VB. 1995. Apluteal development of the sea urchin Holopneustes

purpurescens Agassiz (Echinodermata: Echinoidea: Euechinoidea).

Zool J Linn Soc 114:349–364.

14. Emlet RB. 1995. Larval spicules, cilia, and symmetry as remnants of

indirect development in the direct developing sea urchin Heliocidaris

erythrogramma. Dev Biol 167:405–415.

15. Byrne M, Emlet RB, Cerra A. 2001. Ciliary band structure

in planktotrophic and lecithotrophic larvae of Heliocidaris species

(Echinodermata: Echinoidea): conservation and change. Acta Zool 82:

189–199.

16. Okazaki K, Dan K. 1954. The metamorphosis of partial larvae of

Peronella japonica Mortensen, a sand dollar. Biol Bull 106:83–99.

17. Emlet RB. 1986. Facultative planktotrophy in the tropical echinoid

Clypeaster rosaceus (Linnaeus) and a comparison with obligate

planktotrophy in Clypeaster subdepressus (Gray) (Clypeasteroida:

Echinoidea). J Exp Mar Biol Ecol 95:183–202.

18. Parks AL, Bisgrove BW, Wray GA, Raff RA. 1989. Direct development in

the sea urchin Phyllacanthus parvispinus (Cidaroidea): phylogenetic

history and functional modification. Biol Bull 177:96–109.

19. Amemiya S, Emlet RB. 1992. The development and larval form of an

echinothurioid echinoid, Asthenosoma ijimai, revisited. Biol Bull 182:15–

30.

20. Olsen RR, Cameron JL, Young CM. 1993. Larval development (with

observations on spawning) of the pencil sea urchin Phyllacanthus

imperialis: a new intermediate larval form? Biol Bull 185:77–85.

21. Hart MW. 1996. Evolutionary loss of larval feeding: development, form

and function in a facultatively feeding larva, Brisaster latifrons. Evolution

50:174–187.

What the papers say

BioEssays 26.4 347