PDF(Pubmed)
Abstract：
Golden moles (Chrysochloridae) and marsupial moles (Notoryctidae) are textbook examples of convergent evolution. Both taxa are highly adapted to subterranean lifestyles and have powerful limbs for digging through the soil/sand, ears that are adapted for low-frequency hearing, vestigial eyes that are covered by skin and fur, and the absence of optic nerve connections between the eyes and the brain. The eyes of marsupial moles also lack a lens as well as retinal rods and cones. Two hypotheses have been proposed to account for the greater degeneracy of the eyes of marsupial moles than golden moles. First, marsupial moles may have had more time to adapt to their underground habitat than other moles. Second, the eyes of marsupial moles may have been rapidly and recently vestigialized to (1) reduce the injurious effects of sand getting into the eyes and (2) accommodate the enlargement of lacrimal glands that keep the nasal cavity moist and prevent the entry of sand into the nasal passages during burrowing. Here, we employ molecular evolutionary methods on DNA sequences for 38 eye genes, most of which are eye-specific, to investigate the timing of relaxed selection (=neutral evolution) for different groups of eye-specific genes that serve as proxies for distinct functional components of the eye (rod phototransduction, cone phototransduction, lens/cornea). Our taxon sampling included 12 afrothere species, of which two are golden moles (Amblysomus hottentotus, Chrysochloris asiatica), and 28 marsupial species including two individuals of the southern marsupial mole (Notoryctes typhlops). Most of the sequences were mined from databases, but we also provide new genome data for A. hottentotus and one of the two N. typhlops individuals. Even though the eyes of golden moles are less degenerate than the eyes of marsupial moles, there are more inactivating mutations (e.g., frameshift indels, premature stop codons) in their cone phototransduction and lens/cornea genes than in orthologous genes of the marsupial mole. We estimate that cone phototransduction recovery genes were inactivated first in each group, followed by lens/cornea genes and then cone phototransduction activation genes. All three groups of genes were inactivated earlier in golden moles than in marsupial moles. For the latter, we estimate that lens/cornea genes were inactivated ~17.8 million years ago (MYA) when stem notoryctids were burrowing in the soft soils of Australian rainforests. Selection on phototransduction activation genes was relaxed much later (5.38 MYA), during the early stages of Australia\'s aridification that produced coastal sand plains and eventually sand dunes. Unlike cone phototransduction activation genes, rod phototransduction activation genes are intact in both golden moles and one of the two individuals of N. typhlops. A second marsupial mole individual has just a single inactivating mutation in one of the rod phototransduction activation genes (PDE6B). One explanation for this result is that some rod phototransduction activation genes are pleiotropic and are expressed in extraocular tissues, possibly in conjunction with sperm thermotaxis.
摘要：
金痣（Chrysochilidae）和有袋痣（Notoryctidae）是收敛进化的教科书示例。这两个类群都高度适应地下生活方式，并且有强大的四肢可以挖掘土壤/沙子，适合低频听力的耳朵，被皮肤和毛皮覆盖的残留眼睛，眼睛和大脑之间没有视神经连接。有袋痣的眼睛也缺乏晶状体以及视网膜杆和视锥细胞。已经提出了两个假设来解释有袋痣眼睛的退化性比金色痣更大。首先,有袋痣可能比其他痣有更多的时间来适应它们的地下栖息地。第二,有袋痣的眼睛可能已经迅速和最近被恢复，以(1)减少沙子进入眼睛的伤害作用，(2)适应泪腺的扩大，保持鼻腔湿润，防止沙子进入鼻腔在挖洞过程中通道。这里,我们对38个眼部基因的DNA序列采用分子进化方法，其中大多数是眼睛特有的，为了研究不同组的眼睛特异性基因的放松选择（=中性进化）的时机，这些基因充当眼睛不同功能成分的代理（视杆光转导，锥形光传导，晶状体/角膜）。我们的分类群采样包括12种非洲物种，其中两个是金色痣(Amblysomushotentotus，积雪草)，和28种有袋动物，包括南部有袋痣的两个个体（Notoryctestyphlops）。大多数序列是从数据库中挖掘的，但我们也提供了一种新的基因组数据。即使金色痣的眼睛比有袋痣的眼睛退化得少，有更多的失活突变(例如，移装者，过早停止密码子）的视锥光转导和晶状体/角膜基因比有袋痣的直系同源基因。我们估计视锥光转导恢复基因在每组中首先失活，其次是晶状体/角膜基因，然后是视锥光转导激活基因。所有三组基因在金痣中比在有袋痣中更早失活。对于后者,我们估计约1780万年前（MYA），当茎状线虫在澳大利亚雨林的软土中挖洞时，晶状体/角膜基因已失活。光转导激活基因的选择放宽得多(5.38MYA)，在澳大利亚干旱化的早期阶段，产生了沿海沙平原，最终形成了沙丘。与视锥细胞光转导激活基因不同,杆状光转导激活基因在金色痣和斑驳虫的两个个体之一中都是完整的。第二个有袋痣个体在一个杆状光转导激活基因（PDE6B）中只有一个失活突变。对此结果的一种解释是，一些杆状光转导激活基因具有多效性，并在眼外组织中表达，可能与精子热轴有关。
