Troglobites (obligate cave-dwelling organisms) are population examples of regressive evolution, the repeated degeneration or loss of derived traits. For example, many independent lineages have lost their eyes and are depigmented, such as millipedes, crustaceans, fishes, and salamanders. These phenotypic changes are thought to have evolved in response to the challenging environmental conditions found in subterranean habitats, particularly the absence of light.
Despite the long recognition of regressive evolution in subterranean organisms, there has been surprisingly little evidence for regressive evolution at the molecular in genes associated with regressive characters (e.g., genes involved in the development and function of the eye). The best examples include studies of eye genes in subterranean diving beetles , marsupial moles , naked mole rats , Astanyax cavefish , and cave-roosting bats . Several studies of perhaps the most well-studied eye genes, the retinal photoreceptor gene rhodopsin, in subterranean organisms have yet to uncover compelling evidence for loss of gene functionality at the molecular level. For example, Kim et al.  discovered at least 19 genes associated with visual perception that were either lost or showed evidence for loss of function in the naked mole rat but rhodopsin was not one of these loci.
What might account for apparent functionality of rhodopsin in the subterranean organisms studied previously? One hypothesis is weak selection for the retention of light-sensing abilities [3, 6]. Another possible explanation is that rhodopsin has other pleiotropic functions, such as in circadian rhythms [7-9]. Support for this latter hypothesis was demonstrated in a recent study by Shen et al.  who discovered that invertebrate rhodopsin was important in initiating thermosensory-signaling cascades in Drosophila fruit flies.
An alternative hypothesis to explain presumed functionality in rhodopsin is that the gene no longer has a function in aphotic environments but insufficient time has passed for the accumulation of loss of function (LOF) mutations to accumulate and render the gene nonfunctional. This hypothesis assumes that selection is not the mechanism behind loss of functionality . If neutral processes are responsible for loss of functionality, then integrity of a gene may persist for a considerable period of time due to chance alone . Consequently, inferred maintenance of a visual gene may be the result of recent subterranean colonization rather than retained or pleiotropic functionality.
My colleagues (Drs. Ben Fitzpatrick, Premal Shah, Lars Schmitz and Tom Near) and I investigated this latter hypothesis by testing for loss of selective constraint in rhodopsin of amblyopsid cavefishes in a recently accepted paper in the journal Evolution . We first generated a fossil-calibrated, multilocus molecular phylogeny based on nine genes to provide a temporal and phylogenetic context for elucidating the evolutionary history and potential loss of rhodopsin functionality. Our sampling included every recognized species of amblyopsid cavefish (family Amblyopsidae) as well as related outgroup taxa within the order Percopsiformes (families Aphredoderidae and Percopsidae). The resulting phylogeny (Figure 1) differed considerably from previously phylogenetic hypotheses [13-15], with the most notable difference being the phylogenetic placement of the surface/facultative cave-dwelling genus Forbesichthys. Forbesichthys was nested within a clade of all obligate cave-dwelling lineages and diverged some 5.7 Mya. The most recent common ancestor (MRCA) of all amblyopsids (cave and surface lineages) dates to 12.2 Mya in the Miocene while the MRCA of all cave lineages dates to 10.3 Mya. However, much of the diversification within genera occurred in the Pleistocene.
The intriguing phylogenetic placement of Forbesicthys suggests that a surface-dwelling life history and eye functionality may have reevolved from a cave-dwelling ancestor with degenerate eyes. In fact, ancestral character reconstructions of eye functionality in amblyopsid cavefishes strongly support reevolution in Forbesichthys from a subterranean ancestor. Cave-dwelling organisms are often viewed as "evolutionary dead-ends" that are incapable of recolonizing or adapting to surface habitats. However, the reevolution of a surface form is not novel, as this hypothesis has been proposed for eyed Gammarus amphipods living in karst windows  as well as in cave scorpions , cavefishes  and cave salamanders .
However, patterns of molecular evolution in rhodopsin suggest a different evolutionary history (Figure 2). We found no evidence of loss of selective constraint or functionality in rhodopsin of surface lineages, which was to be expected given that surface-dwelling species actively respond to photic stimuli. However, three cave lineages (Troglichthys rosae, Amblyopsis spelaea and Typhlichthys cf. subterraneus TN) exhibited strong evidence for loss of rhodopsin functionality at the molecular level including ten novel LOF mutations (six deletions, one insertion, and three mutations resulting in premature stop codons). These LOF mutations were highlighted by an 111 amino acid deletion in the federally listed Troglichthys rosae. In addition, rhodopsin in cave lineages (even those without LOF mutations) showed increased rates of nonsynonymous (amino acid changing) mutations compared to surface lineages. These mutations were more likely to have a significant deleterious effect on the structure and subsequently function of rhodopsin in cave lineages, as inferred from several physicochemical protein properties.
As a final test to determine if rhodopsin in cave lineages was evolving at a different evolutionary rate (i.e., under neutrality versus balancing or positive selection) compared to surface lineages, we compared a series of branch-based models of selection using the multilocus phylogeny. Interestingly, the best model corresponded to independent evolution of relaxation of selection and loss of functionality in cave lineages rather than reevolution of rhodopsin functionality in Forbesichthys that was suggested based on ancestral character reconstructions. In fact, not a single nonsynonymous mutation was observed in the 4.6 My interval from the MRCA of all cave amblyopsids and the MRCA of Forbesichthys and Amblyopsis (node b-e in Figure 1), as would be expected if there was a single subterranean colonization event and subsequent reevolution of eyes in Forbesichthys. Other lines of evidence also support independent evolution of loss of eye (and rhodopsin) functionality over a reevolution scenario, including patterns of degeneration of individual eye structures based on histological data, biogeographical and geological evidence, and phylogenetic studies of other cave organisms.
In summary, our study provides compelling evidence for repeated loss of selective constraint in rhodopsin in cave amblyopsids. Although we cannot rule out selection as the primary mechanism behind eye degeneration (i.e., selection might act upstream of rhodopsin or at other important eye developmental loci), our results are consistent with the neutral accumulation of mutations responsible of loss of functionality in rhodopsin. Rather than some unknown pleiotropic function, we hypothesize that presumed functionality of rhodopsin in other cave organisms is due to recent subterranean colonization and the random nature of mutation accumulation.
1. Leys R, Cooper SJB, Strecker U, Wilkens H. 2005. Regressive evolution of an eye pigment gene in independently evolved eyeless subterranean diving beetles. Biol. Lett. 1: 496-499.
2. Springer MS, Burk A, Kavanagh JR, Wadell VG, Stanthope MJ. 1997. The interphotoreceptor retoid binding protein gene in therian mammals: implications for higher level relationships and evidence for loss of function in the marsupial mole. Proc. Natl. Acad. Sci. USA 94: 13754-13759.
3. Kim EB et al. 2011. Genome sequencing reveals insights into physiology and longevity of the naked mole rat. Nature 479: 223-227.
4. Yokoyama S, Meany A, Wilkens H, Yokoyama R. 1995. Initial mutational steps toward loss of opsin gene function in cavefish. Mol. Biol. Evol. 12: 527-532.
5. Zhao H, Rossiter SJ, Teeling E, Li C, Cotton JA, Zhang S. 2009. The evolution of color vision in nocturnal mammals. Proc. Natl. Acad. Sci. USA 106: 8980-8985.
6. Culver DC, Pipan T. 2009. Biology of Caves and Other Subterranean Habitats. Oxford University Press.
7. Crandall KA, Hillis DM. 1997. Rhodopsin in the dark. Nature 387: 667-668.
8. Janssen JWH et al. 2000. A fully functional rod visual pigment in a blind mammal. A case for adaptive functional reorganization? J. Biol. Chem. 275: 38674-38679.
9. Li Z, He S. 2009. Relaxed purifying selection of rhodopsin gene within a Chinese endemic cavefish genus Sinocyclocheilus (Pisces: Cypriniformes). Hydrobiologia 624: 139-149.
10. Shen WL, Kwon Y, Adegbola AA, Luo J, Chess A, Montell C. 2011. Function of rhodopsin in temperature discrimination in Drosophila. Science 331: 1333-1336.
11. Marshall CR, Raff EC, Raff RA. 1994. Dollo's law and the death and resurrection of genes. Proc. Natl. Acad. Sci. USA 91: 12283-12287.
12. Niemiller ML, Fitzpatrick BM, Shah P, Schmitz L, Near TJ. In press. Evidence for repeated loss of selective constraint in rhodopsin of amblyopsid cavefishes (Teleostei: Amblyopsidae). Evolution.
13. Woods LP, Inger RF. 1957. The cave, spring, and swamp fishes of the family Amblyopsidae of central and eastern United States. Am. Midl. Nat. 58: 232-256.
14. Swofford DL. 1982. Genetic variability, population differentiation, and biochemical relationships in the family Amblyopsidae. Master's thesis. Eastern Kentucky University, Richmond, KY.
15. Niemiller ML, Poulson TL. 2010. Subterranean fishes of North America: Amblyopsidae. Pp. 169-280 in Trajano E, Bichuette ME, Kappor BG (eds.). The Biology of Subterranean Fishes. Science Publishers, Enfield, NH.
16. Culver DC, Kane TC, Fong DW. 1995. Adaptation and Natural Selection in Caves: the Evolution of Gammarus minus. Harvard University Press, Cambridge, UK.
17. Prendini L, Francke OF, Vignoli V. 2010. Troglomorphism, trichobothriotaxy and typhlochactid phylogeny (Scorpiones, Chactoidea): more evidence that troglobitism is not an evolutionary dead-end. Cladistics 26: 117-142.
18. Dillman CB, Bergstrom DE, Noltie DB, Holtsford TP, Mayden RL. 2011. Regressive progression, progressive regression or neither? Phylogeny and evolution of the Percopsiformes (Teleostei, Paracanthopterygii). Zool. Scr. 40: 45-60.
19. Trajano E, Cobolli M. 2012. Evolution of lineages. Pp. 295-304 in White WB, Culver DC (eds.). Encyclopedia of Caves, 2nd ed. Academic Press, Oxford.