Trabajo Práctico Nº 1 - ege.fcen.uba.ar · TRABAJO PRÁCTICO 1 : SEMINARIO Lea atentamente los...

EVOLUCIÓNDEPARTAMENTO DE ECOLOGÍA, GENÉTICA Y EVOLUCIÓN

FACULTAD DE CIENCIAS EXACTAS Y NATURALES

UNIVERSIDAD DE BUENOS AIRES

1° CUATRIMESTRE 2018

GUÍA DE TRABAJOS PRÁCTICOS YSEMINARIOS

MODULO 7

EVOLUCIÓN Y DESARROLLO

(EVODEVO)

TRABAJOS PRÁCTICOS

EVOLUCIÓN Y DESARROLLO (EVODEVO)

“Evolution occurs when ontogeny is altered in one of two ways: when new characters are introduced at any stage of development with varying effects upon subsequent stages, or when

characters already present undergo changes in developmental timing”

S.J.Gould en “Ontogeny and Phylogeny” (1977)

BREVE INTRODUCCIÓN

El término “evolución” fue usado por los preformacionistas del siglo XVIII para describir el proceso por el cual un embrión crece hasta convertirse en un organismo adulto. Esta es la primera connotación biológica del término y difiere claramente de su significado actual. Los preformacionistas creían que el embrión estaba ya ubicado dentro de la célula germinal con todas sus partes ya formadas y simplemente crecía en tamaño hasta alcanzar el estado adulto. El ejemplo extremo de este punto de vista es el del homúnculo, el hombrecito con proporciones de adulto dibujado dentro del esperma. Malpighi y Bonnet (ambos representantes del Preformacionismo) habían estudiado el desarrollo del embrión de pollo y conocían el aumento de complejidad aparente al microscopio (poco coherente con las ideas preformacionistas). A pesar de ello afirmaban que lo que veían era ilusorio y estaba sesgado por los instrumentos usados y los sentidos.

En el siglo XIX, con el nacimiento de la Embriología comparada, comienzan a oírse otras voces que hablan del desarrollo. Ètienne Geoffroy SaintHilaire sugiere que la homología de una estructura puede ser inferida por las relaciones con otras estructuras durante el desarrollo embrionario. A diferencia de Cuvier, que dividía los animales en cinco clases sin relación entre sí, Geoffroy sostenía que todos los animales compartían grandes similitudes en su anatomía.

En el mismo siglo, Haeckel expone su visión recapitulacionista. Considera que la filogenia es la que determina la ontogenia, ya que los animales pasan durante su desarrollo por las formas adultas de otras especies “inferiores”. El cigoto humano representa al protista adulto, la blástula humana representa a los protistas coloniales, el estadio humano de arcos branquiales al pez adulto, etc. Las nuevas especies evolucionan adicionando un paso más en el desarrollo. Igualmente, Haeckel manifiesta que es factible un acortamiento del desarrollo en algún tramo para evitar un infinito tiempo de gestación. En 1828, Karl Ernst Von Baer enuncia algunos puntos importantes en sus leyes del desarrollo:

1. Las características generales de un gran grupo de animales aparecen antes enel embrión que las características particulares de un grupo menor.

2. Los animales no pasan durante su desarrollo a través de formas adultas deotros organismos “inferiores” (contradice a los recapitulacionistas).

3. Los embriones de especies emparentadas comparten estadios tempranos dedesarrollo pero divergen en los estadios más tardíos.

2

4. Los caracteres más especializados se desarrollan a partir de estructurasgenerales compartidas por distintas especies.

El trabajo de Von Baer influyó fuertemente en el pensamiento de Darwin, quien consideró las similitudes en el desarrollo de distintas especies como evidencia de ancestralidad común. Fritz Müller (un acérrimo defensor del Darwinismo) y Francis Balfour resaltan en sus escritos la importancia de la Embriología Comparada como herramienta para dilucidar las relaciones filogenéticos. Afirman también, que la selección natural debe actuar tanto en los estadios larvales como en los adultos.

En el siglo XX, el advenimiento de la “Teoría Sintética de la Evolución” acerca la Genética a la Evolución en detrimento de la Biología del Desarrollo. Esta última es uno de los capítulos faltantes en la “Síntesis”. Las causas de esta importante ausencia pueden haber sido muchas, pero la más importante parecería ser la falta de entendimiento entre genetistas de poblaciones y embriólogos. En los ‘40s Richard Goldschmidt, sostiene que la evolución puede ocurrir sólo por cambios heredables en los genes que regulan el desarrollo. Para él los genes son más que loci y alelos, son unidades de desarrollo. Las nuevas especies se originan como “monstruos esperanzados” (“hopeful monsters”) que resultan de mutaciones sistémicas que afectan el desarrollo del animal. Su visión, al ser contraria a la neodarwinista, no es tenida en cuenta en su época. En 1977, Stephen Jay Gould, publica el libro Ontogeny and Phylogeny y reaviva en el ambiente científico el interés por la Biología del Desarrollo como herramienta para estudiar la evolución.

En 1978, E.B. Lewis descubre los genes Hox (los encargados de dar identidad a cada parte del animal a lo largo del eje anteroposterior) y unos años más tarde comienza a conocerse en profundidad el desarrollo embriológico de Drosophila melanogaster, Caenorhabditis elegans y del ratón. Paralelamente, la Biología Molecular avanza a pasos agigantados y provee nuevas herramientas de análisis. Estos hechos recientes marcan el comienzo de la EvoDevo moderna.

En la actualidad la interfase entre Evolución y Desarrollo atraviesa un renacimiento. En él convergen disciplinas como la Biología del Desarrollo, la Ecología, la Embriología, la Paleontología, la Genómica y el Análisis Filogenético. Esta combinación de enfoques y metodologías tiene como objetivo desentrañar los mecanismos y patrones del desarrollo que originan la diversidad de formas vivientes que existen en nuestro planeta.

En este momento existen dos revistas científicas cuyos artículos tratan exclusivamente temas de Evolución y Desarrollo; estas son: “Evolution and Development” y “Development, Genes and Evolution”.

3

BIBLIOGRAFÍA CONSULTADA

Carroll S.B., Grenier J.K. and Weatherbee S.D. 2005. From DNA to diversity, Second Edition. Blackwell Publishing.

Gilbert S.F. 2003. Developmental Biology, 7th Edition. Sinauer Associates.

Gilbert S.F. 2003. The morphogenesis of evolutionary developmental biology. Int.J.Dev.Biol. 47(78):46777.

Gould S.J. 1977. Ontogeny and Phylogeny. Belknap Press of Harvard University Press.

Hall B.K. 2003. EvoDevo: evolutionary developmental mechanisms. Int.J.Dev.Biol. 47(78):4915.

Wilkins A.S.2002. The evolution of developmental pathways. Sinauer Associates.

4

TRABAJO PRÁCTICO 1 : SEMINARIO

Lea atentamente los siguientes trabajos y responda las preguntas:

1. Gerhing, W.J. & Ikeo, J. 1999. PAX 6: Mastering eye morphogenesis and eyeevolution. TIG 5(9):473479.

1.1. Comente las dos hipótesis sobre el origen y evolución del ojo. Utilice el ejemplo de Arca, Cardium y Pecten para ilustrar estas dos hipótesis.

1.2. Nombre las evidencias que sustentan a Pax6 como un gen maestro en el desarrollo de los distintos ojos de los animales.

1.3. Según los autores la universalidad de dos importantes componentes del ojo de los metazoos como son Pax6 y la rodopsina sugiere fuertemente la existencia de un protoojo y por lo tanto la monofilia de esta estructura. Nilsson (Current Biology 15: R94R95 (2005)) critica esta idea porque no tiene en cuenta que los ojos de insectos y vertebrados tienen distintos orígenes embriológicos y usan distintos tipos de células fotorreceptoras para la visión (células rabdoméricas en insectos y células ciliadas en vertebrados).

¿Qué opina ud.? Muchas veces las definiciones no permiten la existencia de grises. ¿Pueden ser los ojos de los metazoos parcialmente homólogos?

1.4. Analice la figura 4, ¿qué plantea la idea de “intercalary evolution”?

1.5. Gehring e Ikeo sugieren que el protoojo pudo haber surgido por azar. Si los autores estuviesen equivocados y el ojo surge a partir una estructura que tiene la capacidad de sensar la luz pero no es un ojo propiamente dicho (no tiene una célula fotorreceptora y una célula pigmentada) ¿Cuál podría ser la función de esta estructura ancestral?

1.6. En un trabajo reciente (Arendt et al. 2004 Science 306:869871) se descubre que el anélido Platynereis (que supuestamente conserva características del ancestro de los animales bilaterales) tiene células rabdoméricas y células ciliadas asociadas a la fotopercepción. ¿Qué le sugiere este hallazgo? Recuerde los tipos de fotorreceptores (pregunta 3) que usan vertebrados e insectos.

2. Wagner, G.P. 2007. The Developmental genetics of homology. Nat.Genet. 8:473479.

2.1. ¿Cuál es la base del concepto de homología empleado por el autor para genes y caracteres morfológicos?

2.2. Analice las implicancias de la figura 2.

2.3. ¿Qué analogías propone el autor entre genes y caracteres morfológicos?

2.4. ¿Qué propone acerca de las redes genéticas regulatorias (GRN)? ¿Cómo se relacionan con las redes de identidad de caracteres (ChIN)?

5

2.5. ¿Cuál, y de qué tipo, es la relación de homología entre: alas delanteras, alas posteriores, élitros y halterios? ¿Qué ChIN estaría involucrado? ¿Cómo?

2.6. Explique la morfogénesis del ojo en vertebrados y en Drosophila. Compare con la propuesta de Gehring e Ikeo. Indique las hipótesis propuestas para explicar las diferencias.

3. Galis, F. 1999. Why do almost all mammals have seven cervical vertebrae?Developmental constraints, Hox genes and cancer. J.Exp.Zool. 285:926.

3.1. ¿Cuál es la restricción (constraint) al cambio sobre la que versa el trabajo? ¿Existe variación intraespecífica del número de vértebras en las distintas regiones de la columna de los mamíferos?

3.2. ¿Qué distingue una vértebra torácica de una cervical? ¿Qué es una costilla cervical? ¿Cuáles son las patologías asociadas con las costillas cervicales?

3.3. ¿Cuál es la función conservada de los genes Hox en animales bilaterales? ¿Qué patologías tienen los ratones mutantes de algunos genes Hox o Polycomb?

3.4. La restricción al cambio en el número de vértebras cervicales: ¿está dada por ausencia de variación para ese carácter o por selección en contra de variantes de ese carácter? ¿Cómo se presenta en humanos la selección estabilizadora?

3.5. ¿Cuál es la conclusión del autor en cuanto a la relación entre el número conservado de vértebras cervicales en mamíferos, el cáncer y los genes Hox? ¿Por qué esta relación no se presentaría en reptiles y anfibios?

6

Explaining the evolution of an organ as perfect as the eyeis a great challenge for all evolutionary biologists. In his

theory ‘The Origin of Species’ Charles Darwin devoted anentire chapter to the problem. Darwin freely admitted thatthe idea of an eye that is capable of adjusting the focus todifferent distances, of admitting different amounts of lightand of correcting spherical and chromatic aberration, couldhave been formed by natural selection seems intuitivelyabsurd. However, he subsequently found a way out of thisdilemma by postulating a simple and imperfect eye, a proto-type, from which the more perfect visual organs might havearisen gradually, by variation (mutation) and by naturalselection. Darwin assumed the prototype to consist of atleast two cells: an ‘optic nerve’ (photoreceptor cell) and apigment cell shielding the photoreceptor cell from one side,covered by translucent skin, but without any lens or otherrefractive body. Such primitive eyes are found, for example,in some planarians (Fig. 1). Comparative anatomists havediscovered numerous intermediates between this most primi-tive type of eye and the vertebrate eye, such as: eye cups;pinhole eyes; camera-type eyes with a single lens; reflectingmirror eyes; and compound eyes with numerous ommatidia,all of which lends support to Darwin’s theory.

On the basis of comparative anatomical and ultrastruc-tural studies of the various types of eye and photoreceptorcells, it has been postulated by Salvini-Plawen and Mayr1,two strong proponents of darwinism, that photoreceptororgans have originated independently in at least 40, butpossibly up to 65 or more different phyletic lines. However,there are some critical facts that are not consistent with thisconclusion, and we would like to challenge this idea andargue for a monophyletic rather than a polyphyletic originof the metazoan eye. Salvini-Plawen and Mayr argue purelyon morphological grounds. Their section on ‘the multipleorigin of eyes’ begins with the comment that ‘it requires lit-tle persuasion to become convinced that the lens eye of avertebrate and the compound eye of an insect are indepen-dent evolutionary developments’. This point has beentaught to biology students for over a hundred years.However, in a later section (p. 237) these authors describethe observation that, in clams, all three major eye-types [thecamera eye with a single lens (in the heart shell, Cardium),the mirror eye with a lens and a reflecting mirror (in the

scallop, Pecten), and compound eyes that consist of 10–80ommatidia each (in Noah’s arc, Arca noae)], are found inthe same phylogenetic class, the Bivalvia (Fig. 1). All ofthese types of eye are located at the same anatomical pos-ition – the edge of the mantle. The compound eyes of Arcaare similar to those of arthropods, but they have only a sin-gle photoreceptor cell per ommatidium, whereas insects andcrustaceans generally have eight or nine visual cells per unit.Salvini-Plawen and Mayr interpret the compound eyes ofArca as new formations, but an equally valid interpretationof these data is to assume that the camera-, mirror- andcompound eyes of clams have evolved monophyleticallyfrom a common ancestral precursor. A monophyletic originfor the eye is also supported by the observation that allmetazoans share the same visual pigment, rhodopsin.

Darwin was highly self-critical in his discussion of theeye prototype and admits that the origin of the prototypecannot be explained by natural selection, because selectioncan only drive the evolution of an eye once it is partlyfunctional and capable of light detection. Therefore, selec-tion cannot explain the origin of the eye prototype, whichfor Darwin represents the same problem as the origin oflife. Therefore, both the origin of life and the origin of theeye prototype must have been very rare events, and a poly-phyletic origin in over 40 different phyla is not compatiblewith Darwin’s theory. In this review, we discuss morerecent evidence in favor of a monophyletic origin of theeye and propose a new hypothesis explaining how morphogenetic pathways might have evolved.

Pax 6 is a master control gene for eyemorphogenesis and evolutionHomeotic mutations in Drosophila have resulted in theidentification of several master control genes that specifythe body plan by controlling anterior–posterior polarity,segmental identity, organogenesis and identity of individ-ual cells in great detail. The term ‘master control genes’was introduced by Lewis2 for the homeotic genes of theBithorax Complex, and, perhaps the most impressivedemonstration of their role in development has been thegenetic construction of four-winged and eight-legged flies3.Targeted expression of the homeotic Antennapedia generesults in complete middle legs being induced in the antennal

PerspectivesEye morphogenesis and evolution

TIG September 1999, volume 15, No. 90168-9525/99/$ – see front matter © 1999 Elsevier Science Ltd. All rights reserved. PII: S0168-9525(99)01776-X

Walter J. [email protected]

*Kazuho [email protected]

Department of CellBiology, Biozentrum,University of Basel,Klingelbergstrasse 70,4056 Basel, Switzerland.*Center for InformationBiology, NationalInstitute of Genetics,Mishima, Yata 411, Japan.

371

Pax 6 genes from various animal phyla are capable of inducing ectopic eye development, indicating that Pax 6is a master control gene for eye morphogenesis. It is proposed that the various eye-types found in metazoa arederived from a common prototype, monophyletically, by a mechanism called intercalary evolution.

Pax 6mastering eye morphogenesisand eye evolution

7

discs of Drosophila3. Another striking example of a mastercontrol gene is Pax 6. This gene was first cloned in themouse4,5 and in humans6 and subsequently shown to beaffected in the mouse mutant, Small eye, and in humanAniridia patients. In humans and mice, eye defects are as-sociated with Pax 6 mutations in heterozygotes. Thehomozygous Pax 6 mutation is lethal to mouse embryos:they lack eyes and a nose, and also have brain damage. Pax6 is expressed from the earliest stages of eye morphogenesisin the optic vesicle, giving rise to the retina and pigmentretina, as well as in the overlying ectoderm that later formsthe lens and the cornea. However, Pax 6 is also expressedin the nasal epithelium, in specific regions of the brain andthe spinal cord, and not exclusively in eye primordia.

Pax 6 encodes a transcription factor that contains apaired domain and a homeodomain. The Pax gene familyclearly illustrates that novel genes are generated in the

course of evolution by recombining parts of pre-existinggenes in a process that Jacob called evolutionary tinkering7.The different Pax genes contain various combinations ofpaired domains, with homeodomains, a sequence calledoctapeptide, or parts of the homeodomain and paireddomain, respectively. The murine and human PAX 6 pro-teins are identical in amino acid sequence. A Pax 6 homologin Drosophila8 was subsequently discovered and this alsoshows extensive sequence similarity, both in the paireddomain (94% identity), and in the homeodomain (90%identity). More surprising is the finding that the DrosophilaPax 6 homolog is the eyeless (ey) gene known by a mutationaffecting the eyes since 1915 (Ref. 9). This was unexpectedbecause of the long-standing dogma, mentioned above, thatthe insect compound eye was non-homologous to the verte-brate camera eye, and that the two types of eye had evolvedindependently. The observation that Pax 6 homologs of

Perspective Eye morphogenesis and evolution

TIG September 1999, volume 15, No. 9372

FIGURE 1. Eye evolution

A hypothetical scheme of the evolution of various eye-types from a common ancestral prototype. As a first step, photosensitive cells with a light receptor (opsin)have evolved. Under the control of the Pax 6 gene, the photosensitive cell assembles with a pigment cell to form an organ, the prototype eye. By divergent, paralleland convergent evolution, the various eye-types are generated from the prototype: the compound eye of insects; the camera-type eye of vertebrates; and the largespectrum of eye-types in molluscs ranging from the primitive camera-type eye in Cardium, the mirror-plus-lens eye of Pecten, the compound eye of Arca to the highlyevolved cephalopod eye, that greatly resembles the vertebrate camera-type eye.

8

both mammals and insects are essential for eye morpho-genesis led to the idea that Pax 6 might be the universalmaster control gene for eye morphogenesis and evolution8.

We tested the master control gene hypothesis by con-structing a gain-of-function mutation. In wild-type larvae,ey is expressed exclusively in the eye-antennal disc from theearliest stages when the disc primordia are formed in theembryo. Therefore, we used the Gal-4-system to target geneexpression into imaginal discs other than eye discs10. By theuse of different genomic enhancer lines, we were able toinduce ectopic eyes on the legs, wings, halteres and theantennae of the fly, and recent electrophysiological exper-iments show that the ectopic eyes on the antenna can gener-ate a normal electroretinogramme, which indicates thatthey are functional (P. Callaerts and W. Gehring, unpub-lished). This illustrates the role of ey as a master controlgene that is capable of switching on a cascade of some 2500genes required for eye morphogenesis10. Of course, eye morphogenesis cannot be induced in any tissue of the fly at any stage of development, but at least it does occur in all imaginal discs up to a certain stage of differentiation.The master control gene first has to interact with subordi-nate control genes to repress the resident genetic pro-gramme and to install the eye programme. If the cells haveproceeded too far along their pathways and are firmlylocked into a different pathway, the ectopic expression of eyhas no effect.

Our next query was whether the mammalian Pax 6gene can functionally substitute for the Drosophilahomolog. The ectopic expression of mouse Pax 6 inDrosophila induces ectopic compound eyes10, suggestingthat Pax 6 has a universal function of gene regulation ineye morphogenesis. The reciprocal experiment has notbeen completed yet, but it has been reported that XenopusPax 6 is capable of inducing ectopic eye lenses11. However,by changing the timing and site of Pax 6 RNA injectioninto the Xenopus embryo, it is possible to induce completeectopic eyes (R. Chow, C. Altmann, R. Lang and A. Hemmati-Brivanlou, pers. commun.). These findingsclearly indicate that Pax 6 is a master control gene for eyemorphogenesis in both insects and vertebrates.

The protein-coding regions of Pax 6 are highly con-served in evolution, as are some of the regulatorysequences in the promoters and enhancers. Consequently,the regulatory mechanisms that direct ocular expressionare also conserved between flies and mice. The eye-specificenhancer region of the Drosophila ey gene8,12, wheninserted upstream of either of the two mouse Pax 6 pro-moters (P1 or P0), directs eye- and CNS-specific expres-sion in transgenic mice that accurately reproduces featuresof endogenous Pax 6 expression13. In a reciprocal exper-iment, the mouse P1 element is able to direct lacZ reportergene expression into the eye imaginal discs of Drosophila.Here, the expression is restricted to the photoreceptorcells, although lacZ expression is delayed and occurs onlyposterior to the morphogenetic furrow, whereas endogen-ous ey expression is confined to the undifferentiated cellsanterior to the morphogenetic furrow. However, theDrosophila ey enhancer itself shows the same spatio–temporal expression pattern as the mouse promotor, that could reflect perdurance of b-galactosidase or lack of regulatory sequences that confer repression posterior tothe morphogenetic furrow12. Overall, there is evidence for conservation of Pax 6 gene regulation, but there isuncertainty about the extent of the conservation.

Genuine Pax 6 genes have now been isolated from:mammals; amphibians; fish; amphioxus; sea squirts; seaurchins; squid; nematodes; ribbonworms; and planarians(Fig. 2). In Cnidarians the situation is less clear, because thegenes found so far are either precursors of Pax 6 or havediverged too far to be clearly identified as Pax 6 homologs.In any case, this survey shows that Pax 6 was present in thelast common ancestor of all these triploblastic phyla, muchlike the rhodopsin gene. In addition to the mammalian Pax6 gene, its homologs from the sea squirt Phallusia and thesquid Loligo are also capable of inducing ectopic eyes inDrosophila. With the exception of sea urchins andCaenorhabditis elegans (which presumably have lost theireyes during evolution because eyes are found in other echino-derms and nematodes), all Pax 6 genes examined so far areexpressed prominently in the developing eyes, includingthose of planarians, which come close to the darwinianprototype. Furthermore, Pax 6 is specifically expressed inthe differentiated eyes of the ribbonworm Lineus14 and par-ticularly during eye regeneration15, strengthening the corre-lation between eye morphogenesis and Pax 6 expression.

The evolution of Pax 6: twin of eyelessMore recently, a second Pax 6 gene homolog inDrosophila called twin of eyeless (toy) was identified16. Itshares 91% sequence identity in the paired domain and90% in the homeodomain with the human and murinePAX 6 proteins (Fig. 2), compared with 95% and 90% forEY. Outside of these highly conserved domains, TOY ismore similar to the mammalian proteins than EY, particu-larly in its overall length and at the C-terminus, where itshares a transcriptional activation domain with other PAX6 proteins that is absent in EY. A survey by polymerasechain reaction (PCR) shows that two Pax 6 genes are onlyfound in holometabolous insects (Drosophila andBombyx) and not in hemimetabolous (grasshopper) orapterygote insects (springtail), nor in all other phylatested16. This indicates that the gene-duplication eventleading to the two paralogs occurred during insect evolu-tion, a conclusion that is also supported by the molecularphylogenetic analysis (Fig. 3). Besides the sequence simi-larity, the localization of the intron splice sites clearly indi-cates that both paralogs are bona fide Pax 6 genes (Fig. 2).The first splice site at the N-terminus of the paired domainis missing in toy, but present in ey, whereas the secondsplice site in the homeodomain is present in toy and absentin ey, indicating that the ancestral gene had all four splicesites in the two boxes. The same four splice sites are alsofound in the nematode Caenorhabditis elegans and threeout of four can be traced back to platyhelminths(Dugesia). This indicates that these introns are very old(precambrian) and that a bona fide Pax 6 gene must havebeen present in the last common ancestor of triploblasticanimals. Vertebrates share a splice site at codon 44/45 thatis vertebrate-specific and is used for differential splicing inthe paired box. It is absent in amphioxus and ascidians,indicating that this intron arose later in evolution, aftervertebrates had separated from invertebrates.

Following gene duplication during insect evolution, thetwo paralogs ey and toy began to diverge in function. Inparticular, toy is expressed much earlier, at the blastodermstage, when the Drosophila body plan is laid down,whereas ey is expressed only later, during germbandextension. The spatial patterns at later stages are verysimilar although not identical. This earlier divergence with


TIG September 1999, volume 15, No. 9 373

9

respect to temporal rather than spatial patterns of geneexpression has been found in other duplicated develop-mental control genes, like sloppy-paired 1 and 2 (Ref. 17),and might be a more general feature of evolution. Like ey,toy is also capable of inducing ectopic eyes in Drosophila,but toy requires a functional ey gene to induce eyes, suggesting that toy is upstream of ey in the genetic cascadecontrolling eye morphogenesis. Epistasis experiments, aswell as biochemical and transgenic analyses, support thenotion that toy acts upstream of ey in the eye develop-

mental pathway18 by directly regulating the eye-specificenhancer of the ey gene12,16. This observation reveals an interesting facet of the evolution of morphogeneticpathways: the single Pax 6 in vertebrates is autoregulatedby a positive feedback loop in which the PAX 6 proteinbinds to the enhancer in its own gene and activates itstranscription19. In Drosophila, after gene duplication this positive autocatalytic feedback loop appears to have evolved into a heterocatalytic loop in which one of the paralogs regulates the other, leading to the

Perspectives Eye morphogenesis and evolution


FIGURE 2. Metazoan Pax 6 proteins

Comparison of the amino acid sequences for PAX 6 proteins from various metazoa. The paired domains are indicated in (a) and the homeodomains in (b). PAX 6protein-specific amino acids are shaded more darkly. The positions of the intron splice sites are indicated by arrowheads. These have not yet been determined forAmphioxus, Paracentrotus and Loligo. The numbers indicate the percentage amino acid sequence identity as compared with the mouse and human proteins. Forcomparison the closely related Pax sequences from the mouse (m) are shown. Pax 2, 5 and 8 have only partial homeodomains. a, a-helices; b, b-sheets.

b1(a) b2 a1 a2 a3

a4 a5 a6

a1(b) a2 a3 / 4

1 10 20 30 40 50 60 70 ��

..................C........K.K..........��

80 90 100 110 120 130��

9

70

1 10 20 30 40 50 60 ��

trends in genetics

SHSGVNQLGGVFVNGRPLPDSTRQKIVELAHSGARPCDISRILQVSNGCVSKILGRYYETGSIRPRAIGGSKPRVA �G............G.........R.......Q.........L.................................. �G.........M..........I......F..N......................A......T.............. �G........................................................................... �G.......................R................................................... �G.......................R.........................T......................... �G............G.............................................................. �G...I......Y...................................................K............ �G.T.................A...R..D...K.C.......L............C....S.T.............. �G...I.....I.........V...R.I..SQ.......................C........K.K.......... ��N .G.I.....T.......IEPV.R.......Q.V.......Q.R..H.......S.F.....V...V......K.. � L.S......L........LD...Q..Q..IR.M.......S.K..............R..VLE.KC.......L. �R.G.................VV..R......Q.V.......Q.R..H................K.GV......K.. �G.G.................VV..R......Q.V.......Q.R..H................K.GV......K.. �G.G.L.....A........EVV..R..D...Q.V.......Q.R..H..................GV......K..

Homo, MusAmphioxusPhallusiaParacentrotusLoligoLineusDrosophila EYDrosophila TOYCaenorhabditisDugesia

Hydra (B)Pax4Pax2Pax5Pax8

TPEVVSKIAQYKRECPSIFAWEIRDRLLSEGVCTNDNIPSVSSINRVLRNLASEKQQMGADG---MYDKLRMLNGQ 100 �.....A....F....................I...E.................GEKNTL-(X) -.LE...L...N 92 �..Q..N...M..................N.A..NAE...............NG.NGRVFSEGNSPNKNHLPDDWSS 87 �..H..TR..H..................A.KI.NQE...................TMGHGD----.F......... 91 �.....Q....F.......................Q.................G.N.KVLGQ.-TTT....GL.... 96 �.....G...H...................DA..NQ...................N.KQLGQSS--.....GL.... 93 �.A......S...................Q.N.....................AQ.E.QS-(X) -I.E...L..T. 95 �.TP..Q...D....................Q..NS..................Q.E.QAQQQNESV.E....F... 91 �.SD..E..ED...DQ..........K..ADNI.N.ET...............AK.E.VTMQTE--L..RI.IVDNF 80 �.NT..R.VTI..Q.S..M........P.QD...NQ..L..I.....I..S..N.SPSSNQTFKSSLSNSHQLSLSN 77 ��. .S..A..QE..QHN.TM.......K....QI.DS.SV........IV..RLGS 68 �..A..AR...L.D.Y.AL.....QHQ.CT..L..Q.KA...........A.QED 70 �..K..D...E...QN.TM..........A..I.D..TV........II.TKVQQ 77 �..K..E...E...QN.TM..........A.R..D..TV........II.TKVQQ 77 �..K..E..GD...QN.TM..........A....D..TV........II.TKVQQ 75



DEAQMRLQLKRKLQRNRTSFTQEQIEALEKEFERTHYPDVFARERLAAKIDLPEARIQVWFSNRRAKWRREEKLRNQRR 100 �....A..R....................................................................... 100 �KDTNA...............S...V......................S.........................M.H..G 95 �ED..A..R.............AQ...E....................Q............................... 93 �TDE...IR.............AA.......G................HQ............................P. 92 �S.E...IR.............NA........................Q............................... 93 �EDD.A..I.............ND..DS....................G..G............................ 90 �EDS....R............SN...DS....................D..G......................M.T... 90 �.D.AA.MR..............V...S....................Q..Q......................M..K.S 93 �RYSNTESK.SK.S..S.....ND..NL..............S..K.SQNLKVA.T..................SEENNM 73 ��

M R.V..T.SL..RR...DA..K.P...AEQ..EISIQC....P.V......K...L..QD 55 �SH...AI.SPG.A.......Q.GQ...SV..GK...ATS...DTVR............Q. 62 �

...Q.L...DRV...PS..... �

...Q.L.V.DRV...Q..S.I. �

.S.HHL....CP...Q...EAY ��



10

integration of ey into the eye developmental pathway underneath toy.

The genetic cascade specifying the eyedevelopmental pathwayFollowing the discovery of ey as a master control gene, several groups have embarked upon the analysis of thegenetic cascade leading to eye morphogenesis by identifyingtarget genes and genetic interactions. However, the directnature of a given genetic interaction and the molecular basisof the interaction has been demonstrated in only a fewcases. In Drosophila, evidence for a direct activation of eytranscription by binding of TOY protein to the eye-specificenhancer of the ey gene has been described above. This putsthe toy gene on top of the hierarchy and ey underneath16.The toy gene requires ey to induce eye formation; in turn, eyinduces and requires sine oculis (so) and eyes absent (eya)for the induction of ectopic eyes18. There is strong evidencethat so is a direct target for EY protein20. However, as moreand more pieces are filled into the puzzle, the simple linearpathways turn into a complex network and several othergenes have been found to be capable of ectopic eye induc-tion. The so gene encodes a homeodomain protein that isrequired for the development of the entire visual system inDrosophila21,22. The eya gene encodes a novel type ofnuclear protein involved in the development of the visualsystem as well as in the somatic gonadal precursors23,24.

A gene called dachshund (dac) encodes a novel nuclearprotein that is required for differentiation of the ommatidia,but is also essential for leg development25,26. The ectopicexpression of eya or dac alone or in combinations of eyawith so or dac induces ectopic eye formation, but also activates ey expression. The ey, eya and dac genes are allactivated during eye induction18 and there is evidence thatthe EYA protein forms a complex with SO (Ref. 27) andDAC (Ref. 28) proteins. Taken together, these findings canbe explained by a model in which ey induces the initialexpression of so and eya that regulates the activity of allfour genes by positive feedback loops required for eyeinduction16.

Targeted expression of the gene teashirt (tsh), whichwas shown to be required for the specification of the trunksegments in the Drosophila embryo, can also induceectopic eyes29. This gene encodes a transcription factorwith zinc-finger motifs and induces the expression of ey,so and dac. In turn, ey induces the expression of tsh, indi-cating that tsh is also a member of the regulatory networkof genes that are connected to each other by positive feed-back loops. However, it should be emphasized that ey is amuch more potent inducer of ectopic eyes than any singlegene in the later group, suggesting that no single gene canrecapitulate the entire spectrum of ey activity, reinforcingthe master control gene status of Pax 6.

A second Pax gene, eyegone (eyg) apparently acts inparallel with ey in determining Drosophila eye develop-ment30. This gene contains only a partial paired domain,but a complete homeodomain. Loss-of-function mutationslead to a reduction of the eyes similar to ey, and ectopicexpression leads to the induction of ectopic eyes. The twogenes eyg and ey seem to have complementary functionsbecause their coexpression leads to a synergistic enhance-ment of ectopic eye formation. The expression of eyg isnot regulated by ey at the transcriptional level, nor does itregulate ey expression. However, homozygous ey:eygdouble mutants are lethal, which indicates that the two

genes interact. It has been proposed that the two proteinproducts can form a heterodimer, which is compatiblewith the findings mentioned above30.

One of our aims is to compare the genetic cascade fromDrosophila with that of the mouse or other vertebrates tofind out how many other genes besides Pax 6 and therhodopsin gene have been conserved during evolution.Several homologs for so and eya have been identified in ver-tebrates, and a second so-like gene has also been isolatedfrom Drosophila31. However, sequence conservation of theprotein-coding region does not necessarily imply that thefunction in eye morphogenesis is also conserved in evolution.For example, the mouse Rx gene that belongs to the paired-like class of homeobox genes was shown to be expressedboth in the developing retina and forebrain. Loss-of-functionmutants in mice do not form optic cups and, as a conse-quence, lack eyes32. Furthermore, misexpression of Rxinduces ectopic retinal tissue in frogs32. However, aDrosophila homolog of Rx that has 100% sequence identityin the homeodomain is expressed only in the developingbrain, but not in the embryonic or the larval eye primordia33.Eventually, it will be interesting to find how many new genesmust be recruited into the eye-developmental pathway togenerate either a mouse or a Drosophila eye, and how manyof these genes are common. However, the major changesoccurring during evolution are likely to occur at the level ofgene regulation, and very different types of eye might be generated by the same set of regulatory genes.

The evolution of the different types of eyeThe evolution of light-sensitive cells is intimately con-nected to the evolution of the visual pigment rhodopsin.Rhodopsin is the molecule of ultimate sensitivity becauseit is capable of sensing a single light quantum. Absorptionof a single quantum of light converts all-trans retinal, thatis covalently bound to the opsin protein molecule, into 11-cis retinal. This conversion causes a conformationalchange of the protein that is amplified by transducin, a G-protein and results in an electrical nerve impulse34.



FIGURE 3. Phylogenetic tree of the Pax 6 genes

The neighbor-joining method was used to generate a phylogenic tree of the Pax 6 genes from variousmetazoa. Note that Drosophila melanogaster eyeless and twin of eyeless are closely related. The scaleshows the number of amino acid substitutions per site. The monophyly of the eyeless/Pax 6 group ofgenes is strongly supported by the phylogenetic analysis of Jacobs et al.42

trends in genetics

Dugesia

Caenorhabditis elegans

Phallusia

toy

eyelesssquid

ribbon worm

quail

Drosophila melanogaster

mouse

human

Xenopus laevis

zebrafish

medaka fish

amphioxus

sea urchin

0.05

11

Rhodopsins are present in some bacteria, and some ofthese proteins also serve a sensory function. However,there is very little sequence conservation between bacterio-rhodopsins and rhodopsins of higher organisms, eventhough both are structurally similar membrane proteinswith seven transmembrane domains.

Protists have also developed visual systems that are basedon rhodopsin. The unicellular green alga Chlamydomonashas developed a visual system that allows it to measure lightintensity, as well as to determine the direction of the incidentlight. These abilities confer a strong selective advantage for anorganism that depends on photosynthesis35. The direction ofthe incident light is determined with the help of the eyespot, acarotenoid-containing vesicle that presumably operates as aninterference reflector. The action spectra for phototaxis andflash-induced phobic responses have a maximum close to550 nm like rhodopsin, and in blind retinal-deficient cells,positive phototaxis can be restored by supplying the cells withall-trans retinal. Chlamydorhodopsin has recently beencloned36, and it shows some sequence homology to inverte-brate rhodopsins. However, it is not a typical seven-trans-membrane receptor, and looks instead rather like an ionchannel. Therefore, this primitive plant rhodopsin probablydiverged from animal opsin early in evolution. In all verte-

brates and invertebrates analyzed so far, typical rhodopsinsbelonging to one and the same gene family have been found.

The visual system of unicellular organisms is an organelle,rather than an organ, and it is formed by intracellular assembly processes, whereas the eyes of metazoa are organsmade up of cells of at least two different types or of differenttissues, as already pointed out by Darwin. There is accu-mulating evidence that Pax 6 is the universal master controlgene for eye morphogenesis in metazoa ranging from platy-helminths to humans. The universality of rhodopsin and Pax6 suggests that the different types of eye found in metazoaare derived from a single prototypic eye and are, therefore,of monophyletic origin. Pax 6 serves as a regulatory gene toassemble the different cell-types, such as photoreceptor cellsand pigment cells, into a light-sensing organ. This new con-cept of eye evolution is illustrated in Fig. 1. Originating froma precambrian prototype, the various types of eye arethought to have evolved by divergent, parallel and conver-gent evolution by recruiting numerous additional genes into the eye-developmental pathways, as discussed in the following section.

In higher metazoa, the eyes are connected to the brain,where visual information is processed and transmitted to theeffector organs, such as muscles. In the more primitive(ancestral) cnidarians, such as cubomedusae (which do nothave a brain, but only a nerve ring around the umbrella), theeyes are directly connected to the muscles in the tentacles.This suggests that the eye evolved as an information-gather-ing organ before the brain, the information-processingorgan.

Evolution of biosynthetic and morphogeneticpathwaysHorowitz37 has proposed a mechanism for the evolutionof biosynthetic (or biochemical) pathways that is based onthe idea of retrograde evolution (Fig. 4a). This hypothesisassumes that, for example, the nine enzymes in histidinebiosynthesis evolved in a retrograde fashion. Presumably,primitive organisms had to take up histidine from theenvironment. The organisms that evolved the last enzymein the pathway presumably had a strong selective advan-tage when the supply of histidine (Z) in the environmentwas exhausted, because it could use compound Y and con-vert it to Z. The next step was the evolution of enzymeand so on, until all nine enzymes had evolved that made itpossible to achieve histidine biosynthesis from PRPP andATP. A similar mechanism of retrograde evolution hasbeen proposed for the evolution of the sex-determinationpathway38.

Based on a similar kind of logic, we propose that morphogenetic (or developmental) pathways evolve byintercalary evolution (Fig. 4b). Prerequisite is the prior evolution of rhodopsin and of Pax 6 to generate the proto-typic eye. The prototype, as pointed out by Darwin, cannotbe explained by selection, because selection can drive evo-lution only when the eye can function at least to a smallextent. Once the prototype has evolved, presumably by sto-chastic events, selection can optimize it by a mechanismthat can be called intercalary evolution to distinguish itfrom retrograde evolution mentioned above. The proto-type has acquired two key genes, Pax 6 on the top, andrhodopsin at the bottom of the genetic cascade. In-creasingly complex and more-sensitive eyes can be gener-ated by the intercalation of genes into the cascade (Fig. 4b).At least three genetic mechanisms for intercalation are

Perspectives Eye morphogenesis and evolution


FIGURE 4. Retrograde and intercalary evolution

(a) Hypothetical retrograde evolution of histidine biosynthesis as proposed by Horowitz37. The lastenzyme (E9) of the biosynthetic pathway evolves first, followed by E8 in a second step. This proceedsuntil all nine enzymes are lined up in a linear pathway. (b) Proposed intercalary evolution ofmorphogenetic pathways. First a rhodopsin-containing photosensitive cell has to evolve, that underthe control of Pax 6 is assembled with a pigment cell to form a functional eye prototype. The top of the cascade is formed by a master control gene (Pax 6), the bottom by essential structural genes,such as rhodopsin. In the course of evolution, new genes are intercalated between the top and bottomof the cascade: regulatory genes, such as eyeless downstream of twin of eyeless; and structuralgenes, such as the lens crystallin genes. The morphogenetic pathway is not linear but, rather, acomplex network.

trends in genetics

(a)

Histidine biosynthesis

(b)

Eye morphogenesis

Intercalation of genes

A

Evolution

(0) Histidine has to be taken up from the environment

(1) Step

(2) Step

(3) Step

Z

Z

Z

Z

Y

Y

Y

X

Y

Z

RhodopsinPax 6

X

XW

ATP + PRPP (9 enzymes) Histidine

E1B

A B

C

C

D

WE2 E7

XE8

YE9

E7

E8

E9

Z

12

known. First, gene duplication and divergence, as de-scribed for ey and toy. The original autocatalytic feedbackloop is converted to a heterocatalytic loop, where toyregulates ey, the latter becoming intercalated into the eyemorphogenetic path-way downstream of toy16. Second,recruitment of novel genes into the morphogenetic path-way by fusion of the coding region of a gene to an eye-specific enhancer or promoter. Piatigorsky39 has describedseveral examples of this kind. Genes encoding enzymes likeenolase or lactate dehydrogenase, or small heat shock pro-teins are recruited into the eye morphogenetic pathway aslens proteins called crystallins. This evolutionary process

has been termed gene sharing or recruitment. Third, therecombination of various coding and regulatory regions ofdifferent genes by‘evolutionary tinkering’ might also leadto recruitment and intercalation into a new morphogeneticpathway.

These considerations clearly have a bearing on our concepts of homology. Homology is not an all-or-nothing phenomenon, because two different types of eye might only be partially homologous and they can alsohave acquired analogous features as proposed byZuckerkandl40. This will resolve discrepancies in the inter-pretation41 of these new findings in eye evolution.



References1 Salvini-Plawen, L. and Mayr, E. (1961) in Evolutionary Biology,

(Vol. 10) (Hecht, M.K., Steere, W.C. and Wallace, B., eds), pp. 207–263, Plenum Press

2 Lewis, E.B. (1992) Clusters of master control genes regulate thedevelopment of higher organisms. J. Am. Med. Assoc. 267,1524–1531

3 Schneuwly, S. et al. (1987) Redesigning the body plan ofDrosophila by ectopic expression of the homeotic geneAntennapedia. Nature 325, 816–818

4 Walther, C. and Gruss, P. (1991) Pax 6, a murine paired box gene,is expressed in the developing CNS. Development 113, 1435–1449

5 Hill, R.E. et al. (1991) Mouse Small eye results from mutations ina paired-like homeobox-containing gene. Nature 354, 522–525

6 Ton, C.C.T. et al. (1991) Positional cloning and characterizationof a paired box- and homeobox-containing gene from the Aniridiaregion. Cell 67, 1059–1074

7 Jacob, F. (1997) Evolution and tinkering. Science 196, 1161–11668 Quiring, R. et al. (1994) Homology of the eyeless gene of

Drosophila to the Small eye gene in mice and Aniridia in humans.Science 265, 785–789

9 Hoge, M.A. (1915) Another gene in the fourth chromosome ofDrosophila. Am. Nat. 49, 47–49

10 Halder, G. et al. (1995) Induction of ectopic eyes by targetedexpression of the eyeless gene in Drosophila. Science 267,1788–1792

11 Altmann, C.R. et al. (1997) Lens induction by Pax 6 in Xenopuslaevis. Dev. Biol. 185, 119–123

12 Hauck, B. et al. Functional analysis of an eye specific enhancer of theeyeless gene in Drosophila. Proc. Natl. Acad. Sci. U. S. A. (in press)

13 Xu, P-X. et al. (1999) Regulation of Pax 6 expression is conservedbetween mice and flies. Development 126, 383–395

14 Loosli, F. et al. (1996) Isolation of a Pax 6 homolog from theribbon worm Lineus sanguineus. Proc. Natl. Acad. Sci. U. S. A. 93,2658–2663

15 Tarpin, M. et al. Reverse homeosis in homeotically reconstructedribbonworms. Nature (in press)

16 Czerny, T. et al. (1999) Twin of eyeless, a second Pax 6 gene ofDrosophila, acts upstream of eyeless in the control of eyedevelopment. Molecular Cell 3, 297–307

17 Cadigan, K.M. et al. (1994) Functional redundancy: the respectiveroles of the two sloppy paired genes in Drosophila segmentation.Proc. Natl. Acad. Sci. U. S. A. 91, 6324–6328

18 Halder, G. et al. (1998) Eyeless initiates the expression of bothsine oculis and eyes absent during Drosophila compound eyedevelopment. Development 125, 2181–2191

19 Plaza, S. et al. (1993) Quail Pax 6 (Pax-QNR) encodes atranscription factor able to bind and trans-activate its ownpromoter. Cell Growth Differ. 4, 1041–1050

20 Niimi, T. et al. Direct regulatory interaction of the eyeless proteinwith an eye-specific enhancer in the sine oculis gene during eyeinduction in Drosophila. Dev. Biol. (in press)

21 Cheyette, B.N.R. et al. (1994) The Drosophila sine oculis locusencodes a homeodomain-containing protein required for thedevelopment of the entire visual system. Neuron 12, 977–996

22 Serikaku, M.A. and O’Tousa, J.E. (1994) sine oculis is a homeoboxgene required for Drosophila visual system development.Genetics 138, 1137–1150

23 Bonini, N.M. et al. (1993) The eyes absent gene: genetic control ofcell survival and differentiation in the developing Drosophila eye.Cell 72, 379–395

24 Bonini, N.M. et al. (1997) The Drosophila eyes absent gene directsectopic eye formation in a pathway conserved between flies andvertebrates. Development 124, 4819–4826

25 Mardon, G. et al. (1994) dachshund encodes a nuclear proteinrequired for normal eye and leg development in Drosophila.Development 120, 3473–3486

26 Shen, W. and Mardon, G. (1997) Ectopic eye development inDrosophila induced by directed dachshund expression.Development 124, 45–52

27 Pignoni, F. et al. (1997) The eye-specification proteins So and Eyaform a complex and regulate multiple steps in Drosophila eyedevelopment. Cell 91, 881–891

28 Chen, R. et al. (1997) Dachshund and Eyes Absent proteins form

a complex and function synergistically to induce ectopic eyedevelopment in Drosophila. Cell 91, 893–903

29 Pan, D. and Rubin, G.M. (1998) Targeted expression of teashirtinduces ectopic eyes in Drosophila. Proc. Natl. Acad. Sci. U. S. A.95, 15508–15512

30 Jang, C-C. et al. Two Pax genes, eye gone and eyeless, act in parallelin determining Drosophila eye development. Development (in press)

31 Toy, J. et al. (1998) The Optx2 homeobox gene is expressed inearly precursors of the eye and activates retina specific genes.Proc. Natl. Acad. Sci. U. S. A. 95, 10643–10648

32 Mathers, P.H. et al. (1997) The Rx homeobox gene is essential forvertebrate eye development. Nature 387, 603–607

33 Eggert, T. et al. (1998) Isolation of a Drosophila homolog of thevertebrate homeobox gene Rx and its possible role in brain andeye development. Proc. Natl. Acad. Sci. U. S. A. 95, 2343–2348

34 Khorana, H.G. (1992) Rhodopsin, photoreceptor of the rod cell.J. Biol. Chem. 267, 1–4

35 Foster, K.W. et al. (1984) A rhodopsin is the functionalphotoreceptor for phototaxis in the unicellular eukaryoteChlamydomonas. Nature 311, 756–759

36 Deininger, W. et al. (1995) Chlamyrhodopsin represents a newtype of sensory photoreceptor. EMBO J. 14, 5849–5858

37 Horowitz, N.H. (1945) On the evolution of biochemical syntheses.Proc. Natl. Acad. Sci. U. S. A. 31, 153–157

38 Wilkins, A.S. (1995) Moving up the hierarchy: a hypothesis on theevolution of a genetic sex determination mechanism. BioEssays17, 71–11

39 Piatigorsky, J. and Wistow, G.J. (1989) Enzyme/crystallins: genesharing as an evolutionary strategy. Cell 57, 197–199

40 Zuckerkandl, E. (1994) Molecular pathways to parallel evolution:I. Gene nexuses and their morphological correlates. J. Mol. Evol.39, 661–678

41 Abouheif, E. et al. (1997) Homology and developmental genes.Trends Genet. 13, 432–433

42 Jacobs, D.K. et al. (1998) in Molecular Approaches to Ecology andEvolution (De Salle, R. and Schierwater, B., eds), pp. 323–357,Birkhäuser

The Internet Section

The Internet section is a regular column of news and information about webresources for researchers in genetics and development (pp. 378–379).Internet is compiled and edited with the help of: Steve Brenner (Departmentof Structural Biology, Stanford University, Fairchild D-109, Stanford, CA94305-5400, USA; [email protected]); Fran Lewitter (ScientificComputing, Whitehead Institute for Biomedical Research, Nine CambridgeCenter, Cambridge, MA 02142-1479, USA; [email protected]); and Laurie

Iten (Department of Biological Sciences, Purdue University, 1392 Lily W. Lafayette, IN 47907-1392,USA; [email protected]).

If you would like to announce or publicize an internet resource, please contact: [email protected]

13

Characters found in different species are homologous if they are derived from the same character in their most recent common ancestor (MRCA), regardless of similarity in form or function (FIG. 1). Whenever we compare two or more species, or use a model organism to learn about the molecular basis of human disease, we implicitly need to identify corresponding body parts and functional systems; that is, we make assess-ments about homology. Intuitively, one would expect that the historical continuity of morphological characters is underpinned by the continuity of the genes that govern the development of these characters. However, things are not that simple: one of the most important results of the past 15 years of molecular developmental genetics is the realization that homologous characters can have different genetic and developmental bases1–3. This seems paradoxical, because the historical continuity of morphological characters implies continuity of the (genetic) information about the characters4. But where else should we look for this continuity, other than in the genes? Here I review some of the conceptual issues, and the experimen-tal results that suggest a solution to this conundrum. I argue that the continuity of morphological characters could be under-written by homologous regulatory networks of co-adapted transcription factor genes, whereas other aspects of their development

can be variable. These networks control the execution of character-specific developmental programmes, which allow for quasi-independent variation of characters5 with respect to other parts of the body.

What does homology mean?

In the eighteenth and nineteenth centuries, it became clear that the similarities and differ-ences among organisms are not random, but follow patterns that call for an explanation. Most intriguing are the similarities among some body parts that cannot be explained by shared functional necessity. For example, the tetrapod limb shows a highly stereotypical pattern of bony elements (FIG. 2), regardless of whether it is used for running, flying, swimming or grasping. This pattern was conceptualized by Richard Owen as homol-ogy6, paving the way for the theory of evolutionary change. Even today, the exist-ence of homologous body parts in different animals and plants is cited as standard evidence for this process7.

The easy case: homology of genes. Since the time of Owen and Darwin, the idea of homology has been extended to other bio-logical entities, such as genes, nucleotides, physiological processes and behavioural patterns7,8. What it means to speak of homologous genes is well understood9, and I recapitulate these ideas here because they

can guide us in a similar understanding of homology among morphological structures. Two genes are homologous as long as they are derived from the same gene in a com-mon ancestor, regardless of whether they have the same function and regardless of the extent of similarity in their nucleotide sequences. The gene retains its identity despite evolutionary change in its function and sequence, as long as all changes result from mutations at the same genomic locus. The basis of gene identity is the historical continuity of the locus undergoing evolu-tionary change. Of course, things become more complicated with gene duplications and loss or fusion of parts of genes (for example, exon shuffling) to form new genes, and when extensive sequence divergence erases the evidence of shared ancestry. The mode of evolution that preserves the histori-cal identity of a gene is the replacement of alleles at the same genomic locus.

Homology of morphological characters: what does ‘sameness’ mean? The homol-ogy of morphological characters is also a case of historical continuity in the face of descent with modification. In a population, a character exists in different states of size, shape or colour. Evolutionary change usu-ally proceeds by changing the frequency of these character states in the population, eventually leading to the replacement of the ancestral character state by a derived character state (FIG. 1). Sameness, then, by the definition of homology, does not refer to similarity of structure or function as such, but to historical continuity through inherit-ance with modification. In other words, the homology concept can be applied to anything that forms a lineage10–13. Of course, things become more complicated when new characters arise (novelties) or characters duplicate like genes (for example, teeth or fins).

At the formal phenomenological level, a morphological character corresponds to a genomic locus, and a character state to an allele14 (BOX 1). Hence, in a more technical sense, a character is a unit of evolutionary change at the phenotypic or morphological level in the same way that a gene is the unit of evolutionary change at the genetic level15.

O P I N I O N

The developmental genetics of homologyGünter P. Wagner

Abstract | Homology is an essential idea of biology, referring to the historical

continuity of characters, but it is also conceptually highly elusive. The main

difficulty is the apparently loose relationship between morphological

characters and their genetic basis. Here I propose that it is the historical

continuity of gene regulatory networks rather than the expression of individual

homologous genes that underlies the homology of morphological characters.

These networks, here referred to as ‘character identity networks’, enable the

execution of a character-specific developmental programme.

NATURE REVIEWS | GENETICS VOLUME 8 | JUNE 2007 | 473

PERSPECTIVES

© 2007 Nature Publishing Group

14

a

bc

Developmental genetics and homology?

The semi-conservative mode by which DNA replicates ensures that genes directly give rise to copies of themselves, and is therefore the mechanistic basis for their historical continuity. In the case of morphological characters, however, the situation is more complicated, because morphological characters and even cell types do not usually directly spawn copies of themselves between generations, but are recreated in each generation from a single cell, the zygote16. The recreation of a character is controlled by developmental genes, so it is tempting to speculate that the continuity of morphological characters can be explained by the continuity of genetic information. Since the beginning of experimental developmental biology in the early twentieth century, the emerging picture has been disappointing and confusing2,17–19. For instance, there is no question that body segments in all orders of insects are homologous and derived from a single common ancestor — there is not a single lineage of organisms more closely related to an insect group than to

other animals that is not segmented and, furthermore, the insects are nested in an even larger clade, the arthropods, which consists exclusively of segmented animals. Yet some genes that are essential for segmentation in Drosophila melanogaster20, for example, the pair-rule genes fushi tarazu (ftz) and even skipped (eve), do not have pair-rule function in the grasshopper Schistocerca americana, but are instead expressed in the developing CNS21,22. Clearly, the way in which segments are formed in development has changed since the MRCA of crown-group insects.

A solution to this conundrum can be found in the fact that developmental variation in homologous characters is not randomly distributed, but affects some aspects of development more than others. For example, in D. melanogaster, segmenta-tion proceeds through three stages that are controlled by particular genes: gap genes, which determine larger body regions, the pair-rule genes, which divide the embryo into stripes of alternating half segments, and the segment-polarity genes, which activate the actual morphogenetic process of

segment formation2. Surprisingly, the most extensive interspecific variation has been found in the higher levels of the segment-ation hierarchy, namely the gap genes and the pair-rule genes2,23. Examples are the pair-rule genes ftz and eve, mentioned above, and the gap gene bicoid (bcd), which exists only in the higher Diptera, not even in the dipteran mosquito Anopheles. By contrast, the segment-polarity gene network, which includes the interaction of engrailed (en) and wingless (wg), seems to be invariant, at least among insects2. This suggests that the genetic regulatory network (GRN) that con-trols the execution of the segment-specific morphogenetic processes is less variable than the upstream processes that activate it.

If the pattern that is suggested by the data on insect segmentation can be generalized, it seems that the most conservative parts of the developmental process are the GRNs that control the developmental programme that specifies the identity of the character; that is, the character identity network (ChIN). For example, individual cell types are determined by a characteristic set of regulatory genes over vast evolutionary distances24–27. Another example is the genetic network for the endomesoderm that starfish and sea urchins share28. By contrast, other aspects of development, from early pattern-ing to the execution of the developmental programme, are more variable2.

Here I review evidence that shows that these networks determine character identity rather than character state, that non-homologous morphological characters are determined by non-homologous ChINs, and that the genes participating in a ChIN are co-adapted for their task; that is, they are functionally non-equivalent to orthologues in species that do not have the character, and to paralogues that do not participate in the development of that character.

ChIN genes determine character identity

The idea that the genes that control character identity are distinct from the genes that determine the special shape and state of a character has been well documented in the case of Ultrabithorax (Ubx) function in insect wing development. Ancestrally, pterygote (or winged) insects have two pairs of proper wings associated with T2 and T3 (the second and third segments in the thorax), as seen, for instance, in honey bees, grasshoppers and most spectacularly in butterflies (FIG. 3a). Dipterans have only one pair of wings, which is localized on T2; that is, they are forewings. A homologue of the hind wing lies on T3 but does not take

Figure 1 | Homology of morphological characters. The hand of a human (part a) differs greatly in

terms of detailed structure and function from the wing of a bird (part b), but both are considered

homologous because they arose from a corresponding character in the tetrapod common

ancestor(part c) through descent with modification. This figure illustrates that morphological

characters form lineages of descent in the same way that genes form unique lineages of descent,

as long as they are not duplicated. Homology is the historical continuity of characters in multiple

lineages despite variations in their character state.

P E R S P E C T I V E S

474 | JUNE 2007 | VOLUME 8 www.nature.com/reviews/genetics


15

BatHumanLionHorseFrogWhale Bird

the shape of a wing blade. Instead, it is a small, club-like appendage called the haltere (FIG. 3b). Beetles also have only one pair of wing blades, but as they are associated with T3 they are hind wings. The forewings in beetles have been transformed into a pair of highly scleratized structures, called elytra, that function as protective covers (FIG. 3c). So, butterflies, flies and beetles all have two pairs of dorsal appendages that are homologous, because they are nested within a larger clade of winged insects, almost all of which have two pairs of wings on their T2 and T3: the forewings, which are flying organs in flies and butterflies but protective organs in beetles, and the hind wings, which form functional wing blades in butterflies and beetles but are sensory organs (halteres) in dipteran insects.

Hence, morphologically, we can distin-guish between two kinds of entities. On the one hand, there are two character identities: forewings and hind wings. On the other hand, there are various character states that insect wings can assume: the forewing can be a wing blade or an elytra, and the hind wing can be a wing blade or a haltere. Distinguishing between character identi-ties and character states also removes the confusion that is inherent in the character concept29 between parts, such as wings and legs (character identities), and attributes of parts, such as size, shape and colour (character states).

Experimental evidence shows that this distinction between character identities and character states is the result of different genetic underpinnings. ChIN genes, like Ubx and abdominal-A (abdA), determine character identity (forewing versus hind wing) across species, regardless of their character state30. In D. melanogaster, a loss-of-function mutation of Ubx leads to the development of a second set of wing blades (FIG. 4a). This does not mean that Ubx function is to suppress wing development; Warren and collaborators demonstrated that the four-winged butterfly also expresses Ubx in its T3 (REF. 31). So, in general Ubx determines hind wing identity, regardless of whether the hind wing is shaped as a wing blade or a haltere, as in dipterans32. This was confirmed in the flour beetle Tribolium castaneum33. As in the butterfly, the beetle expresses Ubx in T3, and suppression of Ubx function by RNAi leads to a second set of elytra on T3 (FIG 4b). Clearly, Ubx does not determine the shape of a wing but determines hind-wing identity “…regardless of form and function”6,30. The character state can change in evolution, but it remains under the control of Ubx.

The experimental evidence that is cited here pertains to only one gene, Ubx, but it is unlikely that Ubx is acting alone. One would expect that Ubx is part of a small network that also includes abdA and other transcription factor genes.

ChINs of non-homologous characters

The GRN underlying eye development is a celebrated example of evolutionary conservation, as it suggests that there is homology between vertebrate and insect eyes34. This assertion largely comes from the common role of paired-box gene 6 (Pax6) in the two systems, but detailed studies show that the rest of the network is strikingly different (see below). The most parsimoni-ous interpretation is that Pax6 is part of the ancestral cell-differentiation pathway for photoreceptors and was then separately incorporated into the ChINs for both types of image-forming eyes28.

Figure 2 | Homologous characters can have different shapes and func-tions. Forelimbs of seven tetrapod species exemplify the fact that

corresponding body parts have a similar design but can serve different

functions, from swimming to flying. Hence, functional necessity cannot

explain the similarity of the basic construction of homologous characters.

This remains one of the standard arguments in favour of evolution; that

is, that species derive from common ancestors by a process of descent

with modification.

Glossary

OrthologueTwo genes are orthologues if their lineages are connected

through a speciation event and without a duplication event.

Paralogue Two genes are paralogues if their lineages are connected

through a gene duplication event.

Pro-orthologueFor example, when one species has two copies of a gene,

say Ga and Gb, and another species has a single copy G,

and the speciation event that separated the species

lineages occurred earlier than the gene duplication event,

G is the pro-orthologue of Ga and Gb.

Semi-orthologueFor example, if G is the pro-orthologue of Ga and Gb, then

both Ga and Gb are the semi-orthologues of G.




16

a b c

In D. melanogaster, the gene eyeless (ey; with homology to the Pax gene family in vertebrates) is necessary for eye development and is sufficient to induce eyes35. In fact, ey is part of a small network that includes another transcrip-tion factor, sine oculis (so; with homology to the Six gene family in vertebrates) and two transcriptional cofactors, eyes absent (eya) and dachshund (dac), and is activated by a paralogue of ey called twin of eyeless (toy) (FIG. 5a). Surprisingly, it was found that homologous genes, most notably Pax6, which is the pro-orthologue36 of ey, are involved in eye development in all animals that have been examined34,37–39, and can induce ectopic eyes in D. melanogaster35 as well as in Xenopus laevis40.

All the genes in the D. melanogaster eye ChIN are members of larger gene families (TABLE 1). In vertebrates, genes from these four gene families (Pax, Six, Eya and Dach) are also involved in the development of several other organs and tissues, such as muscle and ear41,42.

However, the GRN of eye morphogen-esis in vertebrates is not the same as that in D. melanogaster37,41,43 (FIG. 5). For instance, Pax6/ey is upregulated by non-homologous genes in the two systems43: in insects, ey is regulated by toy, whereas in vertebrates, the transcription factor gene retinal home-obox (Rx; also known as Rax) is upstream of Pax6. Although a D. melanogaster homologue of Rx does exist, it is not involved in eye development. Similarly, in vertebrates, Eya1,2,3 (homologues of eya)

do not regulate Dach1, the homologue of D. melanogaster dac.

There are two possible reasons for the dissimilarity between the GRNs of eye develop-ment in D. melanogaster and vertebrates. It could be that the MRCA of flies and mam-mals had a GRN that involved members from all four gene families, but this network later changed. Alternatively, it could be that, just as the image-forming eye structures are inde-pendently derived, the regulatory interactions among these genes in eye development also evolved independently, and the GRN was not present in the MRCA of flies and mammals.

The gene lineages of the Six gene family suggest that the latter is the case. Although Six and Pax genes tend to be expressed together in various organs and cell types41,42, the different Six genes that are involved in D. melanogaster and vertebrate eye develop-ment are not orthologues. According to a phylogenetic analysis of Six genes from animals and unicellular flagellates, so of D. melanogaster and optix from vertebrates are ancient paralogues predating the origin of multicellular animals44,45. Hence, it is most parsimonious to assume that so and the optix genes were independently recruited into eye development in the vertebrate and insect lineages just as their morphological eye structures are independently derived. This shows that ChINs differentiate as they assume control of the development of a new character, so that different characters are controlled by non-orthologous sets of genes. Furthermore, the ciliary, vertebrate-type photoreceptor and the rhabdomeric, insect-type photoreceptor coexist in a polychaete worm, Platynereis dumerilii, showing that vertebrate and insect eyes derive from ‘paral-ogous’ cell populations25,46 and are therefore likely to be non-homologous.

Box 1 | Equivalent terms for genes and morphological characters

Morphological characters are equivalent to a genetic locus that undergoes evolutionary modification. Different instantiations of a gene are called alleles and different instantiations of a character are called character states. Genes that were inherited from a common ancestor without duplication are orthologues, whereas body parts in two species that were inherited from a common ancestor are homologues. Different instances of a gene caused by gene duplication are called paralogues, and repeated instances of a morphological character in the same organisms are called serial homologues. New genes can arise through gene duplication and divergence or from the fusion of parts of genes. For morphological characters, a new character that creates a new lineage of descent with modifications is called an evolutionary novelty67. These can arise in various ways, for instance, by duplication and differentiation, much like for new genes, or by de novo origination.

Genetic terms and their equivalents for morphological characters:

Genetic term Morphological equivalent

Locus Character

Allele Character state

Orthology Special homology

Paralogy Serial homology

Origin of new genes Evolutionary novelty

Figure 3 | Characters and character states. The insects shown all have two

pairs of dorsal appendages, forewings and hind wings. They can both be wing

blades that function in producing lift, as in the case of butterflies (panel a),

or only one might form a proper wing blade, as in Diptera (panel b) or beetles

(panel c). In the Diptera (flies, mosquitos and so on), the proper wing blade

is the forewing, whereas the hind wing is transformed into a club-shaped

appendage, termed the haltere (indicated by a black arrow). In beetles, the

proper wing blade is the hind wing, whereas the forewing is transformed

into a protective cover called the elytra (indicated by a white arrow). Images

in panels b,c courtesy of J. Tanis and A. Andrasi, respectively.




17

ba

Unfortunately, little is known about the genetics of cephalopod eyes47,48, the third type of independently derived complex light-sensory organs. If the ChIN concept is correct, one would predict that the GRN of cephalopod eye development includes Pax6 but is otherwise independently derived, and different from both the insect and the vertebrate eye GRN.

Co-adapted transcription factors

What mechanism keeps the GRNs that determine character identity more con-served than other aspects of development? A possible answer to this question derives from the fact that transcription factor proteins do not remain equivalent, but undergo functionally important changes during evolution, particularly with respect to protein–protein interactions. This is a well documented but often overlooked fact of the evolution of gene regulation. It is now well established that the functional specificity of transcription factors is not solely related to DNA binding specificity, which is low, but also involves interactions with other transcription factor proteins49, and differences in the functional activity of a transcription factor depend on parts of the protein that are not engaged in DNA binding. These functionally important differences in transcription factor proteins might lead to co-adaptation among the transcription factor genes that partake in the same GRNs. Although several lines of evidence have created the misleading impression that transcription factors are functionally invariant50, there is a growing amount of compelling evidence51 that fly transcription factor genes like Ubx52–55, tinman (tin)56, and ftz57, as well as vertebrate Hox genes58,59, do not remain functionally

equivalent in evolution, and that these differences are not due to changes in DNA binding, but rather to changes in protein–protein interactions among transcription factors.

A particularly well investigated case of functional non-equivalence is that of Ubx in D. melanogaster compared with its homologues in the velvet worm Acanthokara kaputensis52,53 and the brine shrimp Artemia franciscana55. Grenier and Carroll52 compared the in vivo activ-ity of A. kaputensis Ubx (Onychophora: O-Ubx) with that of D. melanogaster Ubx1a (D-Ubx1a) in a misexpression approach. O-Ubx and D-UBX1A have similar homeodomains (97% identical amino-acid sequence), but their overall similarity is low. O-UBX is only 214 amino acids long, compared with 380 amino acids in D-UBX1A. Like D-UBX1A, O-Ubx expressed in flies can transform antenna towards a leg phenotype and forewings into halteres. More specifically, O-Ubx can repress Surf wings (Srf) in the wing disc and drive the expression of decapentaplegic (dpp) in the visceral mesoderm, both of which are specific targets of D-UBX1A. But other typical effects of D-Ubx cannot be reproduced by O-Ubx, such as repression of Distal-less (Dll) in the leg rudiments. These D-UBX1A-specific activ-ities are caused by differences outside the homeodomain. The authors suggest that D-UBX1A can engage in protein–protein interactions specific for these characters that O-Ubx cannot, although the specific interaction partners have not been identi-fied. In a later paper, the sequence that is responsible for the D-UBX1A-specific activity was identified as a repressor domain on the carboxy (C)-terminal side

of the protein, with a QAQAQK(A)n motif (QA motif)53. Similar results were obtained in a comparison of D-UBX1A and A. franciscana Ubx (Af-Ubx) by McGinnis and collaborators55. In another study in D. melanogaster, in which the wild-type Ubx allele was replaced with an allele in which the QA motif was deleted (UbxΔQA)54, it was shown that QA has an additive effect together with other peptide motifs60 and ABDA protein activity. This series of stud-ies shows that functional non-equivalence of UBX evolved not through changes in the DNA binding activity of the transcription factor, but most likely through changes in protein–protein or protein–RNA interactions.

Two important studies show that there are also specific functional differences between the homeodomains of Hox genes within the same organism (the mouse in this case). Zhao and Potter58,59 replaced the HoxA11 homeobox with the HoxA13 homeobox (creating HoxA11(a13Hd)) and found that this allele can rescue the HoxA11–/– phenotype in the development of the body axis, kidney and male reproduc-tive tract. Interestingly, the HoxA11(a13Hd) allele antagonizes normal HoxA11 function in limb development and causes homeotic transformation of the uterine epithelium into cervical epithelium.

This study was followed up with homeo-domain swapping experiments between HoxA11, HoxA10 and HoxA4 (REF. 59). HoxA10 is even more similar to HoxA11 than HoxA13, and the HoxA11(a10Hd) allele was functionally quasi-equivalent even in the female reproductive tract, in addition to the body axis, kidney and male repro-ductive tract. By contrast, replacing the HoxA11 homeodomain with that of HoxA4 led to near-recessive null function in all but the axial development.

The pattern emerging from these stud-ies is that functional non-equivalence is more likely for more derived functions, such as the mammalian female reproduc-tive tract, than for more ancient characters, such as body axis. That is, the older the paralogues and the more derived the characters, the less likely it is that the transcription factors are still functionally equivalent. These results are interesting in light of the recent findings of adaptive evolution of the homeodomain following Hox gene duplication61, and the fact that paralogue-specific amino-acid residues in the homeodomain are likely to function in protein–protein interactions rather than DNA binding62.

Figure 4 | Ubx determines character identity, not character states. Loss of Ultrabithorax (Ubx) func-

tion leads to development of a forewing in the position of the hind wing (in body segment T3), regardless

of the character state. In Drosophila melanogaster (panel a), a loss-of-function mutation of Ubx leads

to two forewings, whereas in Tribolium castaneum (panel b) the knockdown of Ubx leads to a second set

of elytra33. So, Ubx is necessary to determine hind wing identity but does not determine character

state30. Part b reproduced with permission from Nature REF. 33 © (2005) Macmillan Publishers Ltd.




18

ET Rx Pax6

dac eya

ey

Tll

Eya1,2,3

Six 3

( Dach1 )

so(Rx)

Optx2 Lhx2toy

Drosophila melanogaster Xenopus laevis

The few examples reviewed above show that a transcription factor that becomes involved in a novel character has a tendency to become modified to carry out that novel function. This probably happens because of co-adaptation among transcription factors in the same GRN63, but because there is no published example in which the evolution of both protein–protein interaction partners has been documented, we cannot be sure that this is the case. It could be that we see the result of adaptation of one transcription factor to another without both of them evolving. In either case, the transcription factors in a ChIN for a specific character are specialized for this task and remain associated with each other and with their character. They thus share historical continuity with the character they control.

ChINs and other GRNs. In a recent syn-thesis of the current knowledge of GRN evolution, Eric Davidson also noted that interspecific variation in GRNs is non-random64. He coined the term ‘kernel’ for the most conserved set of GRNs28, and it is legitimate to ask whether the kernel concept is equivalent to the ChIN concept proposed here. Davidson defines a kernel as “…[a] conserved subcircuit consisting of genes which interact with one another

and which are dedicated to a specific developmental function.”28 Although instances of kernels certainly overlap with examples of ChINs, the conceptual goal differs. From Davidson’s examples, it is clear that kernels are only the most con-served GRNs regardless of their biological function. For instance, Hox GRNs do not constitute kernels because they are not old or invariant enough. By contrast, ChINs are defined as GRNs that subscribe the execu-tion of a character-specific developmental programme, regardless of how old they are.

Certainly, Hox-gene-based GRNs are likely candidates for ChINs, but are not kernels according to Davidson’s use of the term28. Any new character is expected to have its ChIN, regardless of how old it is, whereas kernels are supposed to be associated with and be invariant in the oldest clades in the animal kingdom. Kernels and ChINs are both classes of GRN, but they serve differ-ent conceptual ends and are only partially overlapping.

Future directions

Consistent with modern views of homo-logy10,11,13,65,66, character identity is not tied to particular manifest features, like struc-ture, composition and shape. Instead, homologues have a single historical origin, form a lineage of descent with modification, and can go extinct. From a developmental point of view, character identity and thus homology requires the ability to express an evolutionarily variable developmental programme that is different from those in other parts of the body. In this paper, I propose that this capability is underwritten by GRNs of co-adapted transcription factor genes. This concept implies a number of predictions that can be tested with the aid of standard gene knockdown and rescue experiments and functional genomic tools. For instance, the knockdown of a transcrip-tion factor gene that is part of a ChIN should affect the majority of genes that are regulated during the development of that character. In addition, because the members of a ChIN are assumed to be jointly neces-sary for the development of the character, the knockdown of different members is predicted to affect largely overlapping sets of downstream genes. The assertion that members of a ChIN are co-adapted can be tested by rescue experiments.

Figure 5 | Gene regulatory networks for eye development. The non-homologous eyes of insects

and vertebrates are controlled by non-homologous genetic regulatory networks37,39. In Drosophila

melanogaster, the five genes eyeless (ey), twin of eyeless (toy), eyes absent (eya), dachshund (dac) and

sine oculis (so) are jointly necessary for eye development. Four of these are also sufficient to induce

eye development and can mutually induce each other’s expression: toy, ey, eya and dac. The retinal

homeobox (Rx) transcription factor gene, which is essential in vertebrate eye development, is also

present in the D. melanogaster genome but is not involved in eye development. dac is essential in

D. melanogaster eye development, but its semi-orthologue, Dach1, is not essential in vertebrate (for

example, Xenopus laevis) eye development. For vertebrates, similar genes are involved in eye devel-

opment but the functional relationships among them are different and involve non-orthologous

genes. Orthologous genes are in red boxes and paralogous genes are in blue boxes. Instead of the

vertebrate semi-orthologue of sine oculis, the genes sine oculis homeobox homologue 3 (Six3) and

Optx2 (also known as Six6) are involved in eye development. The gene duplication event that

created the sine oculis and the Six3/Optx2 paralogues occurred long before the most recent com-

mon ancestor of insects and vertebrates44,45. This shows that the genetic regulatory networks of

insect and vertebrate eye development are most likely not homologous but derived independently.

ET, transcription repression factor Tbx3; Lhx2, LIM homeobox 2; Pax6, paired-box 6; Tll, nuclear

receptor subfamily 2, group E, member 1, nr2e1-A, Xtll.

Table 1 | Gene families in eye development: the Pax, Ey, Six, Eya and Dach families

Cnidarians Drosophila melanogaster Vertebrates

PaxA Pox neuro

PaxB sparkling (also known as shaven) Pax2, Pax5, Pax8

PaxC eyeless, twin of eyeless Pax4, Pax6

PaxD Pox meso, gooseberry, gooseberry neuro, paired Pax3, Pax7

Six1/2 sine oculis Six1, Six2

Six3/6 Optix Six3, Optx2

Six4/5 myotonix Six4, Six5

? eyes absent Eya1, Eya2, Eya3, Eya4

? dachshund Dach1, Dach2

See REFS 41,45,68,69. Note that the sine oculis and Optix genes duplicated before the most recent common

ancestor of cnidarians and the bilaterian animals. ? indicates that family members have not been identified so far.

Dach, dachschund family; Ey, eyeless family; Eya, eyes absent family; Pax, paired-box family; Six, sine oculis family.




19

Co-adaptation predicts that the knockdown effects can be reversed with a gene from the clade that shares this character, but not by genes from species that diverged before the origin of that character. High-throughput methods make these predictions testable at a scale that would allow assessment of the utility of the ChIN concept in explaining homology of morphological characters.

Günter P. Wagner is at the Department of Ecology and Evolutionary Biology,

Yale University, POB 208106, New Haven, Connecticut 06520-8106, USA.

e-mail: [email protected]

doi:10.1038/nrg2099

Published online 8 May 2007

1. Hall, B. K. Evolutionary Developmental Biology (Chapman & Hall, London, 1998).

2. Wilkins, A. S. The Evolution of Developmental Pathways (Sinauer, Sunderland, 2002).

3. Butler, A. B. & Saidel, W. M. Defining sameness: historical, biological, and generative homology. BioEssays 22, 846–853 (2000).

4. VanValen, L. Homology and causes. J. Morphol. 173, 305–312 (1982).

5. Lewontin, R. C. Adaptation. Am. Sci. 239, 156–169 (1978).

6. Owen, R. On the Archetype and Homologies of the Vertebrate Skeleton (Voorst, London, 1848).

7. Futuyma, D. J. Evolutionary Biology (Sinauer, Sunderland, 2005).

8. Brigant, I. Homology in comparative, molecular, and evolutionary developmental biology: the radiation of a concept. J. Exp. Zool. B Mol. Dev. Evol. 299, 9–17 (2003).

9. Graur, D. & Li, W.-H. Fundamentals of Molecular Evolution (Sinauer, Sunderland, 2000).

10. Geeta, R. Structure trees and species trees: what they say about morphological development and evolution. Evol. Dev. 5, 609–621 (2003).

11. Oakley, T. H. The eye as a replicating and diverging, modular developmental unit. Trends Ecol. Evol. 18, 623–627 (2003).

12. Roth, V. L. Homology and hierarchies: problems solved and unresolved. J. Evol. Biol. 4, 167–194 (1991).

13. Ghiselin, M. T. Homology as a relation of correspondence between parts of individuals. Theor. Biosci. 124, 91–103 (2005).

14. Wagner, G. P. The biological role of homologues: a building block hypothesis. N. Jb. Geol. Paläont. Abh. 195, 279–288 (1995).

15. Wagner, G. P. in The Character Concept in Evolutionary Biology (ed. Wagner, G. P.) 1–10 (Academic, San Diego, 2001).

16. Wagner, G. P. The biological homology concept. Ann. Rev. Ecol. Syst. 20, 51–69 (1989).

17. Hall, B. K. in Evolutionary Biology Vol. 28 (eds Hecht, M. K., MacIntyre, R. J. & Clegg, M. T.) 1–30 (Plenum, New York, 1994).

18. Wray, G. A. & Abouheif, E. When is homology not homology? Curr. Opin. Genet. Dev. 8, 675–680 (1998).

19. Hall, B. K. Descent with modification: the unity underlying homology and homoplasy as seen through an analysis of development and evolution. Biol. Rev. Camb. Philos. Soc. 78, 409–433 (2003).

20. Pankratz, M. J. & Jackle, H. in The Development of Drosophila melanogaster (eds Bate, M. & Martinez-Arias, A.) 467–516 (Cold Spring Harbor Laboratory Press, Plainview, 1993).

21. Patel, N. H., Ball, E. F. & Goodman, C. S. Changing role of even-skipped during the evolution of insect pattern formation. Nature 357, 339–342 (1992).

22. Dawes, R., Dawson, I., Falciani, F., Tear, G. & Akam, M. Dax, a locust Hox gene related to fushi-tarazu but showing no pair-rule expression. Development 120, 1561–1572 (1994).

23. Peel, A. D., Chipman, A. D. & Akam, M. Arthropod segmentation: beyond the Drosophila paradigm. Nature Rev. Genet. 6, 905–916 (2005).

24. Arendt, D. Evolution of eyes and photoreceptor cell types. Int. J. Dev. Biol. 47, 563–571 (2003).

25. Arendt, D. Genes and homology in nervous system evolution: comparing gene functions, expression patterns, and cell type molecular fingerprints. Theor. Biosci. 124, 185–197 (2005).

26. Allan, D. W., Park, D., Pierre, S. E. S., Taghert, P. H. & Thor, S. Regulators acting in combinatorial codes also act independently in single differentiating neurons. Neuron 45, 689–700 (2005).

27. Shirasaki, R. & Pfaff, S. L. Transcriptional codes and the control of neuronal identity. Ann. Rev. Neurosci. 25, 251–281 (2002).

28. Davidson, E. H. The Regulatory Genome: Gene Regulatory Networks in Development and Evolution 144–159, 28 (Academic, Amsterdam, 2006).

29. Ghiselin, M. ‘Definition, ’ ‘character’ and other equivocal terms. Syst. Zool. 33, 104–110 (1984).

30. Deutsch, J. Hox and wings. BioEssays 27, 673–675 (2005).

31. Warren, R. W., Nagy, L., Senegue, J., Gates, J. & Carroll, S. Evolution of homeotic gene regulation and function in flies and butterflies. Nature 372, 458–461 (1994).

32. Wagner, G. P., Chiu, C.-H. & Laubichler, M. Developmental evolution as a mechanistic science: the inference from developmental mechanisms to evolutionary processes. Am. Zool. 40, 819–831 (2000).

33. Tomoyasu, Y., Wheeler, S. R. & Denell, R. E. Ultrabithorax is required for membranous wing identity in the beetle Tribolium castaneum. Nature 375, 58–61 (2005).

34. Gehring, W. J. & Ikeo, K. Pax6 mastering eye morphogenesis and eye evolution. Trends Genet. 15, 371–377 (1999).

35. Halder, G., Callerts, P. & Gehring, W. J. Induction of ectopic eyes by targeted expression of the eyeless gene in Drosophila. Science 267, 1788–1792 (1995).

36. Holland, P. W. H. in Homology Vol. 222 (eds Bock, G. R. & Cardew, G.) 226–236 (Wiley & Sons, Chichester, 1999).

37. Donner, A. L. & Maas, R. L. Conservation and non-conservation of genetic pathways in eye specification. Int. J. Dev. Biol. 48, 743–753 (2004).

38. Kozmik, Z. Pax genes in eye development and evolution. Curr. Opin. Genet. Dev. 15, 430–438 (2005).

39. Zuber, M. E., Gestri, G., Viczian, A. S., Barsacchi, G. & Harris, W. A. Specification of the vertebrate eye by a network of eye field transcription factors. Development 130, 5155–5167 (2003).

40. Chow, R. L., Altmann, C. R., Lang, R. A. & Hemmati-Brivanlou, A. Pax6 induced ectopic eyes in a vertebrate. Development 126, 4213–4222 (1999).

41. Kardon, G., Heanue, T. A. & Tabin, C. J. in Modularity in Development and Evolution (eds Schlosser, G. & Wagner, G. P.) 59–80 (Univ. Chicago Press, Chicago and London, 2004).

42. Schlosser, G. Evolutionary origins of vertebrate placods: insights from developmental studies and from comparisons with other deuterostomes. J. Exp. Zool. B Mol. Dev. Evol. 304, 347–399 (2005).

43. Bailey, T. J. et al. Regulation of vertebrate eye development by Rx genes. Int. J. Dev. Biol. 48, 761–770 (2004).

44. Bebenek, I. G., Gates, R. D., Morris, J., Hartenstein, V. & Jacobs, D. K. sine oculis in basal Metazoa. Dev. Genes Evol. 214, 342–351 (2004).

45. Stierwald, M., Yanze, N., Bamert, R. P., Kammermeier, L. & Schmid, V. The sine oculis/Six class family of homeobox genes in jellyfish with and without eyes: development and regeneration. Dev. Biol. 274, 70–81 (2004).

46. Arendt, D., Tessmar-Raible, K., Snyman, K., Dorresteijn, A. W. & Wittbrodt, J. Cilliary photoreceptors with a vertebrate-type opsin in an invertebrate brain. Science 306, 869–871 (2004).

47. Nilsson, D.-E. Eye evolution: a question of genetic promiscuity. Curr. Opin. Neurobiol. 14, 407–414 (2004).

48. Ogura, A., Ikeo, K. & Gojobori, T. Comparative analysis of gene expression for convergent evolution of camera eye between octopus and human. Genome Res. 14, 1555–1561 (2004).

49. Chan, S. K., Jaffe, L., Capovilla, M., Botas, J. & Mann, R. S. The DNA binding specificity of Ultrabithorax is modulated by cooperative interactions with extradenticle, another homeoprotein. Cell 78, 603–615 (1994).

50. McGinnis, N., Kuziora, M. A. & McGinnis, W. Human Hox-4.2 and Drosophila deformed encode similar regulatory specificities in Drosophila embryos and larvae. Cell 63, 969–976 (1990).

51. Hsia, C. C. & McGinnis, W. Evolution of transcription factor function. Curr. Opin. Genet. Dev. 13, 199–206 (2003).

52. Grenier, J. K. & Carroll, S. B. Functional evolution of the Ultrabithorax protein. Proc. Natl Acad. Sci. USA 97, 704–709 (2000).

53. Galant, R. & Carroll, S. B. Evolution of a transcriptional repression domain in an insect Hox protein. Nature 415, 910–913 (2002).

54. Hittinger, C. T., Stern, D. L. & Carroll, S. B. Pleiotropic functions of a conserved insect-specific Hox peptide motif. Development 132, 5261–5270 (2005).

55. Ronshaugen, M., McGinnis, N. & McGinnis, W. Hox protein mutation and macroevolution of the insect body plan. Nature 415, 914–917 (2002).

56. Ranganayakulu, G., Elliott, D. A., Harvey, R. P. & Olson, E. N. Divergent roles for NK-2 class homeobox genes in cardionenesis in flies and mice. Development 125, 3037–3048 (1998).

57. Lohr, U., Yussa, M. & Pick, L. Drosophila fishi tarazu: a gene on the border of homeotic function. Curr. Biol. 11, 1403–1412 (2001).

58. Zhao, Y. & Potter, S. S. Functional specificity of the Hoxa 13 homeobox. Development 128, 3197–3207 (2001).

59. Zhao, Y. & Potter, S. S. Functional comparison of the Hoxa 4, Hoxa 10, and Hoxa 11 Homeoboxes. Dev. Biol. 244, 21–36 (2002).

60. Tour, E., Hittinger, C. T. & McGinnis, W. Evolutionarily conserved domains required for activation and repression functions of the Drosophila Hox protein Ultrabithorax. Development 132, 5271–5281 (2005).

61. Lynch, V. J., Roth, J. J. & Wagner, G. P. Adaptive evolution of Hox-gene homoedomains after cluster duuplications. BMC Evol. Biol. 6, 86 (2006).

62. Sharkey, M., Graba, Y. & Scott, M. P. Hox genes in evolution: protein surfaces and paralog groups. Trends Genet. 13, 145–151 (1997).

63. Wagner, G. P. & Lynch, V. J. Molecular evolution of evolutionary novelties: the vagina and uterus of therian mammals. J. Exp. Zool. B Mol. Dev. Evol. 304, 580–592 (2005).

64. Davidson, E. H. & Erwin, D. H. Gene regulatory networks and the evolution of animal body plans. Science 311, 796–800 (2006).

65. Wagner, G. P. & Stadler, P. F. Quasi-independence, homology and the unity of type: a topological theory of characters. J. Theor. Biol. 220, 505–527 (2003).

66. Wiley, E. O. Homology, identity and transformation in Mesozoic fishes (eds Arrtia, G. & Schultze, H.-P.) (in the press).

67. Müller, G. B. & Wagner, G. P. Novelty in evolution: restructuring the concept. Ann. Rev. Ecol. Syst. 22, 229–256 (1991).

68. Miller, D. J. et al. Pax gene diversity in the basal cnidarian Acropora millepora (Cnidaria, Anthozoa): implications for the evolution of the Pax gene family. Proc. Natl Acad. Sci. USA 97, 4475–4480 (2000).

69. Hadrys, T., DeSalle, R., Sagasser, S., Fischer, N. & Schierwater, B. The Trichoplax PaxB gene: a putative proto-PaxA/B/C gene predating the origin of nerve and sensory cells. Mol. Biol. Evol. 22, 1569–1578 (2005).

AcknowledgementsThe author gratefully acknowledges the financial support for the research that led to the ideas expressed in this article: NSF IBN#0321470, OIB#0445971 and the Research Prize from the Alexander von Humboldt Foundation. I also want to thank the reviewers of this paper as well as M. Laubichler, M. Ghiselin, E. Wiley, M. Donaghue, A. Monteiro and the members of my laboratory for discussion and inspiration.

Competing interests statementThe author declares no competing financial interests.

DATABASESThe following terms in this article are linked online to:Entrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query.

fcgi?db=gene

abdA | bcd | dac | Dach1 | dpp | en | eve | ey | eya | ftz | HoxA4 |

HoxA10 | HoxA11 | HoxA13 | Pax6 | Rx | so | Srf | tin | toy |

Ubx | wg

FURTHER INFORMATIONWagner laboratory homepage: http://pantheon.yale.edu/%7Egpwagner/index.html

Access to this links box is available online.




20

JOURNAL OF EXPERIMENTAL ZOOLOGY (MOL DEV EVOL) 285:19�26 (1999)

© 1999 WILEY-LISS, INC.

Why Do Almost All Mammals Have Seven CervicalVertebrae? Developmental Constraints, Hox Genes,and Cancer

FRIETSON GALIS*Institute for Evolutionary and Ecological Sciences, University of Leiden,2300RA Leiden, The Netherlands

ABSTRACT Mammals have seven cervical vertebrae, a number that remains remarkably con-stant. I propose that the lack of variation is caused by developmental constraints: to wit, changesin Hox gene expression, which lead to changes in the number of cervical vertebrae, are associatedwith neural problems and with an increased susceptibility to early childhood cancer and still-births. In vertebrates, Hox genes are involved in the development of the skeletal axis and thenervous system, among other things. In humans and mice, Hox genes have been shown also to beinvolved in the normal and abnormal (cancer) proliferation of cell lines; several types of cancer inyoung children are associated with abnormalities in Hox gene expression and congenital anoma-lies. In these embryonal cancers the incidence of a cervical rib (a rib on the seventh cervicalvertebra, a homeotic transformation of a cervical vertebra towards a thoracic-type vertebra) ap-pears to be increased. The minimal estimate of the selection coefficient acting against these muta-tions is about 12%.

In birds and reptiles variations in the number of cervical vertebrae have frequently occurredand there is often intraspecific variability. A review of the veterinary literature shows that cancerrates appear lower in birds and reptiles than in mammals. The low susceptibility to cancer inthese classes probably prevents the deleterious pleiotropic effect of neonatal cancer when changesin cervical vertebral number occur.

In mammals there is, thus, a coupling between the development of the axial skeleton and otherfunctions (including the proliferations of cell lines). The coupling of functions is either a conservedtrait that is also present in reptiles and birds, but without apparent deleterious effects, or thecoupling is new to mammals due to a change in the functioning of Hox genes. The cost of thecoupling of functions in mammals appears to be an increased risk for neural problems, neonatalcancer, stillbirths, and a constraint on the variability of cervical vertebral number. J. Exp. Zool.(Mol. Dev. Evol.) 285:19�26, 1999. © 1999 Wiley-Liss, Inc.

The exceedingly low level of interspecific varia-tion in the number of cervical vertebrae of mam-mals has puzzled biologists for more than 150 years.In birds, reptiles, and amphibians the number ofcervical vertebrae varies considerably, and in mam-mals the number of vertebrae in other vertebralregions is variable as well (Lebouck, 1898; Schulz,�61). Swans� long necks have a striking 22�25 cer-vical vertebrae, while ducks have 16 (Woolfenden,�61), and swifts 13 (Starck, �79). Giraffes and drom-edaries, however, have only seven vertebrae (Fig.1), as do the Dugong (Fig. 2) and whales with theirshort necks (Starck, �79). There are only three gen-era with an exceptional number of cervical verte-brae, manatees (Trichechus) and sloths (Bradypusand Choloepus). Thus, there seems to be an evolu-tionary constraint towards the development of vari-ability in the cervical region in mammals.

Intraspecific variations in the number of cervi-cal vertebrae in mammals are extremely rare,whereas intraspecific variations in the number ofmore caudal vertebrae are common, especially ofthe lumbar, sacral, and coccygeal regions (e.g.,Lebouck, 1898; Schulz, �61). However, one varia-tion of cervical vertebrae does occur infrequently:cervical ribs. A cervical rib is on the seventh cer-vical vertebra, is a partially or wholly homeotictransformation of the seventh cervical vertebrainto the first thoracic vertebra and, thus, reducesthe number of cervical vertebrae (and increases

*Correspondence to: Frietson Gailis, Institute for Evolutionary andEcological Sciences, University of Leiden, PO Box 9516, 2300RALeiden, The Netherlands. E-mail: [email protected]

Received 28 October 1998; Accepted 16 December 1998.

21

20 F. GALIS

the number of thoracic vertebrae). Further studyof this naturally occurring variation seems rel-evant with respect to the evolutionary constrainton cervical vertebral number, and more specifi-cally, to the study of the selective factors againstthis variation.

PATHOLOGIES ASSOCIATED WITHCERVICAL RIBS

Consideration of the pathologies in humans thatare associated with cervical ribs reveals two types,thoracic outlet syndrome (TOS) and early childhoodcancer. TOS involves pressure on the nerves of thebrachial plexus and on the subclavian artery, some-times leading to severe degenerative symptoms inthe arm (Fig. 3; Makhoul and Machleder, �92; Roos,�96). Often surgery is performed to relieve symp-

toms. Research on this syndrome has revealed thatcervical ribs are invariably associated with changesin the brachial plexus (a different contribution ofmotor and sensory nerves to the brachial plexus)and other structural abnormalities (Makhoul andMachleder, �92; Roos, �96). The correlation ofsymptoms must be due to mutual influences ofthe notochord, neural tube, neural crest, andsomites at the time of somite formation (Gosslerand Hrabe de Angelis, �98).

Early childhood cancer is a considerably moreserious pathology. Childhood cancers tend to re-sult from aberrant developmental processes andare generally embryonal in origin. They are asso-ciated with a high incidence of congenital abnor-malities. This association is assumed to be causedby a common underlying genetic abnormality

Fig. 1. Skeleton of a dromedary (Camelus dromedarius). Note the large cervical verte-brae. From Owen (1866).

Fig. 2. Skeleton of a dugong (Dugong dugon). Note the small cervical vertebrae. FromOwen (1866).

22

CERVICAL VERTEBRAE: HOX GENES AND CANCER 21

(Schumacher et al., �92; Anbazhagan and Raman,�97). A high incidence of vertebral anomalies, es-pecially cervical ribs, was found in a study spe-cifically devoted to finding vertebral anomalies,of 750 children with embryonal cancers (Schu-macher et al., �92). An incidence of around 25% cer-vical ribs was found for the following embryonalcancers: neuroblastoma, brain tumour (astrocytomaand medulloblastoma), acute lymphoblastic andmyeloid leukemia, soft tissue sarcoma, Wilms� tu-mour and Ewing sarcoma (Table 1). This findingconfirms the observations by Adson and Coffey(�27), who found that cervical ribs are sometimesdiscovered in children because of the presence ofa tumor in the neck. In addition, a high correla-tion between malformations of ribs (without fur-ther specification) and cancer of all types wasfound in a large study on childhood cancers (Narodet al., �97).

In agreement with the hypothesis of a commongenetic abnormality underlying both early child-hood cancer and cervical ribs is the observationthat the relation between congenital anomaliesand cancer is stronger in infants than in olderchildren (Brodeur, �95; Breslow et al., �96; Gurneyet al., �96). Many infants with cancer demonstrate

unique epidemiologic, clinical, and genetic char-acteristics compared with cancers that occur inolder children. Some of the early onset cases arefamilial cases, which are rare and generally char-acterized not only by an early onset, but also by aworse prognosis (Brodeur, �95; Breslow et al., �96;Gurney et al., �96). This phenomenon is explainedby Knudson�s (�84) model for embryonal childhoodcancers in which two (or only a few) mutationalevents occur before the onset of cancer. In famil-ial cases one of these mutations has occurred inthe germ line and is transmissible to the offspring.The germ-line mutation has been identified forfamilial retinoblastoma (reviewed in Brodeur, �95).The timing of mutational events should influencethe incidence and type of congenital anomaly andthese differences in timing can, thus, explain thatnot all cases of childhood cancer have congenitaldefects and that the anomalies are variable.

THE ROLE OF HOX GENES INPATTERNING OF THE SKELETAL AXIS

AND IN CELL PROLIFERATION

Hox genes play an important role in the pat-terning of the axial skeleton in all vertebrateclasses (Krumlauf, �94). Hox gene mutants display

Fig. 3. Illustration of the thoracic outlet in a person witha cervical rib showing how arteries and axons are compressedwhen the m. anterior scalaenus contracts. The cervical rib isincomplete and fused with the first thoracic rib. (Reproduced

by permission from Adson AW. 1947. Surgical treatment forsymptoms produced by cervical ribs and the scalenus anticusmuscle. Surg Gynecol Obstet 85:687�700.)

TABLE 1. The incidence of a cervical rib in children with embryonal cancers1

Type of childhood cancer Number of cases Incidence of a cervical rib

Neuroblastoma 88 33%Brain tumour 234 27.4%Leukemia 227 26.8%Soft tissue sarcoma 98 24.5%Wilms tumour 68 23.5%Ewing sarcoma 35 17.1%1Data from Schumacher et al. (�92).

23

22 F. GALIS

abnormalities of the vertebral column. Particu-larly common phenotypic abnormalities in micemutants are cervical ribs. At least four knock-outmutants of hox genes in mice have an increasedincidence of cervical ribs (Hoxa-4, Hoxd-4, Hoxa-5 and Hoxa-6) (reviewed in Horan et al., 95). Inaddition, transgenic mice overexpressing Hoxb-7or Hoxb-8 and mice mutants lacking the polycomb-group genes bmi-1 and mel-18 (involved in theregulation of Hox genes) display cervical ribs(McLain et al., �92; Charité et al., �94; Akasaki etal., �96; van der Lugt et al., �96). Thus, the forma-tion of cervical ribs is a process that seems to beparticularly susceptible to perturbations in Hoxgene expression (Horan et al., �95). Most of thesemutant mice have a severely impaired viability.

At the same time Hox genes have been shownto be involved in the proliferation of cell lines inmice and humans (e.g., Corte et al., �93; Lawrenceet al., �96; Anbazhagan et al., �97). In a study inwhich cells of the myeloid, macrophage, erythroid,and B- and T-lymphoid lingeages were investi-gated for expression of homeotic genes, up to 20different Hox genes were found to be activated(Kongsuwan et al., �88). Some of the genes wereubiquitously expressed, while others were re-stricted to particular cell lineages or lines (see alsoLawrence et al., �96). When the cells were inducedto differentiate, the pattern of Hox gene expres-sion changed. Changes in Hox gene expressionhave been demonstrated for several types of can-cer, including some childhood cancers that werefound to have a high incidence of vertebral anoma-lies: neuroblastoma, Wilms� tumour, and leukemia(Corte et al., �93; Lawrence et al., �96; Manohar etal., �96; Anbazhagan et al., �97). The coupling be-tween these two functions of Hox genes is clearlydemonstrated in mice with mutations of thePolycomb- and trithorax-group genes (Pc-G andtrx-G genes). The evolutionary-conserved Pc-G andtrx-G genes are involved in the maintenance ofexpression of homeobox genes including Hom andHox genes. Mice lacking or overexpressing Pc-Gand trx-G genes have altered expression areas ofHox genes and display both vertebral anomalies(including cervical ribs and other changes in thenumber of cervical vertebrae) and leukemia or re-lated cancers (Corte et al., �93; van der Lugt et al.,�94; Yu et al., �95; Akasaki et al., �96; Schumacheret al., �96; Coré et al., �97). One of these genes is thetrx-G gene Mll, the most commonly involved genein infant leukemias (Pui et al., �95). Mice heterozy-gous for the knock-out allele of the caudal geneCdx2, which is involved in the regulation of Hox

genes (Epstein et al., �97), also display both ver-tebral abnormalities and a predisposition for in-testinal cancer (He et al., �97). Furthermore,rostral overexpression of Hoxb-8 leads to cervicalribs in mice, whereas overexpression in bone mar-row is associated with leukemia (Perkins andCory, �93) and overexpression in fibroblasts withfibrosarcoma (a cancer) (Aberdam et al., �91).

Thus, in mammals Hox genes are involved inpatterning of the skeletal axis and in the prolif-eration of cell lines (among other functions) andaberrations in the regulation of Hox genes maylead to abnormalities in both these functions.

Selection against cervical ribs

The occurrence of cervical ribs in various mam-malian species and the particularly frequent occur-rence of cervical ribs in experimental mice mutantsindicate that there is not a lack of genetic variationfor this phenotype. Thus, there must be strong sta-bilizing selection against the establishment of thistrait. The correlated incidence of cervical ribs andchildhood cancer presents a strong case for ap-parent selection against cervical ribs due to del-eterious pleiotropic effects. This correlation isstrengthened by the mice mutants that not onlydisplay variations in cervical vertebral number, butalso have cancer and a much reduced fitness(Akasaki et al., �96; van der Lugt et al., �96;Schumacher et al., �96; Coré et al., �97; He et al.,�97). The incidence of cervical ribs in the generalhuman population averaged over several large stud-ies (Adson and Coffey, �27; Etter, �44; Sycamore, �44;Crimm, �52; Menárguez Carretero and CampoMuñoz, �67) is approximately 0.2% (347 cases outof 220,026; percentages varied from 0.03�0.5). Thefrequency of these embryonal cancers added to-gether in the U.S. and Europe is approximately0.1% (0.01% Wilms� tumour (15), 0.033�0.075% leu-kemia (Stiller and Parkin, �96), 0.014% neuroblas-toma (Gurney et al., �96); braintumours 0.02�0.06%).

Assuming a chance for embryonal cancers of 0.1%and a chance for cervical ribs associated with em-bryonal cancers of 25% (Shumacher et al., �92) im-plies a 0.025% chance for children to have both acervical rib and early childhood cancer. Assuming afrequency of cervical ribs of 0.2% in the generalpopulation after early childhood and an average sur-vival of 60% for early childhood cancers (Miller etal., �95) implies that the total incidence of cervicalribs at birth is 0.21%, of which 11.9% will developan embryonal cancer. This suggests that childrenwith embryonal cancers have a 125-fold increasedincidence of cervical ribs (25% vs. 0.2%), and that

24


children born with a cervical rib have an almost120-fold chance of early childhood cancer (11.9% vs.0.1%). Thus, neonatal cancer alone seems to presentsufficient apparent selection against the establish-ment of cervical ribs.

In addition, the symptoms of TOS will enhancenatural selection against cervical ribs by directstabilizing selection. The seriousness of the symp-toms is correlated with the amount of manuallabour that is being performed. Therefore, undernatural circumstances the selective disadvantagewill be larger than in the sheltered present-dayhuman environment. Adults with a rudimentaryfirst rib (a partial transformation towards eightcervical vertebrae) often have TOS, suggestingnatural selection against this variation in cervi-cal vertebral number as well (Gelabert et al., �97).

A further selection factor against cervical ribscould be an increased chance of stillbirths. A largeminority (>30%) of fetuses between 49 and 150 mmhas ossification centers in the seventh cervicalprevertebra (Peters, �27; Noback, �51; Meyer, �78).These ossification centers appear in the same posi-tion as those of thoracic prevertebrae�s future ribs.An explanation of this phenomenon could be thatthe high percentage of ossification centers (cervicalribs) is related to the causes that have led to thepremature death of these fetuses. Again the inter-active nature of the early processes which involveHox genes may present a link between cervical ribsand other abnormalities, as it is unlikely that theossification centers themselves cause stillbirths. Itis possible that the problems in the proliferation ofcell lines that lead to neonatal cancer are also caus-ally related to the stillbirths.

A COMPARISON OF MAMMALS,REPTILES AND AMPHIBIANS

The number of cervical vertebrae is variable inamphibians, reptiles, and birds, in strong contrastto mammals (in fishes no cervical vertebral re-gion is distinguished). The selection against suchvariation in the number of cervical vertebrae mustbe considerably weaker or absent in these othervertebrate classes. In necropsy studies of zoo ani-mals, cancer rates of birds and reptiles are lowcompared to mammals (Fox, �12; Ratcliffe, �33;Ippen, �59; Lombard and Witte, �59; Effron et al.,�77). The low susceptibility to cancer in reptilesmakes intuitive sense because of their low meta-bolic rate, which leads to an expectation of lowoxidative DNA damage (cf. Adelman et al., �88;Perez-Campo et al., �98). The low susceptibility inbirds may seem surprising given their high meta-

bolic rate (McNab, �88; Ricklefs et al., �96). How-ever, there is evidence (from canaries and pigeons)that birds have a remarkably low free radical pro-duction and, thus, a low amount of oxidative dam-age (Perez-Campo, �98).

In addition, cancer in birds, especially in youngbirds, is generally believed in the majority of casesto be induced by viruses (Effron et al., �77; Reece,�96; Misdorp and Kik, personal communication).In mammals viral cancers are estimated to occurin 15% of cases, mainly liver cancer, cervical can-cer, and Hodgkin�s disease in children (Pisani etal., �97). A survey of 343,600 young chickensshowed that none developed a non-virally associ-ated cancer in the first five weeks of life whereas53 developed a virally associated cancer (Helmsley,�66). This pattern, confirmed by Reece (�96), is instriking contrast to that in human infants wherein the first month of life almost all cancers are non-virally associated embryonal cancers, predomi-nantly neuroblastoma (35%; Gurney et al.,�96).

In reptiles the viral induction of cancer has beenstudied much less. However, reptilian cancersseem more similar to cancers in birds than inmammals (Effron et al., �77) and the viruses thatinduce cancer in reptiles also seem more similarto those in birds than in mammals (Trubcheninovaet al., �77). In addition, reptiles with cancer atnecropsy are usually very old, and one study hasshown that snakes with cancer are even older onaverage than snakes without cancer (Ramsay etal., �96). In amphibians the situation is even lesswell documented; however, the one type of cancerthat is well documented, Lucké�s tumour in Ranapipiens, is a virally induced cancer (McKinnell andCarlson, �97).

There are a few mammalian species with an ab-errant number of cervical vertebrae: manatees andsloths. Sloths especially show a spectacular break-down of the constraint on variation as the num-ber of cervical vertebrae varies from 6 to 9 (Giffinand Gillett, �96). There is no explanation for theseexceptions, but I suggest as hypothesis that theextremely low metabolic rate of manatees andsloths (e.g., McNab �88; Gallivan and Best �89;Koteja �91; Hammond and Diamond, �97) is asso-ciated with low oxidative DNA damage and, thus,with a low susceptibility to cancer (Adelman etal., �88; Shigenaga and Ames, �93). This hypoth-esis needs to be tested.

CONCLUSIONS

It appears, therefore, that the cause of the con-servation of seven cervical vertebrae should be

25

24 F. GALIS

sought (1) in a genetic link between early child-hood cancer and stillbirths and variation in cer-vical vertebrae number, and (2) in the neuronalproblems leading to the thoracic outlet syndromein adults associated with cervical ribs. The in-volvement of Hox genes in the cancers that areassociated with cervical ribs in mice and menpoints to a coupling between functions of Hoxgenes that appears to be lacking in birds, reptiles,and amphibians, or at least has no apparent con-sequences when cervical vertebral number ischanged. There are two possible explanations forthe observed coupling in mammals: (1) the cou-pling of functions of Hox genes has newly ap-peared in mammals due to a change in thefunctioning of Hox genes (e.g., a new function inproliferation in mammals); and (2) the couplingof functions was already present in reptiles, buthidden because of the low susceptibility to can-cer. This coupling has only become detectable inmammals because of an increase in susceptibilityto cancer. And this increase in cancer susceptibil-ity can be the direct result of the increase in meta-bolic rate, which is associated with an increase inoxidative damage (Adelman et al., �88; Shigenagaand Ames, �93).

The increase in cancer susceptibility and thepresumed increase in stillbirths are pleiotropicdeleterious effects, whereas the neuronal prob-lems are a direct consequence of the change incervical vertebral number. The fact that the pleio-tropic effect of cancer recurs for what presum-ably are a large number of different mutationsallows us to classify these collectively as a devel-opmental constraint. To further understand theconstraint on changes in the number of cervicalvertebrae that exists in virtually all mammals, astudy of the function of Hox genes in cell prolif-eration and carcinogenesis in birds, reptiles, andamphibians is urgently needed. Furthermore, itshould be an interesting experiment to select forcomplete cervical ribs in a mammalian species tosee whether a healthy strain can be produced, orwhether this would lead to the predicted increasein susceptibility for cancer, stillbirths, and neu-ronal problems.

ACKNOWLEDGMENTS

I thank Hans Metz, Rogier Versteeg, GünterWagner, Jacques van Alphen, and Adam Wilkinsfor stimulating discussions and ideas, and MajaKik, Elliott Jacobson, Hans Feuth, and ProfessorMisdorp for medical and veterinary information.Russ Lande, Gerard Mulder, Louise Roth, Menno

Schilthuizen, Jan Sevenster, Elisabeth van Ast-Gray, Günter Wagner, Adam Wilkins, Ole See-hausen and an anonymous referee gave manyhelpful comments on the manuscript. Thanks toDavid Povel and Frank Alders for their help incollecting literature and to Adri �t Hooft and Mar-tin Brittijn for help with the figures.

LITERATURE CITEDAberdam D, Negreanu V, Sachs L, Blatt C. 1991. The onco-

genic potential of an activated Hox-2.4 homeobox gene inmouse fibroblasts. Mol Cell Biol 11:554�557.

Adelman R, Saul RL, Ames BN. 1988. Oxidative damage toDNA: relation to species metabolic rate and life span. ProcNatl Acad Sci USA 85:2706�2708.

Adson AW. 1947. Surgical treatment for symptoms producedby cervical ribs and the scalenus anticus muscle. SurgGynecol Obstet 85:687�700.

Adson AW, Coffey JR. 1927. Cervical rib. Ann Surg 85:839�857.

Akasaki T, Kanno M, Balling R, Mieza MA, Taniguchi M,Koseki HA. 1996. A role for mel-18, a polycomb group�re-lated vertebrate gene, during the anteroposterior specifica-tion of the axial skeleton. Development 122:1513�1522.

Albertazzi E, Cajone F, Lakshmi MS, Sherbet GV. 1998. Heatshock modulates the expression of the metastasis associ-ated gene MTS1 and proliferation of murine and humancancer cells. DNA Cell Biol 17:1�7.

Alkema MJ, Lugt NMT, Bobeldijk RC, Berns A, van LohuizenM. 1995. Transformation of axial skeleton due to over-expression of bmi-1 in transgenic mice. Nature 374:724�727.

Anbazhagan R, Raman V. 1997. Homeobox genes: molecularlink between congenital anomalies and cancer. Eur J Can-cer 33:635�637.

Bardeen CR. 1904. Numerical vertebral variation in the hu-man adult and embryo. Anat Anz 25:497�519.

Breslow NE, Olsen J, Moksness J, Beckwith JB, Grundy P.1996. Familial Wilms� tumor: a descriptive study. MedPediatr Oncol 27:398�403.

Brodeur GM. 1995. Genetics of embryonal tumours of child-hood: retinoblastoma, Wilms� tumour and neuroblastoma.Cancer Surv 25:67�99.

Charité J, de Graaff W, Deschamps J 1994. Ectopic expres-sion of Hoxb-8 causes duplication of the ZPA in the fore-limb and homeotic transformation in axial structures. DevDyn 204:13�21

Crimm PD. 1952. Evaluation of a five year minifilm program.Am J Roentg 68:240�246.

Coré N, Bel S, Gaunt SJ, Aurrand-Lions M, Pearce J, FisherA, Djabali M. 1997. Altered cellular proliferation and me-soderm patterning in Polycomb-M33-deficient mice. Devel-opment 124:721�729.

Corte G, Airoldi I, Briatat P, Corsetti MT, et al. 1993. Thehomeotic gene products in the control of cell differentiationand proliferation. Cancer Detect Prev 17:261�266.

Dongen PAM van. 1998. Brain size in vertebrates. In:Nieuwenhuys R, ten Donkelaar HJ, Nicholson C, editors.The central nervous system of vertebrates, volume 3. Ber-lin: Springer Verlag. p 2099�2134.

Effron M, Griner L, Benirschke K. 1977. Nature and rate ofneoplasia in captive wild mammals, birds and reptiles atnecropsy. J Natl Cancer Inst 59:185�198.

Epstein M, Pillemer G, Yelin R, Yisraeli JK, Fainsod A. 1997.

26


Patterning of the embryo along the anterior-posterior axis:the role of the caudal genes. Development 124:3805�3814.

Etter LE. 1944. Osseous abnormalities of the thoracic cageseen in forty thousand consecutive chest photoroentgeno-grams. Am J Roentg 5:359�363.

Fox H. 1912. Observations upon neoplasms in wild animalsin the Philadelphia Zoological Gardens. J Path Bact 17:217.

Gelabert HA, Machleder HI. 1997. Diagnosis and manage-ment of arterial compression at the thoracic outlet. Ann VascSurg 11:359�366.

Gossler A, Hrabe de Angelis M. 1998. Somitogenesis. CurrTop Dev Biol 38:225�287.

Gurney JG, Davis S, Severson RK, Fang J-Y, Ross JA, RobisonLL. 1996. Trends in cancer incidence among children in theU.S. Cancer 78:532�541.

He T-C, Costa LT, Thiagalingam S. 1997. Homeosis and poly-posis: a tale from the mouse. Bioessays 19:551�555.

Helmsley LA. 1966. The incidence of tumours in young chick-ens. J Path Bact 92:91�96.

Hoeven F van der, Sordino P, Fraudeau N, Izpisúa-Belmon-teand JC, Duboule D. 1996. Teleost HoxD and HoxA genes:comparison with tetrapods and functional evolution of theHoxD complex. Mech Dev 54:9�21.

Horan GSB, Nagy Kovacs E, Behringer RR, Featherstone MS.1995. Mutations in paralogous Hox genes result in overlap-ping homeotic transformations of the axial skeleton: evi-dence for unique and redundant function. Dev Biol 169:359�372.

Ippen R. 1985. Geschwülste. In: Ippen R, Zwart P, SchröderH-D, editors. Handbuch der zootierkrankheiten, band 1,reptilien. Berlin: Akademie-Verlag. p 270�301.

Knudson AG. 1984. Genetic predisposition to cancer. CancerDet Prev 7:1�8.

Kongsuwan K, Webb E, Housiaux P, Adams JM 1988. Ex-pression of multiple homeobox genes within diverse mam-malian haematopoietic lineages. EMBO J 7:2131�2138.

Krumlauf R. 1994. Hox genes in vertebrate development.Cell78:191�201.

Kusewitt DF, Reece RL, Miska KB. 1997. S-100 immunore-activity in melanomas of two marsupials, a bird, and a rep-tile. Vet Pathol 34:615�618.

Lawrence HJ, Sauvageau G, Humphries RK, Largman C.1996. The role of Hox homeobox genes in normal and leu-kemic hematopoiesis. Stem Cells 14:281�291.

Leboucq H 1898. Recherches sur les variations anatomiquesde la première côte chez l�homme. Archiv Biol 15:9�178.

Lombard LS, Witte EJ. 1959. Frequency and types of tu-mors in mammals and birds of the Philadelphia zoologicalgarden. Cancer Res 19:127�141.

Lugt NMT van der, Domen J, Linders K, van Roon M,Robanus-Maandog E, Riele H, van der Valk M, DeschampsY, Sofronio M, van Gohuinen M, Berns A. 1996. Posteriortransformation, neurological abnormalities, and severe he-matopoietic defects in mice with a targeted deletion of thebmi-1 proto-oncogene. Genes Dev 8:757�769.

Makhoul RG, Machleder HI. 1992. Developmental anomaliesat the thoracic outlet: an analysis of 200 consecutive cases.J Vasc Surg 16:534�545.

Manohar CF, Salwen HR, Furtado MR, Cohn SL. 1996. Up-regulation of Hoxc6, Hoxd1, and Hoxd8 homeobox gene ex-pression in human neuroblastoma cells following chemicalinduction of differentiation. Tumor Biol 14:34�47.

McLain K, Schreiner C, Yager KL, Stock JL, Potter SS. 1992.Ectopic expression of Hox-2.3 induces craniofacial and skel-etal malformations in transgenic mice. Mech Dev 39:3�16.

Maris JM, Brodeur GM. 1997. Are certain children more likelyto develop neuroblastoma? J Pediatr 131:656�657.

McKinnell RG, Carlson DL. 1997. Lucké renal adenocarci-noma, an anuran neoplasm: studies at the interface of pa-thology, virology, and differentiation competence. J CellPhysiol 173:115�118.

McNab BK. 1983. Energetics, body size, and the limits toendothermy. J Zool Lond 199:1�29.

Meyer DB. 1978. The appearance of �cervical ribs� during earlyhuman fetal development. Anat Rec 190:481.

Menárguez Carretero L, Campo Muñoz M. 1967. Estudioradiologico y tipos morfologicos de costillas cervicales en elsexo femenino. Enferm Torax 16:285�308.

Miller RW, Young JL, Novakovic B. 1995. Childhood cancer.Cancer 75(suppl):395�405.

Narod SA, Hawkins MM, Robertson CM, Stiller CA. 1997.Congenital anomalies and childhood cancer in Great Brit-ain. Am J Hum Genet 60:474�485.

Noback CR, Robertson GG. 1951. Sequences of appearanceof ossification centers in the human skeleton during thefirst five prenatal months. Am J Anat 89:1�28.

Perez-Campo R, López-Torres M, Cadenas S, Rojas C, BarjaG. 1998. The rate free radical production as a determinantof the rate of aging: evidence from the comparative ap-proach. J Comp Physiol B 168:149�158.

Perkins AC, Cory S. 1993. Conditional immortalization of mousemyelomonocytic megakaryocytic and mast cell progenitors byte Hox-2.4 homeobox gene. EMBO J 12:3835�3846.

Peters H. 1927. Varietäten der Wirbelsäule menschlicherEmbryonen. Gegenbaurs Morph Jahrbuch 58:440�477.

Pisani P, Parkin DM, Muñoz M, Ferlay J. 1997. Cancer andinfection: estimates of the attributable fraction in 1990. Can-cer Epidemiol Biomarkers Prev 6:387�400.

Pui C-H, Kane JR, Crist WM. 1995. Biology and treatmentof infant leukemias. Leukemia 9:762�769.

Ramsay EC, Munson L, Lowenstine L, Fowler L, Fowler ME.1996. A retrospective study of neoplasia in a collection ofcaptive snakes. J Zoo Wildl Medicine 27:28�34.

Ratcliffe HL. 1933. Incidence and nature of tumors in cap-tive wild mammals and birds. Am J Cancer 17:116�135.

Reece RL. 1996. Some observations on naturally occurringneoplasms of domestic fowls in the state of Victoria Austra-lia 1977�1987. Avian Pathol 25:407�447.

Ricklefs RE, Konarzeweski M, Daan S. 1996. The relation-ship between basal metabolic rate and daily energy expen-diture in birds and mammals. Am Nat 147:1047�1071.

Roos DB. 1996. Historical perspectives and anatomic consid-erations. Semin Thorac Cardiovasc Surg 8:183�189.

Satgé D, Sasco AJ, Carlsen NLT, Stiller CA, et al. 1998. Alack of neuroblastoma in Down Syndrome: a study from 11European countries. Cancer Res 58:448�452.

Schulz AH. 1961. Vertebral column and thorax. In: Hofer H,Schultz AH, Starck D, editors. Primatologia, handbook ofprimatology. S. Karger. p 5/1�5/66.

Schumacher A, Faust C, Magnuson T. 1996. Positional clon-ing of a global regulator of anterior-posterior patterning inmice. Nature 383:250�253.

Schumacher R, Mai A, Gutjahr P. 1992. Association of ribanamalies and malignancy in childhood. Eur J Pediatr151:432�434.

Shigenaga MK, Ames BN. 1993. Oxidants and mitogen-esis as causes of mutation and cancer: the influence ofdiet. In: Bronzetti G, editor. Antimutagenesis and anti-carcinogenesis mechanisms III. New York: Plenum Press.p 419�436.

27

26 F. GALIS

Starck D. 1979.Vergleichende Anatomie der Wirbeltiere. Ber-lin: Springer Verlag.

Stiller CA, Parkin DM. 1996. Geographic and ethnic varia-tions in the incidence of childhood cancer. Brit Med Bull52:682�703.

Sycamore LK. 1944. Common congenital anomalies of thebony thorax. Am J Roentg 51:593�599.

Trubcheninova LP, Khutoryansky AA, Svet-Moldavsky GJ,Kuznetsova LE, et al. 1977. Body temperature and tumor

virus infection: tumorigenicity of Rous sarcoma virus forreptiles. Neoplasma 24:3�19.

Turner HN. 1847. Observations on the distinction betweenthe cervical and dorsal vertebrae in the class mammalia.Proc Zool Soc Lond 15:110�114.

Woolfenden GE. 1961. Postcranial morphology of the water-fowl. Bull Florida State Museum Biol Sci 6:1�129.

Yu BD, Hess JL, Horning SE, Brown GA, Korsmeyer SJ. 1995.Altered Hox expression and segmental identity in Mll-mu-tant mice. Nature 387:505�508.

28

TRABAJO PRÁCTICO 2 : EJERCITACIÓN

Problema 1.

Responda verdadero o falso y justifique TODOS los casos.

1) La homología serial a nivel estructural es un concepto análogo a la duplicación ydivergencia a nivel genético. A manera de ejemplo, se asume que a partir de una vértebra ancestral se generó un mayor número de vértebras en los taxa derivados a través de su duplicación, y análogamente el origen de una familia multigénica de varios genes parálogos a través de la duplicación de un gen ancestral con la posterior divergencia.

2) Las heterocronías son cambios en el tiempo de desarrollo de una estructura dada endiferentes taxa. Es decir que, si dos grupos hermanos tienen el mismo tiempo de desarrollo de la estructura estudiada, no hubo procesos heterocrónicos a lo largo de la evolución del linaje.

3) La ontogenia recapitula la filogenia.

Problema 2.

La evolución de los apéndices es uno de los temas más estudiados bajo una perspectiva filogenética, de embriología comparada y experimental. En particular dentro de los tetrápodos, los anfibios son un grupo de gran diversidad en cuanto al número de dígitos y falanges.

Entre los anfibios, existen dos grandes órdenes: los Anuros (ranas) y los Urodelos (salamandras), ambos, por lo general, poseen 5 dígitos en sus apéndices posteriores. Diferentes estudios sugieren que el ancestro de estos dos grupos poseía un apéndice posterior caracterizado por 2 falanges en el primer dígito, 3 en el segundo, 4 en el tercero, 4 en el cuarto y 3 en el quinto (formula del número de falanges: 23443).

Por lo general, las especies de Anuros son muy conservadas en el número de falanges con una fórmula de falanges de 22343 (ver Figura 1). Por el contrario, en los Urodelos la fórmula del número de falanges es mucho más variable que en los Anuros.

29

En la Tabla I se esquematiza la secuencia de diferenciación de las falanges durante la ontogenia en dos especies modelo Xenopus laevis y Ambystoma mexicanus, anuro y urodelo, respectivamente. De acuerdo a esta información:

2.1. ¿es sincrónico el desarrollo de los dígitos en cada especie?

2.2. ¿Qué clase de homología está implícita en esta comparación? Defínala.

30

Pere Alberch, gran embriólogo evolutivo, abordó de manera experimental el problema de la variación en el número de falanges en los anfibios, tratando con un inhibidor mitótico (colchicina) durante las etapas tempranas del desarrollo de los apéndices posteriores de ambas especies. Como resultado de este experimento se observó una reducción en el número de falanges en los 5 dígitos de cada especie. La información de dicho experimento se resume en las Figuras 1 y 2.

2.3. ¿Encuentra algún tipo de patrón que relacione la reducción del número de dígitos con el tiempo de desarrollo de los mismos? Formule una hipótesis que vincule ambos resultados.

En la naturaleza se puede observar especies de anfibios con 4 dedos, tanto en el orden de los Anuros como en el de los Urodelos. Tal es el caso de la especie Psyllophryne didactyla (Anuros) cuya fórmula de numero de falanges es: 02341, como la especie Salamandrina terdigitata (Urodelo) cuya fórmula es de: 12320. Si este patrón natural concuerda con la hipótesis planteada anteriormente por Ud.:

2.4. ¿El dedo “pulgar” (el primer dígito con falange) de ambas especies es homólogo (primario)? Justifique su respuesta utilizando los criterios de homología. Defina los términos alometría evolutiva y alometría ontogenética e identifique un ejemplo de ambos en el caso de estudio.

31

Problema 3.

Las alas son una adaptación muy importante de los insectos, la cual les permite escapar de los predadores, explotar recursos y dispersarse a nuevos nichos. Pese a las supuestas ventajas evolutivas de las alas, casi todos los órdenes de insectos alados poseen linajes con reducción parcial o total de las alas. En este sentido, cientos de transiciones independientes de formas aladas a no aladas han ocurrido durante la evolución de las alas. Sin embargo, una reversión de una forma sin alas a una con alas nunca ha sido demostrada claramente, dado que tal reversión sería muy improbable dadas las complejas interacciones entre nervios, músculos, y escleritos que se requieren para el vuelo.

3.1. Bajo esta hipótesis inicial: ¿son homólogas las alas de los distintos insectos? ¿Y el primer y segundo par dentro de cada especie? Indique que información utilizaría para estudiar alometrías estáticas, ontogenéticas y evolutivas del ala dentro de algún género en particular.

3.2. Un estudio reciente de la evolución de las alas dentro del orden Phasmatodea (los llamados “bichos palo”) mostró el siguiente resultado:

Referencias. Estrella: ganancia de alas. Triángulo: pérdida de alas. Circulo vacío: especies con alas.

Grupos externos

32

¿Modifica este resultado su respuesta anterior? Justifique.

3.3. Un tercer estudio mostró que los genes involucrados en el patrón de desarrollo de las alas están presentes en todos los insectos, incluso los ápteros. Proponga una hipótesis para explicar este resultado y (de ser necesario) reconciliar las hipótesis anteriores.

Problema 4.

Adams y Nistri (BMC Evolutionary Biology 2010) estudiaron la morfología de las patas de ocho especies de salamandras europeas del género Hydromantes para poner a prueba la hipótesis de que la morfología del adulto constituye una adaptación relacionada con sus hábitos arborícolas. Estos autores observaron que cinco especies habitantes de la isla de Cerdeña alcanzan un tamaño mayor que las otras tres especies pertenecientes al continente. Sin embargo, todas las especies muestran valores similares y relativamente bajos de sinuosidad lo que implica la presencia de más tejido interdigital (Fig. 1).

Figura 1: Esquemas de la pata de las salamandras estudiadas que muestran la mayor sinuosidad y menor presencia de tejido interdigital en individuos pequeños (a la izquierda) en comparación con individuos grandes (a la derecha).

Cuando estos autores analizaron la morfología a lo largo del desarrollo de cada especie, encontraron patrones muy diferentes (Tabla 1). Por un lado, la mayoría de la especies no mostró cambios en el nivel de sinuosidad (o en la abundancia de tejido interdigital) a través de la ontogenia. Por otro lado, en tres especies habitantes de la isla, se observó que el nivel de sinuosidad disminuye (o que la cantidad de tejido interdigital aumenta) a medida que los animales crecen.

33

Tabla 1: Principales resultados de los análisis de regresión de la sinuosidad versus el tamaño total. Para cada especie se muestra el valor de la pendiente de la recta de regresión y la significación asociada. p < 0,05 indicado en negrita, c: especie continental, i: especie isleña.

a) Interprete los resultados de la Tabla 1. Identifique el/los tipo/s de alometría/sanalizados.

A continuación, para comprender mejor la evolución de las trayectorias ontogenéticas analizadas, los autores utilizaron una filogenia del grupo y encontraron que el patrón de independencia entre caracteres es el estado ancestral más probable para el linaje europeo (Fig. 2).

34

Figura 2: Relaciones filogenéticas de las especies de salamandras europeas estudiadas basadas en una filogenia molecular realizada utilizando datos de secuencias de ADN mitocondrial y nuclear. Los esquemas muestran el patrón de alometría asociado a cada especie. c: especie continental, i: especie isleña.

b) En base a estos resultados, ¿qué conclusiones puede extraer acerca de la homologíadel patrón de crecimiento alométrico observado en las tres especies isleñas? ¿Cómo podría explicar estos resultados considerando el concepto de heterocronía?

Posteriormente, se estudió la trayectoria ontogenética de una especie de salamandra norteamericana (H. platycephalus) la cual exhibe un crecimiento alométrico similar al observado en las especies isleñas. Cuando se incluyó a H. platycephalus en el análisis, los patrones descriptos anteriormente no cambiaron pero si lo hizo la condición ancestral de todo el género la cual pasó a ser alométrica.

c) ¿Modifica este resultado su hipótesis de homología expuesta en el punto anterior?Justifique.

35

Finalmente, los autores extendieron estos análisis a más especies de salamandras pertenecientes a otros géneros y observaron que las trayectorias ontogenéticas de las especies europeas se asemejan, por un lado, a especies que poseen hábitos similares (es decir, especies trepadoras) y, por otro lado, a especies cavadoras las cuales presentan una posición más basal en la filogenia de las salamandras.

d) ¿Cómo podría explicar estos resultados en términos de exaptación?

Problema 5.

Relacione cada uno de los siguientes gráficos con uno de los enunciados a continuación. Justifique sus respuestas.

a. El estudio comparado de las alometrías a nivel craneofacial no permite inferir procesos heterocrónicos dado que en la muestra se hallan subrepresentados algunos estadios ontogenéticos de ambas especies. También se sospecha que una de las muestras ontogenéticas incluye ejemplares de más de una especie.

b. Un proceso heterocrónico de tipo pedomórfico actuó a nivel craneofacial en humanos mientras que no se puede postular un cambio en los tiempos de desarrollo de dichas estructuras en chimpancés.

c. La relación entre la forma del neurocráneo y la forma de la cara en humanos presenta un desarrollo desacelerado desde la adolescencia a la adultez con respecto a estadios anteriores de su ontogenia.

d. El estudio de las alometrías a nivel craneofacial no permite inferir procesos heterocrónicos en la divergencia entre los linajes de humanos y chimpancés.

e. Nuestra especie se caracteriza por una acentuada aceleración de los ritmos de desarrollo.

36

Trabajo Práctico Nº 1 - ege.fcen.uba.ar · TRABAJO PRÁCTICO 1 : SEMINARIO Lea atentamente los...

Documents

Transcript of Trabajo Práctico Nº 1 - ege.fcen.uba.ar · TRABAJO PRÁCTICO 1 : SEMINARIO Lea atentamente los...