Although horses, elk and lions were at one time “model species” for medical research (Logan, 2002
), modern science has settled on some standard species. The chart in depicts a distribution used in neurodevelopmental research in articles published 2005–2006 in nine model mammals. The base of knowledge developed for each species itself quickly becomes a factor in the choice of which species to use, particularly if there is no convenient way to closely compare results between species.
The chart depicts the proportion of recent studies performed in nine of the most commonly used species (2005– 2006 to date; Medline title search).
Each experimental species has its own advantages. Rodent dams have large litters that are easy to care for, generally disease-resistant, and have no agricultural uses. Rat pups are born after a short gestation (22.5 d), and have long been the general species of choice such that rat macro- and micro-neuroanatomy, neurophysiology, and assessments of behavior are well-mapped. Mice (gestation 19.5d) are currently considered most amenable to genetic manipulations, initially chosen because manipulations were clearly reflected in their physical appearance (Nishioka, 1995
). Hamster pups (gestation 15.5d) are born earlier in their somatic development than other rodents so it is easier to study their early neurological development. These three species are considered altricial - born at relatively underdeveloped stages with eyes shut, and many neurogenic events occur postnatally.
Ferrets (gestation 42d) are also born early in development, but have larger brains, more comparable to humans. Cats, also with large brains, have visual systems that resemble humans, and are born at a somewhat later developmental point (gestation 65d). Rabbits (gestation 31d) were the initial species of choice for toxicology studies because the absence of tear ducts permits contaminant responses to develop quickly.
In contrast, guinea pigs (gestation 65 d) and spiny mice (gestation 40d) are “precocial” –close to independence at birth. Born with eyes open, guinea pigs even shed their baby teeth in utero
, and are useful for studies of behavioral abilities that may develop without experience. Rhesus macaques, the leading laboratory primate species (gestation 165d ), are also considered precocial. Babies are born with eyes open and relatively advanced motoric abilities; they are most directly related to humans with (95% genome homology) (Rogers et al., 2006
Researchers are thus required to assimilate a perplexing quantity of data collected at varying times across a wide developmental spectrum () and relate it to developing humans (gestation 280 d). The nature of human development further confuses any comparison with laboratory models. Human newborns, if classified by the immature development of their body and motoric skills, should be considered an altricial species, but the relatively advanced development of the human brain and many aspects of perceptual systems at birth clearly places human neural development in a precocial category (Clancy et al., 2000
; Verley, 1977
The graph depicts the wide gestation range for mammalian species commonly used as experimental models and studied to satisfy a variety of scientific contingencies.
Moreover, a significant calibration problem is developing in the rodent literature. Although rats dominate, the recent surge of genomic studies in mice means they represent an increasing proportion of experimental studies (8 years ago, mice accounted for 23% of neurodevelopmental studies, currently they make up 39%). Length of gestation varies in these two rodent species (mice 19.5d, rats 22.5d), but are the 3 additional days mainly employed for rat brain development? Is the solution to cross-species translation between these two as simple as subtracting the difference in gestational ages or as overwhelming as repeating all the experiments done in rats again in mice?
The fact that researchers pick laboratory animal models based on diverse practical grounds arises from an assumption of generalizability across species (Logan, 2002
). The conjunction of evolutionary and developmental biology shows that the timing and sequence of early events in brain development are remarkably conserved across mammals (Finlay and Darlington, 1995
). In fact, the critical periods of prenatal and early postnatal development may be the ideal time in the life span to make the most accurate cross-species comparisons, because conversions become more variable in adult animals with widely different “life histories” (such as seasonal breeding, difficult habitats, etc.). A once promising “rate of living theory” suggested that the total number of heartbeats, breaths and matings are constrained across a lifetime i.e.
the faster a species lived, the faster it would die. But although life span is not unrelated to metabolism (Economos, 1981
), the rate of living theory remains controversial (Burns, 2004
) and is generally discounted (Lints, 1989
Below we review some of the ways the research community has attempted to equate brain development across members of the mammalian species, dividing these studies into 3 general categories 1) morphological comparisons, 2) “rules of thumb” based on susceptibility patterns and 3) event-based comparisons. We include our novel “neuroinformatics” technique to the latter approach (Clancy et al., 2000
; Clancy et al., 2001
; Clancy et al., 2006
; Finlay and Darlington, 1995
), based on three related strengths: 1) acquisition and integration of large databases of multiple data types, 2) analyses using standard multivariate techniques made simpler by increased computing power, and 3) public availability through Web-based interfaces. These tools allow us to make accurate predictions of cross species developmental sequences based on multiple events in multiple species.