Striedter G F
Department of Psychobiology and Center for the Neurobiology of Learning and Memory, University of California at Irvine, Irvine, CA 92697-4550, USA.
Brain Behav Evol. 1998;52(4-5):218-31. doi: 10.1159/000006565.
The reunification of embryology with evolutionary biology is impeded by the perception that a phylogenetic view of homology is incompatible with a developmental approach. This dichotomy disappears when developmental information is viewed not as pre-existing within the zygote but as being constructed during development. Developmental information can be depicted as the surface of an epigenetic landscape. An epigenetic landscape, in turn, can be viewed as a series of aligned energy landscapes that change shape and become more complex as development proceeds. In this view, individual valley bottoms are attractors in state space that tend to reappear reliably in successive generations. Phylogeny can therefore be conceptualized as a succession of epigenetic landscapes, and homologues can be identified as corresponding valleys that have reappeared reliably since their origin in a single ancestral population. Epigenetic homologues can be robust to phylogenetic changes in developmental mechanisms, precursors, and lower level characters. Although application of the epigenetic homology concept is complicated by the lack of explicit information about the topography of epigenetic landscapes, comparative biologists can learn to identify recurring ontogenetic patterns in a manner that is analogous to the identification of input patterns by attractor neural networks. The correspondence of epigenetic valleys is therefore not defined by any essential criteria but by their overlap in multidimensional state space. Whether corresponding valleys are homologous to each other must be determined by a phylogenetic analysis using cladistic methods. Among the general implications of epigenetic homology for comparative neurobiology is that the concept of 'field homology' should be used with caution when dealing with novel characters. A case study, applying an epigenetic perspective to understand the variation in monkey visual cortex observed after developmental perturbations, is presented in a final section to make the concept of epigenetic homology more concrete.
胚胎学与进化生物学的重新统一受到这样一种观念的阻碍,即系统发育视角下的同源性与发育方法不相容。当发育信息不被视为预先存在于合子中,而是在发育过程中构建时,这种二分法就消失了。发育信息可以被描绘为表观遗传景观的表面。反过来,表观遗传景观可以被看作是一系列对齐的能量景观,它们随着发育的进行而改变形状并变得更加复杂。按照这种观点,各个谷底是状态空间中的吸引子,往往会在连续几代中可靠地重现。因此,系统发育可以被概念化为一系列表观遗传景观的连续,同源物可以被识别为自其起源于单一祖先种群以来可靠重现的相应谷底。表观遗传同源物对于发育机制、前体和较低层次特征的系统发育变化可能具有稳健性。尽管由于缺乏关于表观遗传景观地形的明确信息,表观遗传同源性概念的应用变得复杂,但比较生物学家可以学会以类似于吸引子神经网络识别输入模式的方式来识别反复出现的个体发育模式。因此,表观遗传谷底的对应关系不是由任何本质标准定义的,而是由它们在多维状态空间中的重叠来定义的。相应的谷底彼此是否同源必须通过使用分支分类方法的系统发育分析来确定。表观遗传同源性对比较神经生物学的一般影响之一是,在处理新特征时应谨慎使用“场同源性”概念。最后一部分给出了一个案例研究,应用表观遗传学观点来理解发育扰动后观察到的猴子视觉皮层的变化,以使表观遗传同源性概念更加具体。