A pragmatic approach for selecting evodevo model species in amniotes

Bearded Dragon Secret Manual Review

Complete Bearded Dragon Care

Get Instant Access

athanasia c. tzika and michel c. milinkovitch

One major classical justification of using a model metazoan species for experimentation has been that discoveries of biological phenomena in that species could be extrapolated to other multicellular species. Because the chances that this extrapolation is valid in humans depend on the phylogenetic distance between humans and the model species, many researchers have somewhat sacrificed the major benefits of small size, short generation time and ease of manipulation that characterise some invertebrates in order to use species that humans can more readily relate to, such as the laboratory mouse (Mus musculus). However, the community of biologists has continued to use additional model species because each of the selected taxa have specific features that make experimental manipulation easier (e.g. easy-to-score morphological variation and giant polytene chromosomes in Drosophila melanoga-ster, or accurate description of the largely invariant complete cell lineage and full neural connectivity in the roundworm Caenorhabditis elegans).

Ever since the molecular genetic revolution, a constant concern has been the possibility of manipulating the genome of model species. For example, generations of Drosophila scientists have developed and applied ingenious approaches that allow, in principle, screening for any phenotype at any stage of development (reviewed in St Johnston 2002). Even for the mouse model, multiple techniques, such as homologous recombination, tissue-specific activation/inactivation techniques, cloning and RNA interference (RNAi), have been developed for performing genotype- or phenotype-driven experiments. Furthermore, recent access to full genome sequences makes genome engineering of some

Evolving Pathways: Key Themes in Evolutionary Developmental Biology, ed. Alessandro Minelli and Giuseppe Fusco. Published by Cambridge University Press. # Cambridge University Press 2008.

model species easier. However, one of the most important limiting factors to the utility of genome engineering approaches is the difficulty with which phenotypes can be identified. Indeed, many genes exert functions that cannot be investigated by simple examination of a few general morphological and/or physiological parameters. Fortunately, imaging technologies and physiological measuring techniques have recently been miniaturised and adapted for use with small species. Another source of difficulty for phenotyping is due to the epistatic effects among genes. For example, multiple studies have demonstrated that a full knock-out mouse for a gene supposedly essential for a given major process can yield no visible phenotype because another gene has become involved to fulfil the function of the invalidated gene.

medicine-driven models (searching for evolutionary conservation) vs. evo-devo-driven models

(understanding evolutionary change)

As indicated above, most of the research performed so far with model species has been justified by the potential power of these species for understanding human biology. However, in the context of evo-devo, it is the massive realm of living species that should, ideally, be opened to genome manipulation and phenotypic investigation. Indeed, the interests of evolutionary developmental biologists go well beyond conserved physiologies and developmental processes or patterns as they seek to understand the generative mechanisms underlying biological diversity (Minelli 2003). Uncovering these mechanisms will require the merging of several disciplines (Milinkovitch and Tzika 2007), including molecular developmental biology, evolutionary molecular genetics, palaeontology (Wagner and Larsson 2003) and ecology (Dusheck 2002, Gilbert and Bolker 2003). Hence, one major challenge in evo-devo will be to adapt the tremendous knowledge and sophisticated technologies accumulated on 'classical' model species to model organisms from a wider assortment of lineages on the tree of life in a wider set of environmental conditions. However, as demonstrated in the past, promoting the use of the same set of model species increases the efficiency with which techniques and analytical approaches are developed simply through collaboration, emulation and establishment of public databases. For example, given the necessity to optimise the interplay between in-silico analyses and in-vivo experiments, it is crucial to establish the list of preferred species for which the genome will be sequenced in priority, and this should be done with more input from the evo-devo community (Milinkovitch and Tzika 2007).

criteria for choosing new model species

Beside the practical criteria of small size, short generation, abundant progeny, ease of manipulation and of housing/breeding, accessibility of phe-notyping and genome manipulation techniques, etc., there are other parameters that should be considered when listing preferred model species.

An intuitive and simple criterion to guide the choice of model species is the evolutionary divergence among these species (Figure 7.1). However, this phylogenetic distance criterion is limited by, among others, two parameters: (1) the rate of phenotypic transformation is highly variable among lineages and (2) variation worth investigation exists at multiple phylogenetic levels (one should not focus only on major transformations). Intermingled with the phylogenetic distance criterion, the ancestrality of a model species is a decisive factor. For example, the zebrafish is often considered a canonical vertebrate (Fishman 2001) because the common ancestor of all vertebrates was fish-like. Although intuitively appealing, this statement is of limited value because, a priori, no extant species is intrinsically more ancestral than any other. So, the real, non-trivial, question is: what is the considered species a model of (i.e. at what hierarchical level(s) of the phylo-geny)? The answer depends heavily on what characters one is interested in. For example, as far as anatomy is concerned, the zebrafish might be a better model for teleost fishes than for vertebrates because the species exhibits multiple characters that seem ancestral for the former group, whereas it is very derived (as all teleost fishes are) in respect to the vertebrate ancestor (Metscher and Ahlberg 1999). Furthermore, although we agree that structural and genomic simplicity can help in guiding the choice of a model species (as in the case of the cephalochordate amphioxus; Holland et al. 2004), it should be carefully investigated whether the observed simplicity is ancestral for the group and not a secondary (derived) simplification, as is likely to be the case for flatworms and myzostomes, for example (Bleidorn et al. 2007).

The availability of robust and extensive (molecular) phylogenetic hypotheses constitutes an important criterion for choosing evo-devo model species: evolutionary trees constitute the basic framework on which character changes are mapped. For example, mapping of phenoty-pic characters on robust molecular phylogenies have demonstrated extensive and multiple convergences of ecologically specialised species

Amniote Evolutionary Tree

Figure 7.1 Right: cladogram showing the best estimate of the phylogeny of amniotes based on molecular data. The Eutheria cladogram is based on Bayesian and Maximum Likelihood analyses of 16 397 bp of DNA from 19 nuclear and 3 mitochondrial genes from 42 placental and 2 marsupial extant species (Murphy et al. 2001b). All nodes are supported by posterior probabilities >95% except the nodes indicated with an asterisk. Note that current analyses of morphological data still disagree with a significant portion of this cladogram. Left: phylogram showing the time of divergence among major amniote lineages. Divergence times (and 95% credibility intervals, grey bars) within Eutheria were obtained (Springer et al. 2003) by analysing the 16 397 bp dataset cited above under a Bayesian approach allowing rate variation among lineages and incorporating fossil information in the form of constraints on divergence times (Kishino et al. 2001). Some palaeontologists favour a model of radiation of placental mammals after the Mesozoic/Cenozoic boundary. Grey bars on the Marsupials versus Eutherians, and Monotremes versus Theria splits are the lowest and highest estimates of divergence time (i.e. considering standard errors) obtained with two molecular data sets of 2 793 bp for 21 taxa and 10 773 bp from 5 taxa, respectively (van Rheede et al. 2006). Divergence time for other nodes on the tree (arrows) are minimum age estimates obtained from palaeontological data (Benton and Donoghue 2007). Dotted lines indicate that no reliable estimate for the divergence of snakes from lizards is available. Two lineages of lizards are indicated to underline the likely paraphyly of that group (with respect to snakes). Geological eras are indicated (Palaeozoic, 570-251 mya; Mesozoic, 251-65 mya; Cenozoic, 65-0 mya).

(ecomorphs), e.g. of cichlid fishes (Kocher et al. 1993, Ruber et al. 1999), ranid frogs (Bossuyt and Milinkovitch 2000), Anolis lizards (Losos et al. 1998) and mammals (e.g. between some afrotherian and eulipotyphlan insectivores, see below). Similarly, the snake-like body form has evolved multiple times independently in squamate reptiles (Wiens et al. 2006). Investigating the development of such convergent traits in different lineages could form the basis for understanding possible general mechanisms involved in convergence (see Jenner, Chapter 6 of this volume, for a discussion on nomothetic versus idiographic approaches to evo-devo). Such analyses will require a strategy of choosing model organisms based on their traits rather than phylogenetic position per se (Figure 7.1).

a multidisciplinary pragmatic optimisation approach to the selection of model species

As further developed elsewhere (Milinkovitch and Tzika 2007), the criteria that are relevant to the choice of a set of model species are multiple and can even be contradictory. This is intrinsic to the highly multidisciplinary nature of evo-devo. For example, a lineage can be characterised by a unique and dramatic set of derived character states that makes it particularly appealing for evo-devo studies, but the group might lack representatives that could reasonably constitute a widespread model species in the laboratory (e.g. take the extreme case of cetaceans). We think that the only possibility is the use of a pragmatic (and partly subjective) optimisation approach, incorporating criteria such as phylogenetic position (Figure 7.1) as well as number and nature of the ancestral/derived character states of the model species, level of diversity within a relevant higher taxon to which the chosen species belongs, ease with which the representative species can be handled, housed and bred, and their protection status. Compromises will have to be made, as it is simply impossible to find species that combine all possible advantageous features. We apply such an approach below for a set of species that could serve as the workhorses for evo-devo research within amniotes. We shortly discuss the advantages and disadvantages of some candidate species and hope this will be used as a starting point for an in-depth analysis with input from morphologists, palaeontologists, animal breeders, physiologists, developmental biologists and molecular phylogeneticists.


Several lineages within Reptilia (including birds) are more genetically diverse than mammals such that using a single reptilian species from a single lineage seems insufficient, especially given that the Mammalia is already represented by several model species (see below).



This clearly monophyletic group (turtles, tortoises and terrapins) includes only about 260 extant species in 13 families and is characterised by several synapomorphies of which the most obvious is the shell (enclosing front and hind limb girdles) comprising a dorsal carapace (made from dermal bones associated with endochondral modified vertebrae and ribs) and a ventral plastron (made of clavicles, interclavicles and abdominal ribs). They have relatively low metabolic rates and are dependent on external sources of body heat, although the leatherback turtle (Dermochelys coriacea) seems to show some degree of endothermy. High longevity might also be a character of interest. Long generation time is an obstacle to using testudines as models: in most species, males and females reach sexual maturity not before 3 and 5 years, respectively. Analysis of development is feasible as eggs can be incubated artificially (Gilbert et al. 2001).

Two species have been suggested as candidates for full genome sequencing (www.reptilegenome.com): Trachemys scripta (red-eared slider) and Chrysemys picta (painted turtle). Individual size, and housing and handling techniques, are very similar for the two species but the latter is much more difficult to breed and much less prolific than the former. Shell morphogenesis has been investigated in Trachemys (Gilbert et al. 2001), whereas a BAC library is available for Chrysemys (www.nsf.gov/bio/pubs/awards/bachome.htm), aiding the sequencing of its genome. For sure, only one of these two species should be sequenced as they are redundant phylogenetically (they belong to the same subfamily Deirochelyinae, in the family Emydidae). These species are closely related to the tortoises (family Testudinidae) (Krenz et al. 2005), of which several species in the genus Testudo are successfully bred around the world and detailed husbandry information is available (Lapid et al. 2004).

We think that an additional species, Pelodiscus sinensis (Chinese soft-shell turtle, see Figure 7.2A), deserves attention from the evo-devo community. This species is the smallest soft-shell turtle, and belongs to a superfamily (Trionychoidea) phylogenetically distant from the species discussed above. Housing and breeding of Pelodiscus sinensis are

Figure 7.2 Some of the proposed new model species for amniotes: A, Pelodiscus sinensis, the Chinese soft-shell turtle; B, Pogona vitticeps, the bearded dragon; C, Elaphe guttata, the corn snake; D, Crocodylus niloticus, the Nile crocodile; E, Petaurus breviceps, the sugar glider (the left and right insets show the patagium and the embryo in the mother pouch, respectively); F, Echinops telfairi, the lesser hedgehog tenrec; G, Atelerix albiventris, the African pygmy hedgehog. A colour version of this figure is available at www.ulb.ac.be/sciences/ueg/modelsp.htm

Figure 7.2 Some of the proposed new model species for amniotes: A, Pelodiscus sinensis, the Chinese soft-shell turtle; B, Pogona vitticeps, the bearded dragon; C, Elaphe guttata, the corn snake; D, Crocodylus niloticus, the Nile crocodile; E, Petaurus breviceps, the sugar glider (the left and right insets show the patagium and the embryo in the mother pouch, respectively); F, Echinops telfairi, the lesser hedgehog tenrec; G, Atelerix albiventris, the African pygmy hedgehog. A colour version of this figure is available at www.ulb.ac.be/sciences/ueg/modelsp.htm

very well documented because numerous farms exist, mainly in Asia, with a production exceeding six million individuals per year. Furthermore, Hox genes have been characterised and their expression patterns investigated (Ohya et al. 2005), details of the shell development are known (Kuraku et al 2005) and a turtle-chicken chimera has been generated (Nagashima et al 2005). In addition, several cDNA libraries have been developed (Kuraku et al 2005) and ideal incubation conditions have been identified (Du and Ji 2003).

All species mentioned above belong to the Cryptodira. Less information is available for members of the Pleurodira. Possibly the best candidate as a Pleurodira model species is Emydura subglobosa, the pink-bellied sideneck, a mid-sized (about 25 cm) aquatic turtle native to Australia and Papua New Guinea. The species is robust and prolific, and is becoming an increasingly popular pet around the world.



Sequencing of the full genome of Anolis carolinensis, the green anole lizard, has recently been initiated (www.genome.gov/10002154). Individuals are small (average 13 cm for females and 17 cm for males), and their life span is 2-7 years. Their maintenance is, however, not trivial. As most lizards, they are oviparous although their reproduction presents a peculiar feature: females produce a single egg every 1 to 2 weeks with successive alternation of the offspring's sex, a phenomenon that seems hormonally controlled by the female (Lovern and Passek 2002). However, only an average of seven eggs are laid per year (the maximum being 20), limiting the material available for developmental biology studies. Extensive literature on the species (especially behavioural and ecological studies) has accumulated since the 1960s, mainly owing to the animal's small size and abundance of the species in North America (see Lovern et al. 2004 for a review).

We consider, however, that an alternative species, Pogona vitticeps (bearded dragon; see Figure 7.2B), deserves attention from the evo-devo community. Bearded dragons are larger than anoles as they have an average adult size of about 40 cm, and they have recently become a common pet all around the world. Indeed, these animals are tame, easy to handle and feed, and much information is accessible on their maintenance and reproduction. Female bearded dragons are significantly more prolific than the green anole as the former can lay about 24 eggs up to six times a year.


The Reptilian Genomics Working Group has suggested the sequencing of the full genome from Thamnophis sirtalis (the common garter snake). This is a relatively small-sized species (46-137 cm) easy to maintain in captivity. Individuals reach sexual maturity at about 55 cm (about 2 years old) and live in captivity up to 10 years. After a three-month gestation, females produce clutches of 4 to 80 offspring, depending on the size of the mother. Given its accessibility in nature in multiple areas of the United States, this species has been widely used for physiology and ecology research programs, but not in developmental biology. A key reproductive feature could prove problematic: Thamnophis sirtalis is ovo-viviparous. Although ovoviviparity (and associated placental membranes) is by itself a subject worth investigating (Krohmer 2004), this feature seriously limits the accessibility of embryos at any developmental stages unless the mother is sacrificed.

This is why we suggest that Elaphe guttata (the corn snake, see Figure 7.2C) constitutes a valid alternative to the common garter snake. It has the great advantage of being oviparous, such that embryos can be incubated artificially for investigation at any stage of the 55-70 days of development after egg laying (i.e. about 50 days after fertilisation) (Kohler 2005).



The large size and long generation time (sexual maturity depends on size but in no species do individuals reach sexual maturity before the age of 8) is a discouraging factor for their use in evo-devo studies, but we think the task is not impossible because eggs are available in crocodile farms all around the world, mostly for Alligator mississippiensis and Crocodylus niloticus. Little information is available on the development of crocodiles (Guillette and Gunderson 2001, Tissir et al. 2003, Milnes et al. 2004), and their sister-group relationship with birds makes this group especially interesting for evo-devo studies. The need to collaborate with crocodile farms is probably the most important constraint on the choice of species. Not surprisingly, the US-based Reptilian Genomics Working Group recommends the sequencing of the Alligator mississippiensis (American alligator) genome, but this species is unfortunately common only in the United States, whereas Crocodylus niloticus (Nile crocodile, see Figure 7.2D) is much easier to find in European and African farms and zoos and is phylogenetically closely related to Crocodylus porosus, the species most studied by Australian laboratories. Furthermore, A. mississippiensis could be advantageously replaced by another representative of the Alligatorinae subfamily: the Cuvier's dwarf caiman (Paleosuchus palpebrosus), which has the great advantage of reaching a small adult size of about 130 cm. However, its generation time is not significantly shorter than those of larger species and it remains a dangerous animal to handle (Grenard 1991).


Birds, the sister group to Crocodylomorpha, constitute a highly diverse group comprising about 10 000 extant species in about 30 orders. It is very likely that Gallus gallus, the domestic fowl (Galliformes), will remain the focal representative model species of Aves because it exhibits a short generation time (sexual maturity is reached at the age of about 5 months), its reproductive output is staggering (a single hen can lay about 365 eggs a year and a single rooster can produce 100 000 offspring a year), much information is available on the embryonic development, captive holding and breeding techniques incorporating automated feeding and egg collection are very well documented and used all around the world, numerous inbred strains (exhibiting different pheno-types) are available, and regulations associated with captive holding/ breeding are well known and much simpler than for other farm animals. The domestic fowl has served as a model system since Aristotle, and new technologies, such as in vivo electroporation and transgenesis, have been continuously incorporated. The development of chicken embryonic stem (ES) cells is an active topic and the genome of the species has been sequenced (Burt 2005, Stern 2005).

Despite the overwhelming advantages discussed above for the domestic fowl, other species have been used as models. For example, multiple species and strains of quails (genus Coturnix, Galliformes) have been used in behavioural studies (Mills et al. 1997) and reproductive ageing studies (Ottinger et al 2004). The Passeriformes, accounting for about 60 % of the extant bird species (Poole 1999), also includes several model species, especially popular with students of sexual selection and animal communication, such as Sturnus vulgaris (starling) (Gentner et al. 2006), Passer domesticus (house sparrow) (Bonneaud et al. 2006) and Taeniopygia guttata (zebra finch) (Johnson and Whitney 2005, Birkhead 1995). Captive handling information is available for representatives of a few additional groups, which should aid comparative developmental studies undertaken by evo-devo scientists: Tyto alba (barn owl) and Athene noctua (little owl) of the Strigiformes are successfully bred in captivity, the widespread Columba livia (pigeon) would be an obvious model species for Columbiformes, whereas Branta bernicla (brent goose) and numerous pet parrot species could be used as models of Anseri-formes and Psittaciformes, respectively (Poole 1999). Finally, Struthio camelus (the ostrich, Struthioniformes) is now bred in farms worldwide, making it possible to investigate development in this species (e.g. in relation to the loss of flight abilities).



Egg-laying mammals include only two genera comprising one species of Ornithorhynchidae and two species of Tachyglossidae. Ornithorhynchus anatinus (the platypus) is very difficult to maintain and breed in captivity despite efforts to do so starting as early as the 1800s (Temple-Smith and Grant 2001). On the other hand, Tachyglossus aculeatus (short-beaked echidna) is a reasonably common animal in zoos throughout the world and can be bred reasonably easily in captivity. Echidnas could serve as model species not only for their peculiar reproductive features but also to investigate other spectacular traits, such as the presence of spines and the characters associated with their ability to feed on ants (Jackson 2003). Obviously, the accessibility to a Monotremata species for evo-devo studies would be of tremendous interest given their phylo-genetic position and the retention of ancestral traits in their anatomy in general and their reproductive system in particular. Unfortunately, many zoos are disinclined to provide material given the restricted availability of animals outside Australia. The low-coverage assembly of the platypus genome has recently been made available.


Monodelphis domestica (the grey short-tailed opossum) is the marsupial species most often used in laboratories because of its small size, large reproductive output and short generation time (males and females reach sexual maturity before 9 months). The species has been used to study the development of some marsupial traits, such as the formation of the placenta (Freyer et al. 2002), of the median vagina and the pouch (Regli and Kress 2002, Kress et al. 2004), of the neural plate, neural crest and facial region (Smith 2001), and of male morphology (Wilson et al. 2002). It is also used as a model for studying cutaneous melanoma induced by ultraviolet radiation (Wang et al. 2004). One of the great advantages of using marsupials as model species is the accessibility, from an early developmental stage, of the embryos attached to the mother's nipples. Finally, Monodelphis domestica has the great advantage of having had its genome completely sequenced (www.broad.mit.edu/ mammals/opossum). Disadvantages of the species are its aggressive behaviour beyond the age of 4 months, its preference for solitariness (Poole 1999) and the difficulty of obtaining animals outside the United States.

We suggest that Petaurus breviceps (the sugar glider, see Figure 7.2E) should also be considered as an additional valid model species. Its size and reproductive output are similar to those of the grey short-tailed opossum but, unlike the latter, it is a social animal so that large groups can be maintained in captivity. Sugar gliders can breed all year around through artificial regulation of the photoperiodicity associated with their reproductive cycle (Jackson 2003). As they are becoming popular pets all around the world, they are much easier to find than short-tailed opossums. Finally, sugar gliders have a skin membrane, called patagium, extending from the forelimb to the hind foot and used to glide between trees. This anatomical feature is of interest in the evo-devo context because it evolved independently in three marsupial families, two rodent families and the flying lemurs.



The representatives of this major clade (Figure 7.1) of mammals exhibit a spectacular diversity of sizes and morphologies, from the tiny Microgale species (about 7 grams) to the Asian and African elephants (about 7 tons). Despite the fact that there is, to our knowledge, no known single morphological character that would clearly support that clade, analyses of multiple molecular sequence datasets unambiguously support its existence (Stanhope et al. 1998, Murphy et al. 2001a, van Dijk et al. 2001) as a lineage grouping elephants, sea cows, hyraxes, aardvarks, elephant shrews, golden moles and tenrecs. Literature on afrotherian mammals is scarce, but reliable data (originating mainly from zoos all around the world) are available on their maintenance and reproduction, such that evo-devo studies would almost certainly require collaborations with zoological institutions. Given their size and long generation time, the Tubulidentata (aardvark), Proboscidea (elephants) and Sirenia (dugongs, sea cows and manatees) will obviously be the most difficult to investigate. On the other hand, representatives of the Afrosoricida (tenrecs), Macroscelidea (elephant shrews) and Hyracoidea (hyraxes) are much more amenable to housing and breeding in the laboratory. Tenrecs are particularly interesting as they retain some primitive mammalian features, such as the lack of scrotum in the male or the presence of a cloaca in females (as in marsupials and monotremes). In addition, several species of tenrecs have spines, a character that has convergently evolved in representatives of other mammalian lineages. Finally, tenrecs could constitute useful models for understanding the mechanisms associated with endothermy as their thermoregulation is less efficient than in other eutherians. We suggest Echinops telfairi (lesser hedgehog tenrec, see Figure 7.2F) as the model species of choice because (1) it is one of the smallest species for which breeding colonies already exist in research facilities, (2) it has spines macroscopically very similar to those of hedgehogs, and (3) extensive studies have been undertaken on the development of the placenta (Carter et al. 2004), as well as the brain structure and physiology (Kunzle 2006). Furthermore, the genome of E. telfairi has been recently sequenced.

For elephant shrews (19 living species), only some anatomical and physiological studies are available (e.g. organisation of the somatosen-sory cortex in Elephantulus edwardii [Dengler-Crish et al. 2006] and thermoregulation in E. myurus [Mzilikazi and Lovegrove 2004, 2006]). Very little information on the reproduction of these species is available, although they have been bred in the past (Rathbun et al. 1981). An alternative could be Macroscelides proboscideus; less information is available on its biology but established colonies can be found in numerous zoos around the world.

For the Hyracoidea, Procavia capensis seems the obvious choice given that it is successfully bred in zoos (although gestation takes about 7 months), it is relatively small, it is a social animal, and its full genome might soon be sequenced (www.genome.gov/10002154).


Representatives of this mammalian group (anteaters, sloths and armadillos) are confined to South America and are highly specialised animals. Anteaters and sloths are very difficult to study owing to their large size and special requirements for maintenance, whereas armadillos could be bred in laboratory facilities. The armour shell of armadillos, a structure consisting of bony scutes covered with keratinous plates, is of interest for evo-devo research, as is an idiosyncrasy of the reproductive strategy of the genus Dasypus: litters typically consist of four clones as the offspring develop from a single fertilised egg. Breeding in captivity has proven difficult, although not impossible (Carvalho et al. 1997). The nine-banded armadillo (Dasypus novemcinctus) genome sequence has recently been made available, although many genes are interrupted or missing because of the low-coverage (2x) assembly: each base in the final sequence is present, on average, in two reads only, as opposed to more than six reads for other sequenced species such as human or mouse.


This large grouping (Figure 7.1) includes the best-studied mammals as it comprises the Rodentia (mice, rats, squirrels, gophers, porcupines, beavers, etc.), Lagomorpha (rabbits, hares and pikas), Primates (lemurs, tarsiers, new and old-world monkeys), Scandentia (tree shrews) and Der-moptera (flying lemurs). Only the latter group has not been extensively studied given the difficulty of keeping these animals in captivity.

Rodentia. Obviously, the laboratory mouse and rat dominate the literature in numerous fields of experimental biology but quite a series of additional species are used as model species for medical and academic research such as, among others: Microtus agrestis (short-tailed field vole), Meriones unguiculatus (laboratory gerbil), and Mesocricetus auratus and Cricetus grisues (Syrian and Chinese hamster, respectively) (Poole 1999, Cavanagh et al. 2004, Mand et al. 2006, Shimozuru et al. 2006). Finally, members of the genus Cavia (guinea pigs) have been used in sociophysiology studies (Sachser 1998) and are very valuable models for ascorbic acid metabolism because they are the only non-primate laboratory animals that require a dietary source of vitamin C (Burk et al. 2006). Guinea pigs are of particular interest in evo-devo studies because they are representatives of the hystricomorph rodents with the largest amount of available information regarding their biology as well as captive holding and breeding. As well as the high-coverage genome sequences of the laboratory mouse and rat, the low-coverage genome sequence of Cavia porcellus is now also available.

Lagomorpha. This group is represented by a widely used model species, Oryctolagus cuniculus (the laboratory rabbit) for which the full genome sequence is available. Rabbits are widely used for antibody production and have served as a model organism in many physiological and immunological studies. An emerging additional lagomorph model species is Ochotona curzoniae (the black-lipped pika) which could be useful for comparative studies given that it belongs to a different family from the laboratory rabbit. Females can breed and produce litters every three months in the summer, with the size of the litter depending on environmental conditions (Dobson et al. 1998). Little empirical information is available regarding its captive breeding, but specimens are available in the pet trade. Genome sequencing has been proposed for this species (www.genome.gov/10002154).

Primates. The most frequent and obvious justification for the use of primate species as models is their close phylogenetic relationship with humans. However, beside multiple and significant ethical problems associated with their use as model species (owing to this close relationship with humans), the high demands in maintenance of primates, as well as their vulnerable status in nature, make them difficult and expensive to acquire, breed and rear. Full genome sequences are available for human, the chimpanzee (Pan troglodytes), the rhesus macaque (Macaca mulatta) and the bushbaby (Otolemur garnettii).

Scandentia. Tree shrews form a small group (20 species in five genera) of small-sized mammals that are native to tropical Southeast Asia. The vast majority of experimental work has been done on their visual system because tree shrews are considered good models for understanding the neural organisation of the early primate visual system (Dren-haus et al. 2006). The focal tree shrew model species has always been Tupaia belangeri, colonies of which have been successfully established in laboratories (Poole 1999). Its value as a medical research model animal and its phylogenetic position as an outgroup of primates are the two main reasons for Tupaia belangeri being a genome sequencing target.


This large grouping (Figure 7.1) comprises Carnivora, Pholidota, Perisso-dactyla, Cetartiodactyla, Chiroptera and Eulipotyphla.

Pholidota. Pangolins are very difficult (if not impossible) to breed in captivity, so that observations can only be based on opportunistic sampling. The situation is particularly unfortunate from an evo-devo perspective because the representatives of this extraordinary group (with only seven living species found in the tropical regions of Africa and Asia) show spectacular features such as, among others, a highly modified skull, an extraordinarily long and muscular tongue, and a set of overlapping scales covering their head, back and tail.

Perissodactyla (horses, tapirs and rhinoceroses) and Cetartiodactyla (camels, ruminants, pigs, hippopotamuses, whales, dolphins and porpoises). The large size of most of these mammals complicates their use as evo-devo model species. The situation is, however, easier for families represented by domesticated species (horses, cows, sheep, goats) for which multiple inbred lines, much developmental information, genome engineering techniques and full genome sequences (for pig and cow) are available. Furthermore, a recent investigation (Thewissen et al. 2006) of gene expression during early development of hind-limb buds in embryos of the pantropical spotted dolphin, Stenella attenuata (providing data on the molecular basis for hind-limb loss during cetacean evolution), demonstrates that a good museum collection of embryo specimens can partially but efficiently compensate for the lack of laboratory animal models.

Carnivora. The diversity of phenotypes among the multiple inbred lines of dogs (Canis familiaris) has already been successfully exploited in evo-devo studies (e.g. Fondon and Garner 2004), and representatives of other families (such as mustelids) are becoming increasingly available as possible laboratory model species. Dog and cat full genome sequences are available in public databases.

Eulipotyphla. True insectivores are small, highly mobile animals with long, narrow and often elaborate snouts, adapted to eating insects. Low-coverage assemblies of the European shrew (Sorex araneus) and the western European hedgehog (Erinaceus europaeus) genomes have recently been made available. The African pygmy hedgehog (Atelerix albiventris, Figure 7.2G) would have probably been a better choice of target species for genome sequencing because representatives of this species are small (average size of 20 cm) and tame animals that can be bred all year around in laboratory conditions. They are partially covered with spines, opening the possibility of investigating the development of this trait convergent with the situation observed in some tenrec species (see above). In fact, multiple convergences have evolved between representatives of Eulipotyphla and Afrotheria: Erinaceidae (hedgehogs) versus Echinops (Madagascar 'hedgehogs'), Soricidae (shrews) vs. Microgale (shrew-like tenrecs), and Talpidae (moles) vs. Chrysochloridae (golden moles). Finally, it would be particularly appealing to identify a model species for the subterranean insectivores. These species exhibit derived traits such as reduced eyes (sometimes covered by skin) and lack of external ears, as well as modified limb morphology. Unfortunately, no information is available on the maintenance of breeding colonies for any of these species, although it is possible to keep them for long periods in captivity (Borroni et al. 1999).

Chiroptera. Bats have acquired a number of spectacular evolutionary innovations associated with their adaptation to active flight. Microbats have a number of interesting features including heavily modified limbs (Sears et al. 2006), the ability to echolocate (Smotherman and Metzner 2005), precise control over both thermoregulation and gestation (Badwaik and Rasweiler 2001), and an unexpectedly long life span (Dobson 2003). Thus far, the little brown bat (Myotis lucifu-gus) has been proposed as a model species because of its small size and its availability. BAC libraries already exist for the little brown bat, and its full genome sequence has recently been assembled with low coverage. An alternative valid model species for bats would be Carollia perspicillata (short-tailed fruit bat) whose maintenance can be readily adjusted to laboratory facilities and whose reproduction takes place at any time of the year (Kiefer 2006). Its embryonic developmental stages have been recently described and illustrated, a fundamental tool for evo-devo studies (Cretekos et al. 2005). This basic information allowed further analyses of the development of bat wing digits, providing a plausible scenario as to how bats acquired powered flight soon after their divergence from other mammals about 65 million years ago (Sears et al. 2006).

Was this article helpful?

0 0


  • Joshua
    Are sugar gliders model organisms?
    8 years ago

Post a comment